Journal of Small Business Management

Editor George T. Solomon, D.B.A. Editor-in-Chief, Journal for Small Business Management (JSBM) Associate Professor of Management, The George Washington University, USA Director, Center For Entrepreneurial Excellence (CFEE) Associate Editors Alain Fayolle, Ph.D., Emlyon Business School, France Alan Carsrud, Ph.D., Åbo Akademi University, Finland David Smallbone, Ph.D., Kingston University, UK Eddy Laveren, University of Antwerp, Belgium Hermann Frank, Ph.D., Vienna University of Economics and Business, Austria Hiroyuki Okamuro, Ph.D., Hitotsubashi University, Japan Howard Van Auken, Ph.D., Iowa State University, USA Joanne Larty, Ph.D., Lancaster University, UK Kate Lewis, Ph.D., Massey University, New Zealand Kelly Shaver, Ph.D., College of Charleston, USA Ki-Chan Kim, Ph.D., The Catholic Unversity of Korea, South Korea Lene Foss, Ph.D., University of Tromso, Norway Marco Cucculelli, Ph.D., Universita Politecnica delle Marche, Italy Mark Freel, Ph.D., University of Ottawa, Canada Mark Schenkel, Ph.D., Belmont University, USA Marko Grunhagen, Ph.D., Eastern Illinois University, USA Martie-Louise Verreynne, Ph.D., The University of Queensland, Australia Massimo G. Colombo, Politecnico di Milano, Italy Norris Krueger, Ph.D., Entrepreneurship Northwest, USA Ulla Hytti, Ph.D., University of Turku, Finland William E. Jackson III, Ph.D., University of Alabama, USA Business Manager Michael Battaglia Business Manager, Journal of Small Business Management (JSBM) Operations Manager, International Council for Small Business (ICSB) The George Washington University, Washington DC, USA Phone: 001-202-994-0704 Email: [email protected]; [email protected] International Research Board (IRB) Ayman El Tarabishy, Ph.D., George Washington University, USA Brian Gibson, Ph.D., University of New England, Australia Charles Matthews, Ph.D., University of Cincinnati, USA Donald F. Kuratko, Ph.D., Indiana University-Bloomington, USA Friederike Welter, Ph.D., Jonk€ oping€ International Business School, Sweden G. Dale Meyer, Ph.D., University of Colorado-Boulder, USA Geralyn McClure Franklin, Ph.D., Stephen F. Austin State University, USA Howard Lin, Ph.D., Ryerson University, Canada Hugo Kantis, Ph.D., Uniersidad Nacional de General Sarmiento, Argentina J. Hanns Pichler, Ph.D., Vienna University of Economics and Business Administration, Austria Jeffrey Cornwall, D.B.A., Belmont University, USA Jerome Katz, Ph.D., St. Louis University, USA Karen L. Orengo Serra, Ph.D., University of Puerto Rico, PR Niels Bosma, Ph.D., Utrecht University, Netherlands Patricia Greene, Ph.D., Babson College, USA Roy Thurik, Ph.D., Erasmus University Gallen, Netherlands Thierry Volery, Ph.D., University of St-Gallen, Switzerland Victoria Calvert, Mout Royal University, Canada Zulma Quinones, Ph.D., Inter American University of Puerto Rico-San German, PR

In March 2006, the Journal of Small Business Management earned the distinction of having the highest percent increase in total citations from January 1995 to December 2005 in the field of Economics & Business, according to Thomson Scientific’s Journal Citation Report for the Social Sciences Edition

Winner of the 1999 Anbar Abstracts Golden Page Award for Originality

Named best publication in 1997 in the continuing publications category by the Association for University Business and Economic Research

The Journal of Small Business Management strives to be the internationally preeminent entrepreneurship and small firm research periodical. JSBM publishes work from around the world that is applicable to scholarly research inquiry in entrepreneurship and small business.

VC 2016 International Council for Small Business

1 Information for subscribers The Journal of Small Business Management is published in 4 issues per year. Institutional subscription prices for 2016 are: Print & Online: US$503 (The Americas), US$777 (Rest of World), £395 (UK), e503 (Europe). Prices are exclusive of tax. Asia-Pacific GST, Canadian GST/HST and European VAT will be applied at the appropriate rates. For more information on current tax rates, please go to www.wileyonlinelibrary.com/tax-vat. The price includes online access to the current and all online back files to January 1st 2011, where available. For other pricing options, including access information and terms and conditions, please visit www.wileyonlinelibrary.com/access. Delivery Terms and Legal Title Where the subscription price includes print issues and delivery is to the recipient’s address, delivery terms are Delivered at Place (DAP); the recipient is responsible for paying any import duty or taxes. Title to all issues transfers FOB our shipping point, freight prepaid. We will endeavour to fulfil claims for missing or damaged copies within six months of publication, within our reasonable discretion and subject to availability. Journal Customer Services: For ordering information, claims and any enquiry concerning your journal subscription please go to www.wileycustomerhelp.com/ask or contact your nearest office. Americas: Email: [email protected]; Tel: +1 781 388 8598 or +1 800 835 6770 (toll free in the USA & Canada). Europe, Middle East and Africa: Email: [email protected]; Tel: +44 (0) 1865 778315. Asia Pacific: Email: [email protected]; Tel: +65 6511 8000. Japan: For Japanese speaking support, Email: [email protected]; Tel: +65 6511 8010 or Tel (toll-free): 005 316 50 480. Visit our Online Customer Get-Help available in 7 languages at www.wileycustomerhelp.com/ask Publisher The Journal of Small Business Management is published by Wiley Periodicals, Inc., Commerce Place, 350 Main Street, Malden, MA 02148; Telephone: 781 388 8200; Fax: 781 388 8210. Production Editor Llany Carigara (email: [email protected]). Electronic Access This journal is available online at Wiley Online Library. Visit wileyonlinelibrary.com to search the articles and register for table of contents e-mail alerts. Access to this journal is available free online within institutions in the developing world through the AGORA initiative with the FAO, the HINARI initiative with the WHO WHO, the OARE initiative with UNEP, and the ARDI initiative with WIPO. For information, visit www.aginternetwork.org, www.who.int/hinari/en/, www.oaresciences.org, www.wipo.org/int/ardi/edn. Wiley’s Corporate Citizenship initiative seeks to address the environmental, social, economic, and ethical challenges faced in our business and which are important to our diverse stakeholder groups. Since launching the initiative, we have focused on sharing our content with those in need, enhancing community philanthropy, reducing our carbon impact, creating global guidelines and best practices for paper use, establishing a vendor code of ethics, and engaging our colleagues and other stakeholders in our efforts. Follow our progress at www.wiley.com/go/citizenship Back issues Single issues from current and recent volumes are available at the current single issue price from cs-journals@ wiley.com. Earlier issues may be obtained from Periodicals Service Company, 351 Fairview Avenue Ste 300, Hudson, NY 12534, USA. Tel: +1 518 822 9300, Fax: +1 518 822 9305, Email: [email protected]. Postmaster: Send all address changes to the Journal of Small Business Management, John Wiley & Sons Inc., C/O The Sheridan Press, PO Box 465, Hanover, PA 17331. Microform The journal is available on microfilm. For microfilm service, address inquiries directly to University Microfilms Library Services, Ann Arbor, MI 48106, USA. Copyright and Photocopying VC 2016 International Council for Small Business. All rights reserved. No part of this publication may be reproduced, stored or transmitted in any form or by any means without the prior permission in writing from the copyright holder. Authorization to copy items for internal and personal use is granted by the copyright holder for libraries and other users registered with their local Reproduction Rights Organisation (RRO), e.g., Copyright Clearance Center (CCC), 222 Rosewood Drive, Danvers, MA 01923, USA (www.copyright.com), provided the appropriate fee is paid directly to the RRO. This consent does not extend to other kinds of copying such as copying for general distribution, for advertising or promotional purposes, for creating new collective works or for resale. Special requests should be addressed to: [email protected]. Advertising: For advertising enquiries, please email Kristin McCarthy (email: [email protected]). Abstracting/Indexing The Journal is indexed by ABI/Inform Database (ProQuest), Academic ASAP, Business ASAP (Thom- son Gale), Business Periodicals Index/Abstracts (HW Wilson), CatchWord, CSA Environmental Sciences & Pollution Manage- ment Database (CSA/CIG), Current Abstracts (EBSCO), Current Contents Collections/ Business (Thomson ISI), Current ContentsVR /Social & Behavioral Sciences (Thomson ISI), EBSCO Online (EBSCO), Emerald Management Reviews (Emerald), InfoTrac, Journal Citation Reports/Social Science Edition (Thomson ISI), OMNIFILE Full Text Mega Edition (OMNIFILE), PAIS: Public Affairs Information Service (CSA/CIG), Proquest 5000 (ProQuest), Proquest Discovery (ProQuest), Proquest Platinum (ProQuest), Proquest Research Library (ProQuest), PsycINFO, Social Sciences Citation IndexVR (Thomson ISI), Social SciSearchVR (Thomson ISI), Student Resource Center College (w/ Academic ASAP), Student Resource Center Gold (w/ Academic ASAP), Student Resource Center Silver (w/ Academic ASAP), Wilson Business Abstracts (HW Wilson), Wilson OmniFile V (HW Wilson). Disclaimer The Publisher, International Council for Small Business, and Editors cannot be held responsible for errors or any consequences arising from the use of information contained in this journal; the views and opinions expressed do not necessarily refl ect those of the Publisher, International Council for Small Business, and Editors, neither does the publication of advertisements constitute any endorsement by the Publisher, International Council for Small Business, and Editors of the products advertised. For submission instructions, subscription and all other information visit: http://onlinelibrary.wiley.com/journal/10.1111/(ISSN)1540-627X ISSN 0047-2778 (Print) ISSN 1540-627X (Online) Printed in the USA by The Sheridan Group The Journal of Small Business Management accepts articles for Open Access publication. Please visit http://olabout.wiley .com/WileyCDA/Section/id-406241.html for further information about OnlineOpen.

2 Journal of Small Business Management*

CONTENTS January 2016 Volume 54 Number 1

External Determinants and Financial Outcomes of an Eco-friendly Orientation in 5 Smaller Manufacturing Firms by Leonidas C. Leonidou, Paul Christodoulides, and Des Thwaites The Resource-Based View in Entrepreneurship: A Content-Analytical Comparison 26 of Researchers’ and Entrepreneurs’ Views by Franz Kellermanns, Jorge Walter, T. Russell Crook, Benedict Kemmerer, and Vadake Narayanan About Gender Differences and the Social Environment in the Development of 49 Entrepreneurial Intentions by Francisco J. Santos, Muhammad Azam Roomi, and Francisco Lin~an Vertical Coopetition and the Sales Growth of Young and Small Firms 67 by Christian Lechner, Birthe Soppe, and Michael Dowling Examining Entrepreneurs’ Presentation Effectiveness in Generating Stakeholder 85 Interest: Observations From the Medical Device Industry by Scott Latham and Steven F. Tello Shaping the Entrepreneurial Mindset: Entrepreneurial Intentions of 102 Business Students in Croatia by Sanja Pfeifer, Natasa Sarlija, and Marijana Zekic´ Susac Effects of Relational Capabilities and Power Asymmetry on Innovativeness and 118 Flexibility of Sub-Sahara Africa Small Exporting Firms by Margaret Jekanyika Matanda, Nelson Oly Ndubisi, and Ferry Jie Linking Processes and Dynamic Capabilities of International SMEs: The Mediating 139 Effect of International Entrepreneurial Orientation by Bernhard Swoboda and Edith Olejnik Managerial Perceptions of SMEs in the Wood Industry Supply Chain on Corporate 162 Responsibility and Competitive Advantage: Evidence from China and Finland by Ning Li, Anne Toppinen, and Marja Lantta The Impact of Entrepreneurship Education: A Study of Iranian Students’ 187 Entrepreneurial Intentions and Opportunity Identification by Saeid Karimi, Harm J. A. Biemans, Thomas Lans, Mohammad Chizari, and Martin Mulder How Does Race and Ethnicity Affect Persistence in Immature Ventures? 210 by Robert E. Freeland and Lisa A. Keister

*A quarterly journal of the International Council for Small Business, JSBM is published in January, April, July, and October. Inquiries regarding manuscripts should be addressed to the Editors, JSBM, George Washington Uni- versity, 2201 G St. NW, Suite 315, Washington, DC 20052, USA; telephone: (202) 994-0704; fax: (202) 994-4930; email: [email protected]; website: www.jsbm.org. Inquiries about circulation, subscription, renewals, back copies, bulk sales, article reprints, manuscripts, advertising, and permission requests should be addressed to Wiley- Blackwell.We are grateful to Office Depot Inc. for their generous support. ISSN: 0047-2778

3 Costly Ties: Social Capital as a Retardant of Network-Level Entrepreneurial Orientation 229 by Joakim Wincent, Sara Thorgren, and Sergey Anokhin How Small Business Advisory Program Delivery Methods (Collective Learning, 244 Tailored, and Practice-Based Approaches) Affect Learning and Innovation by Sukanlaya Sawang, Rachel Parker, and Damian Hine Small and Medium-Sized Enterprise’s Entrepreneurial versus Market Orientation 262 and the Creation of Tacit Knowledge by Ana Perez-Luno,~ Patrick Saparito, and Shanti Gopalakrishnan The Franchise Dilemma: Entrepreneurial Characteristics, Relational Contracting, 279 and Opportunism in Hybrid Governance by Heiner Evanschitzky, Barbara Caemmerer, and Christof Backhaus The When and Why: Student Entrepreneurial Aspirations 299 by Caleb Kwong and Piers Thompson Entrepreneurs as Sophisticated Iconoclasts: Rational Rule-Breaking 319 in an Experimental Game by Richard J. Arend Linking Small-Scale Commercial Activities and Women’s Health: The Jamu System 341 in Urban Areas of Java, Indonesia by Maria Costanza Torri Valuation of Angel-Backed Companies: The Role of Investor Human Capital 356 by Veroniek Collewaert and Sophie Manigart Linking Unlearning with Work–Life Balance: An Initial Empirical 373 Investigation into SMEs by Juan-Gabriel Cegarra-Navarro, Maria-Eugenia Sanchez-Vidal, and David Cegarra-Leiva

4 JOURNAL OF SMALL BUSINESS MANAGEMENT Journal of Small Business Business Management Management 2016 2014 54(1), ••(••), pp. pp. 5–25 ••–•• doi: 10.1111/jsbm.12121 External Determinants and Financial Outcomes of an Eco-friendly Orientation in Smaller Manufacturing Firms* by Leonidas C. Leonidou, Paul Christodoulides, and Des Thwaites

With the significant growth of problems relating to the natural environment, an increasing number of firms are engaged in environmentally friendly activities. This paper reports the results of a study conducted among 153 small-sized manufacturing units located in Cyprus, focusing on the external drivers and financial outcomes of their eco-friendly orientation. The findings confirmed the instrumental role of environmental regulations, environmental public concern, competitive intensity, and market dynamism in harnessing an eco-friendly orientation within the small firm. They also underline the critical role of this orientation in enhancing the firm’s financial results, although this link was found to be stronger when the firm possesses adequate resources and capabilities committed to environmental activities. Our study contributes to the literature by putting together, in a theoretically anchored, integrative, and causal fashion, concepts and ideas that touch upon important aspects of small firms’ environmental behavior that have been only tangentially studied in the past, namely the role of external forces in stimulating sensitivity to green issues, the dynamics of forming an ecological orientation within the organization, the positive link between eco-friendly thinking and financial performance, and the importance of supporting ecological actions with appropriate organizational resources/ capabilities. Critical implications for small business managers and public policymakers are also derived from the study findings.

and declining biodiversity (Banerjee 2002; Introduction Gadenne, Kennedy, and McKeiver 2009; The rapid pace of industrialization in recent Shrivastava 1995). Although the individual decades has been responsible for the signiﬁ- effect of small manufacturing ﬁrms on harming cant growth of problems associated with the the environment is negligible, their collective natural environment, such as global warming, impact is substantial, estimated at approxi- water pollution, soil erosion, ozone depletion, mately two-thirds of total industrial pollution

*The authors would like to thank Yiota Apostolidou, Anna-Maria Christoﬁ, and Anna-Maria Piperidou for their assistance in collecting the study data. Leonidas C. Leonidou is professor of Marketing in the School of Economics and Management at the University of Cyprus. Paul Christodoulides is senior lecturer in Applied Mathematics at the Cyprus University of Technology. Des Thwaites is senior lecturer in Marketing at the Leeds University Business School. Address of correspondence to: Des Thwaites, Leeds University Business School, Maurice Keyworth Building, Western Campus, Leeds LS2 9JT, UK. Tel: +44 113 3432625; fax: +44 113 3434885; e-mail: [email protected]

LEONIDOU, CHRISTODOULIDES, AND THWAITES THWAITES 51 (Hillary 2000). However, there are indications The intensification of competition is also that small firms are unaware of their own an important driver of organizational eco- environmental impact, have low eco-literacy orientation, since many firms capitalize on levels and limited understanding of the benefits green issues to achieve positional competitive accrued from eco-friendly actions, consider advantages (Simpson, Taylor, and Barker green practices as too costly and risky to 2004). Lastly, but not least, is the fact that implement, are inadequately informed of the rapidly changing market conditions concerning prevailing environmental legislation and its green issues provide opportunities for small repercussions, and resist voluntary initiatives firms to exploit, because of their more flexible aiming to promote environmental self- and less formalized structure compared with regulation within their specific industrial sector their larger counterparts (Lefebvre, Lefebvre, (Revell and Rutherfoord 2003). and Talbot 2003).1 Various obstacles are responsible for pre- An eco-friendly orientation is a core organi- venting small firms from engaging in environ- zational value that denotes (1) respect and mentally friendly business practices, such as (1) responsibility for the environment, as well as the absence of an appropriate organizational recognition that the firm has to reduce any structure, culture, and policy that will support harmful effects on it; (2) setting standards of green approaches (del Brio and Junquera 2003; ethical behavior and long-lasting commitment Masurel 2007); (2) the possession of limited to protecting the environment; (3) understand- financial, human, technological, informational, ing and responding to the needs of external and allied resources that are vital in supporting stakeholders (e.g., regulators, communities, ecological operations (Tilley 1999); (3) the buyers) with regard to the environment; and (4) adoption of a short-term business perspective, acting as a good corporate citizen with the which does not allow for the anticipation of responsibility to sustain the environment for new trends in the industry, such as future envi- future generations and care about the well- ronmental regulations, new clean technologies, being of society at large (Banerjee 2002). and innovative green products/services Although the issue of eco-orientation has been (Verheul 1999); (4) reliance on a single owner/ extensively studied within the context of larger manager who is confined by preexisting values firms (e.g., Menon and Menon 1997; Miles and that, in many cases, are guided by profit rather Munilla 1993; Stone and Wakefield 2000), it has than sensitivity to environmental issues been only peripherally tackled in the case of (Schaper 2002); and (5) heavy dependency on small business units, probably because of the external organizations and/or associations to misconception that the latter are less aware of implement environmental initiatives, especially the negative effects of their operations and the when these involve a significant amount of concomitant lack of interest in pursuing eco- investment (Verheul 1999). friendly activities (Patton and Worthington Despite these obstacles, several compelling 2003). However, there are indications that the reasons are pressing small firms to adopt a adoption of such an orientation is expected to strategic orientation toward environmental yield significant gains for the small firm, such activities. First, legislation regulating environ- as those pertaining to cost savings, improved mental practices is increasingly becoming more reputation, and customer attraction/retention strict, and in many cases imposing severe pen- (Gadenne, Kennedy, and McKeiver 2009; Lee alties on firms not abiding by the law (Patton 2009). and Worthington 2003). The emergence of a The intention of our study is to shed light on growing segment of consumers who reward this important issue by developing and testing companies that undertake green imperatives, a model of the external drivers of small-firm but punish those that harm the natural environ- environmental orientation and how this, in ment, is another key stimulating factor turn, affects financial performance. Specifically, (Langerak, Peelen, and van der Veen 1998). our aim is to investigate (1) the effect of key

1Some authors (e.g., Aragón-Correa et al. 2008; Lefebvre, Lefebvre, and Talbot 2003; Walley and Taylor 2002) conceive green market opportunities as corresponding to “niche” markets, which are particularly suitable for small ﬁrms because of their ability to have a more focused and/or ﬂexible approach in their business activities.

26 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT external forces, namely regulatory framework, studies (e.g., Revell and Blackburn 2007; environmental public concern, competitive Revell, Stokes, and Chen 2010; Simpson, intensity, and market dynamism, in determin- Taylor, and Barker 2004; Worthington and ing eco-friendly orientation within the small Patton 2005) highlight the fact that managers/ firm; (2) the extent to which this environmental owners do indeed recognize that their compa- orientation has a favorable effect on the firm’s nies could be harmful to the environment and financial outcomes; and (3) the role of both for this reason are willing to take the respon- organizational resources and capabilities in sibility of solving environmental problems, at moderating the relationship between environ- the same time they show that these managers mental orientation and financial performance. are not convinced that, by taking environmen- The article is organized into seven sections. tal initiatives, they will win customers and/or Following this introductory section, we review reduce their costs. This implies that a “green the literature on small-firm environmental managerial attitude” does not necessarily lead behavior. We then illustrate the conceptual to actual environmentally friendly behavior model of the study and develop the research (Schaper 2002). Because of this discrepancy, hypotheses. This is followed by an explanation adjustments in structures, policies and strate- of the research methodology used to carry out gies to accommodate environmental issues the study. In the next section, we present the were found in only a low proportion of small results with regard to testing our hypotheses. firms (Lee 2009; Williamson and Lynch-Wood Finally, the key findings of the study are sum- 2001). Schaper (2002) tries to explore whether marized and conclusions drawn for company the demographic profile of the owner/manager and public policymakers. in small business units (namely, age, gender, educational level) could be responsible for Literature on Small-Firm shaping green attitudes and yielding high envi- Environmental Issues ronmental performance, but no significant Research on environmental issues relating to results were revealed. small firms has taken a number of different The third line of research dealt with factors directions. The first stream focused on environ- that drive/stimulate environmentally friendly mental awareness, which, although found to be behavior in small firms. Revell, Stokes, and relatively high among small firms, was rarely Chen (2010) distinguish between “push” translated into real commitment to environ- motives (e.g., environmental legislation, techni- mental issues (Williamson and Lynch-Wood cal standards, local community rules) and 2001; Worthington and Patton 2005). Among “pull” motives (e.g., cost savings, new custom- the most frequent sources of environmental ers, good publicity). The most common forces information reported by small firms were driving eco-friendly response were environ- members of the supply chain (Williamson mental legislation (McKeiver and Gadenne and Lynch-Wood 2001), trade associations 2005; Revell, Stokes, and Chen 2010; Rowe and (Worthington and Patton 2005), and regulatory Hollingsworth 1996; Worthington and Patton agencies (Gadenne, Kennedy, and McKeiver 2005), cost reductions (Revell, Stokes, and 2009). However, there was a great disparity in Chen 2010; Rowe and Hollingsworth 1996), the degree of awareness with regard to specific customer demand (Revell, Stokes, and Chen environmental aspects. For example, whereas 2010; Rowe and Hollingsworth 1996; most small firms were found to be well aware Worthington and Patton 2005), and better of environmental legislation (especially that employee motivation/performance (Masurel relating to packaging, duty of care, and inte- 2007; Revell, Stokes, and Chen 2010). Cleaner grated pollution prevention/control) (Simpson, environmental conditions, improved local envi- Taylor, and Barker 2004), the opposite was true ronment, and safer working conditions were with regard to other dimensions of environ- also mentioned in some studies (e.g., McKeiver mental friendly practices, such as possible cost and Gadenne 2005) but at a low frequency rate. savings (Gadenne, Kennedy, and McKeiver The fourth stream of studies focused on the 2009) and performance measurement (Rowe factors obstructing the adoption of environmen- and Hollingsworth 1996). tal initiatives by small firms, such as increased The second group of studies attempted to costs (e.g., Revell, Stokes, and Chen 2010; shed light on the environmental attitudes of Worthington and Patton 2005), loss of market managers in small firms. Although a number of competitiveness (e.g., Revell, Stokes, and Chen

LEONIDOU, CHRISTODOULIDES, AND AND THWAITES THWAITES 7 3 2010; Verheul 1999), and lack of staff time in small firms (McKeiver and Gadenne 2005). (e.g., Revell, Stokes, and Chen 2010). A number In those cases where they did exist, they were of studies (e.g., Dean, Brown, and Stango 2000; mainly of an informal (e.g., changing processes Petts et al. 1999; Williamson and Lynch-Wood to reduce waste) than a formal (e.g., ensuring 2001) also point to the fact that environmental compliance with all environmental laws and regulations may pose a problem for small firms, regulations) nature (McKeiver and Gadenne in the sense that (1) it is not very clear and/or 2005). Some of the most common environmen- obvious what one is actually required to do, tal management systems/practices found to especially as regards what constitutes a compli- be adopted by small firms included recycling ant versus a noncompliant state; (2) there is activities (Cordano, Marshall, and Silverman some disparity between owner/manager’s con- 2010; Revell, Stokes, and Chen 2010), energy cerns about the environment and the perceived conservation (Cordano, Marshall, and role played by legislation to protect the envi- Silverman 2010; Revell, Stokes, and Chen ronment; and (3) the high costs incurred 2010), and carbon emission reduction (Revell, because of environmental regulations (espe- Stokes, and Chen 2010). In Rutherfoord, cially as regards health and safety and waste Blackburn, and Spence’s (2000) study, it was management), which put the firm in a disad- found that the environmental practices of small vantageous position. firms are influenced by national stereotypes The fifth group of studies dealt with and that countries adopt different approaches external/internal drivers influencing the green (e.g., when appealing to their social responsi- strategic behavior of small firms. With regard to bility role) to mobilize firms to engage in envi- external forces, some researchers stressed the ronmentally beneficial activities. role of legislation, suppliers, customers, The seventh group of studies focused on the and institutions in adjusting organizational formal environmental certification program of procedures, developing environmental policies, ISO 14000 and particularly its ISO 14001 using environmental audits, and seeking variant.2 The convergence of the findings of accreditation from external environmental stan- empirical studies on the subject revealed that dards (Gadenne, Kennedy, and McKeiver 2009; the overwhelming majority of small firms have Tilley 1999; Williamson and Lynch-Wood 2001). both a low awareness (Lee 2009; Revell, Stokes, In the case of internal forces, these focused and Chen 2010; Worthington and Patton 2005) mainly on the availability of time and informa- and limited adoption (Cordano, Marshall, and tion resources of small firms (Schaper 2002), as Silverman 2010; Hillary 2004; Lefebvre, well the managerial risk-taking behavior, which Lefebvre, and Talbot 2003; McKeiver and is vital in positively responding to ecological Gadenne 2005; Revell, Stokes, and Chen 2010) intelligence (Stone and Wakefield 2000). More- of this type of environmental management over, a study conducted by Aragón-Correa et al. system. Although smaller firms were found to (2008) confirmed the role of certain organiza- acknowledge the existence of both internal tional capabilities, namely shared vision (i.e., (e.g., quality improvement, cost savings, boost- the owners-founder’s vision and his or her ing employee motivation and morale) and close interaction and communication with external (e.g., gaining a competitive advantage, other employees), strategic proactivity (i.e., the improved environmental performance, creating firm’s entrepreneurial orientation and a positive public image) benefits from the use innovativeness), and stakeholder management of this program, a number of obstacles (e.g., (i.e., organizational flexibility in managing resource constraints, lack of market rewards, inter-firm and external relationships), in devel- implementation complexities) are often respon- oping proactive environmental strategies. sible for inhibiting its adoption (Heras and The sixth stream of research referred to envi- Arana 2010; Hillary 2004). In light of this, ronmental management system/practices, several studies (e.g., Ammenberg and Hjelm which in the majority of cases were not in place 2003; Halila 2007; Halila and Tell 2013)

2The ISO 1400 consists of a series of standards that focus on environmental management systems, environmental auditing, eco-labeling, environmental product speciﬁcations, and environmental performance evaluation (Miles, Munilla, and McClurg 1999). The ISO 1401 provides a speciﬁc standard within this category, because it is intended for registration by a nongovernmental third party (Cordano, Marshall, and Silverman 2010).

48 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT explored the feasibility of using an existing Tilley (2000) investigated the ethical aspects of network of small firms to jointly initiate and small firms and found that (1) they are mostly implement an ISO 14001 certification project, driven by a value system that stresses economic revealing that, although possible and beneficial prosperity as a goal; (2) they are characterized for improving relationships with various stake- by non-responsible behavior and, therefore, holders (e.g., customers), it is very difficult to external regulations are critical to constrain this keep the firms in the network working together behavior; (3) they regard the environmental as a homogeneous group. challenge from a shallow ecology perspective, The eighth set of studies examined the envi- because they are operating within the conven- ronmental behavior and performance of tional ethical discourse system. small firms. For instance, Williamson and A number of observations can be made Lynch-Wood (2001) found that very few of with regard to the previous review of the lit- these firms have an environmental manager in erature on the green activities of small firms: place, most do not have environmental poli- first, it is very fragmented, taking many dif- cies, and they do not produce environmental ferent directions, which, however, cover envi- reports/audits. Bianchi and Noci (1998) also ronmental issues in a rather shallow way; showed that small firms have a greater ten- second, it is relatively descriptive in nature, dency to adopt reactive rather than proactive with only a few studies trying to establish and environmental behavior, because, on the one test causal associations between the environ- hand, they lack the necessary financial means, mental behavior of small firms and its ante- technological equipment, and personnel skills cedents and outcomes; third, with a few required to develop green initiatives, and, on exceptions, the analysis of the various envi- the other, they are less committed to environ- ronmental issues did not rely on sophisticated mental activities on the grounds that these will quantitative methods but was rather simplistic; not yield adequate returns in the short term. fourth, it is relatively atheoretic in nature, Finally, with regard to the environmental per- with most studies lacking a sound theoretical formance of small firms, the following obser- foundation and conceptual development; vations were made: (1) the actual measurement finally, important issues that have been of environmental performance is limited widely studied in the broader environmental and/or not properly conducted (Williamson business literature, such as the role of exter- and Lynch-Wood 2001); (2) there is a disparity nal influences, internal resources/capabilities, between perceived and actual environmental the firm’s eco-friendly orientation, and finan- performance levels (Rowe and Hollingsworth cial implications of environmental initiatives, 1996); and (3) the outcome of most green have only been tangentially tackled. activities is measured in terms of business performance (Williamson and Lynch-Wood 2001). Research Model and The final group of studies dealt with various Hypotheses specialized green topics pertaining to small To fill some of the gaps identified from the firms. For example, Paton (1994) focused on previous review of the literature on the environmental auditing in small firms and environmental-related aspects of small firms, we revealed that environmental audits were rare, have developed a conceptual model particularly because they were considered either irrelevant focusing on external forces, eco-friendly orien- to the running of the business or required tation, financial performance, and organiza- excessive amounts of cost and time, whereas tional resources/capabilities (see Figure 1). the cases in which audits were undertaken Specifically, our model indicates that external were more prevalent in industries harmful to forces (i.e., environmental regulations, environ- the environment, such as minerals and energy. mental public concern, competitive intensity, Noci and Verganti (1999) explored green market dynamism) have a positive impact on the product innovation and found that this was not development of an eco-friendly orientation in a marginal issue for most small firms (even for the small firm, which, in turn, has a positive those not directly affected by environmental effect on its financial performance. It also shows regulations), thereby stressing the need to that the eco-orientation → financial perfor- select a proper research and development strat- mance path is moderated by the existence of egy that would explicitly account for the eco- appropriate organizational resources and efficiency of product technologies. Finally, capabilities. In total, there are five main

LEONIDOU, CHRISTODOULIDES, AND AND THWAITES THWAITES 9 5 Figure 1 The Conceptual Model

Regulatory

framework a s e ion (RFR )

) c r at z u i o (ORE an g res r

H2 O

Environmental public concern H6 (EPC)

Eco−friendly H5 Financial orientation performance (EFO) (FIP)

H3 Competitive H7 intensity (CTI)

l a s e i ion lit at i

H4 z b i a (OCA) an

Market g cap r

dynamism O (MRD)

hypothesized associations and two moderating mental management systems; and (3) they are hypotheses, which are elaborated in the follow- often associated with fines and penalties that in ing sections. many cases are excessively high compared with the limited financial means of small firms Main Hypotheses (McKeiver and Gadenne 2005; Revell, Stokes, The environmental regulatory framework and Chen 2010; Simpson, Taylor, and Barker has been considered in many studies (e.g., 2004). Thus, we may hypothesize that Baylis, Connell, and Flynn 1998; Rutherfoord, Blackburn, and Spence 2000; Williamson and H1: High levels of environmental regulatory Lynch-Wood 2001) as a major stimulus for the intensity will lead to the adoption of an adoption by small firms of an eco-friendly eco-friendly orientation by the small approach to their business. Environmental firm. regulations may cover a wide array of issues, ranging from clean technologies and technical Increasing environmental public concern has standards to package recycling and waste man- been a major driving force behind the adoption agement (Banerjee, Iyer, and Kashyap 2003). of ecological thinking by many firms. Such Most of these regulations aim to increase social public concern can be of either an economic and economic benefits through, for example, (e.g., customers) or political (e.g., local commu- the lowering of illness and mortality, reduction nity) nature (Banerjee, Iyer, and Kashyap 2003). of damage to vegetation, improvement of aes- Irrespective of its origin, to be able to generate thetics, and enhancement of recreational an eco-friendly orientation within the organiza- opportunities (Carlton and Perloff 1990). Envi- tion, the public has to fulfill the following pre- ronmental regulations stimulate eco-friendly conditions: (1) to express awareness, sensitivity, orientation in smaller firms in three ways: (1) and commitment to environmental issues; (2) to they generate awareness of environmental pro- develop a certain level of expectation about tection issues, such as packaging recycling, company environmental activities; and (3) to energy efficiency, and reduction in carbon exert influence on the firm should it not conform emissions; (2) they help toward establishing to their environmental expectations (Langerak, formalized procedures with regard to environ- Peelen, and van der Veen 1998; Menon et al.

610 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT 1999). Although the ecological behavior of large consumer attitudes toward eco-friendly firms, firms is more likely to attract the attention of the buyer shifts toward purchasing and using eco- public because of greater exposure, higher vis- logical products, and the increasing power of ibility, and wider impact of their actions, smaller the consumerism movement are some of the firms can also be affected, especially when they elements characterizing today’s markets in operate in industries harmful to the environ- many parts of the world, which are responsible ment (Peretz, Bohm, and Jasienczyk 1997). for instilling eco-friendly thinking in an ever Based on the previous discussion, we may greater number of firms (Menon and Menon hypothesize that 1997). The flexible and flat organization structure of small firms allows them to grasp eco- H2: High levels of environmental public logical trends in the market quickly, although concern will lead to the adoption of an eco- resource constraints may prevent them from friendly orientation by the small firm. fully adopting and implementing eco-friendly initiatives (Aragón-Correa et al. 2008). The exis- The competitive environment has often been tence of dynamic conditions in the market can described as a key stimulus for the firm’s envi- facilitate first mover advantage for those small ronmental actions (Menon et al. 1999). Specifi- firms that first conceive and swiftly respond to cally, under conditions of high competitive green market changes (Baker and Sinkula intensity, small firms try to find ways to differ- 2005). The following hypothesis can, therefore, entiate their offerings from rival firms, and one be made: way to achieve this is through the adoption of eco-friendly activities, such as recyclable pro- H4: High levels of market dynamism will lead cedures, use of clean technologies, and waste to the adoption of an eco-friendly orienta- management protocols (Menon and Menon tion by the small firm. 1997). Although such an environmental orientation will help large firms to stay ahead of An eco-friendly orientation is an important competitors by establishing new sets of envi- prerequisite for formulating and implementing ronmental norms (especially when following environmentally friendly business practices voluntary/proactive programs), in the case of (e.g., product stewardship, waste management, small firms it will ensure that they reach the pollution prevention) within the organization minimum environmental requirements set by (Menon and Menon 1997; Miles and Munilla the industry (Arora and Cason 1995). Small 1993). Environmentally oriented organizations firms that take competitive movements into are characterized, inter alia, by (1) objectives, consideration seriously are more likely to per- focusing on producing high-value eco-friendly ceive green issues as an opportunity they products that will create consumer brand pref- can exploit and be in a position to anticipate erence; (2) systems, monitoring corporate sen- future environmental pressures more quickly sitivity toward ecological market trends, (Perez-Sanchez, Barton, and Bower 2003). especially as regards the green market segment; Hence, we can posit that (3) values, that are centering on providing customer satisfaction by offering green products H3: High levels of competitive intensity will and other eco-friendly business activities; lead to the adoption of an eco-friendly ori- and (4) behavior, such as tracking ecological entation by the small firm. intelligence, sharing environmental information across functional areas, and responding to con- Another important external determinant of sumer green needs (Miles and Munilla 1993; ecological orientation refers to market dyna- Stone, Joseph, and Blodgett 2004). Obviously, mism, that is, the rate at which market forces in top management has a decisive role to play in the firm’s operating environment (e.g., con- instilling such an orientation within the organi- sumer preferences, demand levels, product zation, since their ideas, values, and beliefs characteristics) are changing (Achrol and Stern determine to a great extent the nature and scope 1988). Under conditions of high market dyna- of the firm’s social responsibility (Menon et al. mism, firms have to better understand buyers’ 1999). This is more evident in the case of small needs, absorb market information faster, and firms, whereby the owner/manager is usually realign their competitive strategies (Jaworski the single decision-maker and the driving force and Kohli 1993). The cultivation of favorable behind any strategic initiative within the

LEONIDOU, CHRISTODOULIDES, AND AND THWAITES THWAITES 11 7 organization (Schaper 2002). Growing evidence new areas of ecological success based on previ- in the literature (e.g., Aragón-Correa et al. 2008; ous experience, exploiting economies of scale in Menon and Menon 1997; Stone and Wakefield sharing environmental costs, and providing a 2000) indicates that acting in an ecologically sound technological basis for sustainable eco- oriented manner will yield better financial friendly programs (Russo and Fouts 1997). results for the company. This is because the Organizational resources can influence both firm’s responsiveness to ecological issues helps, strategic (e.g., green technology) and operating on the one hand, to reduce costs (e.g., recycling (e.g., reverse logistics) dimensions of the firm’s material waste, energy/water savings, avoiding ecological behavior, and their degree of use will green penalties) and, on the other, to increase depend on the specific problem that the firm has benefits (e.g., increasing customer satisfaction, at hand (Barney 1991). Although small firms, by improving firm reputation, enhancing brand default, have limited organizational resources, loyalty). This leads us to the following those that are in a position to make wise use of hypothesis: appropriate resources for environmentally friendly purposes are very likely to achieve H5: High levels of company eco-friendly orien- superior performance (Aragón-Correa et al. tation will lead to high levels of financial 2008). Hence, we may posit that performance. H6: High levels of organizational resources Moderation Hypotheses associated with environmental activities will So far, we have seen the role of external have a strong positive effect on the link factors in determining environmental orienta- between eco-friendly orientation and finan- tion and how this, in turn, leads to superior cial performance, and vice versa. financial performance. However, for eco- friendly orientation to produce financial Organizational capabilities assemble, inte- results, it is important to understand the mod- grate, and manage resources in order to erating effect of certain factors internal to the address the challenges of the external business firm. Specifically, drawing on the resource- environment and meet changing market based theory, which adopts an inward-looking demands (Eisenhardt and Martin 2000). Such perspective regarding the firm (Barney capabilities are essential in activating and coor- 1991), we explore the role of organizational dinating resources to perform their function, resources and capabilities on the association update their current status, and preserve their between the firm’s eco-orientation and finan- existence (Teece et al. 1997). They are cial performance.3 expressed in the form of inward (e.g., new Organizational resources are tangible (e.g., product development), outward (e.g., market finance, equipment, installations) or intangible sensing), or spanning (e.g., relationship build- (e.g., technical know-how, reputation, experi- ing) processes, that help to organize and get ence) assets controlled by the firm that help to things done within the organization in an effec- design and implement strategies that will tive and efficient way (Day 1994). The pertinent improve business performance (Grant 1991). literature points to specific capabilities, such as They are cumulative in nature, socially complex, shared vision, cross-functional coordination, and tacit (Barney 1991). However, to be useful, and technology sensing/response, that are these resources must be valuable, rare, imper- crucial to the adoption and implementation of fectly imitable, and non-substitutable (Barney sustainable business programs (Hart 1995; 1991; Bharadwaj, Varadarajan, and Fahy 1993). Shrivastava 1995). Although small firms are Organizational resources are crucial in support- usually less endowed with organizational capa- ing environmental orientation by, for example, bilities compared with their larger counter- covering the cost of green initiatives, seeking parts, the literature shows that the proper use

3This theoretical paradigm stresses the instrumental role of organizational resources (and capabilities) in achieving positional competitive advantage and superior performance (Barney 1991). It also stresses the mediating role of business strategy formulation and implementation, by exploiting environmental opportunities, accommodating external threats, capitalizing on internal strengths, and limiting the impact of company weaknesses (Bharadwaj, Varadarajan, and Fahy 1993).

812 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT of the right capabilities that are instrumental to work” and “environmental public concern” eco-efficient practices will enhance business comprised six items, each taken from Banerjee, performance (Aragón-Correa et al. 2008). Thus, Iyer, and Kashyap (2003). Jaworski and we can hypothesize that Kohli’s (1993) study was the source of the four-item scale for “competitive intensity,” H7: High levels of organizational capabilities whereas the three-item scale for “market dyna- associated with environmental activities will mism” was adopted from Baker and Sinkula have a strong positive effect on the link (2005). “Eco-friendly orientation,” the central between the firm’s eco-friendly orientation construct of our model, consisted of 11 items and financial performance, and vice versa. taken from Fraj-Andres, Martinez-Salinas, and Matute-Vallejo (2009). Seven items, derived Methods mainly from Leonidou et al.’s (2013) study, were To test our hypotheses, we conducted a employed to measure the firm’s “financial per- study in Cyprus, which is characterized mainly formance.”4 Finally, with regard to moderating by a light manufacturing sector dominated by factors, the scale of “organizational resources” firms of small size. The existence of numerous comprised five items taken from Buysse and smaller-sized firms in the country can be attrib- Verbeke 2003, whereas “organizational capabili- uted to the family nature of most businesses, ties” was a six-item scale extracted from Sharma, the underdeveloped capital market system, and Aragón-Correa, and Rueda-Manzanares (2007). the limited size of the domestic market. Since it Scale items were measured on a seven-point joined the European Union in 2004, the country Likert scale, ranging from strongly disagree (1) has had to adopt stringent legislation relating to to strongly agree (7), whereas some of the the protection of the environment, such as that items were reversed. The survey instrument concerning the recycling of packaging materi- contained additional questions referring to als, clean production technologies, energy and organizational demographics, such as establish- water conservation mechanisms, and waste ment year, number of employees, and type of management processes. goods produced. The questionnaire was first The target population was firms listed in the developed in English and then translated into most recent directory of manufacturers pub- Greek, which is the official language of the lished by the Cyprus Chamber of Commerce country. A back-translation procedure revealed and Industry (CCCI, 2011). These were that the questionnaire did not suffer from any screened out on several criteria, namely to have linguistic problems. To ensure the workability been in operation for at least three years, to be of the questionnaire, it was pretested with five of a size greater than 10 employees but less managers of small companies, revealing no par- than 250, to be of an indigenous character ticular problems. rather than branches of international firms, and Those firms that showed a willingness to to belong to the private, as opposed to the participate in the study were sent a mail ques- public, sector of the economy. Those firms that tionnaire, accompanied by a covering letter fulfilled our eligibility criteria were contacted stating the purpose of the study and explaining by telephone to explore their interest in par- how to administer the questionnaire. In some ticipating in the study, as well as to identify key cases, the questionnaire was hand-delivered to informants. The outcome of this process was to the firm’s premises, whereas in others the secure 513 firms expressing an initial interest in respondents were provided with an electronic participating in the study. version. A series of reminder letters, telephone Constructs were operationalized based on contacts, and even personal visits was scales established in the literature and were employed in order to increase participation in further refined after discussions with a panel of the study. The outcome of these efforts was to managers from small manufacturing units collect 161 questionnaires (31.4 percent (see Appendix). Analytically, “regulatory frame- response rate) within a period of three months.

4Similar multiple scales of financial performance have also been used in key articles in the wider environmental business literature, such as those by Menon and Menon (1997) (which focused on market share and ROI), Judge and Douglas (1998) (which focused on profit growth, sales growth, market share change, and ROI), and Menguc and Ozanne (2005) (which focused on sales growth, market share, and profit).

LEONIDOU, CHRISTODOULIDES, AND AND THWAITES THWAITES 13 9 Of these, only 153 were adequately completed average variance extracted (Fornell and Larcker for data analysis purposes. The existence of 1981) (see Table 2). Construct reliability was possible nonresponse bias was tested using also estimated, with all constructs satisfying Armstrong and Overton’s (1977) procedures, recommended thresholds in terms of whereby the answers given by early respon- Cronbach’s alpha coefficients (>.7), composite dents were compared with those of late respon- reliability scores (>.69), and average variance dents, revealing no statistically significant extracted (>.5) (Bagozzi and Yi 1988; Fornell differences. and Larcker 1981). Common method bias was tested using Data Analysis and Findings Harman’s one-factor test (Podsakoff and Organ Using structural equation modeling (SEM) 1986). First, a principal components analysis on based on EQS, we performed three major types all items examined in this study was performed. of analysis with regard to our conceptual The unrotated solution resulted in six factors model, namely measurement model validation, with eigenvalues greater then 1.0, accounting structural model estimation, and moderating for 66.3 percent of the variance. Second, a CFA effects assessment. model was estimated, where the measurement items and the factors included in the structural Measurement Model Validation model were restricted to load on a single factor. The psychometric properties of our con- The results obtained from this analysis indicate 2 structs were evaluated using the iterative puri- a poor fit (χ (324) = 1868.72, p < .001, NFI = .53, fication process recommended by Gerbing NNFI = .54, CFI = .57and RMSEA = .18.), imply- and Anderson (1988). First, we carried out an ing that common method bias does not pose a exploratory factor analysis on the scales problem in our investigation. employed and any items exhibiting low factors loadings, low item-to-total correlations, or Structural Model Estimation heavy loadings on more than one factor were Table 3 presents the standardized path coef- dropped. With the remaining items, a confirma- ficients for each main hypothesis in the model, tory factor analysis (CFA) was used to verify the together with the corresponding t-values. The hypothesized factor structure and assess con- research hypotheses were tested by estimating vergent validity and discriminant validity. To the structural model based again on the elliptical perform this analysis, we used the elliptical re-weighted least-square method. Although the re-weighted least square procedure, whereby chi-square for this model was found to be sta- 2 each measurement item was restricted to load tistically significant (χ (372) = 642.10, p = .00), only on its a priori specified factor. The chi- the ratio between chi-square and degrees of square statistic of this model was found to be freedom was satisfactory (χ2/df = 1.73) and 2 significant (χ (362) = 573.49, p < .01), which was the values of all alternative fit indices were expected, due to the sensitivity of this statistic within acceptable levels (NFI = .90, NNFI = .94, to sample size (Bagozzi and Yi 1988). However, CFI = .94, RMSEA = .07). the ratio of chi-square to degrees of freedom H1, which connects regulatory framework (χ2/df = 1.58), as well as the other alternative fit and eco-friendly orientation, was confirmed indices (NFI = .90, NNFI = .94, CFI = .94, and (β = .32, t = 2.90, p = .00). This finding is in RMSEA = .06) indicate a good fit to the data harmony with the findings of other studies (e.g., (see Table 1). Banerjee, Iyer, and Kashyap 2003; Menon and With regard to convergent validity, this was Menon 1997), which also underline the instru- met because the t-value for each item was mental role of environmental regulations in always significant, all standard errors of the instilling green thinking within the organization. estimated coefficients were very low, and the This finding is in line with that of Baylis, average variance extracted for each construct Connell, and Flynn (1998), who report environ- was above the threshold level of .50 (Hair et al. mental regulations as the most common source 2011). Discriminant validity was also observed of environmental motivation of small firms. because the confidence interval around the cor- Although complying with such regulations has relation estimate for each pair of constructs often been associated with a reactive approach investigated never included 1.00 (Anderson and to environmental business (Sharma and Gerbing 1988), whereas the squared correlation Vredenburg 1998), our study revealed that these for each pair of constructs never exceeded their can seriously stimulate awareness and interest

1014 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Table 1 Measurement Model Results

Constructs Scale Standardized t-Value αρAVE Mean Standard Items Items Items Loadings Score Deviation (S.D.) Means S.D.

Regulatory Framework RFR1 .72 * 0.71 0.69 0.42 3.95 1.46 4.30 1.88 RFR2 .69 5.75 3.60 2.02 ENDU HITDUIE,ADTWIE 11 THWAITES AND CHRISTODOULIDES, LEONIDOU, ENDU HITDUIE,ADTWIE 15 THWAITES AND CHRISTODOULIDES, LEONIDOU, RFR3 .59 5.23 3.63 2.08 RFR6 .58 5.18 4.29 2.02 Environmental Public Concern EPC4 .66 * 0.81 0.76 0.54 4.44 1.39 4.37 1.53 EPC5 .85 7.34 4.32 1.84 EPC6 .81 7.20 4.59 1.82 EPC9 .59 5.61 4.49 1.74 Competitive Intensity CIN2 .59 * 0.74 0.72 0.66 5.75 1.23 5.51 1.50 CIN3 .99 3.51 5.99 1.37 Market Dynamism MDY2 .96 * 0.73 0.69 0.63 5.40 1.22 5.60 1.32 MDY3 .59 3.62 5.20 1.54 Eco-friendly Orientation EFO1 .67 * 0.91 0.89 0.52 5.02 1.24 4.66 1.87 EFO2 .73 7.35 5.18 1.54 EFO3 .81 8.10 5.05 1.49 EFO4 .80 8.02 5.11 1.60 EFO5 .89 8.76 5.16 1.50 EFO6 .77 7.78 5.36 1.52 EFO7 .55 5.68 4.49 1.98 EFO9 .63 6.51 5.14 1.74 EFO10 .60 6.14 4.44 1.81 EFO11 .65 6.69 5.57 1.59 Financial Performance FIP1 .75 * 0.93 0.89 0.66 4.34 1.34 4.38 1.70 FIP2 .94 11.19 4.39 1.58 FIP3 .88 10.41 4.18 1.58 FIP4 .85 9.97 4.25 1.57 FIP5 .85 9.94 4.48 1.61 FIP6 .72 8.23 4.67 1.45 FIP7 .67 7.68 4.01 1.72

*Item ﬁxed to set the scale. Fit statistics: chi-square (χ2) = 573.49, p = .000; df = 362; ratio chi-square to d.f. (χ2/df) = 1.58; Normed Fit Index (NFI) = .90; Non-normed Fit Index (NNFI) = .94; Comparative Fit Index (CFI) = .94; Root Mean Squared Error of Approximation (RMSEA) = .06; 90 percent conﬁdence interval of RMSEA = (.05, .07). Table 2 Correlation Matrix

Constructs 1. 2. 3. 4. 5. 6.

1. Regulatory framework 1 2. Environmental public concern .26 1 3. Competitive intensity .30 .11 1 4. Market dynamism .22 .20 .40 1 5. Eco-friendly orientation .40 .42 .15 .17 1 6. Financial performance .11 .04 .02 .00 .15 1

Note: Correlations greater than |± 0.21| are signiﬁcant at the .01 level. Correlations greater than |± 0.16| are signiﬁcant at the .05 level.

Table 3 Structural Model Results

Hypothesis Hypothesized Association Standardized t-Value p-Value Status Path Coefﬁcient

H1 Regulatory framework → .32 2.90 .00 Accepted eco-friendly orientation H2 Environmental public concern .36 3.45 .00 Accepted → eco-friendly orientation H3 Competitive intensity → .18 1.86 .06 Accepted eco-friendly orientation H4 Market dynamism → .20 2.11 .03 Accepted eco-friendly orientation H5 Eco-friendly orientation → .19 2.05 .04 Accepted ﬁnancial performance

Fit statistics: Chi-square (χ2) = 642.10, p = .000; df = 372; ratio chi-square to d.f. (χ2/df) = 1.73; Normed Fit Index (NFI) = .90; Non-normed Fit Index (NNFI) = .94; Comparative Fit Index (CFI) = .94; Root Mean Squared Error of Approximation (RMSEA) = .07; 90 percent confidence interval of RMSEA = (.06, .08). among small firms toward taking a more strate- groups in making organizations more sensitive gic perspective in their green business activities. to ecological issues (Kassinis and Vafeas 2006; To some extent this can be attributed to the fact Menon and Menon 1997). This is in accord with that sensitivity to green issues by smaller firms the findings of earlier research on eco- has increased dramatically in recent years, as a orientation conducted within the domain of result of Cyprus’ membership of the European large firms (e.g., Banerjee, Iyer, and Kashyap Union and the concomitant introduction of a 2003; Langerak, Peelen, and van der Veen wide range of rules and regulations for protect- 1998; Stone, Joseph, and Blodgett 2004). ing the environment. Despite the relatively low individual impact of With regard to H2, the positive impact of small firms on the environment, it seems that environmental public concern on green busi- the views of the public about ecological pro- ness orientation was also verified (β = .36, tection are inextricably linked with their busi- t = 3.45, p = .00), stressing the importance of ness operations (Peretz, Bohm, and Jasienczyk citizens, buyers, local communities, and other 1997). This is particularly evident in countries

1216 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT characterized by high educational standards, as Finally, our study confirmed H5, which in the case of Cyprus, where people are more stated that that the cultivation of an eco- aware of environmental problems and, there- friendly orientation could yield positive finan- fore, expect suppliers of goods to behave in a cial results for the firm (β = .19, t = 2.05, more socially responsible manner (Leonidou, p = .04). This implies that thinking (and subse- Leonidou, and Kvasova 2010). quently acting) in an eco-friendly way helps to Significant results were also found in the case make significant cost savings for the small firm, of H3, indicating that competitive intensity can while at the same time boosting sales and indeed be responsible for cultivating a green improving market share. Although this positive orientation within the organization (β = .18, association between eco-orientation and busi- t = 1.86, p = .06). This confirms the view that ness performance has been repeatedly cited in one way for the firm to cope with intensified the literature with regard to large firms (e.g., competition is by adopting an ecological per- Baker and Sinkula 2005, Menon and Menon spective in its business activities (Dechant and 1997, Stone and Wakefield 2000), our study Altman 1994; Menon and Menon 1997). confirms that such a relationship is also plau- Although such an eco-friendly orientation will sible in the case of their smaller counterparts. help the small firm to attain the minimum envi- This finding contradicts the negative predispo- ronmental standards set by the industry within sition expressed by some small business man- which it operates, it can create a positive differ- agers (not only in Cyprus, but also in other entiation from its rival firms (Arora and Cason parts of the world) that eco-friendly initiatives 1995; Langerak, Peelen, and van der Veen 1998). are costly to adopt and take time to yield sat- It would also help to create a more proactive isfactory results. stance toward green issues, with all the benefits that this may entail, such as taking the lead in a Moderating Effects specific niche market, strategically handling Moderating effects were tested through competitors’ movements, and capitalizing on multigroup analysis, where the initial sample ecological product innovations (Aragón-Correa was split into two groups based on the median 1998; Dechant and Altman 1994). This is of (see Table 4). Specifically, for each moderator, particular relevance for Cypriot manufacturing we ran two different models: whereas in the concerns, which are increasingly facing stiff first model all parameter estimates were free to competition not only from other local produc- vary between the two subsamples, in the ers, but also from foreign firms (especially from second model an equality constraint was set on Europe and Asia). the hypothesized moderated link between the Support was also given to the hypothesis two groups. linking market dynamism with eco-friendly ori- With reference to H6, our results indicate entation (H4) (β = .20, t = 2.11, p = .03). that the possession of adequate organizational Indeed, in highly dynamic markets, small firms resources committed to environmental activi- are bound to adapt quickly to changes relating ties positively affects the association between to consumer habits/attitudes and competitors’ eco-friendly orientation and financial perfor- 2 movements, which to a great extent nowadays mance (Δχ (1) = 2.80, p < .10). Analytically, incorporate ecological elements (Baker and when the firm deploys a higher amount of Sinkula 2005). Such changes have become resources, the adoption of an eco-friendly ori- more profound in the Cyprus market recently, entation will have a statistically significant as a result of growing consumer sophistication, impact on financial performance (β = .25, involvement in European business practices, t = 2.25, p = .02), though this association and exposure to worldwide financial problems becomes nonsignificant when there are (Leonidou, Leonidou, and Kvasova 2010). resource constraints in the organization Despite unconvincing evidence to support the (β = .07, t = .58, p = .56). This finding is in line direct link between market turbulence and eco- with the resource-based view paradigm, which orientation within the domain of large firms stresses the decisive role of resources in sup- (Baker and Sinkula 2005), it seems that, within porting the firm’s initiatives to create sustain- the context of small firms, market turbulence able competitive advantage (Barney 1991). It has an important role to play in harnessing also dilutes the prevailing myth that small eco-friendly commitment because of their firms, because of their size constraints, are by inherent greater adaptive capability. default hesitant in deploying appropriate

LEONIDOU, CHRISTODOULIDES, AND AND THWAITES THWAITES 17 13 Table 4 Moderating Effects Results a. Organizational resources as a moderator

Main effect Hypothesized High organizational Low organizational Δχ2 (Δdf =1) moderating effect resources group resources group EFO → FIP H6: Effect is stronger β = 0.25 β = 0.07 2.88 when committing t = 2.25 t = 0.58 (p < .10) more organizational resources b. Organizational capabilities as a moderator

Main effect Hypothesized High organizational Low organizational Δχ2 (Δdf =1) moderating effect capabilities group capabilities group EFO → FIP H7: Effect is stronger β = 0.29 β = 0.08 2.80 when committing t = 2.66 t = 0.77 (p < .10) more organizational capabilities

resources to support environmentally friendly eco-friendly-oriented behavior within the small purposes (Aragón-Correa et al. 2008). firm, which in turn can improve financial per- Our analysis also confirmed H7, since orga- formance. It has also shown that the positive nizational capabilities associated with environ- effect of environmentally friendly orientation mental activities were found to play a on performance becomes stronger when the significant moderating role in the eco-friendly small firm commits adequate resources and orientation → financial performance path capabilities to green initiatives. These findings 2 (Δχ (1) = 2.88; p < .10). Specifically, although suggest that (1) as in the case of larger firms, with the possession of lower organizational small units are also likely to cultivate green capabilities, the relationship between eco- thinking, provided that these are properly orientation and financial performance is statis- stimulated by certain external forces, namely tically nonsignificant (β = .08, t = .77, p = .56), environmental regulations, environmental this relationship becomes significant when the public concern, competitive intensity, and firm possesses a higher amount of these capa- market dynamism; (2) although many small bilities (β = .29, t = 2.66, p = .01). Again, this is firms believe that taking environmental initia- in harmony with the resource-based view tives are prohibitively costly, such initiatives theory, which considers capabilities as vital in can enhance financial results (e.g., profits, effectively and efficiently combining resources sales, market share); and (3) despite the fact to support the implementation of the firm’s that eco-orientation can improve financial per- strategies (Barney 1991). Our results are in formance, the strength of this link will be con- accord with those of Aragón-Correa et al. tingent on the deployment of appropriate (2008), who found that small firms possessing resources (e.g., financial) and capabilities (e.g., certain capabilities (e.g., shared vision, strate- adaptability). gic proactivity, stakeholder management) are This study puts together, in a theoretically very likely to successfully implement environ- anchored, integrative, and causal fashion, con- mental strategies. cepts and ideas that touch upon important aspects of small firms’ environmental behavior. Summary and Conclusions It particularly adds to the small business disci- Our study has shown that external environ- pline in four possible ways. First, it identifies mental forces can be conducive to building a set of key external factors that have been

1418 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT examined under different situations in prior sory, educational and other support to firms research to stimulate the small firm’s sensitivity involved in environmental initiatives, and to environmental issues. Second, it focuses on recognize/reward companies that excel in their the dynamics of forming an ecological orienta- environmental actions. However, it is crucial to tion within the small firm, an issue that has take into consideration the great heterogeneity been tangentially tackled in the environmental that exists among small firms with regard to business literature. Third, it associates environ- their information sources, barriers/stimuli, and mental issues with financial performance, a link attitude/behavior with regard to environmental that, although studied within the context of issues. Most importantly, small firm managers/ large firms, has been virtually neglected by owners have to be convinced that engagement previous studies focusing on the green behav- in environmental actions can enhance commer- ior of small firms. Finally, building on the cial benefits, reduce costs, and create positional resource-based theory, the study underscores advantages. Finally, it is vital to proceed with the critical role of both organizational resources the necessary institutional reforms in educa- and capabilities in enhancing the impact tional systems, regulatory frameworks, and of small firm eco-orientation on financial business infrastructure that will facilitate the performance. smoother adoption of eco-friendly initiatives by The study findings point to the need for small firms. small firms to adopt an ecological perspective in their business as one way to enhance finan- References cial performance. This is imperative nowadays Achrol, R. S., and L. W. Stern (1988). “Environ- due to a growing ecological sensitivity by both mental Determinants of Decision-Making in internal and external stakeholders. However, to Marketing Channels,” Journal of Marketing enhance an eco-oriented spirit in the firm, it is Research 25(1), 36–50. essential to monitor changes in its external Ammenberg, J., and O. Hjelm (2003). “Tracing environment closely with regard to regulatory Business and Environmental Effects of Envi- frameworks, environmental public concern, ronmental Systems—A Study of Networking competitive intensity, and market dynamism. Small and Medium-sized Enterprises Using a For this purpose, it is necessary, on the one Joint Environmental Management System,” hand, to implant policies (e.g., waste minimi- Business Strategy and the Environment 12, zation) and procedures (e.g., environmental 163–174. audits) for better understanding these forces Anderson, J. C., and D. W. Gerbing (1988). and, on the other, to cultivate the right culture “Structural Equation Modeling in Practice: A within the organization that is sensitive to eco- Review and Recommended Two-Step logical matters. Despite resource/capability Approach,” Psychological Bulletin 103(3), constraints by small firms, it seems that direct- 411–423. ing some of them toward green activities Aragón-Correa, J. A. (1998). “Strategic and/or seeking their acquisition from external Proactivity and Firm Approach to the Natural sources (e.g., trade associations, government Environment,” Academy of Management agencies, consulting organizations) will facili- Journal 41(5), 556–566. tate the successful implementation of these Aragón-Correa, J. A., N. Hurtado-Torres, S. activities. Since previous research has under- Sharma, and V. J. García-Morales (2008). lined the decisive role of the top manager in “Environmental Strategy and Performance in taking green initiatives, managers who are both Small Firms: A Resource-Based Perspective,” sensitive to ecological issues and ready to Journal of Environmental Management commit adequate time and effort toward adopt- 86(1), 88–103. ing and implementing environmental programs Armstrong, J. S., and T. S. Overton (1977). need to be appointed. “Estimating Nonresponse Bias in Mail Governments, regulatory bodies, and other Surveys,” Journal of Marketing Research nongovernmental organizations have a role to 14(3), 396–402. play in harnessing an eco-friendly orientation Arora, S., and T. N. Cason (1995). “An Experi- among small business units. For instance, they ment of Voluntary Environmental Regula- could improve awareness of ecological issues, tion: Participation in EPA’s 33/50 Program,” explain the monetary benefits from engaging in Journal of Environmental Economics and eco-friendly activities, provide financial, advi- Management 28(3), 271–286.

LEONIDOU, CHRISTODOULIDES, AND AND THWAITES THWAITES 19 15 Bagozzi, R. P., and Y. Yi (1988). “On the Evalu- Day, G. S. (1994). “The Capabilities of Market- ation of Structural Equation Models,” Driven Organizations,” Journal of Marketing Journal of the Academy of Marketing Science 58(3), 37–52. 16(1), 74–94. Dean, T. J., R. L. Brown, and V. Stango (2000). Baker, W. E., and J. M. Sinkula (2005). “Envi- “Environmental Regulation as a Barrier to ronmental Marketing Strategy and Firm the Formation of Small Manufacturing Estab- Performance: Effect on New Product Perfor- lishments: A Longitudinal Study,” Journal of mance and Market Share,” Journal of the Environmental Economics and Management Academy of Marketing Science 33(4), 461– 40(2), 56–75. 475. Dechant, K., and B. Altman (1994). “Environ- Banerjee, S. B. (2002). “Corporate Environmen- mental Leadership: From Compliance to talism: The Construct and Its Measurement,” Competitive Advantage,” Academy of Man- Journal of Business Research 55(3), 177–191. agement Executive 8(2), 7–20. Banerjee, S. B., E. S. Iyer, and R. K. Kashyap Eisenhardt, K. M., and J. A. Martin (2000). (2003). “Corporate Environmentalism: Ante- “Dynamic Capabilities: What Are They?” cedents and Inﬂuence of Industry Type,” Strategic Management Journal 21(10–11), Journal of Marketing 67(2), 106–122. 1105–1121. Barney, J. (1991). “Firm Resources and Sus- Fornell, C., and D. F. Larcker (1981). “Evaluat- tained Competitive Advantage,” Journal of ing Structural Equation Models with Unob- Management 17(1), 99–120. servable Variables and Measurement Error,” Baylis, R., L. Connell, and A. Flynn (1998). Journal of Marketing Research 28(1), 39–50. “Company Size, Environmental Regulation Fraj-Andres, E., E. Martinez-Salinas, and J. and Ecological Modernization: Further Matute-Vallejo (2009). “Factors Affecting Analysis at the Level of the Firm,” Business Corporate Environmental Strategy in Spanish Strategy and the Environment 7, 285– Industrial Firms,” Business Strategy and the 296. Environment 18(8), 500–514. Bharadwaj, S. G., R. P. Varadarajan, and J. Fahy Gadenne, D. L., J. Kennedy, and C. McKeiver (1993). “Sustainable Competitive Advantage (2009). “An Empirical Study of Environ- in Service Industries: A Conceptual Model mental Awareness and Practices in SMEs,” and Research Propositions,” Journal of Mar- Journal of Business Ethics 84(1), 45– keting 57(4), 83–99. 63. Bianchi, R., and G. Noci (1998). “ ‘Greening’ Gerbing, D. W., and J. C. Anderson (1988). “An SMEs’ Competitiveness,” Small Business Eco- Updated Paradigm for Scale Development nomics 11, 269–281. Incorporating Unidimensionality and Its Buysse, K. and A. Verbeke (2003). “Proactive Assessment,” Journal of Marketing Research Environmental Strategies: A Stakeholder 25(2), 186–192. Management Perspective,” Strategic Man- Grant, R. M. (1991). “The Resource-Based agement Journal 24, 453–470. Theory of Competitive Advantage: Implica- Carlton, D. W., and J. M. Perloff (1990). Modern tions for Strategy Formulation,” California Industrial Organization. Glenview, IL: Scott Management Review 33(3), 114–135. Foresman/Little Brown Higher Education. Hair, J. F., W. C. Black, B. J. Babin, R. E. Cordano, M., R. S. Marshall, and M. Silverman Anderson, and R. L. Tatham (2011). Multi- (2010). “How Do Small and Medium Enter- variate Data Analysis. Englwood Cliffs, NJ: prises Go ‘Green’? A Study of Environmental Pearson, Prentice-Hall. Management in the U.S. Wine Industry,” Halila, F. (2007). “Networks as a Means of Journal of Business Ethics 92(3), 463– Supporting the Adoption of Organizational 478. Innovations in SMEs: The Case of Environ- Cyprus Chamber of Commerce and Industry mental Management Systems (EMSs) Based (2011). Directory of Manufacturers. Nicosia, on ISO 14001,” Corporate Social Responsibil- Cyprus: CCCI. ity and Environmental Management 14, del Brio, J. A., and B. Junquera (2003). “A 167–181. Review of the Literature on Environmental Halila, F., and J. Tell (2013). “Creating Syner- Innovation Management in SMEs: Implica- gies between SMEs and Universities for ISO tions for Public Policies,” Technovation 14001 Certiﬁcation,” Journal of Cleaner Pro- 23(12), 939–948. duction 48, 85–92.

1620 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Hart, S. L. (1995). “A Natural-Resource-Based Masurel, E. (2007). “Why SMEs Invest in Envi- View of the Firm,” Academy of Management ronmental Measures: Sustainability Evidence Review 20(4), 986–1014. from Small and Medium-Sized Printing Heras, I., and G. Arana (2010). “Alternative Firms,” Business Strategy and the Environ- Models of Environmental Management in ment 16, 190–201. SMEs: The Case of Ekoscan vs. ISO 14001,” Menguc, B., and L. K. Ozanne (2005). “Chal- Journal of Cleaner Production 18, 726–735. lenges of the ‘Green Imperative’: A Natural Hillary, R. (2000). Small and Medium-Sized Resources Based Approach to the Environ- Enterprises and the Environment: Business mental Orientation-Financial Performance Imperatives. Shefﬁeld: Greenleaf Publishing. Relationship,” Journal of Business Research ——— (2004). “Environmental Management 58(4), 430–438. Systems and the Smaller Enterprise,” Journal Menon, A., and A. Menon (1997). of Cleaner Production 12, 561–569. “Enviropreunerial Marketing Strategy: The Jaworski, B. J., and A. K. Kohli (1993). Emergence of Corporate Environmentalism “Market Orientation: Antecedents and Conse- as Marketing Strategy,” Journal of Marketing quences,” Journal of Marketing 57(3), 53–70. 61(1), 51–67. Judge, W. Q., and T. J. Douglas (1998). “Per- Menon, A., A. Menon, J. Chowdhury, and J. formance Implications of Incorporating Jankovich (1999). “Evolving Paradigm for Natural Environmental Issues into the Stra- Environmental Sensitivity in Marketing Pro- tegic Planning Process: An Empirical Assess- grams: A Synthesis of Theory and Practice,” ment,” Journal of Management Studies Journal of Marketing Theory and Practice 35(2), 241–262. 7(2), 1–15. Kassinis, G., and N. Vafeas (2006). “Stakeholder Miles, M. P., and L. S. Munilla (1993). “Eco- Pressures and Environmental Performance,” orientation: An Emerging Business Philoso- Academy of Management Journal 49(1), phy?” Journal of Marketing Theory and 145–159. Practice 1(2), 43–50. Langerak, F., E. Peelen, and M. van der Veen Miles, M. P., L. S. Munilla, and T. McClurg (1998). “Exploratory Results on the Anteced- (1999). “The Impact of ISO 14000 Environ- ents and Consequences of Green Market- mental Management System on Small and ing,” Journal of the Market Research Society Medium Sized Enterprises,” Journal of 40(4), 323–335. Quality Management 4(1), 111–122. Lee, K.-H. (2009). “Why and How to Adopt Noci, G., and R. Verganti (1999). “Managing Green Management into Business Organiza- ‘Green’ Product Innovation in Small Firms,” tions? The Case Study of Korean SMEs in R&D Management 29(1), 3–15. Manufacturing Industry,” Management Deci- Paton, R. (1994). “Power in Organizations,” in sion 47(7), 1101–1121. Cases, Issues Concepts. Eds. R. Arson and R. Lefebvre, E., L. A. Lefebvre, and S. Talbot Paton. London: PCP, 194. (2003). “Determinants and Impacts of Envi- Patton, D., and I. Worthington (2003). “SMEs ronmental Performance in SMEs,” R&D Man- and Environmental Regulations: A Study of agement 33(3), 263–283. the UK Screen-Printing Sector,” Environ- Leonidou, L. C., C. N. Leonidou, and O. mental and Planning C: Government and Kvasova (2010). “Antecedents and Outcomes Policy 21(4), 549–566. of Consumer Environmentally-Friendly Atti- Peretz, J. H., R. A. Bohm, and P. D. Jasienczyk tudes and Behavior,” Journal of Marketing (1997). “Environmental Policy and the Management 26(13–14), 1319–1344. Reduction of Hazardous Waste,” Journal of Leonidou, L. C., C. N. Leonidou, T. A. Fotiadis, Policy Analysis and Management 16(4), and A. Zeriti (2013). “Resources and Capa- 556–574. bilities as Drivers of Hotel Environmental Perez-Sanchez, D., J. R. Barton, and D. Bower Marketing Strategy: Implications for Com- (2003). “Implementing Environmental Man- petitive Advantage and Performance,” agement in SMEs,” Corporate Social Respon- Tourism Management 35, 94–110. sibility and Environmental Management 10, McKeiver, C., and D. Gadenne (2005). “Envi- 67–77. ronmental Management Systems in Small Petts, J., A. Herd, S. Gerrard, and C. Horne and Medium Businesses,” International (1999). “The Climate and Culture of Environ- Small Business Journal 23(5), 513–537. mental Compliance within SMEs,” Business

LEONIDOU, CHRISTODOULIDES, AND AND THWAITES THWAITES 21 17 Strategy and the Environment 8(1), 14– Service Sector: An Analysis of North Ameri- 30. can and European Ski Resorts,” Canadian Podsakoff, P. M., and D. W. Organ (1986). “Self Journal of Administrative Sciences 24(4), Reports in Organizational Research: Prob- 268–283. lems and Prospects,” Journal of Manage- Shrivastava, P. (1995). “The Role of Corpora- ment 12(4), 531–544. tions in Achieving Ecological Sustainability,” Revell, A., and R. Blackburn (2007). “The Busi- Academy of Management Review 20(4), 936– ness Case for Sustainability? An Examination 960. of Small Firms in the UK’s Construction and Simpson, M., N. Taylor, and K. Barker (2004). Restaurant Sectors,” Business Strategy and “Environmental Responsibility in SMEs: the Environment 16, 404–420. Does It Deliver Competitive Advantage?” Revell, A., and R. Rutherfoord (2003). “UK Business Strategy and the Environment Environmental Policy and the Small Firm: 13(3), 156–171. Broadening the Focus,” Business Strategy Stone, G., M. Joseph, and J. Blodgett (2004). and the Environment 12, 26–35. “Toward the Creation of an Eco-oriented Revell, A., D. Stokes, and H. Chen (2010). Corporate Culture: A Proposed Model of “Small Business and the Environment: Internal and External Antecedents Leading Turning Over a New Leaf?” Business Strategy to Industrial Firm Eco-orientation,” Journal and the Environment 19(5), 273–288. of Business & Industrial Marketing 19(1), Rowe, J., and D. Hollingsworth (1996). 68–84. “Improving the Environmental Performance Stone, G. W., and K. L. Wakefield (2000). “Eco- of Small- and Medium-Sized Enterprises: A orientation: An Extension of Market Orien- Study in Avon,” Eco-Management and tation in an Environmental Context,” Journal Auditing 3, 97–107. of Marketing Theory & Practice 8(3), 21–31. Russo, M. V., and P. A. Fouts (1997). “A Teece, D. J., G. Pisano, and A. Shuen (1997). Resource-Based Perspective on Corporate “Dynamic Capabilities and Strategic Manage- Environmental Performance and Profitabil- ment,” Strategic Management Journal 18(7), ity,” Academy of Management Journal 40(3), 509–533. 534–559. Tilley, F. (1999). “The Gap between the Envi- Rutherfoord, R., R. A. Blackburn, and L. J. ronmental Attitudes and the Environmental Spence (2000). “Environmental Management Behavior of Small Firms,” Business Strategy and the Small Firm: An International Com- and the Environment 8(4), 238–248. parison,” International Journal of Entrepre- ——— (2000). “Small Firm Environmental neurial Behavior and Research 6(6), 310– Ethics: How Deep Do They Go?” Business 325. Ethics: A European Review 9(1), 31–41. Sarin, S., and V. Mahajan (2001). “The Effect of Verheul, H. (1999). “How Social Networks Reward Structures on the Performance of Influence the Dissemination of Cleaner Cross-Functional Product Development Technologies to SMEs,” Journal of Cleaner Teams,” Journal of Marketing 65(2), 35– Production 7, 213–219. 53. Walley, L., and D. Taylor (2002). “Opportunists, Schaper, M. (2002). “Small Firms and Environ- Champions, Mavericks? A Typology of Green mental Management,” International Small Entrepreneurs,” Greener Management Inter- Business Journal 20(3), 235–249. national 38, 31–35. Sharma, S., and H. Vredenburg (1998). “Proac- Williamson, D., and G. Lynch-Wood (2001). “A tive Corporate Environmental Strategy and New Paradigm for SME Environmental the Development of Competitively Valuable Practice,” The TQM Magazine 13(6), 424– Organizational Capabilities,” Strategic Man- 432. agement Journal 19(8), 729–753. Worthington, I., and D. Patton (2005). “Strate- Sharma, S., J. A. Aragón-Correa, and A. gic Intent in the Management of the Green Rueda-Manzanares (2007). “The Contingent Environment within SMEs: An Analysis of Influence of Organizational Capabilities on the UK Screen-Printing Sector,” Long Range Proactive Environmental Strategy in the Planning 38, 197–212.

1822 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Appendix Operationalization of Constructs

Constructs Item Item Description Source Code

Regulatory RFR1 Government regulations have influenced Banerjee, Iyer, and Framework very much our firm’s environmental Kashyap (2003) strategy. RFR2 Environmental legislation affects the growth of our firm. RFR3 Strict environmental regulations is a major reason for our firm to worry about its impact on the environment. RFR4 More strict regulations are required so that environmental responsible firms to be able to grow and survive. RFR5 The environmental efforts of our firm can determine future environmental legislation for our industry. RFR6 Our industry is influenced by strict environmental regulations. Environmental EPC1 The public worries too much about the Banerjee, Iyer, and Public destruction of the environment. Kashyap (2003) Concern EPC2 The public worries more about the economy rather then the protection of the environment (R). EPC3 The public shows great concern for environmental issues. EPC4 Our customers consider the protection of the environment as a critical issue facing the world nowadays. EPC5 Our customers increasingly demand products and services that are friendly to the environment. EPC6 Our customers expect our company to be friendly to the environment. EPC7 Our stakeholder (e.g., banks, suppliers) consider environmental protection a critical issue facing the world. EPC8 Our stakeholders are increasingly pressing our firm to produce goods friendly to the environment. EPC9 Our stakeholders expect from our company to be friendly to the environment.

LEONIDOU, CHRISTODOULIDES, AND AND THWAITES THWAITES 23 19 Appendix Continued

Constructs Item Item Description Source Code

Competitive CIN1 Firms in our industry spend a large part of Sarin and Mahajan Intensity their sales on marketing efforts because (2001) of growing competition. CIN2 In our industry, companies firms compete fiercely in order to maintain their market share. CIN3 The competition in our industry is intense. CIN4 Firms in our industry follow a philosophy of peaceful coexistence (R). Market MDY1 The production technology in our market Baker and Sinkula Dynamism has changed in the last three years. (2005) MDY2 The level of competitive intensity in our industry is high. MDY3 The rate of market change in our industry is high. Eco-friendly EFO1 The environmental issues are very relevant Fraj-Andres, Orientation to the basic function of our company. Martinez-Salinas, EFO2 In our firm, we put an effort in making and Matute-Vallejo each employee understand the meaning (2009) of environmental protection. EFO3 We try to promote environmental protection as the objective of all departments in our company. EFO4 Our firm has a clear policy to promote environmental conscious in all business areas. EFO5 Environmental protection is a top priority issue in our company. EFO6 The protection of the environment is a central value in our company. EFO7 The natural environment has an impact on the business activity of our firm. EFO8 The good financial situation of our firm depends on the condition of the natural environment. EFO9 In our firm, the protection of the environment contributes to a great extent to maintain its good image. EFO10 Environmental protection is of vital importance for the survival of our firm. EFO11 Our firm tries to have the image of an environmentally responsible organization.

2024 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Appendix Continued

Constructs Item Item Description Source Code

Financial FIP1 Profits Leonidou et al. performance FIP2 Profit growth (2013) FIP3 Return on assets FIP4 Return on investment FIP5 Sales FIP6 Sales growth FIP7 Cash flow Organizational ORE1 Our firm has made investments in the Buysse and Verbeke Resources production processes that are related to (2003) environmental skills. ORE2 Our firm has made investments in the environmental abilities of its employees. ORE3 Our firm has made investments in developing the environmental skills of the top management. ORE4 Our firm has made investments in organizational abilities that are related to environmental issues. ORE5 Our firm has made investments in research and development that are relevant to environmental issues. Organizational OCA1 Our firm has the ability to seek solutions Sharma, Capabilities for environmental issues from different Aragón-Correa, and angles. Rueda-Manzanares OCA2 Our firm pays great attention in satisfying (2007) customer demands. OCA3 Our firm focuses on having at its disposal pioneering, flexible, and innovative technologies. OCA4 In our firm, the managers and employees always agree to adopting the right environmental procedures. OCA5 In our firm, there are formal/informal systems for better coordinating green issues among departments. OCA6 Our firm always expands its knowledge regarding the interaction between business and physical environment.

Note: The sign (R) indicates a reversed scale. Measurement scales were based on a 7-point Likert scale, ranging from 1: strongly disagree to 7: strongly agree.

LEONIDOU, CHRISTODOULIDES, AND AND THWAITES THWAITES 25 21 Journal of of Small Small Business Business Management Management 2016 2014 54(1), ••(••), pp. pp.26–48 ••–•• doi: 10.1111/jsbm.12126 The Resource-Based View in Entrepreneurship: A Content-Analytical Comparison of Researchers’ and Entrepreneurs’ Views by Franz Kellermanns, Jorge Walter, T. Russell Crook, Benedict Kemmerer, and Vadake Narayanan

The resource-based view (RBV) is one of the most influential perspectives in the organizational sciences. Although entrepreneurship researchers are increasingly leveraging the RBV’s tenets, it emerged in strategic management. Despite some important similarities between entrepreneurship and strategic management, there are also important differences, raising questions as to whether and to what extent the RBV needs to be adapted for the entrepreneurship field. As a first step toward answering these questions, this study focuses on resources as the fundamental building block of the RBV and presents a content-analytical comparison of researchers’ and practicing entrepreneurs’ resource conceptualizations to derive similarities and differences between established theory and entrepreneurial practice. We find that although the two conceptualizations exhibit some overlap, there are also important differences in the emphasis on different dimensions of resources and ownership requirements, as well as in the understanding of how those resources shape outcomes. These results suggest important contextual conditions when applying the RBV’s tenets within the field of entrepreneurship.

fields in the organizational sciences to under- Introduction stand entrepreneurs and entrepreneurial ven- As the field of entrepreneurship matures tures (Ireland, Webb, and Coombs 2005). The (Busenitz et al. 2003), entrepreneurship resource-based view (RBV) has grown into one researchers continue to leverage theoretical of the most influential theoretical perspectives perspectives from other, more established in the organizational sciences (Barney, Wright,

Franz W. Kellermanns is Addison H. & Gertrude C. Reese Endowed chair and professor of management in the Belk College of Business at the University of North Carolina—Charlotte and holds a joint appointment with the Center for Family Business at the WHU-Otto Beisheim School of Management (Germany). Jorge Walter is associate professor in the Department of Strategic Management & Public Policy at the School of Business at The George Washington University. T. Russell Crook is associate professor in the Department of Management at the College of Business Administration at the University of Tennessee. Benedict Kemmerer is head of Product Management Electric Cooktops at BSH Bosch und Siemens Hausgeräte GmbH. Vadake Narayanan is Deloitte Touche Jones Stubbs professor and associate dean for research at the LeBow College of Business at Drexel University. Address correspondence to: Franz W. Kellermanns, Department of Management, Belk College of Business, University of North Carolina—Charlotte, Charlotte, NC 28223. E-mail: [email protected].

26 JOURNAL OFKELLERMANNS SMALL BUSINESS ET AL. MANAGEMENT 1 and Ketchen 2001), and entrepreneurship research likely focuses on resources that are researchers have built on insights from this more relevant to larger, more established orga- theory to understand the determinants of entre- nizations. Indeed, Stevenson (1983) defined preneurial venture performance (e.g., Alvarez entrepreneurship as “the pursuit of opportu- and Busenitz 2001; Chandler and Hanks 1994; nity beyond the resources that you currently Chrisman, Bauerschmidt, and Hofer 1998; control.” Accordingly, entrepreneurial ventures Wiklund and Shepherd 2003). Indeed, within might require different resources, or use entrepreneurship, Google Scholar’s citation these resources differently, to survive and counts show that Barney’s (1991) and prosper compared with larger, more estab- Wernerfelt’s (1984) seminal RBV articles con- lished organizations (Unger et al. 2011; tinue to attract more and more interest. This Wiklund and Shepherd 2009). For example, suggests that the RBV is being leveraged with certain resources (e.g., slack, scientific, and greater frequency in entrepreneurship and that technical resources) have been identified as it is becoming increasingly influential. highly useful for high-technology ventures The RBV emerged in the field of strategic when faced with adverse shocks that threaten management, however, which tends to study their survival (De Carolis et al. 2009); and larger, more established organizations. More- Chrisman, Chua, and Kellermanns (2009) over, the RBV was intended to help researchers showed that resources have a differential understand why some firms enjoy a competi- impact in family and nonfamily firms. Without tive advantage, and thereby outperform other addressing this difference, only limited theo- firms (Barney 1991). According to the RBV, retical progress can be made within the field “strategic” resources—resources that are valu- as well as potentially less impactful prescrip- able, rare, inimitable, and nonsubstitutable— tions for entrepreneurs (Bettis 1991; Whetten are the key differentiators between firms that 1989). have advantages vis-à-vis those that do not Because of such differences, it remains (Barney 1991). A growing body of evidence unclear whether or not “the tenets of the supports this notion, and researchers continue resource-based view are applicable to both to identify the types of resources that meet entrepreneurial ventures and established firms” these criteria (Barney and Arikan 2001; Crook (Hitt et al. 2002, p. 4). Instead, when entrepre- et al. 2008). neurship researchers apply theories developed Despite important similarities between the in other fields, such as the RBV, they should fields of strategic management and entrepre- “discuss how the assertions/assumptions neurship, and despite strategic management remain the same or change when used to form theories, such as the RBV, offering important theory-driven testable relationships dealing insights into entrepreneurs and entrepreneur- with entrepreneurship questions” (Ireland, ial ventures, two issues persist. First, though Webb, and Coombs 2005, p. 124). the RBV has become increasingly popular, it Given the increasing influence of the RBV in has been criticized as resources remain ill- entrepreneurship, its origin in a different field, defined, inconsistent, and even contradictory and potential differences between entrepre- across studies (e.g., Bromiley and Fleming neurial ventures and larger, established organi- 2002; Priem and Butler 2001a, 2001b). In zations in their understanding of resources, we other words, researchers have not yet arrived set out to close the gap between what we know at a consensus definition of resources and and what we should know about the RBV in their dimensions, leaving some RBV research- the context of entrepreneurship. In particular, ers puzzled as to what exactly constitutes a our study empirically derives and compares resource (Kraaijenbrink, Spender, and Groen researchers’ and entrepreneurs’ conceptualiza- 2010). Second, there are important differences tions of resources by utilizing content analysis between the fields of strategic management to distill resource definitions from a sample of and entrepreneurship. One key difference is 117 published articles in influential academic that entrepreneurship researchers study journals and from a sample of 201 practicing younger and smaller ventures that are in a entrepreneurs. Our results show both some pursuit of growth, whereas strategic manage- overlap between the two conceptualizations, ment researchers study larger, established but also identify a number of important differ- organizations (Carland et al. 1984; Ireland, ences between researchers’ and entrepreneurs’ Webb, and Coombs 2005). Thus, extant RBV resource definitions. Moreover, both researcher

2 JOURNAL OFKELLERMANNS SMALL BUSINESS ET AL. MANAGEMENT 27 and entrepreneur samples, albeit with different (Barney 1991, 2001), and a recent meta-analysis emphases, included outcomes associated with of the available empirical evidence supports this resources—such as the creation of products assertion (Crook et al. 2008). and/or services, of value/success, and of a com- The RBV developed initially in the field of petitive advantage—as an integral part of their strategic management. Though there is consid- resource conceptualizations. erable overlap between strategic management We intend to make three contributions. First, and entrepreneurship, the underlying domains by decomposing a representative sample of are distinctive. A consensus definition of strate- scholarly resource definitions into their differ- gic management developed by Nag, Hambrick, ent dimensions and by comparing the impor- and Chen (2007, p. 944) states that “strategic tance of each dimension across definitions, we management deals with the major intended and shed light on agreements and disagreements emergent initiatives taken by general managers among researchers on what constitutes the on behalf of owners, involving utilization of main elements of an academic resource defini- resources to enhance the performance of firms tion and thereby address the criticisms of in their external environment.” Thus, though the RBV regarding the conceptualization of strategic management deals with managers, resources (e.g., Kraaijenbrink, Spender, and entrepreneurship deals with “individuals or Groen 2010; Priem and Butler 2001a, 2001b). groups of individuals, acting independently, or Second, by comparing researchers’ and entre- as part of a corporate system, who create new preneurs’ resource definitions, we shed light on organizations, or instigate renewal within an agreements and disagreements between the existing organization” (Sharma and Chrisman RBV and entrepreneurial practice. This allows 1999, p. 17). Though entrepreneurs can exist in us to investigate whether resources are viewed large established organizations (often investi- to have the same or different dimensions and to gated under the umbrella of corporate entrepre- differentiate among resources that are consid- neurship), the focus of our paper is on the most ered more “strategic” to entrepreneurs. Third, prevalent form of entrepreneurship, namely and more broadly, our study helps increase individuals acting independently (Chrisman and awareness about the unique conditions when Kellermanns forthcoming). Entrepreneurship applying the RBV’s tenets within the domain of research in our focal domain focuses on organi- entrepreneurship. zations that (1) are typically smaller and newer (Carayannopoulos 2009; Carland et al. 1984); (2) Literature Review are more reliant on interorganizational relation- The RBV traces its intellectual roots to Edith ships (Chua et al. 2011; Wiklund and Shepherd Penrose (1959), who focused on the role of 2009); and (3) do not yet have established resources in enabling or constraining organiza- reputations (Fischer and Reuber 2007). More- tional growth. She defined resources as “the over, entrepreneurs themselves differ from the physical things a firm buys, leases, or produces general population (Zhao, Seibert, and Lumpkin for its own use, and the people hired on terms 2009) and focus on aggressive growth (Carland that make them effectively part of the firm” et al. 1984). (Penrose 1959, p. 60). Over more than 50 years, Given the differences between entrepreneur- researchers have built on Penrose’s insights, and ial ventures and larger, more established orga- as the RBV evolved, researchers have focused nizations, it is likely that there are important more specifically on “strategic resources” (Amit differences involving how practicing entrepre- and Schoemaker 1993). Strategic resources are neurs view resources and how they are viewed those resources that (1) have value, such that by the RBV. This suggests that for the RBV to they can be leveraged to increase customer continue to evolve within the domain of entre- value or cut costs; (2) are rare, such that com- preneurship, there might be a need to study petitors do not have access to the same or a very whether different resources are needed by similar resource to compete away the value; and entrepreneurs to succeed (Alvarez and Busenitz (3) are difficult to substitute and/or imitate, 2001; Zhao, Seibert, and Lumpkin 2009). Rec- which allows the organization to stay ahead of ognizing this, there have been recent calls for competitors (Barney 1991). The central asser- the RBV to be contextualized (e.g., Siqueira and tion within the RBV is that organizational Bruton 2010)—particularly for entrepreneur- advantages are enhanced to the extent that an ship research. Yet, without an analysis of organization possesses strategic resources differences between the RBV and entrepre-

28 JOURNAL OFKELLERMANNS SMALL BUSINESS ET AL. MANAGEMENT 3 neurs’ views on resources, there will be linger- and Groen’s (2010, p. 359) recent review of the ing ambiguity about potential boundary RBV literature, in which they summarize their conditions and the contextualization of the RBV assessment that “we are left puzzled about the when applied to entrepreneurship research RBV’s core concept [of resources].” This (Busenitz et al. 2003; Welter 2001). Our analy- problem is aggravated as the RBV was devel- sis of the RBV in the field of entrepreneurship oped in the realm of strategic management therefore represents an “evocative study,” research, albeit informed by economics (Barney which deals with “specific domains where and Arikan 2001), which is predominantly general theoretical frameworks may be avail- focused on larger, more established organiza- able, but operationalization of concepts and tions. In sum, we do not yet have a clear specification of linkages among the concepts understanding of and/or agreement on how are still unknown,” and for which “we need resources are conceptualized, which is a signifi- approaches and techniques that fill the gaps cant shortcoming of a theory as influential as the between qualitative identification and quantita- RBV. In addition, we do not know if and how tive verification, and that evoke the constructs this conceptualization contrasts with those of and linkages particular to the specific domain” practicing entrepreneurs, who are mostly active (Nelson et al. 2000, p. 482). In the next section, in smaller, less established organizations. we will review prior attempts at defining resources, before outlining our derivation of a Practicing Entrepreneurs’ consensus definition. Resource Conceptualizations Though research has conceptualized Researchers’ Resource resources in a variety of ways, the way practi- Conceptualizations tioners actually view resources has virtually not Ever since Barney (1991, p. 110) defined been addressed. Very few empirical studies resources as “all assets, capabilities, organiza- have attempted to elicit practitioners’ percep- tional processes, firm dimensions, information, tions of their resources. A rare exception in knowledge, etc. controlled by a firm that enable the context of established organizations is the firm to conceive of and implement strategies Stevenson (1976, 1984), who interviewed 50 that improve its efficiency and effectiveness,” managers in six companies regarding their researchers have complained that “virtually any- strengths and weaknesses. He found an almost thing associated with the firm can be a resource” even distribution between organizational, per- (Priem and Butler 2001a, p. 32, emphasis in the sonnel, marketing, and technical issues, with original). If resources are considered to be an financial strengths and weaknesses having all-encompassing concept, however, resources fewer references. He concluded, however, that become essentially meaningless as a way to “[d]efinitions of strengths and weaknesses gen- explain organizational advantage and above- erally applicable for whole organizations were average performance (Conner 1991). For not found” (Stevenson 1976, p. 68). In contrast example, the conceptualization by Wernerfelt to this open-ended approach, Hall (1992) speci- (1984, p. 172) of resources as “anything which fied 12 intangible resources and then asked could be thought of as a strength or weakness of CEOs in the United Kingdom to specify to a given firm” has been criticized as “subjective which extent those resources made a contribu- and vague, because individuals can differ in tion to the overall success in their business. He what characteristics they think of as a ‘strength’ found that answers were surprisingly uniform or ‘weakness’” (Bromiley and Fleming 2002, across business types and focused on reputa- p. 324). Not surprisingly, this conceptual vague- tion and know-how. ness also translates into a plethora of conceptu- Conceptualizations of resources in an entre- alizations (as we will outline further) and preneurial context are equally rare. Brush and measurements (for an overview of resource colleagues (Brush et al. 1997; Greene, Brush, measures, see Hoopes, Madsen, and Walker and Brown 1997) studied small business 2003). Accordingly, in his reflection on the 10 owners’ relative favorability ratings of resource years since the publication of his seminal paper types (i.e., human, social, organizational, physi- on the RBV, Wernerfelt (1995, p. 172) concludes cal, and financial resources). Later, Lichtenstein that “‘resources’ remain an amorphous heap to and Brush (2001) tracked three ventures over a most of us.” That this critique is still valid can be nearly one-year time frame and repeatedly seen, for instance, in Kraaijenbrink, Spender, asked which resources the ventures had been

4 JOURNAL OFKELLERMANNS SMALL BUSINESS ET AL. MANAGEMENT 29 acquiring or should have been acquiring at this stage in the company’s development. They Methods found that the most salient resources were RBV’s Resource Definitions mostly intangible. To identify a representative sample of how Taken together, there is initial evidence sug- resources have been defined in the extant gesting that resources may have differential RBV—and to minimize subjectivity and effects in small business and entrepreneurial arbitrariness—we built on the approach devel- context. Indeed, we know that entrepreneurs do oped by Nag, Hambrick, and Chen (2007). Our not seem to follow the prescriptions of the RBV goal was to capture a representative set of when they evaluate resources (Kemmerer et al. definitions; it was neither to be all- 2011); yet, we implicitly assume that entrepre- comprehensive, in the sense of including all neurs conceptualize resources the same way articles on the RBV, nor to include tangential academics do. This is unwarranted, however, as articles that reference the RBV. We thus recent research shows. Achtenhagen, Naldi, and excluded a number of articles examining spe- Melin’s (2010) study shows that entrepreneurs cific types of resources without discussing the and scholars understand and conceptualize general concept, such as articles focusing growth quite differently. Indeed, they suggest exclusively on human or managerial resources that different conceptualizations and measure- (e.g., Castanias and Helfat 1991) or specific ment may have to be considered for different technological resources (e.g., Miller 2004). types of organizations. Such a potential incon- Though these specific resources are certainly gruence between scholars and practitioners may worth studying, our goal here was to compile a be even more concerning in the realm of the representative list of generic resource defini- RBV, as this theory is readily applied to an tions used in extant research. entrepreneurial context. Yet, we may arrive at We identified relevant articles with the fol- false prescriptions, which likely would have lowing set of criteria (for a detailed overview more severe consequences for fledging busi- on this process, please refer to Table 1): (1a) nesses. Thus, without congruence, it remains we performed a search of the EBSCO Host unknown whether the findings of prior RBV database for all published articles in research studies have focused on resources that are con- journals1 that were identified as “influential” in sidered important in entrepreneurial contexts, the field of management by Podsakoff et al. or whether the findings cannot be applied (2005);2 (1b) we also included articles from the beyond the boundaries of the strategic manage- EconLit database to adequately cover journals ment field. Accordingly, we investigate the fol- in this adjacent field; (2) we employed two lowing research question: filters for those articles that contain at least one Research Question: What are the similari- primary keyword3 in their title or abstract; and ties and differences in the RBV’s and practicing (3) for those that did not contain at least one of entrepreneurs’ resource conceptualizations? 16 additional keywords4 in their title or

1By restricting our search to scholarly journal articles (as opposed to book chapters or unpublished works), we enhanced quality control because of the rigorous peer review process to which articles published in such journals are subjected prior to publication David and Han (2004). 2As our interest lies in identifying researchers’ resource deﬁnitions, we excluded practitioner-oriented journals, such as California Management Review, Harvard Business Review, Sloan Management Review, and the Journal of Vocational Behavior. To further enhance the comprehensiveness of our sample, we also added the journals Entrepreneurship: Theory & Practice and Organization Science. Our results remained unchanged, however, with and without the inclusion of entrepreneurship journals. 3Following Newbert Newbert (2007), our primary key words were resource-based* and RBV* (the asterisk at the end of a search word allows for different sufﬁxes). 4Our additional keywords also follow Newbert (2007) and include: competitive advantage, perform*, valu*, capability*, intangib*, heterogen*, rare*, imitab*, inimitab*, immob*, non-substitutab*, substitutab*, tangib*, Barney, competenc*, and organiz*.

30 JOURNAL OFKELLERMANNS SMALL BUSINESS ET AL. MANAGEMENT 5 Table 1 Summary of Selection Criteriaa

Filter Type Description EBSCO Host EconLit Total Results Results

Substantive Article must appear in one of 65,885 the scholarly journals selected as relevant for our analysis Substantive All articles with 322 188 510 “resource-based*”, “RBV*”, or “RBT*” in title or abstract Substantive At least one of 16 keywordsb 269 121 390 must also appear in title or abstract Substantive Article must appear in a journal 267 (2 sole pubs) 70 (51 sole pubs) 337 that has returned more than one item from the filters above Substantive Remaining abstracts read for 251 (16 excl.) 59 (11 excl.) 310 substantive relevance Substantive Remaining full articles read for 93 (158 excl.) 26 (33 excl.) 119 both substantive relevance and definition of terms “resource” or “resources” Duplicates Consolidating both databases by (2 excl.) 117 removing duplicate articles that appeared in both aThe selection filters used herein are adapted from those developed by David and Han (2004) and Newbert (2007). bThe 16 keywords are: competitive advantage, perform*, valu*, capability*, intangib*, heterogen*, rare*, imitab*, inimitab*, immob*, non-substitutab*, substitutab*, tangib*, Barney, competenc*, and organiz* Newbert (2007).

abstract, that is, we ensured substantive rel- We need to note that for analytical purposes, evance of the articles; (4) we further eliminated we extracted the definitions in their entirety. As articles that appeared in journals in which only we will describe in more detail in the results one article appeared overall, which makes and discussion sections, the definitions con- these articles more likely to be removed from tained both different dimensions and outcomes the core tenets of the RBV (Newbert 2007); (5) of resources. Though it was not our initial we further verified the substantive relevance of intent to capture these outcomes, analyzing all remaining articles by reading their abstracts; researchers’ resource definitions suggests that and (6) we consolidated results from the they are an integral part of resource definitions EBSCO Host and EconLit databases to eliminate and we thus include them in our subsequent duplicate articles. Our search process and the analyses. series of filters distilled a sample of 230 relevant articles. By reading through these articles Entrepreneurs’ Resource Definitions in their entirety, we managed to extract 117 Given our research objective, we strived to researcher definitions of resources (with the identify practicing entrepreneurs who pos- remaining 113 articles not providing any sessed deep knowledge about the resources resource definitions). needed to help an entrepreneurial venture

6 JOURNAL OFKELLERMANNS SMALL BUSINESS ET AL. MANAGEMENT 31 function. Thus, we did not choose our respon- Table 2 dents on a random basis but instead targeted Survey Respondent respondents who possessed such knowledge (Kumar, Stern, and Anderson 1993). Our Profile—Venture Characteristics sample consists of individuals currently or formerly enrolled in the FastTrac Planning Stage of Venture program, a 2- to 4-month comprehensive entre- Planning Stage 5.0 percent preneurship education program that teaches <1 Year 5.0 percent entrepreneurs business and leadership skills 1–5 Years 33.8 percent and provides them with networking opportuni- >5 Years 53.3 percent ties. FastTrac classes are offered in 150 cities in No Longer Exists 2.9 percent 38 states through local organizations such as Industry universities, business development councils, or Retail 20.3 percent chambers of commerce. The FastTrac Planning Service 44.5 percent program our study focused on was adminis- Wholesale 1.6 percent tered directly by the Kauffman foundation and Manufacturing 13.6 percent specifically targeted individuals in the start-up Construction 14.8 percent phase or with a young business to help them Other 4.9 percent transform and grow their venture. As such, we Number of Employees assured that our participants ran entrepreneur- Range 0–110 ial ventures and not nongrowth-oriented, small Mean 6.0 businesses as described in the literature Std. Dev. 11.7 (Carland et al. 1984). Of note, no RBV-related Median 2.0 subjects were taught as part of the program, so Proportion of Ventures 34.2 percent it is unlikely that respondents’ assessments with at Least 5 Employees would be subject to a social desirability bias, or Entrepreneurial Involvement a priming effect where respondents were cued Full-time 75.5 percent (Podsakoff et al. 2003). Part-time 19.9 percent We conducted a number of formal pilot inter- No Longer Involved 4.6 percent views with entrepreneurs or former entrepreneurs active in the Midwest region of the United States both to gain insights into entrepreneurs’ conceptualizations of resources and to test and percent between 25 and 34 years old; 51.8 refine the wording of our survey. As part of a percent between 35 and 49; 36.1 percent larger study, we then sent out 1,600 individual between 50 and 65; and .6 percent over 65. A surveys (31 of which were not received) and percentage of 34.9 held a business or business- handed out 33 additional surveys to FastTrac related degree. Not including their present participants. The envelopes and cover letters business, an average respondent had founded used Kauffman Foundation logo and letterhead 1.4 other businesses and had 14.4 years of respectively, and the letter was sent out in the work experience. A percentage of 20.3 of the name of the manager of the national FastTrac ventures were active in retail, 44.5 percent in program to further increase the perceived legiti- service, 1.6 percent in wholesale, 13.6 percent macy of the mailing. Of 1,602 potential respon- in manufacturing, 14.7 percent in construction, dents, 41 individuals declined to respond. Two and 4.9 percent in other industries. The average hundred forty-two at least partly filled out respondents’ venture had six employees (not surveys, resulting in a response rate of 15.5 including the respondent), and slightly more percent, and 202 usable definitions were than half of the ventures were more than five obtained, leading to a final response rate of years old. The focus of our sample is in line 12.55 percent, which is comparable with other with the wider entrepreneurship literature, recent studies on the RBV and on entrepreneurs which focuses on small or new businesses (e.g., Chua et al. 2011; Ray, Barney, and (Shane and Venkataraman 2000). Table 2 pro- Muhanna 2004; Sullivan and Marvel 2011; vides additional characteristics of the ventures Ucbasaran, Westhead, and Wright 2009). that are similar compared with other entrepre- A percentage of 46.7 of respondents were neurship studies (e.g., Haynie, Shepherd, and female. The age range of respondents was: 11.4 McMullen 2009).

32 JOURNAL OFKELLERMANNS SMALL BUSINESS ET AL. MANAGEMENT 7 Table 3 Comparison of Early and Late Respondents

Employees No. of Ventures Gender Involvement Founded

Mann–Whitney U 6158 5761 6478 6559.5 Wilcoxon W 12599 12089 13618 13114.5 Z −.238 −1.764 −.556 −.58 Asymp. Sig. ( two-tailed) .812 .078† .578 .562

Bus. Basis of Age Venture Degree Competition Stage

Mann–Whitney U 6550 6669.5 5268.5 6588.5 Wilcoxon W 13690 13224.5 11709.5 13029.5 Z −.416 −.248 −3.186 −.294 Asymp. Sig. .677 .804 .001*** .769

†p < .10 *p < .05 **p < .01 ***p < .001

We compared proportions and response extended narratives was infeasible because of rates for those characteristics for which infor- time and resource constraints barring us from mation on the full sample was available. These interviewing large numbers of respondents to analyses showed that women were dispropor- obtain a decent-size sample of such narratives. tionately more willing to complete the survey Instead, we asked participants to think about (p < .001), which is in line with previous their most important organizational resources, studies (Green 1996) and that there were some and then to answer the following survey ques- geographic effects of marginal statistical signifi- tion: “If somebody asks you to briefly explain cance (p = .06); for instance, respondents with what you mean by the term ‘resources’ in the zip codes of 7xxxx were more than twice as context of your venture, what would you likely to respond as those with a zip code of answer?” Furthermore, the area designated for 0xxxx. To further mitigate concerns of sample the answer started with the prompt: “Resources bias, we conducted a test between early and are . . . .” The prompt was designed to make it late respondents, as late respondents are pre- easier for respondents to provide their resource sumed to be more similar to nonrespondents conceptualizations. (e.g., Kanuk and Berenson 1975). Our findings (reported in Table 3) showed that respondents Analyses only differ in age and marginally in the number Our analytical method closely follows the of ventures founded. procedure outlined by Nag, Hambrick, and The methodological strategy to elicit resource Chen (2007) in their analysis of a consensus conceptualizations from our respondents was definition of the strategic management field. guided by the desire to preserve as much as In particular, we performed two identical possible the meanings and natural language of but separate analyses of researchers’ and the respondents themselves. Conceptualizations entrepreneurs’ resource definitions. With the of resources, however, are impossible to effec- help of the computer-aided text analysis tively study through revealed behavior. Simi- software Concordance (Watt 2004), we con- larly, revealing conceptualizations through ducted a content analysis (Neuendorf 2002)

8 JOURNAL OFKELLERMANNS SMALL BUSINESS ET AL. MANAGEMENT 33 to identify the most frequently recurring, dis- inductive and iterative hermeneutic (Forster tinct dimensions and features of both 1999; King 1998). First, repeated reading of the researchers’ and entrepreneurs’ conceptualiza- list of root words enabled us to let tentative tions of resources. We chose to examine indi- dimensions emerge from our two sets of defi- vidual words rather than entire phrases, nitions, based upon conceptual clusters of root which minimizes researchers’ biases of inject- words. In the entrepreneurial sample, for ing a priori judgments as to the word combi- instance, the root words “knowledge,” “skill,” nations that might be sought out (Nag, “information,” and others were consolidated Hambrick, and Chen 2007). into the dimension “human capital.” For the To make the large number of distinct words sake of parsimony, we wanted to keep the contained in the definitions analytically trac- number of dimensions as small as possible, but table, we restricted our analysis to those words we also needed to ensure that all root words that appeared three times or more among all assigned to a specific dimension were inter- the definitions (Nag, Hambrick, and Chen nally coherent. Finally, to maintain simplicity, 2007). We also excluded proper nouns, prepo- we assigned any given root word to only one sitions, articles, and common descriptors such dimension, even though it might reasonably as “very,” “much,” and “many,” and all very belong to other, additional dimensions. common verbs and adjectives, such as “get,” After agreement on the dimensions was “keep,” “strong,” and “high” which, by them- reached, two of the authors independently selves, have no inherent or consistent meaning. assigned root words to the dimensions. Inter- We then re-read the identified root words rater agreement between them was 92 percent within their original context to ensure that the and 90 percent (with associated Cohen 1960 meaning of the root words was similar across kappa values of .92 and .89) for the research definitions in the same sample, and subse- and entrepreneurial samples, respectively.6 quently excluded all words that had ambiguous Consensus exists among the various guidelines meanings across definitions.5 And finally, we for the interpretation of Cohen’s kappa that a consolidated all variations of a root word— kappa of .8 or above indicates good to excel- such as singular/plural, present tense/past lent reliability (Neuendorf 2002). This means tense, or variations of the same word, such as that the dimensions identified here are cap- “bank,” “banker,” “banking”—and treated them tured in a way that allows for replication and is collectively. This consolidation process led to not merely spurious. 106 and 99 distinct root words researchers and To statistically compare research and entre- entrepreneurs, respectively, used to define preneurs’ definitions, we performed a binary resources. These root words then formed the logistic regression analysis, where the pre- basis for imputing emergent resource defini- dicted value represents the probability that a tions for both groups. given definition is from the research (versus the Though our overall approach was content entrepreneurial) sample. Thus, the dependent analytic—focused on the reliable, quantitative variable was coded “1” for all researcher defi- identification of particular dimensions and fea- nitions and “0” for all entrepreneurial defini- tures of the definitions—the way the emergent tions. Each of the dimensions represents one resource definitions were derived was more predictor variable and is coded “1” if a given

5In the practitioner sample, for example, the word “name” was used in the sense of “Marketing gimmicks— Who is the best at getting your name out” (respondent #10053); “copyright protected or trademarked technology or name” (respondent #12062); and “clients, productivity, and services to name a few” (respondent #13026). 6Cohen’s kappa corrects for the role of chance agreement by scaling agreement in such a way that a kappa of 1 denotes perfect agreement and a kappa of 0 denotes agreement at chance level. More formally,

PAOE− PA Cohen’ s kappa = , where PAO represents proportion of inter-rater agreement observed and PAE 1 − PAE denotes the proportion of agreement expected by chance.

34 JOURNAL OFKELLERMANNS SMALL BUSINESS ET AL. MANAGEMENT 9 definition mentions one or more of the root respect to our researcher sample, this reso- words associated with each dimension and “0” nates with previous theoretical treatments of otherwise.7 the RBV that have cautioned against the potential for circular reasoning inherent in distin- Results guishing resources from nonresources by their Our two content analyses led to the emer- performance implications (e.g., Bromiley and gence of 12 distinct dimensions, which together Fleming 2002; Priem and Butler 2001a). It is constitute the two resource definitions. To remarkable, however, that the entrepreneurs in distill the two consensus resource definitions, our sample, without being prompted, also we retained those dimensions for each perspec- included resource outcomes as part of their tive that were referenced in at least 20 percent resource conceptualizations. of the definitions, or 24 and 40 times, respec- Though the two consensus definitions indi- tively, in the research and entrepreneurial cate a substantial overlap between both per- samples (Tables 4 and 5 provide details). In the spectives, the results from our binary logistic researcher sample, this led to the exclusion of regression analysis also indicate a number of the dimensions “products/services” (16 refer- significant differences between researchers and ences) and “sustainable” (11 references); in the entrepreneurs (see Table 6 for details). Our entrepreneurial sample, we excluded the overall model is highly significant, explains 37 dimensions “organizational capital” (24 refer- percent of variation in our resource definitions, ences), “ownership” (12 references), “goals” and correctly categorizes slightly more than 81 (26 references), and “competitive advantage” (5 percent of definitions. The significance level of references). the Wald statistic for each dimension deter- For the consensus definitions of resources, mines its usefulness for classifying definitions see Tables 4 and 5. Among researchers, the into the researcher or entrepreneurial category. most referenced dimensions are human capital Exp(B) represents the ratio-change in the odds (280 references), creation (195 references), of a definition being associated with a and firm (145 references), followed by physical researcher for a one-unit change in the predic- capital (141 references) and assets (140 tor. When Exp(B) is less than 1, increasing references). The most commonly referenced values of the dimension correspond to increas- journal articles in resource definitions are ing odds of the definition being part of our Barney (1991), with 56 references, followed entrepreneurial sample (and decreasing odds of by Wernerfelt (1984) with 30 references, the definition being part of our researcher Grant (1991) with 19 references, Amit and sample). Conversely, when Exp(B) is greater Schoemaker (1993) with 15 references, and than 1, increasing values of the dimension cor- Penrose (1959) with seven references. Among respond to increasing odds of the definition entrepreneurs, the most referenced dimensions being part of our researcher sample (and are creation (222 references), human capital decreasing odds of the definition being part of (205 references), and firm (163 references), fol- our entrepreneurial sample). lowed by relationship capital (149 references) In particular, assets (p < .001), human capital and assets (115 references). Interestingly, both (p < .01), organizational capital (p < .01), physi- researchers’ and entrepreneurs’ consensus defi- cal capital (p < .05), ownership (p < .001), and nitions also contain references to the outcomes competitive advantage (p < .001) are statistically associated with resources, such as products/ significant dimensions for classification and services (47 references in the entrepreneurs indicate researcher definitions. Conversely, sample), value/success (65 and 93 references, the dimensions relationship capital (p < .10), respectively, in the researchers and entrepre- products/services (p < .05), and value/success neurs samples), and competitive advantage (32 (p < .05) are also statistically significant dimen- references in the researchers sample). With sions but indicate entrepreneurial definitions.

7As a robustness test, we also used an alternative specification in which each predictor variable is a count of how many of the root words associated with each dimension are mentioned in a given definition. In this case, a predictor represents a dimension’s weight for a given definition. The results remained substantively unchanged, however, so we present the results of the binary predictors instead.

10 JOURNAL OFKELLERMANNS SMALL BUSINESS ET AL. MANAGEMENT 35 Table 4 Researchers’ Resource Deﬁnitions

Assets (140) Human Organizational Financial Physical Relationship Capital (280) Capital (90) Capital (35) Capital (141) Capital (64)

Asset (55) Capability (46) Process (27) Financial (30) Physical (37) Reputation (17) Intangible (37) Knowledge (40) Routine (10) Equity (5) Technology (33) Market (9) Tangible (25) Human (39) System (10) Equipment (17) Relation (9) Factor (10) Skill (25) Structure (8) Plant (11) Available (6) Bundle (6) Brand (14) Culture (7) Material (9) Access (5) Complex (4) Information (14) Planning (6) Stock (9) Contract (5) Observable (3) Competency (11) Coordination (5) Land (7) Customer (5) Experience (8) Procedure (5) Location (6) Network (5) Patent (8) Team (5) Geographic (5) Loyalty (3) Employee (7) Activities (4) Machine (4) Individual (7) Reporting (3) Building (3) Ability (6) Capacity (6) Learn (6) Intelligence (5) Personnel (5) Right (5) Training (5) Insight (4) Judgment (4) Legal (4) License (4) Worker (4) Labor (3)

Resources are. . . tangible or intangible assets—such as. . . human capital,. . . organizational capital,. . . ﬁnancial capital,. . . physical capital,. . . and relationship capital—. . .

Robustness Tests entrepreneur had a business degree. The To assess potential contingencies affecting results of these analyses, however, showed our results, we performed two sets of robust- that there were no statistically significant dif- ness tests.8 First, to determine whether there ferences with respect to resource definitions were any significant differences between the across venture stages, venture industries, resource definitions of entrepreneurs, we per- and entrepreneurs with/without business formed several analyses of variance comparing degrees. all predictor variables from Table 6 for different Second, we also compared the resource defi- venture stages (see Table 2 for venture stage nitions of specialized entrepreneurship journals categories), ventures’ industries (see Table 2 for (i.e., Entrepreneurship: Theory & Practice, Inter- ventures’ industries), and for whether or not an national Entrepreneurship and Management

8We thank an anonymous reviewer for suggesting these robustness tests.

36 JOURNAL OFKELLERMANNS SMALL BUSINESS ET AL. MANAGEMENT 11 Table 4 Continued

Ownership (81) Firm Creation Value/Success Competitive (145) (195) (65) Advantage (32)

Control (29) Firm (84) Use (38) Value (16) Competitive (15) Own (17) Organization (46) Develop (17) Efficient (12) Advantage (13) Tied (15) Business (4) Implement (17) Effective (9) Superior (4) Semipermanently (8) Collective (4) Manage (17) Improve (8) Possess (7) Operation (4) Input (15) Economic (6) Internal (5) Company (3) Make (14) Potential (5) Enable (13) Strength (5) Production (10) Performance (4) Conceive (9) Combine (7) Result (6) Source (6) Utilize (5) Achieve (4) Deploy (4) Transform (4) Add (3) Draw (3) Generate (3) that are owned by. . . a firm,. . . and that enable the firm to create. . . value/success and a competitive advantage.

Journal, and Journal of Business Venturing) neurial practice and have uncovered a number versus other academic journals and found that of similarities as well as a few key diffe- the only significant difference with respect to rences in researchers’ and entrepreneurs’ resource definitions was that the specialized views. entrepreneurship journals were more likely to Our study therefore improves our under- emphasize “products/services” in their resource standing of the nature of different dimensions of definitions (p < .05). Based on these analyses, resources, each dimension’s primacy to we conclude that our samples’ resource researchers and entrepreneurs, and how those definitions remained robust across venture resources are viewed to shape outcomes, which stages, industries, entrepreneurs’ education, and were an integral part of both researchers’ and types of journal. entrepreneurs’ resource definitions. In particular, the resource dimensions extracted from Discussion research journals show some overlap with those Because the RBV is becoming increasingly offered by the practicing entrepreneurs, high- important within entrepreneurship but was lighting the importance of those resources to developed in another field with a number of organizational functioning and performance. potentially important differences, we set out However, the relative importance that each to better understand whether and to what group placed on each resource dimensions dif- extent the established RBV can be applied to fered, and there were differences regarding the entrepreneurial venture research. In doing so, implications of resources for outcomes. Further, we have investigated how resources as we discuss the key similarities and differences, the fundamental tenets of the RBV are con- outline implications, and describe areas for ceptualized both in research and entrepre- future inquiry.

12 JOURNAL OFKELLERMANNS SMALL BUSINESS ET AL. MANAGEMENT 37 Table 5 Practicing Entrepreneurs’ Resource Deﬁnitions

Assets (115) Human Financial Physical Relationship Capital (205) Capital (48) Capital (94) Capital (149)

Thing (26) People (48) Money (22) Tool (21) Customer (20) Tangible (25) Knowledge (23) Financial (13) Equipment (18) Supply (18) Intangible (22) Information (15) Bank (8) Material (16) Client (17) Asset (14) Skill (14) Cash (5) Technology (16) Market (11) Item (13) Experience (13) Physical (6) Relation (11) Component (8) Idea (11) Computer (5) Available (10) Element (7) Ability (9) Building (4) Network (10) Employee (8) Machine (4) Access (9) Individual (7) Ofﬁce (4) Contact (6) Creativity (6) Community (5) Human (6) Sale (5) Staff (6) Association (4) Talent (6) Contract (4) Capability (5) Outside (4) Concept (5) Partner (4) Education (5) Referral (4) Intellectual (4) Sell (4) Labor (4) Advice (3) Learn (4) Personnel (3) Training (3)

Resources are. . . tangible or intangible assets—such as. . . human capital,. . . ﬁnancial capital,. . . physical capital,. . . and relationship capital—. . .

Similarities the resource–performance relationship (for Both researchers and practicing entrepre- example, see Eddleston, Kellermanns, and neurs conceptualized tangible and intangible Sarathy 2008 and Sirmon, Gove, and Hitt 2008; assets as key components of resources. In addi- see also Sirmon, Hitt, and Ireland 2007 for a tion, both defined resources in terms of human, theoretical treatment). Moreover, the processes financial, physical, and relationship capital and of leveraging resources may be different in the defined these assets as a necessary, but not realm of strategy and entrepreneurship, which sufficient condition for organizational advan- only future research would be able to examine. tages. Both researchers’ and entrepreneurs’ con- Related to that argument, both researchers and ceptualizations further suggest that it may not be entrepreneurs stressed the multidimensionality resources themselves, but the leverage of such of resources (tangible versus intangible; human, resources, that is important for organizational financial physical, and relationship capital; etc.) advantages (indicated by the emphasis both (e.g., Brinckmann and Hoegl 2011; Clarysse, groups place on the dimension “creation”). So Bruneel, and Wright 2011). As these resources far, however, only very few studies have empiri- likely do not only have main but also joint cally investigated how resource use influences outcome effects (Powell and Dent-Micallef

38 JOURNAL OFKELLERMANNS SMALL BUSINESS ET AL. MANAGEMENT 13 Table 5 Continued

Firm (163) Creation Products/Services Value/Success (222) (47) (93)

Business (98) Use (39) Product (26) Success (40) Venture (26) Make (25) Service (21) Proﬁt (18) Company (22) Help (19) Growth (11) Operation (7) Provide (16) Value (10) Firm (5) Run (12) Effective (4) Organization (5) Contribute (10) Improve (4) Source (10) Vital (3) Develop (9) Wealth (3) Necessary (9) Draw (8) Support (8) Create (7) Operate (7) Achieve (6) Add (5) Generate (5) Foundation (4) Manage (4) Obtain (4) Carry (3) Combine (3) Production (3) Rely (3) Result (3) which allow a ﬁrm. . . to create. . . products and/or services. . . in its pursuit of value/success.

1997), future research might want to investigate Key Differences such resource interactions. Capital Resources. In defining resources, An important implication of these similari- entrepreneurs put significantly less emphasis ties is that, broadly speaking, the main tenets of on human, organizational, and physical capital. the RBV are representative of entrepreneurial Although human capital is an important practice. However, we also found germane dif- resource (Crook et al. 2011), entrepreneurs ferences regarding the relative weight that is may simply take human capital, which in our placed on different dimensions of resources sample is most likely their own human capital, and how they enable the creation of outcomes, for granted and thus cognitively attribute less which suggests the necessity of contextual or importance to it. Another potential explanation boundary conditions when leveraging the RBV is that the entrepreneurs view themselves— in entrepreneurship research (Busenitz et al. rightly or wrongly—as the key resources of 2003). We will discuss these differences in the their firms. If this is the case, future research following sections. might want to separate the entrepreneur’s from

14 JOURNAL OFKELLERMANNS SMALL BUSINESS ET AL. MANAGEMENT 39 Table 6 Binary Logistic Regression Analysis

Researcher versus Entrepreneurial Resource Deﬁnitions

Dimension References References in Estimated S.E. Exp in Researcher Entrepreneurial Coefﬁcient (B) Sample (%) Sample (%) (B)

Constant −2.72*** .51 .07 Assets 71 39 1.11*** .34 3.05 Human Capital 83 56 1.12** .39 3.08 Organizational Capital 42 11 1.13** .39 3.10 Financial Capital 27 22 .10 .39 1.11 Physical Capital 54 35 .71* .33 2.04 Relationship Capital 41 44 −.59† .35 .56 Ownership 44 7 2.12*** .43 8.32 Firm 85 69 .60 .39 1.82 Creation 70 69 −.42 .35 .66 Products/Services 9 17 −1.08* .54 .34 Value/Success 37 41 −.71* .34 .49 Competitive Advantage 15 3 3.20*** .67 24.57 χ2 (chi square) 146.86*** -2log-likelihood 271.53 Pseudo-R2 (Nagelkerke) .37 Sample Size 318 Percentage of Deﬁnitions 81.10 Correctly Classiﬁed

†p < .10 *p < .05 **p < .01 ***p < .001 employees’ human capital in the venture and assets and organizational advantages. In a seek to better understand the performance hypercompetitive world, such assets are some- implications of each type of human capital, and times viewed as limiting flexibility, and thus, as how each can be created and better managed detrimental to organizational advantages by the entrepreneur. This suggests that work is (Mosakowski 2002). Another potential explana- needed that blends insights involving high- tion for these findings is that the lack of entre- performance work practices and systems with preneurs’ emphasis on human, organizational, the RBV (cf. Combs et al. 2006) and that helps and physical capital is compensated by their to further understand how human capital is reliance on contracted resources, which we will orchestrated within entrepreneurial ventures to discuss further. enhance performance (Sirmon et al. 2011). Interestingly, entrepreneurs also put less Relationship Capital versus Ownership. The emphasis on organizational and physical traditional RBV—and most of our researcher capital. Such a reduced focus on these dimen- definitions (e.g., Amit and Schoemaker 1993; sion of resources seem to be justified in light of Barney 1991; Wernerfelt 1984)—imply that the recent findings by Newbert (2007) and resources are owned or at least controlled by Crook et al. (2008), which found only partial the firm. Entrepreneurs, in contrast, tend to support for the relationships between physical emphasize the importance of relationship

40 JOURNAL OFKELLERMANNS SMALL BUSINESS ET AL. MANAGEMENT 15 capital—such as contractual relations with sup- and Ruefli 2002), our entrepreneurs’ definitions pliers and partners—which might be able to indicate that very few firms have a competitive substitute, or at least complement, the firm’s advantage or outperform their peers. Although own resource base (Poppo and Zenger 1998). entrepreneurs’ put more emphasis on resource- The idea of relationship capital as part of firms’ enabled value creation and success than resource base also resonates with more recent researchers, entrepreneurs do not conceptual- theoretical developments, such as the relational ize resources in relation to their competition, view (Dyer and Singh 1998) and studies on which is a cornerstone of the idea of perfor- interfirm collaborations (Gulati 1999), which mance advantage (Barney 1991). Success in suggest that “firms essentially use alliances to entrepreneurs’ eyes therefore seems to be gain access to other firms’ valuable resources” defined by sufficient levels of value creation (Das and Teng 2000, p. 33). and success, levels that can be achieved in a A key implication for future inquiry is to state of competitive parity (Hitt et al. 2011). In recognize that entrepreneurial ventures tend to short, entrepreneurs define resource-related be more reliant on relationships (i.e., other organizational advantages in absolute terms, firms) and that these relationships represent and not relative to their competition. An impor- important strategic resources that are exten- tant implication is that future entrepreneurship sions to entrepreneurial ventures. Given this, research might want to account for entrepre- future RBV-based inquiry within entrepreneur- neurs’ objectives and how they factor into the ship should not only investigate the resources (resource-related) management of their ven- that are “controlled by a firm” (Barney 1991, p. tures. For instance, future research might want 110) but recognize that ownership or control of to avoid asking entrepreneurs to compare their resources is not a necessary condition for orga- venture’s performance to others and instead nizational advantages (Wiklund and Shepherd ask them about different metrics that more 2009). In short, for entrepreneurial ventures, directly capture resource-related organizational there is a weaker condition of resource acces- advantages. sibility. Future research might seek to shed More broadly, these results also corroborate light on the extent to which accessibility— the argument that resource outcomes likely which establishes the right to leverage other differ with the stage of the entrepreneurial firms’ resources—allows entrepreneurial ven- venture, and thus the measured outcomes need tures to capture their associated benefits, and to be contextualized with respect to the life the extent to which they are appropriated by cycle of the organization in order to provide the actual resource owners (Dyer and Singh meaningful results (Sirmon et al. 2011). For 1998). These differences in the emphasis of example, though the mere survival of the relationship capital suggests the intersection venture is important for start-up phases, when between the RBV and the relational view the entrepreneurial firm matures, other con- (Dyer and Singh 1998) as a fruitful area for cerns, such as the pursuit of nonfinancial goals future inquiry in entrepreneurship research. (Gómez-Mejía et al. 2007) or profit (Sirmon Though recent research has generated some et al. 2011), might become increasingly impor- insights for larger, more established organiza- tant. In sum, our findings suggest the potential tions (e.g., Schreiner, Kale, and Corsten 2009), of a comprehensive treatment of resource out- our results suggest that relational resources comes, including nonfinancial outcomes, for might matter even more and look different in the development of a RBV in the realm of an entrepreneurship context (see also Foss entrepreneurship (for reviews of performance et al. 2008). outcomes, see Shepherd and Wiklund 2009).

Resource Outcomes—Value, Success, and Com- Resource Outcomes—Products and Ser- petitive Advantage. As mentioned previously, vices. Another important difference is that it came as a surprise to us that not only entrepreneurs emphasize more proximate out- researches’ but also entrepreneurs’ resource comes of resources, such as products and ser- deﬁnitions, without being prompted to do so, vices, in contrast to researchers’ focus on more contained resource-related outcomes as an inte- distant outcomes, such as value creation and gral part of their resource conceptualizations, competitive advantage. In this way, our study albeit with a number of important differences. complements recent research that shows that Consistent with extant research (cf. Wiggins entrepreneurs and scholars operationalized

16 JOURNAL OFKELLERMANNS SMALL BUSINESS ET AL. MANAGEMENT 41 growth very differently (Achtenhagen, Naldi, alizations, our study enables future researchers and Melin 2010). Specifically, our results to build on the field’s consensus definition of suggest that finding potential uses for resources, and thereby represents an important resources—in terms of improving product and step toward conceptual clarity. Moreover, our service offerings—are viewed as essential for empirical comparison of researchers’ and prac- achieving organizational advantages. This is ticing entrepreneurs’ definitions provides consistent with Ray, Barney, and Muhanna important insights into the theoretical versus (2004), who asserted that measures of organi- practical view on what is arguably the most zational advantages (e.g., returns on assets or crucial building block of organizational advan- stock prices) are overaggregated, and impor- tages. In particular, our study subjects the RBV tant intermittent outcomes (e.g., new products in the field of entrepreneurship to an epistemo- created) might be missed when left unmea- logical analysis (Narayanan and Zane 2011), sured. Ray, Barney, and Muhanna’s (2004) and thereby represents a first step in advancing theory suggests that firms could have a both its empirical verification as well as its resource-based advantage in one area (e.g., application in practice. Indeed, understanding marketing) and a disadvantage in another (e.g., how resources are conceptualized can help human resources) and that the advantage and advance both discovery and creation theories disadvantage could potentially offset each (see Alvarez and Barney 2007). other when linked directly to measures of orga- On a fundamental level, entrepreneurs (i.e., nizational advantages. A key implication for individuals) and organizations are not congru- future inquiry is that entrepreneurship ent, particularly if the organization has evolved researchers need to link strategic resources to beyond the individual ownership stage (Bruyat product and service outcomes, and then link and Julien 2000). As such, an RBV for entrepre- those outcomes to dependent variables at the neurship needs to recognize the individual organizational level. Doing so should provide a decision-maker (see also Haynie, Shepherd, more complete picture of resources’ implica- and McMullen 2009; Kemmerer et al. 2011). tions for organizational advantages by allowing Alvarez and Busenitz (2001), for example, have researchers to capture evidence of important noted that the boundary condition of the RBV intermediate outcomes that are vital to the sur- in relationship to entrepreneurship needs to vival and prosperity of entrepreneurial ventures include the individual cognitive abilities of the (Ray, Barney, and Muhanna 2004; Rosenbusch, entrepreneur. Indeed, it is the entrepreneur Brinckmann, and Bausch 2011). who creates benefits by using resources differently (Shane and Venkataraman 2000). This is General Implications reflected in our results, as tangible and intan- The development of the RBV into a strong gible resources are emphasized quite differ- theory—for both the fields of entrepreneurship ently by entrepreneurs and scholars. Similarly, and strategic management—makes it necessary, the RBV in larger, more established organiza- although not sufficient, to have clearly defined tions needs to more fully integrate the organi- variables (cf. Sutton and Staw 1995). Moreover, zational context of the manager and their researchers’ resource definitions should reflect relationship with the owners of the organiza- how practitioners conceptualize resources and tion (e.g., Nag, Hambrick, and Chen 2007). attempt to build their firms around them, that Our study also supports the long-held is, they should attain operational validity assumption that entrepreneurship is driven by (Thomas and Tymon 1982). In her review of the pursuit of opportunities and does not the RBV literature, however, Montgomery require the actual ownership of the resources (1995, p. 257) concludes that the “characteriza- (Stevenson 2000; Stevenson and Jarillo 1990). tion of a perfect form of a resource is very This has implications for the core building useful from both a theoretical and practical blocks of the RBV, as resources that are not standpoint. At the same time, it is important to owned by the entrepreneurs may be easier to know something about the size of the gap imitate and substitute, requiring the entrepre- between the idealized version of a form and neurs to emphasize different processes than what is seen in reality. This is where the managers in larger, more established firms to resource-based literature falls dangerously maintain a resource-based advantage (see also short.” With a content analysis of a comprehen- Stevenson 1985); for example, economies of sive sample of researchers’ resource conceptu- scale and vertical integration are staples of

42 JOURNAL OFKELLERMANNS SMALL BUSINESS ET AL. MANAGEMENT 17 larger, more established firms (Chandler 1977) consisted mainly of individuals who did not but may not be equally feasible for fledgling have a formal business education, who were start-ups. predominantly involved in the start-up of small Irrespective of their differences in emphases, businesses and who were being associated with it is further remarkable that both researchers the FastTrac program. Because of this, future and entrepreneurs included resource outcomes research might investigate other groups of as integral parts of their resource definitions. respondents in order to generalize our findings. Though this may fuel additional criticism of the Not only may the demographics of the entre- RBV’s tendency for circular of even tautological preneurs be important, but also the context in reasoning (e.g., Bromiley and Fleming 2002; which entrepreneurship takes place. The start- Priem and Butler 2001a), the differences our ups our sample was comprised of and our study has uncovered in researchers’ and entre- study focused on can be classified as small preneurs’ definitions with respect to resource businesses, which constitute the majority of outcomes also suggest that the RBV for entre- entrepreneurial start-ups (Carland et al. 1984). preneurial firms needs to develop appropriate Yet, an investigation of the differences between outcomes. Though, theoretically, the RBV serial entrepreneurs or high-growth ventures focuses on competitive advantage, and our (sometimes labeled “gazelles”) compared with study supports this focus, entrepreneurial firms our respondent group may be insightful. Simi- tend to focus more on actual value creation and larly, an investigation of how resources are profit. Ironically, the empirical side of the RBV defined in corporate venturing settings may be has somewhat anticipated this theoretical very useful. A second limitation is that despite development, by more directly focusing on per- our findings about the importance of certain formance rather than on competitive advan- dimensions of resources, recent work on the tages (Crook et al. 2008). RBV suggests the need to look at how Our research also provides valuable guid- resources are managed (Sirmon, Hitt, and ance for entrepreneurs. By eliciting resource Ireland 2007; Sirmon, Gove, and Hitt 2008). definitions from entrepreneurs and contrasting Due to the nature of our research design, we them with academic definitions, we highlight could not investigate how the resource dimen- potential blind spots in entrepreneurs’ mental sions interact or how they are managed to maps. Indeed, recent research has begun to shape performance; however, future research question that entrepreneurs behave according into this area seems warranted. A third limita- to the normative prescriptions advanced by the tion is that not all organizations have a purely RBV and has found, for instance, that entrepre- financial focus. For example, family firms may neurs tend to overemphasize the resource attri- focus on nonfinancial goals in addition to per- butes value and inimitability while largely formance advantages such as economic value neglecting nonsubstitutability and rareness creation (Eddleston, Kellermanns, and Sarathy (Kemmerer et al. 2011). Our study comple- 2008), and these goals can further vary based ments and extends these previous findings by on the life cycle stage the firm is in (Hoy and identifying potential blind spots with respect to Sharma 2010). Thus, there might be other human, organizational, and physical capital important outcomes—other than value resources, resource ownership, as well as com- creation—that entrepreneurial ventures are petitive advantage versus products/services as more concerned with but that we did not resource outcomes. Though these blind spots capture given our research methodology. are not necessarily detrimental per se—for Lastly, we need to comment on the fit of our example, not every resource has to be owned logistic regression model. Though fit indices in to create a competitive advantage (e.g., Dyer logistic regression models are generally difficult and Singh 1998)—entrepreneurs should be to interpret (Pedhazur 1997), it has never been aware of all the attributes and outcomes asso- the intention of our study to maximize the ciated with resources and evaluate them correct classifications but merely to show what accordingly in order to make high-quality similarities and differences between practitio- resource judgments and investment decisions. ner and scholarly dimensions exist. In conclusion, this study focused on decision- Limitations and Conclusion makers’ conceptualizations of resources and A first limitation of our study is that our their implications for organizational advantages. sample of nascent/practicing entrepreneurs Our findings revealed a number of important

18 JOURNAL OFKELLERMANNS SMALL BUSINESS ET AL. MANAGEMENT 43 similarities and differences between researchers Brinckmann, J., and M. Hoegl (2011). “Effects and practicing entrepreneurs and are in line of Initial Teamwork Capability and Initial with the general criticism that research and Relational Capability on the Development of practice operate independent of each other New Technology-Based Firms,” Strategic (Rynes, Bartunek, and Daft 2001). The differ- Entrepreneurship Journal 5(1), 37–57. ences highlight the need for researchers to con- Bromiley, P., and L. Fleming (2002). “The sider the uniqueness of the entrepreneurial Resource-Based View of Strategy: A Behav- venture context when applying the RBV’s theo- iorist Critique,” in The Economics of Choice, retical tenets. Doing so will require researchers Change, and Organizations: Essays in to examine different resource dimensions and Memory of Richard M. Cyert. Eds. M. Augier more intermediate outcomes, such as the cre- and J. G. March. Cheltenham, UK: Elgar Pub- ation of new products and services. If research- lishing, 319–336. ers continue to focus on distant performance Brush, C. G., P. G. Greene, M. M. Hart, and L. advantages at the expense of more intermediate F. Edelman (1997). “Resource Conﬁgurations outcomes, researchers might miss exactly what Over the Life Cycle of Ventures,” in Frontiers is being created. of Entrepreneurship Research. Eds. P. D. Reynolds, W. D. Bygrave, N. M. Carter, P. Davidsson, W. B. Gartner, C. M. Mason, and References P. P. McDougall. Wellesley, MA: Center for Achtenhagen, L., L. Naldi, and L. Melin (2010). Entrepreneurial Studies, Babson College, “‘Business Growth’—Do Practitioners and 315–329. Scholars Really Talk About the Same Bruyat, C., and P.-A. Julien (2000). “Deﬁning Thing?,” Entrepreneurship Theory and Prac- the Field of Research in Entrepreneurship,” tice 34(2), 289–316. Journal of Business Venturing 16, 165–180. Alvarez, S. A., and J. B. Barney (2007). “Discov- Busenitz, L. W., G. P. West, D. Shepherd, T. ery and Creation: Alternative Theories of Nelson, G. N. Chandler, and A. Zacharakis Entrepreneurial Action,” Strategic Entrepre- (2003). “Entrepreneurship Research in Emer- neurship Journal 1(1–2), 11–26. gence: Past Trends and Future Directions,” Alvarez, S. A., and L. W. Busenitz (2001). Journal of Management 29(3), 285–308. “The Entrepreneurship of Resource-Based Carayannopoulos, S. (2009). “How Technology- Theory,” Journal of Management 27, 755– Based New Firms Leverage Newness and 775. Smallness to Commercialize Disruptive Amit, R., and P. Schoemaker (1993). “Strategic Technologies,” Entrepreneurship Theory Assets and Organizational Rent,” Strategic and Practice 33(2), 419–438. Management Journal 14, 33–46. Carland, J. W., F. Hoy, W. R. Boulton, and J. C. Barney, J. (1991). “Firm Resources and Sus- Carland (1984). “Differentiating Entrepre- tained Competitive Advantage,” Journal of neurs from Small Business Owners: A Con- Management 17(1), 99–120. ceptualization,” Academy of Management Barney, J., M. Wright, and D. J. Ketchen (2001). Review 9(2), 354–359. “The Resource-Based View of the Firm: Ten Castanias, R. P., and C. E. Helfat (1991). “Mana- Years After 1991,” Journal of Management gerial Resources and Rents,” Journal of 27(6), 625–641. Management 17(1), 155–171. Barney, J. B. (2001). “Is the Resource-Based Chandler, A. D. (1977). The Visible Hand: The ‘View’ a Useful Perspective for Strategic Managerial Revolution in American Busi- Management Research? Yes,” Academy of ness. Cambridge, MA: Belknap Press. Management Review 26(1), 41–57. Chandler, G. N., and S. H. Hanks (1994). Barney, J. B., and A. M. Arikan (2001). “The “Market Attractiveness, Resource-Based Resource-Based View: Origins and Implica- Capabilities, Venture Strategies, and Venture tions,” in The Blackwell Handbook of Strate- Performance,” Journal of Business Ventur- gic Management. Eds. M. A. Hitt, R. E. ing 9, 331–349. Freeman, and J. S. Harrison. Oxford, UK: Chrisman, J., and F. W. Kellermanns (forthcom- Blackwell Publisher Ltd, 124–188. ing). “Entrepreneur,” in Blackwell Encyclo- Bettis, R. A. (1991). “Strategic Management and pedia of Entrepreneurship Research. Eds. M. the Straightjacket: An Editorial Essay,” Orga- Morris and D. Kuratko. Oxford: Blackwell nization Science 2(3), 315–319. Publishing.

44 JOURNAL OFKELLERMANNS SMALL BUSINESS ET AL. MANAGEMENT 19 Chrisman, J. J., A. Bauerschmidt, and C. W. De Carolis, D. M., Y. Yang, D. L. Deeds, and E. Hofer (1998). “The Determinants of New Nelling (2009). “Weathering the Storm: The Venture Performance: An Extended Model,” Benefit of Resources to High-Technology Entrepreneurship Theory and Practice 23(1), Ventures Navigating Adverse Events,” Stra- 5–29. tegic Entrepreneurship Journal 3(2), 147– Chrisman, J. J., J. H. Chua, and F. W. 160. Kellermanns (2009). “Priorities, Resource Dyer, J. H., and H. Singh (1998). “The Rela- Stocks, and Performance in Family and Non- tional View: Cooperative Strategies and Family Firms,” Entrepreneurship Theory and Sources of Inter-Organizational Competitive Practice 33(3), 739–760. Advantage,” Academy of Management Chua, J. H., J. J. Chrisman, F. W. Kellermanns, Review 23(4), 660–679. and Z. Wu (2011). “Agency Problems of Bor- Eddleston, K., F. W. Kellermanns, and R. rowed Social and Financial Capital: Resolu- Sarathy (2008). “Resource Configuration in tion in New Ventures Through Family Family Firms: Linking Resources, Strategic Involvement,” Journal of Business Venturing Planning and Technological Opportunities 26(4), 472–488. to Performance,” Journal of Management Clarysse, B., J. Bruneel, and M. Wright (2011). Studies 45(1), 26–50. “Explaining Growth Paths of Young Fischer, E., and R. Reuber (2007). “The Good, Technology-Based Firms: Structuring the Bad, and the Unfamiliar: The Challenges Resource Portfolios in Different Competitive of Reputation Formation Facing New Firms,” Environments,” Strategic Entrepreneurship Entrepreneurship Theory and Practice Journal 5(2), 137–157. 31(1), 53–75. Cohen, J. (1960). “A Coefficient of Agreement Forster, N. (1999). “The Analysis of Company for Nominal Scales,” Educational and Documentation,” in Qualitative Methods in Psychological Measurement 20(1), 37– Organizational Research: A Practical Guide. 46. Eds. G. E. Symon and C. E. Cassell. London, Combs, J. G., Y. Liu, A. Hall, and D. J. Ketchen UK: Sage, 147–166. (2006). “How Much Do High-Performance Foss, N. J., P. G. Klein, Y. Y. Kor, and J. T. Work Practices Matter? A Meta-Analysis of Mahoney (2008). “Entrepreneurship, Subjec- Their Effects on Organizational Perfor- tivism, and the Resource-Based View: mance,” Personnel Psychology 59, 501– Toward a New Synthesis,” Strategic Entre- 528. preneurship Journal 2(1), 73–94. Conner, K. R. (1991). “A Historical Comparison Gómez-Mejía, L. R., K. T. Haynes, M. of Resource-Based Theory and Five Schools Núñez-Nickel, K. J. L. Jacobson, and H. of Thought Within Industrial Organization Moyano-Fuentes (2007). “Socioemotional Economics: Do We Have a New Theory of Wealth and Business Risk in Family- the Firm?,” Journal of Management 17(1), Controlled Firms: Evidence from Spanish 121–154. Olive Oil Mills,” Administrative Science Crook, T. R., D. J. Ketchen, J. G. Combs, and S. Quarterly 52(1), 106–137. Y. Todd (2008). “Strategic Resources and Grant, R. M. (1991). “The Resource-Based Performance: A Meta-Analysis,” Strategic Theory of Competitive Advantage: Implica- Management Journal 29(11), 1141–1154. tions for Strategy Formulation,” California Crook, T. R., S. Y. Todd, J. G. Combs, D. J. Management Review 33(3), 114–135. Woehr, and D. J. Ketchen (2011). “Does Green, K. E. (1996). “Sociodemographic Factors Human Capital Matter? A Meta-Analysis of and Mail Survey Response,” Psychology and the Relationship Between Human Capital Marketing 13(2), 171–184. and Firm Performance,” The Journal of Greene, P. G., C. G. Brush, and T. E. Brown Applied Psychology 96(3), 443–456. (1997). “Resources in Small Firms: An Das, T. K., and B.-S. Teng (2000). “A Resource- Exploratory Study,” Journal of Small Busi- Based Theory of Strategic Alliances,” ness Strategy 8(2), 25–38. Journal of Management 26(1), 31–61. Gulati, R. (1999). “Network Location and Learn- David, R. J., and S.-K. Han (2004). “A System- ing: The Influence of Network Resources atic Assessment of the Empirical Support for and Firm Capabilities on Alliance Forma- Transaction Cost Economics,” Strategic Man- tion,” Strategic Management Journal 20, agement Journal 25, 39–58. 397–420.

20 JOURNAL OFKELLERMANNS SMALL BUSINESS ET AL. MANAGEMENT 45 Hall, R. (1992). “The Strategic Analysis of Intan- Research Using Key Informants,” Academy gible Resources,” Strategic Management of Management Journal 36(6), 1633–1651. Journal 13, 135–144. Lichtenstein, B. M. B., and C. G. Brush (2001). Haynie, J. M., D. A. Shepherd, and J. S. “How Do ‘Resource Bundles’ Develop and McMullen (2009). “An Opportunity for Me? Change in New ventures? A Dynamic Model The Role of Resources in Opportunity Evalu- and Longitudinal Exploration,” Entrepre- ation Decisions,” Journal of Management neurship Theory and Practice 25(3), 37–58. Studies 46(3), 337–361. Miller, D. J. (2004). “Firms’ Technological Hitt, M. A., R. D. Ireland, S. M. Camp, and D. L. Resources and the Performance Effects of Sexton (2002). “Strategic Entrepreneurship: Diversiﬁcation: A Longitudinal Study,” Stra- Integrating Entrepreneurial and Strategic tegic Management Journal 25, 1097–1119. Management Perspectives,” in Strategic Montgomery, C. A. (1995). “Of Diamonds and Entrepreneurship: Creating a New Mindset. Rust: A New Look at Resources,” in Eds. M. A. Hitt, R. D. Ireland, S. M. Camp, Resource-Based and Evolutionary Theories and D. L. Sexton. Malden, MA: Blackwell, of the Firm: Towards a Synthesis. Ed. C. A. 1–16. Montgomery. Norwell, MA: Kliver, 251–267. Hitt, M. A., R. D. Ireland, D. G. Sirmon, and Mosakowski, E. (2002). “Overcoming Resource C. A. Trahms (2011). “Strategic Entre- Disadvantages in Entrepreneurial Firms: preneurship: Creating Value for Individuals, When Less Is More,” in Strategic Entrepre- Organizations, and Society,” Academy neurship: Creating a New Integrated of Management Perspectives 25(2), 57–75. Mindset. Eds. M. A. Hitt, R. D. Ireland, S. M. Hoopes, D. G., T. L. Madsen, and G. Walker Camp, and D. L. Sexton. Oxford, UK: Black- (2003). “Guest Editor’s Introduction to the well Publishing, 106–126. Special Issue: Why Is There a Resource- Nag, R., D. C. Hambrick, and M.-J. Chen (2007). Based View,” Strategic Management Journal “What Is Strategic Management, Really? 24(10), 889–902. Inductive Derivation of a Consensus Deﬁni- Hoy, F., and P. Sharma (2010). Entrepreneurial tion of the Field,” Strategic Management Family Firms. Upper Saddle River, NJ: Pren- Journal 28, 935–955. tice Hall. Narayanan, V. K., and L. Zane (2011). “Current Ireland, R. D., J. W. Webb, and J. E. Coombs Theoretical Debates in Management (2005). “Theory and Methodology in Entre- Research: Epistemological Analysis in Strate- preneurship Research,” in Research Method- gic Management,” in Challenges and Contro- ology in Strategy and Management. Eds. D. versies in Management Research. Eds. C. J. Ketchen and D. D. Bergh. Bingley, UK: Cassell and B. Lee. New York: Routledge. Emerald Group, 111–141. Nelson, K. M., S. Nadkarni, V. K. Narayanan, Kanuk, L., and C. Berenson (1975). “Mail and M. Ghods (2000). “Understanding Soft- Surveys and Response Rate: A Literature ware Operations Support Expertise: A Review,” Journal of Marketing Research 22, Revealed Causal Mapping Approach,” MIS 440–453. Quarterly 24(3), 457–507. Kemmerer, B., J. Walter, F. W. Kellermanns, Neuendorf, K. A. (2002). The Content Analysis and V. K. Narayanan (2011). “A Judgment- Guidebook. Thousand Oaks, CA: Sage. Analysis Perspective on Entrepreneurs’ Newbert, S. L. (2007). “Empirical Research on Resource Evaluations,” Journal of Business the Resource-Based View of the Firm: An Research 65, 1102–1108. Assessment and suggestions for Future King, N. (1998). “Template Analysis,” in Quali- Research,” Strategic Management Journal tative Methods and Analysis in Organiza- 28, 121–146. tional Research: A Practical Guide. Eds. G. Pedhazur, E. J. (1997). Multiple Regression in Symon and C. Cassell. London: Sage, 118– Behavioral Research. Explanations and Pre- 134. dictions. Fort Worth, TX: Harcourt Brace. Kraaijenbrink, J., J.-C. Spender, and A. J. Groen Penrose, E. T. (1959). The Theory of the Growth (2010). “The Resource-Based View: A of The Firm. Oxford, GB: Blackwell. Review and Assessment of Its Critiques,” Podsakoff, P. M., S. B. MacKenzie, J.-Y. Lee, Journal of Management 36(1), 349–372. and N. P. Podsakoff (2003). “Common Kumar, N., L. W. Stern, and J. C. Anderson Method Biases in Behavioral Research: A (1993). “Conducting Interorganizational Critical Review of the Literature and Recom-

46 JOURNAL OFKELLERMANNS SMALL BUSINESS ET AL. MANAGEMENT 21 mended Remedies,” The Journal of Applied Sharma, P., and J. J. Chrisman (1999). “Toward Psychology 88(5), 879–903. a Reconciliation of the Definitional Issues in Podsakoff, P. M., S. B. Mackenzie, D. G. the Field of Corporate Entrepreneurship,” Bachrach, and N. P. Podsakoff (2005). “The Entrepreneurship: Theory & Practice 23(3), Influence of Management Journals in the 11–27. 1980s and 1990s,” Strategic Management Shepherd, D., and J. Wiklund (2009). “Are We Journal 26, 473–488. Comparing Apples with Apples or Apples Poppo, L., and T. Zenger (1998). “Testing with Oranges? Appropriateness of Knowl- Alternative Theories of the Firm: Trans- edge Accumulation Across Growth Studies,” action Cost, Knowledge-Based, and Entrepreneurship Theory and Practice 33(1), Measurement Explanations for Make-or- 105–123. Buy Decisions in Information Services,” Siqueira, A., and G. D. Bruton (2010). Strategic Management Journal 19, 853– “High-Technology Entrepreneurship in 877. Emerging Economies: Firm informality Powell, T. C., and A. Dent-Micallef (1997). and Contextualization of Resource-Based “Information Technology as Competitive Theory,” IEEE Transactions on Engineering Advantage: The Role of Human, Business, Management 57(1), 39–50. and Technology Resources,” Strategic Man- Sirmon, D. G., M. A. Hitt, and R. D. Ireland agement Journal 18, 375–405. (2007). “Managing Firm Resources in Priem, R. L., and J. E. Butler (2001a). “Is the Dynamic Environments to Create Value: Resource-Based ‘View’ a Useful Perspective Looking Inside the Black Box,” Academy of for Strategic Management Research?,” Management Review 32(1), 273–292. Academy of Management Review 26(1), Sirmon, D. G., S. Gove, and M. A. Hitt (2008). 22–40. “Resource Management in Dyadic Competi- ——— (2001b). “Tautology in the Resource- tive Rivalry: The Effects of Resource Bun- Based View and the Implications of Exter- dling and Deployment,” Academy of nally Determined Resource Value: Further Management Journal 31(5), 919–935. Comments,” Academy of Management Sirmon, D. G., M. A. Hitt, R. Duane Ireland, and Review 26(1), 57–65. B. A. Gilbert (2011). “Resource Orchestra- Ray, G., J. B. Barney, and W. A. Muhanna tion to Create Competitive Advantage: (2004). “Capabilities, Business Processes and Breadth, Depth, and Life Cycle Effects,” Competitive Advantage: Choosing the Journal of Management 37(5), 1390–1412. Dependent Variable in Empirical Tests of the Stevenson, H. H. (1976). “Defining Corporate Resource-Based View,” Strategic Manage- Strengths and Weaknesses,” Sloan Manage- ment Journal 25, 23–37. ment Review 17(Spring), 51–68. Rosenbusch, N., J. Brinckmann, and A. Bausch ——— (1983). A Perspective on Entrepreneur- (2011). “Is Innovation Always Beneficial? A ship. Boston, MA: Harvard Business School. Meta-Analysis of the Relationship Between ——— (1984). Resource Assessment: Defining Innovation and Performance in SMEs,” Corporate Strengths and Weaknesses. Journal of Business Venturing 26(4), 441– Boston, MA: Harvard Business School Press. 457. ——— (1985). “The Heart of Entrepreneur- Rynes, S. L., J. M. Bartunek, and R. L. Daft ship,” Harvard Business Review March– (2001). “Across the Great Divide: Knowledge April, 85–94. Creation and Transfer Between Practitioners ——— (2000). “Why Entrepreneurship has and Academics,” Academy of Management Won!,” Coleman White Paper (USASBE Journal 44(2), 340–355. Plenary Address). Schreiner, M., P. Kale, and D. Corsten (2009). Stevenson, H. H., and J. C. Jarillo (1990). “A “What Really Is Alliance Management Capa- Paradigm of Entrepreneurship: Entrepre- bility and How Does It Impact Alliance Out- neurial Management,” Strategic Manage- comes and Success?,” Strategic Management ment Journal 11, 11–27. Journal 30, 1395–1419. Sullivan, D., and M. Marvel (2011). “How Entre- Shane, S., and S. Venkataraman (2000). “The preneurs’ Knowledge and Network Ties Promise of Entrepreneurship as a Field of Relate to the Number of Employees in new Research,” Academy of Management Review SMEs,” Journal of Small Business Manage- 25(1), 217–226. ment 49(2), 185–206.

22 JOURNAL OFKELLERMANNS SMALL BUSINESS ET AL. MANAGEMENT 47 Sutton, R. I., and B. M. Staw (1995). “What ——— (1995). “The Resource-Based View of Theory Is Not,” Administrative Science the Firm: Ten Years After,” Strategic Man- Quarterly 40, 371–384. agement Journal 16, 171–174. Thomas, K. W., and W. G. Tymon (1982). “Nec- Whetten, D. A. (1989). “What Constitutes a essary Properties of Relevant Research: Theoretical Contribution?,” Academy of Lessons from Recent Criticisms of the Orga- Management Review 14, 490–495. nizational Sciences,” Academy of Manage- Wiggins, R. A., and T. W. Rueﬂi (2002). “Sus- ment Review 7, 345–352. tained Competitive Advantage: Temporal Ucbasaran, D., P. Westhead, and M. Wright Dynamics and the Incidence and Persistence (2009). “The Extent and Nature of Opportu- of Superior Economic Performance,” Orga- nity Identiﬁcation by Experienced Entrepre- nization Science 13(1), 82–105. neurs,” Journal of Business Venturing 24, Wiklund, J., and D. Shepherd (2003). “Aspiring 99–115. for, and Achieving Growth: The Moderating Unger, J. M., A. Rauch, M. Frese, and N. Role of Resources and Opportunities,” Rosenbusch (2011). “Human Capital and Journal of Management Studies 40(8), 1919– Entrepreneurial Success: A Meta-Analytical 1941. Review,” Journal of Business Venturing Wiklund, J., and D. A. Shepherd (2009). “The 26(3), 341–358. Effectiveness of Alliances and Acquisitions: Watt, R. J. C. (2004). “Concordance (3.2 ed) The Role of Resource Combination Activi- (software),” http://www.concordancesoft ties,” Entrepreneurship Theory and Practice ware.co.uk/ [accessed on 1 October 2012]. 33(1), 193–212. Welter, F. (2001). “Contextualizing Zhao, H., S. E. Seibert, and G. T. Lumpkin Entrepreneurship-Conceptual Challenges (2009). “The Relationship of Personality to and Ways Forward,” Entrepreneurship Entrepreneurial Intentions and Performance: Theory & Practice 35(1), 165–184. A Meta-Analytic Review,” Journal of Man- Wernerfelt, B. (1984). “A Resource-Based View agement 36(2), 381–404. of the Firm,” Strategic Management Journal 5(2), 171–180.

48 JOURNAL OFKELLERMANNS SMALL BUSINESS ET AL. MANAGEMENT 23 Journal of Small Business Business Management Management 2016 2014 54(1), ••(••), pp. pp. 49–66 ••–•• doi: 10.1111/jsbm.12129 About Gender Differences and the Social Environment in the Development of Entrepreneurial Intentions by Francisco J. Santos, Muhammad Azam Roomi, and Francisco Liñán

This study analyzes the interplay between gender differences and the social environment in the formation of entrepreneurial intentions. Data were obtained from two different European regions. The results show that the formation of entrepreneurial intentions is similar for men and women. At the same time, men consistently exhibit more favorable intentions than women do. Nevertheless, the perception of the social legitimation of entrepreneurship only serves to reinforce male entrepreneurial intentions, and not those of women. This holds for both regions and probably is a consequence of women feeling entrepreneurship to not be an acceptable career option for them. The implications of these results are discussed.

interpretation derived from the alternative- Introduction gendered perspectives (Bird and Brush 2002). The existence of a gap between men and As a result, perceptions such as self-efficacy women in entrepreneurship has long been differ by gender (Kickul et al. 2008). Women are acknowledged, and it is attracting increasing also found to perceive fewer opportunities and academic attention (Hughes et al. 2012). Thus, to identify higher financial barriers than their the proportion of any country’s adult female male counterparts (Langowitz and Minniti 2007; population participating in entrepreneurship Minniti and Nardone 2007). is lower than that of men (Hindle, Klyver, and Second, only a small proportion of research Jennings 2009). However, more research is presently considering the socioeconomic is needed to fully explain the gender gap in context of female entrepreneurship and, in this entrepreneurial activity, at least in two respects: sense, comparative works from different coun- individual perceptions and environmental influ- tries and regions are recommended (Ahl 2006). ences (Neergaard, Shaw, and Carter 2005). Cultural values and beliefs play a role in First, individual cognitions and self- shaping the institutions of a country (Verheul, perceptions may help explain whether (and van Stel, and Thurik 2006). Hence, they may why) women interpret the reality around influence the decision to become self-employed them differently from the way men do (de (Mueller and Thomas 2001). Bruin, Brush, and Welter 2007). In this sense, This paper aims to fill these two gaps in the some authors stress the differences in world literature. To do so, perceptions of both males

Francisco J. Santos is associate professor in the Department Economia Aplicada I at the University of Seville. Muhammad Azam Roomi is senior lecturer in the Bettany Centre for Entrepreneurship at Cranfield at the Cranfield School of Management. Francisco Liñán is associate professor in the Department Economia Aplicada I at the University of Seville. Address correspondence to: Francisco Liñán, University of Seville, Economia Aplicada I, Seville, Spain E-41018. E-mail: fl[email protected].

SANTOS, ROOMI, AND LI LIÑÁNN~AN 49 1 and females from two different European not—of the individual’s firm-creation decision regions (southern Britain and southern Spain) by the people in his or her closer environment. will be analyzed. Specifically, this research will This social-norm element captures the influence focus on attitudes, capacities, and intentions of the society around the individual (Ajzen toward business start-ups. This will allow the 1991). consideration of new ideas about gender- The literature has found strong empirical specific perceptions of entrepreneurship (Bird evidence supporting the TPB, especially in the and Brush 2002). It will also help to explain case of the influence of PA and PBC on inten- why the level of entrepreneurial intentions of tions (Armitage and Conner 2001). Neverthe- women is found to be lower than that of men less, some studies have found the direct (Hindle, Klyver, and Jennings 2009). influence of perceived SN on entrepreneurial To achieve these objectives, a cognitive intention to be quite weak (Autio et al. 2001; approach has been followed based on two ele- Krueger, Reilly, and Carsrud 2000). This has ments. First, Ajzen’s (1991) well-known theory led some authors to exclude SN from the analy- of planned behavior (TPB) is used to explain sis (Fitzsimmons and Douglas 2011). Other entrepreneurial intentions (Krueger and authors have, instead, suggested SN to be a Carsrud 1993). Second, the role of both the way to “channel” the influence of the perceived micro- and the macrosocial environments on closer and social environments on personal perceptions and intentions of men and women perceptions (Ferreira et al. 2012; Liñán, is considered (Busenitz and Lau 1996; Etzioni Urbano, and Guerrero 2011), thus mediating 1987). this relationship. SN is, then, an anticipation of the expected rewards or sanctions by people in Literature Review the individuals’ closer environment if the and Hypotheses behavior was performed (Meek, Pacheco, and The TPB York 2010). Entrepreneurial intention has long been recognized as a key precursor of new venture Gender Differences in creation (Bird 1988). The TPB (Ajzen 1991) and Entrepreneurial Intentions the entrepreneurial event theory (Shapero As mentioned previously, women are said to 1982) have received special attention (Hindle, present some weaknesses in the context of Klyver, and Jennings 2009). The latter explains entrepreneurial activity in comparison with intention as a function of desirability, feasibil- men. Some of these weaknesses are fewer finan- ity, and propensity to act (Shapero 1982). The cial, human, and network resources TPB, in turn, was proposed to explain planned (Becker-Blease and Sohl 2007; Brush et al. 2002; behavior in general (Ajzen 1991) and has fre- Carter and Allen 1997; Fabowale, Orser, and quently been applied to entrepreneurship Riding 1995; Marlow and Patton 2005; (Kolvereid 1996; Krueger and Carsrud 1993; Smith-Hunter 2006), or less management expe- Liñán and Chen 2009). In practice, both models rience (Brush et al. 2004; Loscocco et al. 1991). are considered as highly compatible (Krueger, Nevertheless, once some of these variables— Reilly, and Carsrud 2000) and as having such as the starting capital or hours worked— substantial commonalities (Fitzsimmons and are statistically controlled for, researchers Douglas 2011). have found more similarities than differences The TPB considers entrepreneurial intentions between male and female businesses to be directly influenced by three perceptions (Neergaard, Shaw, and Carter 2005; Watson (Ajzen 1991; Kolvereid 1996; Krueger and 2002). Carsrud 1993; Krueger, Reilly, and Carsrud Traits or demographic variables, such as 2000). According to Hindle, Klyver, and risk-taking propensity, have been used to Jennings (2009), entrepreneurial personal atti- explain the specificities of female entrepreneur- tude (PA) is the degree of attraction toward ship (Masters and Meier 1988; Sexton and becoming an entrepreneur (very similar to desir- Bowman-Upton 1990). However, this approach ability), whereas entrepreneurial perceived has been criticized (Krueger, Reilly, and behavioral control (PBC) refers to the ability to Carsrud 2000; Robinson et al. 1991). Recently, develop the entrepreneurial behavior (very cognitive elements have been proposed to similar to feasibility). Finally, perceived subjec- explain the gender gap in entrepreneurial activ- tive norm (SN) refers to the approval—or ity (Arenius and Minniti 2005; Fernández,

250 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Liñán, and Santos 2009; Krueger, Reilly, and Social capital includes both strong ties Carsrud 2000). In this sense, feminist/social (among members of a family or ethnic group) feminist theory argues that research so far has and weak ties (Woolcock and Narayan 2000). been carried out following a masculine para- Cognitive social capital refers to types of under- digm (Bird and Brush 2002). standings that develop among individuals Both at the aggregate and the individual depending on a shared meaning of language, levels of analysis, research has shown that codes, and culture (Farr-Wharton and Brunetto there is a gender gap in entrepreneurial inten- 2007; Naphiet and Ghoshal 1998). From a cog- tions and perceptions, regardless of the level of nitive perspective, both types of social capital economic development (Langowitz and Minniti (strong and weak ties) play a different and 2007; McGee et al. 2009; Minniti and Nardone complementary role in transmitting values and 2007; Verheul, van Stel, and Thurik 2006). Simi- ideas that will influence perceptions and inten- larly, there are gender differences in the tion (De Carolis and Saparito 2006; Simon, manner in which self-beliefs and attitudes Houghton, and Aquino 2000). about entrepreneurship are processed and As Fayolle, Basso, and Bouchard (2010) point developed (Kickul et al. 2008). out, it is important to consider the interplay One important perception influencing entre- between different levels of social influence in preneurial intentions is entrepreneurial self- explaining the entrepreneurial orientation. The efficacy (ESE). This refers to the belief that one social influence on entrepreneurial attitudes is capable of performing an activity (Bandura and behaviors is exerted at both the macro- 1997). Specifically, women were found to have and micro-levels (Morris and Schindehutte a lower ESE and lower entrepreneurial inten- 2005). tions than men have (Mueller and Dato-On Thus, the microsocial or closer environment 2008; Wilson, Kickul, and Marlino 2007). Nev- derives from links with family, friends, or ertheless, it seems the effect of self-efficacy on acquaintances (Uphoff 2000). Participation in intentions may be stronger for women (Kickul this closer-environment network will provide, et al. 2008). among other things, advice, support, and legiti- Likewise, some research has found that macy (Hindle, Klyver, and Jennings 2009). In women perceive fewer opportunities, a higher this sense, closer valuation (CV) refers to the fear of failure, and higher financial barriers way individuals perceive the entrepreneurial than their male counterparts (Langowitz and activity to be valued in their closer environment Minniti 2007; Minniti and Nardone 2007). Other (family, friends, and ethnic group). This influ- studies, on the other hand, argue that these ence received from the closer environment differences are in the contrasting way in which values (CV) contributes to the generation of women and men recognize those opportunities more favorable perceptions toward start-up (DeTienne and Chandler 2007). At least part of (Cooper and Dunkelberg 1987; Scherer, these differences could be due to the dissimilar Brodzinsky, and Wiebe 1991). Therefore, the effect of environmental influences on the indi- value assigned to entrepreneurship in this vidual perceptions of men and women (Byrne closer environment (CV) is likely to promote a and Fayolle 2010). In this sense, a distinction more positive perception of personal support if between biological sex (man/woman) and the individual decides to start a venture (SN) socialized perspective (masculine/feminine) is (Neergaard, Shaw, and Carter 2005). At the advocated (Bird and Brush 2002). same time, these perceived valuations may increase self-confidence in the ability to suc- The Influence of the Social Environment cessfully start a venture (entrepreneurial PBC) Social cognitive theory (Bandura 2001) sug- and the desirability toward the entrepreneurial gests that the social environment around indi- career (entrepreneurial PA) (Rimal and Real viduals plays an important role in shaping their 2003). cognition and, ultimately, behavior (De Carolis The macrosocial environment, however, is and Saparito 2006). The social status of entre- made up of the social values and culture shared preneurship (Begley and Tan 2001) or it being by the society (Thornton, Ribeiro-Soriano, and a respected career path (Busenitz, Gómez, and Urbano 2011). The value society puts on entre- Spencer 2000) will raise the individuals’ interest preneurship will manifest itself in the form of a in entrepreneurship and new venture creation higher social status of entrepreneurship or a (Morris and Schindehutte 2005). greater admiration for entrepreneurs (Begley

SANTOS, ROOMI, AND AND LI LIÑÁNN~AN 51 3 and Tan 2001; Busenitz, Gómez, and Spencer tries and regions with very low income levels 2000). Thus, social valuation (SV) refers to the and high female unemployment, it may also be way individuals perceive the entrepreneurial true that women tend to undertake very mar- activity is valued in society, as a consequence ginal subsistence activities, thus showing of macrosocial values and culture (Liñán, apparently higher start-up rates than those of Urbano, and Guerrero 2011). The underlying men (García-Cabrera and García-Soto 2008; system of values pertaining to a specific group Verheul, van Stel, and Thurik 2006). or society shapes the development of person- It may be argued, then, that the environment ality perceptions (Davidsson and Wiklund exerts a different influence depending on the 1997; Zahra, Jennings, and Kuratko 1999), level of economic and social development modeling normative (SN), affective (PA) and (Iakovleva, Kolvereid, and Stephan 2011; Liñán also ability (PBC) perceptions toward the entre- and Chen 2009). preneurial activity (Thomas and Mueller 2000). It is expected, therefore, that potential entre- Research Model and Hypotheses preneurs will be aware of what the SV of entre- Figure 1 presents the final research model to preneurship is and their intentions will be be tested, considering the four elements of the shaped accordingly. This influence comes from TPB model along with the two additional social social legitimation and the promotion of certain perceptions defined previously: CV and SV. As positive values regarding firm creation may be seen, it specifically hypothesizes that (Busenitz and Lau 1996; Davidsson and CV and SV influence entrepreneurial PA, entre- Wiklund 1997; Etzioni 1987). preneurial PBC, and SN (Liñán, Urbano, and From a gender perspective, some studies Guerrero 2011). find women perceive their task environment as Following the research model depicted in less suitable for entrepreneurial activity (Zhao, Figure 1, and taking into account the gender Siebert, and Hills 2005). The normative support gap in entrepreneurial intentions and percep- from the females’ closer environment seems to tions found by the recent entrepreneurship lit- be embedded in overall attitudes about entre- erature (Kickul et al. 2008; Langowitz and preneurship and gender equality. Only when Minniti 2007; Minniti and Nardone 2007; the normative support is strong is the influence Wilson, Kickul, and Marlino 2007), the follow- on the start-up rate of women positive ing hypotheses are derived: (Baughn, Chua, and Neupert 2006). Similarly, it has also been pointed out that H1: Across different regions, women exhibit, cognitive differences in entrepreneurial behav- when compared with men: iors are explained by gender stereotypes and socially conditioned perceptions of what it H1a: lower intentions of becoming means to be masculine or feminine (Bird and entrepreneurs. Brush 2002; Byrne and Fayolle 2010; Gupta et al. 2009; Mueller and Dato-On 2008). In H1b: lower entrepreneurial PBC. general, the masculine stereotype based on aggressiveness, competitiveness, and risk- H1c: lower entrepreneurial PA. taking behavior has been assigned to men, and it has been considered very important for entre- H1d: lower subjective norm (SN) of becoming preneurship and the economic success of entrepreneurs. nations (Bird and Brush 2002; Byrne and Fayolle 2010; Kickul et al. 2008). H1e: lower closer valuation (CV) of becoming In the case of less developed countries, entrepreneurs. research has also focused on the influence of the national or regional environment on female H1f: lower social valuation (SV) of becoming entrepreneurship. In these environments, tradi- entrepreneurs. tional attitudes and values transmitted through family and social links could specifically be Nevertheless, despite these expected differ- behind the lower entrepreneurial activity of ences, the influence of basic perceptions on women with respect to men (Bertaux and intention-model elements should be similar for Crable 2007; Roomi and Parrot 2008; Wells, both genders and in different contexts (Arenius Pfantz, and Byrne 2003). Conversely, in coun- and Minniti 2005; Minniti and Nardone 2007).

452 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Figure 1 Research Model

Hence, based on the theory, the following (Eddleston and Powell 2008; Gupta et al. 2009; hypotheses are proposed: Kickul et al. 2008; Mueller and Dato-On 2008; Watson and Newby 2005; Zhao, Siebert, and H2: The following relations hold for both Hills 2005). This influence tends to be weaker genders across different regions: for women (Matthews and Moser 1996; Verheul, Uhlaner, and Thurik 2005; Watson and Newby H2a: Entrepreneurial PA has a positive impact 2005). These differences in environmental influ- on entrepreneurial intentions. ences have been found to be stronger in less developed regions or countries (Bertaux and H2b: Entrepreneurial PBC has a positive Crable 2007; Roomi and Parrot 2008; Wells, impact on entrepreneurial intentions. Pfantz, and Byrne 2003). This therefore leads to the following hypotheses being proposed: H2c: SN has a positive impact on entrepreneurial intentions. H3: There are differences in the following relationships depending on gender (stronger for H2d: SN has a positive impact on entrepreneur- men) and on the region: ial PA. H3a: Closer valuation has a positive influence H2e: SN has a positive impact on entrepreneur- on entrepreneurial PA. ial PBC. H3b: Closer valuation has a positive influence However, the literature also points out that on entrepreneurial PBC. socio-environmental elements exert a different influence on male and female perceptions H3c: Closer valuation has a positive influence and intentions regarding entrepreneurship on SN.

SANTOS, ROOMI, AND AND LI LIÑÁNN~AN 53 5 Table 1 Descriptive Characteristics of the Subsamples (%)

Description Full Bedfordshire Seville Sample Subsample Subsample

Gender Male 50.4 53.2 47.3 Female 49.6 46.8 52.7 Age 18–24 67.8 57.3 75.1 25–30 22.7 31.1 17.7 >31 5.8 11.6 2.4 Total (number) 516 267 249

Table 2 Basic Economic Data in the Two Regions

Indicator Bedfordshirea Sevilleb

Income Per Capita 2007(GDP PPS Per Capita) 31,600 20,200 Activity Rate (2003–2006) 70.9 64.6 Unemployment Rate (2009) 5.9 25.4 Female Unemployment Rate (2009) 4.7 27.1 Male Unemployment Rate (2009) 6.9 24.1 Employment in High-Tech Sectors 8.08 2.43 (2008, % of Total Employment) Individuals Regularly Using Internet 86.0 52.0 (2010, % of Individuals Who Accessed the Internet, on Average, at Least once a Week) aData for Bedfordshire, Hertfordshire. bData for Andalusia. Source: Eurostat, Regional Statistics NUTS 2.

H3d: Social valuation has a positive influence tionnaires were collected: 267 British students on entrepreneurial PA. from the University of Bedfordshire in Luton and 249 Spanish students from the University H3e: Social valuation has a positive influence of Seville (see Table 1). on entrepreneurial PBC. The two regions differ in some of their economic characteristics. Bedfordshire H3f: Social valuation has a positive influence is located near London; it is very well con- on SN. nected with the British capital, and its income level is one of the highest in the EU-15.1 On Research Methodology the other hand, Seville is located in southern Data Spain; it is one of the least industrialized Data come from a survey on final-year busi- regions of the country, and its income level is ness undergraduate students of two different therefore one of the lowest in the EU-15 European regions. Five hundred sixteen ques- (Table 2).

1EU-15 means that the comparison is made between the 15 countries that were members of the European Union before the accession of several East European states in 2004.

654 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Regarding entrepreneurship, the Eurobaro- Data Analysis meter data show that the entrepreneurial activity Given the relationships between different index in the United Kingdom is above the EU-15 perceptions and the entrepreneurial intention, average but, on the contrary, this index for structural equation modeling has been chosen Spain is below the EU-15 average (European for the analysis. In particular, Partial Least Commission 2007). This result is consistent with Squares (PLS) is applied, and PLS Graph V. 3.00 a higher entrepreneurship rate and a lower Build 1126 software is used for the data analy- proportion of business failures in the United sis (Chin and Frye 2003). This multivariate sta- Kingdom. tistical technique is suitable when exploratory Cultural values differ considerably between studies are carried out and relatively small both countries, with the U.K. scoring higher samples are used (Sánchez-Franco and Roldán than Spain in individualism and masculinity, 2005). whereas lower in uncertainty avoidance and To test H1a to H1f, a PLS model was built power distance (Hofstede 2003). In this sense, a using data from the full sample (Bedfordshire more pro-entrepreneurial set of values is present and Seville). Then, with the resulting constructs in the United Kingdom (Mueller, Thomas, and (entrepreneurial intention, entrepreneurial PBC, Jaeger 2002). Additionally, the United Kingdom entrepreneurial PA, SN, CV, and SV) we per- also has a higher proportion of individuals with formed an analysis of variance (ANOVA) a low perception of financial difficulties for the test to check for the existence of possible start-up, a high risk tolerance, and a high prob- gender differences in the constructs of the two ability of starting the business as a result of an subsamples. opportunity. To test H2a to H2e, two PLS models for the full sample were built (Bedfordshire and Seville together): one for men and the other for women. Scales Finally, regarding H3a to H3f, a dichotomous The Entrepreneurial Intention Question- control variable (BED) was included in the two naire (EIQ) developed by Liñán, Urbano, previous PLS models (for men and women) to and Guerrero (2011) was used to test the reflect the influence of the regional environment hypotheses proposed. The questionnaire was (value 1 for the Bedfordshire subsample and built to avoid some of the most common value 0 for the Seville subsample). Then, a problems in this kind of analyses, such as multigroup analysis was performed to look for common method bias, evaluation apprehen- statistically significant differences in path coef- sion, or acquiescence bias. Additionally, in the ficients (Chin 1998). empirical analysis, reliability and validity analyses will be repeated to confirm the pre- Results vious results and ensure the instrument’s The analysis of the measurement model for appropriateness. the full sample found low loadings for a small The questionnaire uses Likert-type scales to number of items. They were removed, and the measure each construct (response range is 1 to model was run again. Scores regarding item 7, with 4 being the central value). Thus, 20 reliability, construct reliability and convergent, items in the first part of the questionnaire and discriminant validity were then satisfactory measure the four core constructs of the TPB (see Tables 3 and 4). (entrepreneurial intention, entrepreneurial PA, When we repeated the analysis for each entrepreneurial PBC, and entrepreneurial SN). region individually, the results were similar. A sample item for entrepreneurial intention is: Entrepreneurial intention levels are notably “I am determined to create a business venture higher in Bedfordshire (mean value of 4.95) in the future.” On the other hand, the EIQ also than in Seville (3.94), as was expected given the provides measures of CV and SV. The following economic and cultural differences between are example items: “My friends value entrepre- both regions. This is in accordance with neurial activity above other activities and respondents in Bedfordshire showing more careers” (for CV) and “The culture in my positive PA (5.41 versus 4.85) and PBC (4.53 country is highly favorable toward entrepre- versus 3.73) than those in Seville. Then, neurial activity” for (SV). In the present study, ANOVA analyses were performed in each all six constructs have been measured through region to compare male and female scores for reflective indicators. the six constructs.

SANTOS, ROOMI, AND AND LI LIÑÁNN~AN 55 7 Table 3 Reliability and Convergent Validity Analysis for the Full Sample (Bedfordshire and Seville N = 516)

Construct Items Loadings Composite Average Variance Reliability Extracted (AVE)

Entrepreneurial Intention A04 0.7591 0.914 0.680 A06 0.7835 A13 0.8864 A17 0.8713 A19-rev- 0.8156 Entrepreneurial PA A10 0.8443 0.880 0.709 A15 0.8680 A18 0.8129 Social Norms A03 0.7989 0.859 0.672 A08 0.7758 A11 0.8800 Entrepreneurial PBC A01 0.7961 0.847 0.582 A07 0.7756 A14 0.8112 A20 0.6579 Closer Valuation C1 0.7704 0.846 0.647 C4 0.8055 C7 0.8355 Social Valuation C2 0.8676 0.856 0.748 C6 0.8622

Table 4 Convergent and Discriminant Validity of Constructs for the Full Sample (Bedfordshire and Seville, N = 516)

Entrep. Entrep. Social Entrep. Closer Social Intention PA Norms PBC Valuation Valuation

Entrep. Intention 0.824 Entrep. PA 0.790 0.842 Social Norms 0.354 0.398 0.819 Entrep. PBC 0.669 0.603 0.367 0.762 Closer Valuations 0.427 0.381 0.266 0.419 0.804 Social Valuations 0.308 0.283 0.128 0.372 0.469 0.864

Diagonal elements (bold) are the square root of the average variance extracted (AVE) between the constructs and their measures. Off-diagonal elements are correlations between constructs. For discriminant validity, diagonal elements should be larger than off-diagonal elements in the same row and column.

856 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT As seen in Table 5, men always had higher dents stating a high intention (5 or higher in entrepreneurial intentions and perceptions in the 1–7 scale) is 53.6 percent for males (65.6 each region individually considered (the mean percent in Bedfordshire and 39.6 percent in is higher for men than for women). Regarding Seville) and 30 percent for women (41.2 entrepreneurial intentions, the share of respon- percent in Bedfordshire and 19.5 percent in

Table 5 ANOVA to Test Gender Differences in Entrepreneurial Intentions and Perceptions in Bedfordshire and Seville Subsamples

Gender NM SD SSq d.f. MSq Fp

Entrep. Inten. Bed. Men 128 5.2938 1.27309 Inter 31.569 1 31.56 15973 .000 Bed. Women 114 4.5702 1.54146 Intra 474.334 240 1.976 Total 505.903 241 Sev. Men 111 4.2559 1.48009 Inter 20.931 1 20.93 10.69 .001 Sev. Women 123 3.6569 1.32183 Intra 454.135 232 1.957 Total 475.066 233

H1a Supported Entrep. PA Bed. Men 135 5.6247 1.24883 Inter 13.350 1 13.35 8.160 .005 Bed. Women 119 5.1653 1.31260 Intra 412.290 252 1.636 Total 425.640 253 Sev. Men 112 5.1429 1.28806 Inter 17.896 1 17.89 Sev. Women 124 4.5914 1.34339 Intra 408.123 234 1.744 Total 426.019 235 10.26 .002

H1b Supported Entrep. PBC Bed. Men 134 4.6791 1.10476 Inter 5.920 1 5.920 4.696 .031 Bed. Women 121 4.3740 1.14251 Intra 318.967 253 1.261 Total 324.887 254 Sev. Men 111 3.9797 1.03415 Inter 13.490 1 13.49 13.20 .000 Sev. Women 122 3.4980 0.98883 Intra 235.954 231 1.021 Total 249.444 232

H1c Supported SN Bed. Men 132 5.0000 1.20079 Inter 4.508 1 4.508 3.140 .078 Bed. Women 121 4.7328 1.19538 Intra 360.360 251 1.436 Total 364.868 252 Sev. Men 111 5.2643 1.32659 Inter 0.023 1 0.023 0.013 .909 Sev. Women 122 5.2842 1.33035 Intra 407.731 231 1.765 Total 407.754 232

H1d Not supported CV Bed. Men 141 4.5201 1.22952 Inter 3.397 1 3.397 2.550 .112 Bed. Women 123 4.2927 1.06148 Intra 349.101 262 1.332 Total 352.498 263 Sev. Men 112 4.0298 1.08549 Inter 0.411 1 0.411 0.319 .573 Sev. Women 114 3.9462 1.17593 Intra 300.286 234 1.286 Total 301.286 235

H1e Not supported SV Bed. Men 142 4.7500 1.25159 Inter 0.007 1 0.007 0.005 .944 Bed. Women 143 4.7398 1.08142 Intra 363.550 263 1.382 Total 363.557 264 Sev. Men 110 3.7227 1.22397 Inter 0.000 1 0.000 0.000 .995 Sev. Women 124 3.7218 1.29775 Intra 370.444 232 1.597 Total 370.444 233

H1f Not Supported

SANTOS, ROOMI, AND AND LI LIÑÁNN~AN 57 9 Seville). However, the ANOVA test showed that To take into account the regional context, a these gender differences were only significant dummy variable (BED) was included in the for three of the four central elements of Ajzen’s model. As shown in Figure 3, the path coeffi- model: entrepreneurial intention, entrepreneur- cients are broadly similar and the explained ial PA, and entrepreneurial PBC. Therefore, variance for entrepreneurial intention is also H1a, H1b, and H1c are supported, but H1d, notably high (68.7 percent for men and 68.3 H1e, and H1f are not. percent for women). The same as before, the Second, the models were tested separately links between SN and the entrepreneurial on the subsample for all men (Bedfordshire intention were not significant, whereas the and Seville) and the subsample for all women links between all other constructs of the TPB (Bedfordshire and Seville). Results showed that model were significant. the relationships were significant for both men Regarding the links between the dummy and women in the two regions. The model variable (BED) and the six constructs, all path explained 66.1 percent (for men) and 67.6 coefficients were significant in both models percent (for women) of the variance in entre- (women and men). This means that the region- preneurial intentions (see Figure 2). Only the specific characteristics exert an influence on relationship between SN and entrepreneurial the perceptions and intentions regarding the intentions was not significant, in accordance start-up, being in general more positive in with results by other researchers (Autio et al. Bedfordshire. This is in accordance with its 2001; Krueger, Reilly, and Carsrud 2000; Liñán higher development level and more pro- and Chen 2009). Therefore, H2a, H2b, H2d, entrepreneurial culture. Only SN was perceived and H2e are supported whereas H2c is not. more negatively in Bedfordshire.

Figure 2 Structural Models for Women and Men in the Two Regions (Nwomen = 250 and Nmen = 254)

1058 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Figure 3 Structural Model for Women and Men of the Two Samples Including Regional Effect (Nwomen = 250 and Nmen = 254)

Finally, to statistically test H3, a multigroup Regarding their views of the environment analysis was carried out (Table 6). As may be around them, both genders have similar seen, only two path-coefficient differences perceptions about their macrosocial (SV), were significant: the one between SV and entre- microsocial, or closer environment (CV) and preneurial PA, and the one between SV and support toward the entrepreneurial activity entrepreneurial PBC. Therefore, this result (SN). Therefore, perceived SV does not differ leads to the rejection of H3a, H3b, H3c, and by gender. Instead, what is different is the way H3f, whereas H3d and H3e are supported. SV affects personal perceptions (PA and PBC). Thus, for males, more positive SV leads them Discussion to feel entrepreneurship as more attractive and The empirical analysis carried out in this feasible. In the case of women, perceived SV paper has yielded two main results. First, entre- has no effect on personal perceptions. Follow- preneurial intention is, both for females and ing the so-called social feminism view of entre- males, the result of socialization processes in preneurship (Ahl 2006; Byrne and Fayolle which personal perceptions about entrepre- 2010), it may be argued that females do not see neurship (entrepreneurial PA and PBC) play a entrepreneurship as a career option for them key role. Thus, this paper once again confirms (Bird and Brush 2002). As a consequence, the applicability of the TPB model to entrepre- women’s personal perceptions and intentions neurship, irrespective of gender. Men are found are not affected by the value society puts on to exhibit higher entrepreneurial intentions this activity. than women do, but this is the logical conse- Interestingly enough, these relationships quence of their more favorable PA and PBC. hold for two notably different regions. In

SANTOS, ROOMI, AND AND LI LIÑÁNN~AN 59 11 Table 6 t-Tests for Multigroup Analysis: Men and Women from the Full Samplea

Links Path Path Path Standard Standard SP t-statistic Men Women Difference Error Error Men Women

Ent PA-EntInt 0.5680 0.6050 −0.0370 0.0452 0.0526 0.7760 −0.6702 ns PBC-EntInt 0.2670 0.2300 0.0370 0.0469 0.0493 0.7620 0.6826 ns SN-EntInt 0.0410 0.0630 −0.0220 0.0354 0.0487 0.6728 −0.4596 ns SN-Ent. PA 0.3360 0.3930 −0.0570 0.0738 0.0546 1.0309 −0.7773 ns SN-PBC 0.2900 0.4120 −0.1220 0.0621 0.0515 0.9052 −1.8947 ns CV-Ent PA 0.1900 0.2230 −0.0330 0.0717 0.0632 1.0718 −0.4328 ns CV-SN 0.2500 0.2770 −0.0270 0.0747 0.0653 1.1127 −0.3411 ns CV-PBC 0.1840 0.2230 −0.0390 0.0787 0.0584 1.1004 −0.4982 ns SV-Ent PA 0.1640 −0.0090 0.1730 0.0730 0.0687 1.1236 2.1645 * SV-SN 0.1480 0.0570 0.0910 0.0877 0.0757 1.2994 0.9845 ns SV-PBC 0.2100 0.0440 0.1660 0.0690 0.0579 1.0105 2.3093 * aMultigroup analysis of links between BED and the six indicators has been omitted for clarity. Differences between the path coefﬁcients were not signiﬁcant. t(0.001; 502) = 3.32834; t(0.01; 502) = 2.59487; t(0.05; 502) = 1.96913.

***p < .001; **p < .01; *p < .05; ns, not signiﬁcant (based on t(502), two-tailed test).

Bedfordshire, the perceived valuation of dummy being included or not. Therefore, both entrepreneurship in the wider (SV) and closer males’ and females’ personal entrepreneurial (CV) environments is higher, as are the perso- perceptions are similarly affected by the valu- nal levels of PA, PBC, and entrepreneurial ation of entrepreneurship in their closer envi- intentions. However, this does not affect the ronment. In this case, there is no gender nature of these relationships. The results are the difference. This result is in line with that of same for both regions. In fact, once the country Verheul, van Stel, and Thurik (2006), who dummy is included (Figure 3), the inﬂuence of found the effect of “importance of family” to be SV on PA, PBC, and SN is much clearer and more the same for males and females. consistent (compared with Figure 2). The other A great majority of men probably exhibit a relationships in the model remain essentially masculine stereotype, and they do not feel unaffected however. gender discrimination (Bird and Brush 2002; This should make researchers be especially Byrne and Fayolle 2010). They tend to consider careful when analyzing data from different entrepreneurship as a way to win social pres- social environments. The social and economic tige or recognition. In contrast, women are situation matters. Failure to recognize this may more worried about access to some relevant yield biased and/or misleading results. As some resources because they feel more barriers for authors point out, more comparative studies the entrepreneurial activity (Becker-Blease and are needed to fully understand the sociocultural Sohl 2007; Brush et al. 2002; Carter and Allen inﬂuence on female entrepreneurship (Ahl 1997; Fabowale, Orser, and Riding 1995; 2006; Verheul, van Stel, and Thurik 2006). Marlow and Patton 2005; Smith-Hunter 2006). In turn, when the family, friends, and ethnic Summarizing, one thing is whether women feel group (CV) are considered, the value they put their social environment values the entrepre- on entrepreneurship does not have any differ- neurial activity, and another very different one ential effect by gender. As the comparison of is whether they feel the social environment Figures 2 and 3 clearly shows, results are essen- values their initiatives with the same intensity tially the same, regardless of the country as those of men.

1260 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Previous results have found that less entre- increase the visibility of entrepreneurship as a preneurial societies face a shortage of entrepre- career option for women. At the same time, neurship because the SV effect is not present policies must continue focusing on providing (Liñán, Urbano, and Guerrero 2011). Only fami- women with a higher infrastructure of tangible lies providing a favorable CV (because there is and intangible support to facilitate their deci- already an entrepreneur within it, for instance), sion to set up a firm (Marlow and Patton 2005). will promote higher entrepreneurial intentions However, this action is even more necessary in among their members, but these “entrepreneur- the case of relatively backward regions, such as ial families” are comparatively scarce. However, Seville. the relative participation of women in entrepre- Likewise, higher education at universities neurship needs not be very different from that in can play an important role in the promotion of more entrepreneurial societies. That is, those female entrepreneurship (Kickul et al. 2008; families and ethnic groups positively valuing Wilson, Kickul, and Marlino 2007). Entrepre- entrepreneurship will provide a supporting neurship education should be designed not environment for both males and females only to overcome actual discrimination (in (Verheul, van Stel, and Thurik 2006). practical knowledge or access to resources, for At most, it is the step from intention to instance), but also to take into account the action that may differ. That is, if women per- particular perceptions and motivations of ceive higher barriers than men, a lower fraction women (Bird and Brush 2002; Byrne and of them will try to start-up. In turn, it may be Fayolle 2010; Liñán, Rodríguez-Cohard, and argued that in areas with a more positive SV of Rueda 2011). The inclusion of female role entrepreneurship, social institutions are shaped models as guest speakers is a relevant measure to facilitate start-ups, and therefore, females in this respect (Kickul et al. 2008). (and also males) will find fewer barriers. Thus, a higher fraction of women will attempt to start Limitations their ventures. The generalizability of these results should not be taken for granted. A number of limi- Implications tations may have affected the results. The use The results of this paper show that women of student samples is the first one, as they are not born with lower entrepreneurial inten- may not be fully representative of the general tions than men (Wilson, Kickul, and Marlino adult population. However, despite some 2007). Rather, they perceive the entrepreneur- criticism regarding the use of student samples ial role as being less adequate for them. This (Robinson et al. 1991), some research has makes them perceive a lower entrepreneurial shown that the entrepreneurial intentions of PA and PBC which, in turn, explains why their university students remain quiet stable after intention levels are lower. Therefore, actions to graduation (Audet 2004; Liñán, Rodríguez- increase female’s perceived attraction and fea- Cohard, and Guzmán 2011), as they are sibility toward entrepreneurship will have an at the stage of making a decision about effect on intentions and, eventually, on actual their professional careers (Fitzsimmons and start-ups (Kickul et al. 2008). Douglas 2011; Shepherd and DeTienne 2005). On the theoretical side, this result calls for Additionally, such a population is repeatedly the need to fully understand why a more posi- used in entrepreneurship research, facilitating tive entrepreneurial SV does not lead to higher comparisons (Autio et al. 2001; Kickul et al. PA and PBC in women. It probably has to do 2009; Krueger, Reilly, and Carsrud 2000; Liñán with entrepreneurship being considered a and Chen 2009; Zhao, Siebert, and Hills “male” career option. If this is true, active 2005). policy measures to change this view are Another limitation derives from the geo- needed. Measures to increase the perceived SV graphic scope of this analysis. It is possible for of entrepreneurship in general will help samples coming from different countries or promote entrepreneurship (Liñán, Urbano, and regions to yield conflicting results. In particu- Guerrero 2011) but especially that of men. lar, this sample comes from two developed They will have little or no effect on the entre- countries. The results may not be equally appli- preneurial activity of women. cable to developing economies. Therefore, In turn, the specific promotion of “women more research is necessary to confirm or refute entrepreneurs” clubs or associations will these results in alternative settings.

SANTOS, ROOMI, AND AND LI LIÑÁNN~AN 61 13 Conclusions A Multicountry Study,” Entrepreneurship We consider that this paper has contributed Theory and Practice 30(5), 687–708. to an advance in the understanding of the Becker-Blease, J. R., and J. E. Sohl (2007). “Do interplay between gender differences and the Women-Owned Businesses Have Equal social environment in entrepreneurship. It has, Access to Angel Capital?,” Journal of Busi- first, confirmed that women and men form ness Venturing 22(4), 503–521. their intention to start a venture in the same Begley, T. M., and W. L. Tan (2001). “The manner. Thus, women have lower entrepre- Socio-Cultural Environment for Entrepre- neurial intentions, because they see this neurship: A Comparison between East-Asia option as being less attractive and less feasible and Anglo-Saxon Countries,” Journal of than men do. International Business Studies 32(3), 537– It has also shown that a more positive per- 553. ception about the SV of entrepreneurship leads Bertaux, N., and E. Crable (2007). “Learning men, but not women, to increase their attrac- About Women, Economic Development, tion and sense of feasibility toward the entre- Entrepreneurship and the Environment in preneurial activity. Hence, women may feel India: A Case Study,” Journal of Develop- starting a venture is highly valued by the mental Entrepreneurship 12(4), 467– society, but do not think this is an acceptable 478. option for them. This needs to be changed if a Bird, B. (1988). “Implementing Entrepreneurial substantial increase in the share of female Ideas: The Case for Intention,” Academy of entrepreneurship is sought. Management Review 13(3), 442–453. Bird, B., and C. G. Brush (2002). “A Gendered References Perspective on Organizational Creation,” Ahl, H. (2006). “Why Research on Women Entrepreneurship Theory and Practice 26(3), Entrepreneurs Needs New Directions,” 41–65. Entrepreneurship Theory and Practice 30(5), Brush, C. G., N. M. Carter, P. G. Greene, M. M. 595–621. Hart, and E. Gatewood (2002). “The Role of Ajzen, I. (1991). “The Theory of Planned Social Capital and Gender in Linking Finan- Behavior,” Organizational Behavior and cial Suppliers and Entrepreneurial Firms: A Human Decision Processes 50(2), 179–211. Framework for Future Research,” Venture Arenius, P., and M. Minniti (2005). “Perceptual Capital: An International Journal of Entre- Variables and Nascent Entrepreneurship,” preneurial Finance 4(4), 305–323. Small Business Economics 24(3), 233– Brush, C. G., N. M. Carter, E. Gatewood, P. G. 247. Greene, and M. M. Hart (2004). Clearing the Armitage, C. J., and M. Conner (2001). “Efficacy Hurdles: Women Building High-Growth of the Theory of Planned Behavior: A Meta- Businesses, Financial Times Prentice Hall Analytic Review,” The British Journal of Books. Upper Saddle River, NJ: FT/Prentice Social Psychology 40(4), 471–499. Hall. Audet, J. (2004). “A Longitudinal Study of the Busenitz, L. W., and C. M. Lau (1996). “A Cross- Entrepreneurial Intentions of University Stu- Cultural Cognitive Model of New Venture dents,” Academy of Entrepreneurship Creation,” Entrepreneurship Theory and Journal 10(1), 3–16. Practice 20(4), 25–39. Autio, E., R. H. Keeley, M. Klofsten, G. G. C. Busenitz, L. W., C. Gómez, and J. W. Spencer Parker, and M. Hay (2001). “Entrepreneurial (2000). “Country Institutional Profiles: Intent among Students in Scandinavia and in Unlocking Entrepreneurial Phenomena,” the USA,” Enterprise and Innovation Man- Academy of Management Journal 43(5), agement Studies 2(2), 145–160. 994–1003. Bandura, A. (1997). Self-Efficacy: The Exercise Byrne, J., and A. Fayolle (2010). “A Feminist of Control. New York: Freeman. Inquiry into Entrepreneurship Training,” in ——— (2001). “Social Cognitive Theory: An The Theory and Practice of Entrepreneur- Agentic Perspective,” Annual Review of Psy- ship. Eds. D. Smallbone, J. Leitao, M. chology 52(1), 1–26. Raposo, and F. Welter. Cheltenham: Edward Baughn, C. C., B. L. Chua, and K. E. Neupert Elgar Publishing, 76–100. (2006). “The Normative Context for Carter, N. M., and K. R. Allen (1997). “Size Women’s Participation in Entrepreneurship: Determinants of Women-Owned Businesses:

1462 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Choice or Barriers to Resources?,” Entrepre- Farr-Wharton, R., and Y. Brunetto (2007). neurship and Regional Development 9(3), “Women Entrepreneurs, Opportunity Recog- 211–220. nition and Government-Sponsored Business Chin, W. W. (1998). “The Partial Least Squares Networks: A Social Capital Perspective,” Approach to Structural Equation Modelling,” Women in Management Review 22(3), 187– in Modern Methods for Business Research. 207. Ed. G. A. Marcoulides. Mahwah, NJ: Law- Fayolle, A., O. Basso, and V. Bouchard (2010). rence Erlbaum Associates, 295–336. “Three Levels of Culture and Firms’ Entre- Chin, W. W., and T. Frye (2003). Pls-Graph, preneurial Orientation: A Research Agenda,” Version 3.00 (Build 1126). Houston, TX: Entrepreneurship and Regional Develop- University of Houston. ment 22(7–8), 707–730. Cooper, A. C., and W. C. Dunkelberg (1987). Fernández, J., F. Liñán, and F. J. Santos (2009). “Entrepreneurial Research: Old Questions, “Cognitive Aspects of Potential Entrepre- New Answers and Methodological Issues,” neurs in Southern and Northern Europe: An American Journal of Small Business 11(3), Analysis Using Gem-Data,” Revista de 11–23. Economía Mundial 23, 151–178. de Bruin, A., C. G. Brush, and F. Welter (2007). Ferreira, J. J., M. L. Raposo, R. G. Rodrigues, A. “Advancing a Framework for Coherent Dinis, and A. do Paço (2012). “A Model of Research on Women’s Entrepreneurship,” Entrepreneurial Intention: An Application Entrepreneurship Theory and Practice 31(3), of the Psychological and Behavioral 323–339. Approaches,” Journal of Small Business Davidsson, P., and J. Wiklund (1997). “Values, and Enterprise Development 19(3), 424– Beliefs and Regional Variations in New Firm 440. Formation Rates,” Journal of Economic Psy- Fitzsimmons, J. R., and E. J. Douglas (2011). chology 18(2–3), 179–199. “Interaction between Feasibility and Desir- De Carolis, D. M., and P. Saparito (2006). ability in the Formation of Entrepreneurial “Social Capital, Cognition, and Entrepreneur- Intentions,” Journal of Business Venturing ial Opportunities: A Theoretical Frame- 26(4), 431–440. work,” Entrepreneurship Theory and García-Cabrera, A. M., and M. G. García-Soto Practice 30(1), 41–56. (2008). “Cultural Differences and Entre- DeTienne, D. R., and G. N. Chandler (2007). preneurial Behaviour: An Intra-Country “The Role of Gender in Opportunity Identi- Cross-Cultural Analysis in Cape Verde,” ﬁcation,” Entrepreneurship Theory and Entrepreneurship and Regional Develop- Practice 31(3), 365–386. ment 20(5), 451–483. Eddleston, K. A., and G. N. Powell (2008). “The Gupta, V. K., D. B. Turban, S. A. Wasti, and A. Role of Gender Identity in Explaining Sex Sikdar (2009). “The Role of Gender Stereo- Differences in Business Owners’ Career Sat- types in Perceptions of Entrepreneurs and isﬁer Preferences,” Journal of Business Ven- Intentions to Become an Entrepreneur,” turing 23(2), 244–256. Entrepreneurship Theory and Practice 33(2), Etzioni, A. (1987). “Entrepreneurship, Adapta- 397–417. tion and Legitimation: A Macro-Behavioral Hindle, K., K. Klyver, and D. F. Jennings Perspective,” Journal of Economic Behavior (2009). “An ‘Informed’ Intent Model: Incor- and Organization 8(2), 175–189. porating Human Capital, Social Capital, European Commission (2007). Flash Euro- and Gender Variables into the Theoretical barometer: Entrepreneurship Survey of the Model of Entrepreneurial Intentions,” in EU, (25 Member States), United States, Understanding the Entrepreneurial Mind: Iceland, and Norway. Analytical Report. Opening the Black Box. Eds. A. Carsrud Luxembourg: DG for Enterprise and Indus- and M. Brännback. New York: Springer, try, European Union and The Gallup Orga- 35–50. nization. Hofstede, G. (2003). Culture’s Consequences: Fabowale, L., R. Orser, and A. Riding (1995). Comparing Values, Behaviors, Institutions “Gender, Structural Factors, and Credit and Organizations across Nations, 2nd ed. Terms between Canadian Small Businesses Newbury Park, CA: Sage Publications. and Financial Institutions,” Entrepreneur- Hughes, K. D., J. E. Jennings, C. G. Brush, S. ship Theory and Practice 19(4), 41–65. Carter, and F. Welter (2012). “Extending

SANTOS, ROOMI, AND AND LI LIÑÁNN~AN 63 15 Women’s Entrepreneurship Research in New Liñán, F., D. Urbano, and M. Guerrero (2011). Directions,” Entrepreneurship Theory and “Regional Variations in Entrepreneurial Cog- Practice 36(3), 429–442. nitions: Start-up Intentions of University Stu- Iakovleva, T., L. Kolvereid, and U. Stephan dents in Spain,” Entrepreneurship and (2011). “Entrepreneurial Intentions in Devel- Regional Development 23(3&4), 187–215. oping and Developed Countries,” Education Loscocco, K., J. Robinson, R. H. Hall, and J. K. and Training 53(5), 353–370. Allen (1991). “Gender and Small Business Kickul, J., F. Wilson, D. Marlino, and S. D. Success: An Inquiry into Women’s Relative Barbosa (2008). “Are Misalignments of Per- Disadvantage,” Social Forces 70(1), 65–83. ceptions and Self-Efficacy Causing Gender McGee, J. E., M. Peterson, S. L. Mueller, and J. Gaps in Entrepreneurial Intentions among M. Sequeira (2009). “Entrepreneurial Self- Our Nation’s Teens?,” Journal of Small Busi- Efficacy: Refining the Measure,” Entrepre- ness and Enterprise Development 15(2), 321– neurship Theory and Practice 33(4), 965– 335. 988. Kickul, J., L. K. Gundry, S. D. Barbosa, and L. Marlow, S., and D. Patton (2005). “All Credit to Whitcanack (2009). “Intuition versus Analy- Men? Entrepreneurship, Finance, and sis? Testing Differential Models of Cognitive Gender,” Entrepreneurship Theory and Style on Entrepreneurial Self-Efficacy and Practice 29(6), 717–735. the New Venture Creation Process,” Entre- Masters, R., and R. Meier (1988). “Sex Differ- preneurship Theory and Practice 33(2), 439– ences and Risk-Taking Propensity of Entre- 453. preneurs,” Journal of Small Business Kolvereid, L. (1996). “Prediction of Employ- Management 26(1), 31–35. ment Status Choice Intentions,” Entrepre- Matthews, C. H., and S. B. Moser (1996). “A neurship Theory and Practice 21(1), 47– Longitudinal Investigation of the Impact of 57. Family Background and Gender on Interest Krueger, N. F., and A. L. Carsrud (1993). “Entre- in Small Firm Ownership,” Journal of Small preneurial Intentions: Applying the Theory Business Management 34(2), 29–43. of Planned Behavior,” Entrepreneurship and Meek, W. R., D. F. Pacheco, and J. G. York Regional Development 5(4), 315–330. (2010). “The Impact of Social Norms on Krueger, N. F., M. D. Reilly, and A. L. Carsrud Entrepreneurial Action: Evidence from the (2000). “Competing Models of Entrepreneur- Environmental Entrepreneurship Context,” ial Intentions,” Journal of Business Ventur- Journal of Business Venturing 25(5), 493– ing 15(5–6), 411–432. 509. Langowitz, N., and M. Minniti (2007). “The Minniti, M., and C. Nardone (2007). “Being in Entrepreneurial Propensity of Women,” Someone Else’s Shoes: The Role of Gender Entrepreneurship Theory and Practice 31(3), in Nascent Entrepreneurship,” Small Busi- 341–364. ness Economics 28(2–3), 223–238. Liñán, F., and Y. W. Chen (2009). “Develop- Morris, M., and M. Schindehutte (2005). “Entre- ment and Cross-Cultural Application of a preneurial Values and the Ethnic Enterprise: Specific Instrument to Measure Entrepre- An Examination of Six Subcultures,” Journal neurial Intentions,” Entrepreneurship Theory of Small Business Management 43(4), 453– and Practice 33(3), 593–617. 479. Liñán, F., J. C. Rodríguez-Cohard, and J. Mueller, S. L., and M. C. Dato-On (2008). Guzmán (2011). “Temporal Stability of “Gender-Role Orientation as a Determinant Entrepreneurial Intentions: A Longitudinal of Entrepreneurial Self-Efficacy,” Journal of Study,” in Entrepreneurship Research in Developmental Entrepreneurship 13(1), Europe: Evolving Concepts and Processes. 3–20. Eds. O. J. Borch, A. Fayolle, P. Kyrö, and E. Mueller, S. L., and A. S. Thomas (2001). Ljunggren. Cheltenham, UK: Edward Elgar, “Culture and Entrepreneurial Potential: A 34–55. Nine Country Study of Locus of Control and Liñán, F., J. C. Rodríguez-Cohard, and J. M. Innovativeness,” Journal of Business Ventur- Rueda (2011). “Factors Affecting Entrepre- ing 16(1), 51–75. neurial Intention Levels: A Role for Educa- Mueller, S. L., A. S. Thomas, and A. M. Jaeger tion,” International Entrepreneurship and (2002). “National Entrepreneurial Potential: Management Journal 7(2), 195–218. The Role of Culture, Economic Development

1664 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT and Political History,” in Managing Trans- Decide to Start Companies,” Journal of Busi- national Firms: Resources, Market Entry and ness Venturing 15(2), 113–134. Strategic Alliances. Eds. M. A. Hitt and Smith-Hunter, A. E. (2006). Women Entrepre- J. L. C. Cheng. Amsterdam: JAI Press, 221– neurs Across Racial Lines, New Horizons in 257. Entrepreneurship Series. Ed. S. Shane. Chel- Naphiet, J., and S. Ghoshal (1998). “Social tenham: Edward Elgar. Capital, Intellectual Capital, and Organiza- Thomas, A. S., and S. L. Mueller (2000). “A Case tional Advantage,” Academy of Management for Comparative Entrepreneurship: Assess- Review 23, 242–266. ing the Relevance of Culture,” Journal of Neergaard, H., E. Shaw, and S. Carter (2005). International Business Studies 31(2), 287– “The Impact of Gender, Social Capital and 301. Networks on Business Ownership: A Thornton, P. H., D. Ribeiro-Soriano, and Research Agenda,” International Journal of D. Urbano (2011). “Socio-Cultural Factors Entrepreneurial Behaviour and Research and Entrepreneurial Activity,” International 11(5), 338–357. Small Business Journal 29(2), 105–118. Rimal, R. N., and K. Real (2003). “Understand- Uphoff, N. (2000). “Understanding Social ing the Influence of Perceived Norms on Capital: Learning from the Analysis and Behaviors,” Communication Theory 13(2), Experience of Participation,” in Social 184–203. Capital, a Multifaceted Perspective. Eds. P. Robinson, P. B., D. V. Stimpson, J. Huefner, Dasgupta and I. Serageldin. Washington: and H. K. Hunt (1991). “An Attitude The World Bank, 215–249. Approach to the Prediction of Entrepreneur- Verheul, I., L. Uhlaner, and R. Thurik (2005). ship,” Entrepreneurship Theory and Practice “Business Accomplishments, Gender and 15(4), 13–31. Entrepreneurial Self-Image,” Journal of Busi- Roomi, M. A., and G. Parrot (2008). “Barriers to ness Venturing 20(4), 483–518. Progression of Women Entrepreneurs in Verheul, I., A. van Stel, and R. Thurik (2006). Pakistan,” Journal of Entrepreneurship “Explaining Female and Male Entrepreneur- 17(1), 59–72. ship at the Country Level,” Entrepreneurship Sánchez-Franco, M. J., and J. L. Roldán (2005). and Regional Development 18(2), 151– “Web Acceptance and Usage Model,” Inter- 183. net Research-Electronic Networking Applica- Watson, J. (2002). “Comparing the Performance tions and Policy 15(1), 21–48. of Male and Female Controlled Businesses: Scherer, R. F., J. D. Brodzinsky, and F. A. Wiebe Relating Outputs to Inputs,” Entrepreneur- (1991). “Examining the Relationship ship Theory and Practice 26(3), 91–100. between Personality and Entrepreneurial Watson, J., and R. Newby (2005). “Biological Career Preference,” Entrepreneurship and Sex, Stereotypical Sex-Roles and Sme Owner Regional Development 3, 195–206. Characteristics,” International Journal of Sexton, D. L., and N. Bowman-Upton (1990). Entrepreneurial Behaviour and Research “Female and Male Entrepreneurs— 11(2), 129–143. Psychological Characteristics and Their Role Wells, B., T. J. Pfantz, and J. Byrne (2003). in Gender-Related Discrimination,” Journal “Russian Women Business Owners: Evi- of Business Venturing 5(1), 29–36. dence of Entrepreneurship in a Transition Shapero, A. (1982). “Social Dimensions of Economy,” Journal of Developmental Entre- Entrepreneurship,” in The Encyclopedia of preneurship 8(1), 59–73. Entrepreneurship. Eds. C. A. Kent, D. L. Wilson, F., J. Kickul, and D. Marlino (2007). Sexton, and K. Vesper. Englewood Cliffs, NJ: “Gender, Entrepreneurial Self-Efficacy, and Prentice Hall, 72–90. Entrepreneurial Career Intentions: Implica- Shepherd, D. A., and D. R. DeTienne (2005). tions for Entrepreneurship Education,” “Prior Knowledge, Potential Financial Entrepreneurship Theory and Practice 31(3), Reward, and Opportunity Identification,” 387–406. Entrepreneurship Theory and Practice 29(1), Woolcock, M., and D. Narayan (2000). “Social 91–112. Capital: Implications for Development Simon, M., S. M. Houghton, and K. Aquino Theory, Research and Policy,” The World (2000). “Cognitive, Biases, Risk Perception Bank Research Observer 15(2), 225– and Venture Formation: How Individuals 249.

SANTOS, ROOMI, AND AND LI LIÑÁNN~AN 65 17 Zahra, S. A., D. F. Jennings, and D. F. Kuratko Zhao, H., S. E. Siebert, and G. E. Hills (2005). (1999). “The Antecedents and Consequences “The Mediating Role of Self-Efﬁcacy in the of Firm-Level Entrepreneurship: The State of Development of Entrepreneurial Intentions,” the Field,” Entrepreneurship Theory and The Journal of Applied Psychology 90(6), Practice 24(2), 45–63. 1265–1272.

1866 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Journal of Small Business Business Management Management 2016 2014 54(1), ••(••), pp. pp. 67–84 ••–•• doi: 10.1111/jsbm.12131 Vertical Coopetition and the Sales Growth of Young and Small Firms by Christian Lechner, Birthe Soppe, and Michael Dowling

Vertical coopetition describes a situation where a firm has a vertical exchange relationship with a direct competitor. Drawing on resource-based perspectives and resource dependence theory, we examine specific characteristics of vertical coopetition that affect the sales growth of young and small firms. We propose that the size of the competitor and mutual dependence as well as overdependence among the cooperating rivals affect the growth potential of young and small firms differently. Our research makes a contribution to the literature by disentangling some of the benefits and threats related to vertical coopetition for young and small firms.

contractor relationships with a direct competi- Introduction tor, where the competitors are simultaneously Firms frequently engage in vertical, noncom- involved in both noncompetitive and competi- petitive relationships with their direct competitive relationships with each other. tors (Dowling and Lechner 1998; Ketchen, Although the establishment of relationships Snow, and Hoover 2004).1 Such vertical with buyers and suppliers, as a form of exchange relationships—supplier or buyer rela- social capital (Lin 2002), appears to be vital for tionships between competitors—have been the development of young and small firms called vertical coopetition (Dowling et al. (Beekman and Robinson 2004; Venkataraman 1996), inspired by a quote from Ray Noorda, and Van de Ven 1998), one could argue that CEO of Novell, who used the term coopetition teaming up with direct competitors would be the to refer to a business relationship with licens- least preferred choice. So, why should young ees that were, at the same time, Novell’s com- and small firms engage in and benefit from such petitors. Following Dowling et al. (1996), we relationships? Theory generally assumes that the define vertical coopetition as an ongoing and potential appropriation hazard is high in trans- cooperative vertical exchange relationship action relationships with competitors (Alvarez between competing firms. Vertical coopetition and Barney 2004; Wyld, Pugh, and Tyrrall 2012). can take the form of buyer, supplier, or sub- Proponents of coopetition suggest, however,

Christian Lechner is professor of strategy and entrepreneurship at the School of Economics and Manage- ment, Free University of Bolzano and afﬁliated professor at Université de Toulouse 1, Toulouse Business School. Birthe Soppe is associate professor of entrepreneurship at the Centre for Entrepreneurship, University of Oslo. Michael Dowling is professor of innovation and technology management at the Chair of Innovation and Technology Management, University of Regensburg. Address correspondence to: Christian Lechner, School of Economics and Management, Free University of Bolzano, Piazza Università 1, 39100 Bolzano, Italy. E-mail: [email protected]. 1For example, Samsung created a joint venture with LCD-TV competitor Sony for the joint manufacturing of LCD displays. In addition, Samsung supplies Apple with NAND chips for its iPhone and media processor slots for its iPods, although Samsung competes with Apple with smartphones, e-book readers, and MP3 players.

LECHNER, SOPPE, AND DOWLING DOWLING 67 1 that firms can derive valuable resources from view of the firm traditionally sustained that such relationships (Gnyawali and Madhavan firms can exploit resource quality differentials 2001; Lado, Boyd, and Hanlon 1997). The to increase a firm’s performance. Recent contri- implicit assumption is that value creation from butions have argued that value differentials can this type of social capital outweighs potential also derive from differing complementarities appropriation hazards. However, the research between resources to be acquired and those results are rather mixed and focus (1) on hori- already held in the organization (e.g., zontal coopetition (such as R&D alliances), (2) Adegbesan 2009; Wernerfelt 2011). In the case on large firms (e.g., Bengtsson and Kock 2000; of vertical coopetition, competitors generally Bonel and Rocco 2007; Gnyawali, He, and hold resources that are highly relevant for the Madhavan 2006), similar to research on buyer– other due to them sharing similar strategic supplier relationships in general (for a discus- industry factors (Gnyawali and Park 2009), sion, see Adams, Khoja, and Kauffman 2012), potentially reinforcing the complementarity and (3) on the comparison of firms engaged in effects between buyers and suppliers (Weigelt coopetition with those that are not. Previous 2013). Resource dependence theory also sug- work has been beneficial for improving our gests that business relationships can be valu- understanding of vertical coopetition, but it still able. An increasing interdependence between leaves unanswered questions of how young two parties is usually associated with a benefi- and small firms can gain advantages from cial intensification of the relationship (Beier exchange relationships with direct competitors and Stem 1969; Crook and Combs 2007; (Adams, Khoja, and Kauffman 2012) and grow. Frazier 1983; Zacharia, Nix, and Lusch 2009), Growth differentials of firms engaged in vertical except when the relationship becomes too coopetition have not been researched before. asymmetrical (Anderson and Weitz 1989; Chung Dagnino (2009), for example, suggested that a 2012). firm’s capability to manage both cooperation In this paper, we advance the understand- and competition better than its peers might ing of how vertical coopetition can create sales explain differences between firms. However, growth differentials for young and small differences can also stem from the characteris- firms based on the combined analysis of the tics of the transaction partners (Stuart 2000). We resource-based view and resource dependence argue that the type of transaction partners and theory. Our research makes a theoretical con- different forms of dependence influence the tribution by disentangling some of the benefits effect of vertical coopetition. Important consid- and threats associated with vertical coopetition erations for small and young firms before enter- for young and small firms. In our empirical ing such relationships are: with whom should analysis, we investigate how specific aspects of they engage in a relationship, how many such relationships—the size of the cooperating resources should be exchanged, and how much partner and mutual dependence as well as they should rely on an exchange partner in order overdependence in vertical coopetition—affect to further develop the business through increas- sales growth. We test our hypotheses with a ing sales (Beekman and Robinson 2004). sample of venture capital financed firms. We Given the dilemma of value creation and based our choice on the following reason: the appropriation through transaction relationships sales growth of a young and small firm depends (Alvarez and Barney 2004; McCarter and on the effectiveness of its important strategic Northcraft 2007), the primary purpose of this choices. Growth beyond its initial status is vital paper is to analyze the ambiguous effects of for these firms. Because we were interested in vertical coopetition. Resource-based perspec- understanding better some mechanisms of sales tives help us to understand how resources and growth, we searched for firms with some (a resource combinations drive performance, priori) growth potential, as evidenced by whereas resource dependence theory helps us venture capital selection. This selection largely to develop insights into the benefits and threats rules out lack of growth intentions as an expla- of interfirm relationships. The resource-based nation for differential growth.2 We focused on

2It should be noted that in this research, we do not account directly for entrepreneurial intentions or entrepreneurial behavior (for a discussion, see e.g., Carsrud and Brännback 2007), but investigate factors driving the growth of small and young ﬁrms.

268 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT sales growth as an effectiveness measure as Shah and Tripsas 2007; Terpend et al. 2008; it is strongly linked to firm development and, Weigelt 2013). most likely, directly and more immediately However, relationships with competitors are affected by firm choices concerning interfirm risky, and competitors might have an interest relationships than efficiency measures such as in appropriating most of the value created profitability (Stuart 2000). Understanding the (Alvarez and Barney 2004; McCarter and benefits as well as the threats for young and Northcraft 2007; Park and Russo 1996), espe- small firms in vertical coopetition is particularly cially if the partner is larger (Chung 2012). So, important because the failure or success of why should supplier or buyer relationships such interactions among firms can have a with a competitor be more valuable than those lasting impact on the sales growth of small and with other suppliers or buyers? First, competi- young firms. tors share the same context, are subject to similar threats and opportunities, and, thus, Theory and Hypotheses they display a similar need for basic resources Vertical Coopetition, Resource and knowledge to be able to operate success- Complementarity, and Growth fully in the market. Competitors share similar The resource-based view describes how the strategic industry factors, which represents an possession of heterogeneous resources leads to equivalent concept of firm resources, but at the performance differentials (Barney 1986, 1991; industry level (Amit and Schoemaker 1993). Peteraf 1993). However, the context of young Resources are “determined at the market level and small firms is different; they are affected by through complex interactions among the firm’s liabilities of newness and smallness (Baum competitors, customers, regulators, innovators 1996; Stinchcombe 1965). They are character- external to the industry, and other stakehold- ized by a lack of legitimacy, knowledge, and ers” (Amit and Schoemaker 1993, p. 36). size, leading to both an imperative to learn and Therefore, competitors possess highly relevant grow, and growth with insufficient resources resources for each other (Gnyawali and Park (Lechner and Dowling 2003). Therefore, the 2009). These resources are particularly valu- development of firms depends largely on attract- able because they can be part of a firm’s ing and accessing resources, as well as develop- product offering; the underlying resources ing and possessing at least some important show a high degree of complementarity. resources. Social capital, defined as the Recent contributions to the resource-based per- resources an individual or an organization can spective have argued that performance differ- access through relationships (Lin 2002; Payne entials might not only be due to heterogeneity et al. 2011), is thus critical for young and small in terms of resource quality but also due to firms (Florin, Lubatkin, and Schulze 2003; complementarities between acquired resources Townsend and Busenitz 2008; Venkataraman and resources already held by the acquiring and Van de Ven 1998), but poses a challenge: firm (Adegbesan 2009; Townsend and Busenitz “the twin tasks of gaining access to those 2008; Wernerfelt 2011). In the case of buyers resources to generate rents associated with a and suppliers who are, at the same time, direct market opportunity, and doing so in a way that competitors, these effects should be even enables this economic actor to appropriate at stronger (Petersen, Handfield, and Ragatz least some of the rents that were generated, can 2005; Weigelt 2013). become quite complicated” (Alvarez and Barney We sustain that vertical coopetition is a 2004, p. 622). The growth of a young and small source of social capital for accessing more of firm is thereby the outcome of developing and the highly relevant resources needed for the maintaining value-generating relationships firm’s development and for effectively exploit- (Venkataraman and Van de Ven 1998) while ing resource complementarities that lead to extracting (Weigelt 2013; Wu 2008) and appro- stronger sales growth. However, the research priating part of the value (Alvarez and Barney findings have been rather mixed, calling for 2004). Research streams confirm the value of research that investigates the relationship buyers and suppliers, which can be summarized between social capital and the growth of firms under the wider concept of relational resources (Payne et al. 2011). Although there might be and, thus, social capital (Adams, Khoja, and strong arguments that the value creation poten- Kauffman 2012; Dyer and Singh 1998; tial is high, the appropriation hazard might also Nicholls-Nixon and Woo 2003; Payne et al. 2011; be high. We propose, in line with related

LECHNER, SOPPE, AND AND DOWLING DOWLING 69 3 research (Stuart 2000), that specific elements of relevant products faster. In addition, vertical the relationship and partner characteristics can relationships with larger competitors can increase value creation and decrease value provide young and small firms with a higher appropriation hazards. stability of sales and a more continuous and reliable supply of resources, which will even- Vertical Coopetition with Larger tually influence firms’ sales and free up time for Competitors and the Sales Growth of other business activities (Lechner and Dowling Young and Small Firms 2003). Overall, engaging with a large competi- Because any form of interfirm relationships tor should—everything else being equal—lead deals with actors and nodes, it is only logical to to more social capital. investigate some attributes of the exchange However, research on unbalanced relation- partner as well as the constituencies of the ships deals with the benefits and costs associ- exchange relationship. The balance between ated with asymmetries in the sizes of the firms two partnering firms in vertical coopetition that engage in cooperative relationships seems to play an important role for young and (Alvarez and Barney 2001; Ross and Robertson small firms (Alvarez and Barney 2001; Ross and 2007; Wyld, Pugh, and Tyrrall 2012). This Robertson 2007). They can cooperate with research tends to suggest that large firms—in a larger, more established firms in their business sort of power play—capture more value from context, or with smaller or similar-sized part- interfirm relationships than young and small ners. When specifically considering the context firms (Chung 2012). This behavior might, for for young and small firms, it is evident that example, translate (on a minor scale) in late similar-sized competitors share similar con- payment practices (Paul and Boden 2011). In straints. Larger competitors are usually not such asymmetric constellations, new ventures affected by these constraints and most often might easily lose control over their resources, already possess considerable legitimacy, repu- which makes the performance and survival tation, experience, and other resources. The prospects of young and small firms suffer while implicit transfer of status and reputation may large firms appropriate most of the relationship reduce uncertainty about the start-up’s quality value (Alvarez and Barney 2001; Chung 2012; and increases the potential for future growth Ketchen, Ireland, and Snow 2007). opportunities. Increasing reputation also facili- This line of research, though, analyzes asym- tates the mobilization of required resources and metry mainly in horizontal alliances. In hori- the access to other stakeholders such as new zontal alliances, firms typically bring in some of suppliers and buyers who are willing to engage their resources for a specific common purpose. with young and small firms if the perceived risk Although the overlap of the resource base is of an exchange relationship is lower (Stuart small for large firms, logically, it is much larger 2000; Stuart, Hoang, and Hybels 1999). Coop- for small firms. Thus, young and small firms do erating with a more experienced competitor risk more and can potentially lose more can also provide firms with more opportunities (Ketchen, Ireland, and Snow 2007). The to acquire advanced know-how (Baum, context of vertical exchange relationships with Calabrese, and Silverman 2000; Shepherd, competitors, however, differs from horizontal McMullen, and Jennings 2007). This know-how alliances. It is more likely that they enter such comprises the management of business pro- relationships because of higher resource cesses, knowledge about competition and a complementarity, leading to mutual depen- firm’s environment (Cope 2005), efficient dence among firms. For larger firms, higher buying and selling practices, as well as relevant resource complementarity, combined with operational issues (Padula and Dagnino lower overlap with a firm’s total resource base, 2007). would mean that the common benefits are Furthermore, large and established firms greater than the private benefits. As a conse- (with, as a consequence, higher sales) have quence of this particular situation, we argue more experience in knowing what sells and that it is less risky for young and small firms how to sell it. Young and small firms usually to work with larger than with similar-sized lack experience and most likely need to competitors. develop this type of knowledge. The interac- In essence, we propose that it is more ben- tion with these experienced firms should help eficial and less risky for young and small firms young and small firms to develop and sell more to work with larger than with similar-sized

470 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT companies. The resulting social capital that and Lusch 2009). The higher the intensity of this leads to resource access is greater with larger mutual dependence, the more the exchange companies than with similar-sized companies, partners will engage in the relationship (Crook and partnerships with larger competitors will and Combs 2007; Kumar, Scheer, and help new ventures develop and sell more rel- Steenkamp 1995) and the higher the relative evant products faster. exchange volume between the parties. From the point of view of the young and small firm, the H1: Vertical coopetition with larger competitors higher the exchange volume relative to the total will have a positive effect on the sales growth exchange volume, the higher the dependence of young and small firms. within the vertical coopetition. Exchange volume is a relatively symmetrical concept of Mutual Dependence in Vertical mutual dependence as the increase on the Coopetition and Sales Growth of Young supply side corresponds to an increase on the and Small Firms demand side. Therefore, the willingness to One reason why firms choose to enter supply more matches the willingness to buy relationships with competitors is the access to more. This indicates—ceteris paribus—that with complementary resources or other necessary increasing relative exchange volume, partners assets (Ahuja 2000; Gnyawali and Park become less replaceable and, thus, higher 2009; Lechner, Dowling, and Welpe 2006; mutual dependence among the competitors Quintana-García and Benavides-Velasco 2004). exists. This effect should be increased through This situation creates some form of dependence higher resource complementarity (Zacharia, on the other party’s resources, which—in any Nix, and Lusch 2009). The higher this interde- interfirm relationship—potentially carries a risk pendence is, the higher the risk of mutual nega- of opportunism. This risk might be higher in tive consequences of opportunistic behavior, relationships with direct competitors (Park and and thus the less likely the actual occurrence of Russo 1996). In the case of young and small opportunistic behavior (Beier and Stem 1969; firms, the risks and disadvantages of vertical Crook and Combs 2007; Frazier 1983; Heide and coopetition might even be reinforced as young John 1988; McCarter and Northcraft 2007). and small firms are generally more vulnerable. Hence, higher interdependence also implies For instance, a critical aspect of firm relation- higher exit barriers from the relationships. High ships is the potential for conflict between the exit barriers increase the incentive for both allied firms. The source of the conflict is the parties to maintain and strengthen the relation- conflicting roles of the exchange partners. If ship, which will increase the trust in the rela- the exchange partner uses destructive conflict tionship and reduce potential conflicts resolution methods such as domination or coer- (Kohtamäki et al. 2012; Kumar, Scheer, and cion, then the young and small firm is likely to Steenkamp 1995). be harmed severely (Chung 2012; Deutsch We therefore suggest that higher relative 1969). In essence, opportunism and destructive exchange volume—as an indicator of the conflicts are potential threats for young and intensity of interdependence among firms— small firms in vertical coopetition. Marketing will strengthen and stabilize the relationship, channel research that investigated supplier reducing opportunism and conflicts (Crook and buyer relationships has showed mixed and Combs 2007). Resource complementarities results on the effects of a firm’s dependence on and co-specialization increase, and young and performance (Kumar, Scheer, and Steenkamp small firms that engage in vertical coopetition 1995). Whether these relationships exercise can thereby optimize their value proposition either a positive or negative influence on (Rivkin and Siggelkow 2007; Townsend and firm growth is largely determined by the Busenitz 2008) and better address customer intensity and form of dependence (Kohtamäki needs (Priem 2007). This, in turn, might posi- et al. 2012; Kumar, Scheer, and Steenkamp tively influence the sales growth of young and 1995). small firms (Beekman and Robinson 2004). It We assume that two firms will engage in should be noted, however, that according to vertical coopetition because of expectations of this logic, low exchange volume as an indica- higher resource complementarities. Resource tor of low interdependence would actually complementarities lead to a mutual, positive favor opportunism and conflicts. We propose form of dependence among firms (Zacharia, Nix, thus

LECHNER, SOPPE, AND AND DOWLING DOWLING 71 5 H2: Increasing relative exchange value in ver- and the loss of control over the relationship tical coopetition will have a positive effect on and the partner’s actions. Asymmetrical depen- the sales growth of young and small firms. dence can therefore become a handicap, which may eventually result in lower sales growth Overdependence in Vertical Coopetition (Alvarez and Barney 2001; Ketchen, Ireland, and Sales Growth of Young and and Snow 2007). Small Firms Thus, if young and small firms perceive that Resource access through interfirm relation- the purchased products from a competitor are ships is an option for firms to overcome highly important for their competitive advan- resource constraints, but it also has its costs. tage and less important for the competitor or, Although we suggest that mutual dependence vice versa, if supplying a competitor with spe- among competitors through a relative high cialized products is critical and necessary for the exchange volume will positively affect sales young and small firm but less important for the growth, other forms of dependence might be competing buyer, then this asymmetrical depen- more harmful. Asymmetrical relationships are dence will harm a firm’s sales growth potential. characterized by a strongly unequal dependence, even an overdependence, of one party H3: Perceived overdependence in vertical on the other, that is, one party is highly coopetition will have a negative effect on the important to the other but not vice versa sales growth of young and small firms. (Chung 2012; McCarter and Northcraft 2007). In contrast, as argued before, symmetric relationships with mutual forms of dependence Research Design and are more stable than asymmetrical ones Methodology (Anderson and Weitz 1989; Kohtamäki et al. Sample and Survey 2012; McCarter and Northcraft 2007). In asym- The hypotheses were tested using data from metrical vertical coopetition, common interests a pretested survey, which was specifically are low, thus increasing conflictual behavior designed for this study and sent to CEOs and in both parties (Chung 2012; Lawler, Ford, founders of venture capital (VC)-financed firms and Blegen 1988; Molm 1989). Although it is in Germany. Because there was no comprehen- obvious that the less dependent firm tends to sive database on VC-financed companies in not care about the other party’s interests, the Germany, we developed our own database. Firm more dependent party has strong grounds for information was primarily identified through the reducing its dependence, for altering the rela- German Venture Capital Association and the tionship, for exiting the relationship, or for “German High-tech Gründerfonds,” which led anticipating conflictual or opportunistic behav- to a population of 812 VC-backed companies. ior (Lawler, Ford, and Blegen 1988). The We restricted the sample firms in the following absence of conflicts favors trust in the rela- way. The firms had to be less than 10 years old. tionships, whereas high conflict potential The rationale behind choosing VC-financed undermines trust. Higher mobility of the other firms is that in this study, we are concerned with party (through lower exit barriers), higher linking vertical coopetition to firm growth dif- conflict potential, and a low trust environment ferentials among small and young firms. Firms favor harmful opportunistic behavior (Kumar, that acquire venture capital have growth poten- Scheer, and Steenkamp 1995; McCarter and tial and tend to grow as much as possible in Northcraft 2007). order to satisfy the venture capitalists’ return Overdependence is thereby an “open invita- targets. Thus, limiting the sample of small and tion” for opportunism and conflicts. These will new firms to VC-backed firms means choosing a reduce the potential benefits of the relation- sample of firms that are eager and willing to ship, enable only partial resource access, and grow, where the likely effects of CEOs’ and slow down organizational processes. Partial founders’ varying growth motivations on new resource access requires the search for alterna- venture growth are strongly limited. Hence, this tives, thus reducing the speed of firm develop- sample might be very suitable for investigating ment, resulting in negative effects on sales how young and small firms can benefit from growth. Hence, there are risks and disadvan- coopetitive relationships in terms of sales tages of vertical coopetition for firms, such as a growth, depending on certain relationship attri- strong dependence on a competitor’s resources butes. Additionally, VC-backed firms are not

672 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT subject to corporate sponsorship and can there- electronic, services, media, and energy indus- fore be considered independent ventures tries. Fifty-three of the firms are involved in (Robinson 1999). vertical coopetition, which represents 65 A survey was an appropriate means for col- percent of the final sample. lecting data on vertical coopetition because secondary resources did not contain detailed Dependent Variable information on firm–buyer and firm–supplier In this paper, we are concerned with the link relationships. All information about vertical between the growth of young and small firms coopetition was self-reported, but sales data and the attributes of coopetitive relationships were matched with objective data, acquired such firms are engaged in, and seek to predict from Amadeus and Kompass—databases.3 To growth differentials of new and small firms ensure that a high proportion of the answers engaged in vertical coopetition. Because was valid, the questionnaires were sent to the growth can occur in a multitude of a firm’s CEOs and/or founders of the start-ups, using a aspects, many different growth measures have key informant approach (Brush and been discussed in the literature, including abso- Vanderwerf 1992; Chandler and Hanks 1993). lute versus relative sales, profitability, the We also controlled for common method bias. number of employees, market share, and physi- We conducted a Harman one-factor test cal output (Delmar, Davidsson, and Gartner (Podsakoff and Organ 1986). Not a single factor 2003; Murphy, Trailer, and Hill 1996; Shepherd emerged; the principal factor did not explain and Wiklund 2009; Weinzimmer, Nystrom, and the majority of the variance. In addition, we Freeman 1998). We chose average sales growth adopted a marker variable technique (Lindell as the dependent variable. and Whitney 2001). As a marker variable, we First, although there is no best measure of used the question of whether the firms engaged new venture growth, the use of sales seems to in vertical coopetition wished to receive the be a widely used and highly legitimate measure results of the study. We have no theoretical for firm growth in empirical growth studies, argument concerning the explanatory variables particularly for small and young firms (Delmar, to believe that there is a theoretical correlation Davidsson, and Gartner 2003; Hoy, McDougall, between the response to this question and and Dsouza 1992; Lebrasseur, Zanibbi, and the three explanatory variables. The highest Zinger 2003; Robinson 1999; Weinzimmer, correlation was 0.07 (with overdependence), Nystrom, and Freeman 1998). It indicates the and there were no significant correlations. We increasing acceptance of a firm’s products and can therefore reasonably exclude a common services by customers (Gilbert, McDougall, and method bias. Audretsch 2006). Although other growth indi- The questionnaires were sent to 600 firms. cators such as market share and production Of the 96 questionnaires we received, 82 met output are problematic when comparing firms all sample selection criteria and were suffi- across different industries, it has been argued ciently complete. This translates to an effective that sales are highly suitable for comparing and acceptable response rate of about 15 firms across varying industry contexts, such as percent. To control for nonresponse bias, we the firms in our sample (Delmar, Davidsson, followed the recommendations by Miller and and Gartner 2003). Smith (1983) to compare early and late respon- Second, although sales are not the perfect dents. Tests for mean differences did not show indicator of growth for all purposes, each significant differences for any of the research alternative measure of growth represents a variables. Thus, the final sample consists of 82 different type and a particular aspect of VC-financed start-ups that had to be less than growth (Shepherd and Wiklund 2009). There- 10 years old and have a sales track record. fore, the different measures cannot be used as Sample firms came mainly from the information substitutes (McKelvie and Wiklund 2010), and technology (IT), pharmaceutical, engineering, the preferred choice of the dependent variable

3We had also asked the respondents to indicate their sales data. We were able to obtain about 60 percent of the sales data from objective sources. The correlation between the subjective data and objective data was greater than 0.99. These results indicate the great accuracy of the data obtained through the questionnaire and also strengthened our conﬁdence concerning self-reported data about the coopetitive relationships.

LECHNER, SOPPE, AND AND DOWLING DOWLING 73 7 (whether a single growth measure or a mix of such an approach might smooth out irregulari- different measures of firm growth) should be ties in a firm’s growth pattern (for a discussion guided by the purpose and theoretical back- of this issue, see Delmar, Davidsson, and ground of the study (Davidsson and Wiklund Gartner 2003; McKelvie and Wiklund 2010), 2000; Murphy, Trailer, and Hill 1996). We argue given that our study encompasses cross- that sales as a measure of growth is particularly sectional data, we considered the use of mean appropriate in the research context of our study, sales growth as more appropriate than the use as the engagement of young and small firms in of absolute yearly sales volume because some business relationships with competitors (either of the independent variables might logically buyers, suppliers, or subcontractors) builds on correlate highly with sales volume (such as size the rationale that such relationships affect firm of the exchange partner), but not with absolute growth by increasing the recognition and sales growth. approval of a firm’s products and services by Our decision to transform the variable was customers and suppliers. Other measures such based on issues of skewness and kurtosis of as profits are, compared with sales, not only the dependent variable, nonnormality, and difficult to obtain for young firms, but the impact heteroskedasticity. Screening continuous vari- of a certain business strategy, such as vertical ables for normality and homoskedasticity is an coopetition, on a firm’s profitability tends to important early step in almost every multivariate become only apparent in the long run (Delmar, analysis. However, an assessment using the raw Davidsson, and Gartner 2003). Although a suc- dependent variable revealed that our analysis cessful vertical coopetitive partnership might seemed to violate these assumptions and there- also improve a young firm’s financial perfor- fore would offer the potential risk of leading to mance (e.g., by lowering a firm’s production less efficient regression estimates. To meet the cost), it may take longer for such processes to assumptions of normality (and the associated materialize in higher profits.4 Similar to previous problem of heteroskedasticity), a commonly rec- research on interorganizational relationships ommended data transformation, particularly in (Stuart 2000) and vertical relationships in par- the case of substantially right skewed data, is the ticular (Petersen, Handfield, and Ragatz 2005), use of logarithmic transformation (see e.g., we expect increases or decreases in sales from Tabachnick and Fidell 2007). Data transforma- coopetitive relationships to occur with a strong tion provides the principal means to correct and more or less immediate effect, and therefore nonnormality and heteroskedasticity. argue for sales growth as the preferential The mathematical justification for transform- measure of firm growth in the context of this ing the dependent variable was the following. study. Both the Levine’s and the Kolomogorov– Third, we chose the natural log-transformed Smirnov test significantly indicated nonnor- average sales growth over three consecutive mality; but after the log-transformation of the years because we understand growth as a variable, there was no longer an issue of process rather than as a snapshot of a certain nonnormality. Before the log-transformation, point in time (Delmar, Davidsson, and Gartner the raw variable was highly right skewed and 2003). Relative growth measures tend to kurtotic; but after the transformation, this accredit impressive growth rates in relative problem was resolved. We also ran two tests for terms to young and small firms because of their heteroskedasticity, the Breusch–Pagan test and initial size (Delmar, Davidsson, and Gartner the Koenker test (the latter being especially 2003; Shepherd and Wiklund 2009). In order to appropriate for smaller sample sizes). Before the avoid such a magnifying effect and following log-transformation, the tests were significant, previous research (e.g., Florin, Lubatkin, and indicating therefore heteroskedasticity, an issue Schulze 2003; McDougall et al. 1994), we mea- that was resolved after the log-transformation. sured sales growth as mean growth rate Therefore, in our final analysis, we used the between two different points in time (Delmar natural log transformation of average sales 1997), using a three-year time span. Although growth as our dependent variable.

4For a discussion of the differences between ﬁrm performance and growth, see for example, Kiviluoto, Brännback, and Carsrud (2011).

874 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Independent Variables reliance on the coopetitor will lead to asym- Competitor Size. With vertical coopetition, we metrical dependence that will harm a firm’s only considered the impact of the partner that position and its growth potential. The per- was perceived by the respondents as the most ceived importance of the exchanged resources important one, neglecting the presence of other by the young and small firms thus indicates coopetitors. Given the particularity of the rela- whether they tend to be overdependent on tionship, we considered the marginal returns of their transaction partners. In the survey, additional partnerships as, at best, diminishing respondents were asked how important the (McFayden and Cannella 2004). The respon- exchange of specific goods or services with the dents were specifically asked to indicate the transaction partner is for the perceived com- size of the main competitor they are in an petitive advantage of the focal firm. We used a exchange relationship with, in comparison with Likert scale from 1 to 5 to indicate the impor- their own size. We asked them to report tance of the products or services. whether their firm is smaller, equally sized, or larger than the competitor’s firm in terms of Vertical Coopetition. We used a dummy vari- total assets. From a social capital perspective, able indicating whether firms were engaged in we expect increasing effects with the relative vertical coopetition or not. We used this variable firm size of the partner and used, therefore, the in a first model with a sample of 83 firms, of ordinal scaled variable for smaller, equally which 53 engaged in vertical coopetition (the sized, and larger coopetitors. results are shown in Table 2 later in this article).

Relative Exchange Value. The relative Control Variables exchange value reflects the intensity of mutual Firm age and number of employees were dependence between the young and small included in the regression analysis as control firms and their most important coopetitor. We variables. Firm age was used as a control operationalized this variable by asking the because a firm’s age might influence its finan- respondents to indicate the percentage of real- cial performance and growth stability. Data on ized exchange value in vertical coopetition in firm age were collected as the number of years terms of their total sales or supplies. Thus, since the firm was founded. We also controlled depending on the nature of the relationship, for the number of employees as a measure of the respondents reported the percentage of firm size. For this variable, we also applied the sales volume from sales realized with a com- natural log transformation, that is, we used the petitor (in firm–buyer relationships), or the per- natural log-transformed variable of the number centage of supply value spent for competitors’ of employees as the control variable. We products or services (in firm-supplier relation- decided to not include industry control vari- ships). A low percentage with the main ables for the following reasons. First, though coopetitor might also generally indicate the industry characteristics as well as firm charac- degree of mutual dependence, even in multiple teristics could matter for the existence of relationships (because all other relationships coopetition, we had previously conducted have lower percentages). Assuming diminish- empirical research (Dowling, Lechner, and Bau ing returns (McFayden and Cannella 2004) and 1998) where we had assumed that these char- a reduced interdependence through multiple acteristics indeed would matter. In order to test sourcing (with lower relation-specific percent- if coopetition was more frequent in different ages), the logic does not change. settings (industry criteria) or across firms (small versus large, national versus interna- Perceived Importance of the Exchange for tional, etc.), we collected specific coopetition the Focal Firm’s Competitive Advantage. We data. Surprisingly, we did not find any differ- argued that if young and small firms perceive ences concerning industry characteristics or that the purchased products from a competitor firm size. The results were throughout insig- are highly important for their competitive nificant: industry characteristics (concentration, advantage and less important for the competi- type of technology) as well as firm character- tor or, vice versa, if supplying a competitor istics (size, degree of internationalization, etc.) with specialized products is critical and neces- did not matter at all. Coopetition simply sary for the young and small firm but less appeared to be a frequent phenomenon in all important for the competing buyer, then over- industries and for all types of firms. Therefore,

LECHNER, SOPPE, AND AND DOWLING DOWLING 75 9 we logically opted against the inclusion of also tested an alternative model in order to industry controls. Second, empirically, in order address the heteroskedasticity issue: we used to preserve degrees of freedom, we did not heteroskedasticity-consistent standard error include a control variable for industry effects in estimators in OLS regression which allows the the model. Particular industry effects on the raw variable to be kept (Hayes and Cai 2007). dependent variable sales growth would statis- The alternative model did not show any signifi- tically mean that, on the one hand, there were cant differences compared with the log- significant mean differences between indus- transformed model (with the latter having the tries and, on the other hand, that the variances advantage that it also addresses issues of kur- within the industry would be much smaller tosis and skewness). Therefore, the model with compared with the whole sample. At first the log-transformed dependent variable glance, this appeared not to be the case. In the appears to be correctly specified since the pat- testing for mean differences, there were no terns of heteroskedasticity disappeared. significant differences between industries. We An alternative method would have been to also tested for variance differences, both with use generalized least squares (GLS) regression the Levene’s test (using means) and the techniques. This method will also correct Brown–Forsythe test (using medians). Both the problem of bias in the standard errors tests were not significant (p = .406 and and will lead to efficient regression estimates. p = .592). We ran a model with all industry However, GLS regression requires more strin- dummies on the dependent variable, but the gent assumptions about the error term distribu- model was not significant. tion and some knowledge about the functional We also tried to reduce the number of form of the heteroskedasticity. If transforma- dummy variables to be used in the full model. tion of some variables can reduce the problem There was one industry with an average mean significantly, then this is commonly the pre- growth that was slightly lower than the ferred option. average. We included this dummy variable in Given the small sample size, we proceeded the full model, assuming that this variable cap- to test the robustness of our findings with an tured the only significant effect and, on the alternative approach. As mentioned, in a pre- other hand, absorbed all the other (nonsignifi- vious coopetition study (Dowling, Lechner, cant) effects (through the zero-coding). We and Bau 1998), we had collected also finally constructed another single industry coopetition data. To see if our results would dummies (high versus low growth) as well as a fundamentally change, we used the previously variable with the mean sales growth per indus- collected data, picked all small and medium- try and included them in three additional sized firms in the sample (the sample consisted research models (including the other research of firms of all sizes from a variety of indus- variables), but the results did not fundamen- tries), and added them to the original sample. tally change. Therefore, we opted for the more The small and medium-sized enterprises parsimonious model without industry controls. (SMEs) from the old sample were not VC-backed. We then ran the regressions and Model Specifications compared them with the original sample. We tested our hypotheses using hierarchical Although the explanation power of the merged ordinary least squares (OLS) regression. In step sample is weaker (not surprisingly, given that one, we tested the influence of control vari- two heterogeneous samples from different time ables, that is, firm age and size (employees), on periods were merged and the original sample sales growth with the sample of the 53 firms was rather homogeneous for starting condi- engaged in vertical coopetition. In step two, we tions), the fundamental message of the analysis entered the independent variables “size of did not change. All three hypotheses would be the transaction partner” and “realized relative confirmed by this merged sample. The com- exchange value with the transaction partner” parison of the results, however, shows the (as a measure of mutual dependence) and “per- robustness of our findings. ceived importance of the exchange for the focal firm’s competitive advantage.” We ran a series Results of robustness checks for our specified model. Descriptive Results First, though we have reasonable justification Table 1 presents the results of the descrip- to log-transform the dependent variable, we tive analysis in terms of means, standard

1076 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Table 1 Descriptive Statistics and Variable Correlationsa

Variable Mean S.D. 1 2 3 4 5

1. Average sales growth (log) 13.41 1.31 2. Firm age 8.11 3.43 0.26* 3. Firm size (log) 3.23 1.42 0.52*** 0.37** 4. Competitor size 1.59 0.65 0.41*** 0.18 0.19 5. Relative exchange value 15.47 15.61 0.25* −0.31* −0.16 0.08 6. Importance of exchanged resources 3.15 1.44 −0.28* −0.13 −0.06 −0.21 0.05 aReported for all firms that are engaged in vertical coopetition (n = 53). *p < .05. **p < .01. ***p < .001. deviations, and correlations for the dependent as well as all independent variables for the Table 2 restricted sample of 53 firms. To control for OLS Regression: Vertical multicollinearity, we calculated variance infla- Coopetition and Small tion factors (VIF) for the variables used in each Firm Growth regression model (not shown). All VIF scores were less than 1.3, meeting neither the critical Variablea Model 1 Model 2 value of 10 suggested by Neter, Wasserman, and Kutner (1989) nor the more stringent value of 2 suggested by Cohen et al. (2003). Thus, (Constant) multicollinearity does not appear to influence Control variables our results. Firm age 0.135 0.101 Table 2 shows the models for the full sample Firm size (log) [number 0.479*** 0.455*** of 83 firms with vertical coopetition as explana- of employees] tory variable. Coopetition literature suggests Independent variable that vertical coopetition would have a positive Vertical coopetition 0.176 2 impact on firm growth. As our results show, the R 0.306 0.334 2 relationship is not significant (p < .1), in line Adjusted R 0.288 0.308 with our expectations that coopetition alone Significance model 0.000 0.000 2 would not appropriately explain differences in Significance change R — 0.072 sales growth. We expected that the character- df 2/79 3/78 istics of the relationships would matter more n =82 n =82 and were therefore interested in what drives sales growth differentials among firms engaged aDependent variable: average sales growth in vertical coopetition. (log). *p < .05. Hypotheses Tests **p < .01. We report here the steps of the hierarchical ***p < .001. OLS regression analysis. Table 3 shows the OLS, ordinary least squares. results for the restricted sample containing only young and small firms engaged in vertical coopetition. The first model includes only the with the explanatory variables is significant control variables; the second model includes with a relatively high adjusted R2 of 0.50. The the independent variables “competitor size” change in the F-statistic is also significant, and “relative exchange volume.” The model which means that the introduced research

LECHNER, SOPPE, AND AND DOWLING DOWLING 77 11 Table 3 established competitor would have a positive OLS Regression: Characteristics impact on the sales growth of young and small firms. Our findings show a positive impact of of Vertical Coopetition and the size of the competitor with whom the focal Small Firm Growth firm is engaged on sales growth. Thus, this result indicates that vertical coopetition with Variablea Model 1 Model 2 larger competitors is increasingly beneficial for sales growth, which confirms our first hypoth- (Constant) esis. H2 analyzed the relative exchange value Control variables as a measure of mutual dependence in vertical Firm age 0.078 0.164 coopetition. The relationship between relative Firm size (log) 0.489*** 0.456*** exchange value and sales growth is positive [number of and significant. Thus, we find support for H2. employees] H3 predicted a negative relationship between Independent variables the perceived importance of the exchanged Competitor size 0.274** resources in vertical coopetition and sales Relative exchange 0.399*** growth. Indeed, we found a significant negative value relationship. H3 can therefore also be Importance of −0.216* confirmed. exchanged The findings from our analysis suggest resources that all our hypotheses concerning vertical R2 0.273 0.548 coopetition were confirmed. Engaging in verti- Adjusted R2 0.244 0.500 cal coopetition is generally not beneficial for Significance model 0.000 0.000 the sales growth of small and young firms; a Significance of R2 — 0.000 more fine-grained understanding is necessary change as some characterizing attributes of vertical n =53 n =53 coopetition are positive, whereas others, such as overdependence, can lead to negative outcomes for young and small firms. aDependent variable: average sales growth (log). Discussion Although there is empirical evidence of the *p < .05. phenomenon of vertical coopetition occurring **p < .01. in established as well as in young and small ***p < .001. firms, most research on vertical coopetition to OLS, ordinary least squares. date has focused on large, established firms (e.g., Bengtsson and Kock 2000; Bonel and variables added prediction power. The sample Rocco 2007; Gnyawali et al. 2006). Motivated size appears to be sufficient. The variable ratios by this phenomenon of practical and theoreti- are higher than the absolute acceptable cal importance, we attempted to disentangle minimum level of 5 (Tabachnick and Fidell some important benefits and threats associated 2007), or even 10, but what matters most is the with vertical coopetition for young and small statistical power (Green 1991). Despite the firms, and empirically tested different influenc- small sample size (n = 53), the statistical power ing attributes of vertical coopetition on sales of the full model (Faul et al. 2007) is greater growth. We adopted a rational choice position, than 0.99.5 that is, firms can decide to enter into vertical With respect to the hypotheses,6 H1 pre- coopetition or renounce it. The current study dicted that cooperating with a larger, more suggests that depending on accurate attributes

5Alternatively, we estimated the minimum sample size needed with the actual effect sizes. The result indicated a sample size below the actual sample size. 6Because our dependent variable is log-transformed, the results need to be interpreted accordingly. Unit— changes in the untransformed independent variables lead to (100 ×β) percent—changes in the (log- transformed) dependent variable; percent—changes in the log-transformed independent variable (in our case, the number of employees as a measure for ﬁrm size) lead to percent—changes in the dependent variable.

1278 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT of vertical coopetition, such as the size of the understand vertical coopetition as an important competitor and the nature of dependence source of social capital. Competitors possess among the cooperating rivals, interfirm rela- highly relevant resources and coopetition can be tionships with direct competitors can have an effective means to access these; coopetition is ambiguous effects on young firms. therefore an important strategy for young and small firms to overcome challenges related to Theoretical Implications resource acquisition. The primary purpose of this paper was to By investigating the nature of dependence advance our understanding of how vertical among firms, we found that interdependence coopetition is related to sales growth. We inves- (expressed as relatively high exchange volume tigated why and under what circumstances in vertical coopetition) positively affects vertical coopetition—understood as buyer, sup- growth. Relatively high exchange volume with plier, or subcontractor relationships with direct the competing transaction partner positively competitors—can influence the sales growth of influences the growth prospects of young and young and small firms. We drew on resource- small firms. This insight is not trivial because based and dependence theory to understand young and small firms might be tempted to why and how vertical coopetition can posi- gain advantages through the opportunity of tively or negatively affect growth. We argued vertical coopetition and, at the same time, try to that young and small firms do benefit from reduce the perceived risk by lowering stakes, vertical coopetition and proposed these inter- engagement, and interdependence (McCarter firm relationships as a form of social capital and Northcraft 2007). However, this behavior (Lin 2002; Payne et al. 2011), with competitors would create, rather, a situation that triggers enabling access to necessary resources (Ahuja conflict and opportunism (Kohtamäki et al. 2000; Zacharia, Nix, and Lusch 2009). Vertical 2012). coopetition can be considered as a source of There is, however, a fine line between posi- value creation (Dyer and Singh 1998) as com- tive mutual dependence and (asymmetrical) petitors dispose of highly relevant and comple- overdependence for the competitiveness of the mentary resources (Gnyawali and Park 2009). young and small firm. If a young and small firm However, relationships with direct rivals are depends too strongly on its supplier, buyer, or also considered to be highly risky; they hold subcontractor, then this situation creates all threats, especially for young and small firms potential costs and risks inherent in vertical (Chung 2012). Thus, a deeper investigation of coopetition (Chung 2012). The descriptive sta- the ambiguous effects of vertical coopetition on tistics also show some correlation between the entrepreneurial firm growth is necessary. We degree of dependence and overdependence. It analyzed, therefore, how exchange partners is thus important for young and small firms to as well as constituencies of the exchange rela- build competitive advantages on the base of tionship can influence firm growth. We argued their own resources and capabilities, and to that a larger competitor as business partner and complement and augment this position through mutual dependence among the competitors complementary resources and learning oppor- are positively related to growth until young tunities. These findings also have important and small firm dependence becomes too implications for research on social capital, and asymmetrical. contribute particularly to the “dark side” of We found that young firms can benefit from social capital (Gargiulo and Benassi 2000). unbalanced vertical coopetition, that is, young Although coopetition can be considered an and small firms gain advantages by engaging in important source of social capital, coopetitive vertical coopetition with relatively larger com- relational resources seem to be dynamic and petitors. These findings provide evidence to fragile. Our findings indicate that beneficial support the theory that vertical coopetitive rela- gains through social capital seem to be strongly tionships with larger competitors lead to more dependent on the degree of dependence among social capital than relationships with similar the competitors and require careful manage- sized or smaller competitors. Our findings ment in terms of both the choice of business suggest that signaling and reputation effects are partner and the specific goods or services that greater if young and small firms are engaging are exchanged. All in all, our findings suggest with larger firms and that these effects positively that future studies on vertical coopetition and influence sales growth. Thus, we propose to the phenomenon of coopetition should not only

LECHNER, SOPPE, AND AND DOWLING DOWLING 79 13 examine the effects of vertical coopetition or complementarity would provide a better under- coopetition as a “holistic” concept, but also standing of the relationship between vertical investigate the underlying characteristics and coopetition and firm growth. Lastly, our study specific attributes of such transaction relation- focused on the coopetitive relationship with the ships and transaction partners. We encourage most important partner; future research should future research to refine our understanding of investigate multiple coopetition and study if, as the impact of vertical coopetition on young and suggested, diminishing returns or even nega- small firm performance and the diverse charac- tive effects set in (McFayden and Cannella teristics of vertical coopetition. 2004).

Practical Implications Conclusions and For entrepreneurs and managers, the find- Future Research ings of our study suggest that exchange rela- Drawing on insights from resource-based tionships with direct competitors can be perspectives as well as dependence theory, our beneficial for young and small firms if they model proposes that vertical coopetition can enter and design vertical coopetition carefully. either stimulate or hinder growth, depending For example, choosing an established and on important attributes of the relationship. larger exchange partner appears to foster sales Being a mix between noncompeting, interfirm growth. Therefore, choosing a business partner relationships and competition, young and small may involve more than the resources at hand. firms in particular are most likely subject to In addition, young and small firms might have positive or negative dependencies in vertical an interest in increasing the exchange value coopetition. Our findings help us to better with both competing buyers and suppliers, understand the complex nature of vertical without basing the competitive fate on the com- coopetition and its beneficial or costly conse- petitor. Exploiting resource complementarities quences for young and small firms. The logic appears to be important for young and small of advantages and disadvantages is not the firms; however, an important challenge for same for all exchange relationships. It rather young firms is to balance their efforts at depends on the size of the competitors and the strengthening the relationship with their busi- intensity and form of dependence within the ness partners, while avoiding a situation of exchange relationship. overdependence. Our results may help young In conclusion, our study offers evidence that and small firms to be aware of the pervasive the effects of vertical coopetition are not uni- influence of vertical coopetition on firm lateral and simple, and are subject to a series of growth, and improve the management of such contingencies. Future research should address relationships. more of these contingencies. Furthermore, we did not investigate motives for young firms Limitations taking part in vertical coopetition, nor the best The limitations of our study must be taken organizational or governance forms for dealing into account when evaluating our results. First, with vertical coopetition. Given the risks of we tested our hypotheses with a relatively vertical coopetition, investigating the reasons small sample and a relatively low but accept- why young and small firms enter vertical able response rate. It would thus be desirable coopetition and how they manage various to replicate our study in different settings and forms of vertical coopetition would be another with larger sample sizes. Second, though the fruitful stream of future research to better financial and demographic firm data were understand this specific form of interfirm rela- matched with nonbiased and objective sources, tionships. With regard to our theorizing of ver- other measures of this study were self-reported tical coopetition as an important source of by a single respondent. We controlled for social capital, future work could build on this common method bias and did not find any bias. study and explore this research area in more Third, our arguments are strongly tied to depth. For instance, an interesting line of future resource-based logic and to the concept of research would be to apply a contingency per- resource complementarity. In our hypotheses spective to better understand the conditions and models, resource complementarity remains under which coopetitive relationships present a a latent variable. The formal inclusion in source of social capital and enhance firm future models of variables estimating resource growth, and under which they are more a

1480 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT liability than an asset for young and small firms. ——— (1991). “Firm Resources and Sustained Another interesting area would be to examine Competitive Advantage,” Journal of Man- more explicitly the direct effect of vertical agement 17(1), 99–120. coopetition on young firms’ reputation and Baum, J. (1996). “Organizational Ecology,” in legitimacy gains, or their ability to access rel- Handbook of Organization Studies. Eds. S. evant information and financial capital. Clegg, C. Hardy, and S. Nord. London: Sage, Our study’s goal was to advance research on 77–114. vertical coopetition and to show under which Baum, J., T. Calabrese, and B. Silverman (2000). circumstances this configuration can create “Don’t Go It Alone: Alliance Network Com- value for young and small firms. Implications position and Startups’ Performance in Cana- can be derived concerning how and under what dian Biotechnology,” Strategic Management circumstances firms can receive competitive Journal 21(3), 267–294. advantage from vertical coopetition by minimiz- Beekman, A., and R. Robinson (2004). “Sup- ing their disadvantages and costs. This study plier Partnerships and the Small, High- offers some interesting insights into the perva- Growth Firm: Selecting for Success,” Journal sive nature of vertical coopetition for young and of Small Business Management 42(1), 59–77. small firms. We hope that our empirical study Beier, F., and L. Stem (1969). “Power in Distri- also motivates other researchers to further inves- bution Channels,” in Distribution Channels: tigate the ambiguous impacts of vertical Behavioral Dimensions. Ed. L. Stem. Boston: coopetition for theory and practice. Houghton Mifflin Co., 92–116. Bengtsson, M., and S. Kock (2000). References “Coopetition in Business Networks—To Adams, J., F. Khoja, and R. Kauffman (2012). Cooperate and Compete Simultaneously,” “An Empirical Study of Buyer–Supplier Industrial Marketing Management 29(5), Relationships within Small Business Organi- 411–426. zations,” Journal of Small Business Manage- Bonel, E., and E. Rocco (2007). “Coopeting to ment 50(1), 20–40. Survive; Surviving Coopetition,” Interna- Adegbesan, A. (2009). “On the Origins of Com- tional Studies of Management and Organi- petitive Advantage: Strategic Factor Markets zation 37(2), 70–96. and Heterogeneous Resource Complemen- Brush, C., and P. Vanderwerf (1992). “A Com- tarity,” Academy of Management Review parison of Methods and Sources for Obtain- 34(3), 463–475. ing Estimates of New Venture Performance,” Ahuja, G. (2000). “The Duality of Collaboration: Journal of Business Venturing 7(2), 157– Inducements and Opportunities in the For- 170. mation of Interfirm Linkages,” Strategic Carsrud, A., and M. Brännback (2007). Entre- Management Journal 21(3), 317–343. preneurship. Westport, CT: Greenwood. Alvarez, S. A., and J. B. Barney (2001). “How Chandler, G., and S. Hanks (1993). “Measuring Entrepreneurial Firms Can Benefit from Alli- the Performance of Emerging Businesses: A ances with Large Partners,” Academy of Validation Study,” Journal of Business Ven- Management Executive 15(1), 139–148. turing 8(5), 391–408. ——— (2004). “Organizing Rent Generation Chung, J.-E. (2012). “When and How Does Sup- and Appropriation: Toward a Theory of the plier Opportunism Matter for Small Retailers’ Entrepreneurial Firm,” Journal of Business Channel Relationships with the Suppliers?,” Venturing 19(5), 621–635. Journal of Small Business Management Amit, R., and P. Schoemaker (1993). “Strategic 50(3), 389–407. Assets and Organizational Rent,” Strategic Cohen, J., P. Cohen, S. G. West, and L. S. Aiken Management Journal 14(1), 33–46. (2003). Applied Multiple Regression/ Anderson, E., and B. Weitz (1989). “Determi- Correlation Analysis for the Behavioral Sci- nants of Continuity in Conventional Indus- ences. Mahwah, NJ: Lawrence Erlbaum. trial Channel Dyads,” Marketing Science Cope, J. (2005). “Toward a Dynamic Learning 8(4), 310–323. Perspective of Entrepreneurship,” Entrepre- Barney, J. (1986). “Types of Competition and neurship Theory and Practice 29(4), 373– the Theory of Strategy: Toward an Integra- 397. tive Framework,” Academy of Management Crook, T., and J. Combs (2007). “Sources and Review 11(4), 791–800. Consequences of Bargaining Power in

LECHNER, SOPPE, AND AND DOWLING DOWLING 81 15 Supply Chains,” Journal of Operations Man- Frazier, G. (1983). “On the Measurement of agement 25(2), 546–555. Interfirm Power in Channels of Distribu- Dagnino, G. (2009). Coopetition Strategy: tion,” Journal of Marketing Research 20(2), Theory, Experiments, Cases. London: 158–166. Routledge. Gargiulo, M., and M. Benassi (2000). “Trapped Davidsson, P., and J. Wiklund (2000). “Concep- in Your Own Net? Network Cohesion, Struc- tual and Empirical Cahllenges in the Study of tural Holes, and the Adaptation of Social Firm Growth,” in Handbook of Entrepre- Capital,” Organization Science 11, 183– neurship. Eds. D. Sexton and H. Landstrom. 196. Oxford: Blackwell, 26–44. Gilbert, B., P. McDougall, and D. Audretsch Delmar, F. (1997). “Measuring Growth: Meth- (2006). “New Venture Growth: A Review and odological Considerations and Empirical Extension,” Journal of Management 32(6), Results,” in Entrepreneurship and SME 926–950. Research: On Its Way to the Next Millen- Gnyawali, D., and R. Madhavan (2001). nium. Eds. R. Donckels and A. Miettinen. “Cooperative Networks and Competitive Surrey, UK: Ashgate Publishing, 199–216. Dynamics: A Structural Embeddedness Per- Delmar, F., P. Davidsson, and W. Gartner spective,” Academy of Management Review (2003). “Arriving at the High-Growth Firm,” 26(3), 431–445. Journal of Business Venturing 18(2), 189– Gnyawali, D., and B. Park, Jr. (2009). “Co- 216. opetition and Technological Innovation in Deutsch, M. (1969). “Conflicts: Productive and Small and Medium-Sized Enterprises: A Mul- Destructive,” The Journal of Social Issues tilevel Conceptual Model,” Journal of Small 25(1), 7–41. Business Management 47(3), 308–330. Dowling, M., and C. Lechner (1998). Gnyawali, D., J. He, and R. Madhavan (2006). “Kooperative Wettbewerbsbeziehungen: “Impact of Co-opetition on Firm Competitive Theoretische Ansätze und Management- Behavior: An Empirical Examination,” strategien,” Die Betriebswirtschaft 58(1), Journal of Management 32(4), 507–530. 86–102. Green, S. (1991). “How Many Subjects Does It Dowling, M., D. Roering, B. Carlin, and J. Take to Do a Regression Analysis?,” Multi- Wisnieski (1996). “Multifaceted Relation- variate Behavioral Research 26(3), 499–510. ships under Coopetition: Description and Hayes, A., and L. Cai (2007). “Using Theory,” Journal of Management Inquiry Heteroskedasticity-Consistent Standard Error 5(2), 155–167. Estimators in OLS Regression: An Introduc- Dowling, M., C. Lechner, and F. Bau (1998). tion and Software Implementation,” Behav- “Kooperative Wettbewerbsbeziehungen— ior Research Methods 39(4), 709–722. Eine Empirische Untersuchung,” in Heide, J., and G. John (1988). “The Role of Innovationsforschung und Dependence Balancing in Safeguarding Technologiemanagement. Eds. N. Franke Transaction-Specific Assets in Conventional and C.-F. von Braun. Berlin: Springer, 165– Channels,” Journal of Marketing 52(1), 174. 20–35. Dyer, J., and H. Singh (1998). “The Relational Hoy, F., P. McDougall, and D. Dsouza View: Cooperative Strategy and Sources of (1992). “Strategies and Environments of Interorganizational Competitive Advantage,” High Growth Firms,” in The State of the Art Academy of Management Review 23(4), 660– of Entrepreneurship. Eds. D. Sexton and 679. J. D. Kasarda. Boston: PWS-Kent, 341– Faul, F., E. Erfelder, A. Lang, and A. Buchner 357. (2007). “G*Power 3: A Flexible, Statistical Ketchen, D., C. Snow, and V. Hoover (2004). Power Analysis Program for the Social, “Research on Competitive Dynamics: Recent Behavioral and Biomedical Sciences,” Accomplishments and Future Challenges,” Behavior Research Methods 39(2), 175– Journal of Management 30(6), 779–804. 191. Ketchen, D., D. Ireland, and C. Snow (2007). Florin, J., M. Lubatkin, and W. Schulze (2003). “Strategic Entrepreneurship, Collaborative “A Social Capital Model of High-Growth Innovation, and Wealth Creation,” Strategic Ventures,” Academy of Management Journal Entrepreneurship Journal 1(3–4), 371– 46(3), 374–384. 385.

1682 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Kiviluoto, N., M. Brännback, and A. Carsrud McDougall, P., J. Covin, R. Robinson, and L. (2011). “Are Firm Growth and Performance Herron (1994). “The Effects of Industry the Same or Different Concepts in Entrepre- Growth and Strategic Breadth on New neurship Studies ? An Analysis of the Depen- Venture Performance and Strategy Content,” dent and Independent Variables,” in Strategic Management Journal 15(7), 537– Entrepreneurship, Growth and Economic 554. Development. Eds. M. Raposo, D. Smallbone, McFayden, A., and A. Cannella (2004). “Social K. Balaton, and L. Hortovanyi. Cheltenham: Capital and Knowledge Creation: Diminish- Edward Elgar, 11–29. ing Returns of the Number and Strength of Kohtamäki, M., J. Vesalainen, S. Henneberg, P. Exchange Relationships,” Academy of Man- Naudé, and M. Ventresca (2012). “Enabling agement Journal 47(5), 735–746. Relationship Structures and Relationship Per- McKelvie, A., and J. Wiklund (2010). “Advanc- formance Improvement: The Moderating ing Firm Growth Research: A Focus on Role of Relational Capital,” Industrial Mar- Growth Mode Instead of Growth Rate,” keting Management 41, 1298–1309. http:// Entrepreneurship Theory and Practice 34(2), dx.doi.org/10.1016/j.indmarman.2012.08.001 261–288. Kumar, N., L. Scheer, and J. Steenkamp (1995). Miller, L., and K. Smith (1983). “Handling “The Effects of Perceived Interdependence Nonresponse Issues,” Journal of Extension on Dealer Attitudes,” Journal of Marketing 21(5), 45–50. Research 32(3), 348–356. Molm, L. (1989). “Punishment Power: A Balanc- Lado, A., N. Boyd, and S. Hanlon (1997). ing Process in Power-Dependence Rela- “Competition, Cooperation and the Search tions,” The American Journal of Sociology for Economic Rents: A Syncretic Model,” 94(6), 1392–1418. Academy of Management Review 22(1), 110– Murphy, G., J. Trailer, and R. Hill (1996). “Mea- 141. suring Performance in Entrepreneurship Lawler, E., R. Ford, and M. Blegen (1988). Research,” Journal of Business Research “Coercive Capability in Conflict: A Test of 36(1), 15–23. Bilateral Deterrence versus Conflict Spiral Neter, J., W. Wasserman, and M. H. Kutner Theory,” Social Psychology Quarterly 51(2), (1989). Applied Linear Regression Models. 93–107. Homewood, IL: Irwin. Lebrasseur, R., L. Zanibbi, and T. Zinger (2003). Nicholls-Nixon, C., and C. Woo (2003). “Tech- “Growth Momentum in Early Stages of Small nology Sourcing and Output of Established Business Start-ups,” International Small Firms in a Regime of Encompassing Techno- Business Journal 21(3), 315–330. logical Change,” Strategic Management Lechner, C., and M. Dowling (2003). “Firm Net- Journal 24(5), 651–666. works: External Relationships as Sources for Padula, G., and G. B. Dagnino (2007). “Untan- the Growth and Competitiveness of Entre- gling the Rise of Coopetition,” International preneurial Firms,” Entrepreneurship & Studies of Management and Organization Regional Development 15(1), 1–26. 37(2), 32–52. Lechner, C., M. J. Dowling, and I. Welpe (2006). Park, S., and M. V. Russo (1996). “When Com- “Firm Networks and Firm Development: The petition Eclipses Cooperation: An Event Role of the Relational Mix,” Journal of Busi- History Analysis of Joint Venture Failure,” ness Venturing 21(4), 514–540. Management Science 42(6), 875–890. Lin, N. (2002). Social Capital: A Theory of Social Paul, S., and R. Boden (2011). “Size Matters: Structure and Action. Cambridge: Cam- The Late Payment Problem,” Journal of bridge University Press. Small Business and Enterprise Development Lindell, M., and D. Whitney (2001). “Account- 18(4), 732–747. ing for Common Method Variance in Cross- Payne, T., C. Moore, S. Griffis, and C. Autry Sectional Research Designs,” The Journal of (2011). “Multilevel Challenges and Opportu- Applied Psychology 86(1), 114–121. nities in Social Capital Research,” Journal of McCarter, M., and G. Northcraft (2007). “Happy Management 37(2), 491–520. Together? Insights and Implications of Peteraf, M. A. (1993). “The Cornerstones of Viewing Managed Supply Chain as Social Competitive Advantage: A Resource-Based Dilemma,” Journal of Operations Manage- View,” Strategic Management Journal 14(3), ment 25, 498–511. 179–191.

LECHNER, SOPPE, AND AND DOWLING DOWLING 83 17 Petersen, K., R. Handfield, and G. Ragatz (2005). Growth and Innovation Rates in a High- “Supplier Integration into New Product Technology Industry,” Strategic Manage- Development: Coordinating Product, Process ment Journal 21(8), 791–811. and Supply Chain Design,” Journal of Opera- Stuart, T., H. Hoang, and R. Hybels (1999). tions Management 23, 371–388. “Interorganizational Endorsements and the Podsakoff, P., and D. Organ (1986). “Self- Performance of Entrepreneurial Ventures,” Reports in Organizational Research: Prob- Administrative Science Quarterly 44(2), lems and Prospects,” Journal of Management 315–349. 12(4), 531–544. Tabachnick, B., and L. Fidell (2007). Using Mul- Priem, R. (2007). “A Consumer Perspective on tivariate Statistics. Boston: Pearson. Value Creation,” Academy of Management Terpend, R., B. Tyler, D. Krause, and R. Review 32(1), 219–235. Handfield (2008). “Buyer–Supplier Relation- Quintana-García, C., and C. Benavides-Velasco ships: Derived Value over Two Decades,” (2004). “Cooperation, Competition, and Journal of Supply Chain Management 44(2), Innovative Capability: A Panel Data of Euro- 28–55. pean Dedicated Biotechnology Firms,” Townsend, D., and L. Busenitz (2008). “Factor Technovation 24(12), 927–938. Payments, Resource-Based Bargaining, and Rivkin, J., and N. Siggelkow (2007). “Patterned the Creation of Firm Wealth in Technology- Interaction in Complex Systems: Implica- Based Ventures,” Strategic Entrepreneurship tions for Exploration,” Management Science Journal 2(4), 339–355. 53(7), 1068–1085. Venkataraman, S., and A. Van de Ven (1998). Robinson, K. (1999). “An Examination of the “Hostile Environmental Jolts, Transaction Influence of Industry Structure on Eight Set, and New Business,” Journal of Business Alternative Measures of New Venture Perfor- Venturing 13(3), 231–255. mance for High Potential Independent New Weigelt, C. (2013). “Leveraging Supplier Capa- Ventures,” Journal of Business Venturing bilities: The Role of Locus of Capability 14(2), 165–187. Development,” Strategic Management Ross, W., and D. Robertson (2007). “Compound Journal 34, 1–21. doi:10.1002/smj.1998. Relationships Between Firms,” Journal of Weinzimmer, L., P. Nystrom, and S. Freeman Marketing 71(7), 108–123. (1998). “Measuring Organizational Growth: Shah, S., and M. Tripsas (2007). “The Acciden- Issues, Consequences and Guidelines,” tal Entrepreneur: The Emergent and Collec- Journal of Management 24(2), 235– tive Process of User Entrepreneurship,” 262. Strategic Entrepreneurship Journal 1(1–2), Wernerfelt, B. (2011). “The Use of Resources in 123–140. Resource Acquisitions,” Journal of Manage- Shepherd, D., and J. Wiklund (2009). “Are We ment 37(5), 1369–1373. Comparing Apples With Apples or Apples Wu, W. (2008). “Dimensions of Social Capital With Oranges? Appropriateness of Knowl- and Firm Competitiveness Improvement: edge Accumulation Across Growth Studies,” The Mediating Role of Information Sharing,” Entrepreneurship Theory and Practice 33(1), Journal of Management Studies 45(1), 122– 105–123. 146. Shepherd, D., J. McMullen, and P. Jennings Wyld, J., G. Pugh, and D. Tyrrall (2012). “Can (2007). “The Formation of Opportunity Powerful Buyers ‘Exploit’ SME Suppliers?,” Beliefs: Overcoming Ignorance and Reduc- Journal of Small Business and Enterprise ing Doubt,” Strategic Entrepreneurship Development 19(2), 322–334. Journal 1(1–2), 75–95. Zacharia, Z., N. Nix, and R. Lusch (2009). “An Stinchcombe, A. (1965). “Social Structure and Analysis of Supply Chain Collaborations and Organizations,” in Handbook of Organiza- Their Effect on Performance Outcomes,” tion. Ed. J. March. Chicago, IL: Rand Journal of Business Logistics 30(2), 101– McNally, 142–193. 124. Stuart, T. (2000). “Interorganizational Alliances and the Performance of Firms: A Study of

1884 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Journal of Small Business Business Management Management 2016 2014 54(1), ••(••), pp. pp. 85–101 ••–•• doi: 10.1111/jsbm.12132 Examining Entrepreneurs’ Presentation Effectiveness in Generating Stakeholder Interest: Observations From the Medical Device Industry by Scott Latham and Steven F. Tello

Entrepreneurs make countless pitches and presentations to prospective stakeholders that are critical to securing needed resources. Our study examines whether or not certain aspects of visual presentations are more likely to elicit stakeholder interest. By applying precepts of information visualization theory, we empirically examine 42 early-stage venture poster presentations to determine the impact design and content have on attracting stakeholder interest. Our ﬁndings suggest that visualization matters; posters that included speciﬁc visual devices garnered a higher level of interest among prospective stakeholders, including investors and collaborators.

presentations run the gamut of formats and Introduction settings. For example, an entrepreneur’s pre- When starting a new venture, entrepreneurs sentation might involve a chance encounter are challenged with gathering the various between the entrepreneur and a potential inves- resources required to move their venture tor at an airport lounge, or it might involve a forward. In order to gather needed resources, formal, scheduled presentation at a venture entrepreneurs reach out to various stakeholders capitalist’s (VC) ofﬁce. Although the nature of —potential investors, partners, and collabora- the presentation varies, the motivation and tors who have knowledge and skills critical to objective are similar. In each instance, the entre- the new venture’s emergence and success. preneur is trying to persuade a potential stake- However, entrepreneurs do not do this in iso- holder that the venture, and all its various lation, as Chen, Yao, and Kotha (2009) detail, components, have merit and warrant further entrepreneurs are tasked with the job of per- investigation (Clark 2008; Gregoire, de Koning, suasion; that is, among a crowded sea of entre- and Oviatt 2008). Thus, presentation is critical preneurs all seeking resources, they must in early-stage entrepreneurial ventures, yet, as convince potential stakeholders of the merit of scholars have put forth, many questions remain their venture. Much of this persuasion occurs as to the underlying dynamics regarding entre- during the numerous presentations that entrepreneurial presentation effectiveness (see e.g., preneurs deliver. As past research suggests, Mason and Harrison 2003).

Scott Latham is associate professor and department chair of Management and Strategy, Manning School of Business, University of Massachusetts Lowell. Steven F. Tello is associate vice chancellor of Entrepreneurship and Economic Development at the University of Massachusetts Lowell. Address correspondence to: Steven F. Tello, Management & Entrepreneurship, University of Massachusetts Lowell, One University Avenue, Lowell, MA 01854. E-mail: [email protected].

LATHAM AND TELLO TELLO 85 1 Our study sought to determine whether or contributes to the ability of visual posters to not visual devices used in the entrepreneur’s create interest. By examining information visu- presentation help an entrepreneur to stand out alization, we seek to ascertain whether or not among a crowded sea of competitors, thus certain visual devices play a role in influencing potentially increasing stakeholder interest in potential stakeholder interest. Such a theoreti- the venture. Several researchers have exam- cal bridge is appropriate, as potential investors ined the impact of entrepreneurs’ “live” pre- and partners in entrepreneurial ventures are sentation skills in relation to their ability to similarly bombarded with a myriad of informa- build interest in their venture (Baron and tion, presented in multiple formats across Brush 1999; Chen, Yao, and Kotha 2009; Clark numerous ventures. 2008; Gregoire, de Koning, and Oviatt 2008). Unlike past examinations of the informa- Their findings indicate that the entrepreneur’s tion visualization tenets, we examine various level of preparedness for the presentation and aspects of information visualization in a their social communication skills are positively context that is far removed from experimental related to stakeholder interest. Although the settings. More specifically, we undertake a role that the entrepreneur plays in the process contextualization of information visualization. has been investigated, one aspect that requires Theory contextualization calls for the “proving” more scholarly attention is the nature of the and “breaking” of widely accepted theoretical actual presentation. Undoubtedly, entrepre- tenets in contexts that are diverse and appro- neurs are armed with countless visual devices, priate. Bamberger (2008) makes the case that such as business cards, presentations, posters, greater theoretical advances might be made slide decks, overheads, brochures, and other by contextualizing theory. He states: “Context materials that are meant to garner attention theorizing requires a researcher to build situ- and generate interest. However, few studies ational and/or temporal conditions directly into have examined the role that these visual theory and, just as importantly, to explicate the devices play in affecting potential stakeholder mechanisms either linking these situational interest. Clarke (2011) commented on the poor and temporal conditions to embedded phenom- understanding of the ways in which entrepre- ena, or governing the conditioning of relation- neurs use visual devices to influence resource ships between phenomena by these situational providers, stating that “Entrepreneurs commu- and temporal conditions” (Bamberger 2008, p. nicate to resource providers using a combina- 841). We concur with Bamberger’s assessment tion of language of visual cues, and future that contextualization can enhance theoretical research in this area should be designed to investigation, and suggest that our study ensure that the visual aspect of meaning- enhances the understanding of information making can be examined” (p. 1384). Our study visualization through its application within a aligns with this research call, specifically exam- unique context. ining the comparative effectiveness of various To test this presumption, we analyzed the types of visual devices in entrepreneurial visual posters presented by 42 entrepreneurs at presentations. a major inventor showcase event. Such events In assessing the outcome variance associated are common in technical and scientific fields, with visual devices in entrepreneurs’ presenta- such as biotechnology, nanomanufacturing, tions, we draw on the field of information visu- and medical devices. At these events, entrepre- alization. Previous studies have established that neurs display a three-by-four foot poster that information visualization plays a critical role in contains technical, scientific, and market infor- fields where complex data, concepts, and mation about their inventions (see Figure 1 for dynamics are being related (Johansson et al. a poster example). 2006; Spence 2007; Ware 2004). Tufte (1997) The primary reason for participating in these explains in great detail how the visual display events is to attract investment funds that will of information in various formats can be used support the growth of the venture. Inventors to either share and express multiple levels of attend the conference in the hope of attracting information, or hide or obfuscate important investors, as well as other stakeholders details. We consider how preattentive process- that might provide tangible and intangible ing, the manner in which individuals subcon- resources. Potential stakeholders walk the floor sciously focus on specific visual information reviewing the posters, often without speaking (Janiszewski 1998; Spence 2007; Ware 2004), to the entrepreneur. After reviewing the poster,

286 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Figure 1 Example of Entrepreneur’s Poster

interested parties may initiate a discussion We first deliver a brief review of the litera- with the entrepreneur, who is usually in ture on entrepreneurial presentation, and then close proximity, or they may follow up after we invoke information visualization theory, the event. Given the dynamics of a poster paying specific attention to the elements of session presentation, where participants can preattentive process. Next, we propose our choose whether or not to engage the entrepre- hypothesis by integrating aspects of informa- neur, the design and content of the poster is tion visualization theory with the entrepreneur- possibly a critical factor in generating interest ial literature. We then detail the methods in the venture and attracting investors to the employed, specifically discussing the event, table. sample, variables, and analysis. Finally, we Our methods employed a two-step process discuss our findings, as well as the research and that tested the relationship between variance in managerial implications of our paper. poster presentation and potential stakeholder interest in the venture. First, we analyzed the Literature Review posters of entrepreneurs from the showcase The relationship between an entrepreneur’s event in order to categorize both the type of presentation and its subsequent impact on information presented and the visual design of stakeholder behavior is of particular interest to the poster. Second, we utilized postconference entrepreneurship scholars. Existing theory has survey data to determine the relationship attributed presentation to the early-stage ven- between the nature of the presentation in ture’s ability to attract resources (Clarke 2011). accordance with information visualization Such a relationship is intriguing from a practi- theory and stakeholder interest. cal standpoint because it demonstrates that less

LATHAM AND TELLO TELLO 87 3 prominent aspects of the venture, such as pre- ticipated in an hour-long focus group where sentation style, affect stakeholder decision- the video was discussed, providing the making, along with more widely accepted researchers with more insight into the investor factors, such as the specific product or business decision-making process. model. Therefore, a practical implication of this Based on 30 business angel responses, the research is ensuring that entrepreneurs pay authors broke down evaluation across four cri- equal or adequate attention to smaller details of teria: presentation, market, product, and other. their ventures. As such, it may be worthwhile to Of importance to this manuscript is the finding explore the possibility that entrepreneurs can that of the previously mentioned criteria, pre- better affect stakeholder interest by making sentation was by far the most important aspect manageable adjustments to their presentations considered by potential investors. Presentation rather than larger-scale, more complex adjust- was cited 41 percent by evaluators as critical ments to their proposals. to support, whereas product and market considerations garnered 23 percent and 10 Understanding the Role and Importance percent, respectively. Most notable was that 90 of Entrepreneurs’ Presentations percent of the comments regarding the presen- Several studies offer guidance in support of tation were negative; subsequently, 80 percent this study’s efforts. In a study of the impact of of the respondents stated that they would not entrepreneurs’ social skills in presentations, pursue further investment opportunities. It Baron and Brush (1999) found that entrepre- was the presentation content, style, and struc- neurial presentations characterized by a high ture that most influenced potential stakeholder level of social competence attracted a higher decision-making. level of interest. In commenting on their study, Similarly, Baron, Markman, and Bollinger the authors state: “It is precisely in this kind of (2006) investigated the role of entrepreneurs’ environment, we suggest, that a high level of physical attractiveness in influencing individu- social competence may be especially crucial to als’ perceptions of the venture’s viability. The entrepreneurs, helping to ‘tip the balance’ in authors conducted three studies that drew from their favor” (Baron and Brush 1999, p. 460). the fields of fitness-related evolutionary theory Thus, their study demonstrates that apart from and the affect infusion model. In their first entrepreneur and venture, the efficacy of the study, the authors provided black and white presentation, in this instance whether or not photos of the entrepreneur, as well as limited the entrepreneur demonstrates social compe- information about the entrepreneurial venture. tency, plays a critical role. Raters were then asked to rate the idea and Mason and Harrison (2003) examined the entrepreneur. The physical attractiveness of the response of 41 angel investors and institutional entrepreneur had a favorable effect on the investors to an entrepreneur’s presentation, entrepreneur and the product or service being focusing on the presentation’s content, pitched. In the second study, raters viewed product, and market. In summarizing their two-minute videotapes of entrepreneurs that research intent, Mason and Harrison (2003) contained more information about the venture. state: “Our concern in this article is with pre- Again, the findings suggested that there was a sentation. Knowing how to present an oppor- strong relationship between attractive physical tunity effectively to potential investors can be appearance and favorability. Finally, the third regarded as one specific aspect of being ‘inves- study linked entrepreneurs’ attractiveness with tor ready’ ” (p. 30). greater financial success. The authors videotaped the investment pitch The study authors conclude: “Thus, it seems of a senior manager at a small software orga- possible that people meeting entrepreneurs for nization. Subsequently, this videotape was the first time who must make important deci- shown to two mixed groups of institutional sions about them (e.g. whether to provide them investors and angel investors. Data were gath- with financial support or purchase their new ered in a three-stage manner. First, video products) may be influenced by entrepreneurs’ observers were asked to comment on a pro- attractiveness without being fully aware of such vided form as they watched the video. Second, effects” (Baron, Markman, and Bollinger 2006, they were asked to indicate on the form p. 477). On a subconscious level, factors unre- whether or not they would pursue an invest- lated to the venture, like attractiveness, had a ment opportunity. Third, video observers par- significant impact on the evaluation of the

488 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT overall venture. Although the Baron, Markman, ings. Even if the underlying proposition is and Bollinger (2006) study examined attributes sound a business may still fail to raise finance if related to the entrepreneur, such an examina- the business plan is poorly constructed and tion is in direct line with our study: the seem- presented. This includes shortcomings in busi- ingly inconsequential details related to the ness plans and other written documents that overall presentation may sway beholders’ are aimed at investors and also deficiencies in impressions of the entrepreneur and the ‘pitches’ at investment forums” (Mason and venture. Kwok 2010, p. 6). Indeed, Chen, Yao, and Kotha (2009) found The literature is consistent in suggesting that that the presentation might be the factor that entrepreneurs’ presentations are important for tips the scale in favor of some entrepreneurs. garnering stakeholder interest. It is also clear The authors conducted two independent from the literature that many aspects require studies that demonstrated that investors’ per- further investigation, such as the content and ception of an entrepreneur’s presentation pre- format of presentations themselves. In other paredness was significantly related to the words, even if the entrepreneur’s venture has a investment decision. In summarizing their sound business plan and business model, it can study, the authors state: “Some might argue be undermined by presentation. Hence, we feel that all the objective criteria earlier research has that it would be useful to examine how visual included (e.g. market opportunities, competi- presentation devices affect stakeholders’ inter- tion, and strategies) already address the issue est in entrepreneurial ventures. Are potential of preparedness. However, our results suggest stakeholders influenced by visual presentation that perceived preparedness might be the devices, independent of the entrepreneur and ‘missing link’ between a set of objective criteria venture, when deciding on which ventures to VCs may or may not use in their investment investigate? In line with such inquiry, we decisions” (Chen, Yao, and Kotha 2009, p. 212). present a theoretical perspective derived from Again, the preceding quote demonstrates the information visualization theory to examine the potential effect of the presentation in addition effect of visual presentation devices in attract- to the objective criteria. ing potential stakeholder interest to the Most recently, Clarke (2011) focused on the venture. role of visual symbols during entrepreneur presentations. She conducted a very novel study of three entrepreneurial ventures and their inter- Drawing from the Field of Information actions with employees, customers, and inves- Visualization Theory tors; specifically, the study focused on Information visualization is a field that entrepreneurs’ use of visual symbols such as examines the manner in which individuals setting, props, dress, and expressiveness. Of process, assimilate, and act upon the visual note, Clarke (2011) found that one entrepre- presentation of data, information, and knowl- neur displayed awards, grants, and patents edge (Ackoff 1989; Spence 2007; Ware 2004; on the walls of the office to demonstrate Zeleny 1987). More generally, it seeks to under- legitimacy. Again, such findings demonstrate stand the manner in which individuals process the importance of presentation in generating the multitude of data that they see in everyday interest. life. In large part, information visualization Finally, Mason and Kwok (2010) also cite the focuses on the presentation of data in defined importance of presentation, separate and inde- formats that ensures specific information is pendent of the business model or opportunity. drawn from a larger field of information. In a conceptual paper, the authors put forth a Studies demonstrate that information visualiza- framework for investment readiness. The tion plays a critical role in fields where complex authors argue that the failure to generate data, concepts, and dynamics are related venture investment is more related to entrepre- (Johansson et al. 2006; Kelton, Pennington, and neurs’ lack of investment readiness than lack of Tuttle 2010; Vessey 1991). available funding. As part of the framework, Within the field of information visualization, they cite one of the primary reasons for lack of substantial research focuses on preattentive investment readiness as “presentational fail- processing. Preattentive processing describes ings.” The authors state: “The third dimension the manner in which individuals automatically of investment readiness is presentational fail- and subconsciously focus on specific visual

LATHAM AND TELLO TELLO 89 5 Figure 2 tion. Accordingly, another aspect of informa- Preattentive Processing tion visualization theory we draw upon is the integration of words and images. Substantial literature examines whether words in the form of text and tables, or images, in the form of symbols and graphs, are more effective in presenting information (Anderson and Kaplan 1992; Anderson and Reckers 1992; Bertin 1983; Dickson, DeSanctis, and McBride 1986; Kelton, Pennington, and Tuttle 2010; Tergan, Keller, and Burkhard 2006). Graphics are best used to convey complex relationships between data or actions where they help the viewer to mentally information (Triesman 1986). Figure 2 offers a record and process information (Bertin 1983). very basic depiction of preattentive processing. However, simpler tasks or relationships that Both visuals are identical, with the exception require precise data comparisons might be that in the second diagram the number “3” is better represented through text-based tables surrounded by a thicker outline. According to (Benbast and Dexter 1986; Hard and Vanecek the precepts of preattentive processing, inevi- 1991). Text alone may be the best vehicle for tably, if individuals were asked to pick a presenting an abstract concept such as number after viewing these two diagrams, most freedom. Similarly, when complex information individuals would pick the number three. such as scientific information is being pre- Preattentive processing works on a subcon- sented, it is best conveyed by linking text-based scious level, that is, individuals do not under- explanation with visual representation. stand why their attention was drawn to the number three. Hypotheses Development Preattentive devices are typically broken By relying on information visualization down into four categories: form, color, motion, theory, we hope to ascertain whether or not, and spatial position (Ware 2004). Form specific presentations are more effective in gar- includes such attributes as line width, line nering stakeholder interest. Our hypotheses length, grouping, highlighting, and numeros- address three aspects of information visualiza- ity. Highlighting is a simple example of form- tion: preattentive processing, amount of based preattentive processing. Highlighted text information, and information complexity. In draws the viewer’s attention to a specific piece developing our hypotheses, we believe it is of data. Hue and intensity are examples of critical to review our study context. The color-based preattentive processing. For sample was selected from an inventor–investor example, shading is a preattentive attribute forum conducted in the U.S. medical device that influences human cognition and recogni- sector. Although we expand more on the tion (Irani, Slonowsky, and Shajahan-Human sample in the Methodology section next, it is 2006). In general, experimental studies have important to understand the nature of informa- demonstrated that individuals recognize and tion shared in these events. Medical device process shaded information more quickly than start-ups require a variety of resources and adjacent, nonshaded entities, objects, or text. partners to move forward, including investors, Motion and position are also preattentive medical-regulatory experts, engineering and devices. Flicker and direction of motion are product development specialists, and others. common preattentive devices used in multime- The entrepreneur must present a range of dia and computer presentations (but not rel- complex information related to the sometimes evant for static visual poster displays). Finally, unique interests of potential partners, includ- the position of an object, data point, or infor- ing clinical pathway, competitive environment, mation piece relative to other information market size, patent status, product specifica- has also been found to induce preattentive tions, and regulatory requirements. As a result, processing. an entrepreneur’s ability to effectively convey As just detailed, information visualization complex information in a concise manner is seeks to better understand the manner in which central to our study, as is reflected in our individuals process complex, visual informa- hypothesis development.

690 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Venture Evaluation Criteria and more interest than a poster that did not apply Preattentive Processing preattentive visual devices to these criteria? In the process of securing resources, entre- As discussed, preattentive processing occurs preneurial ventures are evaluated by a wide when information is altered in form, color, array of potential partners, including angel motion, and/or its spatial position relative to investors, corporate development staff, VCs, other information. Form devices include line banks, and private equity firms. These potential size, highlighting of text, grouping of text, and partners typically look for information that helps numerosity, a device where presenters use them evaluate the venture’s merits. Although grouping of marks or symbols to represent and scholars note potential differences in evaluation contrast numeric values. Color devices include criteria across various investor groups (Mason the use of color shading, color intensity, and and Stark 2004), we focused on factors that hue. The use of a visual printed poster lends appear to have the strongest agreement across itself to the use of form and color devices. Form investors, specifically angel investors and VCs. devices can be used to highlight headings, indi- This is consistent with the event setting we cate important information, and group and examined, an inventor showcase event that organize related information. Color devices can attracts regional angel investors and venture be used to attract attention to specific informa- capital firms. Zacharakis and Meyer (2000) tion such as market size or revenue projections, found that VCs base their investment decisions and to contrast information, like device perfor- on the capabilities of the entrepreneur and team, mance compared with one’s competitors. attractiveness of the product or service, market Spatial placement can be used to contrast size, and competitive conditions, and the potential performance, and related metrics. return on the investment. Such findings are The effectiveness of preattentive devices is supported by other literature, placing these dependent on the degree of difference between factors into one of four well-defined categories: target text or data, and nontarget text or data market opportunity (Douglas and Shepherd (Duncan and Humphreys 1989; Quinlan and 2002), business opportunity (Chesbrough and Humphreys 1987; Ware 2004). Their findings Rosenbloom 2002), entrepreneurial manage- indicate that in addition to attracting attention, ment team (Douglas and Shepherd 2002), and preattentive devices subconsciously help in the nature of technology (Astebro 2004). Mason and organization and comprehension of text and Stark (2004) found that VCs and angel investors data. If the intent of the visual poster presen- relied on similar criteria, though they differ in tation is to attract and focus the attention of two specific regards. Angel investors tend to potential investors and collaborators, the use of place a higher premium on management capa- preattentive devices may increase the entrepre- bilities and emotional attachment. The former neur’s likelihood of starting this conversation. (management capabilities) is captured in many Based on this discussion, our first hypothesis of the presentations evaluated, whereas emo- states: tional attachment is not inherently part of static visual presentations. Thus, the commonality H1: Within entrepreneurs’ visual presenta- across VCs and angel investors speaks to market tions, there will be a positive relationship size, business potential, management team between the preattentive processing of key experience, and the nature of the product or evaluation criteria and stakeholder interest. technology. These are the key evaluation criteria upon which we integrate and build our first Quantity of Information hypothesis. Also drawing from the field of information Although the research suggests that poten- visualization, our second hypothesis addresses tial investors rely on such criteria to evaluate the issue of information overload, specifically early-stage ventures, our study asked: “If the as it relates to preattentive processing. Gener- preceding well-established criteria were sub- ally, the notion is less is better: when presented jected to information visualization devices, with a field of highly complex, diverse infor- would the likelihood of interest from potential mation, studies have shown that individuals stakeholders increase?” For example, if an tend to assimilate a finite level of information entrepreneur’s poster relied on preattentive (Ware 2004). Item can be preattentively pro- processing to call attention to market size or cessed by surrounding them with less informa- market opportunity, would that poster illicit tion. Under such pretense, individuals can

LATHAM AND TELLO TELLO 91 7 discern and filter potential information or data 2004). It holds that the use of graphs and points when they are embedded in a discrete images to break up text and convey complex field of surrounding information. As the nature information can assist in both attracting viewer and multitude of information rises, then infor- attention and comprehension. mation overload begins. Specifically, individu- Such a perspective is very relevant within als are unable to process information beyond a the context of our study. The information pre- certain point (Stewart 1988). sented in the posters of early-stage medical Very few studies have examined this device ventures typically includes market data, dynamic in an entrepreneurial context. In his clinical performance data, technology features, examination of judgment analysis procedures, entrepreneurial team information, patent data, Stewart (1988) found that most experts rely and opportunity information. These data are on three to seven pieces of information incredibly complex. The literature suggests that when making important decisions, whereas tables and text are best suited to presenting Zacharakis and Meyer (2000) found that inves- scientific information, such as clinical data. tors were limited in their ability to process Thus, our third hypothesis, is formally stated excess information. They found that as more next: information was provided to investors, they were less likely overall to view a venture as H3: Within entrepreneurs’ visual presenta- viable. From the perspective of an entrepre- tions, there will be a positive relationship neur’s attempt to recruit venture support, these between the use of tables to convey scientific earlier findings suggest that the entrepreneur and clinical data and stakeholder interest. must be selective in providing information in visual, written, and oral presentations. The In line with the rationale just mentioned, entrepreneur must address minimum critical past studies demonstrate that graphs and information points, but must not present too images are best used for conveying complex much information, or otherwise she risks over- relationships (Benbast and Dexter 1986; Bertin loading the potential stakeholder. Too little 1983; Hard and Vanecek 1991; Ratwani, information may limit the impact of a presen- Trafton, and Boehm-Davis 2008). Within the tation, but the efficacy of information presen- context of our study, the potential ventures tation also reaches a breaking point; such a involve very complex medical devices, often relationship approximates an inverse “U.” entailing the description of how surgical Thus, our second hypothesis: devices interact with human and other medical systems. Discussions regarding medical supply H2: Within entrepreneurs’ visual presenta- and distribution chains are also complicated. tions, there will be an inverse curvilinear Accordingly, simple text descriptions are often relationship between the amount of infor- inadequate or difficult to understand. In line mation presented and stakeholder interest. with existing literature, our expectation is that entrepreneurs who invoke a higher level of Integration of Graphs, Images, graphs and images to communicate such infor- and Tables mation will garner a higher level of initial inter- Our next two hypotheses predict the efficacy est from potential stakeholders. From this of integrating visual representation of scientific perspective, we offer our fourth hypothesis: and complex data. Substantial literature has examined whether words (e.g., text or tables) H4: Within entrepreneurs’ visual presenta- or images (e.g., symbols, pictures, or graphs) tions, there will be a positive relationship are more effective in presenting information between the use of graphs and images to (Anderson and Kaplan 1992; Anderson and convey complex information regarding Reckers 1992; Bertin 1983; Dickson, DeSanctis, product, market, and/or competitors and and McBride 1986; Kelton, Pennington, and stakeholder interest. Tuttle 2010; Tergan, Keller, and Burkhard 2006). Additionally, information visualization Methodology research finds that the degree of difference Study Context between target text or data and nontarget text This study examined the visual poster pre- or data is important (Duncan and Humphreys sentations and subsequent event interactions of 1989; Quinlan and Humphreys 1987; Ware 42 medical device inventors attending an

892 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT inventor–investor showcase event hosted by a document that includes useful information major university in the greater Boston region. regarding the stage of venture, composition of Inventor/entrepreneurs display a three-by-four- team, past awards, and related data. The foot poster detailing their invention, as well as researchers reviewed each poster for content other information they view as critical to related to the criteria discussed previously, attracting the interest of potential investors and including the market, business model, technol- partners. At the event, entrepreneurs are sta- ogy, and entrepreneurial team. The majority of tioned at their posters. They may proactively posters addressed market criteria (57.14 engage attendees or wait for attendees to percent) and technology criteria (64.29 approach them concerning their venture. The percent), whereas fewer addressed business poster plays many roles. First, it serves as a model criteria (16.67 percent) and venture team hook to attract investors that focus on a specific criteria (11.90 percent). product or market. Second, it helps convey The researchers coded each poster for the very technical and scientific data. Third, it presence of specific preattentive devices rela- allows an investor to view the entrepreneur’s tive to market size, market opportunity, entre- device without speaking to the investors. preneurial management team, and nature of Finally, the poster serves to support the entre- technology. Coding was also conducted on the preneur’s pitch, if they do speak directly with use of tables, graphs, and images. Using an investor. Thus, the poster is central to gen- Figure 1, we offer an example of the coding erating interest among event attendees. process. In Figure 1, the independent coders These events are quite common in regions addressed the question: “Was a graph or image known for their success in bringing new used to convey market opportunity on the medical devices from the lab to market. With its poster?” As Figure 1 illustrates, a graph is high concentration of research hospitals and prominently displayed relative to the medical universities, and established investment com- devices market opportunity. The researchers munity, Massachusetts ranks second in the then coded “yes” or “1,” indicating that a graph country (behind California) for Food and Drug was used to convey market opportunity. Simi- Administration (FDA) premarket notifications larly, coders then asked: “Was a table used to and approvals of medical devices (John Adams convey scientific or medical data?” Again, with Innovation Institute 2010). This particular event Figure 1 as a reference, there are no tables attracted over 200 participants. It was spon- present in the poster; thus, the researchers sored by the university’s medical device center, coded the poster “no” or “0.” Importantly, three the state’s premier medical device industry independent researchers manually coded the group, and the state’s innovation and technol- posters in Excel using a previously agreed- ogy center. The university medical device upon coding scheme based on the preattentive center hosting the event has assisted medical and visual processing literature discussed pre- device start-up companies in Massachusetts to viously. Once the initial coding was conducted, raise over $8 million in public and private the researchers met to review their results and funds. Participants at the event included angel judge interrater agreement. A kappa coefficient investors from the region’s medical community, of 0.75, p < .001, was obtained, indicating a established venture capital investment firms, substantial level of agreement between coders along with other partners critical to the devel- (Kvalseth 1989; Landis and Koch 1977; Stemler opment of medical device ventures (e.g., 2004). Good agreement is measured at the product development specialists, clinical regu- range between 0.60 and 0.80. Table 1 summa- latory experts, market strategists). Participant rizes the data resulting from this exercise. registration is required, providing the research- The second phase of data collection ers an opportunity to confirm participation of involved the administration of a 24-item online potential investors and partners. survey that examined inventor/entrepreneur prior experience, stage of venture, occurrence Data Collection and nature of stakeholder interactions at and Data collection was conducted in two dis- after the event, and information regarding the tinct phases. After the inventor showcase event, design of the poster. A series of e-mail requests each inventor’s poster was collected in both a explaining the purpose of our study and solic- print and electronic format. We also collected iting participation in the online survey were the inventor’s application to attend the event, a sent to all 42 inventor/entrepreneurs following

LATHAM AND TELLO TELLO 93 9 Table 1 Descriptive Data for Poster Use of Visual Devices

No. of No. of No. of No. of No. of Words Images Graphs Tables Headings

Mean 472.83 5.43 1.79 0.64 8.76 Maximum 1200 24 8 6 19 Minimum 46 0 0 0 2

the event. Thirty entrepreneurs completed the approach was adopted for the number of head- survey, providing a 71.43 percent response ings. The first variable (Number_of_Headings) rate. The data from the online survey were then measured the total number of headings on matched to the researcher coding data for each the poster, whereas the second variable poster. The survey was critical in determining (Curvi_Headings) was the squared product of whether or not the entrepreneur had generated the number of headings. stakeholder interest at the event or after the In testing H3, we constructed a variable that event. Survey items were also used to measure determined if tables were used to present the nature of stakeholder interest. medical or clinical data (Table_ClinicalData). The intent of the survey was twofold. First, it Each variable was coded “1” if the poster con- allowed for a distinct information gathering tained tabular clinical or scientific data; each effort not attached to the inventor–investor variable was coded “0” if no tabular data were showcase. We were ascertaining whether or not present. In testing H4, we constructed a variable the entrepreneur had been successful in gar- that determined if graphs or images were used nering interest primarily from investors, but to relay usage, market, product, or competitive also other resource providers, such as potential information (Graph/Image_ComplexData). As corporate partners. Second, it helped overcome with the number of words, a similar coding any common-method bias that could result if procedure was utilized. independent and dependent measures were obtained from the same data source. Dependent Variable Survey data were primarily used to measure Independent Variables stakeholder interest. By relying on the survey H1, examining the effect of preattentive pro- data and consistent with our hypothesis, we cessing on stakeholder interest, was measured defined a dependent variable called total stake- by the independent coding of each poster by holder interest (Total_Interest). Table 2 summa- the researchers as notedpreviously. We focused rizes the items used to measure stakeholder on two primary preattentive devices, color interest. If the respondent answered in the affir- shading (Color_Shading) and form highlighting mative, then the item was coded “1,” indicating (Form_Highlighting). Each variable was coded interest in the venture on the part of a potential “1” if the poster shaded or highlighted infor- stakeholder. Summarizing the score on these mation relative to the four venture criteria. five items created a total stakeholder interest Each variable was coded “0” if preattentive index. Accordingly, the total score had a poten- processing was not applied to the four venture tial range from 0 to 5, with the higher score criteria. indicating that the entrepreneur had experi- In line with H2, we constructed linear and enced a higher level of interest from potential curvilinear measures of the number of words stakeholders. We used Cronbach’s alpha to esti- and headings. With regard to the number of mate the dependent variable reliability. In our words, two variables were constructed. The first study, Cronbach’s alpha across the five items variable (Number_of_Words) measured the total related to total stakeholder interest was 0.72. number of words on the poster, whereas the With an item scale less than 20 items, such a second variable (Curvi_Words) was the squared score is viewed as good measure of a common product of the number of words. A similar construct (Cortina 1993).

1094 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Table 2 Survey Items Ascertaining Stakeholder Interest

• Did you speak with any potential investors during the poster session? • After the event ended, were you contacted by any potential investors who learned about your device/company through the poster session? • Did you speak with any potential collaborators (other than investors) during the poster session? • After the event ended, were you contacted by any potential collaborators (other than potential investors) who learned about your device/company through the poster session? • Did you make any other contacts that helped your venture progress as a result of conversations you had during the poster session?

Control Variables and the control variables are displayed in We included three control variables: whether Table 3, along with the correlations between or not the venture had been externally validated the various variables. prior to the event (External Venture Valida- A multiple regression analysis was con- tion), prior entrepreneurial experience (First ducted to evaluate the impact of the Venture), and prior team marketing experience preattentive device variables and visual device (Mark_Exp). Although our factor variables mea- variables on the dependent variable, total sured aspects relative to information visualiza- stakeholder interest. We followed an iterative tion, our controls were intended to capture estimation procedure, also called a stepwise variance relative to the technology and entrepre- model. First, we ran a control-only model, then neurs. External Venture Validation determined we interjected each independent variable one whether the quality of the venture idea exerted by one into the model. We then ran a fully influence on stakeholder interest. Researchers specified model, which included all control reviewed application materials provided by variables and all independent variables. Such each entrepreneur to determine whether or not an approach allows for an understanding of the the venture had been validated by some type of model estimation. After including all indepen- external process, such as the award of public or dent variables, the results demonstrate a much private grant funding, award of business plan more robust model. All hypotheses were tested competition, and/or angel or venture capital against the full model. Table 4 illustrates the investment. A venture was coded “1” if it had results of this regression analysis. been validated by an external process, and “0” if The results offer support for H1, which states it had not been validated by an external process. that posters characterized by a higher level of First Venture asked the entrepreneur to identify preattentive devices associated with the four whether or not they had previously tried to decision-making factors would demonstrate a commercialize any other type of invention or higher level of potential stakeholder interest. device; a yes response was coded as “2.” Mar- The Form_Highlighting device (t = 2.60; p < .05) keting experience was controlled under the was significant. Color_Shading, (t = 1.76; assumption that a team with marketing exper- p < .10) was also significant. This finding sug- tise might potentially design a more visually gests that stakeholders were attracted to posters attractive poster. Researchers reviewed applica- that utilized highlighting and shading to draw tion materials provided by each entrepreneur to attention to the established evaluation criteria— determine whether or not the team had a market size, business opportunity, entrepre- member with a formal marketing background. A neurial team, and nature of technology. venture was coded “1” if it had a team member In line with aspects of information visualiza- with marketing experience, and “0” if no mar- tion theory, H2 predicted a curvilinear relation- keting experience was noted. ship between the amount of information presented and potential stakeholder interest. Results As detailed previously, we relied on two vari- The means and standard deviations for the ables that captured the linear and curvilinear independent variables, dependent variables, effect of the number of words and headings,

LATHAM AND TELLO TELLO 95 11 12 96

Table 3 ORA FSALBSNS MANAGEMENT BUSINESS SMALL OF JOURNAL ORA FSALBSNS MANAGEMENT BUSINESS SMALL OF JOURNAL Means, Standard Deviations, and Correlations of Study Variables

Variable Mean Std. Dev 1 2 3 4 5 6 7 8 9 10 11 12

1 Ext_Venture_Validation 0.33 0.47 1 2 First Venture 1.53 0.50 −0.05 1 3 Form_Highlighting 0.43 0.50 0.00 0.21 1 4 Color_Shading 0.26 0.44 0.07 −0.34 0.07 1 5 Number_of_Words 472 292 0.06 0.00 0.25 0.01 1 6 Curvi_Words 307277 368875 0.05 0.14 −0.23 −0.02 0.96** 1 7 Number_of_Headings 8.76 3.35 −0.1 0.19 0.03 0.25 0.38* 0.30* 1 8 Curvi_Headings 87.76 69.40 −0.14 −0.17 0.08 0.02 0.30 0.24 0.97** 1 9 Table_ClinicalData 0.20 0.40 0.44** −0.2 −0.1 −0.03 0.03 −0.01 −0.12 −0.12 1 10 Graph/Im_ComplexData 0.60 0.49 −0.02 0.19 0.43* 0.46** 0.20 0.24 0.1 −0.12 −0.1 1 11 Mark_Exp 14.00 0.35 0.00 −0.22 0.03 0.03 0.09 0.03 0.05 0.04 0.04 −0.08 1 12 Total_Interest 2.13 1.35 0.17 0.09 0.42* 0.22 0.04 0.04 −0.38* −0.37* 0.07 0.34* −0.11 1

n = 30. *p < .10. **p < .05. Table 4 Effect of Visual Presentation on Stakeholder Interest

I II III IV V VI

Constant 1.66 (0.90) 0.53 (0.64) 4.08 (2.53) 1.74 (0.94)* 1.12 (0.75) 2.15 (1.02)** Ext_Venture_Validation 0.50 (0.52) 0.52 (0.35) 0.42 (0.53) 0.48 (0.59) 0.42 (0.53) 0.22 (0.36) First Venture 0.21 (0.52) 0.17 (0.35) 0.27 (0.61) 0.22 (0.54) 0.27 (0.61) 0.02 (0.38) AHMADTLO97 TELLO AND LATHAM AHMADTLO13 TELLO AND LATHAM Mark_Exp −0.43 (0.77) −0.68 (0.53) −0.70 (0.83) −0.43 (0.78) −0.70 (0.83) −0.37 (0.52) Form_Highlighting 0.78 (0.37)** 1.04 (0.40)** Color_Shading 1.62 (0.37)*** 0.74 (0.42)** Number_of_Words 0.006 (0.005) 0.005 (0.003) Curvi_Words −0.000003 (0.000) −0.000002 (0.000) Number_of_Headings −0.74 (0.60) −0.21 (0.38) Curvi_Headings 0.02 (0.03) 0.001 (0.019) Table_ClinicalData 0.05 (0.73) 0.09 (0.44) Graph/Image_ComplexData 1.62 (0.43)*** 0.88 (0.41)** F Statistic 0.47 7.24*** 0.97 0.34 4.04** 5.68*** R2 0.05 0.60 0.23 0.05 0.39 0.64

n = 30; df = 28. Robust standard errors in parentheses. *p < .10. **p < .05. ***p < .01. respectively. Neither the linear or curvilinear on a subconscious level. Within the context of components were significant. Thus, our data our event, potential stakeholders did not arrive suggest that the amount of information con- at a poster and exclaim: “I like this poster tained on the posters does not affect stakeholder because it’s colorful!” However, on some level interest. that is exactly what happened. Aspects of visu- H3 predicted a positive relationship between alization theory partially explain why an entre- the use of tables to convey scientific and clini- preneur garnered particular interest. Of all the cal data and further stakeholder interest. We factors that explain what sets the powerful did not find support for the inclusion of tables engine of entrepreneurship in motion, our and scientific data. However, we did discover study suggests that the simple highlight func- strong support for H4. Visual presentations that tion in PowerPoint contributes to this effort. All contained images and graphs to convey else being equal between two start-ups com- complex information about the venture, the peting for resources, the venture that visually product, and/or the market garnered a higher leverages elements of information visualization level of potential stakeholder interest (t = 2.13; theory, specifically highlighting and color p < .05). This is consistent with information shading, appears to get the second look. Addi- visualization theory, especially given the nature tionally, entrepreneurs within scientific fields of the technology in this sample, highly that took the extra step of including graphs and complex, technical, medical devices. images that explained complex dynamics attracted more interest than those who did not Discussion include these items. Similarly, our findings Our study offers several contributions to the provide entrepreneurs with tangible, manage- field. First, we examine the tangential aspects able strategies for improving their likelihood of of the entrepreneurship process, such as the attracting attention and, potentially, securing details of an entrepreneur’s presentation. investment partners. Presentation design is a Building on previous efforts (Baron and Brush teachable skill that yields immediate and 1999; Clark 2008; Gregoire, de Koning, and positive returns for entrepreneurs without sig- Oviatt 2008), our study suggests that it is not nificant investment on the part of the entrepre- only the business opportunity or the nature of neurial team. the technology that attracts stakeholder atten- Second, as Fried and Hisrich (1994) detail, tion, but also the entrepreneur’s ability to the investment decision in an entrepreneurial format and highlight specific venture details. venture often follows a well-known decision As Baron, Markman, and Bollinger (2006) path. Within the context of this process, our suggest, often small, unnoticeable dynamics paper addresses the beginning of the process influence a potential stakeholder’s perception. when entrepreneurs are competing to secure It is quite possible that an entrepreneur who the attention of potential investors. Fried and calls more visual attention to the size of the Hisrich (1994) call this stage origination: when market or her team’s strong background will investors work within known industries direc- attract a higher level of stakeholder interest tories and groups to find deals. As detailed, than an entrepreneur that allows the text alone such a description fits our study context quite to do the talking. Mason and Harrison (2003) well, as medical device investors in our study allow for such a prospect, as they argue that attended an industry event looking for early- not all investor decisions are likely to be stage ventures. Accordingly, our research grounded in rational decision-making. More sought to determine which factors in the entre- so, it is the common frame of reference or preneur’s presentation helped the venture to shared understanding of the organization, its stand out in the origination phase. product, its market, and its environment that is Third, this study offers substantive guidance created by the presentation, that is likely to for early-stage entrepreneurs. As Hayward, draw investor interest, and potentially subse- Shepherd, and Griffin (2006) detailed, entrepre- quent investment. neurs, more often than not, believe that their Although one might argue that our findings idea or abilities will speak for themselves, a merely confirm that highlighting and shading concept called entrepreneurial hubris. Our attract attention, we respectfully suggest that study offers further proof that the entre- this greatly understates our contribution. preneurs’ presentations play a significant Indeed, information visualization theory occurs role in attracting interest. Entrepreneurs must

1498 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT diligently review the specific aspects that make contribution in building upon recent work that their venture unique and attractive to potential stresses importance of presentation in generat- stakeholders, and then preattentively process ing interest. those aspects of their venture. For example, Despite these limitations, this study makes a patent protection and clinical regulatory useful contribution to the literature by squarely pathway may be critical to investors in the addressing the relationship between an entre- medical device industry, whereas computer preneur’s presentation and his/her ability to gaming investors might be more concerned generate interest in his/her venture. To state about potential market segment growth and that presentation effects potential stakeholder technology factors. From a new venture per- interest is obvious and clear; less clear is the spective, preattentive processing is not only manner in which the presentation works. This about attracting attention but rather focusing is the spirit of our study. We are not looking to attention on factors that make a venture unique determine the linkage between presentation and fundable. and investment, but the linkage between pre- These findings are also informative for the sentation and interest. It is in line with other targets of these presentations, the investors and recent studies that have examined the smaller, partners the entrepreneur hopes to attract. ancillary details in entrepreneurs’ presentation. Information visualization theory operates on a Baron, Markman, and Bollinger’s (2006) subconscious level, meaning that many of the finding that physical appearance may affect devices that influence individual data assimila- stakeholder interest and Clarke’s (2011) finding tion occur without the individual recognizing that entrepreneurs establish legitimacy among the dynamic. Prospective stakeholders should investors by hanging a patent on the wall are recognize that their interest in a specific tech- each incredibly insightful in this regard. These nology or venture might be solely dictated by studies suggest that investor interest is not only the merits of the technology or business oppor- shaped by product, entrepreneurial experience, tunity, but rather by presentation design. and other well-known entrepreneurial traits, Although substantial research has examined but also underlying, less obvious factors not the evaluation criteria of potential stakeholders entirely related to the venture’s merits. In both such as investors, our study allows for the instances, these scholars demonstrate that possibility that stakeholders are significantly seemingly nonrelated factors, such as appear- influenced by small visual details. ance and wall decorations, affect the ability of Our study does have limitations. First, an entrepreneur to attract stakeholder interest. readers may question the generalizability of the Similarly, our findings demonstrate that the sample; we only examined one event within smallest formatting details applied to a presen- the medical device industry. However, we tation may elicit further stakeholder interest. believe that our findings will apply to other Given the competitive nature of the early-stage sectors with similar attributes. Entrepreneurs entrepreneurial process, in which countless across technology sectors are faced with the entrepreneurs vie for limited resources, the challenge of communicating very complex effect of information visualization devices products and services. Our study offers some within an entrepreneur’s presentation cannot guidance to entrepreneurs in all technology be discounted. We believe our study provides sectors, not just medical device entrepreneurs. insight into the manner in which certain entre- Second, we did not control for the effect of preneurs are able to attract a higher level of supporting materials, such as business plans, interest, and potentially resources that aid in prototypes, and other materials and props. This moving their ventures forward. could, and should, be an avenue for additional research. Third, our regression analysis was References limited in size. Although certainly a limitation, Ackoff, R. L. (1989). “From Data to Wisdom,” Johnson, Parker, and Wijbenga (2006) make Journal of Applied Systems Analysis 16, the case that any small business research 3–9. requires a higher level of flexibility in its Anderson, J. C., and S. E. Kaplan (1992). “An methods and analysis than the traditional large Investigation of the Effect of Presentation business research model that relies heavily on Format on Auditors’ Non-Investigation publicly available data. Despite these limita- Region Judgments,” Advances in Accounting tions, we believe our paper offers a notable Information Systems 1, 71–88.

LATHAM AND TELLO TELLO 99 15 Anderson, J. C., and P. M. J. Reckers (1992). Clarke, J. (2011). “Revitalizing Entrepreneur- “An Empirical Investigation of the Effects of ship: How Visual Symbols Are Used in Entre- Presentation Format and Personality on preneurial Performances,” Journal of Auditor’s Judgment in Applying Analytical Management Studies 48(6), 1366–1390. Procedures,” Advances in Accounting 10, Cortina, J. M. (1993). “What Is Coefficient 19–43. Alpha? Examination of Theory and Applica- Astebro, T. (2004). “Key Success Factors for tions,” The Journal of Applied Psychology 78, Technological Entrepreneurs’ R&D Projects,” 98–104. IEEE Transactions on Engineering Manage- Dickson, G. W., G. DeSanctis, and D. J. McBride ment 51(3), 314–321. (1986). “Understanding the Effectiveness of Bamberger, P. (2008). “Beyond Contextuali- Computer Graphics for Decision Support: A zation: Using Context Theories to Narrow Cumulative Experimental Approach,” Com- the Micro-Macro Gap in Management munications of the ACM 29(1), 40–47. Research,” Academy of Management Journal Douglas, E. J., and D. A. Shepherd (2002). 51(5), 839–846. “Exploring Investor Readiness: Assessment Baron, R. A., and C. G. Brush (1999). “The by Entrepreneurs and Investors in Australia,” Role of Social Skills in Entrepreneurs’ Venture Capital 4(3), 219–236. Success: Evidence from Videotapes of Entre- Duncan, J., and G. Humphreys (1989). “Visual preneurs’ Presentations,” Frontiers of Search and Stimulus Similarity,” Psychologi- Entrepreneurship Research, 1999. http:// cal Review 96, 433–458. fusionmx.babson.edu/entrep/fer/papers99/ Fried, V. H., and R. D. Hisrich (1994). “Toward I/I_A/I_A%20Text.htm (accessed January 15, a Model of Venture Capital Investment Deci- 2011). sion Making,” Financial Management 23(3), Baron, R. A., G. D. Markman, and M. Bollinger 28–37. (2006). “Exporting Social Psychology: Effects Gregoire, D., A. de Koning, and B. M. Oviatt of Attractiveness on Perceptions of Entrepre- (2008). “Do VCs Evaluate ‘Live’ Presenta- neurs, Their Ideas for New Products, and tions Like They Evaluate Business Plans?,” Their Financial Success,” Journal of Applied Frontiers of Entrepreneurship Research, Pro- Social Psychology 36, 467–492. ceedings of the Babson College Entrepre- Benbast, I., and A. S. Dexter (1986). “An Inves- neurship Research Conference, June 2008. tigation of the Effectiveness of Color and http://digitalknowledge.babson.edu/cgi/ Graphical Information Presentation Under viewcontent.cg1?article=1230&context=fer Varying Time Constraints,” Management (accessed August 4, 2014). Information Systems Quarterly 10(10), Hard, N. J., and M. T. Vanecek (1991). “The 59–83. Implications of Tasks and Format on the Use Bertin, J. (1983). Semiology of Graphics. of Financial Information,” Journal of Infor- Madison, WI: The University of Wisconsin mation Systems 5(2), 35–49. Press. Hayward, M. L. A., D. A. Shepherd, and D. Chen, X. P., X. Yao, and S. Kotha (2009). Griffin (2006). “A Hubris Theory of Entrepre- “Entrepreneur Passion and Preparedness in neurship,” Management Science 52, 160–172. Business Plan Presentations: A Persuasion Irani, P., D. Slonowsky, and P. Shajahan- Analysis of Venture Capitalists’ Funding Human (2006). “Perception of Structure in Decisions,” Academy of Management Shaded Space-Filling Visualizations,” Infor- Journal 52(1), 199–214. mation Visualization 5, 47–61. Chesbrough, H., and R. A. Rosenbloom (2002). Janiszewski, C. (1998). “The Influence of “The Role of the Business Model in Captur- Display Characteristics on Visual Explor- ing Value from Innovation: Evidence from atory Search Behavior,” The Journal of Con- Xerox Corporation’s Technology Spin-Off sumer Research 25(3), 290–301. Companies,” Industrial and Corporate Johansson, J., P. Ljung, M. Jern, and M. Cooper Change 11(3), 529–555. (2006). “Revealing Structure in Visualiza- Clark, C. (2008). “The Impact of Entrepreneurs’ tions of Dense 2D and 3D Parallel Coordi- Oral ‘Pitch’ Presentation Skills on Business nates,” Information Visualization 2006(5), Angels’ Initial Screening Investment 125–136. Decisions,” Venture Capital 10(3), 257– John Adams Innovation Institute (2010). 2009 279. Index of the Massachusetts Innovation

16100 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Economy. Westborough, MA: Massachusetts Graph Comprehension,” Journal of Experi- Technology Collaborative. mental Psychology. Applied 14(1), 36–49. Johnson, P. S., S. C. Parker, and F. Wijbenga Spence, R. (2007). Information Visualization: (2006). “Nascent Entrepreneurship Research: Design for Action, 2nd ed. Essex, England: Achievements and Opportunities,” Small Pearson Education Limited. Business Economics 27(1), 1–3. Stemler, S. E. (2004). “A Comparison of Con- Kelton, A., R. Pennington, and B. Tuttle (2010). sensus, Consistency, and Measurement “The Effects of Information Presentation Approaches to Estimating Interrater Reliabil- Format on Judgment and Decision Making: ity,” Practical Assessment, Research & Evalu- A Review of the Information Systems ation 9(4), 1–18. http://PAREonline.net/ Research,” Journal of Information Systems getvn.asp?v=9&n=4 (accessed February 10, 24(2), 79–105. 2011). Kvalseth, T. O. (1989). “Note on Cohen’s Stewart, T. R. (1988). “Judgment Analysis Pro- Kappa,” Psychological Reports 65, 223–226. cedures,” in Human Judgment: The SJT Landis, J. R., and G. G. Koch (1977). “The View. Eds. B. Brhemer and C. Joice. North Measurement of Observer Agreement for Holland: Elsevier, 41–74. Categorical Data,” Biometrics 33(1), 159– Tergan, S. O., T. Keller, and R. Burkhard 174. (2006). “Integrating Knowledge and Infor- Mason, C. M., and R. T. Harrison (2003). “Audi- mation: Digital Concept Maps as Bridging tioning for Money: What Do Technology Technology,” Information Visualization Investors Look For at the Initial Screening 2006(5), 167–174. Stage?,” The Journal of Private Equity 6(2), Triesman, A. (1986). “Features and Objects in 29–42. Visual Processing,” Scientiﬁc American Mason, C. M., and J. Kwok (2010). “Investment 255(2), 114–125. Readiness Programmes and Access to Tufte, E. (1997). Visual Explanations: Images Finance: A Critical Review of Design Issues,” and Quantities, Evidence and Nature. Local Economy 25(4), 269–292. Cheshire, CT: Graphics Press. Mason, C. M., and M. Stark (2004). “What Do Vessey, I. (1991). “Cognitive Fit: A Theory- Investors Look For in a Business Plan? A Based Analysis of the Graphs Versus Tables Comparison of the Investment Criteria of Literature,” Decision Sciences 22, 219– Bankers, Venture Capitalists, and Business 240. Angles,” International Small Business Ware, C. (2004). Information Visualization, Journal 22, 227–246. 2nd ed. San Francisco, CA: Elsevier. Quinlan, P., and G. Humphreys (1987). “Visual Zacharakis, A. L., and G. D. Meyer (2000). “The Search for Targets Deﬁned by Combinations Potential of Actuarial Decision Models: Can of Color, Shape and Size: An Examination of they Improve the Venture Capital Invest- Task Constraints on Feature and Conjunc- ment Decision,” Journal of Business Ventur- tion Searchers,” Perception and Psychophys- ing 15, 323–346. ics 41(5), 455–472. Zeleny, M. (1987). “Management Support Ratwani, R. M., J. G. Trafton, and D. A. Systems: Towards Integrated Knowledge Boehm-Davis (2008). “Thinking Graphically: Management,” Human Systems Management Connecting Vision and Cognition during 7(1), 59–70.

LATHAM AND TELLO TELLO 101 17 Journal of Small Business Business Management Management 2016 2014 54(1), ••(••), pp. pp. 102–117 ••–•• doi: 10.1111/jsbm.12133 Shaping the Entrepreneurial Mindset: Entrepreneurial Intentions of Business Students in Croatia by Sanja Pfeifer, Nataša Šarlija, and Marijana Zekic´ Sušac

Business students from a public university in Croatia participated in an international study on entrepreneurial self-efficacy, identity, and education. The results of this preliminary empirical research indicate that the main predictors of the entrepreneurial intentions in Croatia are strength of entrepreneurial identity aspiration and entrepreneurial self-efficacy. These two main constructs mediate the number of personal, situational, or contextual factors, including education. Empiri- cal analysis supports the majority of Social Cognitive Career Theory hypothesized interaction between control variables and main constructs such as self-efficacy, positive outcome expectations, and entrepreneurial identity. These findings thus fill the gap in the empirical evidence of the theoretical framework validity derived from different contexts.

seek a fresh supply of new entrepreneurs. Introduction Policy makers are searching for answers about Renewed interest in intention-based models what makes a person willing to become an of entrepreneurial behavior derives from entrepreneur, how these influencing factors several avenues. First, empirical evidence sug- can be amplified, and how the number of gests that intention-based models are still a potential or actual entrepreneurs can be highly relevant theoretical framework for increased in order to provide more significant exploring entrepreneurial behavior. It has been economic growth. Intention-based models can argued that intention is still the single best offer more insight into these issues. Third, predictor of future behavior. However, major industries’ restructuring, and public sectors’ intention-based models still produce divergent downsizing reduce hiring and job openings. An or narrowly focused results. Therefore, inte- alarming number of young adults are faced grated, cross-disciplinary approaches, more with unemployment, including graduates from sophisticated methodologies, more robust higher education institutions. Higher education instruments for measuring, and more evidence institutions have been increasingly called upon from multinational and diverse populations are to provide more enterprising orientation welcome. Second, the global economic crisis among graduates (Henry, Hill, and Leicht and slowdowns in national economic growth 2005). Therefore, there is a need to critically rates all over the world urge policy makers to evaluate how entrepreneurship education

Sanja Pfeifer is professor of Management at the Department of Economics, J.J. Strossmayer University. Nataša Šarlija is professor of Methodology of Entrepreneurship Research at the Department of Economics, J.J. Strossmayer University. Marijana Zekic´ Sušac is professor of Methodology of Entrepreneurship Research at the Department of Economics, J.J. Strossmayer University. Address correspondence to: Sanja Pfeifer, Department of Economics, J.J. Strossmayer University, Gajev trg7, Osijek 31000, Croatia. E-mail: [email protected].

102 JOURNALPFEIFER, OF SMALL ŠARLIJA, BUSINESS AND ZEKIC MANAGEMENT´ SUŠAC 1 contributes to the development of career inter- main results are presented and discussed. The est, intentions, and behavior. final part of the paper focuses on the implica- Croatia is a postsocialist, south-eastern Euro- tions of this preliminary research and its pean country in transition from an efficiency to limitations. an innovation-driven economy. Fostering entrepreneurship through entrepreneurship educa- Theoretical Background tion programs has recently become a part of Integration of the Intention-Based the national strategic agenda. Croatian higher Models education, especially in the field of business Understanding what compels individuals to or social studies, has been introducing become entrepreneurs remains an important entrepreneurship-related courses, modules, or question in entrepreneurship research (Shane even official programs since 2005. Studies con- and Venkataraman 2000). Krueger and Brazeal ducted on the population of Croatian higher (1994) indicated that “before there can be education students indicated that students entrepreneurship there must be the potential report various barriers to entrepreneurship, for entrepreneurs.” In pursuit of the elusive including the lack of attitudes or capabilities to entrepreneurial potential or potential entrepre- successfully launch and manage a new venture. neurs’ determinants, academic research has Entrepreneurship education has been per- been evolving through related fields of inquiry ceived as a way of promoting enterprising anchored in motivational theories such as the attitudes and professional capabilities for mas- Theory of Planned Behavior (TPB) (Ajzen tering an entrepreneurial career after gradua- 1991) and social cognition theories such as the tion. However, there is little research on the Social Cognitive Career Theory (SCCT) (Lent, effects of entrepreneurship education on the Brown, and Hackett 2002). different antecedents of intentions. In addition, The intention-based models derived around there remains a lack of research about the the TPB suggest that entrepreneurial intentions generalizability of the intent-based models and are a function of the perceived feasibility entrepreneurship education impact coming and perceived desirability of the entrepreneurial from developing or entrepreneurially young act (Krueger, Reilly, and Carsrud 2000). The contexts. SCCT also offers a comprehensive framework The aim of this paper is to explore what in which the choice of career is a function explains entrepreneurial intentions of Croatian of dynamic interaction among self-efficacy, students, and how well recently hypothesized outcome expectations, and personal goals (Lent, organizing framework on entrepreneurial inten- Brown, and Hackett 2002). These constructs tions fits in the Croatian cultural and socioeco- fully or partially mediate the influence of per- nomic context. In addition, the paper will sonal (demographic, attitudes, experience) or investigate the influence of formal entrepreneur- external (supports, barriers) factors, and explain ship courses on attitudes, self-efficacy, and posi- (1) the development of interest for career, (2) tive outcome expectations that usually precedes actual career choices, and (3) stability of the the formation of the entrepreneurial intentions performance. The general overview of the entre- or career choice. Data were collected from the preneurial career as an interaction of these key business students of a public university in constructs in the SCCT model is presented in Croatia who participated in an international Figure 1. survey, “Entrepreneurship Education Project: The SCCT model suggests that the develop- Enhancing Entrepreneurial Self-Efficacy and ment of entrepreneurial aspirations depends Identity,” during 2010–2012. on interaction among self-efficacy and outcome The paper is structured as follows: It begins expectations. The interaction of these key with a short overview of the theoretical frame- constructs leads to the formation of the entre- work of entrepreneurship intentions. Empirical preneurial intentions. The entrepreneurial evidence of previous research is presented in intentions are more likely to develop if a order to justify the selection of the main con- person: (1) feels capable to successfully structs for the model and the presumptions conduct entrepreneurial tasks or activities; (2) behind their interactions. The paper proceeds anticipates positive outcomes from entrepre- with a description of the sample, measurement neurial activity; and (3) has a high personal instruments, and main dependent, indepen- interest or aspiration toward entrepreneurship. dent, and control variables. After that, the High entrepreneurial intentions increase the

2 JOURNALPFEIFER, OF SMALLSARLIJA, BUSINESS AND ZEKIC MANAGEMENT´ SUSAC 103 Figure 1 Development of Entrepreneurial Interests and Career Choices

Source: Adopted from Social Cognitive Career Theory (Lent, Brown, and Hackett 2002) likelihood of actual nascent entrepreneurial theoretical framework validity, we assumed behavior and new venture creation. The per- that self-efficacy and outcome expectation have sonal or external factors such as aptitudes, past a strong direct influence on entrepreneurial learning experiences, social norms, support, intentions in the sample of Croatian students. and barriers may directly or indirectly lead to The entrepreneurial self-efficacy is a capability changes in entrepreneurial intentions or of the successful conducting of particular entre- nascent behavior. The actual behavior or per- preneurial tasks such as recognizing opportu- formance serves as feedback for personal or nity. It has been perceived as one of the environmental influences. strongest predictors of entrepreneurial inten- Winkel, Vanevenhoven, and Ehrhardt (2011) tions and, consequently, behavior (Zhao, Hills, recently proposed the new, integrative organiz- and Seibert 2005). Positive expectations to get ing framework for intention-based research financial benefits, more security and autonomy, anchored in the SCCT theory. The overall or other outcomes from entrepreneurial activity generalizability of their organizational frame- are assumed to be positively related to the work has not been empirically validated in the entrepreneurial intentions (Carter et al. 2003). context of a transitional, postsocialist, entre- Segal, Borgia, and Schoenfeld (2002) found that preneurially young culture such as Croatia. self-efficacy and outcome expectations together Some studies conducted with international explained over half the variance of students’ samples suggest that entrepreneurial intention entrepreneurial intentions. models are time or cultures invariant (Liñán Croatia as a transitional country faced with and Chen 2009). Other studies suggest that the economic downturn, displays a specific relationship among the key constructs may socioeconomic and environmental conditions diverge from those proposed by theoretical that may influence the perception of entrepre- framework (Liñán, Urbano, and Guerrero neurial intentions. In addition to the straightfor- 2011) and show a different configuration ward and joint impact of self-efficacy and among the key constructs from those hypoth- outcome expectation on intentions, it is pos- esized. Therefore, in order to confirm portabil- sible that some additional path or different ity of the SCCT model, we kept the main configuration of the key constructs plays a more drivers of intentions in the model, also keeping explicit and more central role. For example, an open door to the possibility of the new Nabi and Linan (2011) found that entrepreneur- configuration or some additional paths that ial intentions are higher for the graduates in may make the SCCT model more indigenous to developing countries when compared with the Croatian context. developed ones. Developing countries have unfavorable economic or institutional frame- SCCT Model Indigenization work conditions which usually diminish the Key Constructs Configuration. In order to outcome expectation or appropriation of the contribute to the empirical base of the SCCT valued outcomes. Therefore, we assumed that

104 JOURNALPFEIFER, OF SMALL ŠARLIJA, BUSINESS AND ZEKIC MANAGEMENT´ SUŠAC 3 additional construct that will describe the inter- H2: Personal factors are related to (a) entrepre- est or attractiveness of the entrepreneurial neurial self-efficacy and (b) entrepreneurial career may be introduced, along with self- intentions. efficacy and positive outcome expectations. Recently, Farmer, Yao, and Kung-Mcintyre It has been assumed that self-efficacy, (2009) raised the issue of entrepreneurial iden- outcome expectation, or even identity aspira- tity aspiration as a strong precursor of indi- tions are acquired and influenced through vidual engagement or actions. Perception of the situational factors such as: (1) personal accom- “possible self as an entrepreneur can be an plishments; (2) learning experiences from important motivational mechanism in the either personal or family business, entrepre- nascent entrepreneurial process” (Farmer, Yao, neurial exposures; and (3) social persuasions and Kung-Mcintyre 2009). The issue of identity (social norms about entrepreneurship). Posi- aspirations has been rarely treated explicitly in tive experiences usually raise self-efficacy, the SCCT. Bandura (1986) states that one’s par- identity, and outcome expectations. Empirical ticular pattern of likes or dislikes regarding evidence suggests particularly that role various occupations should have a more impor- models, prior family, or own business expo- tant position in the formation of the career sure, and academic entrepreneurship pro- interest. Downsizing of job openings in the grams may have both direct impact on sectors that traditionally employ a majority of intentions and indirect impact on key con- business study graduates (such as public sector structs of self-efficacy, identity, or outcome or large private companies) makes an entrepre- expectation (Lent, Brown, and Hackett 2002; neurial career more attractive. The presence of Zellweger, Sieger, and Halter 2011; Zhao, unemployment poses a specific triggering effect Hills, and Seibert 2005). and may make the aspiration to “entrepreneur The impact of formal academic programs in identity” stronger. The number of graduates entrepreneurship on SCCT key constructs is who are faced with unemployment in Croatia is particularly perplexing. Gerry, Marques, and increasing. For example, in 2011, Croatia regis- Nogueira (2008) found no significant impact of tered 36.1 percent of young unemployed adults entrepreneurship-related training on students’ aged 18–25, and according to the Eurostat data- entrepreneurial intentions, whereas Fayolle, base this was the second highest rate in Europe. Gailly, and Lassas-Clerc (2006) found that These specific circumstances led us to the entrepreneurship training programs may have a assumption that the strength of entrepreneurial direct positive impact as well as a counter- identity aspiration may have a more prominent effect on entrepreneurial intentions. Some pre- role in explaining entrepreneurial intentions liminary evidence of the samples of Croatian among business students in Croatia. students shows that university-based entrepre- In order to evaluate whether these hypoth- neurship education has a significant direct esized key constructs of intentions fit with impact on developing entrepreneurial capacity Croatian context, we propose the following and mindsets (Kružic´ and Pavic´ 2010). Bilic´, hypothesis: Prka, and Vidovic´ (2011) provided evidence of a rather weak entrepreneurial orientation H1: Entrepreneurial self-efficacy, entrepre- toward entrepreneurship among Croatian uni- neurial outcome expectation, and entrepre- versity students (29.8 percent of the students neurial identity aspiration are positively reported that they have entrepreneurial inten- associated with entrepreneurial intentions. tions), and a low correlation between students’ enrollment in entrepreneurship courses and As suggested in motivational or social cogni- their entrepreneurial intentions. tion intent models, there is a multitude of per- It has also been suggested that situational, sonal and contextual factors that affect the educational, or contextual impact might be development of entrepreneurial intentions and funneled primarily through self-efficacy, their key constructs. Personal factors such as outcome expectations, or identity aspirations. gender, age, ethnicity, family wealth, as well as Krueger and Brazeal (1994) suggested that abilities influence intentions either directly or entrepreneurship education should have a indirectly through their impact of self-efficacy positive impact on perceived feasibility for (Lent, Brown, and Hackett 1994). Therefore, we entrepreneurship by increasing abilities to suggest solve entrepreneurship-related tasks. It should

4 JOURNALPFEIFER, OF SMALLSARLIJA, BUSINESS AND ZEKIC MANAGEMENT´ SUSAC 105 also improve the perceived aspirations and desirability and perceived feasibility. They expectations by presenting entrepreneurship as found that key constructs “need not both be a personally rewarding career. On the other high to induce entrepreneurial intentions.” side, Brockner, Higgins, and Low (2004) Therefore, we propose to test interrelations suggest that entrepreneurship education can between the key constructs in the SCCT intent trigger a person’s sensitivity to negative out- model to find whether the intention may be comes and force a person to decrease outcome influenced through either enhancing self- expectation or intention to start a business in efficacy or nurturing expectations. We expect order to avoid the possibility of failure. There- that the interactive and multiplicative relation- fore, we propose that entrepreneurship educa- ships between the key constructs of intention tion impact, as well as situational or contextual development may be assessed through high factors may be mediated through entrepreneur- correlations between key construct variables ial identity, outcome expectation, and self- and presume efficacy effect as follows: H4: Entrepreneurial identity aspirations, self- H3: Situational, educational, and contextual efficacy, and outcome expectation are factors are related to: (a) entrepreneurial highly correlated and interactive. self-efficacy; (b) entrepreneurial identity aspiration; (c) outcome expectation; and (d) In light of the aforementioned, we suggest entrepreneurial intentions. the hypothesized relationship between the key constructs in the SCCT model as presented in Configuration of the Key Constructs Relations Figure 2. and Their Interactions. Lent, Brown, and Hackett (2002) hypothesized that self-efficacy interacts with outcome expectations. A person Methodology During 2010, the Entrepreneurship Educa- who feels capable to carry out specific entre- tion Project: Enhancing Entrepreneurial Self- preneurial activities has a higher positive Efficacy and Identity (http://quipuapps expectation of the outcomes of entrepreneurial .com/EEP/) gathered 467 higher education activities. In addition, the person who feels institutions from 91 countries worldwide. The efficacious in entrepreneurial activities may main purpose of the project was to enable the form higher aspirations and create a stronger longitudinal exploration of the motivational image of entrepreneurial identity, or may processes underlying students’ interest or anticipate more positive outcomes. However, intention in an entrepreneurial career, and the interaction between entrepreneurial self- entrepreneurial education impact on it. The efficacy and outcome expectation or aspirations project resulted in the international database on may be more complex. A person’s intentions student perceptions, values, and drives toward and career choices do not always reflect this an entrepreneurship career. Students were primary and straightforward path. For example, asked to self-administer an online question- a person who aspires and perceives positive naire in English. For the purposes of this study, outcomes from entrepreneurship may choose a we have used the data set for Croatia in order career in entrepreneurship, despite the absence to analyze Croatian students’ attitudes on an of self-efficacy or entrepreneurial identity, entrepreneurship career. simply to seize the emergent opportunity. A person with high self-efficacy and capability to successfully conduct activities connected to Sample Demographics starting or managing a new venture may not The Croatian sample included 504 under- develop high entrepreneurial aspirations in the graduate and graduate students from the presence of other more attractive career Department of Economics, University of Osijek, options. In addition, one may feel capable but Croatia. The survey was conducted in two unsure about outcomes he or she may get out waves: 194 students entered the survey in 2010 of an entrepreneurial career. The interactions and 310 entered the survey in 2012. No differ- of the key constructs may open opportunities ences in the key constructs variables were for a broader set of academic interventions. For found in the two waves of data collection. After example, Fitzsimmons and Douglas (2011) pro- elimination of the incomplete questionnaires, vided evidence for the interaction of perceived the data set of 454 respondents was created.

106 JOURNALPFEIFER, OF SMALL ŠARLIJA, BUSINESS AND ZEKIC MANAGEMENT´ SUŠAC 5 Figure 2 Integrative Model of Entrepreneurial Intention

The sample comprises 67.84 percent women; entrepreneurial self-efficacy as a person’s confi- 82.38 percent Croats; and the average age is dence in his/her ability to perform standard 20.5. entrepreneurial tasks such as identifying opportunity, assessing market potential, mastering The Dependent Variable business plans, leading teams, and so on. In The main dependent variable was entrepre- this paper, the self-efficacy statements were neurial intentions. Recently, Thompson (2009) aggregated in a single construct in order to suggested that entrepreneurial intent has been get a distinctive indicator for the complex defined in a variety of ways: from broad dispo- phenomenon. The same procedure was used sitions or interests, to relatively advanced speci- for all key variables described with multi-item fications of actions that are being taken and measurement scales such as: strength of the connected to entrepreneurial behavior. He sug- entrepreneurial identity (Farmer, Yao, and gested that entrepreneurial intent is not “simply Kung-Mcintyre 2009), the entrepreneurial a yes or no question but a matter of extent outcome expectation (Krueger 2000), and social ranging from very low to very high degree,” norms (Kolvereid and Isaksen 2006). The mul- and proposed a uniform reliable and interna- tidimensional and multi-item instrument for tionally applicable individual entrepreneurial social norms was aggregated to reflect the mul- intent scale for measuring intentions. Conse- tiplicative effect of the positive or negative opin- quently, the seven-point Likert scale measure of ions of the family, relatives, and friends, and students’ perception on “How true is it that you their importance for student’s own opinions. are saving money (have plans to launch, spend Appendix 1 presents descriptive statistics for time learning, etc)?” was adopted from dependent variable, main model constructs, and Thompson (2009), and for this study the origi- a few other multi-item measurement instru- nal six statements were aggregated in a single ments used in our study. construct. This procedure was necessary to The control variables comprise the set of provide the single proxy for various degrees of personal, situational, and educational factors entrepreneurial intentions. (see Appendix 2). One particular variable—the student’s major—is created to allow us to get The Independent Variables preliminary insight in certain aspects of the The selection of the independent variables educational impact on students’ intentions. was based on theoretical framework presented Undergraduates at the time of the survey have earlier in a paper. Recently, McGee et al. (2009) not yet been enrolled in any of the entrepreneur- introduced a refined instrument for measuring ship courses. Majors in the entrepreneurship

6 JOURNALPFEIFER, OF SMALLSARLIJA, BUSINESS AND ZEKIC MANAGEMENT´ SUSAC 107 program have also been enrolled in a number of Additionally, the CART algorithm was used entrepreneurship courses and exposed to a for creating a model for classifying entrepre- variety of activities relevant for entrepreneur- neurial intentions. Questier et al. (2005) sum- ship education. Other majors such as marketing, marized CART steps as: (1) assign all objects to management, finances, or informatics have been root node; (2) split each explanatory variable at exposed to at least two courses in entrepreneur- all possible split points; (3) for each split point, ship. This allows us to track the subsample split the parent node into two child nodes by differences in the perception of entrepreneur- separating the objects with values lower and ship due to the students’ exposure to the courses higher than the split point for the considered with entrepreneurship-related content. explanatory variable; (4) select the variable and split point with the highest reduction of impu- Methods rity; (5) perform the split of the parent node We used descriptive statistics, t-test, analysis into the two child nodes according to the of variance (ANOVA), chi-square test, Pearson selected split point; (6) repeat steps 2–5, using correlation coefficient, logistic regression (LR), each node as a new parent node, until the tree and classification and regression tree (CART). has maximum size; and (7) prune the tree back The t-test was used to test the difference using cross-validation to select the right-size between two means. In order to test the differ- tree. The evaluation function used in this ence between three or more means, ANOVA research for splitting in CART is the Gini index was used. Chi-square was used to test the defined as (Apte and Weiss 1997): dependence on categorical and ordinal vari- () 2 ables. Pearson correlation coefficient was used Gini t=−1 ∑ pi to test the significance of correlation between i continuous variables (Sheskin 2009). LR modeling is widely used for analyzing where t is a current node and pi is the prob- multivariate data involving binary responses ability of class i in t. The CART algorithm con- that we deal with in our research. It provides a siders all possible splits in order to find the best powerful technique analogous to multiple one according to the Gini index. The algorithm regression and ANOVA for continuous can deal with continuous as well as with cat- responses. Because the likelihood function of egorical variables. The aim is to identify the branches that provide the least additional pre- mutually independent variables Y1,...,Yn with outcomes measured on a binary scale is a dictive power per leaf. Some of the advantages member of the exponential family with of this method are: it makes no distributional π π assumptions for dependent and independent log1 ,… , log n as a canonical ()()− π ()− π variables, it is not affected by the outliers, and 111 n it does not vary under a monotone transforma- π parameter ( j is a probability that Yj becomes tion of independent variables (Yohannes and 1), the assumption of the LR model is a linear Webb 1999). relationship between a canonical parameter and the vector of explanatory variables xj Results (dummy variables for factor levels and mea- In order to support H1, we have calculated sured values of covariates): Pearson correlation coefficients between entrepreneurial intention and self-efficacy, outcome ⎛ π j ⎞ τ expectation and entrepreneurial identity (see log ⎜ ⎟ = x jb ⎝ ⎠ Table 1). 1 − π j Students with higher entrepreneurial intentions have higher strength of entrepreneurial This linear relationship between the loga- identity, higher self-efficacy, and higher entre- rithm of odds and the vector of explanatory preneurial outcome expectation. Consequently, variables results in a nonlinear relationship H1 is confirmed. between the probability of Y equals 1 and j In order to test H2, we performed the t-test the vector of explanatory variables (Harrel and ANOVA test, and presented the main 2001): results in Table 2. Personal factors such as gender, minority, π = ()ττbb()+ () j expxxjj1 exp and marital status are not related to the student

108 JOURNALPFEIFER, OF SMALL ŠARLIJA, BUSINESS AND ZEKIC MANAGEMENT´ SUŠAC 7 Table 1 perception of self-efficacy. However, the stu- Correlation dents who are coming from wealthier families and students from urban areas have a higher Coefficients—Entrepreneurial perception of self-efficacy. This partially con- Identity, Self-Efficacy Outcome firms the H2a. Male students have higher entre- Expectation, and Entrepreneurial preneurial intentions, although they do not have higher self-efficacy. Minority, married, Intentions wealthier students coming from urban areas have higher entrepreneurial intentions. Thus, Variable Entrepreneurial the H2b is confirmed. Intentions In order to support H3, we tested how situational (prior family or personal exposure to Entrepreneurial identity 0.551*** business activities, positiveness of that experi- Entrepreneurial self-efficacy 0.374*** ence), educational variables (major, degree, construct student status), and contextual variables (social Entrepreneurial outcome 0.127*** norms) explain the main intention-based model expectation construct constructs. The ANOVA, t-test, and Pearson correlation tests are presented in Table 3. All tested factors are related to entrepreneur- *Significant at 10 percent. ial intentions (H3d). The entrepreneurial self- **Significant at 5 percent. efficacy is related to all mentioned factors, with ***Significant at 1 percent. Table 2 t-Test and ANOVA—Personal Factors According to Intentions and Self-Efficacy

Variable Mean (S.D.)

Entrepreneurial Self-efﬁcacy Entrepreneurial Intention

Gender Female 67.67 (14.96) 4.13 (1.04) Male 66.91 (15.41) 4.39 (1.20) t = 0.50 t = 2.21** Minority Yes 67.37 (12.06) 4.60 (0.89) No 67.99 (14.95) 4.18 (1.12) t = 0.30 t = 2.76** Married Yes 67.87 (16.13) 4.44 (1.15) No 67.77 (14.32) 4.17 (1.09) t = 0.05 t = 1.77* Family wealth Below 62.65 (13.31) 4.33 (0.84) Average 67.64 (14.51) 4.17 (1.10) Above average 72.63 (17.09) 4.78 (1.13) F = 3.15** F = 4.97*** Area Urban 70.11 (15.29) 4.42 (1.09) Rural 66.04 (15.14) 4.07 (1.10) Suburb 65.91 (13.88) 4.00 (1.11) Mixed 66.45 (13.73) 4.17 (1.06) F = 2.55* F = 3.52*

Boldfaced entries indicate significant differences at following levels: *Significant at 10 percent. **Significant at 5 percent. ***Significant at 1 percent.

8 JOURNALPFEIFER, OF SMALLSARLIJA, BUSINESS AND ZEKIC MANAGEMENT´ SUSAC 109 110

Table 3 ANOVA, t-Test and Correlation Coefﬁcients—Key Construct of Intentions According to Situational, Educational, and Contextual Factors

Variable Self-Efﬁcacy Identity Outcome EntrepreneurialIntention H3a Aspiration Expectation H3d

ORA FSALBSNS MANAGEMENT BUSINESS SMALL OF JOURNAL H3b H3c

FIE,ŠRIA N ZEKIC AND ŠARLIJA, PFEIFER, Situational inputs Family business exposure Yes 69.26 (14.35) 3.42 (0.91) 5.64 (1.08) 4.41 (1.09) No 65.86 (15.61) 3.29 (0.87) 5.67 (1.14) 4.03 (1.09) t = 2.39** t = 1.61* t = 0.31 t = 3.63*** Own business exposure Yes 70.24 (14.64) 3.40 (0.91) 5.79 (0.96) 4.32 (1.09) No 64.68 (15.08) 3.29 (0.87) 5.54 (1.22) 4.09 (1.12) t = 3.95*** t = 1.38 t = 2.43** t = 2.22** Family added conﬁdence r = 0.115** r = 0.044 r = −0.072 r = 0.077* Own added conﬁdence r = 0.167*** r = 0.151*** r = 0.054 r = 0.185*** Educational inputs Major Entrepreneurship 68.22 (11.62) 4.14 (0.62) 5.73 (0.76) 4.93 (0.70) Graduate 68.80 (13.66) 3.23 (0.94) 5.92 (0.87) 4.21 (1.15) ´

UA 9 SUŠAC Undergraduate 65.26 (17.10) 3.38 (0.78) 5.24 (1.35) 4.09 (1.02) F = 2.91* F = 13.43*** F = 20.45*** F = 6.96*** Status Full time 67.43 (14.70) 3.33 (0.89) 5.68 (1.06) 4.19 (1.10) Part time 72.32 (16.10) 3.75 (0.92) 5.19 (1.85) 4.66 (1.00) t = 1.48 t = 2.09** t = 1.20 t = 1.94* Contextual inputs Social norms r = 0.156*** r = 0.201*** r = 0.127*** r = 0.171***

Boldfaced entries indicate significant differences at following levels: *Significant at 10 percent. **Significant at 5 percent. ***Significant at 1 percent. the exception of student status (H3a). Not all lists rank the variables according to their hypothesized relations between situational, importance for prediction. educational, or contextual factors and key con- Both models (LR and CART) extracted struct have been supported. For example, the entrepreneurial identity and self-efficacy as the outcome expectations are higher for those who best predictors of entrepreneurial intentions. either have previous own business experiences, From the logistic regression coefficient, it can higher perception of the social norms, or are be seen that increasing entrepreneurial iden- enrolled in graduate programs. Identity aspira- tity, entrepreneurial self-efficacy, as well as tion is higher for students who have family or personal business exposure would increase own business experiences, higher perception probability of entrepreneurial intention. In of social norms, or have entrepreneurship as addition, students above average in family their major field of study. Therefore, H3b and wealth would have a higher probability of H3c are partially confirmed (Table 4). starting a business than those below average Based on this evidence, the LR and CART and average. The CART model also identifies analysis were performed in order to find key entrepreneurial identity and self-efficacy as the overall predictors of the entrepreneurial inten- two most highly ranked variables, whereas the tions. For the purpose of this analysis, the third important predictor extracted by CART is construct that measures entrepreneurial inten- the construct of social norms, followed by tion is transformed into a dichotomous variable family wealth and personal business exposure. where 1 represents entrepreneurial intention Gender, family business exposure, and expo- (students who scored more than 4 in the seven- sure to longer entrepreneurship education point Likert measurement scale), and 0 repre- (graduate students) contribute to the predic- sents no entrepreneurial intention (students tive power of the CART model, although to a who scored 4 or less on the seven-point Likert lesser extent. measurement scale). For modeling purposes, Both models are tested on the same valida- the whole data set was divided into a develop- tion sample. The best extracted LR model ment sample and a validation sample. The regarding standard overall fit measures for LR development sample (75 percent of data) con- (e.g., Pearson χ2 = 435.919***) showed the total sisted of 344 cases, 198 cases with entrepre- hit rate of 71.2 percent, the hit rate for those neurial intentions and 146 cases without with entrepreneurial intention was 79.2 entrepreneurial intentions. The validation percent, and for those without entrepreneurial sample (25 percent of data) consisted of 110 intention, it was 63.2 percent. The CART model cases, out of which 43 have an intention and 67 showed a total hit rate of 69.15 percent, have no intention to start a business. Results of whereas the hit rate for respondents with entre- the best extracted LR and CART are presented preneurial intention was 64.41 percent, and the in Table 5. The logistic regression is presented hit rate for respondents without entrepreneur- with estimated LR coefficients whereas CART ial intention was 75.0 percent.

Table 4 Pearson Correlations Between Self-Efﬁcacy, Aspirations, and Outcome Expectations

Identity Self-Efﬁcacy Outcome Expectations

Identity construct 1 0.32*** 0.19*** Self-efﬁcacy construct 1 0.37*** Outcome expectation construct 1

*Significant at 10 percent. **Significant at 5 percent. ***Significant at 1 percent.

10 JOURNALPFEIFER, OF SMALLSARLIJA, BUSINESS AND ZEKIC MANAGEMENT´ SUSAC 111 Table 5 The Logistic Regression and CART Models

Variable LR estimates CART rank

Entrepreneurial identity 0.929*** 1 Entrepreneurial self-efficacy 0.033*** 2 Personal business exposure added to confidence 0.070** 5 Family wealth Below average 0.281 4 Average −0.051 above average 0.843* Gender: Male 0.190 7 Social norms 0.021 3 Major Entrepreneurship 0.608 6 Other graduate −0.324 Undergraduate 0.018 Family business exposure added to confidence 0.3927 8

Boldfaced entries indicate significant differences at following levels: *Significant at 10 percent. **Significant at 5 percent. ***Significant at 1 percent. CART, classification and regression tree; LR, logistic regression.

constructs in explaining entrepreneurial inten- Discussion tions. In our study, the measuring instruments As suggested in Winkel, Vanevenhoven, and of the SCCT key constructs showed good inter- Ehrhardt (2011), the recently elaborated inte- nal consistency, and were considered as valid. grative framework that “seeks to encapsulate The self-efficacy construct importance was con- the key elements of current theoretical models firmed. Expectation constructs showed interest- and to reduce the conceptual overlaps in dif- ing interaction. Positive outcome expectation is ferent intent-based models,” needs to be vali- correlated with entrepreneurial identity, and dated in different contexts. The purpose of this self-efficacy, allowing either of these constructs study was to examine what explains entrepre- to mediate the effect of positive outcome expec- neurial intentions of Croatian students, and tations. In the context of Croatia, the entrepre- how well-hypothesized SCCT anchored frame- neurial identity aspiration construct took a more work on entrepreneurial intentions fits in the prominent role than theoretically presumed. Croatian cultural and socioeconomic context. Croatia’s transitional past, the present economic LR and CART indicated the importance of downturn, and relatively young entrepreneurial self-efficacy, identity aspirations, personal busi- culture with a weak institutional infrastructure ness exposures, family wealth, and social norms explain why students’ positive outcome expec- in the prediction of Croatian students’ entrepre- tations did not take a more important rank neurial intentions. The entrepreneurial inten- in our prediction model. Even if students tions are higher for students who perceive have high expectations of financial rewards, entrepreneurial identity as an attractive career autonomy, personal rewards, or family security, option, who feel capable of performing they may have doubts about the possibility to entrepreneurship-related tasks, who perceive achieve those outcomes during slowdowns. In that such a career choice enjoys positive valua- addition, they may overemphasize entrepre- tions from family members or other important neurial career desirability in the presence of referrals, and who have a wealthier family back- unemployment and find entrepreneurial iden- ground. The main theoretical assumption was tity highly appealing. Their valuation of the that SCCT constructs such as capability and entrepreneurial identity may be higher and expectations should serve as the most important stronger than their expectations about the

112 JOURNALPFEIFER, OF SMALL ŠARLIJA, BUSINESS AND ZEKIC MANAGEMENT´ SUŠAC 11 appropriation of valuable outcomes. We there- Figure 3 fore implicate that SCCT key constructs such as Distribution of Graduates and self-efficacy or expectations or entrepreneurial aspirations are useful and highly valid for Undergraduates Along Identity explaining intentions. This provides an initial and Self-Efficacy Scores support for generalizability of the SCCT-based model in different contexts. However, further validation in the larger samples and across different countries would be beneficial. The other theory-driven assumption that we were interested in was that entrepreneurial self- efficacy, entrepreneurial identity aspiration, and positive outcome expectations fully or partially mediate the effect of personal or other external factors. Our evidence also supports the mediating role of self-efficacy and entrepreneurial identity. The self-efficacy construct seems to fully or partially mediate the impact of personal or educational factors. Students who are enrolled in graduate programs (either with a major in entrepreneurship or marketing, finances, management, etc.) have higher intentions, self-efficacy, and positive outcome expectation. Moreover, majors in an entrepreneurship program have the highest entrepreneurial iden- neurship education programs may introduce tity aspirations. Education factors (particularly the courses or pedagogies in which inspira- enrollment in graduate programs) have been tions, envisioning, positive attitudes, and emo- related to all main constructs in the model. In tions are emphasized, and a broader range of addition, key SCCT constructs are interactive career paths is welcome. and correlated. There are several implications This study has focused on a relatively narrow we would like to emphasize. Because these array of main effects and generalizability out- main constructs are correlated, we imply that comes. The more detailed and more robust entrepreneurial intentions of the students might exploration of the interaction between educa- be amplified through either one of these key tion and self-efficacy or identity construct goes constructs. The findings suggest that entrepre- beyond the aim of this paper. This study used neurship education programs should take into single constructs as the representative of multi- account the multidimensional nature of entre- item measurement instruments of main depen- preneurial intentions. Croatian students, for dent and independent variables. Future studies instance, show dispersion along the identity and could make more use of these multi-item instru- self-efficacy lines (see Figure 3). ments and use structural equation modeling in The design of entrepreneurship education, order to provide a more robust and accurate at least in Croatia, seems to be more focused on measure of direct and indirect effect, as well as the professional cognitive competences or interactions between the variables in the model. skills, and emphasize the educational impact Future research could also provide educators or on self-efficacy. The impact of education on program designers with more practical sugges- aspirations is highly relevant but often left tions, address the issue of the longitudinal untapped. The exposure of entrepreneurship effects, and explore the broader range of career majors to role models may be accountable to progress indicators (i.e., whether an individual their strong entrepreneurial identity aspiration. actually became an entrepreneur). Students may follow a high aspiration/low self- efficacy route, or low aspiration/high efficacy Conclusion route. We suggest that these less investigated The empirical evidence in this paper sug- interactions of key constructs be treated as gests the validity of the SCCT-based integrative evolving phenomena where education pro- entrepreneurial intentions model in Croatia. grams should exert stronger impact. Entrepre- The paper supports the implication that

12 JOURNALPFEIFER, OF SMALLSARLIJA, BUSINESS AND ZEKIC MANAGEMENT´ SUSAC 113 self-efficacy and entrepreneurial identity Fayolle, A., B. Gailly, and N. Lassas-Clerc mediate the educational impact on intentions. (2006). “Effect and Counter-Effect of Entre- The personal, situational, educational, and preneurship Education and Social Context social backgrounds of the Croatian students on Student’s Intentions,” Estudios de that have direct impact on entrepreneurial Economia Aplicada 24(2), 509–523. intentions are rather comparable with those Fitzsimmons, J. R., and E. J. Douglas (2011). hypothesized by theory or supported in previ- “Interaction Between Feasibility and Desir- ous research. However, the one particular ability in the Formation of Entrepreneurial construct—entrepreneurial identity—seems to Intentions,” Journal of Business Venturing call for more attention in the Croatian sample. 26, 431–440. Contrary to the SCCT-based models where self- Gerry, C., C. S. Marques, and F. Nogueira efficacy serve as the strongest mediator, in (2008). “Tracking Student Entrepreneurial Croatia, a more prominent role is given to the Potential: Personal Attributes and the Pro- identity construct. We assume that self-efficacy pensity for Business Start-ups after Gradua- and identity may be treated as an evolving tion in a Portugese University,” Problems phenomena, and that education can impact and Perspectives in Management 6(4), their formation. The paper contributes to the 45–53. field of SCCT-based intention models by filling Harrel, F. E. (2001). Regression Modeling Strat- the gap in the empirical evidence of the theo- egies with Applications to Linear Models, retical framework validity in the context of Logistic Regression and Survival Analysis. developing entrepreneurially young countries. Berlin: Springer. Henry, C., F. Hill, and C. Leicht (2005). “Entre- References preneurship Education and Training: Can Ajzen, I. (1991). “The Theory of Planned Entrepreneurship be Taught: Parts I–II,” Behavior,” Organizational Behavior and Education and Training 47(2), 98–111. Human Decision Processes 50, 179–211. 47(3), 158–169. Apte, C., and S. Weiss (1997). “Data Mining Kolvereid, L., and E. Isaksen (2006). “New with Decision Trees and Decision Rules,” Business Start-up and Subsequent Entry into Future Generation Computer Systems 13, Self-Employment,” Journal of Business Ven- 197–210. turing 21, 866–885. Bandura, A. (1986). Social Foundations of Krueger, N. F., M. D. Reilly, and A. L. Carsrud Thought and Action: A Social Cognitive (2000). “Competing Models of Entrepreneur- Theory. Englewood Cliffs, NJ: Prentice-Hall. ial Intentions,” Journal of Business Ventur- Bilic´, I., A. Prka, and G. Vidovic´ (2011). “How ing 15, 411–432. Does Education Influence Entrepreneurship Krueger, N. F., Jr. (2000). “The Cognitive Infra- Orientation? A Case Study of Croatia,” Man- structure of Opportunity Emergence,” Entre- agement 16(1), 115–128. preneurship Theory and Practice 24(3), Brockner, J., E. T. Higgins, and M. B. Low 5–23. (2004). “Regulatory Focus Theory and the Krueger, N. F., Jr., and D. V. Brazeal (1994). Entrepreneurial Process,” Journal of Busi- “Entrepreneurial Potential and Potential ness Venturing 19, 203–220. Entrepreneurs,” Entrepreneurship Theory Carr, J. C., and J. M. Sequeira (2007). “Prior and Practice 18, 91–104. Family Business Exposure as Intergenera- Kružic´, D., and I. Pavic´ (2010). “Students’ Entre- tional Entrepreneurial Intent: A Theory of preneurial Characteristics: Empirical Evi- Planned Behavior Approach,” Journal of dence from Croatia,” The Business Review Business Research 60, 1090–1098. 14(2), 216–221. Carter, N. M., W. B. Gartner, K. G. Shaver, and Lent, R. W., S. D. Brown, and G. Hackett E. J. Gatewood (2003). “The Career Reasons (1994). “Toward a Unifying Social Cognitive of Nascent Entrepreneurs,” Journal of Busi- Theory of Career and Academic Interest, ness Venturing 18(1), 13–39. Choice, and Performance,” Journal of Voca- Farmer, S. M., X. Yao, and K. Kung-Mcintyre tional Behavior 45, 79–122. (2009). “The Behavioral Impact of Entrepre- ——— (2002). “Social Cognitive Career neur Identity Aspiration and Prior Entrepre- Theory,” in Career Choice and Development, neurial Experience,” Entrepreneurship 4th ed. Eds. D. Brown and Associates. San Theory and Practice 35(2), 1–29. Francisco, CA: Jossey-Bass, 255–311.

114 JOURNALPFEIFER, OF SMALL ŠARLIJA, BUSINESS AND ZEKIC MANAGEMENT´ SUŠAC 13 Liñán, F., and Y.-W. Chen (2009). “Develop- Shane, S., and S. Venkataraman (2000). “The ment and Cross-Cultural Application of a Promise of Entrepreneurship as a Field of Specific Instrument to Measure Entrepre- Research,” Academy of Management Review neurial Intentions,” Entrepreneurship Theory 25(1), 217–226. and Practice 33(3), 593–617. Sheskin, D. J. (2009). Handbook of Parametric Liñán, F., D. Urbano, and M. Guerrero (2011). and Nonparametric Statistical Procedures. “Regional Variations in Entrepreneurial Washington, DC: Chapman and Hall/CRC. Cognitions: Start-up Intentions of University Thompson, E. R. (2009). “Individual Entrepre- Students in Spain,” Entrepreneurship neurial Intent: Construct Clarification and and Regional Development 23(2–4), 187– Development of an Internationally Reliable 215. Metric,” Entrepreneurship Theory and Prac- McGee, J., M. Peterson, S. Mueller, and J. M. tice 33(3), 669–694. Sequeira (2009). “Entrepreneurial Self- Winkel, D., J. Vanevenhoven, and K. Ehrhardt Efficacy: Refining the Measure,” Entrepre- (2011). “An Organizing Framework for neurship Theory and Practice 33(4), 963– Entrepreneurial Intentions Research: An 988. Integration and Extension of Dominant Nabi, G., and F. Linan (2011). “Graduate Entre- Intent-Based Models Using Social Cognitive preneurship in the Developing World: Inten- Career Theory,” USASBE 2011 Proceedings. tions, Education and Development,” http://sbaer.uca.edu/research/USASBE/ Education and Training 54(5), 325–334. 2011/PaperID72.pdf (accessed October 20, Questier, F., R. Put, D. Coomans, B. Walczak, 2012). and Y. Vander Heyden (2005). “The Use of Yohannes, Y., and P. Webb (1999). “Classifica- CART and Multivariate Regression Trees for tion and Regression Trees, CART: A User Supervised and Unsupervised Feature Selec- Manual for Identifying Indicators of Vulner- tion,” Chemometrics and Intelligent Labora- ability to Famine and Chronic Food Insecu- tory Systems 76, 45–54. rity,” Microcomputers in Policy Research 3, Schwarzer, R., and M. Jerusalem (1995). “Gen- 1–50. eralized Self-Efficacy Scale,” in Measures in Zellweger, T., P. Sieger, and F. Halter (2011). Health Psychology: A User’s Portfolio, Causal “Should I Stay or Should I Go? Career Choice and Control Beliefs. Eds. J. Weinman, S. Intentions of Students with Family Business Wright, and M. Johnston. Windsor, England: Background,” Journal of Business Venturing NFER-NELSON, 35–37. 26, 521–536. Segal, G., D. Borgia, and J. Schoenfeld (2002). Zhao, H., G. E. Hills, and S. E. Seibert (2005). “Using Social Cognitive Career Theory to “The Mediating Role of Self-Efficacy in the Predict Self-Employment Goals,” New Development of Entrepreneurial Intentions,” England Journal of Entrepreneurship 5, The Journal of Applied Psychology 90(6), 47–56. 1265–1272.

14 JOURNALPFEIFER, OF SMALLSARLIJA, BUSINESS AND ZEKIC MANAGEMENT´ SUSAC 115 Appendices Appendix 1. Descriptive Statistics—Multi-Items Measurement Instruments and Scales

Description of Multi-Item Measurement Scale Mean Cronbach’s α Instruments (S.D.)

Personal, situational, contextual factors General self-efficacy Schwarzer and Jerusalem (1995) 1 = strongly disagree 3.79 (0.51) 0.85 10 items aggregate 5 = strongly agree For example: I can always manage to solve difficult problems if I try hard enough . . . Family business Carr and Sequeira (2007) 1 = detracted 1.79 (4.99) exposure added to 3 items aggregate 3 = added to personal confidence Exposure to the businesses started confidence by parents/guardians, siblings, grandparents added or detracted to your own confidence Personal business Carr and Sequeira (2007) 1 = detracted 2.25 (3.73) exposure added to 2 items aggregate 3 = added to personal confidence Personal paying/nonpaying confidence position at a new business Social norms Partly based on Kolvereid and Response scale: Referral 12.05 (6.47) Isaksen (2006) opinion: From extremely 6 items weighted aggregate negative to extreme Referral opinions on favorability Referral importance: of starting a business are Not important to extremely weighted with the referral important importance for respondents. After that, the statements were aggregated. The higher the value, the more positive and more important the social norms are. Main independent Entrepreneurial Farmer, Yao, and Kung-Mcintyre 1 = strongly disagree 3.35 (0.89) 0.94 identity (2009) 5 = strongly agree 6 items aggregate I often think about becoming an entrepreneur . . . Entrepreneurial McGee et al. (2009) Response scale: 0 to 100 67.41 (15.08) 0.85 self-efficacy 20 items aggregate percent “How much confidence do you have in your ability to . . . come up with a new idea for a product or service on your own ...” Entrepreneurial Partly based on Krueger (2000) 1 = not expected at all 5.64 (1.12) 0.86 outcome 4 items 7 = very much expected expectations “To what extent do you expect to achieve . . . Financial rewards . . . Independence . . .” Main dependent Entrepreneurial Thompson (2009) 1 = very untrue 4.21 (1.09) 0.73 intentions 6 items aggregate 7 = very true Thinking of yourself, how true is it that you . . . (search for opportunities, money, learning, plan . . .)

116 JOURNALPFEIFER, OF SMALL ŠARLIJA, BUSINESS AND ZEKIC MANAGEMENT´ SUŠAC 15 Appendix 2. Descriptive Statistics—Personal Factors, Situational and Educational Factors

Variables Measurement Frequencies Instrument/Scale (Percent)

Gender 1 = Male 31.93 2 = Female 67.84 Minority 1 = Yes 8.59 2 = No 82.38 3 = Dk 9.03 Family wealth (SES) 1 = Below 4.41 2 = Average 84.58 3 = Above average 7.71 4 = Nr 3.30 Area 1 = Urban 38.33 2 = Rural 22.91 3 = Suburban 13.44 4 = Mixed 22.69 5 = Dk 2.64 Marital status 1 = Partnered 15.86 2 = Not partnered 79.74 3 = Nr 4.40 Prior family business exposure 1 = Yes 46.26 Did your (parents/guardians, siblings, grandparents . . .) 2 = No 52.42 ever start a new venture? 3 = Dk 1.32 Prior entrepreneurial experience 1 = Yes 3.96 Have you started a business that is currently operating? 2 = No 96.04 Prior own business exposure 1 = Yes 51.98 Have you ever held a paying/nonpaying position in 2 = No 46.04 a new company or entrepreneurial venture? 3 = Dk 1.98 Student status 1 = Full time 91.85 2 = Part time 4.63 3 = Nr 3.52 Major ﬁeld of study: 1 = Graduates in entrepreneurship 5.95 2 = Other graduates 53.08 3 = Undergraduates 39.87 4 = Nr 1.00

Dk, do not know; Nr, not responded.

16 JOURNALPFEIFER, OF SMALLSARLIJA, BUSINESS AND ZEKIC MANAGEMENT´ SUSAC 117 Journal of Small Business Business Management Management 2016 2014 54(1), ••(••), pp. pp. 118–138 ••–•• doi: 10.1111/jsbm.12134 Effects of Relational Capabilities and Power Asymmetry on Innovativeness and Flexibility of Sub-Sahara Africa Small Exporting Firms by Margaret Jekanyika Matanda, Nelson Oly Ndubisi, and Ferry Jie

The objective of this study is to examine the effects of relational capabilities and exercise of power on innovativeness, flexibility, and performance of Sub-Sahara Africa small exporters. Data from a sample of 206 small fresh produce suppliers in Zimbabwe that had long-term relationships with retail buyers in export markets were analyzed using structural equation modeling. Findings suggest that long-term cooperation between small suppliers and buyers negatively influenced innovativeness and flexibility. However, trust and commitment, as well as dependence, positively impacted innovativeness of small suppliers but had negative effects on flexibility. Further, innovativeness negatively influenced firm performance, whereas a positive link emerged between flexibility and performance.

cooperative arrangements between firms that Introduction facilitate exchange and sharing of knowledge as To ensure survival in a highly volatile and well as resources with the intent of developing dynamic business environment, firms need to processes, products, or services (Gulati 1998). adapt and modify the rules of exchange as Popular types of strategic alliances include joint circumstances change (Sezern and Yilmaz ventures, research and development agree- 2007). Maintaining long-term relationships has ments, co-marketing contracts, and significant emerged as one of the tools of enhancing inter- buyer–supplier relationships (Das and Teng national competitiveness. Relational capabilities 2000). allow firms to pool resources and capabilities to Global market changes in dietary habits, achieve compatible goals that may not be easily health, and nutritional requirements have achieved individually (Lambe, Spekman, and resulted in a requirement for all supply chain Hunt 2002). Organizations choose strategic alli- participants to focus on customer needs and ance partners with resources that can be lever- desires (Stevens and Kennan 2000). Addition- aged to enhance their competitiveness (Hitt ally, the rise of private food safety and quality et al. 2000). Strategic alliances are voluntary standards has created new benchmarks for all

Margaret Jekanyika Matanda is senior lecturer and director of the Masters Program in the Department of Marketing, Monash University. Nelson Oly Ndubisi is professor of Marketing, Helsinki Business School, Finland and professor of Entrepreneurship, Covenant Business School, Nigeria. Ferry Jie is senior lecturer of Marketing and Logistics in the School of Business IT and Logistics, College of Business, RMIT University, Melbourne Australia. Address correspondence to: Ferry Jie, Business IT and Logistics, RMIT University, 19 Kirrama Street, Tarneit, Vic. 3029, Australia. E-mail: [email protected]. Margaret Jekanyika Matanda, Department of Marketing, Faculty of Business & Economics, Monash University, P.O. Box, 197, Caulﬁeld East, Vic 3145, Melbourne, Australia. E-mail: [email protected].

118 JOURNALMATANDA, OF SMALL NDUBISI, BUSINESS AND MANAGEMENT JIE 1 participants in global food chains (Ouma 2010). and a major driver of success in both small and Consequently, small exporters in developing large firms (Vermeulen, de Jong, and countries are establishing long-term relation- O’Shaughnessy 2005). According to Romero ships with large retail buyers and distributors in and Martínez-Román (2012), small firms may developed markets, thereby becoming players not have the pre-requisite resources and capa- in market-driven global value chains (GVCs) bilities required for innovation. Consequently, (Humphrey 2006; Pietrobelli and Rabellotti small firms cannot solely rely on their own 2007). These relationships enable small firms to competency developing capabilities and undertake value-adding activities such as bar tapping into the distinctive capabilities of other coding, packaging, and labeling, as well as supply chain partners is crucial for sustaining production of pre-prepared or ready-to-eat competitiveness (Bradley, Meyer, and Gao fresh produce that can be readily transferred 2006). from the farm to supermarket shelves (Daviron Literature suggests that there is also a need and Gibbon 2002; Dolan and Humphrey 2004). for small producers to learn from the leaders in Thus, to effectively participate in GVCs, small GVC (Gereffi, Humphrey, and Kaplinsky 2001). firms need to enhance capabilities and innova- Small firms need to develop an external orien- tion (Daviron and Gibbon 2002; Humphrey tation and the long-term relational capabilities 2006). that will enable them to access resources, and Although most prior research has viewed competencies, facilitating innovation and flex- interfirm relationships as beneficial, some ibility (Hadjimanolis 2000). These relational researchers argue relationship building may capabilities can extend a firms’ knowledge base not be advantageous to all firms under all con- and minimize costs, risk, and investments ditions (Corsaro and Snehota 2011). Other (Freel and Harisson 2006). researchers have specifically called for more Increasingly, buyers in the fresh produce empirical research into the relational and sector are searching for innovative suppliers that power dynamics in GVC. can use new technology and management tools Within GVC are lead firms that engage in to enable the supply chain to quickly deliver quasi-hierarchical relationships with upstream highly customized products to market (Gereffi, and downstream supply chain members that Humphrey, and Kaplinsky 2001). When faced are, to a high degree, reliant on the lead firm’s with environmental uncertainty, flexible organi- decisions (Altenburg 2006). These relationships zations quickly and effectively adjust output to result in power asymmetry. According to Le market demand, and thus create resources that Heron et al. (2001), although prior research has are crucial to attaining a competitive advantage focused on power relationships, limited atten- in dynamic and uncertain environments tion has been paid to imbalance of power from (Sanchez 1995). The acquisition and creation of the perspective of different players in the heterogeneous skills and competencies that supply chain such as retailers, farmers, and enable a firm to adapt to changes in the envi- processors. The role of power in relational ronment and enhance flexibility are crucial for capabilities has received limited attention from the survival of small firms (Wright and Snell relationship marketing and small business 1998). In a buyer’s market (where buyers have researchers. This research gap created the higher access to information about the end impetus for this study, which examines the customer), buyers tend to have more power than effect of long-term relational capabilities and suppliers. This power asymmetry has different the exercise of power on innovativeness, flex- implications on the innovativeness and flexibil- ibility, and performance of small export suppli- ity of different supply chain partners. In GVC, ers in Sub-Sahara Africa. there is also asymmetry in terms of the ability to As small businesses are increasingly faced define what counts as quality and as legitimate with environmental uncertainty resulting from behavior in the marketplace (Ponte and Gibbon fast changing technology, markets, and global 2005; Tallontire 2007). Further, an asymmetry in competition, the importance of innovation in skills and competence (as well as in access to creating and enhancing competitiveness has information) has been identified in most GVCs been widely acknowledged in prior research that are governed by multinational corporations (Cooper 2000; de Jong and Vermeulen 2006). based in developed markets (OECD 2008) Innovativeness is viewed as a core capability of similar to those being examined in the current dynamic organizations (Gatignon et al. 2002) study.

2 JOURNALMATANDA, OF SMALL NDUBISI, BUSINESS AND MANAGEMENT JIE 119 Thus, this study is important in an area that mance, and the research hypotheses are pre- has received limited attention in prior research sented. This is followed by an outline of the as it identifies, for small firms in developing methodology used to develop the sampling countries, predictors of innovativeness and frame, data collection procedures, and an flexibility. Although prior research has focused outline of how the measurements were on investigating the effect of relational vari- operationalized. Finally, the results are dis- ables on relational outcomes, limited attention cussed and the article concludes by outlining has been directed to the effect of relational implications, limitations, and areas for future capabilities and other firm performance attri- research. butes, such as flexibility and innovativeness. As the small business sector forms an integral part Theoretical Background of most countries’ economies (Street and and Literature Meister 2004), especially in those that are Both the capabilities and relational orienta- developing and emerging, research in the tion approaches (Knight and Cavusgil 2004; sector is important to economic development. Leonard-Barton 1992) provide a useful theoreti- Further, the small business sector is the fastest cal underpinning for analyzing the association growing sector in most economies (Tagliavini, between long-term relational capabilities, Ravarini, and Antonelli 2001). innovativeness, and flexibility in small firms. Given the resource constraints of small The capabilities approach is anchored in the firms, the development of an external orienta- resource-based view. According to the tion and long-term relational capabilities pro- resource-based view, resources are heteroge- vides the access to resources and capabilities neous and immobile (Conner 1991), and this needed to compete with more powerful partici- enables resource-rich organizations to outcom- pants, especially in turbulent markets (Carson, pete their competitors. As a result, small Gilmore, and Rocks 2004). This issue is critical firms establish relationships to access scarce in developing countries, where a lack of insti- resources and enhance their competitive advan- tutional support can compound environmental tage. Such relational capabilities can be rare, uncertainty (Ghauri, Lutz, and Tesform 2003). inimitable, and not substitutable at company Therefore, an important strategic asset is the level (Kale, Dyer, and Singh 2002). ability to develop and manage relational capa- When firms develop long-term relationships bilities with supply chain partners whose with firms with complementary resources resources can facilitate performance (Awuah and capabilities, they not only achieve 2001). Kaynak and Kara (2001) argued that superior competitive advantage but also innovative long-term relationships designed to enhance commitment and trust (Sarkar et al. meet the needs of specific target markets and 2001). According to the resource-based view, facilitate identity preservation are especially an organization’s capabilities are its capacity to important for the success of small suppliers organize, manage, coordinate and undertake from developing countries. Indeed, literature specific activities (Teece, Pisano, and Shuen indicates such long-term relationships are 1997) or to deploy resources, using organiza- crucial in GVC that link small exporters in Latin tional processes to attain desired outcomes America, Sub-Sahara Africa, and South and (Day 1994). Eastern Asia as well as retailers and distributors A capability is a knowledge system com- in the United States and Europe (Dolan and posed of complementary behavior and abilities, Humphrey 2000, 2004; Gulati et al. 2007; expressed through organizational processes Pietrobelli and Rabellotti 2010; Reardon et al. that enable organizations to respond to market 2009). For instance, small fresh produce changes (Leonard-Barton 1992). Complemen- exporters in Kenya, South Africa, Ghana, and tary capabilities are crucial in attaining syner- Zimbabwe have become participants in the gistic alliance outcomes (Dyer and Singh 1998; United Kingdom supermarket-driven GVC Emden, Calantone, and Droge 2006) because (Daviron and Gibbon 2002; Humphrey 2006). when appropriate complementary resources The article is structured as follows. First, our and assets are combined, the core competen- theoretical framework, the resource-based view cies of partnering organizations are enhanced. is outlined. A review of literature on power, Indeed, a firm’s innovativeness and flexibility relational capabilities, trust and commitment, can also be viewed as its organizational capa- innovativeness, flexibility, and business perfor- bilities, given they encompass behaviors that

120 JOURNALMATANDA, OF SMALL NDUBISI, BUSINESS AND MANAGEMENT JIE 3 enable an organization to access resources types of products that are offered to customers and to enhance customer responsiveness. (Ebben and Johnson 2005). Research on flex- Relational capabilities also complement techni- ibility has mostly focused on larger firms, rather cal capabilities and increase the possibility of than small ones (Ebben and Johnson 2005). innovation within partnering firms (Tyler Within the fresh produce sector, flexibility is 2001). also closely associated with innovation, denot- ing a firm’s ability to provide made-to-order Firm Performance products that are unique to individual custom- To quickly respond to customer needs and ers or groups (White 2000). Relational capabili- the fast chaining business environment, orga- ties allow the involvement of supply chain nizations in developed countries are continu- partners and end users in the process of new ally establishing supply chain relationships product development, thereby speeding-up with small suppliers in developing countries product development and innovation (Salomo, (Daviron and Gibbon 2002; Humphrey 2006). Steinhoff, and Trommsdorff 2003), as well as The performance of small exporters in GVCs flexibility (Ebben and Johnson 2005). has major repercussions on the performance of Organizational innovation is an outcome of their supply chain partners in both developing organizational processes (Crawford and Di and developed countries. Thus, there is need to Benedetto 2000; Kunz, Schmitt, and Meyer understand the performance of all players in 2011). Innovativeness is a notion of openness GVCs so as to identify where local upgrading is to ideas and, as a characteristic of a firm’s required or hindered (Schmitz and Knorringa culture, promotes new solutions (Hurley and 2000). In this study, just as in extant small firm Hult 1998) and determines the degree of open- research firm performance has been used as a ness to new ideas, processes, or products dependent variable (see Lau, Man, and Chow (Damanpour 1991; Hurley and Hult 1998). 2004; Sadler-Smith et al. 2003; Swierczek and Innovativeness is reflected in an organization’s Ha 2003). Small business performance can be behavior when its faces change or the adoption assessed by such measures as terms of financial of new processes, products, or services (Hult, performance, sales, market share, innovation, Hurley, and Knight 2004; Kunz, Schmitt, and and human resources management (HRM) per- Meyer 2011). Thus, innovativeness within an formance. However, assessing the performance organization reflects the firm’s fundamental of small firms is rather problematic because of openness to break away from established pro- inadequate data availability and data reliability cedures (Kimberly 1979). problems (Acar 1993), and an unwillingness, or For the fresh produce sector, process inno- inability, to provide the desired information vation involves the ability of a supplier to (Birley and Westhead 1990). Performance can undertake year-round production, product be measured in absolute terms, relative to com- assortment and customization, as well as the petitors or in terms of changes from a previous ability to use specialized production and pack- period and may be based on either objective or aging techniques and provide differentiated subjective measures (Sapienza et al. 1988). products. Individual small firms are seldom able to innovate independently, therefore rela- Innovativeness, Flexibility, and Interfirm tional capabilities and strategic alliances are Relational Capabilities crucial in their innovation enhancement (Freel Researchers have observed that supply chain 2003; Powell, Koput, and Smith-Doerr 1996) relationships can enhance business perfor- and flexibility. Indeed, research suggests that mance through flexibility, innovativeness, small firms used relational capabilities to return on assets, return on investment, and manage and respond to uncertain environ- market share growth (Monczka et al. 1998). In ments (Dolan, Humphrey, and Harris-Pascal this sense, flexibility refers to an organization’s 1999). maneuvering capability (Alpkan, Yilmaz, and Powell, Koput, and Smith-Doerr (1996) sug- Kaya 2007), or its capability to quickly identify gested that innovation and learning occur more and respond to changes in the external envi- in long-term relationships than within indi- ronment (Shimizu and Hitt 2004). Flexibility in vidual firms. Long-term relationships enable the fresh produce sector comes not only from small firms to compete effectively by generat- the ability to meet variation in quantity of ing intelligence and accessing information, product provided but also from variation in the resources, and technologies with, and from,

4 JOURNALMATANDA, OF SMALL NDUBISI, BUSINESS AND MANAGEMENT JIE 121 trading partners (Gilmore, Carson, and Rocks less stability, more conflict, and consequently 2006). Within food chains, relational capabili- lower relationship quality, symmetric depen- ties help firms manage new product develop- dence structures foster improved relationship ment by rationalizing product ranges, quality. identifying new opportunities, and confirming Literature suggests that flexible firms can new ideas and innovation (Carson, Gilmore, quickly and easily change procedures and prac- and Rocks 2004; O’Donnell et al. 2001). Gener- tices in response to diverse market and envi- ating benefits for individual small firms, long- ronmental changes (Rowe and Wright 1997). term relational capabilities create chains of For instance, Matanda (2002) found some small distinctive competencies through synergy and suppliers indicated that their buyers sold expanded power (Gulati, Nohria, and Zaheer produce that exceeded their contractual quan- 2000). tities on consignment. Given that relational capabilities are the Thus, we hypothesize the following. organizational assets that facilitate competitiveness in developing and implementing joint H1: In a market characterized by power asym- activities with external partners (Capaldo metry, exercise of power by buyers will 2007), they “lubricate” effective interfirm rela- negatively influence the flexibility of tionships (Hunt, Arnett, and Madhavaram suppliers. 2006) and knowledge transfer between trading partners, stimulating innovation and enhanc- H2: The flexibility of small supplier firms will ing flexibility. Thus, it is beneficial for small positively influence their performance. firms to have long-term relationships with significant supply chain partners, whose The Exercise of Power in Interfirm resources can facilitate market performance Relationships and Innovativeness (Awuah 2001). Innovativeness is one of the most important Various dimensions of relational capabilities drivers of business performance (Hult, Hurley, such as trust and commitment, long-term ori- and Knight 2004). Geroski and Machin (1992) entation, exercise of power, bonding, depen- found that innovative firms often outcompeted dence, commitment, conflict handling, and noninnovators. Inextricably linked to firm per- cooperation have been identified as being formance (Freel 2000), innovativeness is crucial important to enhancing strategic alliances and to attaining competitiveness (Calantone, relational capabilities. However, this study is Cavusgil, and Zhao 2002). However, for small limited to the effects of cooperation, trust and firms with limited resources and capital, commitment, dependence, and power on the innovativeness can be costly and risky (Sivadas innovativeness, flexibility, and performance of and Dwyer 2000). For the fresh produce sector, small exporting firms. These relational dimen- innovation has resulted in products (such as sions have been selected as crucial for small baby corn and mini-carrots) tailored to specific fresh produce exporters. end users in global markets. Thus, we hypothesize the following.

Exercise of Power in Interfirm H3: In a market characterized by power asym- Relationships and Flexibility metry, the exercise of power by buyers will Power, in the context of this article, is the negatively influence the innovativeness of ability of one actor to influence another to suppliers. perform in a manner that he/she would not have otherwise done (Emerson 1962). Accord- Innovativeness is allegedly one of the most ing to the resource-based view, alliance part- important factors for a firm’s survival, growth, ners with control of resources have power over and competitiveness. Hult, Hurley, and Knight other firms seeking those resources (Pfeffer (2004) contended that innovativeness is one of and Salancik 1978). An imbalance in power in the most important predictors of firm perfor- long-term interfirm relationships can create mance. According to Damanpour and Evan opportunities to act opportunistically and exer- (1984), innovativeness can also help organiza- cise coercion (Ireland and Webb 2007). Rokkan tions maintain or improve performance and Haughland (2002) suggested that although levels. However, Petkova (2006) found the asymmetric relationships are associated with innovativeness of suppliers in developing

122 JOURNALMATANDA, OF SMALL NDUBISI, BUSINESS AND MANAGEMENT JIE 5 countries was, at times, inhibited by coopera- H5: In a market characterized by power asym- tion between buyers and small suppliers in metry, long-term cooperation with buyers GVCs. Tripsas and Gavetti (2000) ascertained will negatively influence the innovativeness that innovation can enhance core competence of small supplier firms. in one firm and inhibit it in another. Despite the lack of consensus on the link between H6: In a market characterized by power asym- innovativeness and performance, it is generally metry, long-term cooperation with buyers still considered one of the major determinants will ultimately negatively influence the flex- of firm performance (Cooper 2000). ibility of small supplier firms.

H4: Innovativeness of small supplier firms will Interfirm Trust and Commitment positively influence their performance. In business organizations, trust and commitment are viewed as the most crucial anteced- Interfirm Cooperation ents to cooperative behavior (Cannon et al. Interfirm cooperation can enable organiza- 2010). Trust is a general expectancy that the tions to access new markets, technology, prod- word of an individual (or organization) can be ucts, and skills as well as provide the relied on (Rotter 1967). Thus, trust is the will- information necessary for enhancing customer ingness to rely on a trading partner in whom satisfaction (Palmer 2000). Cooperation facili- one has confidence (Ganesan 1994; Mariotti tates idiosyncratic investments (Rindfleisch and 1999; Monczka et al. 1998; Morgan and Hunt Heide 1997) that are crucial for maintaining 1994; Spekman, Kamauff, and Myhr 1998). and sustaining competitiveness of players in Overall, trust is the degree to which partners the fresh produce sector. Cooperation between perceive each other as credible and benevolent suppliers and buyers can facilitate adaptation (Doney and Cannon 1997; Ganesan 1994; and innovation to suit buyer requirements Kumar, Scheer, and Steenkamp 1995) and is (Metcalf, Frear, and Krishnan 1992). Conse- expected to have a positive effect on the quently, within fresh produce chains, suppliers degree of collaboration in supply chain rela- and buyers also need to collaborate to raise the tionships. Here commitment is defined as the large capital outlays needed. For example, pro- willingness of each partner to exert effort in ducers from Sub-Saharan African countries support of the relationship. Research indicates (such as Ghana, Kenya, Zimbabwe, and South that commitment in interfirm relationships pro- Africa) have cooperated and made joint invest- motes motivation, as perceptions of equity that ments with buyers in Australia, UAE, Europe, evolve from the involvement (White and Lui and Japan (Matanda 2002; Takane 2004). Com- 2005) can improve firm performance (De peting in global food markets is a complex Clercq and Sapienza 2006; Wu and Cavusgil undertaking. As supply chain partners, small 2006). suppliers from developing countries have to Trust and commitment have also been iden- continually react to the demands of wealthier tified as influencing flexibility and firm perfor- and more selective consumers in developed mance (Mirani, Moore, and Weber 2001; Mohr markets (Regmi and Gehlhar 2005). Idiosyn- and Spekman 1994). It is clear that real collabo- cratic and relationship-specific investments ration in supply chain relationships requires can, however, inhibit suppliers’ flexibility as meaningful commitment and trust in interfirm they are unable to switch to other trading part- relationships. Commitment and trust determine ners and alternative uses (Altenburg 2006). the degree to which each party believes it can Subroto, Sivakumar, and Wilkinson (2004) rely on the integrity of the promise offered by argued that the generation of innovation the other. Based on these definitions, trust and in supply chain relationships is a consequence commitment have also been identified as sig- of the interaction between buyers and sellers nificant factors in influencing the manner in and that internal and external relationship which small business managers perceive the factors, such as an asymmetry of power, can potential benefits of long-term alliances moderate such interaction and innovativeness. (Brunetto and Farr-Wharton 2007). In addition, The low bargaining power of small suppliers trust is a key factor in fostering relational capa- may, in fact, reduce innovativeness and bilities (Gilmore, Carson, and Rocks 2006) as it inhibit flexibility. Thus we hypothesize the can alleviate conflict and facilitate mutual following. understanding and bilateral communication in

6 JOURNALMATANDA, OF SMALL NDUBISI, BUSINESS AND MANAGEMENT JIE 123 interfirm relationships (Cullen, Johnson, and Although retailers can exercise their power Sakano 2000; Voss et al. 2006). According to by opening or restricting market access, end Ahuja (2000), interfirm trust is also frequently manufacturers can exercise their power by assumed as central to the acquisition of inno- withholding, imposing the price of a product, vative capabilities as integral to interfirm or dictating the terms by which the product is relationships. traded (Coombs, Harvey, and Tether 2001). It Although it is argued that building trust and is, for instance, an asymmetric power relation- commitment in business relationships will ship when retailers control suppliers’ access to enhance firm performance, some researchers real-time information on sales of their own claim that asymmetric relationships are more products (to monitor product flows and sales dysfunctional than symmetric interdependence trends), and determine a product’s access to its (Kumar, Scheer, and Steenkamp 1995). They final market. Such information asymmetry can argued that trust and commitment decline inhibit suppliers’ operational flexibility. The lit- with increasing interdependence asymmetry, erature suggests that too much dependence (on whereas interfirm conflict increases. Both trust either buyers or suppliers) can inhibit innova- and commitment facilitate learning and the tion by small players in the supply chain exchange of tacit knowledge (Kale, Dyer, and (Romero and Martínez-Román 2012). This sug- Singh 2002; Liu, Luo, and Liu 2009) that can gests that dependence may result in power facilitate innovativeness. The literature also asymmetry that can negatively influence suggests that trust and commitment can limit innovativeness and flexibility. Thus, we mobility by preventing exploration of alterna- hypothesize the following. tive opportunities (Wu and Cavusgil 2006). This lock in effect reduces the flexibility of supply H9: In a market characterized by power asym- chain partners such as small exporters in GVCs. metry, dependence on buyers will negatively Thus, we hypothesize the following. influence the innovativeness of small supplier firms. H7: In a market characterized by power asymmetry, trust and commitment in buyers will H10: In a market characterized by power negatively influence the innovativeness of asymmetry, dependence on buyers will small supplier firms. negatively influence the flexibility of small supplier firms. H8: In a market characterized by power asymmetry, trust and commitment in buyers will The stated hypothesized relationships are negatively influence the flexibility of small shown in a conceptual model in Figure 1. supplier firms. Research Methodology Dependence Research Setting and According to the resource-based view, firms Sampling Procedures are unable to internally generate all the A sampling frame of 595 small horticultural resources that are needed for survival (Pfeffer exporters was drawn from the data supplied by and Salancik 1978). An increased level of depen- the government department in charge of exten- dence between buyers and suppliers leads to an sion services (AGRITEX) and from the horticul- increase in the level of commitment to the tural exporters association in Zimbabwe. Using relationship (Dwyer, Schurr, and Oh 1987). Stra- simple random sampling technique, 234 firms tegic symmetry is determined by the unique mix in the fresh produce sector were interviewed of valuable resources needed to achieve alliance face to face. Twenty-eight of the questionnaires success (Sarkar et al. 2001). However, asymmet- had been completed by members of agricul- ric resource complementarity can lead to power tural cooperatives; this resulted in 206 usable imbalance. Wetzels, de Ruyer, and van Birgelen interview schedules. The definition of small (1998) specifically equate dependence with firms varies across countries and can be based asymmetry in interfirm relationships. In situa- on the employees, structural characteristics, tions where some supply chain members have industry type, company age, total turnover, or fewer resources and are relatively small, depen- total assets (see Brooksbank 1991; Street and dence would most likely result in an exercise of Cameron 2007). In this small firms were power and control. defined as independently owned and operated

124 JOURNALMATANDA, OF SMALL NDUBISI, BUSINESS AND MANAGEMENT JIE 7 Figure 1 suggested by Churchill (1979), the constructs Hypothesized Relationships and measurement items were generated from a review of pertinent literature. Most of the scale between Relational Capabilities, items were adapted from existing literature and Firm Innovativeness, interviews were held with business managers Flexibility and Performance and academics in Zimbabwe to help in the scale adaptation and development process. To ensure questionnaire equivalence, the questionnaire was translated into Shona (a local language) and back translated into English.

Cooperation Trust and Commitment Dependence Exercise of Power Firm Performance. Most prior research on small ﬁrms uses subjective performance measures (Chandler and Hanks 1993; Conant, Mokwa, and Varadarajan 1990). Some researchers have reported a strong correlation between self-reported, perceived measures, and objec-

Innovativeness Flexibility tive measures of ﬁrm performance (Powell 1994). Firm performance was operationalized using subjective measures of performance (market share, sales growth, and proﬁtability) adapted from Baker and Sinkula (1999).

Firm Respondents were asked to compare current Performance marketing and financial performance with that of the previous two years, as recommended by Baker and Sinkula (1999). These measures were used as there were easy for the respondents to comprehend and recall. enterprises that are not dominant in their field and have relatively fewer resources than other Flexibility. To assess flexibility, five items bor- companies in that market (Street and Cameron rowed from Lusch and Brown (1996) were uti- 2007). lized to measure the interviewees’ perception Managers and owners of these firms were of their organization’s adaptability to fluctua- requested to identify two of their major buyers tions in product supply and demand, and the they had been doing business with for over two availability of alternative markets. years and to respond to questions with respect to one of these buyers with whom they had the Innovativeness. Calantone, Cavusgil, and longest relationship. All suppliers exported Zhao (2002) five-item measure was used to most (75 percent or more) of their produce to assess organizational innovativeness. the Netherlands, United Kingdom, Australia, Japan, or UAE. Interfirm Cooperation. Cooperation was measured using a five-item scale adapted from Interview Schedule Development Doney and Cannon (1997). Closed questions were used. The schedule was designed in English and then translated to Interfirm Trust and Commitment. This con- Shona (the major indigenous language in Zim- struct was assessed using measurement items babwe) by language specialists at a local uni- from Cannon et al. (2010). versity. To achieve translation equivalence and linguistic consistency, different people were Exercise of Power. A five-item scale adapted used for the translation and back translation, as from Brown, Lusch, and Nicholson (1995) was suggested by Douglas and Craig (1984). used to assess exercise of power.

Development of Measures Dependence. Dependence was operationaliz- Established psychometric approaches were ed using a six-item scale, derived from Ganesan used to develop the measurements scales. As (1994), which focused on the perceived impact

8 JOURNALMATANDA, OF SMALL NDUBISI, BUSINESS AND MANAGEMENT JIE 125 of discontinuing the business relationship with intended factors (standardized factor loadings the buyer and the degree of profitability attrib- exceeding 0.5), suggesting convergent validity. uted to the buyer by the supplier. The average variance extracted (AVE) was used Firm performance measures were anchored to assess discriminant validity. As suggested by by “1” = much better/much more/significant Fornell and Lackner (1981), discriminant valid- increase to “7” = much worse/much less/ ity is assumed if the AVEs of any two constructs significant decrease. The measures for the are greater than their shared variance. Reliabil- other constructs were on a seven-point Likert ity was assessed by calculating the composite scale anchored by “1” = strongly disagree to reliability and all were above 0.7 as recom- “7” = strongly agree. mended by Nunnally (1978). All the constructs, their measurement items, and the sources of the measures are shown in Common Method Variance Table 1. In survey research, the wording of the questionnaire, scale length and item structure Analysis and Results of the measurements can influence the Measure Purification and Validation response to questions and result in common To analyze the data, SPSS Version 20 (SPSS method bias or variance (Malhotra et al. 2006; Inc., Chicago, IL, USA) and AMOS Version 17 Podsakoff, MacKenzie, and Podsakoff 2012). (SPSS Inc., Chicago, IL, USA) were used. First, To test for common method variance, the exploratory factor analysis was used to assess marker variable technique was used (Lindell internal consistency of the measure. The valid- and Whitney 2001; Malhotra et al. 2006). The ity of the scales was established through the marker variable was conceptually and statisti- assessment of the unidimensionality of the cally unrelated to both our dependent and pre- measurement items of each scale (Gerbing and dictor variables. This marker variable was Anderson 1988). Confirmatory factor analysis introduced in the three CFA models. The χ2 (CFA) using AMOS was then used to assess difference tests were conducted comparing convergent validity of the operational measures models with and without a marker variable. (Churchill 1979). Models with a marker variable had worse To estimate the measurement models, the goodness-of-fit indices than those without. analysis was split into three parts. This proce- Thus, common method bias was assumed to dure is recommended when sample size is have limited effect in this study. small compared with the number of estimated parameters (Hair et al. 2002). The first CFA Results and Discussion model (measurement model 1) was estimated Following the assessment of the measure- for the relational capability variables (trust and ment models, the hypothesized relationships commitment, dependence, exercise of power, were tested using AMOS Version 6. The model and cooperation), see Table 1. A second CFA was estimated as structural, with individual model (measurement model 2) was estimated measurement items loading on latent con- for innovation and flexibility. structs, see Figure 2. The model produced a A third CFA model (measurement model 3) significant χ2 statistics, χ2(655) = 304, p ≤ .05. was estimated for firm performance, and this The other fit measures indicated adequate fit to also indicated a good fit between model data the sample data (RMSEA = 0.05; GFI = 0.90, sets. Convergent validity is assumed if the indi- AGFI = 0.88, CFI = 0.90). The standardized cator’s estimated pattern coefficient exceeds parameter estimates and t-values are reported 0.50 and the overall measurement model pos- in Table 2 and the significant relationships are sesses acceptable fit indices (Bagozzi, Yi, and shown in Figure 2. Phillips 1991). Indicating the model fitted the As predicted, exercise of power by buyers data well, the GFI, AGFI, and CFI were above over small suppliers has a negative effect on the 0.90 benchmark (Browne and Cudeck supplier innovativeness (β = −0.277, t = −3.12, 1993). The root mean squared error of approxi- p ≤ .01) and flexibility (β = −0.279, t = 3.36, mation (RMSEA) was below the 0.06 cut-off p ≤ .001), providing support for H1 and H3. limit, indicating good model fit (Hu and Bentler The results suggest that small fresh produce 1999). Table 1 shows the fit indices of the three suppliers view the exercise of power by large CFA models and also indicates that measure- buyers as inhibiting their innovation and flex- ment items are significantly loaded on the ibility. Most small suppliers had contracts with

126 JOURNALMATANDA, OF SMALL NDUBISI, BUSINESS AND MANAGEMENT JIE 9 Table 1 Source of Measure and Conﬁrmatory Factor Analysis Results

Construct and Source Operational Measures of Construct SFL t-value of Measureb

Measurement Model 1 Model fit indices: χ2(163) = 112; RMSEA = 0.001; Relational Capabilities GFI = 0.99; AGFI = 0.98; CFI = 0.99 Cooperationa Please use the following scale to indicate your extent (Doney and Cannon of agreement about how well each of the following 1997) statements is an accurate description of α = 0.77, AVE = 0.604 inter-organizational relationships between your firm and your major buyer. • We cannot sell the same volume without the 0.47 6.23 cooperation of this buyer.a 0.72 14.84 • This buyer always gives us the support we need 0.65 7.86 in times of difficulty. 0.35 4.82 • We have a good working relationship with this 0.66 7.67 buyer. 0.72 8.69 • This buyer is always understanding of our problems.a • We usually work together with this buyer to solve problems in our relationship • This buyer always acts in our interest. Dependencea • If our relationship was discontinued with this 0.39 5.70 (Ganesan 1994) buyer, we would have difficulty finding a 0.67 8.76 α = 0.74, AVE = 0.71 replacement.a 0.16 2.30 • This buyer is crucial for our future performance. 0.81 9.08 • Dealing with a new buyer would require 0.66 8.29 minimum adjustment in business.a • This is the most profitable buyer we can get. • Our organization generates high sales volume for this buyer. Exercise of Powera • Even if we disagree with this buyer, we have to 0.58 5.56 (Brown, Lusch, and comply with their request. 0.74 7.76 Nicholson 1995) • In case of disagreement, buyer could penalize us. 0.39 3.11 α = 0.71, AVE = 0.73 • This buyer is able to make decisions that can 0.57 5.28 a alter our profit levels. 0.44 4.04 • This buyer can adversely influence the way we operate. • This buyer has at times threatened to terminate the relationship.a Trust and Commitmenta • This buyer has been frank in dealing with us. 0.68 6.01 (Cannon et al. 2010) • We have invested substantial in this buyer 0.67 8.54 α = 70, AVE = 0.72 • Promises made by this buyer are honored 0.80 9.05 • This buyer provides us with all the information 0.56 5.31 we need to make decisions. 0.77 5.83 • The buyer has made sacrifices on our behalf in 0.89 6.20 the past. 0.76 7.84 • The buyer has changed their business operations to suit our requirements • We are committed to our relationship with this buyer.

10 JOURNALMATANDA, OF SMALL NDUBISI, BUSINESS AND MANAGEMENT JIE 127 Table 1 Continued

Construct and Source Operational Measures of Construct SFL t-value of Measureb

Measurement Model 2 Model fit indexes: χ2(20) = 17, RMSEA = 0.04; Innovativenessa GFI = 0.99, AGFI = 0.97, CFI = 0.95 (Calantone, Cavusgil, Please use the following scale to indicate your extent and Zhao 2002) of agreement about how well each of the following α = 70, AVE = 0.66 statements is an accurate description of introduction of new products, ideas, and processes in your firm. • Our firm is able to make new products that are 0.51 7.76 required by our customers. 0.56 7.73 • Our firm is always looking for new ideas on 0.63 9.65 production techniques. 0.82 12.52 • Our firm is keen to get information about new 0.76 11.32 production processes in our industry. • Our firm is keen to get information about new packaging and grading processes in our industry. • Our firm is keen to learn about new technology in our industry. Flexibilitya Please use the following scale to indicate your extent (Lusch and Brown 1996) of agreement about how well each of the following α = 0.74, AVE = 0.72 statements is an accurate description of how your organization is able to respond to changes in the marketplace. • Our firm is always searching for ways to change 0.65 5.76 products to meet the requirements of our buyers. 0.51 3.83 • Our firm seeks information products that are 0.87 8.82 needed to meet the requirements of our buyers.a 0.71 8.01 • Our firm is keen to supply the type of produce 0.66 6.89 our buyers want. • When some unexpected situation arises, we would rather work out a new deal with our buyer rather than hold them to the original terms. • Our firm expects to make adjustments in dealing with the supplier to cope with changing circumstances. Measurement Model 3 Model fit indices: χ2 = 14.98; df = 20; RMSEA = 0.01; Firm Performanceb GFI = 0.99; AGFI = 0.98; CFI = 0.99 (Baker and Sinkula Please use the following scale to indicate your extent 1999) of agreement about your firm performed this year α = 0.84, AVE = 0.68 compared with performance in the past two years on the following issues. i. Sales volume 0.85 10.21 ii. Profit margins 0.89 12.45 iii. Return on assets 0.77 8.44 aMeasures on seven-point Likert scale anchored on 1 = “strongly disagree”; 7 = “strongly agree.” bMeasure on seven-point Likert scale anchored on 1 = “much lower”; 7 = “much higher.” AVE = average variance extracted; GFI = goodness-of-fit index; AGFI = adjusted goodness-of-fit index; CFI = comparative fit index; RMSEA = root mean square of approximation; SFL = standardized factor loadings.

128 JOURNALMATANDA, OF SMALL NDUBISI, BUSINESS AND MANAGEMENT JIE 11 Figure 2 Results

C2 1 C3 Cooperation C4

C5 -0.735 C6 Flx1 Flx2 Flx3 Flx4 Flx5 Dp1 1

Dp2 -0.641 Flexibility 1 -0.423 Dp3 Dependence 0.101 1 pypp1 Dp4 0.315 Firm pypp2 Dp5 Performance pypp3 -0.446 0.196

Ts1 Innovativeness Ts2 0.492 1 Ts3 Inv5 1 Trust and Commitment -0.279 Inv1 Inv2 Inv3 Inv4 Ts4 Ts5 -0.277

Pw1

Pw2 Pw3 1 Exercise of Power Pw4

Pw5

export agents and followed specific prescrip- upstream members in developing countries. In tions from the buyers. This may reduce their line with prior research (Calantone and Dröge own creativity or innovativeness. Thus, power 2006), flexibility emerged as a positive predic- asymmetry may be seen as an inhibitor of inno- tor of small firm performance (β = 0.101, vation. Power imbalance can result in the use t = 2.65, p ≤ .05). Thus, H3 was supported. of coercive power that reduces flexibility as Flexibility enables small suppliers to meet small suppliers would put all their resources in variations in both quantity and types of ensuring the maintenance of contracts with produce. Flexible produce suppliers can con- powerful buyers. tribute to profitability by enhancing quality and These results support prior research that reducing product rejection and its negative con- suggests that relationship-specific investment sequences. Flexible suppliers are also able to and long-term relationships may reduce flex- ascertain alternative markets and viable uses ibility and strengthen the bargaining power of for rejected products, often increasing sales the party, which has not incurred such invest- volume and earning additional income. ment (Altenburg 2006). This view is echoed by Contrary to prediction (H4), innovativeness Gibbon (2003) who contends that the growing had negative effects on firm performance market power of downstream members of GVC (β = −0.196, t = 2.99, p ≤ .05). For fresh produce has inhibited high value-adding activities for supply chains, innovativeness implies continu-

12 JOURNALMATANDA, OF SMALL NDUBISI, BUSINESS AND MANAGEMENT JIE 129 Table 2 Results of Analysis

Hypotheses Constructs and Predicted Inﬂuence Beta t-value Conclusion

H1 In a market characterized by great power −0.279 −3.36** Supported asymmetry, exercise of power by buyers over small suppliers will negatively influence suppliers’ flexibility. H2 . . . flexibility of small supplier firms will 0.101 2.65* Supported positively influence firm performance. H3 . . . exercise of power by buyers over small −0.277 −3.12** Supported suppliers will negatively influence suppliers’ innovativeness. H4 . . . innovativeness of small supplier firms will −0.196 −2.99 Rejected positively influence firm performance. H5 . . . long-term cooperation with buyers will −0.641 −3.68*** Supported negatively influence the innovativeness of small supplier firms. H6 . . . long-term cooperation with buyers will −0.735 −2.79** Supported negatively influence the flexibility of small supplier firms. H7 . . . trust and commitment in buyers will 0.420 3.41*** Rejected negatively influence the innovativeness of small supplier firms. H8 . . . trust and commitment in buyers will −0.446 2.98** Supported negatively influence the flexibility of small supplier firms. H9 . . . dependence on buyers will negatively 0.315 3.41*** Rejected influence the innovativeness of small supplier firms. H10 . . . dependence on buyers will negatively −0.423 −2.70** Supported influence the flexibility of small supplier firms.

Model diagnostics: χ2(655) = 304, p ≤ .05; RMSEA = 0.05; GFI = 0.90, AGFI = 0.88, CFI = 0.90. *p ≤ .05. **p ≤ .01. ***p ≤ .001. ous change in products and processes to meet produce or using these expensive products customer needs. If the outcome deviates from (such as mange tout or baby corn) as animal these specific needs, buyers can reject the feed (Matanda 2002). The results suggest that produce, which means a negative effect on for small suppliers, innovativeness may be too supplier performance. Innovative suppliers demanding and unprofitable, especially when invest in equipment and facilities that reduce they have little or no influence on pricing or shrinkage, improve quality, processes, and end sale of the produce. shelf life. In Sub-Saharan Africa, for instance, We hypothesized with H5 and H6 that long- innovative small suppliers supplied ready-to- term cooperation between small suppliers eat vegetables to European supermarkets and their buyers negatively influences innov- (Takane 2004). When such highly customized ativeness and flexibility. The results suggest that products are rejected by the contracting buyers, the cooperation of small suppliers with their and small suppliers fail to get alternative export buyers had a strong negative influence buyers, some suppliers will resort to dumping on their innovativeness (β = −0.641, t = −3.68,

130 JOURNALMATANDA, OF SMALL NDUBISI, BUSINESS AND MANAGEMENT JIE 13 p ≤ .001) and flexibility (β = −0.735, t = −2.79, innovativeness is the ability to embrace ideas p ≤ .01). Thus, H5 and H6 were supported. and introduce new processes, adapting the The ever-changing customer demands of technology that enables them to meet customer export markets mean cooperation between needs. Since the success of small suppliers in small suppliers and their buyers results in most fresh produce chains depends on their ability to value-adding activities being pushed down the meet the product and quantity specifications of supply chain. Produce production, labeling, export buyers and end users (Barrett et al. and packaging specifications are determined by 1999), they often depend on their large buyers the buyer, and small suppliers must comply. In to provide the resources and information that some instances, large buyers prescribe, or even help develop new products and processes. dictate, the input, tools, and produce growing Consistent with H10, dependence was sig- processes, leaving small suppliers no opportu- nificantly negatively associated with flexibility nity for imagination or creativity. (β = −0.423, t = 2.70, p ≤ .01). Small firms in the As a result, in situations where a small fresh produce sector depend on buyers for export supplier closely cooperates with a single most inputs and equipment needed to enhance major export buyer, working on contract with produce quality, and respond to customer detailed produce specifications for highly cus- demands. But as these are usually under con- tomized produce, the flexibility of the small tractual agreements, they cannot be used for supplier is reduced. This results in the inverse other customers. This would explain the nega- relationship between cooperation and flexibil- tive link between dependence and flexibility. ity this study reveals. Our results indicate that trust and commit- Implications, Limitations, and ment have a significant positive effect on inno- Areas for Future Research vation (β = 0.420, t = 3.41, p ≤ .001), but are Our research results have implications for negatively associated with flexibility (β = firms working in or with supplier and buyer −0.446, t = −2.69, p ≤ .01), providing support for organizations in the fresh produce sector, as H8 and disproving H7. Trust and commitment the findings suggest that long-term interfirm encourage the flow of information between sup- relational variables have a differential effect on pliers and their buyers, and joint investments innovativeness and flexibility. that facilitate innovation. It was projected that in It is crucial for small fresh produce firms a relationship with an unfair distribution of to identify whether they want to pursue power and influence, trust in a more powerful innovativeness or flexibility, and then to focus partner may hinder innovation, but the outcome on specific relational variables and a relation- of the study shows otherwise. It is plausible that ship structure that contributes to their organi- because small suppliers have a limited knowl- zation’s objectives. Our research supports edge of the structure and practice of export earlier work that suggests small firms embed- markets, and may have few, or no, options ded in long-term relationships with a single or of alternative buyers, they rely on their relation- few buyers run the risk of neglecting their own ship with existing buyers to evolve new ideas dynamic capabilities (Stremersch, Wuyts, and and products. This is especially relevant Frambach 2001). As a result, small firms need for small exporters that serve the global to critically evaluate the advantages accruing supermarket-driven supply chains serving Euro- from the development of relational capabilities pean markets as those covered in this study. with specific buyers. As suggested by Freel However, although trust and commitment facili- (2003), both researchers and managers should tate joint investments in relationship-specific be cautious of recommending innovativeness to assets, this reduces supplier flexibility as assets small firms as a way of enhancing competitive- restricted to producing for the trusted buyer ness, as the associated costs may outweigh the cannot be used for other purposes. Our results benefits. The establishment and maintenance of suggest that although the trust and commitment relational capabilities come at a major cost, and of a buyer were thought to facilitate the goal should be relationships within GVC innovativeness, it did not actually enable the that have fairly distributed power between the small supplier to retain flexibility. small firm and its partners. Contrary to H9, dependence emerged as a Finally, buyers working with small busi- positive predictor of innovativeness (β = 0.315, nesses in the food sector need to move away t = 2.98, p ≤ .01). For suppliers in this sector, from the use of coercive power that is common

14 JOURNALMATANDA, OF SMALL NDUBISI, BUSINESS AND MANAGEMENT JIE 131 (see Dolan, Humphrey, and Harris-Pascal Impact,” European Journal of Development 1999; Hingley 2005). By investing in the Research 18(4), 498–521. enhancement of relational variables and reduc- Awuah, G. (2001). “A Firm’s Competence ing or neutralizing the effects of coercive Development Through Its Network of power, they will also benefit from the improved Exchange Relationships,” Journal of Busi- supplier innovativeness and flexibility. ness & Industrial Marketing 16(6/7), 574– This study advances the resource base view 599. and relational theory, and builds on previous Bagozzi, R. P., Y. Yi, and L. W. Phillips (1991). work by testing the effect on firm performance “Assessing Construct Validity in Organiza- of the ability to be flexible and innovative in a tional Research,” Administrative Science model that has received limited attention in Quarterly 36(3), 421–458. prior research. It further contributes to theory Baker, W. E., and J. M. Sinkula (1999). “The by empirically testing the model in small firms Synergistic Effect of Market Orientation and in a developing country in Sub-Sahara Africa. Learning Orientation on Organizational Per- Cunningham (2001) noted there is a serious formance,” Journal of the Academy of Mar- deficiency in any sort of research on marketing keting Science 27(4), 411–427. chains in Africa. Barrett, H. R., B. W. Ilbery, A. W. Brown, and T. The limitations of this study are derived Binns (1999). “Globalization and the Chang- from the fact that it was conducted in a spe- ing Networks of Food Supply: The Importa- cific context, where small fresh produce firms tion of Fresh Horticultural Produce from in a developing country have GVC relation- Kenya into the UK,” Transactions of the ships with large buyers operating in the Institute of British Geographers 24(2), 159– highly market-driven environments of Europe, 174. Japan, and Australia. Hence, the findings may Birley, S., and P. Westhead (1990). “Growth not be generalizable to small suppliers in and Performance Contrasts Between ‘Types’ other contexts. of Small Firms,” Strategic Management It should also be noted that the constructs Journal 11, 535–557. and measurement scales used in this study Bradley, F., B. I. Meyer, and Y. Gao (2006). were adapted from Western literature. “Use of Supplier-Customer Relationships by Although we attempted to minimize the poten- SMEs to Enter Foreign Markets,” Industrial tial problems associated with the cross-cultural Marketing Management 35(6), 652–665. adaptation of measurements, some differences Brooksbank, R. (1991). “Defining the Small in the conceptual meaning of the constructs Business: New Classification of Company could still exist. Size,” Entrepreneurship and Regional Devel- There is also the opportunity for future opment 3, 17–31. research to investigate the effect of other rela- Brown, J. R., R. Lusch, and C. Nicholson (1995). tional capability variables, such as benevolence “Power and Relationship Commitment: Their and bonding, on innovativeness and flexibility. Impact on Marketing Channel Performance,” Journal of Retailing 71(4), 363–392. References Browne, M. W., and R. Cudeck (1993). “Alter- Acar, A. C. (1993). “The Impact of Key Internal native Ways of Assessing Model Fit,” in Factors on Firm Performance: An Empirical Testing Structural Equation Models. Eds. K. Study of Small Turkish Firms,” Journal of A. Bollen and J. S. Long. Newbury Park, CA: Small Business Management 31(4), 86–92. Sage, 136–162. Ahuja, G. (2000). “Collaboration Networks, Brunetto, Y., and R. Farr-Wharton (2007). “The Structural Holes, and Innovation: A Longitu- Moderating Role of Trust in SME Owner/ dinal Study,” Administrative Science Quar- Managers’ Decision—Making about Collabo- terly 45(3), 425–455. ration,” Journal of Small Business Alpkan, L., C. Yilmaz, and N. Kaya (2007). Management 45(3), 362–387. “Market Orientation and Planning Flexibility Calantone, R., and C. Dröge (2006). “Supply in SMEs: Performance Implications and an Chain Flexibility: An Empirical Study,” Empirical Investigation,” International Journal of Supply Chain Management 35(3), Small Business Journal 25(2), 152–172. 16–24. Altenburg, T. (2006). “Governance Patterns in Calantone, R. J., S. T. Cavusgil, and Y. Zhao Value Chains and their Development (2002). “Learning Orientation, Firm Innova-

132 JOURNALMATANDA, OF SMALL NDUBISI, BUSINESS AND MANAGEMENT JIE 15 tion, and Firm Performance,” Industrial Cunningham, D. (2001). “Controlling the Marketing Management 31(6), 515–524. Marketing-Purchasing Interface: Resource Cannon, J. P., P. M. Doney, M. R. Mullen, and Development and Organizational Implica- K. J. Petersen (2010). “Building Long-Term tions,” Supply Chain Management: An Inter- Orientation in Buyer–Supplier Relationships: national Journal 6(5), 212–215. The Moderating Role of Culture,” Journal of de Jong, J., and P. Vermeulen (2006). “Deter- Operations Management 28(6), 506–521. minants of Product Innovation in Small Capaldo, A. (2007). “Network Structure and Firms: A Comparison Across Industries,” Innovation: The Leveraging of a Dual International Small Business Journal 24(6), Network as a Distinctive Relational Capabil- 587–609. ity,” Strategic Management Journal 28(6), Damanpour, F. (1991). “Organizational Innova- 585–608. tion: A Meta-Analysis of Effects of Determi- Carson, D., A. Gilmore, and S. Rocks (2004). nants and Moderators,” Academy of “SME Marketing Networking: A Strategic Management Journal 34(3), 555–590. Approach,” Strategic Change 13(7), 369–373. Damanpour, F., and W. M. Evan (1984). “Orga- Chandler, G. N., and S. H. Hanks (1993). “Mea- nizational Innovation and Performance: The suring the Performance of Emerging Busi- Problem of Organizational Lag,” Administra- nesses: A Validation Study,” Journal of tive Science Quarterly 29, 392–409. Business Venturing 8, 391–408. Das, T. K., and B. Teng (2000). “A Resource- Churchill, G. J. (1979). “A Paradigm for Devel- Based Theory of Strategic Alliances,” oping Better Measure of Marketing Con- Journal of Management 26(1), 31–61. structs,” Journal of Marketing Research Daviron, B., and P. Gibbon (2002). “Global 16(1), 64–73. Commodity Chains and African Export Agri- Conant, J. S., M. P. Mokwa, and P. R. culture,” Journal of Agrarian Change 2(2), Varadarajan (1990). “Strategic Types, Dis- 137–161. tinctive Marketing Competencies and Orga- Day, G. S. (1994). “The Capabilities of Market- nizational Performance: A Multiple Driven Organizations,” Journal of Marketing Measures-Based Study,” Strategic Manage- 58(October), 37–52. ment Journal 11(5), 365–383. De Clercq, D., and H. J. Sapienza (2006). Conner, K. (1991). “Historical Comparison of “Effects of Relational Capital and Commit- Resource-Based Theory and Five Schools of ment on Venture Capitalists’ Perception of Thought within Industrial Organization Eco- Portfolio Company Performance,” Journal of nomics: Do We Have a New Theory of the Business Venturing 21(3), 326–347. Firm?,” Journal of Management 17(1), 121– Dolan, C., and J. Humphrey (2000). “Gover- 154. nance and Trade in Fresh Vegetables: The Coombs, R., M. Harvey, and B. Tether (2001). Impact of UK Supermarkets on the African Analysing Distributed Innovation Processes: Horticulture Industry,” The Journal of Devel- A CRIC Position Paper, Centre for Research opment Studies 37(2), 147–176. on Innovation & Competition. Manchester: ——— (2004). “Changing Governance Patterns University of Manchester. in the Trade in Fresh Vegetables Between Cooper, R. G. (2000). “Doing It Right: Winning Africa and the United Kingdom,” Environ- with New Products,” IVEY Business Journal ment and Planning A 36, 491–509. 64(6), 54–60. Dolan, C., J. Humphrey, and C. Harris-Pascal Corsaro, D., and I. Snehota (2011). “Alignment (1999). “Horticultural Commodity Chains: and Misalignment in Business Relation- The Impact of the UK Market on African ships,” Industrial Marketing Management Fresh Vegetable Industry,” IDS Working 40(6), 1042–1054. Paper, 96, University of Sussex: Institute of Crawford, C. M., and C. A. Di Benedetto (2000). Development Studies. New Products Management, 6th ed. New Doney, P. M., and J. P. Cannon (1997). “An York: McGraw Hill. Examination of the Nature of Trust in Buyer– Cullen, J. B., J. L. Johnson, and T. Sakano Seller Relationships,” Journal of Marketing (2000). “Success Through Commitment and Research 61(2), 35–51. Trust: The Soft Side of Strategic Alliance Douglas, S. P., and S. C. Craig (1984). “Estab- Management,” Journal of World Business lishing Equivalence in Comparative Con- 35(3), 223–240. sumer Research,” in Comparative Marketing

16 JOURNALMATANDA, OF SMALL NDUBISI, BUSINESS AND MANAGEMENT JIE 133 Systems. Eds. E. Kaynak and R. Savitt. New Gerefﬁ, G., J. Humphrey, and R. Kaplinsky York: Praeger Publishers, 93–113. (2001). “Introduction: Globalisation, Value Dwyer, F. R., P. H. Schurr, and S. Oh (1987). Chains and Development,” IDS Bulletin “Developing Buyer–Seller Relationships,” 32(3), 1–8. Journal of Marketing 51(2), 11–27. Geroski, P., and S. Machin (1992). “Do Inno- Dyer, J. H., and H. Singh (1998). “The Rela- vating Firms Outperform Non-Innovators?,” tional View: Cooperative Strategy and Business Strategy Review 3(2), 79–90. Sources of Inter-Organizational Competitive Ghauri, P. N., C. H. Lutz, and G. Tesform Advantage,” Academy of Management (2003). “Using Business Networks to Solve Review 23(4), 660–679. Export Marketing Problems of Small and Ebben, J. J., and A. Johnson (2005). “Efﬁciency, Medium-Sized Manufacturing Firms from Flexibility, or Both? Evidence Linking Strat- Developing Countries,” European Journal of egy to Performance in Small Firms,” Strate- Marketing 37(5/6), 728–752. gic Management Journal 26(13), 1249– Gibbon, P. (2003). “The African Growth and 1259. Opportunity Act and the Global Commodity Emden, Z. R., J. Calantone, and C. Droge Chain for Clothing,” World Development 31, (2006). “Collaborating for New Product 1809–1827. Development: Selecting the Partner with Gilmore, A., D. Carson, and S. Rocks (2006). Maximum Potential to Create Value,” “Networking in SME’s: Evaluating Its Contri- Journal of Product Innovation Management bution to Marketing Activity,” International 23(4), 330–341. Business Review 15(3), 278–293. Emerson, R. M. (1962). “Power-Dependence Gulati, A., N. Minot, C. Delgado, and S. Bora Relations,” in Power in Societies. Ed. M. E. (2007). “Growth in High-Value Agriculture Olsen. New York: Macmillan Publishing, in Asia and the Emergence of Vertical Links 44–53. with Farmers,” in Global Supply Chains, Fornell, C., and D. Lackner (1981). “Evaluating Standards, and Poor Farmers. Ed. J. Structural Equation Models with Unobserv- Swinnen. London: CABI Press. able Variables and Measurement Errors,” Gulati, R. (1998). “Alliances and networks,” Journal of Marketing Research 18(1), 39– Strategic Management Journal 19(1), 293– 50. 317. Freel, M. (2003). “Sectoral Patterns of Small Gulati, R., N. Nohria, and A. Zaheer (2000). Firm Innovation, Networking and Proxim- “Strategic Networks,” Strategic Management ity,” Research Policy 32(5), 751–770. Journal 21(3), 199–201. Freel, M., and R. Harisson (2006). “Innovation Hadjimanolis, A. (2000). “An Investigation of and Cooperation in the Small Firm Sector: Innovation Antecedents in Small Firms in the Evidence from “Northern Britain”,” Regional Context of a Small Developing Country,” Studies 40(4), 289–305. Research and Development Management Freel, M. S. (2000). “External Linkages and 30(3), 235–245. Product Innovation in Small Manufacturing Hair, J. F., R. E. Anderson, R. L. Tatham, and W. Firms,” Entrepreneurship and Regional C. Black (2002). Multivariate Data Analysis. Development 12(3), 245–266. Englewood Cliffs, NJ: Prentice Hall. Ganesan, S. (1994). “Determinants of Long- Hingley, M. K. (2005). “Power Imbalanced Term Orientation in Buyer–Seller Relation- Relationships: Cases from the UK Fresh ships,” Journal of Marketing 58(2), 1–19. Food Supply,” International Journal of Gatignon, H., W. Smith, M. L. Tushman, and P. Retail and Distribution Management 3(8), Anderson (2002). “A Structural Approach to 551–569. Assessing Innovation: Construct Develop- Hitt, M. A., T. Dacin, E. Levitas, J. Arregle, and ment of Innovation Locus, Type, and Char- A. Borza (2000). “Partner Selection in acteristics,” Management Science 48(9), Emerging Market Contexts: Resource Based 1103–1122. and Organizational Learning Perspectives,” Gerbing, D., and J. C. Anderson (1988). “The Academy of Management Journal 43(3), Updated Paradigm for Scale Development 449–467. Incorporating Unidimensionality and Its Hu, L. T., and P. M. Bentler (1999). “Cut-Off Assessment,” Journal of Marketing Research Criteria for Fit Indexes in Covariance Struc- 25(2), 186–192. ture Analysis: Conventional Criteria Versus

134 JOURNALMATANDA, OF SMALL NDUBISI, BUSINESS AND MANAGEMENT JIE 17 New Alternatives,” Structural Equation Mod- Lau, T., T. W. Y. Man, and I. Chow (2004). eling 6(1), 1–55. “Organizational Capabilities and Perfor- Hult, R. F., G. T. Hurley, and G. Knight (2004). mance of SMEs in Dynamic and Stable Envi- “Innovativeness: Its Antecedents and Impact ronments,” The International Journal of on Business Performance,” Industrial Mar- Entrepreneurship and Innovation 5(4), 221– keting Management 33(5), 429–438. 229. Humphrey, J. (2006). “Policy Implications of Le Heron, R., G. Penny, M. Paine, G. Sheath, J. Trends in Agribusiness Value Chains,” Euro- Pedersen, and N. Botha (2001). “Global pean Journal of Development Research Supply Chains and Networking: A Critical 18(4), 572–592. Perspective on Learning Challenges in the Hunt, S., D. Arnett, and S. Madhavaram (2006). New Zealand Dairy and Sheep Meat Com- “The Explanatory Foundations of Relation- modity Chains,” Journal of Economic Geog- ship Marketing Theory,” Journal of Business raphy 1(4), 439–445. & Industrial Marketing 21(2), 72–87. Leonard-Barton, D. (1992). “Core Capabilities Hurley, R. F., and G. T. Hult (1998). “Innova- and Core Rigidities: A Paradox in Manag- tion, Market Orientation, and Organizational ing New Product Development,” Strategic Learning: An Integration and Empirical Management Journal 13(S1), 111– Examination,” Journal of Marketing 62(3), 125. 42–54. Lindell, M. K., and D. J. Whitney (2001). Ireland, R. D., and J. W. Webb (2007). “A “Accounting for Common Method Variance Multi-Theoretic Perspective on Trust and in Cross-Sectional Research Designs,” The Power in Strategic Supply Chains,” Journal Journal of Applied Psychology 86(1), 114– of Operations Management 25(2), 482– 121. 497. Liu, Y., Y. Luo, and T. Liu (2009). “Governing Kale, P., J. H. Dyer, and H. Singh (2002). “Alli- Buyer–Supplier Relationships Through ance Capability, Stock Market Response, and Transactional and Relational Mechanisms: Long-Term Alliance Success: The Role of the Evidence from China,” Journal of Operations Alliance Function,” Strategic Management Management 27(4), 294–309. Journal 23(8), 747–767. Lusch, R., and J. Brown (1996). “Interdepen- Kaynak, E., and A. Kara (2001). “Channel of dency, Contracting, and Relational Distribution in Developing Countries,” Behaviour in Marketing Channels,” Journal Journal of International Marketing and of Marketing 60(4), 19–38. Marketing Research 26(2), 59–75. Malhotra, N., J. Shaw, M. Hall, and P. Kimberly, J. R. (1979). “Issues in the Creation of Oppenheim (2006). Marketing Research: An Organizations: Initiation, Innovation, and Applied Approach. French Forest, NSW, Aus- Institutionalization,” Academy of Manage- tralia: Prentice Hall. ment Journal 22(3), 437–457. Mariotti, J. L. (1999). “The Trust Factor in Knight, G. A., and S. T. Cavusgil (2004). “Inno- Supply Chain Management,” Supply Chain vation, Organizational Capabilities, and the Management Review 35(2), 70–77. Born-Global Firm,” Journal of International Matanda, M. J. (2002). Market Orientation, Business Studies 35(2), 124–141. Business-to-Business Relationships, Cus- Kumar, N., L. K. Scheer, and J. B. Steenkamp tomer Satisfaction, Business and Channel (1995). “The Effects of Perceived Interde- Performance in Horticultural Markets. Mel- pendence on Dealer Attitudes,” Journal of bourne: Monash University. Marketing Research 32(3), 348–356. Metcalf, L. E., C. R. Frear, and R. Krishnan Kunz, W., B. Schmitt, and A. Meyer (2011). (1992). “Buyer–Seller Relationships: An “How Does Perceived Firm Innovativeness Application of the IMP Interaction Model,” Affect the Consumer?,” Journal of Business European Journal of Marketing 26(2), Research 64(8), 816–822. 27–46. Lambe, C. J., R. Spekman, and S. D. Hunt Mirani, R., D. Moore, and J. A. Weber (2001). (2002). “Alliance Competence, Resources, “Emerging Technologies for Enhancing and Alliance Success: Conceptualization, Supplier–Reseller Partnerships,” Industrial Measurement, and Initial Test,” Journal of Marketing Management 30(2), 101–104. the Academy of Marketing Science 30(2), Mohr, J., and R. E. Spekman (1994). “Charac- 141–158. teristics of Partnership Success: Partner-

18 JOURNALMATANDA, OF SMALL NDUBISI, BUSINESS AND MANAGEMENT JIE 135 ship Attributes, Communication Behavior, Ponte, S., and P. Gibbon (2005). “Quality Stan- and Conflict Resolution Techniques,” dards, Conventions and the Governance of Strategic Management Journal 15(2), 135– Global Value Chains,” Economy and Society 152. 34, 1–31. Monczka, R., M. Petersen, R. B. Handfield, and Powell, T. (1994). “The Role of Contingency G. L. Ragatz (1998). “Success Factors in Stra- Factors: Untangling the Relationship tegic Supplier Alliances: The Buying between Strategic Planning and Perfor- Company Perspective,” Decision Science mance,” Revue Canadienne des Sciences de 29(3), 5553–5577. l’Administration 11(2), 124–128. Morgan, R. M., and S. D. Hunt (1994). “The Powell, W., K. W. Koput, and L. Smith-Doerr Commitment—Trust Theory of Relationship (1996). “Interorganizational Collaboration Marketing,” Journal of Marketing 58(3), and the Locus of Innovation: Networks of 20–38. Learning in Biotechnology,” Administrative Nunnally, J. C. (1978). Psychometric Theory. Science Quarterly 41(1), 116–145. New York: McGraw Hill Book Company. Reardon, T., C. B. Barrett, J. A. Berdegué, and O’Donnell, A., A. Gilmore, D. Cummins, and D. J. F. Swinnen (2009). “Agrifood Industry Carson (2001). “The Network Construct in Transformation and Small Farmers in Devel- Entrepreneurship Research: A Review and oping Countries,” World Development Critique,” Management Decision 39(9), 749– 37(11), 1717–1727. 756. Regmi, A., and M. Gehlhar (2005). New Direc- OECD (2008). Enhancing the Role of SMEs tions in Global Food Markets. Washington, in Global Value Chains. Paris: OECD DC: U.S. Department of Agriculture. Agricul- Publishing. ture Information Bulletin No. 794. Ouma, S. (2010). “Global Standards, Local Rindfleisch, A., and J. B. Heide (1997). “Trans- Realities: Private Agrifood Governance and action Cost Analysis: Past, Present, and the Restructuring of the Kenyan Horticulture Future Applications,” Journal of Marketing Industry,” Economic Geography 86(2), 197– 61(4), 30–54. 222. Rokkan, A. L., and S. A. Haughland (2002). Palmer, A. (2000). “Co-operation and Competi- “Developing Relational Exchange: Effective- tion: A Darwinian Synthesis of Relationship ness and Power,” European Journal of Mar- Marketing,” European Journal of Marketing keting 36(1), 211–220. 34(5), 687–704. Romero, I., and J. A. Martínez-Román (2012). Petkova, I. (2006). “Shifting Regimes of Gover- “Self-Employment and Innovation. Exploring nance in the Coffee Market: From Secular the Determinants of Innovative Behavior in Crisis to a New Equilibrium?,” Review of Small Businesses,” Research Policy 41(1), International Political Economy 13, 313– 178–189. 339. Rotter, J. B. (1967). “A New Scale for the Mea- Pfeffer, J., and G. R. Salancik (1978). The Exter- surement of Interpersonal Trust,” Journal of nal Control of Organizations: A Resource Personality 35(4), 651–665. Dependence Perspective. New York: Harper Rowe, W. G., and P. M. Wright (1997). “Related & Row. and Unrelated Diversification and Their Pietrobelli, C., and R. Rabellotti, eds. (2007). Effect on Human-Resource Management Upgrading to Compete: SMEs, Clusters and Controls,” Strategic Management Journal Value Chains in Latin America. Cambridge, 18(4), 329–338. MA: Harvard University Press. Sadler-Smith, E., Y. Hampson, I. Chaston, and Pietrobelli, C., and R. Rabellotti (2010). “Global B. Badger (2003). “Managerial Behavior, Value Chains Meet Innovation Systems: Are Entrepreneurial Style, and Small Firm Perfor- There Learning Opportunities for Develop- mance,” Journal of Small Business Manage- ing Countries?,” World Development 39(7), ment 41(1), 47–67. 1261–1269. Salomo, S., F. Steinhoff, and V. Trommsdorff Podsakoff, P. M., S. B. MacKenzie, and N. P. (2003). “Customer Orientation in Innovation Podsakoff (2012). “Sources of Method Bias Projects and New Product Development in Social Science Research and Recommen- Success—The Moderating Effect of Product dations on How to Control It,” Annual Innovativeness,” International Journal of Review of Psychology 63, 539–569. Technology Management 26(5/6), 442–463.

136 JOURNALMATANDA, OF SMALL NDUBISI, BUSINESS AND MANAGEMENT JIE 19 Sanchez, R. (1995). “Technological Transforma- Industrial Maintenance Market,” Industrial tion and the New Competitive Landscape,” Marketing Management 30(1), 1–12. Strategic Management Journal 16(Special Subroto, R., K. Sivakumar, and I. F. Wilkinson Issue), 135–159. (2004). “Innovation Generation in Supply Sapienza, H. J., K. G. Smith, and M. J. Gannon Chain Relationships: A Conceptual Model (1988). “Using Subjective Evaluations of and Research Propositions,” Journal of the Organizational Performance in Small Busi- Academy of Marketing Science 32(1), 61– ness Research,” American Journal of Small 79. Business 12(3), 45–53. Swierczek, F. W., and T. T. Ha (2003). “Entre- Sarkar, M. B., R. Echambadi, T. S. Cavusgil, and preneurial Orientation, Uncertainty Avoid- P. S. Aulakh (2001). “The Inﬂuence of ance and Firm Performance,” International Complementarity, Compatibility, and Rela- Journal of Entrepreneurship and Innovation tionship Capital on Alliance Performance,” 4(1), 46–58. Journal of the Academy of Marketing Science Tagliavini, M., A. Ravarini, and A. Antonelli 29(4), 358–373. (2001). “An Evaluation for Electronic Com- Schmitz, H., and P. Knorringa (2000). “Learning merce Activities within SMEs,” Information from Global Buyers,” Journal of Develop- Technology and Management 2(2), 211– ment Studies 37(2), 177–205. 230. Sezern, B., and C. Yilmaz (2007). “Relative Takane, T. (2004). “Smallholder and Non- Effects of Dependence and Trust on Flexibil- Traditional Exports Under Economic Liberal- ity, Information Exchange, and Solidarity In ization: The Case of Pineapples in Ghana,” Marketing Channels,” Journal of Business & African Study Monographs 25(1), 29– Industrial Marketing 22(1), 41–51. 43. Shimizu, K., and M. A. Hitt (2004). “Strategic Tallontire, A. (2007). “CSR and Regulation: Flexibility: Organizational Preparedness to Towards a Framework for Understanding Reverse Ineffective Strategic Decisions,” The Private Standards Initiatives in the Agri-Food Academy of Management Executive 18(4), Chain,” Third Year Quarterly 28, 775–791. 44–59. Teece, D. J., G. Pisano, and A. Shuen (1997). Sivadas, E., and R. F. Dwyer (2000). “An Exami- “Dynamic Capabilities and Strategic Manage- nation of Organizational Factors Inﬂuencing ment,” Strategic Management Journal 18(7), New Product Success in Internal and 509–533. Alliance-Based Processes,” Journal of Mar- Tripsas, M., and G. Gavetti (2000). “Capabili- keting 64(1), 31–43. ties, Cognition and Inertia: Evidence from Spekman, R. E., J. W. Kamauff Jr., and N. Myhr Digital Imaging,” Strategic Management (1998). “An Empirical Investigation into Journal 21(10/11), 1147–1161. Supply Chain Management: A Perspective on Tyler, B. B. (2001). “The Complementarity of Partnerships,” Supply Chain Management Cooperative and Technological Competen- 3(2), 53–67. cies. A Resource-Based Perspective,” Stevens, C., and J. Kennan (2000). “Will Africa’s Journal of English Technology Management Participation in Horticulture Chains Survive 18(1), 1–27. Liberalization?,” Working Paper 106. Insti- Vermeulen, P. A., J. P. de Jong, and K. C. tute of Development Studies, Brighton, O’Shaughnessy (2005). “Identifying Key UK. Determinants for New Product Introductions Street, C., and A. Cameron (2007). “External and Firm Performance in Service SMEs,” The Relationships and the Small Business: A Service Industries Journal 25(5), 625–640. Review of Small Business Alliances and Voss, K. E., J. L. Johnson, J. B. Cullen, T. Network Research,” Journal of Small Busi- Sakano, and H. Takenouchi (2006). “Rela- ness Management 45(2), 239–266. tional Exchange in US–Japanese Marketing Street, C. T., and D. B. Meister (2004). “Small Strategic Alliances,” International Marketing Business Growth and Internal Transparency: Review 23(6), 610–635. The Role of Information Systems,” MIS Wetzels, M., K. de Ruyer, and M. van Birgelen Quarterly 28(3), 473–506. (1998). “Marketing Service Relationships: Stremersch, S., S. Wuyts, and R. T. Frambach The Role of Commitment,” Journal of Busi- (2001). “The Purchasing of Full-Service Con- ness & Industrial Marketing 13(4/5), 406– tracts: An Exploratory Study within the 423.

20 JOURNALMATANDA, OF SMALL NDUBISI, BUSINESS AND MANAGEMENT JIE 137 White, H. M. (2000). “Buyer–Supplier Relation- Flexibility in Strategic Human Resource ships in the UK Fresh Produce Industry,” Management,” The Academy of Management British Food Journal 102(1), 6–17. Review 23(4), 756–772. White, S., and S. S. Lui (2005). “Distinguishing Wu, F., and S. T. Cavusgil (2006). “Organiza- Cost of Cooperation and Control in Alli- tional Learning and Joint Value Creation in ances,” Strategic Management Journal Interﬁrm Relationships,” Journal of Business 26(10), 913–932. Research 59, 81–89. Wright, P. M., and S. A. Snell (1998). “Toward a Unifying Framework for Exploring Fit and

138 JOURNALMATANDA, OF SMALL NDUBISI, BUSINESS AND MANAGEMENT JIE 21 Journal of Small Business Business Management Management 2016 2014 54(1), ••(••), pp. pp. 139–161 ••–•• doi: 10.1111/jsbm.12135 Linking Processes and Dynamic Capabilities of International SMEs: The Mediating Effect of International Entrepreneurial Orientation by Bernhard Swoboda and Edith Olejnik

Previous research on the internationalization of small and medium-sized enterprises (SMEs) has highlighted the role of knowledge and learning about foreign markets. However, empirical results on the performance implications of foreign market scanning and planning have been mixed. Following the dynamic capability perspective, we argue that SMEs can capitalize on scanning and planning processes because of their international entrepreneurial orientation. We test our hypotheses with a sample of 604 established SMEs and ﬁnd that entrepreneurial orientation completely mediates the relationship between scanning and planning and international performance. Moreover, the results implicate a bidirectional relationship between processes and international entrepreneurial orientation.

(2005) found that internationalization strate- Introduction gies may be unplanned. Armario, Ruiz, and Internationalization research has high- Armario (2008) demonstrated that SMEs must lighted the role of knowledge and learning in spend more time seeking sources of export the internationalization process of both multi- information to stay competitive. In another national enterprises and small and medium- recent study, however, it was shown that sized enterprises (SMEs) (Johanson and Vahlne market orientation (which includes informa- 1977). Exploring, analyzing, and planning tion acquisition) has no significant influence international activities can create knowledge of the international performance of SMEs and therefore serve as a critical factor in suc- (Frishammar and Andersson 2009). As results cessfully internationalizing firms (Knight and are inconclusive, it seems that little is known Liesch 2002; McGee and Sawyerr 2003). about how SMEs capitalize on foreign market Although this argumentation is theoretically scanning and planning processes to achieve reasonable, empirical studies have shown superior performance. We extend the knowl- inconsistent results. Seringhaus (1993) stressed edge on this relationship by including interna- the importance of information for SMEs, tional entrepreneurial orientation as a dynamic whereas Li, Li, and Dalgic (2004) showed that capability into our reasoning. Thereby, we aim SMEs tend to follow an unsystematic decision- to examine the mediating influence of a capa- making process. Similarly, Crick and Spence bility on the relationship between scanning

Bernhard Swoboda is a professor at the Department of Business Administration at Trier University. Edith Olejnik is a PhD student at the Department of Business Administration at Trier University. Address correspondence to: Bernhard Swoboda, Business Administration, Trier University, Universitätsring 15, Trier 54286, Germany. Tel.: +00496512013050; E-mail: [email protected].

SWOBODA AND OLEJNIK 139 1 and planning processes and the international (p. 101). This definition partly resulted in a performance of SMEs. synonymous usage of the terms. We follow the Consequently, we intend to contribute to perspective that resources, processes, and this field by analyzing the following research capabilities clearly differ from one another. questions. First, why and how do scanning and Processes are actions in which firms engage planning processes contribute to the interna- to accomplish a business purpose or objective tional performance of established international (Ray, Barney, and Muhanna 2004). Scanning SMEs? Second, what role does international processes, on the one hand, are implemented entrepreneurial orientation play in mature to create knowledge about foreign markets and international SMEs? customers (Knight and Liesch 2002). Planning It is theoretically and practically relevant to processes, on the other hand, intend to guide study whether and why scanning and planning the activities of the firm and create adaptive exerts an influence on the international per- thinking (Martinez and Jarillo 1991; Miller and formance of SMEs. As these processes may Cardinal 1994). The implementation of pro- have direct or indirect performance effects, cesses to systematically explore, analyze, and managers should know whether investing plan international activities is not only an scarce resources into these processes makes important management task but also a critical sense. As Cadogan et al. (2006) noted, SMEs factor for success (McGee and Sawyerr 2003). tend to reduce information collection in foreign Hence, processes are viewed as a way to markets if they lack the necessary resources, exploit a firm’s resources to create competitive which is often the case. Responding to the calls advantage as resources cannot be a source of of Jones and Coviello (2005) and Jantunen et al. competitive advantage by themselves (Ray, (2005) to analyze entrepreneurial orientation in Barney, and Muhanna 2004). However, not all studies on international SMEs, we include inter- processes will be a source of competitive national entrepreneurial orientation as an inter- advantage for a firm. The resource-based view vening factor and dynamic capability in our suggests that processes that exploit intangible reasoning. Thereby, we enhance the knowl- firm resources are more likely to be a source of edge on how SMEs make use of processes competitive advantage than processes that leading to competitive advantage. In doing so, exploit tangible firm resources (Barney 1991). we discuss the central role of international Of course, intangible and tangible resources entrepreneurial orientation as a dynamic firm must often be combined to enable the imple- capability in established international SMEs. mentation of a particular process. The remainder of this article is organized as Capabilities refer to a firm’s competence to follows. We use resource-based logic and the make use of resources and processes and dynamic capability perspective to highlight the combine them with other processes and intan- significance of processes and capabilities. gible resources such as knowledge (Lu et al. Then, we discuss the literature on entrepre- 2010). Amit and Schoemaker (1993, p. 35) neurial orientation, scanning, and planning, as defined capabilities as “a firm’s capacity to well as the respective implications concerning deploy resources, usually in combination, using performance. The hypotheses are empirically organizational processes, to effect a desired tested using several structural equation models end.” From the dynamic capability perspective, (SEMs). Finally, the results are discussed, fol- capabilities can be understood as a firm’s lowed by limitations and directions for further orientation to integrate and reconfigure its research. resources and processes and, even more importantly, transform its processes in response to Theoretical Background foreign environments to achieve sustainable The resource-based view suggests that a firm competitive advantage (Wang and Ahmed can attain and sustain competitive advantage 2007). Thereby, the term dynamic refers to the only if the firm creates an idiosyncratic pool of capacity of adapting to changing environments resources (Dhanaraj and Beamish 2003; Lu and finding innovative solutions to new prob- et al. 2010). However, the literature presents lems through the adaptation, integration, and different definitions of resources. Barney reconfiguration of resources and processes (1991) classified resources broadly as “all (Teece, Pisano, and Shuen 1997). Wang and assets, capabilities, organizational processes, Ahmed (2007) described the nature of dynamic firm attributes, information, knowledge, etc.” capabilities as behavioral orientation, whereas

2140 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Teece, Pisano, and Shuen (1997) considered advantage involves not only the resources dynamic capabilities to be an ability or a capac- owned by a firm but also how the firm inte- ity (Teece 2007). As Barreto (2010) pointed out grates, combines, and transforms these more recently, there are many different concep- resources through dynamic capabilities. Hence, tualizations and definitions of what dynamic dynamic capabilities are directly linked to firm capabilities are. Hence, he suggested a defini- performance (Teece, Pisano, and Shuen 1997). tion of dynamic capabilities that accommodates Moreover, scanning and planning processes are important ideas within the field: “A dynamic closely connected to entrepreneurship and the capability is the firm’s potential to systemati- discovery of opportunities (Teece 2007). The cally solve problems, formed by its propensity constructs are strongly related to SME learning to sense opportunities and threats, to make (Baker and Sinkula 2009). Zahra, Sapienza, timely and market-oriented decisions, and to and Davidsson (2006) noted that the creation change its resource base” (p. 271). The author and use of dynamic capabilities correspond to viewed dynamic capabilities as a multidimen- the perception of opportunities by the entre- sional composite construct that highlights the preneur or the principal decision-maker(s). following main thoughts on dynamic capabili- Consequently, we propose that scanning and ties: (1) the propensity to sense opportunity planning are processes that combine tangible and threats (Gilbert 2006); (2) the propensity to and intangible resources within a firm to make timely decisions in order to be ahead of achieve a certain purpose that often contributes competitors (Teece, Pisano, and Shuen 1997); to a small firm’s knowledge development and (3) the potential to systematically solve prob- learning. As these processes are potentially lems as compared with trial-and-error learning; transferrable, we argue that the processes must and (4) the propensity to reconfigure the be transformed through the entrepreneurial ori- resource base of the firm (Eisenhardt and entation of small firms to create and sustain a Martin 2000). Although Eisenhardt and Martin competitive advantage. The entrepreneurial ori- (2000) define dynamic capabilities as pro- entation is viewed as a dynamic capability that cesses, there is a general difference between has the propensity to sense and seize interna- dynamic capabilities and processes in that tional opportunities in an innovative, market- dynamic capabilities cannot be bought or trans- oriented, and timely manner. By applying the ferred (Makadok 2001). Processes, however, capability-building perspective, we enrich the are often explicit combinations of tangible and understanding of the role of entrepreneurial intangible resources and can thus be trans- orientation as a key dynamic capability of inter- ferred. Following a hierarchical perspective national SMEs. (Wang and Ahmed 2007; Winter 2003), we However, we must acknowledge an addi- would argue that resources build the zero order tional but different role of international entre- or base element. Processes are first order preneurial orientation. It has been argued that because they are based on tangible and intan- entrepreneurs have unique insights into the gible resources. Capabilities are third-order ele- value of resources and processes and hence ments because they combine resources and influence the extent to which scanning and processes to achieve a desired end. Dynamic planning processes are carried out. Therefore, capabilities build on (mere) capabilities entrepreneurial orientation as dynamic capabil- because they combine resources and processes ity could be viewed as antecedent factor. This in response to changing environments. reversed causality fits well with the discussion We interpret entrepreneurial orientation as on dynamic capabilities led by Eisenhardt and an essential dynamic capability of SMEs to Martin (2000). Eisenhardt and Martin (2000) reconfigure scanning and planning processes defined dynamic capabilities as specific firm to match the requirements of changing environ- processes and strategic routines by with firms ments by identifying and exploiting interna- achieve new resource combinations. Hence, the tional opportunities. Thereby, we refer to the authors stated: “Dynamic capabilities are the previously stated definition of dynamic capa- antecedent organizational and strategic routines bilities and the capability-building mechanism by with managers alter their resource base” (p. that affects economic profit after the acquisition 1107). Also Barreto (2010) argued that this view of a resource (Makadok 2001). Teece, Pisano, is consistent with previous research. Although and Shuen (1997) suggested a similar distinc- many studies have assumed a direct link tion, arguing that sustainable competitive between dynamic capabilities and performance,

SWOBODA AND OLEJNIK OLEJNIK 141 3 the relationship may as well be of indirect nature. Following this theoretical argument, Hypothesis Development entrepreneurial orientation could be understood International Entrepreneurial Orientation as an antecedent factor that influences resource There are several termini to describe entrepre- acquisition through scanning and planning. neurial firms (e.g., entrepreneurship, entrepre- Given that former research has found entrepreneurial orientation, entrepreneurial posture, neurial orientation to be an antecedent factor to and entrepreneurial management), which are market orientation (Matsuno, Mentzer, and often used interchangeably. However, some Özsomer 2002) and different business processes general definitions have found common (Knight and Cavusgil 2004), we must consider acceptance. Stevenson and Jarillo (1990), for that entrepreneurial orientation also influences example, describe entrepreneurship as pursuing business processes. Covin and Slevin (1991) opportunities. These opportunities exist in argued in their seminal work that the entrepre- domestic and international markets (Kontinen neurial posture (i.e., the propensity to innovate, and Ojala 2011; Zahra and Garvis 2000). Hence, be proactive, and take risks) both influences and entrepreneurship can be found not only in new is influenced by organizational resources and venture creation but also in entering and competences. The authors elaborated not only penetrating foreign markets (Lumpkin and that resources and processes build the basis for Dess 1996). In this study, we focus on interna- entrepreneurship but also that entrepreneurship tional entrepreneurial orientation, which can be influences how resources are used. In their viewed as a precursor to entrepreneurship and conceptual model, Covin and Slevin (1991) entrepreneurial behavior (Jantunen et al. 2005). already introduced the idea of a possibly recip- The construct is fundamentally based in entrepre- rocal relationship assuming that capabilities and neurship research and applied to international resources have a stronger effect on entrepre- firms. According to Knight and Cavusgil (2004), neurial orientation than the other way around. international entrepreneurial orientation is The authors also stressed performance as a characterized by the firm’s overall innov- dependent variable. ativeness and proactiveness in the pursuit of Consequently, we aim to research the international markets. Hence, international entre- mediating effect of international entrepreneur- preneurial orientation is associated with risk ial orientation on the process–performance taking, innovativeness, and proactiveness (Covin relationship. Extending this line of thought, we and Slevin 1989; Miller 1983) and reflects a firm’s test the reciprocal relationship between pro- propensity to engage in innovative and risk- cesses and international entrepreneurial orien- taking behavior to achieve its strategic objectives tation as a dynamic capability (see Figure 1). in international markets (Knight 2001).

Figure 1 Conceptual Framework

Processes

Scanning

International Entrepreneurial Orientation Performance Innovativeness Risk Taking Proactiveness

Planning

4142 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Entrepreneurial orientation has been mation and hence contribute to a firm’s learn- acknowledged to promote corporate competi- ing about foreign markets (Johanson and tiveness. Lumpkin and Dess (1996) and Naman Vahlne 1977). Information-gathering processes and Slevin (1993), for example, noted that are based on intangible knowledge resources entrepreneurship can be used to improve com- within the firm. Using those resources to scan petitive positioning and create value through the environment and generate information on the recognition and pursuit of opportunities. foreign markets can be viewed as rare and Although several studies have suggested that valuable intangible process. In general, scan- the association between entrepreneurship and ning processes have been analyzed under a performance might not be positive (Zahra variety of names, from different perspectives 1993), most researchers argue that entrepre- and in different contexts. Empirical research, neurship leads to the increased performance of however, provides differing results. Although SMEs in both domestic (Zahra 1991) and inter- some authors argue that scanning processes are national contexts (Balabanis and Katsikea 2003; irrelevant for small firms, other studies claim a Covin and Slevin 1991; Dimitratos, Lioukas, positive relationship. According to the litera- and Carter 2004; Robertson and Chetty 2000). ture review of Julien and Ramangalahy (2003), Recent empirical evidence shows that the rela- scanning processes can generate competitive tionship between entrepreneurial orientation advantage. Seringhaus (1993) found that export and performance is inversely U-shaped in new marketing expertise (i.e., information-related ventures but linear positive in established ven- skills concerned with the segmentation of tures (Su, Xie, and Li 2011), which explains the foreign markets) has a significant impact on variance in former studies. As we aim at ana- performance, whereas results on market lyzing established SMEs, we propose: research behavior have been inconclusive. Information acquisition and planning activities H1: International entrepreneurial orientation are significant export success factors, claimed has a direct positive effect on international Walters and Samiee (1990). Similarly, Yeoh performance. (2000) found that global start-ups that actively search for information are also more likely to Scanning have higher export performance. McGee and We define scanning as processes that intend Sawyerr (2003) found that the extent of envi- to generate information relevant for decision- ronmental scanning depends on the age of the making in international markets (Zahra 1991), small firm and the perceived strategic uncer- which may also be labeled as information tainty surrounding the firm. acquisition (Williams 2003), intelligence gen- Scanning helps the international SMEs to eration as part of the market orientation con- acquire relevant information on markets, cus- struct (Jaworski and Kohli 1993), and tomers, competitors, and the industry on acquisition of international market information time. Thereby, scanning processes facilitate the as part of the export market orientation con- accumulation of information that may be of struct (Cadogan and Diamantopoulos 1995). interest to the firm. Scanning also alerts senior Information acquisition relates to the process of executives to threats and opportunities in their searching and obtaining relevant information firms’ environment (Zahra 1991). Through sys- on customers, competitors, and environmental tematic environmental scanning processes, trends. Although Williams (2003) considers opportunities can be identified, which impacts information sources as a part of information the international entrepreneurial orientation. acquisition, we focus on the active part of the Knowledge created through scanning can information-gathering process. Therefore, scan- reveal innovative ideas and lead to the proac- ning refers to efforts to collect, analyze, and tive pursuit of international expansion oppor- interpret data about a firm’s external environ- tunities. In a similar vein, scanning can be ment and competition and encompasses ele- viewed as a process of uncertainty absorption ments such as market research and analysis (Barringer and Bluedorn 1999). Information (Zahra 1991). Particularly for the small interna- generated through scanning can help small tional firm, it is essential to gain appropriate firm managers to reduce the uncertainty con- information by market research and objective nected to foreign market operations. Yeoh data (Brouthers and Nakos 2005). Scanning (2000) claims that “even modest additions to processes provide a firm with objective infor- their databases can lead to dramatic shifts in

SWOBODA AND OLEJNIK OLEJNIK 143 5 their risk perceptions” (p. 54). Less uncertain essentially the capability-building perspective. actions can then be proactively pursued as the Hence, we suggest that international entrepre- risk may be perceived assessable. Hence, neurial orientation provides a filter through we claim that scanning processes have a posi- which firms can direct processes to lead to supe- tive influence on entrepreneurial orientation. rior performance and conclude the following: The empirical literature has provided some support for this relationship. In a national H2: Scanning has a direct positive effect on study of large established firms, Zahra (1991) international entrepreneurial orientation showed that environmental scanning is posi- such that international entrepreneurial ori- tively associated with corporate entrepreneur- entation mediates the relationship between ship. Similarly, Barringer and Bluedorn (1999) scanning and international performance. supported the effect of environmental scanning on corporate entrepreneurship. Miocevic Planning and Crnjak-Karanovic (2011) found that the We define planning as processes that intend effect of export market orientation on export to guide the activities of the firm (Martinez and performance is partly mediated through the Jarillo 1991). This definition includes processes global vision of the small firm manager. More- such as strategic planning, budgeting, and over, we argue that resources and processes formal planning. Hence, planning is aimed at are a basic condition of entrepreneurship creating adaptive thinking and integrating and (Covin and Slevin 1991). Scanning is a process controlling various parts of the firm (Miller and aimed at creating knowledge of foreign Cardinal 1994). In the strategic management markets. Without scanning, entrepreneurs literature especially, the formality of planning would be faced with tradeoffs in undermining has been focused. Although Crick and Spence existing activities to pursue uncertain interna- (2005) highlighted the unplanned international- tional opportunities (Perks and Hughes ization behavior of SMEs, other studies demon- 2008). Firms possess and acquire different strated a positive association between planning stocks of information, and these stocks of and performance. Both Schwenk and Shrader information influence their ability to recognize (1993) and Miller and Cardinal (1994) found a particular opportunities (Kontinen and Ojala positive and significant relationship between 2011; Shane and Venkataraman 2000), which is planning and performance in their meta study. an essential entrepreneurial act. Cooper, Folta, However, neither study focused on international and Woo (1995) claim that gathering informa- SMEs. Walters (1993) showed, however, that tion for decision-making is a critical activity for both size and environmental complexity influ- entrepreneurs as it helps them to identify ence planning activity. Baird, Lyles, and Orris opportunities. As entrepreneurial orientation (1994) found that internationally oriented small is a precursor to entrepreneurship, support- firms have more formal planning systems than ing opportunity recognition in international those who are not. Yip, Biscarri, and Monti markets (Jantunen et al. 2005), we have a (2000) found that firms that systematically plan reason to argue that scanning processes are internationalization steps achieve better perfor- associated with entrepreneurial orientation. mance. Although SMEs might follow fewer Consequently, we propose that scanning is formal planning processes compared with large positively associated with entrepreneurial orien- corporations, planning might separate success- tation. To be precise, we believe that the effect of ful from nonsuccessful international SMEs. scanning on performance is mediated by entre- According to resource-based logic, firms use preneurial orientation. The acquisition of infor- information and knowledge resources to plan mation itself does not necessarily lead to effective future steps, which implies that planning is use of this information (Teece 2007). As the based on intangible resources and carried out to outcome of scanning processes depends on the guide and control the development of the firm. firm’s capability to coordinate, configure, and Researchers have argued either for or against transform the knowledge generated through the usefulness of planning for entrepreneurial scanning, we argue that the nature of the rela- activity. Covin, Green, and Slevin (2006), for tionship is indirect. Entrepreneurial orientation example, proposed that entrepreneurial orien- does not simply magnify the relation; it creates it. tation has more of a positive effect on the sales Therefore, we argue that economic rent is created growth rate when strategies emerge than when after the acquisition of the resource, which is they are planned. However, earlier research

6144 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT argued that planning helps people to overcome entation mediates the relationship between the limits of their cognitive capacities (Simon planning and international performance. 1976). Human beings have limited information- processing capacities that can be exceeded by Reciprocity of Processes and the complexity of foreign activities in uncertain International Entrepreneurial Orientation environments (Gruber 2007). Planning pro- So far, we have deduced that scanning and cesses can simplify task complexity and there- planning processes influence international entre- fore serve as facilitator for entrepreneurial preneurial orientation. However, previous orientation (Covin and Slevin 1991). Barringer research has shown that entrepreneurial orienta- and Bluedorn (1999) showed that planning flex- tion may as well impact scanning and planning ibility, the locus of planning, and strategic processes. Empirical studies showed entrepre- control have a significant positive influence on neurial orientation to be an antecedent factor to corporate entrepreneurship. Planning provides market orientation (Matsuno, Mentzer, and a framework that guides individuals in their Özsomer 2002) and different business processes understanding of their environment and rel- (Knight and Cavusgil 2004). Baker and Sinkula evant strategic issues and helps them to seize (2009) demonstrated that entrepreneurial orien- opportunities (Teece 2007). In other words, tation has a significantly positive influence on planning processes provide SME managers with market orientation, which is understood as complexity reduction and the ability to pursue customer-oriented market intelligence. Knight entrepreneurial activities. Similarly, Delmar and (2001) found that international entrepreneurial Shane (2003) argued that planning helps entre- orientation promotes the development of strate- preneurs to make decisions more quickly than gic competence, internationalization prepara- trial-and-error learning. Planning is also sup- tion, and the acquisition of technology. In posed to be useful in managing resource supply general, international entrepreneurial orienta- and demand in ways that minimize time- tion, understood as the propensity to engage in consuming bottlenecks and more efficiently innovative, proactive, and risk-taking behaviors, turning abstract goals into concrete operational seems to function as a dynamic capability that not activities (Delmar and Shane 2003). Conse- only transforms business processes, leading to quently, we propose that planning processes are higher returns, but also influences and designs positively associated with international entre- processes. Following this perspective, interna- preneurial orientation. We further believe that tional entrepreneurial orientation helps firms to the effect of planning on performance is medi- make decisions on the extent of resources and ated by entrepreneurial orientation. SMEs capi- processes that must be invested to achieve a talize on planning processes through their desired end. Whereas managerial decisions and entrepreneurial capability. If we image two international behavior can be based on pro- firms with similar planning processes, which cesses, processes may in turn be based on entre- firm has the competitive advantage? It will be the preneurial behaviors. Jones and Coviello (2005) firm that has the capability to deploy and recon- argued that innovativeness, proactiveness, and figure scanning and planning processes such risk taking will influence organizational level that international opportunities can be recog- factors. This argument is supported by the nized and exploited. This dynamic capability is general conceptual model of entrepreneurship as given by the international entrepreneurial orien- introduced by Covin and Slevin (1991). Hence, tation. Through entrepreneurial orientation, we conclude the following: firms can make use of knowledge-generating processes. Moreover, Schwenk and Shrader H4: International entrepreneurial orientation (1993) assume an intervening variable to explain has a simultaneous direct positive effect on the variance of effect sizes reported in the (a) scanning processes and (b) planning literature on the performance effects of plan- processes. ning. We propose the intervening variable to be international entrepreneurial orientation and Empirical Study conclude the following: Sample The cross-sectional empirical study is H3: Planning has a direct positive effect on based on managerial perceptions. This international entrepreneurial orientation approach is—in spite of its methodological such that international entrepreneurial ori- shortcomings—widespread in internationaliza-

SWOBODA AND OLEJNIK OLEJNIK 145 7 tion research (Acedo and Jones 2007; Jones and industries with average sales of €20.419 million Coviello 2005). Furthermore, primary data using and 84 employees on average (see Table 1). a key informant survey design have been con- To account for nonresponse bias, we com- sidered due to a lack of secondary data on pared early and late respondents and gathered German SMEs, which are usually not subject to secondary data on the demographics of ran- external publication requirements. domly selected nonrespondents to compare Aiming to research the role of foreign market responding and nonresponding firms (in terms scanning and planning in conjunction with the of size, age, and sales) (Armstrong and Overton international entrepreneurial orientation for the 1977). In both cases, insignificant differences international performance of SMEs, we define were found. As we must rely on the perceptions our population as SMEs based in Germany that of SME owner–managers, we must consider exhibited some form of international selling common method variance. We tried to appre- activity. To develop the sample, we used the hend the possible bias a priori through the Hoppenstedt database. The Hoppenstedt data- design of the questionnaire and ex post through base is a commercial online database containing Harman’s single factor test (Podsakoff et al. several thousand addresses of firms in 2003). As the first factor accounts for only 26.17 Germany. This database has been used by percent of the total variance explained in explor- several authors researching German firms atory factor analysis, common method bias can (Maekelburger, Schwens, and Kabst 2012; be assumed to be reduced within our possibili- Schilke and Goerzen 2010). We focused on ties as we must rely on self-reports of executives highly internationalized manufacturing indus- (Spector 2006). With the focus on the percep- tries such as mechanical engineering, textile, tions of the actual decision-maker being the chemicals, and polymer processing to reach competent person to answer the questionnaire, internationally active SMEs. We extracted the single response bias is assumed to be limited addresses of every fifth firm from industry- (Hughes and Garrett 1990). Following the sug- based lists provided by the Hoppenstedt data- gestion of Kumar, Stern, and Anderson (1993), base to randomize the selection procedure. we attempted to obtain a second respondent to Through this procedure, we identified 3500 validate the data and asked the senior executives SMEs. We disseminated questionnaires by ordi- by phone for a second person with appropriate nary mail and email to the senior managers of knowledge. This resulted in 41 further contact these firms. We received notice that 123 partners, mostly export/sales managers, out of questionnaires/emails were not deliverable. whom we could reach 29 by phone. Testing for The questionnaire could be completed in inter-rater congruence, we received significant writing or electronically, depending on the correlations; thus, the chance for bias is limited. executive’s preference. We sent two reminder Another threat to the quality of survey data emails and made telephone calls two weeks is sample selection bias. As we sampled interna- later. We received 855 responses in total, which tionalized firms only, we omitted SMEs that are equals a response rate of 25.3 percent. not yet international but employ some scanning However, we had to exclude 251 questionnaires and planning processes regarding international from further analysis due to at least one of the markets potentially affecting performance following reasons: frequent missing values, no through entrepreneurial orientation. As we regular international activities, the firm reported defined our population to be internationally international production only, the firm was no active firms only, we cannot claim SME based on the number of employees (250 is anything about the relationship between recommended by the European Commission), the scanning, planning, and performance of and the questionnaire was not filled out by an noninternationalized firms. Nevertheless, we SME owner or managing director. All in all, we tested whether there are significant differences recovered 604 questionnaires suitable for the in the performance of recently internationalized final analysis, which equals an effective firms and the rest. We did not find any significant response rate of 17.7 percent. This figure is in differences. Because our study focuses on inter- line with former research on international SMEs nationalized SMEs from Germany and there is (Baum, Schwens, and Kabst 2011; Dhanaraj and only information available regarding the struc- Beamish 2003). Finally, the questioned tural characteristics of all SMEs in Germany and decision-makers represent internationally active structural data on internationally active SMEs are SMEs from four major German manufacturing lacking, it is difficult to statistically evaluate the

8146 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT representativeness of our sample. However, as we randomized the sample selection procedure and could show that nonresponse bias is not a Percent serious threat, we can assume that we have a good approximation of the population (Yang, n Wang, and Su 2006). Moreover, the sample design corresponds with our research aim and with the procedure applied by other studies in a similar context (Maekelburger, Schwens, and Kabst 2012).

Measurements 9.491 in Million EUR Average: Considering the literature on scale development and modiﬁcation, we derived the measures for our survey from entrepreneurship and SME literature. To assess the face validity of the scaled items, we conducted pretests with 10 Percent Foreign Sales ﬁrm owners. Our pretest resulted in adjustments of the used scales, which we describe n next. The items used in the questionnaire and their respective indicators of reliability and validity are displayed in Table 2. For the measurement of scanning, we referred to Seringhaus (1993) and focused on market research and different forms of foreign 20.419

Million EUR market analyses. Thereby, we adapted several Average: items of Miller and Friesen’s (1982) scale on environmental scanning and Beal’s (2000) extended scale on the scope of scanning in terms Table 1 of customers and competitors in the task envi- Percent Total Sales in ronment. For the measurement of planning,we created a new scale based on the research of Sample Statistics n Robinson and Pearce (1983), Bracker, Keats, and Pearson (1988), and Lyles et al. (1993). Previous research has often classified firms based on different dimensions of planning such as use of different planning instruments (e.g., 83.6 financial and personnel planning), the planning Average: horizon, and the extent of planning (Bracker, Keats, and Pearson 1988; Walters 1993). Follow- ing the research on strategic planning, we measured planning in terms of planning formality Percent Employees (Lyles et al. 1993; Robinson and Pearce 1983). Both planning and scanning exhibit satisfactory n results on reliability and validity. Although scanning and planning are highly correlated, the variance inflation factor (VIF) values are below 2.0, which indicates that multicollinearity is not present. Following the procedure by Anderson and Gerbing (1988), we also ensure that discriminant validity is given between scanning and planning processes. International entrepreneurial orientation was measured with several items on the Industries Mechanical engineeringChemicals 198TextilePolymer 32.8 ProcessingMiscellaneousMissingTotal 87 1–19 126 14.4 103 77 20.9 116 17.1 100–199constituent 12.7 19.2 20–49 141 0 200–250 1–4.9 126 23.3 50–99 604 55 0 100 20.9 179dimensions 121 20–49.9 Missing 9.1 Total 29.6 20.0 5–9.9 127 10–19.9 21.0 0 604 50 + 131of 0–0.9 100 142 0 21.7risk 10–19.9 42 Total 23.5 84 Missing taking, 69 7.0 13.9 1–4.9 5–9.9 11.4 718 41 223 100 20 + 110 6.8 36.9 Total 18.2 Missing 63 10.5 55 604 100 9.1

SWOBODA AND OLEJNIK OLEJNIK 147 9 10 148 Table 2 Reliability and Validity of the Measurements

Item Item-to-Total Cronbach’s Factor AVE CR Operationalization

Correlation α (≥0.7) Loadings λ(z) (CFA) (CFA) Source (≥0.3) (CFA) (≥0.3) (≥0.5) (≥0.7)

Model Constructs Scanninga 0.912 0.675 0.912 Adapted from Miller and ORA FSALBSNS MANAGEMENT BUSINESS SMALL OF JOURNAL ORA FSALBSNS MANAGEMENT BUSINESS SMALL OF JOURNAL Systematic Foreign Market Research 0.745 0.790 Friesen (1982), Beal Analysis of the Foreign Market Structure 0.797 0.835 (2000) Analysis of Market Opportunities 0.822 0.870 Analysis of Customer Developments 0.784 0.837 Analysis of Local Competition 0.736 0.780 Planninga 0.841 0.575 0.843 Own measure based on Scheduling (i.e., We Have Mandatory Deadlines) 0.610 0.679 Robinson and Pearce Strategic Planning (i.e., We Plan Our International 0.725 0.830 (1983) Engagement Well in Advance) We Have a Written Strategic Plan 0.700 0.763 We Have Medium-Term Plans for Our International 0.671 0.749 Engagement (Three Years) International Entrepreneurial Orientationb 0.760 0.276 0.775 Own measure based on When Chances and Risks Are Equally Distributed in 0.309 0.384 Miller and Friesen International Decision-Making Situations, We (1982), Covin and Refrain From The Project (Reverse Coded) Slevin (1989), and We Have a Proclivity for High-Risk Over Low-Risk 0.332 0.429 Knight and Cavusgil Projects (2004) Technology Is Our Special Strength 0.467 0.417 Flexibility and Customer Focus Are Important to Us 0.306 0.321 We Focus on the Development of New Products and 0.468 0.487 Innovations The Enforcement of Unconventional Ideas and the 0.495 0.506 Acceptance of Creative Thinkers Are Important to Us We Are Proactively Growth and Profit Oriented 0.438 0.578 We Act Based on the Belief that Our Firm’s Future 0.437 0.649 Lies in International Markets We Actively Encourage the International Orientation 0.520 0.725 of Our Employees We Consistently Trust in Our Own Strengths 0.459 0.493 International Performancec 0.852 0.676 0.861 Adapted from Cavusgil Sales Growth 0.616 0.655 and Zou (1994) and ROI 0.760 0.864 Evans and Mavondo Profit 0.808 0.953 (2002) Instrumental Constructs Foreign Market Reporting Relevanced 0.905 0.773 0.911 Own measure based on Foreign Market Controlling in General 0.769 0.832 Cadogan et al. (2006) Receiving Foreign Market Reports Regularly 0.873 0.930 Receiving Reports on Market Developments 0.795 0.872 Foreign Market Controlling Figurese 0.769 0.558 0.788 Own measure based on Sales 0.587 0.734 Walters (1993) Turnover/Profits 0.719 0.862 WBD N LJI 149 OLEJNIK AND SWOBODA WBD N LJI 11 OLEJNIK AND SWOBODA Personnel 0.556 0.669 Perceived Distanceb 0.781 0.559 0.789 Own measure based on We Consciously Travel Abroad to Learn about 0.586 0.735 Dichtl, Koeglmayr, Countries and Cultures and Mueller (1990) Asian Cultures Seem Very Distant to Us (Reverse 0.590 0.663 Coded) The Geographic Distance to Overseas Markets Does 0.691 0.869 Not Pose a Problem

Confirmatory model fit: χ2 = 1629.245, df = 413, RMSEA (root mean square error of approximation) = 0.070, CFI (comparative fit index) = 0.872. Confirmatory model fit (international entrepreneurial orientation measured by three parcels): χ2 = 693.770, df = 231, RMSEA = 0.059, CFI = 0.942. aMeasured with seven-point Likert-type scales: To what extent does your firm carry out the following scanning/planning processes? Please answer on a scale between 1 = not at all and 7 = to a very high extent. bMeasured with seven-point Likert-type scales: To what extent do you agree to the following statements? Please answer on a scale between 1 = completely disagree and 7 = completely agree. cMeasured with seven-point Likert-type scales: How successful have you been on average during the last five years? Please answer using one of the following categories: 1 = more than 20 percent decrease, 2 = 11–20 percent decrease, 3 = 1–10 percent decrease, 4 = unchanged, 5 = 1–10 percent increase, 6 = 11–20 percent increase, 7 = more than 20 percent increase. dMeasured with seven-point Likert-type scales: How important are the following activities for your firm? Please answer on a scale between 1 = completely unimportant and 7 = extremely important. eMeasured with seven-point Likert-type scales: To what extent does your firm have monthly information on the following figures abroad? Please answer on a scale between 1 = not at all and 7 = to a very high extent. df, degrees of freedom. innovativeness, and proactiveness (Covin and recent studies on entrepreneurial orientation Slevin 1989). In the pretests, we adapted several (Moreno and Casillas 2008; Ripollés-Meliá, items of the original scales of Miller and Friesen Menguzzato-Boulard, and Sánchez-Peinado (1982) and Covin and Slevin (1989) in terms of 2007) in using three parcels instead of 10 the wording and the international focus of our items to represent the construct (i.e., we use study (Knight and Cavusgil 2004). Because one summed score for each subdimension). Rauch et al. (2009) found in their meta-study Thereby, we account of the three constituents of that different entrepreneurial orientation mea- international entrepreneurial orientation and surements did not significantly differ in their improve the CFA model fit. Finally, discrimi- performance implications, we accepted the nec- nant validity between international entrepre- essary changes to the scales. Another important neurial orientation and the other constructs can aspect of the discussion and analysis of entre- be established. preneurial orientation is the dimensionality of International performance was measured the construct. Previous research has highlighted using a combination of financial and nonfinan- that although the constructs represent different cial indicators, as well as strategic effectiveness. aspects of the entrepreneurial attitude, they are Respondents were asked to indicate the extent empirically related and must be combined to to which a number of financial and nonfinan- model entrepreneurial orientation properly cial indicators had changed in the last five-year (Ibeh 2003; Robertson and Chetty 2000). More- period (Cavusgil and Zou 1994; Evans and over, early entrepreneurship studies have Mavondo 2002). As entrepreneurial orientation defined the entrepreneurial orientation con- is said to mostly relate to financial performance struct as a composite measure because entrepre- (Rauch et al. 2009), we use financial indicators neurial firms must score high on all three to model international performance in this dimensions (Miller 1983). In their comprehen- study only. sive discussion of the dimensionality of the To model the bidirectional relationship entrepreneurial orientation construct, Covin, between processes and entrepreneurial orien- Green, and Slevin (2006) further highlight that tation, we had to include three latent instru- entrepreneurship research does not question mental construct (see Kline 2011 for whether the subdimensions exist as distinct methodological requirements in nonrecursive constructs in reality but whether they must vary models). We used the relevance of foreign independently. As we follow the conceptualiza- market reporting as an antecedent of scanning tion of Miller (1983), we assume that entrepre- processes, as general attitudes often pose as neurial firms are firms that score rather high on determinants of behavior (Ajzen 1991). Simi- all three dimensions (Mostafa, Wheeler, and larly, we used the foreign market controlling Jones 2005). Therefore, we associate the 10 figures as determinants of planning processes. items measuring the three subdimensions of risk Thereby, we assumed that a firm that has infor- taking, innovativeness, and proactiveness with mation on key foreign figures would also be one latent construct. The items used in this study more formal in its planning efforts. Moreover, yield an adequate Cronbach’s alpha (0.760) and we propose that the perceived psychic distance exhibit modest but still acceptable factor load- to overseas markets is an important determi- ings in confirmatory factor analysis. Although nant of the international entrepreneurial orien- the composite reliability shows reasonable tation (Dichtl, Koeglmayr, and Mueller 1990). values, the AVE is below common thresholds, Based on former research (Freeman and implying that the latent construct does not suf- Cavusgil 2007; Nummela, Saarenketo, and ficiently explain its items. This circumstance is Puumalainen 2004), we believe that the less most likely due to the dimensionality issue. distance the SME manager perceives, the more Hence, we performed a principal component entrepreneurial and global his mindset will be, factor analysis with oblique rotation and found which also influences the international entre- that three factors were extracted based on preneurial orientation. Furthermore, discrimi- the Kaiser criterion: one factor for the first nant validity is shown for all constructs. two items (risk taking), one factor for the fol- We included six control variables in the lowing four items (innovativeness), and one study. Considering the efficient allocation of factor for the last four items (proactiveness) (the resources, firms’ age was measured using the items are shown in Table 2). We maintain our number of years of the company’s existence. conceptualization of the construct and follow Firm size was measured using the number of

12150 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT employees. International experience was con- latent constructs (important constructs such as sidered because a higher level of international strategy, orientation, and performance are experience could result in a more successful latent and multidimensional) and their simulta- implementation of processes caused by learn- neous complex relationships (Shook et al. ing effects. We measured international experi- 2004). However, we also must acknowledge ence using the number of years a firm had been that the use of SEM should follow certain stan- operating in foreign countries. The time lag dards to unfold its utility. After a complete between firm founding and first international description of the sample selection procedure activity has been of special interest in recent and the empirical data, measurement reliability years (Acedo and Jones 2007). In testing for its and validity must be evaluated. Next, research- performance implications, we measured time ers must assess both the structural coefficients lag using the time span in years from founding and the model fit. Brown (2006) recommends to the initiation of internationalization (Baum, the use of at least three fit indices: one absolute Schwens, and Kabst 2011). Using country fit index, one absolute fit index with parsimony as a proxy for geographical distance, it was correction, and one comparative fit index.1 coded as a dummy variable with 1 = EU and Accordingly, we will report the χ2 statistic, the 0 = non-EU (Pedersen, Petersen, and Benito RMSEA, and the CFI. Finally, when modeling 2002). Lower distance (e.g., within the EU) is mediated relationships, it is important to supposed to reduce uncertainty and might control for the endogeneity of the constructs. therefore increase performance. Foreign sales Hence, we correlate the disturbance terms of ratio (i.e., the ratio of foreign sales to current the endogenous constructs with each other total sales) was utilized as a proxy for the (Antonakis et al. 2010). In correlating the error degree of internationalization (Baum, Schwens, terms, researchers acknowledge that there and Kabst 2011), which may be performance might be a common cause affecting the mea- related (Dhanaraj and Beamish 2003). surement error of the endogenous constructs.

Method Results The hypotheses are tested with several SEMs First, we test the simultaneous effect of the computed with the statistical program Mplus control variables on performance. Because we (Muthén & Muthén, Los Angeles, CA, USA). In find that only firm size has a significant effect MPLUS, missing values for the scaled items are on performance (p < .05), we decide to main- replaced by the full information maximum- tain size as the only covariate in the subsequent likelihood (FIML) approach. The FIML models to reduce model complexity. Second, approach has been identified as one of the we test the individual direct relationships (see preferred methods of addressing missing data Table 3). Models 1, 2, and 3 exhibit positive (Graham 2009). We chose the SEM approach and significant effects from scanning, planning, because SEM offers much potential for small and international entrepreneurial orientation, firm research, as it is capable of analyzing respectively, on performance. Model 4

1First, Brown recommends an absolute fit index (such as χ2) that evaluates how close the observed variance–covariance matrix is to the estimated matrix. Researchers would desire a result that is not statistically significant (i.e., the observed covariance matrix equals the estimated matrix). However, the χ2 statistic is very sensitive to sample size. For large sample sizes, even small differences between the two matrices lead to significant results (Shook et al. 2004). Second, Brown recommends an absolute fit index with parsimony correction (such as RMSEA) that incorporates a penalty for poor model parsimony. The RMSEA ranges from 0to∞, and according to Hu and Bentler (1999), a cutoff value close to 0.06 is necessary to conclude that there is a relatively good fit between the hypothesized model and the observed data. Hair et al. (2010) recommend a cutoff value of 0.07. Third, Brown recommends a comparative fit index (such as the CFI) that evaluates the fit of the hypothesized model to a null model with covariances = 0. The CFI ranges from 0 to 1 (where 1 indicates perfect fit), and according to Hu and Bentler (1999), a cutoff value close to 0.95 is necessary to assume a good fit. Hair et al. (2010) state that a cutoff value of 0.90 is “usually” (669) acceptable, but the authors distinguish cutoff values based on sample size and number of indicators. For a large sample (N > 250) and more than 30 observed variables, Hair et al. (2010) would expect a significant χ2 and RMSEA < 0.07 with a CFI of 0.900 or higher to demonstrate goodness of fit (672). However, one must note that cutoff values are rules of thumb and do not provide statistical tests of significance.

SWOBODA AND OLEJNIK OLEJNIK 151 13 14 152

Table 3 Results of SEM on the Mediation

Model 1 Model 2 Model 3 Model 4 Model 5a Model 6

ORA FSALBSNS MANAGEMENT BUSINESS SMALL OF JOURNAL βββββ β ORA FSALBSNS MANAGEMENT BUSINESS SMALL OF JOURNAL Paths

Scanning → Performance 0.160 ** −0.060 ns −0.062 ns Planning → Performance 0.197 *** 0.043 ns 0.042 ns International Entrepreneurial Orientation → 0.277 *** 0.280 * 0.280 * 0.258 *** Performance Scanning → International Entrepreneurial 0.308 *** 0.304 *** Orientation Planning → International Entrepreneurial 0.525 *** 0.529 *** Orientation Controls Firm Size 0.105 * 0.094 ns 0.078 ns 0.081 ns 0.088 ns 0.087 ns Model Fit χ2 113.198 *** 58.758 *** 13543 ns 267.314 *** 268.907 *** 269.213 *** df 25 18 12 95 96 97 RMSEA 0.076 0.061 0.015 0.055 0.055 0.054 CFI 0.967 0.975 0.998 0.962 0.962 0.962

*p ≤ .05. **p ≤ .01. ***p ≤ .001. Standardized estimates are shown. aModel estimated without correlation between error terms because the model was not identified otherwise. ns, not significant; SEM, structural equation model. considers the simultaneous direct influences of instrumental variables and the reciprocal rela- scanning, planning, and international entrepre- tionship between processes and international neurial orientation on performance. The results entrepreneurial orientation (see model 7). The show that the performance effect of scanning processes do impact international entrepre- and planning vanishes when the three con- neurial orientation, and international entrepre- structs are simultaneously considered. On the neurial orientation impacts the processes. The basis of these results, we can support H1. effect of entrepreneurial orientation on perfor- Model 5 tests the mediation model, which is mance is significantly positive as well. In con- proposed by H2 and H3. Model 5 exhibits clusion, we can support H4. strong and highly statistically significant paths Concluding, we support H4 in favor of from both scanning and planning to entrepre- a reciprocal relationship. Comparing the neurial orientation and from entrepreneurial reciprocal model (model 7) with a nested model orientation to performance. Further, the direct that does not include the reciprocity (model 8), effects of scanning and planning on interna- we find that model 7 is superior because tional performance are nonsignificant. This it has a significant better model fit (Δχ2 finding statistically supports complete media- [Δdf = 2] = 151.905, p ≤ .001). This result sup- tion (Baron and Kenny 1986). Following the ports the consideration of the bidirectional rela- criteria for model fit assessment provided by tionship between international entrepreneurial Hu and Bentler (1999) and Hair et al. (2010), orientation and scanning and planning. the model fit reaches acceptable values (RMSEA Because the structural relations could be dif- 0.060, CFI 0.961), although the χ2 is statistically ferent considering different performance mea- significant. Therefore, we can conclude that sures, we assessed model 6 with five different both H2 and H3 are supported. performance measurements: (1) perceived To further validate the complete mediation international financial performance, as dis- through SEM, the model fit of the mediated cussed before; (2) perceived overall satisfaction model (model 5) can be compared with a regarding international performance; (3) per- model without the direct effects (model 6). ceived achievement of financial and nonfinan- Thereby, we test whether the direct perfor- cial goals; and (4) foreign sales. We find that mance effects of scanning and planning are the structural relationships stay the same no negligible. If the mediated model (model 5) is matter which performance measure is used. not significantly better than the indirect model Although the effect sizes and the model fit (model 6), then the direct performance effects differ, the relationships remain constant, imply- of scanning and planning can indeed be ing that the performance effect of scanning and neglected. A chi-squared (χ2) difference test planning is mediated by the international entre- suggests that the mediated model (model 5) is preneurial orientation. not significantly better than the indirect model (model 6)—both with and without the error Discussion and Conclusions correlation implemented in model 6. We there- Scanning leads to superior performance, as fore choose the parsimonious model with more do thorough planning processes. Likewise, degrees of freedom (df) because both models fit a high level of international entrepreneurial the data similarly well, which implies that we orientation leads to a high level of interna- could reduce complexity by choosing model 6 tional performance. However, when all three over model 5 without losing information. “success factors” are considered at the same Model 6 proposes that scanning and plan- time, the performance effect vanishes. This ning have indirect effects on performance only. finding can be explained by the intervening Scanning has a significant indirect effect of mechanism of international entrepreneurial ori- β = 0.079 on performance (p ≤ .01). Planning entation that has thus far been mostly disre- also has a significant overall indirect effect of garded. Although other studies have derived β = 0.137 on performance (p ≤ .001). international entrepreneurial orientation as an H4 proposes that international entrepreneur- antecedent factor (Jones and Coviello 2005; ial orientation has a positive effect on scanning Knight and Cavusgil 2004), we showed that the and planning processes and is tested via a effect of scanning and planning on interna- nonrecursive structural model (see Table 4). tional performance is completely mediated We find that all structural coefficients are highly by international entrepreneurial orientation. significant, supporting the relevance of the Hence, the impact that scanning and planning

SWOBODA AND OLEJNIK OLEJNIK 153 15 Table 4 Results of SEM on the Bidirectional Relationship

Model 7 Model 8 Paths ββ

International Entrepreneurial Orientation → 0.139 ** 0.200 *** Performance Scanning → International Entrepreneurial Orientation 0.421 *** 0.211 *** Planning → International Entrepreneurial Orientation 0.481 *** 0.412 *** International Entrepreneurial Orientation → Scanning 0.486 *** International Entrepreneurial Orientation → Planning 0.330 *** Instrumental Variables Foreign Market Reporting → Scanning 0.330 *** 0.745 *** Foreign Market Controlling → Planning 0.405 *** 0.873 *** Geographic Distance → International Entrepreneurial 0.384 *** 0.537 *** Orientation Controls Firm Size 0.103 * 0.101 ns Model Fit χ2 824.058 *** 975.963 *** df 260 262 RMSEA 0.060 0.067 CFI 0.930 0.912

Standardized estimates are shown. *p ≤ .05. **p ≤ .01. ***p ≤ .001. ns, not signiﬁcant; SEM, structural equation model.

have on performance is fully channeled by orientation for internationally established international entrepreneurial orientation such SMEs. that the direct process–performance relation- Increasing the objective knowledge of SME ships are negligible. This extraordinary result managers through foreign market scanning might explain the variance in the results of and planning may either directly impact inter- earlier research on scanning and planning national performance or indirectly affect inter- processes (Julien and Ramangalahy 2003; national performance through changing Schwenk and Shrader 1993) and the “why” of international entrepreneurial orientation. We the process–performance relationship (Baron argued that learning more about relevant and and Kenny 1986). We can conclude that scan- possibly distant markets reduces the perceived ning and planning lead to success because of risks of international business. When the risk is international entrepreneurial orientation, as perceived to be more accessible, the SME international entrepreneurial orientation con- manager may be willing to take more risks. stitutes a dynamic capability that transforms Such learning also opens up new avenues for business processes in response to changing business opportunities that can be proactively environments and thereby creates economic acted upon. Hence, an increase in processes rents. Thus, international entrepreneurial ori- can lead to an increase in international entre- entation explains why and how processes con- preneurial orientation. Thereby, we argued, tribute to ﬁrm performance. This result based on the basic idea of the Stimulus- strengthens the signiﬁcance of entrepreneurial Organism-Response model, that processes are

16154 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT easier to change than behavioral attitudes. and show that the effect that scanning has on Hence, processes can function as stimuli in entrepreneurial orientation is larger than the changing or at least affecting international way around, which gives support to Covin and entrepreneurial orientation. International entre- Slevin’s thesis. However, our model showed preneurial orientation determines international- that planning is influenced by entrepreneurial ization behavior (i.e., whether the firm enters orientation to a greater extent than planning new markets, which markets, which products), affects entrepreneurial orientation. Further- and behavior ultimately impacts performance. more, Covin and Slevin (1991) noted in their Within this mediated model, the conceptualiza- model that resources and competences may tion of international entrepreneurial orientation also influence the relationship between entre- as a dynamic capability played a central role. preneurial orientation and firm performance. We differentiated between resources, pro- Clearly, moderation is another plausible causal cesses, and (dynamic) capabilities, highlighting mechanism for the interplay of resources and their hierarchical ordering. Knight and Cavusgil capabilities. In our case, modeling interactions (2004) argued from a dynamic capability per- would classify the conditions under which spective that international entrepreneurial ori- scanning and planning processes influence per- entation can be viewed as a dynamic capability formance, which represents a different but because it gives rise to successful entry into highly relevant research question. Although new markets. In fostering innovative, proactive, several moderating relationships of the entre- and risk-taking behavior, a firm can constantly preneurial orientation–performance relation- renew its resources and competences to ship have been already considered in the achieve international objectives. Whereas literature, there is potential for further research Knight and Cavusgil (2004) interpreted interna- (Rauch et al. 2009). tional entrepreneurial orientation as antecedent Overall, this study contributes to the to business processes, we followed the research on the performance effects of pro- capability-building perspective (Makadok cesses and dynamic capabilities within estab- 2001). However, we could show that there is a lished international SMEs. We advance theory bidirectional relationship between processes by modeling international entrepreneurial ori- and international entrepreneurial orientation in entation as dynamic capability that intervenes international markets. Thereby, we refer to in the process–performance relationship. Covin and Slevin (1991), who argued in their Thereby, we explain why SMEs can capitalize seminal work that entrepreneurial orientation on processes and show that international entre- both influences and is influenced by organiza- preneurial orientation does not only initiate tional resources and competences. The authors and impact processes of SMEs—international have further elaborated: “A firm’s ability to entrepreneurial orientation is also influenced engage in entrepreneurial behavior will by the very same processes. International entre- depend, in part, on its resources and compe- preneurial orientation must be understood as a tencies. These variables are operationally key dynamic capability transforming and adapt- defined in the broadest sense and are intended ing firm processes in changing environments to to include (. . .) organizational-level capabilities perform well in international markets, not only (for example, ability to get a new product to the as an antecedent factor. Thereby, this study market in a timely fashion), and organizational closes a knowledge gap on how SMEs create systems (for example, marketing research competitive advantage from processes by systems). Resources and competencies provide employing the dynamic capability perspective. the bases for all forms of organizational action. Moreover, with a large sample of international They can serve as either facilitators or deter- SMEs, this study contributes to the knowledge rents of entrepreneurial behavior, and influ- on the international performance–international ence the specific form of entrepreneurship in entrepreneurial orientation relationship. which the firm engages.” (p. 15). In their con- From a practical perspective, SME managers ceptual model, Covin and Slevin (1991) already learn that (1) scanning and planning are crucial introduced the idea of a possibly reciprocal to a firm’s international success because of their relationship assuming that capabilities and international entrepreneurial orientation and resources have a stronger effect on entrepre- that (2) they, as managers, have two main chan- neurial orientation than the other way around. nels by which to influence the probabilities that Our results support a bidirectional relationship they succeed in international markets. This

SWOBODA AND OLEJNIK OLEJNIK 155 17 finding means that the contribution of manag- Therefore, the relationship between scanning ers lies in designing, developing, and construct- and entrepreneurial orientation might not be ing capabilities, and in anticipating the value of completely linear. At some point, further infor- resources and processes. The first perspective mation may no longer contribute to interna- highlights the fact that managers must make tional entrepreneurial orientation. Managers use of resources and processes to adapt to the should be aware of that. Furthermore, planning environment. Processes enable the firm to build is a heterogeneous task. Future research needs a basis for the pursuit of entrepreneurship. to better account for this heterogeneity. One Already, Barringer and Bluedorn (1999) have might want to focus on a specific planning area shown that environmental scanning and some and differentiate between planning intensity, planning dimensions significantly impact cor- planning responsibility, planning flexibility, porate entrepreneurship. The authors summa- planning horizon, or planning locus as not all rized with the following: “What is particularly areas may be of equal importance to interna- instructive about this result is that the pursuit of tional entrepreneurial orientation and ulti- entrepreneurship requires an increase in the mately to firm success (Barringer and Bluedorn intensity of some management practices, such 1999). Another conceptual question is the con- as scanning intensity. Opportunity recognition, ceptualization and dimensionality of entrepre- which is a precursor to entrepreneurial behav- neurial orientation. Although Rauch et al. ior, is often associated with a flash of genius, (2009) found in their meta-analysis that differ- but in reality is probably more often than not ent conceptualizations and specifications of the end result of a laborious process of envi- entrepreneurial orientation do not influence its ronmental scanning and industry awareness. As overall effect on performance, researchers must a result, the fundamental practice of scanning be conscious of the specification they employ the environment to recognize opportunities (a comprehensive discussion can be found in and threats should be a principal concern of Covin, Green, and Slevin 2006). Following entrepreneurially minded firms” (p. 436). Miller (1983), we defined international entre- Hence, the foundation of entrepreneurship can preneurial orientation as being reflected in be established in international entrepreneurial innovativeness, proactiveness, and risk taking. orientation as a dynamic capability that inte- Lumpkin and Dess (1996), however, consider grates and configures resources and processes. two additional dimensions as being part of the Highlighting the second perspective, managers construct. Other studies argue that the dimen- must realize that there are complex feedback- sions must be addressed as independent con- loops. International entrepreneurial orientation structs (Kreiser, Marino, and Weaver 2002), and also impacts the extent to which scanning and some even suggest a formative understanding planning processes are carried out. Managers (Covin, Green, and Slevin 2006). also contribute through their anticipation of the From a methodical perspective, our study value of certain resources and processes. is limited because of our focus on internationally active SME from Germany and the use of Limitations and Directions for perceptual data. Although our sample design Further Research is consistent with our research aim and is There are several limitations to this study, also consistent with other research studies on which we will attend to in the following. In similar research objects (Maekelburger, doing so, we will also discuss avenues for Schwens, and Kabst 2012), we cannot claim the further research. complete generalization of our results. As we We have considered only two firm pro- focused on four main manufacturing industries cesses. Future studies could analyze and test within one single country, we cannot general- the relationship of other processes and firm- ize the results to other sectors or to other level factors with international entrepreneurial countries. Researching SMEs with international orientation (Covin and Slevin 1991). Addition- activities, it is not possible to transfer our ally, resources as a basis for processes could be results to SMEs that only operate in the home included in future studies. Moreover, two market. Furthermore, it would be beneficial for natural limitations of information and hence future studies to collect data on performance scanning should be addressed. First, informa- from a different source which was not possible tion does not have obvious units. Second, infor- for us as German firms are not obliged to mation may not be cumulative (Fiet 1996). publish financial data. Moreover, although our

18156 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT models showed acceptable fit indices in terms Antonakis, J., S. Bendahan, P. Jacquart, and R. of recommended cutoff values, the model fit is Lalive (2010). “On Making Causal Claims: A not perfect and must be taken as a limitation of Review and Recommendations,” The Leader- our study. Similarly, modeling causal relationship Quarterly 21(6), 1086–1120. ships based on cross-sectional data leads to Armario, J. M., D. M. Ruiz, and E. M. Armario tentative rather than conclusive results. Hence, (2008). “Market Orientation and Internation- a longitudinal research design could extend alization in Small and Medium-Sized Enter- findings on the relevance and development of prises,” Journal of Small Business international entrepreneurial orientation, Management 46(4), 485–511. which might support the causal relationships Armstrong, J. S., and T. S. Overton (1977). drawn in this study and the reciprocal effects “Estimating Nonresponse Bias in Mail between international entrepreneurial orienta- Surveys,” Journal of Marketing Research tion and processes. 14(3), 396–402. Finally, our study has highlighted the Baird, I. S., M. A. Lyles, and J. B. Orris (1994). complex and bidirectional relationships “The Choice of International Strategies by between firm processes and dynamic capabili- Small Businesses,” Journal of Small Business ties in creating competitive advantage. As we Management 32(1), 48–57. have interpreted international entrepreneurial Baker, W. E., and J. M. Sinkula (2009). “The orientation as a dynamic capability, future Complementary Effects of Market Orienta- research should analyze the proposed relation- tion and Entrepreneurial Orientation on ships in different settings, different types of Profitability in Small Businesses,” Journal firms, and different environmental conditions. of Small Business Management 47(4), 443– Researching the boundary conditions is essen- 464. tial to determine the limitations of dynamic Balabanis, G. I., and E. S. Katsikea (2003). capabilities (Barreto 2010). We could imagine “Being An Entrepreneurial Exporter: Does It that the role of international entrepreneurial Pay?,” International Business Review 12(2), orientation is most dominant in challenging, 233–252. competitive, and quickly changing environ- Barney, J. B. (1991). “Firm Resources and Sus- ments. Another idea is the differentiation of tained Competitive Advantage,” Journal of service firms. In this study, we cannot draw any Management 17(1), 99–120. conclusion about service firms because we Baron, R. M., and D. A. Kenny (1986). “The focused on the manufacturing industry. As Moderator-Mediator Variable Distinction in service firms must often establish a physical Social Psychological Research: Conceptual, presence in new country markets, they may Strategic, and Statistical Considerations,” have to strongly rely on capability building for Journal of Personality and Social Psychology rent creation. 51(6), 1173–1182. Barreto, I. (2010). “Dynamic capabilities: A References Review of Past Research and an Agenda for Acedo, F. J., and M. V. Jones (2007). “Speed of the Future,” Journal of Management 36(1), Internationalization and Entrepreneurial 256–280. Cognition: Insights and a Comparison Barringer, B. R., and A. C. Bluedorn (1999). between International New Ventures, “The Relationship between Corporate Entre- Exporters and Domestic Firms,” Journal of preneurship and Strategic Management,” World Business 42(3), 236–252. Strategic Management Journal 20(5), 421– Ajzen, I. (1991). “The Theory of Planned 444. Behavior,” Organizational Behavior and Baum, M., C. Schwens, and R. Kabst (2011). “A Human Decision Processes 50(2), 179–211. Typology of International New Ventures: Amit, R., and P. J. H. Schoemaker (1993). “Stra- Empirical Evidence from High-Technology tegic Assets and Organizational Rent,” Stra- Industries,” Journal of Small Business Man- tegic Management Journal 14(1), 33–46. agement 49(3), 305–330. Anderson, J. C., and D. W. Gerbing (1988). Beal, R. M. (2000). “Competing Effectively: “Structural Equation Modeling in Practice: A Environmental Scanning, Competitive Strat- Review and Recommended Two-Step egy, and Organizational Performance in Approach,” Pyschological Bulletin 103(3), Small Manufacturing Firms,” Journal of 411–423. Small Business Management 38(1), 27–47.

SWOBODA AND OLEJNIK OLEJNIK 157 19 Bracker, J. S., B. W. Keats, and J. N. Pearson Dhanaraj, C., and P. W. Beamish (2003). “A (1988). “Planning and Financial Performance Resource-Based Approach to the Study of Among Small Firms in a Growth Industry,” Export Performance,” Journal of Small Busi- Strategic Management Journal 9(6), 591– ness Management 41(3), 242–261. 603. Dichtl, E., H.-G. Koeglmayr, and S. Mueller Brouthers, L. E., and G. Nakos (2005). “The (1990). “International Orientation as a Pre- Role of Systematic International Market condition for Export Success,” Journal of Selection on Small Firms’ Export Perfor- International Business Studies 21(1), 23– mance,” Journal of Small Business Manage- 40. ment 43(4), 363–381. Dimitratos, P., S. Lioukas, and S. Carter (2004). Brown, T. A. (2006). Conﬁrmatory Factor “The Relationship Between Entrepreneur- Analysis for Applied Research. New York: ship and International Performance: The The Guilford Press. Importance of Domestic Environment,” Cadogan, J. W., and A. Diamantopoulos (1995). International Business Review 13(1), 19– “Narver and Slater, Kohli and Jaworski and 41. the Market Orientation Construct: Integra- Eisenhardt, K. M., and J. A. Martin (2000). tion and Internationalization,” Journal of “Dynamic Capabilities: What Are They?,” Strategic Marketing 3(1), 41–60. Strategic Management Journal 21(10/11), Cadogan, J. W., C. C. Cui, R. E. Morgan, and V. 1105–1111. M. Story (2006). “Factors Facilitating and Evans, J., and F. T. Mavondo (2002). “Psychic Impeding the Development of Export Distance and organizational Performance: Market-Oriented Behavior: A Study of Hong An empirical Examination of international Kong Manufacturing Exporters,” Industrial Retailing Operations,” Journal of Interna- Marketing Management 35(5), 634–647. tional Business Studies 33(3), 515–532. Cavusgil, S. T., and S. Zou (1994). “Marketing Fiet, J. O. (1996). “The Informational Basis of Strategy-Performance Relationship: An Entrepreneurial Discovery,” Small Business Investigation of the Empirical Link in Export Economics 8(6), 419–430. Market Ventures,” Journal of Marketing Freeman, S., and S. T. Cavusgil (2007). “Toward 58(1), 1–21. a Typology of Commitment States Among Cooper, A. C., T. B. Folta, and C. Woo (1995). Managers of Born-Global Firms: A Study of “Entrepreneurial Information Search,” Accelerated Internationalization,” Journal of Journal of Business Venturing 10(2), 107– International Marketing 15(4), 1–40. 120. Frishammar, J., and S. Andersson (2009). “The Covin, J. G., and D. P. Slevin (1989). “Strategic Overestimated Role of Strategic Orientations Management of Small Firms in Hostile and for International Performance in Smaller Benign Environments,” Strategic Manage- Firms,” Journal of International Entrepre- ment Journal 10(1), 75–87. neurship 7(1), 57–77. ——— (1991). “A Conceptual Model of Gilbert, C. G. (2006). “Change in The Presence of Entrepreneurship as Firm Behavior,” Entre- Residual Fit: Can Competing Frames preneurship: Theory & Practice 16(1), 7– Coexist?,” Organization Science 17(1), 150– 25. 167. Covin, J. G., K. M. Green, and D. P. Slevin Graham, J. W. (2009). “Missing Data Analysis: (2006). “Strategic Process Effects on the Making It Work in the Real World,” Annual Entrepreneurial Orientation–Sales Growth Review of Psychology 60(NN), 549–576. Rate Relationship,” Entrepreneurship: Gruber, M. (2007). “Uncovering the Value of Theory & Practice 30(1), 57–81. Planning in New Venture Creation: A Crick, D., and M. Spence (2005). “The Process and Contingency Perspective,” Internationalisation of “High Performing” Journal of Business Venturing 22(6), 782– UK High-Tech SMEs: A Study of Planned and 807. Unplanned Strategies,” International Busi- Hair, J. F., R. L. Tatham, R. E. Anderson, and W. ness Review 14(2), 167–185. Black (2010). Multivariate Data Analysis. Delmar, F., and S. Shane (2003). “Does Busi- Upper Saddle River, NJ: Pearson Prentice ness Planning Facilitate the Development of Hall. New Ventures?,” Strategic Management Hu, L., and P. Bentler (1999). “Cutoff Criteria Journal 24(12), 1165–1185. for Fit Indexes in Covariance Structure

20158 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Analysis: Conventional Criteria Versus New and Medium-Sized Family Firms,” Journal of Alternatives,” Structural Equation Modeling: Small Business Management 49(3), 490–514. A Multidisciplinary Journal 6(1), 1–55. Kreiser, P. M., L. D. Marino, and K. M. Weaver Hughes, M. A., and D. E. Garrett (1990). (2002). “Assessing the Psychometric Proper- “Intercoder Reliability Estimation Appro- ties of the Entrepreneurial Orientation Scale: aches in Marketing: A Generalizability A Multi-Country Analysis,” Entrepreneur- Theory Framework for Quantitative Data,” ship: Theory & Practice 26, 4–71. Journal of Marketing Research 27(2), 185– Kumar, N., L. W. Stern, and J. C. Anderson 195. (1993). “Conducting Interorganizational Ibeh, K. I. N. (2003). “Toward a Contingency Research Using Key Informants,” Academy Framework of Export Entrepreneurship: of Management Journal 36(6), 1633– Conceptualisations and Empirical Evidence,” 1651. Small Business Economics 20(1), 49–68. Li, L., D. Li, and T. Dalgic (2004). “Internation- Jantunen, A., K. Puumalainen, S. Saarenketo, alization Process of Small and Medium-sized and K. Kyläheiko (2005). “Entrepreneurial Enterprises: Toward a Hybrid Model of Orientation, Dynamic Capabilities and Inter- Experiential Learning and Planning,” Man- national Performance,” Journal of Interna- agement International Review 44(1), tional Entrepreneurship 3(3), 223–243. 93–116. Jaworski, B. J., and A. K. Kohli (1993). “Market Lu, Y., L. Zhou, G. Bruton, and W. Li (2010). Orientation: Antecedents and Conse- “Capabilities as a Mediator Linking quences,” The Journal of Marketing 57(3), Resources and the International Perfor- 53–70. mance of Entrepreneurial Firms in an Emerg- Johanson, J., and J.-E. Vahlne (1977). “The ing Economy,” Journal of International Internationalization Process of the Firm—A Business Studies 41(3), 419–436. Model of Knowledge Development and Lumpkin, G. T., and G. G. Dess (1996). “Clari- Increasing Foreign Market Commitments,” fying the Entrepreneurial Orientation Con- Journal of International Business Studies struct and Linking It to Performance,” The 8(1), 23–32. Academy of Management Review 21(1), 135– Jones, M. V., and N. E. Coviello (2005). 172. “Internationalisation: Conceptualising an Lyles, M. A., I. S. Baird, J. B. Orris, and Entrepreneurial Process of Behaviour in D. F. Kuratko (1993). “Formalized Planning Time,” Journal of International Business in Small Business: Increasing strategic Studies 36(3), 284–303. Choices,” Journal of Small Business Manage- Julien, P.-A., and C. Ramangalahy (2003). ment 31(2), 38–50. “Competitive Strategy and Performance of McGee, J. E., and O. O. Sawyerr (2003). “Uncer- Exporting SMEs: An Empirical Investigation tainty and Information Search Activities: A of the Impact of Their Export Information Study of Owner–Managers of Small High- Search and Competencies,” Entrepreneur- Technology Manufacturing Firms,” Journal ship: Theory and Practice 27(3), 227–246. of Small Business Management 41(4), 385– Kline, R. B. (2011). Principles and Practice of 401. Structural Equation Modeling. New York: Maekelburger, B., C. Schwens, and R. Kabst The Guilford Press. (2012). “Asset Speciﬁcity and Foreign Market Knight, G. A. (2001). “Entrepreneurship and Entry Mode Choice of Small and Medium- Strategy in the International SME,” Journal sized Enterprises: The Moderating Inﬂuence of International Management 7(3), 155–171. of Knowledge Safeguards and Institutional Knight, G. A., and S. T. Cavusgil (2004). “Inno- Safeguards,” Journal of International Busi- vation, Organizational Capabilities, and the ness Studies 43(5), 458–476. Born-Global Firm,” Journal of International Makadok, R. (2001). “Towards a Synthesis of Business Studies 35(2), 124–141. the Resource-Based and Dynamic-Capability Knight, G. A., and P. W. Liesch (2002). “Infor- Views of Rent Creation,” Strategic Manage- mation Internalisation in Internationalising ment Journal 22(5), 387–401. the Firm,” Journal of Business Research Martinez, J. I., and J. C. Jarillo (1991). “Coordi- 55(12), 981–995. nation Demands of International Strategies,” Kontinen, T., and A. Ojala (2011). “Interna- Journal of International Business Studies tional Opportunity Recognition among Small 22(3), 429–444.

SWOBODA AND OLEJNIK OLEJNIK 159 21 Matsuno, K., J. T. Mentzer, and A. Özsomer and Recommended Remedies,” The Journal (2002). “The Effects of Entrepreneurial Pro- of Applied Psychology 88(5), 879–903. clivity and Market Orientation on Business Rauch, A., J. Wiklund, G. T. Lumpkin, and M. Performance,” The Journal of Marketing Frese (2009). “Entrepreneurial Orientation 66(3), 18–32. and Business Performance: An Assessment Miller, C. C., and L. B. Cardinal (1994). “Strate- of Past Research and Suggestions for the gic Planning and Firm Performance: A Syn- Future,” Entrepreneurship: Theory & Prac- thesis of More than Two Decades of tice 33(3), 761–787. Research,” The Academy of Management Ray, G., J. B. Barney, and W. A. Muhanna Journal 37(6), 1649–1665. (2004). “Capabilities, Business Processes, Miller, D. (1983). “The Correlates of Entrepre- and Competitive Advantage: Choosing the neurship in Three Types of Firms,” Manage- Dependent Variable in Empirical Tests of the ment Science 29(7), 770–791. Resource-Based View,” Strategic Manage- Miller, D., and P. H. Friesen (1982). “Innovation ment Journal 25(1), 23–37. in Conservative and Entrepreneurial Firms: Ripollés-Meliá, M., M. Menguzzato-Boulard, Two Models of Strategic Momentum,” Stra- and L. Sánchez-Peinado (2007). “Entrepre- tegic Management Journal 3(1), 1–25. neurial Orientation and International Com- Miocevic, D., and B. Crnjak-Karanovic (2011). mitment,” Journal of International “Cognitive and Information-Based Capabili- Entrepreneurship 5(3), 65–83. ties in the Internationalization of Small and Robertson, C., and S. K. Chetty (2000). “A Medium-Sized Enterprises: The Case of Croa- Contingency-Based Approach to Under- tian Exporters,” Journal of Small Business standing Export Performance,” Interna- Management 49(4), 537–557. tional Business Review 9(2), 211–235. Moreno, A. M., and J. C. Casillas (2008). “Entre- Robinson, R. B., Jr., and J. A. Pearce (1983). preneurial Orientation and Growth of SMEs: “The Impact of Formalized Strategic Plan- A Causal Model,” Entrepreneurship: Theory ning on Financial Performance in Small & Practice 32(3), 507–528. Organizations,” Strategic Management Mostafa, R., C. Wheeler, and M. Jones (2005). Journal 4(3), 197–207. “Entrepreneurial Orientation, Commitment Schilke, O., and A. Goerzen (2010). “Alliance to the Internet and Export Performance in Management Capability: An Investigation of Small and Medium-Sized Exporting Firms,” the Construct and Its Measurement,” Journal Journal of International Entrepreneurship of Management 36(5), 1192–1219. 3(4), 291–302. Schwenk, C. B., and C. B. Shrader (1993). Naman, J. L., and D. P. Slevin (1993). “Entre- “Effects of Formal Strategic Planning on preneurship and the Concept of Fit: A Model Financial Performance in Small Firms: A and Empirical Tests,” Strategic Management Meta-Analysis,” Entrepreneurship: Theory & Journal 14(2), 137–153. Practice 17(3), 53–64. Nummela, N., S. Saarenketo, and K. Seringhaus, F. H. R. (1993). “A Comparison of Puumalainen (2004). “A Global Mindset–A Export Marketing Behavior of Canadian and Prerequisite for Successful Internationaliza- Austrian High-Tech Firms,” Journal of Inter- tion?,” Canadian Journal of Administrative national Marketing 1(4), 49–69. Sciences 21(1), 51–64. Shane, S., and S. Venkataraman (2000). “The Pedersen, T., B. Petersen, and G. R. G. Benito Promise of Entrepreneurship as a Field of (2002). “Change of Foreign Operation Research,” The Academy of Management Method: Impetus and Switching Costs,” Inter- Review 25(1), 217–226. national Business Review 11(3), 325–345. Shook, C. L., D. J. Ketchen, G. Tomas, M. Hult, Perks, K. J., and M. Hughes (2008). “Entrepre- and K. M. Kacmar (2004). “An Assessment of neurial Decision-Making in Internationaliza- The Use of Structural Equation Modeling in tion: Propositions from Mid-Size Firms,” Strategic Management Research,” Strategic International Business Review 17(3), 310– Management Journal 25(4), 397–404. 330. Simon, H. A. (1976). Administrative Behavior. Podsakoff, P. M., S. B. MacKenzie, L. New York: Free Press. Jeong-Yeon, and N. P. Podsakoff (2003). Spector, P. E. (2006). “Method Variance in “Common Method Biases in Behavioral Organizational Research,” Organizational Research: A Critical Review of the Literature Research Methods 9(2), 221–232.

22160 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Stevenson, H. H., and J. C. Jarillo (1990). “A Winter, S. G. (2003). “Understanding Dynamic Paradigm of Entrepreneurship: Entrepre- Capabilities,” Strategic Management Journal neurial Management,” Strategic Manage- 24(10), 991–995. ment Journal 11(1), 17–27. Yang, Z., X. Wang, and C. Su (2006). “A Review Su, Z., E. Xie, and Y. Li (2011). “Entrepreneurial of Research Methodologies in International Orientation and Firm Performance in New Business,” International Business Review Ventures and Established Firms,” Journal of 15(6), 601–617. Small Business Management 49(4), 558– Yeoh, P.-L. (2000). “Information Acquisition 577. Activities: A Study of Global Start-Up Export- Teece, D. J. (2007). “Explicating Dynamic Capa- ing Companies,” Journal of International bilities: The Nature and Microfoundations of Marketing 8(3), 36–60. (Sustainable) Enterprise Performance,” Stra- Yip, G. S., J. G. Biscarri, and J. A. Monti (2000). tegic Management Journal 28(13), 1319– “The Role of the Internationalization Process 1350. in the Performance of Newly International- Teece, D. J., G. Pisano, and A. Shuen (1997). izing Firms,” Journal of International Mar- “Dynamic Capabilities and Strategic Manage- keting 8(3), 10–35. ment,” Strategic Management Journal 18(7), Zahra, S. A. (1991). “Predictors and Financial 509–533. Outcomes of Corporate entrepreneurship: Walters, P. G. P. (1993). “Patterns of Formal An Exploratory Study,” Journal of Business Planning and Performance in U.S. Exporting Venturing 6(4), 259–285. Firms,” Management International Review ——— (1993). “A Conceptual Model of Entre- 33(1), 43–63. preneurship as Firm Behavior: A Critique Walters, P. G. P., and S. Samiee (1990). “A and Extension,” Entrepreneurship: Theory & Model for Assessing Performance in Small Practice 17(4), 5–21. U.S. Exporting Firms,” Entrepreneurship: Zahra, S. A., and D. M. Garvis (2000). “Interna- Theory & Practice 15(2), 33–50. tional Corporate Entrepreneurship and Firm Wang, C. L., and P. K. Ahmed (2007). “Dynamic Performance: The Moderating Effect of Inter- Capabilities: A Review and Research national Environmental Hostility,” Journal of Agenda,” International Journal of Manage- Business Venturing 15(5–6), 469–492. ment Reviews 9(1), 31–51. Zahra, S. A., H. J. Sapienza, and P. Davidsson Williams, J. E. M. (2003). “Export Information (2006). “Entrepreneurship and Dynamic Use in Small and Medium-Sized Industrial Capabilities: A Review, Model and Research Companies,” International Marketing Agenda,” Journal of Management Studies Review 20(1), 44–66. 43(4), 917–955.

SWOBODA AND OLEJNIK OLEJNIK 161 23 Journal of Small Business Business Management Management 2016 2014 54(1), ••(••), pp. pp. 162–186 ••–•• doi: 10.1111/jsbm.12136 Managerial Perceptions of SMEs in the Wood Industry Supply Chain on Corporate Responsibility and Competitive Advantage: Evidence from China and Finland by Ning Li, Anne Toppinen, and Marja Lantta

Our comparative study investigates the Chinese and Finnish managers’ perceptions of corporate responsibility (CR) and how CR engagement can benefit small and medium-sized enterprises (SMEs) in the mechanical wood processing industry. Based on the semi-structured interviews with 23 line managers between December 2009 and December 2010, our study provides evidence that with a focused stakeholder approach, managers in wood industry SMEs adopt informal corporate social responsibility strategies and tools to meet the specific expectations of their key stakeholders. Legal aspects were the key drivers of CR, whereas advanced production technology, efficient wood procurement, customer focus, and a competent workforce formed the four principal cornerstones in their pursuit of competitive advantage. Despite the two vastly different cultural and operational settings, the importance and the role of CR were found to follow a fairly similar logic in how CR was understood and implemented in the wood industry SMEs seeking for competitive advantage.

definition of CR as it tends to assume different Introduction meaning depending on the context within Corporate responsibility (CR) has become which it is used or applied, a wide range of part of modern business in today’s globalized definitions therefore exist, which reflects the economy (Donaldson 2005) as “business corpo- different perspectives and priorities of the rations have an obligation to work for social stakeholders involved in the CR debate betterment” (Frederick 2006, p. 37). Overlap- (Aupperle 1984; Bansal, Maurer, and Slawinski ping with concepts such as business ethics, 2008; Dahlsrud 2008; Moir 2001; van Marrewijk corporate philanthropy, and social and envi- 2003). ronmental sustainability, CR is a “dynamic and The first Green Paper of the European contestable concept that is embedded in each Union, CR is defined as “a concept whereby social, political, economic, and institutional companies integrate social and environmental context” (Matten and Moon 2004). Given concerns in their business operations and in the absence of a specific cast-in-the-stone their interaction with their stakeholders on a

Ning Li is postdoctoral researcher at the Department of Forest Sciences at the Faculty of Agriculture and Forestry, University of Helsinki. Anne Toppinen is professor of Forest Products Marketing and Management at the Department of Forest Sciences at the Faculty of Agriculture and Forestry, University of Helsinki. Marja Lantta is junior researcher at the Department of Forest Sciences at the Faculty of Agriculture and Forestry, University of Helsinki. Address correspondence to: Ning Li, Department of Forest Sciences, Latokartanonkaari 7, University of Helsinki, P. O. Box 27, 00014 Helsingin Yliopisto, Finland. Tel.: +358 40 9384322. E-mail: cathyrine_lee@ hotmail.com.

162 JOURNALLI, OF TOPPINEN, SMALL BUSINESS AND LANTTA MANAGEMENT 1 voluntary basis” (European Union 2001, p. 6). Contributing over $US 130 billion of gross By integrating social and environmental con- value-added worldwide (Macqueen 2004; cerns in business operations and in interactions Mayers 2006), SMEs comprise 80–90 percent of with stakeholders, CR can be interpreted as an all forestry enterprises and provide 50 percent extra investment in human capital, the environ- or more of the forest-related employment in ment, and stakeholder relations (European many countries (Mayers 2006). Approximately Commission 2001; van Marrewijk 2003). A 20 million people in the world are employed by deeper sense of CR refers to all of a company’s forestry SMEs (Macqueen et al. 2006), and 8 impacts on society and the responsibility to million individuals are directly employed from deal with the impacts on each group of stake- small-scale forestry and wood processing enter- holders (e.g., employees, customers, local comprises alone (Macqueen and Mayers 2006). In munities, and the natural environment) across addition to their contribution to national all product stages from primary processing, economy, a number of shared features can be through use, to disposal, making CR synony- found among SMEs in the forestry sector and mous with sustainable development at the cor- nonforest SMEs (see, e.g., Luo et al. 2009), such porate level (Hopkins and Cowe 2003). Recent as having a strong impact on the environment, studies have shown that by approaching CR securing the livelihoods and well-being of with corporate priorities, strengths, and abili- people (in particular the rural poor living in or ties, firms can develop socially and environ- near forested landscapes), spreading wealth mentally responsible solutions to emerging CR locally, and enabling local innovation. They issues, which could provide operational and also face increasing unique and complex chal- competitive advantages (CAs) (Galbreath 2009; lenges, including great hikes in the prices of Litz 1996; O’Brien 2001; Porter and Kramer raw materials and energy, resource tenure, 2006). preservation of cultural identities and practices, Beside the profound impact of demand for access to finance, export rebate adjustments, global responsibility on large multinational and contradiction between sustainable devel- companies, small and medium-sized enter- opment and commercial profitability. prises (SMEs) are also inevitably under increas- We agree with other researchers that under- ing pressure via legislation, supply chain, trade standing the corporate awareness and motiva- association, and consumer demands for sus- tions for CR engagement will contribute tainable business conduct. Extant literature CR positively to the developing of relevant and have generally focused on large firms with the efficient CR practices, programs, and tools primary thrust of explaining the institutional- (Spence and Rutherford 2003). In this paper, ization of formal policies and the manner in we argue that though CR research in the for- which CR is incorporated into decision-making estry sector is permeated by a focus on large and business practices. Although SMEs account pulp and papermaking companies (Ketola for the majority of business in many countries, 2009; Li et al. 2011; Mikkilä and Toppinen and they are often the most dynamic and long- 2008; Toppinen et al. 2012; Vidal, Bull, and standing players in local communities and of an Kozak 2010), there is a severe gap in our integral part of the value chain, understanding knowledge about the particular organizational of their awareness and motivation for CR is field in which wood industry SMEs are embed- insufficient. Many SMEs still perceive it as a ded, including their awareness and motivations fuzzy concept, lacking clarity in terms of defi- to engage in CR practices. Having a different nition, execution, and the potential benefits point of departure than that of large firms for embraced. As SMEs are often run by their development of competitiveness and business owners who tend to have a personalized style case interests, SMEs need to develop their own of management and lack of formal management ways and logic to integrate CR into their strat- structure with specialized personnel (Bolton egies and operations for such arguments. Firms 1971), their approaches to CR are particularly are embedded in national systems and institu- endogenous, being inherently shaped by the tions, which shape managerial belief systems, psychological characteristics of owner– firm stakeholder response, and overall orienta- managers and a range of contexts, such as tions to social issues. There is a need for spe- cultural difference and values, stakeholder cific knowledge of how such national, local structure, economic development, or strategic institutional contexts affect the way in which cognition (Antal and Sobczak 2007). CR is institutionalized within SMEs. In this

2 JOURNALLI, OF TOPPINEN, SMALL BUSINESS AND LANTTA MANAGEMENT 163 respect, we wish to investigate whether and that integrates CR with six fundamental and how different types of institutional environ- interconnected dimensions of corporate strat- ments influence the views of SME managers of egy: company mission, identification of strate- CR in different context. gic issues, markets, customer needs, resources, Motivated by the above-mentioned con- and CA. When designing strategic CR initia- cerns and the limited understanding of CR in tives, the firm’s interest should be rooted in its SMEs, our comparative study aims to investi- potential contribution to the value chain of gate wood industry SME managers’ awareness the firm, with regard to the competitive envi- and motivation toward CR and whether CR ronment and the possibility of creating new can become a contributor of CA for wood business opportunities (Bhattacharyya et al. industry SMEs in two selected countries. 2008). Based on the resource-based view (RBV) The RBV, on the other hand, contends that theory of the firm and stakeholder theory, the resources are not evenly distributed across all strategic thinking to CR conceptualized by firms, and these resources may not be perfectly Galbreath and Galvin (2008) and Galbreath mobile (Barney 1991). Strategic resources of (2009), and the EU definition of CR (2001), the firm include assets, capabilities, attributes, we aim to investigate SME managers’ under- processes, and knowledge controlled and uti- standing of CR, their cognitive construct of lized by the firm to implement strategies CA, and their perceptions of whether CR can (Barney 1991; Galbreath 2005; Galbreath and be a generator of CA in the wood industry. Galvin 2008). Environmental performance con- Our research questions are thus formulated as stitutes a source of CA (Russo and Fouts 1997). the following: (1) how do wood industry SME Relying upon significant employee involve- managers from China and Finland perceive CR ment, firm-wide coordination, and a forward- and how the EU’s definition of CR is adopted thinking managerial style (Shrivastava 1995), by wood industry SMEs? and (2) how do environmental initiatives may help firms to wood industry SME managers from China and develop new competences, resources, and Finland perceive CR’s contribution to competi- capabilities that are manifested in tangible tive advantage? assets like organizational culture, technology, Our empirical data were collected by semi- structure, and human resources (Orlitzky, structured in-depth interviews between Decem- Schmidt, and Rynes 2003; Russo and Fouts ber 2009 and December 2010 with a total of 23 1997) and to differentiate through process and managers from eight privately owned mechani- product innovation (McWilliams and Siegel cal wood processing SMEs in China and 2000). Positive investments in the early phase Finland. The two countries were chosen based of organizational life cycle can contribute to the on the assertion that the proliferation of CR is firm’s long-term CA. In order to benefit through dependent upon resolution of a myriad of orga- increased innovativeness with a market orien- nizational, economic, political, and legal issues; tation strategy, managers need to efficiently China and Finland are two dominant players in leverage their environmental commitment in the global forest products market, representing resource allocation (Dibrell, Craig, and Hansen distinctive and interesting characteristics in 2011). Resources associated with raw material, terms of institutional and cultural settings. reputation and service, collaboration, and technological know-how also had a positive influ- Literature Review ence on the financial performance of the firm Corporate Responsibility, (Lähtinen et al. 2009). Resource-Based View, and Building on the RBV, Galbreath and Galvin Competitive Advantage (2008) and Galbreath (2009) argued that the There are ample indications that CR cannot firm should consider whether different aspects be separated from corporate strategic manage- and dimensions of CR can constitute a strategic ment. Managerial perception of competition is issue, such as having the potential to affect the regarded as a key determinant of the strategic ability of the firm to achieve its objectives. plans they make and the subsequent outcomes Therefore, expectations by multi-stakeholder (Andrews 1971). As strategic thinking about CR groups on the economic, environmental, and has become increasingly necessary in develop- social performance of the firm can be regarded ing CA in the current turbulent environment, as CR issues; and with scarce resources, the Galbreath (2009) demonstrates a framework firm must identify and acknowledge those CR

164 JOURNALLI, OF TOPPINEN, SMALL BUSINESS AND LANTTA MANAGEMENT 3 issues that are most imperative, and screen and sustainability management does not mean that analyze those strategic CR issues improving or CR strategies are the right choice for all situa- impairing the firm’s ability to execute its tions but rather “. . . something that must be mission. carefully honed to the specific circumstances of Scholars and practitioners have tried in many individual companies operating in unique posi- different ways to seek justification—a sound tions within distinct industries. Successes in economic rationale—for sustainability strategies whole industries and at other companies are within organizations (e.g., Bonifant, Arnold, and useful examples but the case still has to be Long 1995; Elkington 1994; Hart 1995; Holliday, applied to one company at a time” (Reed 2001, Schmidheiny, and Watts 2002; Porter and van p. 4). Managers should be aware that “the busi- der Linde 1995; WBCSD 2002). Key elements of ness case for CR is powerful at an individual the business case for CR revealed in the existing company level and on specific issues . . . There literature include, for example, reputation, risk appear to be some issues where the business management, employee satisfaction, innova- case is either weak or non-existent, and other tion, access to capital, and competitive advan- where there can be no business case until tage. Nevertheless, there is still substantial market conditions changed” (Hopkins and skepticism and uncertainty (Morsing 2003; Cowe 2003, p. 24). Walley and Whitehead 1994) because of information type (opinion versus case studies), rel- Corporate Responsibility in Small and evance, applicability (specificity, which may Medium-Sized Enterprises preclude wider application to other organiza- Given the fact that standards vary from tions), source (business peers, nongovern- country to country, there is no universally mental organizations, or academicians), or accepted definition of SMEs, other than to say underlying attitudes of managers (Salzmann, that they are companies with metrics that fall Ionescu-Somers, and Steger 2005). below a certain threshold (e.g., Ayyagari, Beck, Despite the mixed results over the last and Demirgüç-Kunt 2003). The definition of an decades on the generic causality between finan- SME predictably differs between China and cial performance and environmental/social per- Finland.1 In any case, SMEs play an ever- formance, there is clear evidence of positive growing role in sustainable development not links (see, e.g., meta-analysis by Margolis and only for their significant contribution to the Walsh 2001, 2003), especially when consider- health of national and international economies ing the increased relevance in intangible assets (CCICED 2003; European Commission 2003) such as reputation, knowledge networks, orga- but also for their large share of total pollutants nizational learning, and employee recruitment (Hillary 2000; Su 2004). However, the CR and and retention (e.g., Brammer and Pavelin 2006; CA concepts remain somewhat vague and dif- Shore and Wayne 2003; Turban and Greening ficult for SMEs to enunciate even though they 1996; Whetten, Rands, and Godfrey 2002). Dif- may in fact conduct CR-related practices. In ferent types of CR-related activities have differ- order to succeed in the business context, the ent implications for financial performance concepts need to be translated into practice, (Griffin and Mahon 1997; Hillman and Keim and CR initiatives need to be screened and 2001; Lankoski 2009). Justification for integrated into operation (Jenkins 2006),

1As defined in European Union law, the main factors determining whether a company is an SME are number of employees (micro < 10, small < 50, medium-sized < 250) and either turnover (micro ≤€ 2 million, small ≤€10 million, medium-sized ≤€ 50 million) or balance sheet total (micro ≤ small ≤€ 10 million, medium-sized ≤€43 million) (Commission Recommendation 2003). In addition, the European Union specifies terms of ownership, stating that SMEs must be independent, with less than 25 percent being owned by outside interests (European Commission 2007). In China, the definition of an SME varies among different industries and much larger enterprises are categorized as SMEs in China. For administrative purposes, the Interim Regulations on Small and Medium Enterprises Categorizing Criteria defines SMEs based on three identifying criteria (National Bureau of Statistics of China 2003): payroll (small < 300, medium-sized 300– 2,000), total assets in million RMB (small < 40, medium-sized 40–400), and total revenue in million RMB (small < 40 small, medium-sized 30–300). There is no requirement on the independent ownership of a SME in China (NBSC 2003).

4 JOURNALLI, OF TOPPINEN, SMALL BUSINESS AND LANTTA MANAGEMENT 165 enabling SMEs to recognize and embrace the flexibility (Fiegenbaum and Karmani 1991) and intrinsic benefits of CR. innovation (Lefebvre and Lefebvre 1993). SMEs confront difficulty in defining and executing CR, as well as communicating their Divergences in Institutional best practices to both internal and external Environment: China versus Finland stakeholders (Jenkins 2006; Murillo and Lozano The development of CR is a result of its 2006; Spence and Lozano 2000). Large firms are institutional environment, including the politi- more willing to define and implement CR strat- cal and legal systems, history, culture, and reli- egies (Perrini, Russo, and Tencati 2007) with an gion of a nation or a region (Doh & Duay 2006; integrity strategy in which values are defined Lammers 2003). As our study is concerned with without control or sanction mechanisms SME managers’ perceptions of CR in China and (Graafland, Ven van de, and Stoffele 2003) and Finland, an understanding of national culture, with far more substantial volume and range of rival institutional systems, and CR regimes efforts (Sweeney 2007), whereas SMEs mainly enables us to understand the strategic rel- describe in a sense of ethics with specific con- evance of CR in the two different contexts. tributions to the local community (Sweeney Culture has been specially recognized as one 2007) and favor a dialogue strategy—learning of the determinants of ethical judgments in the from stakeholders to identify the most impor- models of ethical behavior (Bartels 1967; Hunt tant and imperative issues (Graafland, Ven van and Vitell 1986). Most empirical evidence de, and Stoffele 2003). However, pressure from exploring the effects of national culture on CR an upstream supply chain buyer could also is found to agree with the cultural mapping boost SMEs to act as a change agent toward CR proposed by Hofstede (1991). Various factors (Jorgensen and Knudsen 2006). influenced by culture, such as a society’s laws, SMEs’ managers regard CR more as an values, and religious beliefs, are key influences ethical and moral issue rather than a strategic in ethical decision-making (Bartels 1967). Orga- instrument, making formal CR communication nizational culture and the decision-maker’s past with their external stakeholders particularly experiences, as molded by national culture, challenging (Nielsen and Thomsen 2009). also have a significant impact on ethical judg- Fassin (2008) argued that formal reports and ments (Carroll and Gannon 1997). procedures could not validate or prove actual Corporate governance varies widely ethical behavior, and the absence of formal between countries and organizations, because reporting should not be equivalent to the countries influence the costs that firms incur to absence of responsible conduct in SMEs. There bind themselves to good governance and the are indications that owner–managers of SMEs accrued benefits of doing so. There is a large are capable of differentiating the various con- literature showing that a country’s governance cepts of CR and business ethics and simultane- system affects corporate governance decisions ously recognize the interrelationship and of the firm. For example, Allen and Gale (2000) interdependence of these concepts (Fassin, contrasted governance systems across coun- Rossem, and Buelens 2011). tries; John and Kedia (2003, 2004) theoretically In large firms, CR has been traditionally demonstrated how financial development and driven by the mid-level managers in human the quality of the monitoring technologies of a relations or external relations with the support country affect the choice of governance mecha- of the top management, whereas the owner– nisms. Cressy, Cumming, and Mallin (2010), on managers frequently play a dual role—as both the other hand, highlight the significant impli- the driver and the implementer of firm values cations of ownership structure (e.g., public in SMEs (Jenkins 2006). The line managers of listed versus entirely private) for CR determi- smaller firms appear to be a “silent” stake- nants and consequences. holder group (Mattila 2009), although their Table 1 summarizes the institutional and cul- close relationships within the social and busi- tural models and CR regimes in China and ness environment may make them potentially Finland. Finland represents the corporatist highly motivated in securing the survival and institutional model, a mixture of rule- and long-term success of their firms. Beyond hard principle-based societies that prevail in north- economics, the uniqueness and characteristics ern continental Europe (van Tulder and van der of SMEs are inherently more socially beneficial Zwart 2005). The understanding and definition than those of larger firms in terms of output on CR goes beyond the European reference

166 JOURNALLI, OF TOPPINEN, SMALL BUSINESS AND LANTTA MANAGEMENT 5 Table 1 Institutional and Cultural Models and CSR Regimes in China and Finland (Compiled Based on Hofstede 1991; van Tulder and van der Zwart 2005)

China Finland

Rival Institutional Business-statist Corporatist/soc-democratic Model

Interaction Principle Complementarity and Subsidiarity pragmatism Characterization Co-optation/Guanxi Cooperation/corporatism Contracts State-rule-based and Mixture of principle- and rule-based relationship-based Orientation Long-term Medium-term Religious Spiritual Confucianism; Taoism; Protestant Origins Shintoism; Buddhism Profit Sector Medium–large Medium Nonprofit Sector Small Medium Family Controlled Medium–high Medium–high Cooperatives Low High Power Distance Large Small–medium Individualism Low Medium Uncertainty Weak–medium Weak–medium Avoidance Closed/Open Semi-closed Semi-open CSR Regime Corporate-statist (CR Asia) Neo-corporatist (CR Europe) Weak and inactive approach Vary from re-active, active, to Moderately closed: pro/interactive partnering/endorsing, Open: facilitating/partnering; producer oriented employee/consumer/producer Common/communist law; oriented; informal codes; pragmatic Civil law; voluntary and regulation principle oriented codes and reporting; CR regulations are primarily Environmental and labor screens developed in the area of feature prominently; environmental protection, National and regional CSR regimes are particularly the area that formed by stakeholders from all directly affects the three spheres of society (public, internationalization business and civil society) as a rule strategies aimed at High transparency; weak sanctions; developed markets precautionary principle; weak Low transparency; no corporate volunteering and corporate volunteering and philanthropy tradition; philanthropy tradition Large firms are actively encouraged to Often exhibit an inactive CSR do more than what is legally and international CR required (e.g., investment in human orientation on labor and capital, environmental protection, human rights and working sustainability, and stakeholder conditions relations). Very low public advocacy of Moderate-high public advocacy of CSR; negligent SRI CSR; weak SRI.

6 JOURNALLI, OF TOPPINEN, SMALL BUSINESS AND LANTTA MANAGEMENT 167 (European Commission 2001), and 83 percent tions are “largely declaratory and their impact of Finnish SMEs are engaged in CR remain unclear” (Chauhoud 2011, p. 173). (Observatory of European SMEs 2002). SMEs in China are relatively small in size, However, the CR concept is rather vague to the and their production capability and product Finnish small business owner–managers, and it structure are usually unilateral. They often lack is associated with the responsibilities defined transparency in terms of their business perfor- by the law and enforced by the authorities. mance and have weak corporate governance Such vagueness explains why the entrepre- and management skills, primarily because of neurs were reluctant to increase their social their dependence on the model of family- responsibilities above the existing level oriented management (All-China Federation of (Lähdesmäki 2005a). In general, Finnish small Industry and Commerce 2002; International business owner–managers believe that respon- Finance Corporation 2006). Previous studies sible behavior legitimises the firm’s existence have shown that the requirement from custom- and trustful stakeholder relations would be ers, government regulations, and corporate economically profitable (Lähdesmäki 2005a), environmental awareness are the most impor- whereas moral responsibility tends to be more tant factors in CR uptake (Peng and Ren 2003), extensive than responsibility toward customers the exterior barriers of policy and financial and employees, and charity is utilitarian for barriers should be stressed rather than the local well-being and improving firm image, and internal technical and management barriers, the government would be the least important and absence of incentives on economic poli- stakeholder in an ethical sense (Lähdesmäki cies, tax enforcement of environmental regula- 2005b). Lacking the exact knowledge of the tions, and high initial capital cost were the most content and applicability of the concept of CR important hindering factors in adopting clean (Heikinheimo 2005; Lähdesmäki 2005a), technologies (Peng et al. 2005; Shi et al. 2008). Finnish SMEs find it difficult to identify which functions are responsible conduct, making CR Methodology benefits distant to the majority of entrepreneurs Interview has been broadly used to access (Taipalinen and Toivio 2004), revealing a need people’s experiences, their inner perceptions, to develop the communication of CR (Huju attitudes, and feelings of reality in social and 2005), as well as bring up peer networking organizational research (Crewsell and Plano with competitiveness orientation. Clark 2011; Yin 2009). Case study method is Remaining an avowedly socialistic regime, suitable in where the topic is an uncharted China represents a society that the business- territory. Although cases cannot be generalized statist institutional model and corporate-statist to a large population, they are generalizable to CR regime prevail (van Tulder and van der Zwart theory and can provide an effective research 2005). CR was introduced to China in early methodology that permits the examination of 1990s because of the resulting rise in trade, an extensive body of evidence (Yin 1989). As increased demand for ethics programs, and our aim was to investigate managers’ percep- social responsibility audits by global buyers, tions of CR in China and Finland, a qualitative particularly accompanied with China’s acces- inquiry using semi-structured interviews and sion to the World Trade Organization (WTO). case study was considered justified for our Determined by current social development and purpose. It was also believed that evidence socialist system with Chinese characteristics from a multiple-case study could be more com- (Shen and Cheng 2009), CR activities in China pelling and robust (Yin 2009) to enhance are specified by four sources of laws and regu- understanding of the research phenomenon. lations, including (1) the Company Law, enacted Purposive sampling was used in this study, in 2006; (2) the Labour Contract Law in 2008; (3) allowing the researchers to select a case that the Instructions for CR in State-owned Enter- illustrate a feature or a process of interest and prises issued in 2008 by the State-Owned Assets to critical consideration on the parameters of Supervision and Administration Commission of the population (Silverman 2000). Companies the State Council; and (4) Guidelines for CR were to be selected based on two criteria: first, compliance for Foreign-Invested Enterprises their product portfolios and business models issued in 2008 by the Chinese Academy of are nonidentical; second, they should have International Trade and Economic Cooperation more or less publicized their interests through (Chauhoud 2011). However, these CR regula- external communication (e.g., company

168 JOURNALLI, OF TOPPINEN, SMALL BUSINESS AND LANTTA MANAGEMENT 7 website or other forms of corporate disclosure). ing on the relevant knowledge and interest of The latter criterion was based on the need to individual interviewees. Immediately after each identify and promote role models in CR prac- interview, the annotations were written out into tice and the presumption that the number of complete sentences, and clarifying comments companies engaging in CR is expected to grow and notes were also added to the text material in the future. Given the large number of SMEs by the interviewers wherever needed, allowing in China, and there we chose to focus on for and anticipating valuable inputs spontane- Guangxi province, which is one of the major ously introduced by the informants. The mana- bases of fast growing Eucalyptus plantations gerial perceptions of CR and its connection and wood products production in China. with CA are summarized in Table 2. A number of candidate companies were To ensure the overall quality of the study, identified and informed of the study, and inter- the researchers applied those tactics (Crewsell views were requested. The final sample con- 2009; Gibbs 2007; Yin 2009) for case study sists of four medium-sized Finnish companies, throughout the subsequent conduct of the three medium-sized and one small-sized study. In addition to the English language skills Chinese companies. All these eight case com- of the individual researchers, an interceding panies are privately owned. Three managers agreement was clearly predefined in advance from each company were requested interviews. between the researchers, including the research One Finnish manager could not participate in objective, the themed interview questionnaire, the interview as planned because of an unex- the awareness of the perceived cultural biases pected event. of the answers (both in Finn-to-Finn and As already mentioned, the European Union’s Chinese-to-Chinese interviews), the awareness definition of CR in the first Green Paper (2001) of the perceived effects of language barriers in was adopted in our study. Although China has explaining the questions during the interviews basically formed a legal system appropriate for and in interpreting the answers (Finnish to CR, a clear CR policy system has not yet been English and Chinese to English), the methods established. To ensure equal standpoint in for data collection and analysis, the use of research design, the questionnaires (see the logical patterns in interpreting the transcripts, Appendix) were delivered to the Chinese man- as well as the methods for establishing chain agers four weeks prior to the interviews. They evidence and addressing rival explanations. were asked to give a brief description of CR at Replication logic was used to ensure the exter- the beginning of the interviews to make sure nal validity of our multiple case studies. Mul- whether they have a satisfactory understanding tiple sources and cross-checking in data of CR. Similar procedures were carried out for collection by different coders were used to the Finnish interviewees. establish chain of evidence and thus construct Our empirical data were collected between and reliability. Internal validity was achieved December 2009 and December 2010. Two through pattern matching in the interpretation interview methods were chosen based on the and explanation of the findings. Moreover, convenience of the interviewee, time-saving, selected authentic extracts of primary data of and cost-effective purposes. Of the 23 semi- the interviewees were also presented for the structured interviews, 12 face-to-face audiotape reader to determine the consistency of the find- interviews were performed in China by the first ings and the accuracy of the interpretation. author who is a Chinese native speaker; 11 Table 3 illustrates the theoretical framework telephone-researcher interviews with the of our study, forming the basis of our semi- Finnish managers were conducted in Finland structured thematic interview questionnaire. As and transcribed by the second and the third the operationalization of the conceptual frame- authors, who are Finnish native speakers. works of Galbreath and Galvin (2008) and The unit of analysis was an individual infor- Galbreath (2009), it illustrates the allocating of mant. The interviewees were line managers actions and resources into specific activities responsible for specific and various functions and outputs by retrieving European Commis- of their firms, including wood procurement, sion’s definition (2001) for CR. As an effort to production, sales, marketing, and overseas explore areas in where the CR perspective operations. The participants were assured of could be integrated in the firm, the CR activities the confidentiality of their responses. The inter- of a forest industry SME presented in Table 3 views varied from 30 to 100 minutes, depend- are not a comprehensive list containing all

8 JOURNALLI, OF TOPPINEN, SMALL BUSINESS AND LANTTA MANAGEMENT 169 Table 2 Summary of Managers’ Perceptions of CR and Its Connections with CA

China Finland

Wood Procurement Constitution of CR: Constitution of CR: —Obeying the law* —Compliance with the law* —Making a profit & paying taxes* —Making a profit for owners & —Good product quality & service & paying taxes* good price* —Customer-orientation* —Reliable to local small forest —Injecting millions into regional owners economy —Engagement in community —Payment to employees & development contractors CR as a source of CA: —Participation in projects with —Good reputation, trustworthiness schools & reliability CR as a source of CA: —No demand for certified wood —Industrial customers cannot be from customers won over with imagery —No certified wood in procurement —No marketing potential; only if & production wood origin is required —Uncertain influence of FC or CoC on consumer buying Production Constitution of CR: Constitution of CR: —Dependable to business partner* —Local wood procurement & —Promotion regardless age or employment gender CR as a source of CA: CR as a source of CA: —CR is important but often less —Sustained employee motivation visible to outsiders —Continuous employee training & —PEFC certificate as a barrier in loyal employees export markets —First-mover in innovative HRM —Driving forces of consumer buying not always unclear Sales/Marketing/Export/ Constitution of CR: Constitution of CR: Other Functions —Fair & reliable to employees & —Provision of employment customers* —Environmentally friendly practices —Stakeholder-oriented —Investment in regional economy —People-centered & equitable CR as a source of CA: —Economic & social contributions —Possible marketing potential from to community ecological aspects CR as a source of CA: —CoC certificate necessary—but not —Reputation built on consistent aCA performance —CoC-certificate only important in —Trustworthiness, reliability & export markets win–win —Employee welfare is a concern of —Differentiation via normalization foreign buyers —Never forget “where one’s —A watered-down concept with no happiness comes from” potential of CA; cannot be —No demand for FC or CoC; no converted into monetary value potential for CA —CR can only be practiced locally —Stabilized employee retention

*The issues are shared by the informants from the same categories in terms of work function and country.

170 JOURNALLI, OF TOPPINEN, SMALL BUSINESS AND LANTTA MANAGEMENT 9 Table 3 Categorization of CR Activities of Wood Industry SMEs

Internal operations Internal interactions

Production and end-products: efficient Human resources: job security; ongoing training use of round-wood and by-products, and education; performance appraisal; energy efficiency, high quality of acknowledgment and compensation for specific products; customer value for money; skills and know-how; voluntary health and patents, copyrights or trademarks for safety investments environmentally friendly products Organizational culture and communications: flat Processes: quality and environmental management structure, for example, easy management systems; organizational access to management for all employees; routines that consider the effects of efficient daily operational communications; a activities from economic, social and clear business strategy that is communicated environmental perspectives, R&D of throughout the organization; reporting of environmentally advanced processes overall performance to all employees Economic responsibility toward owners: safeguarding a good rate of long-term return to shareholders, for example, by improving productivity and profitability

External operations External interactions

Wood procurement: round-wood Customers: active maintenance of customer dialog; acquisition from certified forests, rapid response to concerns; striving for long-term procurement from the vicinity of the relationships and collaboration; truthful mill, maintaining chain-of-custody of promotion; verbalized mission timber Forest owners: long-term relationships as a result of Supply chain: locality and longevity of collaboration and trust contracts; incorporation of economic, Local community: local recruitment; provision of environmental, and social elements in income for local forest owners and contractors; supplier selection criteria; compliance corporate philanthropy; direct involvement in with upper supply chain requirements community projects and affairs, for example, regarding environmental and social collaboration with local schools or hospital performance; long-term alliances and Society at large: monitoring of societal trends (e.g., collaboration with industrial customers media, industry publications, market research); communication and collaboration with NGOs, educational and research institutions, media; disclosure of overall performance possible actions and outputs of the firm. The Within the four main categories, the CR main stakeholder division and key business activities are arranged under a few fairly broad practices suggested by Spiller (2000) and the themes, which constitute the frame for the CR activities of SMEs identified by Jenkins themed interviews of sawmill managers in (2006) have been utilized whenever applicable. this study (see the Appendix). For instance, the To connect CR with the RBV of the firm, internal operations category has three major the resource-pool of sawmill introduced by themes: production and end products, pro- Lähtinen (2009) and the resource items catego- cesses, and economic responsibility. According rized by Galbreath and Galvin (2008) have to Sawhney, Wolcott, and Arroniz (2006), been adapted for the purposes of this study. processes are the configurations of business

10 JOURNALLI, OF TOPPINEN, SMALL BUSINESS AND LANTTA MANAGEMENT 171 activities a firm employs to manage internal market segmentation. Driven by their normal- operations. Production and technical efficien- ization over market competition, medium-sized cies are treated as a separate theme because of companies were more technology and R&D their obvious importance in the wood products oriented. The following quotes were from the manufacturing sector (Salehirad and Sowlati Chinese managers: 2006). The economic aspects of business, on the other hand, originate from the very core of 90% of our raw materials are business and are generally pursued by busi- eucalyptus—no secret about production nesses, as “the social responsibility of business —not much money for new machine is to increase its profits” (Friedman 1970). or technology...Reliability and better Although financial accountability cannot be services make us better off [than our regarded as an individual operation, nor is it competitors]. restricted to the internal domain, the economic Ensured delivery, stable price, and aspects are categorized in the context of inter- quality are important [sources of CA] to nal operations in our study. Interactions in this us...we want lasting businesses and study entail the activities directly related to frequent and transparent communication people: skills and competencies, communica- are very important for win-win. tions, stakeholder relationships, etc.; all other activities and outputs are considered opera- Despite the variation in means and methods, tions. Obviously, there are overlaps and efficient use of raw materials and efficient pro- interconnectedness between the categories. curement are the predominant methods for cost- The categories are further divided into internal cutting. Given the high raw material cost, strict and external dimensions, that is, the activities environmental regulations and norms, and the taking place at the mill and those taking place advanced technology available to Finnish saw- outside it. This division provides a clear and mills, the Finnish managers interviewed tended straightforward way of categorizing different to integrate a customer-focused strategy with a activities; it will be further utilized in the col- strong technology orientation. Such an lection and analysis of empirical data. approach includes, for example, stacking raw In the following section, instead of four cat- materials and the by-products of sawmilling egories described in the theoretical framework, (e.g., wood chips, sawdust, and bark) near-to- findings of wood SMEs’ CR practices are cat- hand, and upgrading in timber processing egorized as internal and external. The to ensure efficient use of raw materials and to responses of the interviewees are also pre- enhance customer satisfaction on agreed-to sented based on their content and relevance to quality. In addition, the Finnish managers inter- the themes given to maintain clarity. viewed highlighted the competence and skills in pricing and healthy relationships with domestic Results forest owners as a source of CA. The Chinese Internal Operations and managers in our study, on the other hand, Internal Interactions unanimously prioritized efficient wood procure- All the interviewees considered production ment as a source of CA and paid less attention to technology as a source of CA. The Finnish efficient use of round-wood and its by-products managers emphasized advanced and optimized in their firms. They tended to secure their round- production technology as the single contributor wood supply through local forest owners, who to their profitability. Half of the Chinese infor- are convinced to receive better-than-market mants linked production technology to better prices, immediate and full payment. Such a performance and branding, and the others gave divergence could be explained by the differ- more weight to customer services such as ences existing in the two nations as discussed in assured delivery, stable pricing, guaranteed the literature review section. quality, and active communication. Such varia- None of the respondents possessed environ- tions could be explained by the resilience of mental or quality certification like the ISO9001 customer-oriented business models applied by or the ISO14001 during the interviews. The the companies. For example, customer services high cost and the weak legal and/or customer with greater flexibility seemed to be more demand mentioned by the interviewees help important for smaller sawmills because of their explain the reluctance of SMEs in acquiring limited capital and narrower product lines and such certification for their current business.

172 JOURNALLI, OF TOPPINEN, SMALL BUSINESS AND LANTTA MANAGEMENT 11 However, the current absence of environmental training takes time and money, so we certification does not validate or approve that want both our employees and their these SMEs are not aware of such standards. knowledge and expertise stay. Job safety is a MUST...when people At one time customers seemed to ask for stay healthy at work, the machines run, them, but I really don’t think we’ve lost therefore the business continues. a single customer because of not having them . . . they are really expensive for In terms of organizational structure and com- smaller [Finnish] firms like us . . . but we munication, the respondents used “simple,” can set them up quickly, should it turn “explicit,” and “flexible” to highlight the strength out to be necessary. that “prevents obstruction in communication” To [any Chinese SMEs like] us, having and “ensures that information flows freely.” such certificates make no difference but They believed that the ownership structure in cost a lot . . . neither required by the law their firms strengthens “the firm’s face to be nor by our customers. Obey the law is recognizable both on the inside and on the everything... outside of the firm” and ensures that decision- making is swift, flexible, and unanimous. The How to secure the firm’s longevity was interviewees raised their concern over the wide- invariably discussed by the interviewees. The spread use of oral communication in SMEs, Finnish managers were most concerned about arguing that such “unsystematic filing” would how their firms can survive under the pressure hinder organizational learning and growth. of market oversupply and heavy financial burdens and emphasized the importance of [Heavy] dependence on oral communica- profit-making through optimized production tion [in execution of actual instructions] technology, round-wood, and shipping costs. makes knowledge management impos- Their Chinese counterparts, on the other hand, sible . . . We need to file what we do and focused on profit maximization and growth how we do, but no copy & paste of big through low-costs in raw material acquisition, firms’ way of doing things. labor, logistics, favorable export policies, as well as networking with the fast-growing External Operations and Interactions domestic construction and real estate sectors. In terms of external operations, secured In terms of internal interactions, the infor- access to round-wood was considered as a mants in our study acknowledged committed major source of CA by the both Chinese and workforce as a source of CA and the benefits of Finnish managers interviewed. low turnover such as consistent quality and skilled employees. Employees’ knowledge and No roundwood, no sawmill...[Wood] competence were most emphasized by the procurement is never an easy job. Finnish interviewees because of the high auto- mation and the decreasing need for manual The interviewees were asked about their labor in their firms. The Chinese managers perceptions of forest certification, chain-of- interviewed were most concerned about custody (CoC), and the significance of such employee retention in their companies. Like in certification to their business. Some Finnish many other regions in China, the wood pro- managers enunciated their doubt on the impact cessing industry is still labor intensive in of wood of origin: Guangxi province. The salary levels of the industry are less attractive than other export- I really don’t think . . . I mean I doubt oriented manufacturing industries such as that a European customer will buy apparel and electronics. Long-term and stable [wood] furniture [merely] based on a employment, open dialogue and caring, and stamp . . . you really can’t tell what kind enhanced safety at work have been used to of timber are the Asian products made retain employees. of, can you?

Labour is relatively cheap here [in China] In contrast, other Finnish informants in and job-hopping is very common... our study regarded the CoC certiﬁcate as a

12 JOURNALLI, OF TOPPINEN, SMALL BUSINESS AND LANTTA MANAGEMENT 173 necessity that “just has to be there by default” wood acquisition from the vicinity of the mill and believed that the absence of such certifi- and considered healthy relationships with local cate would constitute a problem when dealing forest owners as a source of CA. However, the with buyers in Central Europe and in the Finnish informants commented on the fact that United Kingdom. More specifically, the Finnish Finnish forest holdings are fairly small in the informants addressed their frustration over end (e.g., 20–30 hectares with a 40-year own- forest certification and the reliability of some ership on average), and duplicate and/or fre- individual certificates: quent transactions are not likely to take place. Along with the current “atmosphere of suspi- This is not an issue here in the northern cion” among the forest owners, the result of an coniferous forest belt...itmaybein earlier price cartel formed by the three major Russia...butit doesn’t seem to keep Finnish forest companies in Finland (Stora them from selling timber—they just Enso, UPM-Kymmene, and Metsäliitto Coopera- lower the price as they wish . . . and if tive), the Finnish interviewees noted that local IKEA requires a certificate of wood of sawmills may benefit from their long-standing origin, then anyone holding a pen can presence in the local community. Continuous write one for it. work is thus required to sustain the relationships with those small forest owners, and “the Most of the Chinese managers interviewed firm must keep its words and avoid creating felt suspicious of the market potential of certi- bad blood.” fied wood products, arguing that “many are not The Chinese interviewees noted that the ready to go for” because of the weak industrial current policies (e.g., investment subsidies, demand, limited consumer ability and willing- plantations, and wood acquisition) are in favor ness to pay for the price premium, and the lack of large firms and those major foreign inves- of effective market-based mechanisms in China. tors, who have influence on pricing and local In their opinions, having forest certification government intervention. or a CoC certificate does not seem beneficial for their firms, and therefore not a source Competition [for wood] get more inten- of CA. sified, especially when Stora Enso or In terms of supply chain management, the APP are injected. Fair pricing and reli- Finnish interviewees discussed the signifi- able in payment are the most important cance of business partners at a general level issues [for us] when buying round-wood and focused more on the efficiency of logis- from local forest owners...It’s a small tics. They considered trust as the basis of a community—bad news always spread partnership in the supply chain, where the faster than good news. longevity of their firm is fostered by a business alliance of proven and reliable partners. The Finnish interviewees underlined that Being geographically remote from the Euro- their firms aim at building lasting relationships pean markets, Finnish companies face the with their customers based on mutual trust and challenge of efficient logistics in terms of collaboration and tend to pursue CA through price–quality ratio, transportation mode, and optimized production and technology aided by environmental impact. The Chinese managers other tailored services for their industrial interviewed acknowledged the unique geo- buyers. “Green thinking” seems more transpar- graphical advantages that their companies ent in the marketing of the end products than have. For example, all the Chinese sawmills in in the preceding stages of the supply chain, our study are located in one of the most especially when selling timber to those indus- intensive eucalyptus plantation areas in south- trial customers in the United Kingdom. ern China. The area is also advantageous of However, the actual importance of “the green- diverse intermodal transport modes within an ness” to final customers was doubted. Without ideal distance. All these factors are significant mentioning green products or green marketing, contributors of cost saving in wood procure- the Chinese interviewees noted that their rela- ment and logistics. tionships with customers are based on reputa- In terms of external interaction, both tion, trust, and performance. Consistent and Chinese and Finnish interviewees strongly reliable performance (e.g., quality, delivery, emphasized the dependence of sawmilling on responsiveness, open communication, and

174 JOURNALLI, OF TOPPINEN, SMALL BUSINESS AND LANTTA MANAGEMENT 13 flexibility in conflict situations) are key to the Sawmill Managers’ Perceptions of success of smaller sawmills. Corporate Responsibility and The Finnish interviewees described their Competitive Advantage firms’ contributions to the local and regional Table 2 summarizes the managers’ percep- development as “a responsibility that has to be tions of CR and its connection with CA based recognized by the firm” and believed that their on three categories: wood procurement, pro- economic input manifests itself through direct duction, and sales/marketing/other functions. employment and the income generated for the A vast majority of the managers interviewed forest owners and the supply chain. As such seemed to be familiar with the CR concept and relationships are “matter-of-fact and business- defined CR with more or less individual and like,” the companies are acknowledged for firm-specific emphasis. For example, a Finnish their continuous contribution and cooperation wood procurement manager defined CR as in energy production with the municipality and following: enjoy a good reputation “out of a sheer sense of duty if nothing else.” However, such relation- [CR is] to comply with legislation, to be ships are not entirely simple and problem-free. able to look into people’s eyes while One Finnish manager described the relation- walking down the street and, to maintain ship between his firm and the municipality as good relations with customers . . . our “an amoebic phenomenon,” indicating that the CR is to inject millions of euro into the local politicians were aloof in their relations regional economy—this is more signifi- with the firm for fear of “voter suspicions” on cant than just providing employment the nature of the relationship. Another Finnish and paying taxes. manager emphasized the absence of independence that is important for the firm to take Many Finnish managers interviewed felt that initiatives collaborating with the municipality. CR does not have potential in marketing com- By sponsoring sporting or local events, the munication as European consumers are aware Finnish sawmills practices corporate philan- of the differences between the Scandinavian thropy to a certain extent mainly for maintain- products and those of other countries. Only ing their local presence. However, such one Finnish manager acknowledged the impor- sponsorship “would be the first thing to cut” in tance of CR as a source of CA: a time of economic recession. The Chinese sawmills have relatively shorter business history, CR works at the personal level, too. ranging from 6 to 10 years. The Chinese man- Having the “passion for sustainability” is agers in our study seemed to be aware of their a source of enthusiasm and motivation in firms’ importance to and dependence on the coping with the growing responsible surrounding communities. Regardless of the consumerism. The challenge is how to amount of money, corporate philanthropy is promote our environmentally friendly mainly through sponsoring local events, finan- products. cial support to individual employee families, and donation for natural disasters (e.g., the The interviewees shared the view that CR is snowstorm in 2008 and the Wenchuan earth- “an image of responsibility” and could “hardly quake in 2008). be converted into currency.” Although CR has Corporate tax and compliance with the law been incorporated into corporate communica- were identified as two major forms of CR by all tion, it still remains “a watered-down concept” the managers interviewed. In terms of external to many. A Finnish manager pointed out that sources of CA, the Finnish interviewees his firm is restricted by its limited resources emphasized on accurate forecasting of the when implementing “real CR,” and the future of operating environment, investments in technol- CR should be linked to the contract with the ogy, appropriate customer selection, and effi- customer (in both domestic and overseas cient logistics with long-term partners. In markets). The Finnish interviewees described addition to legal responsibilities, the Chinese the CR practices of their firms through legal interviewees outlined the importance of having and environmental aspects and argued that good relationships with the local govern- “compliance with the law is something so self- ment and authorities to maintain corporate evident that the thought of non-compliance legitimacy. would not even occur to anyone.” The Chinese

14 JOURNALLI, OF TOPPINEN, SMALL BUSINESS AND LANTTA MANAGEMENT 175 interviewees prioritized being law-abiding and ties that go beyond legal compliance, whereas paying taxes, making a profit and providing others remain skeptical with such a “voluntary jobs as the most important obligations over only” definition. As a prime sample of volun- others in their current CR practices. Among the tary CR, the European Commission’s definition fewer interviewees in our study who held a for CR adopted in our study is particularly positive view of CR as a source of CA, one beneficial for classifying into broad categories Finnish production manager noted that CR is for the setting up of public policies. This defi- regarded as “an obvious way to do business” in nition allows for a wide scope of relationships his firm: with stakeholders, all of which may not be relevant for SMEs. In the context of SMEs, a CR is very important . . . perhaps we’re combination of market-driven and regulatory being too modest about the whole thing interventions would guarantee success in eradi- . . . I mean we’re honest and think of cating socially irresponsible behavior while these matters obviously . . . so perhaps it encouraging responsible activities by integrat- is not so self-evident when looking from ing the potential of either. the outside. From the RBV perspective, the managers of small- and medium-sized wood industry A Finnish export manager in our study noted companies in our study identified four main that specific consumer segments have been themes as the sources of CA: advanced produc- developed through CR practices in her firm in tion technology, raw materials acquisition, recent years, and CR is more apparent in some customer-focused products and services, and areas determined by their core business like the human resources within the firm. Advanced use of local raw material and employees. technology was the single contributor to CA, most emphasized by the Finnish and Chinese We buy round-wood and do our sawing interviewees of the medium-sized sawmills, locally—we are here because of the raw and the optimal utilization of raw material— material . . . The [same] business changes producing the optimal product from each completely when you go to Central log—is of growing interest in the sawmilling Europe. industry (Flodin, Oja, and Grönlund 2008). In regard to the perceived importance of Regarding the possible benefits of CR, the advanced technology, our findings are consis- Finnish managers interviewed argued that the tent with the results by Lähtinen et al. (2009). ecological perspective and country of origin of In the case of smaller Chinese sawmills in our Finnish timber should be further promoted in study, the benefits of advanced production marketing. Without mentioning the renewable technology were much less significant and aspects of wood products, the Chinese infor- applicable because of their current business mants in our study did not consider forest strategy, limited financial capital, more labor- certification (even environmental initiatives intensive production, narrower product catego- beyond environmental regulations) or “obeying ries and customer segmentation, etc., which in the law” as a potential source of CA. They turn has led to the formulation of more clearly preferred to incorporate the social aspects of service-oriented business models more suited CR through relationships with employees and to the resource structure of the company. industrial customers. Efficient wood procurement was the second principal source of CA identified by the The environmental impact of our busi- interviewees in our study. According to the ness is not yet concerned by the [local] interviewees, securing access to round-wood at government . . . we don’t have any mar- a reasonable cost is vital to the sawmill’s sur- keting costs, because our reputation is vival and success, as also found in Lähtinen built on word-of-mouth . . . Making a et al. (2009) for Finland and in Johansson profit and improving employee retention (2008) for Sweden, underlining the local nature and welfare are more imperative to us. and strong raw material orientation of the sawmilling industry. Discussion Customer focus, the third principal source There have been various attempts to define of CA acknowledged by the informants, CR, many of which limit the concept to activi- was based on the available round-wood and

176 JOURNALLI, OF TOPPINEN, SMALL BUSINESS AND LANTTA MANAGEMENT 15 technologies facilitating the delivery of the eficiary of CR. Firms do not provide only direct promised quality to customers. Long-term rela- employment and income for their own employ- tionships with industrial buyers were consid- ees, but also a source of income for local forest ered particularly valuable by all the informants; owners and supply chain partners. CR is dem- the demands from the upper supply chain are onstrated in their conduct of employees and in drivers for certain practices, for example, CoC the interactions between the firm and its direct certificates, statements regarding employee stakeholders. The strong connection between working conditions. Regarding stated customer SMEs and the local community has been also orientation, our findings are congruent with the well recognized in previous studies (Jenkins findings of Stendhal, Roos, and Hugosson 2009; Jorgensen and Knudsen 2006; (2007), who argued that industry and market Observatory of European SMEs 2002). changes were the most common driver of However, quite a number of interviewees in product development. our study, especially the Chinese managers, Finally, a competent workforce was the still equated CR with legal compliance and cor- fourth principal source of CA acknowledged by porate taxation, a perception that is often docu- the managers interviewed. The interviewees in mented in earlier studies on SMEs. However, as our study perceived retention of skilled and the definition used in our study maintains that committed employees an imperative in keeping CR constitutes the “voluntary” actions of com- know-how inside the firm and guaranteeing panies (European Commission 2001), what we customers with consistent quality. This finding interpreted as CR should reach beyond legal is congruent with what has been found regard- compliance. Regarding customer focus, several ing the innovativeness in the U.S. forest prod- managers mentioned that CR is executed in ucts industry based on organizational culture their firms through customer-orientation, and commitment to innovation as the two ante- because to them, delivering what is promised cedents of the perceived innovativeness of the to customers is in the core of CR. This percep- firm (Crespell and Hansen 2009). tion of the CR of SMEs is also evident in a By and large, the four principal sources of number of studies in other industry sectors CA commonly acknowledged by the Chinese (see, e.g., Jenkins 2009; Murillo and Lozano and Finnish sawmill managers in our study 2006). were found to be closely linked and inter- When the interviewees were asked to give twined. In their attempt to define CR and illus- their opinion on the potential of CR in the trate how it is being implemented in their pursuit of CA in the context of wood industry companies, the interviewees incorporated CR SMEs, the responses ranged from the extreme through their corporate relationships with three skeptics to those who recognized unharnessed main stakeholder groups: their employees, the marketing potential in the “greenness” and cer- surrounding community, and their customers. tified origin of wood products. As much as a Although most of the managers interviewed did decade ago, European SME managers from not use the word “stakeholder” in what they various industry sectors unanimously agreed said, this fairly clear stakeholder orientation is that engaging in voluntary environmental and also congruent with extant literature on CR in social activities can bring various benefits to the SMEs (see, e.g., Jenkins 2006). firm, including customer loyalty, community In terms of the firm’s responsibility to relationships, and satisfying customer demands employees, the interviewees in our study con- (Observatory of European SMEs 2002). A more sidered providing a workplace, income, and job recent study on SME middle managers similarly security as the key components. Translating noticed that the informants do not necessarily “Enlightened employee policies” (Perrini and consider CR to be a strategic tool, but rather an Minoja 2008) into the owner–managers’ efforts issue of ethics and morals (Nielsen and will continuously increase the level of Thomsen 2009). employee satisfaction by providing economic advantages and ongoing training or enhanced Conclusions value creation of the firm, and mediate the In summary, the findings of our present perceived positive reactions of these stakehold- study suggest that SMEs in the wood industry ers (Hammann, Habisch, and Pechlaner 2009). apparently do not have difficulty in identifying The interviewees considered the surround- various stakeholders as the generators of stra- ing community as a major objective and ben- tegic CR issues or in implementing CR. What

16 JOURNALLI, OF TOPPINEN, SMALL BUSINESS AND LANTTA MANAGEMENT 177 challenges managers of SMEs most seems to be Within the supply chain of the wood indus- their limited ability to recognize their own best try SMEs in our study, customers, employees, practices, which determine whether and to and local forest owners were identified as the what extent their companies can embrace the three most important stakeholder groups in our benefits brought by their own CR stewardship. study. “The lack of resources can result in dif- It is unlikely that the stakeholder expectations ferent administrative and management pro- on the overall performance of SMEs would cesses between large and small businesses, diminish in the future—quite the contrary. which can also be reflected in their abilities and Therefore, wood industry SMEs could and ways to deal with social and ethical issues in should explore the possibility of gaining first- business . . .” (Lähdesmäki 2012, p. 10). Each of mover advantage. At the very least, they are these groups has a major effect on the business advised to keep a keen eye on the develop- operation and success of wood industry SMEs ments in society and in the markets when and therefore represents a key direction of expectations regarding overall corporate per- value creation for wood industry SMEs. By formance are in question. carefully evaluating the companies may achieve While wood industry SMEs have not been competitive CA with a well-defined stakeholder previously a subject of research in the context orientation and develop new resources and of China and Finland, our comparative study capabilities in dealing with company stake- of the managerial perceptions brought up a holder matters. lot of new findings and enabled us to draw To strengthen and convert their relation- several conclusions. First, as flexibility and ships with stakeholders into strategic assets, vagueness in the extant CR literature is a docu- companies must seek to understand the strate- mented problem (Fassin 2008), CR still is gic importance of stakeholders and respect the determined by a wide range of endogenous values they deserve. Companies have tradition- factors applying to the firm. Perceiving CR as ally started their CR path by integrating social a strategic issue prompts us (and the manag- and environmental concerns into human ers) to recognize the presumed reality of SMEs resource management practices (Perrini 2006). that not all activities related to CR can be Along with the growing importance of intan- implemented, yet various aspects of the phe- gible assets for company success, the quality of nomenon are undiscovered or under discus- the workforce has become a critical source sion. Furthermore, categorizations of CR of competitiveness for companies, regardless of activities (see Table 3) provide a substantive the size of firm. Therefore, internal initiatives and comprehensive checklist, enabling SMEs can start with those issues that are to be to screen their internal and external environ- embedded in organizational processes or ments, to identify the most relevant CR issues systems. Such issues may emphasize training in the business, and to determine their optimal on a wide range of issues relevant to the level of CR strategy and action by incorporat- employees, such as occupational health and ing the changes in market and societal devel- safety, product safety, ethics, company policies opment. Our findings thus lend support to and procedures of equality of treatment, fair previous research that argued that, mainly wages, transparency in compensation, incen- driven by the need to secure their license to tives and reward for performance, flexible job operate in the communities, SMEs approach design, and career plan. By giving value and CR with different strategies and socially relevance to employees and their well-being responsible behavior (Russo and Tencati 2009) and stimulating participation, collaboration, and that there is a fairly clear stakeholder ori- and new knowledge creation and exchange, CR entation in SME CR practices (Jenkins 2006; has the potential to turn into operational and Russo and Tencati 2009). We also confirmed competitive benefits, such as creation, manage- that in case of wood industry SMEs in Finland ment, and transfer of knowledge, increased and China, management often practices CR productivity, identification of growth and inno- implicitly without recognizing or articulating it vation opportunities, and efficiency gains (Fassin 2008; Nielsen and Thomsen 2009; through reduced costs. Sweeney 2007), and active screening and com- Furthermore, in consistent with previous mitting only to those most critical CR issues research, the middle-managers in wood indus- are key to CA (Kurucz, Colbert, and Wheeler try SMEs in our study also revealed their famil- 2008). iarity with and consciousness of CR, but the

178 JOURNALLI, OF TOPPINEN, SMALL BUSINESS AND LANTTA MANAGEMENT 17 used CR instruments remain less valuable and and evaluate the performance of SMEs relevant applicable to them, given the SMEs’ unwilling- to them in a given context, and whether major ness to formalize their CR strategies into spe- differences in their expectations emerge in cific and explicit management systems for a comparison to those views expressed at a wide range of factors, such as limited financial managerial level. capital, excessive cost of acquiring formal man- Second, institutional settings affect CR and agement systems, the actual needs for such its implications for companies (McWilliams formalization, and specific stakeholder expec- et al., 2006). Legal analysis of the impact of CR tations (Graafland, Ven van de, and Stoffele has received growing attention from scholars in 2003; Spence and Lozano 2000). Finally, our recent years (see, e.g., Cumming and Johan findings also contribute to enhancing the 2007; Cumming et al. 2010; Johan and Najar understanding of mid-level managerial input, 2010). Because the primary focus of our which Bird and Waters (2002) discussed as “the present study was not corporate ownership, moral muteness of managers,” as managers are corporate governance, or any topics closely often reluctant to describe their actions in related to the financial markets in China and moral terms and prefer using their organiza- Finland, we have only discussed divergences in tional interests and economic objectives as the institutional environment in both countries drivers of their actions, which were in fact from a relatively general perspective. Further defined by moral honor standards. work on the linkage between country-level Under the recognition that CR is inherently legal conditions and CR is needed. shaped by culture, the findings of our cross- Third, investors have the potential to play a country case study between Finland (a small disproportionate role in influencing CR out- Nordic democracy) and China (a leading comes (see, e.g., Cumming et al. 2010; Johan emerging economy) may be also relevant to and Najar 2010). Sufficient financial services those decision-makers seeking best practices of and market access are crucial for SMEs’ sus- CR or aiming to promote a business case for CR tainable development. For example, although by underlining similarities and differences in the Chinese central government has taken con- how CR can be understood and implemented in siderable efforts (e.g., adjusting forestry taxes two different cultural settings. However, and eliminate fees for develop timber planta- though a number of factors other than culture tions and processing, developing direct financ- also influence ethical decision-making, varia- ing mechanisms such as Initial Public tions derived from culture may differ in extent Offerings in the stock market, issuing favor- (Carroll and Ramamoorthy 1995), and based on able indirect financing policies, and establish- our limited data and single-industry context, ing the Small and Medium Enterprise Board our results are inconclusive. Overall, our find- that provides funding for SMEs) to sustain and ings seem to be consistent with previous strengthen the existing capacities of SMEs in research on managerial ethical practices the forestry sector, local SMEs in China still acknowledged in the literature review section suffer from capital shortage because of lack of of this study. financial support (see, e.g., Luo et al. 2009). Nevertheless, a number of questions also Constrained by the objective and scope of the arise driven from both the contributions and present study, although the effect of the inves- the limitations of our present study, stressing tor on the outcome of CR is beyond the dis- the need for further investigations. First, as an cussion here, this topic is worth further exploratory study, we only examined the line thinking. managers’ perceptions of CR in a single wood Fourth, as our aim was to investigate man- industry setting with evidence from a small set agers’ perceptions of CR in China and Finland, of line managers against the full array of com- participant companies were chosen based on plexity of managerial perceptions of CR. More their nonidentical product portfolios and busi- research is thus needed to improve our knowl- ness models, and publicized corporate interests edge of interactions between the focal firm and through external communication to ensure an its most important stakeholder groups, identi- equal standpoint for investigation. It would fied in our present study as employees, custom- have been interesting to present contrasting ers, and the community. For example, it would cases by illustrating different rationales behind be worth investigating how the key stakeholder the Finnish and Chinese small and medium- groups in the manufacturing sector define CR sized sawmills.

18 JOURNALLI, OF TOPPINEN, SMALL BUSINESS AND LANTTA MANAGEMENT 179 Finally, but not least, a more in-depth focus social-responsibility/beyond-good- on the supply chain perspectives in a wider intentions-strategies-for-managing-your-csr- geographical context of the globalizing wood performance#.U-nD8BAvgrM. industry presents a quite promising area for Barney, J. (1991). “Firm Resources and Sus- further research on analyzing managerial per- tained Competitive Advantage,” Journal of ceptions and attitudes to CR toward the key Management 17(1), 99–120. stakeholder groups of SMEs, as spontaneously Bartels, R. (1967). “A Model for Ethics in Mar- identified by the focal company itself. Given keting,” Journal of Marketing 31, 20–36. the exclusive focus of the current CR agenda on Bhattacharyya, S. S., A. Sahay, A. P. Arora, and large companies, SMEs usually act reactively or A. Chaturvedi (2008). “A Toolkit for Design- indirectively as suppliers to large companies or ing Firm Level Strategic Corporate social as the beneficiaries of large companies’ philan- Responsibility (CSR) Initiatives,” Social thropic activities. Taking the increasing supply Responsibility Journal 4(3), 265–282. chain pressures and the nature of trading rela- Bird, F., and J. Waters (2002). “The Moral Mute- tionships between SMEs and their upstream ness of Managers,” in Ethical Issues in Busi- buyers, the ultimate focus should be on study- ness: A Philosophical Approach, 7th ed. Eds. ing how value can be created through respon- T. Donaldson, P. H. Werhane, and M. sible business practices and networks in a Cording. Upper Saddle River, NJ: Prentice- globalizing manufacturing industry. Hall, 302–314. Bolton, J. (1971). Report of the Committee of Acknowledgments Enquiry of Small Firms. London: HMSO. Financial support from Academy of Finland Bonifant, B. C., M. B. Arnold, and F. J. Long (Grant Number 127889) is gratefully acknowl- (1995). “Gaining Competitive Advantage edged. We would also like to express our grati- Through Environmental Investments,” Busi- tude to the Editor and two anonymous ness Horizons July–August, 37–47. reviewers of this journal for their insightful Brammer, S., and S. Pavelin (2006). “Corporate comments. All errors remain our own. Reputation and Social Performance: The Importance of Fit,” Journal of Management References Studies 43(3), 435–455. All-China Federation of Industry and Com- Carroll, S., and M. J. Gannon (1997). Ethical merce (2002). Report on SMEs in China. Dimensions of International Management. International Report, ACFIC (in Chinese). Thousand Oaks, CA: Sage. Allen, F., and D. Gale (2000). Comparing Carroll, S. J., and N. Ramamoorthy (1995). “Do Financial Systems. Cambridge, MA: MIT Business Students from Different Nations but Press. with the Same Business Educational Experi- Andrews, K. (1971). The Concept of Corporate ences Perceive Ethical Situations Differ- Strategy. Irwin, IL: Homewood. ently?,” Working Paper, College of Business Antal, A. B., and A. Sobczak (2007). “Corporate and Management, University of Maryland. Social Responsibility in France: A Mix of China Council for International Cooperation on National Traditions and International Influ- Environment and Sustainable Development ences,” Business and Society 46(1), 9–32. (CCICED) (2003). Recommendations of the Aupperle, K. E. (1984). “An Empirical Measure CCICED to the Government of China. of Corporate social Orientation,” Research in CCICED Phase 3 Second Meeting, Beijing. Corporate Social Performance and Policy 6, Chauhoud, T. (2011). “Policies on Corporate 27–54. Social Responsibility,” in China’s Labour Ayyagari, M., T. Beck, and A. Demirgüç-Kunt Question. Ed. C. Scherrer. Germany: Rainer (2003). “Small and Medium Enterprises Hampp Verlag, 158–181. across the Globe: A New Database,” World Commission Recommendation (2003). “Con- Bank, Development Research Group, cerning the Definition of Micro, Small and Working Paper 3127. Washington, DC. Medium-sized Enterprises,” Official Journal Bansal, P., C. Maurer, and N. Slawinski (2008). of the European Union L 124, 36–41. “Beyond Good Intention: Strategies for Man- Crespell, P., and E. Hansen (2009). “Anteced- aging Your CSR Performance,” Ivey Business ents to Innovativeness in the Forest Products Journal Online 72(1), 9 pp. Available at Industry,” Journal of Forest Products Busi- http://iveybusinessjournal.com/topics/ ness Research 6(1), 20.

180 JOURNALLI, OF TOPPINEN, SMALL BUSINESS AND LANTTA MANAGEMENT 19 Cressy, R., D. Cumming, and C. Mallin (2010). European Union (2001). Green Paper: Promot- “Entrepreneurship, Governance and Ethics,” ing a European Framework for Corporate Journal of Business Ethics 95, 117–120. Social Responsibility. Commission of the Crewsell, J. W. (2009). Research Design: Quali- European Communities, Brussels. Available tative, Qualitative, and Mixed Methods at http://www.csr-in-commerce.eu/data/ Approaches, 3rd ed. Thousand Oaks, CA: files/resources/717/com_2001_0366_en.pdf. Sage. Fassin, Y. (2008). “SMEs and the Fallacy of Crewsell, J. W., and V. L. Plano Clark (2011). Formalizing CSR,” Business Ethics: A Euro- Design and Conducting Mixed Methods pean Review 17(4), 364–378. Research, 2nd ed. Thousand Oaks, CA: Sage. Fassin, Y., A. Rossem, and M. Buelens (2011). Cumming, D. J., and S. A. Johan (2007). “Small-Business Owner-Managers’ Percep- “Socially Responsible Institutional Invest- tions of Business ethics and CSR-Related ment in Private Equity,” Journal of Business Concepts,” Journal of Business Ethics 98, Ethics 75, 395–416. 425–453. Cumming, D. J., S. A. Johan, G. Fleming, and M. Fiegenbaum, A., and A. Karmani (1991). Takeuchi (2010). “Legal Protection, Corrup- “Output Flexibility: A Competitive Advan- tion, and Private Equity Returns in Asia,” tage for Small Firms,” Strategic Management Journal of Business Ethics 95(2), 173– Journal 12(2), 101–114. 193. Flodin, J., J. Oja, and A. Grönlund (2008). “Fin- Dahlsrud, A. (2008). “How Corporate Social gerprint Traceability of Sawn Products Using Responsibility Is Defined: An Analysis of 37 Industrial Measurement Systems for X-ray Definitions,” Corporate Social Responsibility Log Scanning and Sawn Timber Surface and Environmental Management 15(1), Scanning,” Forest Products Journal 58(11), 1–13. 100–105. Dibrell, C., J. Craig, and E. Hansen (2011). Frederick, W. C. (2006). Corporation Be Good! “Natural Environment, Market Orientation, The Story of Corporate Social Responsibility. and Firm Innovativeness: An Organizational Indianapolis, IN: Dogear Publishing. Life Cycle Perspective,” Journal of Small Friedman, M. (1970). “The Social Responsibility Business Management 49(3), 467–489. of Business Is to Increase Its Profits,” New Doh, J., and T. Duay (2006). “Corporate Social York Times 13 (September). Responsibility, Public Policy, and NGO Galbreath, J. (2005). “Which Resources Matter Activism in Europe and the United States: the Most to Firm Success? An Exploratory An Institutional-Stakeholder Perspective,” Study of Resource-Based Theory,” Journal of Management Studies 43, 47– Technovation 25(9), 979–987. 73. ——— (2009). “Building Corporate Social Donaldson, T. (2005). “Define the Value of Responsibility into Strategy,” European Doing Good Business,” Financial Times: Business Review 21(2), 109–127. Mastering Corporate Governance. 2 June. Galbreath, J., and P. Galvin (2008). “Firm Elkington, J. (1994). “Towards the Sustainable Factors, Industry Structure and Performance Corporation: Win-Win Business Strategies Variation: New Empirical Evidence to a for Sustainable Development,” California Classic Debate,” Journal of Business Management Review 36(2), 90–100. Research 61(2), 109–117. European Commission (2001). “Green Paper: Gibbs, G. R. (2007). “Analyzing Qualitative Promoting a European Framework for Cor- Data,” in Book 6 of The Sage Qualitative porate Social Responsibility,” Brussels, Research Kit. Ed. U. Flick. London: Sage. COM(2001) 366 final. Graafland, J., B. Ven van de, and N. Stoffele ——— (2003). “Commission Recommendation (2003). “Strategies and Instruments for Orga- of 6 May 2003: Concerning the Definition of nizing CSR by Small and Large Business in Micro, Small and Medium-Sized Enterprises,” the Netherlands,” Journal of Business Ethics Journal of the European Union L124, 36– 47(1), 45–60. 41. Griffin, J., and J. Mahon (1997). “The Corporate ——— (2007). Enterprise and Industry: SME Social Performance and Corporate Financial Definition: User Guide and Model Declara- Performance Debate: Twenty-Five Years of tion. Enterprise and Industry Publications. Incomparable Research,” Business and ISBN 92-894-7909-4. Society 36(1), 5–31.

20 JOURNALLI, OF TOPPINEN, SMALL BUSINESS AND LANTTA MANAGEMENT 181 Hammann, E., A. Habisch, and H. Pechlaner Johan, S. A., and D. Najar (2010). “The Role of (2009). “Values that Create Value: Socially Law, Corruption and Culture in Investment Responsible Business Practices in SMEs— Fund Manager Fees,” Journal of Business Empirical Evidence from German Compa- Ethics 95(2), 147–172. nies,” Business Ethics: A European Review Johansson, J. (2008). “The Swedish Hardwood 18(1), 37–51. Sawmill Industry: Structure, Present Status Hart, S. (1995). “A Natural-Resource-Based and Development Potential,” Wood Material View of the Firm,” Academy of Management Science and Engineering 3(3), 94–101. Review 20(4), 986–1014. John, K., and S. Kedia (2003). “Institutions, Heikinheimo, K. (2005). “Promoting Corporate Markets, and Growth: A Theory of Compara- Social Responsibility—Views of the tive Corporate Governance,” Working Paper, Finnish Public Sector’s Role and Influence,” New York University. in Vastuullinen Liiketoiminta: Perusky- ——— (2004). “Design of Corporate Gover- symyksiä ja Esimerkkejä. Eds. T. J. Kallio nance: Monitored Debt, Takeovers, and and P. Nurmi. Turku, Finland: Turku School Ownership Structure,” Working Paper, New of Economics and Business Management. York University. Series Discussion and Working Papers, Jorgensen, A. L., and J. S. Knudsen (2006). 10:2005, 87–100. “Sustainable Competitiveness in Global Hillary, R. (2000). Small and Medium-Sized Value Chains: How Do Small Danish Firms Enterprises and the Environment: Business Behave?,” Corporate Governance 6(4), 449– Imperatives. Sheffield, UK: Greenleaf Pub- 462. lishing. Ketola, T. (2009). “Corporate Responsibility for Hillman, A., and G. Keim (2001). “Shareholder Individual Cultural, and Biodiversity,” Man- Value, Stakeholder Management, and Social agement of Environmental Quality 20(3), Issues. What’s the Bottom Line?,” Strategic 239–254. Management Journal 22, 125–139. Kurucz, E., B. Colbert, and D. Wheeler (2008). Hofstede, G. (1991). Culture and Organiza- “The Business Case for Corporate Social tions: Software of the Mind. New York: Responsibility,” in The Oxford Handbook of McGraw Hill. Corporate Social Responsibility. Eds. A. Holliday, C., S. Schmidheiny, and P. Watts Crane, A. McWilliams, D. Matten, J. Moon, (2002). Walking the Talk: The Business Case and D. Siegel. Oxford: Oxford University for Sustainable Development. Sheffield, UK: Press, 83–112. Greenleaf. Lammers, J. C. (2003). “An Institutional Per- Hopkins, M., and R. Cowe (2003). Corporate spective on Communicating Corporate Social Responsibility: Is There a Business Responsibility,” Management Communica- Case? London: ACCA Publication. tion Quarterly 16(4), 618–624. Huju, P. (2005). “Yhteiskuntavastuu Yrityksen Lankoski, L. (2009). “Different Economic Kilpailuetuna,” in Kymmenvuotiskatsaus Impacts of Corporate Responsibility Issues,” 2005: Teemana Yritystoiminta. Ed. J. Autio. Business and Society 48(2), 206–224. Helsinki: Tilastokeskus, 95–97. Lähdesmäki, M. (2005a). “Yrityksen Hunt, S. D., and S. M. Vitell (1986). “A General Yhteiskuntavastuun Käsite Pienyrityksissä— Theory of Marketing Ethics,” Journal of Kilpailuvaltti vai Kilpailun Este?,” Raportteja Macromarketing 6, 5–15. 7, Helsingin Yliopisto, Ruralia-instituutti, International Finance Corporation. (2006). Seinäjoki. “Micro, Small, and Medium Enterprises: A ——— (2005b). “When Ethics Matters— Collection of Published Data,” Database, Interpreting the Ethical Discourse of Small World Bank. Nature-Based Entrepreneurs,” Journal of Jenkins, H. (2006). “Small Business Champions Business Ethics 61, 55–68. for Corporate Social Responsibility,” Journal ——— (2012). Studies on Corporate Social of Business Ethics 67(3), 241–256. Responsibility in the Finnish Small Business ——— (2009). “A ‘Business Opportunity’ Model Context. Published Doctoral Dissertation. of Corporate Social Responsibility for Small Ruralia Institute. University of Helsinki. and Medium-Sized Enterprises,” Business ISBN: 978-952-10-6506-4. Ethics: A European Review 18(1), 21– Lähtinen, K. (2009). Assessing the Resource 36. Usage Decisions and Financial Performance

182 JOURNALLI, OF TOPPINEN, SMALL BUSINESS AND LANTTA MANAGEMENT 21 in Finnish Sawmills within the Resource- Company’s Social and Financial Perfor- Based View Framework. Vantaa, Finland: mance. Mahwah, NJ: Lawrence Erlbaum. Dissertationes Forestales 89. Finnish Forest ——— (2003). “Misery Loves Companies: Research Institute. Rethinking Social Initiatives by Business,” Lähtinen, K., A. Toppinen, P. Leskinen, and A. Administrative Science Quarterly 48, 268– Haara (2009). “Exploring the Connection 305. between Resource Usage Decisions and Matten, D., and J. Moon (2004). “Pan-European Business Success in the Case of Finnish Approach: A Conceptual Framework for Large and Medium-Sized Sawmills,” Journal Understanding CSR,” in CSR across Europe. of Forest Products Business Research 6(1), Eds. A. Habisch, J. Jonker, M. Wegner, and 1–18. R. Schmidpeter. Berlin: Springer, 339–360. Lefebvre, L. A., and E. Lefebvre (1993). “Com- Mattila, M. (2009). “Corporate Social Responsi- petitive Positioning and Innovative Effort in bility and Image in Organizations: For the SMEs,” Small Business Economics 5(4), 297– Insiders or the Outsiders?,” Social Responsi- 305. bility Journal 5(4), 540–549. Li, N., A. Toppinen, A. Tuppura, K. Mayers, J. (2006). “ ’Small- and Medium-Sized Puumalainen, and M. Hujala (2011). “Deter- Forestry Enterprises,’ International Tropical minants of Sustainability Disclosure in the Timber Organization (ITTO),” Tropical Global Forest Industry,” Electronic Journal Forest Update 16(2), 10–11. of Business Ethics and Organization Studies McWilliams, A., D. Siegel, and P. M. Wright 16(1), 33–40. (2006). “Corporate Social Responsibility: Litz, R. A. (1996). “A Resource-Based View of a Strategic Implications,” Journal of Manage- Socially Responsible Firm: Stakeholder Inter- ment Studies 43(1), 1–18. dependence, Ethical Awareness and Issue Mikkilä, M., and A. Toppinen (2008). Responsiveness as Strategic Assets,” Journal “Sustainability Reporting in the World’s of Business Ethics 15(12), 1355–1363. Largest Pulp and Paper Companies,” Forest Luo, X. J., R. Li, L. Y. Lin, X. F. Gao, G. Y. Pan, Policy and Economics 10(8), 500–506. E. L. Xia, and J. Li (2009). Challenge and Moir, L. (2001). “What Do We Mean by Corpo- Opportunities for China’s Small and rate Social Responsibility?,” Corporate Gov- Medium Forest Enterprises (SMFEs), Forest ernance 1(2), 16–22. Connect: Diagnostic Studies on Small and Morsing, M. (2003). “CSR—A Religion with Too Medium-Sized Forest Enterprises. Rome: Many Priests?,” An interview with Michael FAO. Porter. European Business Forum 15, Macqueen, D. (2004). Associations of Small and Autumn 2003. Available at http://www Medium Forest Enterprise: An Initial Review .fsg.org/Portals/0/Uploads/Documents/PDF/ of Issues for Local Livelihoods and CSR_Religion_with_Too_Many_Priests.pdf Sustainability. London: IIED, Briefing ?cpgn=WP%20DL%20-%20CSR%20-%20A Paper. %20Religion%20With%20too%20Many Macqueen, D., and J. Mayers (2006). Forestry’s %20Priests. Messy Middle: A Review of Sustainability Murillo, D., and J. M. Lozano (2006). “SMEs and Issues for Small and Medium Forest Enter- CSR: An Approach to CSR in Their Own prises. London: IIED, Policy Paper. Words,” Journal of Business Ethics 67(3), Macqueen, D., S. Bose, S. Bukula, C. Kazoora, 227–240. S. Ousman, N. Porro, and H. Weyerhaeuser. National Bureau of Statistics of China (2003). (2006). “Working Together: Forest-Linked China Agricultural Statistics 2003. Beijing: Small and Medium Enterprise Associations National Bureau of Statistics Press. and Collective Action,” IIED, Gatekeeper Nielsen, A. E., and C. Thomsen (2009). “CSR Series 125. London, UK. Communication in Small and Medium-Sized McWilliams, A., and D. Siegel (2000). “Corpo- Enterprises: A Study of the Attitudes and rate Social Responsibility and Financial Per- Beliefs of Middle Managers,” Corporate Com- formance: Correlation or Misspecification?,” munications 14(2), 176–189. Strategic Management Journal 21(5), 603– Observatory of European SMEs (2002). Euro- 609. pean SMEs and Social and Environ- Margolis, J. D., and J. P. Walsh (2001). People mental Responsibility. Belgium: European and Profits? The Search for a Link between a Commission.

22 JOURNALLI, OF TOPPINEN, SMALL BUSINESS AND LANTTA MANAGEMENT 183 O’Brien, D. (2001). “Integrating Corporate Canada,” Forest Products Journal 56(11/12), Social Responsibility with Competitive Strat- 25–32. egy,” Center for Corporate Citizenship at Salzmann, O., A. Ionescu-Somers, and U. Steger Boston College. (2005). “The Business Case for Corporate Orlitzky, M., F. L. Schmidt, and S. L. Rynes Sustainability: Literature Review and (2003). “Corporate Social Financial and Per- Research Options,” European Management formance: A Meta-Analysis,” Organizational Review 23(1), 27–36. Studies 24(3), 403–441. Sawhney, M., R. Wolcott, and I. Arroniz (2006). Peng, S. Z., and R. M. Ren (2003). “The “The 12 Different Ways for Companies to Stimuli and Barriers for EMS in the SMEs,” Innovate,” MIT Sloan Management Review Shanghai Management Science 1, 27–28 (in 47(3), 75–81. Chinese). Shen, S. B., and H. E. Cheng (2009). “Economic Peng, S. Z., Y. Liu, H. Shi, and P. Zhong (2005). Globalization and the Construction of “Studies on Barriers for Promotion of Clean China’s Corporate Social Responsibility,” Technology in SMEs of China,” Chinese International Journal of Law and Manage- Journal of Population, Resources and Envi- ment 51(3), 134–138. ronment 3(1), 9–17. Shi, H., S. Z. Peng, Y. Liu, and P. Zhong (2008). Perrini, F. (2006). “SMEs and CSR Theory: Evi- “Barriers to the Implementation of Cleaner dence and Implications from an Italian Per- Production in Chinese SMEs: Government, spective,” Journal of Business Ethics 67(3), Industry and Expert Stakeholders’ Perspec- 305–316. tives,” Journal of Cleaner Production 16(7), Perrini, F., and M. Minoja (2008). “Strategizing 842–852. Corporate Social Responsibility: Evidence Shore, L. M., and S. J. Wayne (2003). “Commit- from an Italian Medium-Sized, Family- ment and Employee Behaviour: Comparison Owned Company,” Business Ethics: A Euro- of Affective Commitment and Continuance pean Review 17(1), 47–63. Commitment With Perceived Organizational Perrini, F., A. Russo, and A. Tencati (2007). Support,” The Journal of Applied Psychology “CSR Strategies of SMEs and Large Firms: 78(5), 774–780. Evidence from Italy,” Journal of Business Shrivastava, P. (1995). “The Role of Corpora- Ethics 74(3), 285–300. tions in Achieving Ecological Sustainability,” Porter, M., and C. van der Linde (1995). “Green The Academy of Management Review 20(4), and Competitive: Ending the Stalemate,” 936–960. Harvard Business Review 73(5), 120– Silverman, D. (2000). Doing Qualitative 134. Research: A Practical Handbook. London: Porter, M. E., and M. R. Kramer (2006). “Strat- Sage. egy and Society: The Link Between Competi- Spence, L., and R. Rutherford (2003). “Small tive Advantage and Corporate Social Business and Empirical Perspectives in Busi- Responsibility,” Harvard Business Review ness Ethics: Editorial,” Journal of Business 84, 78–92. Ethics 47, 1–5. Reed, D. J. (2001). Stalking the Elusive Business Spence, L. J., and J. F. Lozano (2000). “Commu- Case For Corporate Sustainability. Washing- nicating About Ethics with Small Firms: ton, DC: World Resources Institute. Experiences from the U.K. and Spain,” Russo, A., and A. Tencati (2009). “Formal vs. Journal of Business Ethics 27(1), 43–53. Informal CSR Strategies: Evidence from Spiller, R. (2000). “Ethical Business and Invest- Italian Micro, Small, Medium-Sized, and ment: A Model for Business and Society,” Large Firms,” Journal of Business Ethics Journal of Business Ethics 27(1–2), 149– 85(2), 339–353. 160. Russo, M., and P. Fouts (1997). “A Stendhal, M., A. Roos, and M. Hugosson (2007). Resource-Based Perspective on Corporate “Product Development in the Swedish and Environmental Performance and Proﬁtabil- Finnish Sawmilling Industry—A Qualitative ity,” Academy of Management Review 40, Study of Managerial Perceptions,” Journal of 534–559. Forest Products Business Research 4(4), Salehirad, N., and T. Sowlati (2006) “Productiv- 1–24. ity and Efﬁciency Assessment of the Wood Su, Y. (2004). “Policy Recommendations on Industry – A Review with a Focus on Financial Mechanisms for Pollution Control

184 JOURNALLI, OF TOPPINEN, SMALL BUSINESS AND LANTTA MANAGEMENT 23 of SMEs in China,” Consultation and Deci- Appendix: Themed sion 2, 8–11 (in Chinese). Sweeney, L. (2007). “Corporate Social Respon- Interview Questionnaire sibility in Ireland: Barriers and Opportuni- 1. Background information ties Experienced by SMEs when Undertaking CSR,” Corporate Governance 7(4), 516– -Name of informant 523. -Job title and responsibility Taipalinen, J., and T. Toivio (2004). -Work history in the company “Vastuullinen Yritystoiminta Pk-Yritysten -Gender Voimavarana,” KTM julkaisuja 16/2004, -Education KTM elinkeino-osasto, Helsinki. 2. Internal Operations Toppinen, A., N. Li, A. Tuppura, and Y. Xiong When considering the activities that take (2012). “Corporate Responsibility and place at the mill, can you identify specific Strategic Groups in the Forest-Based Indus- sources of competitive advantage? For example: try: Exploratory Analysis Based on the Global Reporting Initiative (GRI) Frame- -Technologies (e.g., raw material use effi- work,” Corporate Social Responsibility and ciency, innovation) Environmental Management 19(4), 191– -Other processes (e.g., continuous R&D, 205. quality system) Turban, D. B., and D. W. Greening (1996). -End-product quality (e.g., price/quality, “Corporate Social Performance and Organi- desired quality) zational Attractiveness to Prospective Employee,” Academy of Management 3. External Operations Journal 40(3), 658–672. When considering the activities outside the van Marrewijk, M. (2003). “Concepts and Defi- mill, can you identify specific sources of com- nitions of CSR and Corporate Sustainability: petitive advantage? For example: Between Agency and Communion,” Journal of Business Ethics 44, 95–105. - Wood procurement (e.g., proximity, “appro- van Tulder, R., and A. van der Zwart (2005). priate” quality) International Business-Society Management: - Logistics (e.g., avoiding unnecessary Linking Corporate Responsibility and Glo- transport) balization. Oxon: Routledge. - Efficient use of by-products (e.g., in wood Vidal, N. G., G. Q. Bull, and R. A. Kozak (2010). panel production and pulp manufacture, “Diffusion of Corporate Responsibility Prac- proximity of business partners) tices to Companies: The Experience of the Forest Sector,” Journal of Business Ethics 4. Internal Interactions 4(4), 553–567. Considering how activities at the mill are Walley, N., and B. Whitehead (1994). “It’s Not organised and how people are managed, can Easy Being Green,” Harvard Business you identify specific sources of competitive Review 72(3), 46–52. advantage? For example: World Business Council for Sustainable Devel- opment (WBCSD) (2002). The Business Case -Personnel (know-how, education, perfor- for Sustainable Development. Geneva: mance appraisal, merit pay, job stability) WBCSD. -Low organizational structure and efficient Whetten, D. A., G. Rands, and P. Godfrey communications (2002). “What Are the Responsibilities of - Awareness of strategy and performance Business to Society?,” in Handbook of Strat- egy and Management. Eds. A. M. Pettigrew, 5. External Interactions T. Howard, and R. Whittington. London: Considering external stakeholders and your Sage, 373–409. relationship with these groups, can you identify Yin, R. K. (1989). Case Study Research: Design specific sources of competitive advantage? For and Methods. London: Sage. example: ——— (2009). Case Study Research: Design and Methods, 4th ed. Thousand Oaks, CA: -Customer relationships (e.g., long-term rela- Sage. tionships, charting customer needs)

24 JOURNALLI, OF TOPPINEN, SMALL BUSINESS AND LANTTA MANAGEMENT 185 -Surrounding community (e.g., employment, -How is CR implemented by the company you voluntary participation in communal work for? activities) -Has its implementation changed during your -Forest owners (e.g., maintaining long-term work history? relationships) -How important do you consider CR is to the -Other stakeholders (e.g., supply chain part- CA of your firm now? ners, other businesses, educational facilities, -How important do you think CR will be to the environmental organizations, society at large) CA of your firm in the future? -Reporting, marketing, other communications -Monitoring societal change (as categorized in about the firm’s economic, environmental and Table 1). social performance

6. Corporate Social Responsibility

-How do you deﬁne CR? -When did you ﬁrst learn about CR?

186 JOURNALLI, OF TOPPINEN, SMALL BUSINESS AND LANTTA MANAGEMENT 25 Journal of Small Business Business Management Management 2016 2014 54(1), ••(••), pp. pp. 187–209 ••–•• doi: 10.1111/jsbm.12137 The Impact of Entrepreneurship Education: A Study of Iranian Students’ Entrepreneurial Intentions and Opportunity Identiﬁcation by Saeid Karimi, Harm J. A. Biemans, Thomas Lans, Mohammad Chizari, and Martin Mulder

Building on the theory of planned behavior, an ex ante and ex post survey was used to assess the impacts of elective and compulsory entrepreneurship education programs (EEPs) on students’ entrepreneurial intention and identification of opportunities. Data were collected by questionnaire from a sample of 205 participants in EEPs at six Iranian universities. Both types of EEPs had significant positive impacts on students’ subjective norms and perceived behavioral control. Results also indicated that the elective EEPs significantly increased students’ entrepreneurial intention, although this increase was not significant for the compulsory EEPs. The findings contribute to the theory of planned behavior and have implications for the design and delivery of EEPs.

can be taught and that education can be con- Introduction sidered one of the key instruments for fostering During the past few decades, entrepreneur- entrepreneurial attitudes, intentions, and ship has become an important economic and competences (Falkang and Alberti 2000; Harris social topic as well as an often-researched and Gibson 2008; Henry, Hill, and Leitch 2005; subject around the world (Fayolle and Gailly KuratKo 2005; Martin, McNally, and Kay 2013; 2008). According to research, entrepreneurship Mitra and Matlay 2004). This view has led to a is an intentional and planned behavior that dramatic rise in the number and status of entre- can increase economic efﬁciency, bring innova- preneurship education programs (EEPs) in col- tion to markets, create new jobs, and raise leges and universities worldwide (Finkle and employment levels (Shane and Venkataraman Deeds 2001; Katz 2003; KuratKo 2005; Matlay 2000). Most empirical studies indicate that 2005); investment in these programs is still on entrepreneurship, or at least some aspects of it, the increase (Gwynne 2008). Nevertheless, the

Saeid Karimi is an entrepreneurship PhD student in the Department of Social Sciences, Education, and Competence Studies Group at the Wageningen University. Harm J. A. Biemans is an associate professor of education in the Department of Social Sciences, Education, and Competence Studies Group at the Wageningen University. Thomas Lans is an assistant professor of entrepreneurship in the Department of Social Sciences, Education, and Competence Studies Group at the Wageningen University. Mohammad Chizari is professor of agricultural education in the Department of Agricultural Education and Extension at the Tarbiat Modares University. Martin Mulder is professor of education in the Department of Social Sciences, Education, and Competence Studies Group at the Wageningen University. Address correspondence to: Saeid Karimi, Department of Social Sciences, Education, and Competence Studies Group, Wageningen University, Wageningen 6702 PJ, The Netherlands. E-mail: [email protected].

KARIMI ET AL. 187 1 impact of these programs has remained largely design to study these effects. And the fourth unexplored (Bechard and Gregoire 2005; contribution is to assess the effect of entrepre- Peterman and Kennedy 2003; Pittaway and neurship education on non-business university Cope 2007; von Graevenitz, Harhoff, and students in a developing country, namely Iran. Weber 2010). Moreover, the results of previous This paper is organized as follows. In the next studies are inconsistent. Some of these studies section, we explain entrepreneurial intentions reported a positive impact from EEPs (e.g., (EIs) and the theory of planned behavior (TPB). Athayde 2009; Fayolle, Gailly, and Lassas-Clerc We then discuss the relationships between 2006; Peterman and Kennedy 2003; Souitaris, intentions, their antecedents, and opportunity Zerbinati, and Al-Laham 2007), whereas others identification, and point out how EEPs may found evidence that the effects are statistically affect these factors. Next, we describe the insignificant or even negative (Mentoor and method and findings. Finally, we discuss our Friedrich 2007; Oosterbeek, van Praag, and results and their implications both for the prac- Ijsselstein 2010; von Graevenitz, Harhoff, tice of entrepreneurship education and for and Weber 2010). future research. Methodological limitations may be the cause of these inconsistent results (von Graevenitz, Theoretical Framework Harhoff, and Weber 2010). Some studies, for EIs. In the social psychology literature, inten- instance, are ex post examinations that there- tions have proved to be the best predictor of fore do not assess the direct impact of an EEP planned individual behaviors, especially when (e.g., Kolvereid and Moen 1997; Menzies and the target behavior is rare, difficult to observe, Paradi 2003), or have small sample sizes (e.g., or involves unpredictable time lags (Krueger, Fayolle, Gailly, and Lassas-Clerc 2006; Jones Reilly, and Carsrud 2000). Entrepreneurship is et al. 2008); this has led Martin, McNally, and a typical example of such planned and inten- Kay (2013) to conclude that entrepreneurship tional behavior (Bird 1988; Krueger and education researchers must include pre- and Brazeal 1994). EI refers to a state of mind that postentrepreneurship interventions. Previous directs and guides the actions of the individual studies also have not differentiated between toward the development and implementation elective and compulsory programs, and of a new business concept (Bird 1988). There is research on the important role of compulsory a vast body of literature arguing that EI plays versus voluntary participation in EEPs has been a very pertinent role in the decision to start a neglected; therefore, Oosterbeek, van Praag, new business (Linan and Chen 2009). As a and Ijsselstein (2010) call for the testing of consequence, in recent years, employment different program variants. In addition, there is status choice models that focus on EI have been no agreement on what would constitute a well- the subject of considerable interest in entrepre- defined method and a suitable conceptual neurship research (e.g., Engle et al. 2010; model for assessing the effects of EEPs Iakovleva, Kolvereid, and Stephan 2011; Karimi (Falkang and Alberti 2000; von Graevenitz, et al. 2013a, 2014). Krueger, Reilly, and Carsrud Harhoff, and Weber 2010). Moreover, non- (2000) found that intention models offer a great business university students have received opportunity to increase our understanding and limited attention in previous studies (Lans et al. predictive ability for entrepreneurship. 2013), despite the fact that this population represents the bulk of young adults pursuing an The TPB. Among intention models, one of the education program. Finally, there is no study most widely researched is the TPB, originally regarding the impact of entrepreneurship edu- presented by Ajzen (1991). This model has been cation for Iranian universities. widely applied in entrepreneurship research, The present study has attempted to reduce and its efficacy and ability to predict EI and these theoretical and methodological gaps and behaviors have been demonstrated in a number make four contributions to the existing litera- of studies on entrepreneurship (for example, ture. First, we applied an intention model to Karimi et al. 2014; Kolvereid and Isaksen 2006). assess the impact of EEPs. As a second contri- The central factor of the TPB is the individual bution, we studied the effects of large-scale intention to perform a given behavior (e.g., the compulsory and elective entrepreneurship intention to become an entrepreneur). Conse- courses at different universities. The third con- quently, the model stresses that intention is tribution is our use of a pretest plus post-test affected by three components or antecedents

2188 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT (Ajzen 1991): (1) subjective norms (SN), refer- sum, these findings together support Ajzen’s ring to perceived social pressures to perform or (1991) assertion that all three antecedents are refrain from a particular behavior (e.g., becom- important, although their explanatory power is ing an entrepreneur); (2) attitudes toward the not the same in every situation and country. behavior, that is, the degree to which a person Therefore, it is hypothesized that has a favorable or unfavorable evaluation about performing the target behavior (e.g., being an H1: (1) SN, (2) ATE, and (3) PBC are positively entrepreneur); and (3) perceived behavioral related to university students’ EI. control (PBC), that is, the perceived difficulty or ease of performing the behavior (e.g., becoming Opportunity Identification. Opportunity iden- an entrepreneur). PBC is conceptually similar to tification or recognition has been defined as perceived self-efficacy as proposed by Bandura the ability to identify a good idea and trans- (1997). In both concepts, the sense of capacity to form it into a business concept (or the consid- perform the activity is important (Ajzen 2002). erable improvement of an existing venture) that adds value to the customer or society and Hypotheses generates revenues for the entrepreneur Researchers have empirically applied the (Lumpkin and Lichtenstein 2005). Opportunity TPB to students’ EI and confirmed the theory’s identification has long been accepted as a key predictions regarding the effects of SN, PBC, step in the entrepreneurial process (Ozgen and attitude toward entrepreneurship (ATE) on and Baron 2007). In fact, without business their intentions (e.g., Engle et al. 2010; Linan opportunity identification, there is no entre- and Chen 2009; Iakovleva, Kolvereid, and preneurship (Short et al. 2010). For this Stephan 2011). However, these findings as a reason, opportunity identification has become whole do not represent a conclusive and con- a required element of scholarly research and sistent picture. Linan and Chen (2009) tested studies of entrepreneurship, and there has the TPB among university students in Spain been considerable interest in studying the and Taiwan. Their results showed that both factors, processes, and dynamics that foster it ATE and PBC had significant effects on EI; (Gregoire, Shepherd, and Lambert 2010). The however, PBC was the strongest predictor of EI literature provides two main theories regard- in Taiwan, whereas in Spain, ATE was the ing opportunity identification: the discovery strongest predictor of EI. Even though SN had theory and the creation theory (Alvarez and no significant direct effect on intention, SN Barney 2007). Recent research has provided indirectly affected intention through ATE and evidence that both the discovery and creation PBC. Engle et al. (2010) tested the ability of the approaches can occur in entrepreneurial prac- TPB to predict EI in 12 countries. The results tice and that research is moving toward a suggested that the TPB model successfully pre- middle ground position (Bhave 1994; Short dicted EI in each of the study countries, et al. 2010). although, as foreseen by Ajzen and just illustrated in empirical work, the significant contrib- The TPB and Opportunity Identification. uting model elements differ among countries. Although three attitudinal antecedents are Engle et al. (2010) reported that SN was a sig- known to influence a wide range of behaviors, nificant predictor of EI in every country, prior studies conducted in different areas (e.g., whereas ATE was a significant predictor in only Bagozzi, Moore, and Leone 2004; Conner and six countries (China, Finland, Ghana, Russia, Armitage 1998; Haustein and Hunecke 2007; Sweden, and the United States), and PBC was a Hsu et al. 2006; Perugini and Bagozzi 2001) significant predictor in only seven countries argued that additional variables could enhance (Bangladesh, Egypt, Finland, France, Germany, the power of the TPB to predict and explain an Russia, and Spain). Finally, Iakovleva, individual’s intention and behavior. Within the Kolvereid, and Stephan (2011) used the TPB to domain of entrepreneurship, opportunity iden- predict EI among students in five developing tification can be added to the TPB as an addi- and eight developed countries. The findings tional fundamental element. As mentioned, provided support for the applicability of the opportunity identification is a crucial compo- TPB in both developing and developed coun- nent of the entrepreneurial process (Ardichvilia, tries. They found the three antecedents to be Cardozob, and Ray 2003; Gaglio and Katz 2001; significantly related to EI in all 13 countries. In Shane and Venkataraman 2000), and it is an

KARIMI ET ET AL. 189 3 intentional process (Krueger, Reilly, and grounded in the TPB (e.g., Dutton and Jackson Carsrud 2000). In fact, the act of entrepreneur- 1987; Krueger 2003). In addition, researchers ship and the creation of a new business firm are have made considerable efforts to understand based on the joint occurrence of two events the antecedents of opportunity identification (Krueger and Brazeal 1994; Reitan 1997a). The (e.g., Ardichvilia, Cardozob, and Ray 2003; first event is the presence of a suitable entrepre- Baron and Ensley 2006; Casson and Wadeson neurial opportunity, whereas the second event 2007; Gaglio and Katz 2001; Ozgen and Baron represents a person who is able and willing to 2007; Shane 2000). These attempts have con- take advantage of an entrepreneurial opportu- tributed greatly to our understanding of oppor- nity. When these two events coincide, entrepre- tunity identification; however, they fall short of neurial behaviour may take place; thus, a new offering a comprehensive understanding of the firm can be founded. According to Reitan process. Dutton and Jackson (1987) first (1997a), “a potential entrepreneur is a person mapped out an elegant model of opportunity who perceives a venture opportunity and/or perception in a study with similarities to the intends to start a new venture, but has not (yet) TPB. They argued that a situation is perceived taken any steps regarding venture start-up.” The as an opportunity when an individual’s percep- argument is that opportunity identification and tion of the outcomes is positive and the situa- EI are key characteristics of potential entrepre- tion is perceived as controllable. Jackson and neurs, and both must be present for new busi- Dutton (1988) tested this model successfully. ness creation to take place. Based on Shapero’s and Sokol (1982) model Edelman and Yli-Renko (2010) also stated and Dutton and Jackson (1987), Krueger (2000, that perceptions and other cognitive factors 2003) and Krueger and Brazeal (1994) devel- play a fundamental role in both the discovery oped a complementary EI model that includes and creation views of entrepreneurship. They the perception of opportunity. According to argued that the perception that opportunities this model, the perception of opportunity is exist in the market rather than the actual envi- dependent on the same two crucial antecedents ronment or the objective changes in technology of EI, perceptions of desirability (attitude in the or consumer needs is important in predicting TPB) and perceptions of feasibility (PBC or efforts to create a new business. In other self-efficacy in the TPB). In other words, if words, perceptions of opportunity will stimu- individuals perceive entrepreneurship as desir- late an individual’s efforts to start a new busi- able and feasible, they are more likely to see an ness. Stronger perceptions will increase the opportunity and, thus, form an EI. Reitan intention to create a new firm and the energy of (1997b) conducted an empirical study and potential entrepreneurs to start a firm (Edelman found that opportunity identification has some and Yli-Renko 2010). A perception of an oppor- of the same antecedents as EI. Specifically, per- tunity can spark an intention-based cognitive ceptions of desirability and feasibility were process that leads to entrepreneurial action strong predictors of both, whereas SN was (Krueger, Reilly, and Carsrud 2000). It has been important for understanding EI only. shown that the opportunity identification per- Although the relationship between OIP and ception (OIP) and EI are closely connected ATE is less clear and research on this relation- (Bird 1988). That is, a person who finds an ship is scant, previous empirical studies indi- opportunity desirable and feasible is likely to cate that PBC may be positively related to OIP. create a business (Bhave 1994). According to Ajzen (2002), PBC includes self- On the basis of the previous discussion and efficacy and controllability. Research has dem- in line with Reitan (1997b) and Edelman and onstrated that self-efficacy (Krueger and Yli-Renko (2010), we propose the following Dickson 1994) and controllability (Dutton hypothesis: 1993) are positively linked to opportunity identification. Studies have also found that self- H2: Those students who have higher OIP will efficacy is a remarkable predictor of OIP have greater intentions to start up a new (Ardichvilia, Cardozob, and Ray 2003; Gibbs business. 2009; Gonzalez-Alvarez and Solis-Rodriguez 2011; Krueger 2000; Mitchell and Shepherd In the last decade, researchers have pre- 2010; Ozgen and Baron 2007; Ucbasaran, sented numerous models of entrepreneurship Westhead, and Wright 2009). For example, the and opportunity identification that are study by Krueger and Dickson (1994) found a

4190 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT direct correlation between an increase in self- (Athayde 2009; Peterman and Kennedy 2003; efficacy and an increase in perceptions of Souitaris, Zerbinati, and Al-Laham 2007; von opportunity. Increasing entrepreneurial self- Graevenitz, Harhoff, and Weber 2010). Several efficacy should increase perceived feasibility of scholars (e.g., Fayolle, Gailly, and Lassas-Clerc starting a business, thus increase perceptions of 2006; Weber 2012) suggest that the TPB is opportunity (Krueger, Reilly, and Carsrud appropriate for the evaluation of EEPs such as 2000). Ozgen and Baron (2007) believe that entrepreneurship courses. The main purpose of individuals with high self-efficacy tend to have such an intervention is to bring about a change broader social networks and to be more in students’ entrepreneurial attitudes and inten- popular due to high self-confidence and self- tions, and the TPB promises to deliver a sound assurance; as a result, these people will receive framework for assessing this change systemati- more information. Therefore, these authors cally. The TPB has been empirically used by believe that high self-efficacy may indeed be some researchers to assess the impact of EEPs linked to opportunity recognition in this on the students’ EI, and its value has been manner. Moreover, individuals with high self- successfully demonstrated (Fayolle, Gailly, and efficacy believe that they can successfully Lassas-Clerc 2006; Souitaris, Zerbinati, and develop the opportunities they discover. As a Al-Laham 2007). As such, the TPB is considered result, they may be more proactive in searching to provide a useful framework for both analyz- for such opportunities (e.g., Gaglio and ing how EEPs might influence students with Katz 2001) and, in particular, in seeking regard to their EI and, in particular, for defining opportunity-relevant information from other and measuring relevant criteria. persons (Ozgen and Baron 2007). Accordingly, their study demonstrates that self-efficacy is Entrepreneurship Education Effects positively related to opportunity recognition. on EIs Drawing on the results and arguments in the Krueger and Carsrud (1993) were the first to studies just mentioned, we propose that stu- apply the TPB in the specific context of entre- dents’ PBC and ATE influence their perception preneurship education. They pointed out that of new business opportunity identification. an education program can have an impact on the antecedents of intention identified by the H3: (1) ATE and (2) PBC will be positively TPB. Fayolle, Gailly, and Lassas-Clerc (2006) related university students’ OIP. found that although entrepreneurship education has a strong and measurable effect on Entrepreneurship Education. Entrepreneurial students’ EI, it has a positive, but not very education is a rapidly growing area and a hot significant, impact on their PBC. Souitaris, topic in colleges and universities all around the Zerbinati, and Al-Laham (2007) used the TPB in world, and its supposed benefits have received order to test the impact of EEPs on the attitudes much praise from researchers and educators. and intentions of science and engineering stu- Nevertheless, the outcomes and effectiveness of dents. They found that EEPs significantly EEPs have remained largely untested (Pittaway increased students’ EI and subjective norms. and Cope 2007; von Graevenitz, Harhoff, and However, they did not find a significant rela- Weber 2010). According to Alberti, Sciascia, tionship between EEPs and attitudes and PBC, and Poli (2004), the first and most important whereas Peterman and Kennedy (2003) and area for further investigation should include Athayde (2009) found a positive effect of EEPs assessing the effectiveness of these programs. on intentions and perceived feasibility, or ATE, However, this raises an important question: among high-school students. Walter and Dohse How should entrepreneurship education be (2012) reported that EEPs were positively assessed? One of the most common ways to related only to ATE, not to SN or PBC. Results evaluate an EEP is to assess individuals’ inten- regarding entrepreneurship education initia- tions to start a new business. Intentionality is tives are therefore somewhat inconclusive, and central to the process of entrepreneurship (Bird more detailed research is needed to get a full 1988; Krueger 1993), and studies show that EI understanding of the relationship between is a strong predictor of entrepreneurial behav- entrepreneurship education and attitudes/ ior. Nonetheless, the impact of EEPs on EI to intentions. Notably, in their recent meta- set up a business is at present poorly under- analysis, Martin, McNally, and Kay (2013) stood and has remained relatively untested found overall positive effects of EEPs on

KARIMI ET ET AL. 191 5 knowledge and skill, perceptions of entrepre- tiﬁcation of entrepreneurial opportunities neurship, and entrepreneurship outcomes. (Shepherd and DeTienne 2005). Thus, we Thus, we propose that propose that

H4: Students who have followed an EEP will H5: Students who have followed an EEP will have higher (a) SN, (b) ATE, (c) PBC, and have higher OIP after the program than (d) EI after the program than before the before the program. program. Elective versus Compulsory Entrepreneurship H4e: Students whose SN, ATE, and PBC have Education. As already mentioned, empirical increased will also have increased their studies have yielded mixed results about EI. the effects of EEPs on entrepreneurship. Oosterbeek, van Praag, and Ijsselstein (2010) and von Graevenitz, Harhoff, and Weber (2010) Entrepreneurship Education Effects on found that the EEPs had a negative impact on Opportunity Identification EI. Both studies examined compulsory EEPs. If entrepreneurs are to be successful in cre- Oosterbeek, van Praag, and Ijsselstein (2010) ating and operating new ventures, they must argued that the effects of EEPs may have been not only develop an EI but also be successful at negative because participation in EEPs was discerning opportunities that others ignore or compulsory. In this study, we assess the effects fail to notice and then exploit these opportuni- of two types of EEPs (voluntary, or elective, ties in a timely and effective manner (Dutta, Li, and compulsory EEPs) on students’ EI. Com- and Merenda 2011). Therefore, developing pulsory programs are given to every student opportunity identification abilities is a key enrolled in a certain degree program; therefore, element of the entrepreneurship process, and they include both those interested and those entrepreneurship education should enhance uninterested in entrepreneurial activity and this competency (Linan, Rodríguez-Cohard, and education. However, participants in elective Rueda-Cantuche 2011; Lumpkin, Hills, and EEPs have an interest in entrepreneurship edu- Shrader 2004). According to the entrepreneur- cation and seek out further knowledge and ship education literature, opportunity identifi- skills in entrepreneurship. Moreover, motivated cation could and should be taught, and it students will more actively participate in learn- should be a central topic in programs that aim ing activities than students forced to take the to train future entrepreneurs (Saks and Gaglio course. Therefore, we can expect that an elec- 2002). Along the same lines, DeTienne and tive EEP has a greater influence on participants Chandler (2004) state that the entrepreneurship than does a compulsory one. classroom is an appropriate place for fostering the skills required to enhance opportunity iden- H6: An elective EEP will have a greater effect tification competency. Despite a growing on students’ ATE, SN, PBC, OIP, and EI com- amount of literature on opportunity identifica- pared with a compulsory EEP. tion and its importance in the entrepreneurship process, there is a dearth of research regarding Research Method the effects of education on students’ ability to EEPs. Over the past decades, many develop- identify business opportunities. The results of a ing countries including Iran have faced various study by DeTienne and Chandler (2004) indi- economic problems, in particular the excessive cate that entrepreneurship education led to the number of university graduates unable to find identification of more opportunities and more government or private sector work opportuni- innovative opportunities. Munoz, Mosey, and ties. Over the last decade, Iran has expressed Binks (2011) also reported that entrepreneur- increasing interest in various entrepreneurship education develops students’ opportunity ship fields (in higher education settings, identification capabilities. Moreover, entrepre- policymaking, and business) as a fundamental neurship education can increase the entrepre- solution for the unemployment problem and neurial knowledge of students (Martin, improving the economy. The government is McNally, and Kay 2013), and it has been indi- spending more than ever to promote and cated that there is a positive relationship encourage entrepreneurship and innovation. between entrepreneurial knowledge and iden- Accordingly, measures and mechanisms have

6192 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT been proposed to develop entrepreneurship ties was the objective of covering a wide range in the public and private sectors as well as in of different class characteristics and of different universities. The first official step was taken in rankings of Iranian universities. As not all the 2000 with the establishment of a comprehen- students in the university were allowed to take sive program for entrepreneurship develop- entrepreneurship courses, respondents for our ment in universities, called KARAD, as part of questionnaire were selected on a purposive the Third Economic and Social Development basis. The students surveyed were told that the Program. The main goals of KARAD were to questionnaires were for research purposes only promote an entrepreneurial spirit and culture and that their answers would not affect their in academic communities; familiarize students curriculum in any way; participation was with entrepreneurship as a career choice and as always presented as a voluntary choice. In the specific facets aimed to encourage and train first survey (t1), 275 students participated them on how to prepare a business plan; and (response rate of 86 percent), and in the second start and manage a new business. To achieve survey (t2), 240 students (response rate of 75 this goal, several programs and strategies were percent). We were able to match the two ques- considered including establishing entrepre- tionnaires (at t1 and at t2) for 205 students. neurship centers and introducing entrepreneur- These represent 64 percent of total enrollment ship courses such as “fundamentals of in the entrepreneurship courses at the selected entrepreneurship” into undergraduate educa- universities. The sample consisted of 86 male tion (Karimi et al. 2010). students (42 percent) and 119 female students “Fundamentals of entrepreneurship” as a (58 percent), with ages ranging from 19 to 31, compulsory or elective course is taught to with a mean of 22.08 years. There is a greater undergraduate students in their last two years proportion of females in the sample because of college in various faculties/departments. It more females than males enroll in the degrees aims to increase university graduates’ knowl- where the data were collected. There was no edge about entrepreneurship, influencing their control group; only students participating in entrepreneurial attitudes and intentions, and the course filled out the two questionnaires. In encourage them to be job creators rather than general terms, the breakdown of the sample job seekers. According to Linan’s (2004) EEP according to college major is as follows: agri- categorization, these criteria allow the course in cultural sciences (49.8 percent), engineering which this study’s survey was conducted to be sciences (21.5 percent), Humanistic Science classified in the category of “entrepreneurial (21.5 percent), and Basic Sciences (7.2 awareness education.” Although the course percent). description is almost the same at every university, educators might use various teaching Measurement of Variables. All construct mea- materials and methods for this course. The sures were adopted from existing scales. All methods most often employed are lectures, items (aside from demographic characteristics) readings, class discussion, business plans, case were measured using a seven-point Likert scale studies, and guest speakers. ranging from 1 representing “strongly disagree” to 7 representing “strongly agree.” These items Participants and Procedures. During the and the sources from which the items were 2010–2011 academic year, an ex ante and ex adopted are summarized in Table 1. Several post survey was used to measure the change in control variables were used in the study: age, student EI and opportunity identification per- gender (coded as 1 = male and 0 = female), uni- ceptions over approximately a four-month versity ranking (coded as 3 = high ranking, period in “fundamentals of entrepreneurship” 2 = intermediate ranking, and 1 = low ranking), courses at six Iranian universities. Our research university (categorical variable for the six used a quantitative method, including a ques- selected universities), and academic major (cat- tionnaire that was handed out at the beginning egorical variable for the four academic majors). of the first session (t1) and at the end of the final session (t2) of the courses. Undergraduate Statistical Analysis. The obtained data were students who enrolled in the entrepreneurship analyzed using SPSS 18 (SPSS Inc., Chicago, courses at six Iranian public universities served Illinois, USA) and AMOS 18 (IBM, New York, as the sample for the study (n = 320). The USA). As a first step, an exploratory factor reason for including several different universi- analysis (EFA) was performed on the items.

KARIMI ET ET AL. 193 7 8 194 Table 1 Details, Reliability, and Validity of the Measures

Construct Research Reference No. of α CR AVE Item Pre Post Pre Post Pre Post

ORA FSALBSNS MANAGEMENT BUSINESS SMALL OF JOURNAL Entrepreneurial Linan and Chen (2009) (e.g., “I have very seriously thought 6 0.84 0.85 0.89 0.90 0.50 0.52 ORA FSALBSNS MANAGEMENT BUSINESS SMALL OF JOURNAL Intentions of starting a firm”) Attitude toward Linan and Chen (2009) (e.g., “Being an entrepreneur implies 5 0.78 0.85 0.86 0.91 0.55 0.66 Entrepreneurship more advantages than disadvantages to me”). Subjective Norm Adopted from Kolvereid (1996), which has been used in 6 0.82 0.91 0.90 0.95 0.58 0.74 Kolvereid and Isaksen (2006), Krueger, Reilly, and Carsrud (2000), and Souitaris, Zerbinati, and Al-Laham (2007). This scale included two separate questions: belief (e.g., “I believe that my closest family thinks that I should start my own business”) and motivation to comply (e.g., “I care about my closest family’s opinion with regard to me starting my own business”). The belief items were recoded into a bipolar scale (from −3 to +3) and multiplied with the respective motivation-to-comply items. The subjective norm variable was calculated by adding the three results and dividing the total score by three. Perceived Behavioral Linan and Chen (2009) (e.g., “Starting a firm and keeping it 6 0.88 0.88 0.93 0.93 0.60 0.61 Control viable would be easy for me.”) Opportunity Selected from the literature on opportunity identification 9 0.83 0.81 0.89 0.88 0.46 0.42 Identification (Hills 1995; Nicolaou et al. 2009; Ozgen and Baron 2007; Perception Singh et al. 1999; Ucbasaran et al. 2003), gauging both the self-perceived ability to recognize opportunities (e.g., “I am able to recognize new business opportunities in the market”) and alertness to opportunities when they exist (“I have a special alertness or sensitivity toward business opportunities in my environment”)

AVE, average variance extracted; CR, construct reliability. EFA helps explain the variability among observ- exceed 6.117 (p < .01), and all loadings are able variables and thus served to eliminate more than 0.5. Moreover, all constructs had a problematic items with significant cross load- construct reliability (CR) value, ranging from ings or loading to the wrong factor; items 0.86 to 0.95, higher than the recommended remaining after this filtering exercise were level of 0.70. With respect to the AVE estimate, selected to build each of the constructs used in the results revealed that the AVE estimate for the structural equation modeling (SEM) in the all constructs is above or close to the recom- second step. SEM was employed to define the mended threshold of 0.50 (Table 1). Discrimi- relationship between EI and its antecedents nant validity was assessed by comparing the (H1) and to test the relationships between PBC, square root of the AVE for a given construct ATE, OIP, and EI (H2 and H3). Furthermore, with the correlations between that construct the paired samples t-test was used to test the and all other constructs. The square roots of the impact of the programs on the students’ entre- AVE of each construct, listed on the diagonal of preneurial attitudes, OIP, and intentions (H4 Table 2, all exceed the correlation shared and H5). Finally, the independent samples between the construct and other constructs in t-test was utilized to compare the effects of the model, indicating adequate discriminant elective and compulsory courses (H6). validity between each construct.

Results The Assessment of the Structural Model SEM With the construct validity and reliability The SEM approach was used to validate the measures established, all the constructs were research model and test the effects in the used as input to form a structural model repre- hypotheses. According to Hair et al. (2006), it is senting the hypothesized model depicted in appropriate to adopt a two-step approach in Figure 1. As shown in Figure 2, the overall SEM: (1) the assessment of the measurement goodness-of-fit statistics show that the structural model and (2) the assessment of the structural model fits the pretest and post-test data well. model. Having assessed the fit indices for the measurement models and structural models, the esti- The Assessment of the mated coefficients of the causal relationships Measurement Model between constructs were examined. Table 3 The first step, involving confirmatory factor shows the coefficient of each hypothesized path analysis, was to test the goodness-of-fit indices, and its corresponding CR (known as the and the reliability and validity of the proposed t-value). It can be seen from this table that the measurement model. The measurement model predictive positive effect of SN on EI is sup- includes 23 items describing five latent con- ported (pretest: β = 0.22, CR = 3.299, p < .001; structs: ATE, SN, PBC, OIP, and EI. Goodness- post-test: β = 0.20, CR = 3.056, p < .01), an effect of-fit indicators suggest a very good fit of the which corresponds to H1a. H1b is also sup- proposed model for the pretest (χ2 = 284.432, ported: ATE has a positive effect on EI p = .001; χ2/df = 1.323; GFI = 0.893; TLI = 0.962; (pretest: β = 0.28, CR = 3.969, p < .001; post-test: CFI = 968; IFI = 0.968; RMSEA = 0.04) and post- β = 0.30, CR = 4.078, p < .001). As the PBC test data (χ2 = 278.022, p = .003; χ2/df = 1.287; also has a significant effect on EI (pretest: GFI = 0.898; TLI = 0.972; CFI = 0.976; IFI = β = 0.45, CR = 5.684, p < .001; post-test: 0.977; RMSEA = 0.038). Therefore, on the basis β = 0.47, CR = 5.212, p < .001), H1c is sup- of the results obtained, the hypothesized model ported. The results also show that OIP positively of five constructs is a suitable measurement influence EI (pretest: β = 0.22, CR = 3.169, model for this study. p < .01; post-test: β = 0.14, CR = 1.970, p < .05), The convergent and discriminant validities supporting H2. H3a and H3b presume that ATE of the constructs can be assessed by referring to and PBC would influence OIP. As hypothesized, the measurement model. According to Fornell the estimate of the paths coefficients of ATE and Larcker (1981), convergent validity is (pretest: β = 0.20, CR = 2.261, p < .05; post- evaluated for the measurement model based on test: β = 0.21, CR = 2.414, p < .05) and PBC three criteria: (1) factor loadings; (2) the scale (pretest: β = 0.31, CR = 3.636, p < .001; post-test: composite or construct reliability (CR); and (3) β = 0.34, CR = 3.481, p < .001) on OIP was posi- the average variance extracted (AVE). The find- tive and statistically significant, which provided ings showed that all items’ critical ratio values support for H3a and H3b. Overall, the TPB

KARIMI ET ET AL. 195 9 10 196

Table 2 The Correlation Matrix and Discriminant Validity

Variable Mean S.D. 123456789101112131415161718 ORA FSALBSNS MANAGEMENT BUSINESS SMALL OF JOURNAL ORA FSALBSNS MANAGEMENT BUSINESS SMALL OF JOURNAL

EI (t1) 4.85 1.43 (0.71) ATE (t1) 5.13 0.953 0.33** (0.74) SN (t1) 2.25 5.67 0.36** 0.11 (0.76) PBC (t1) 4.35 1.32 0.60** 0.21** 0.24** (0.77) OIP (t1) 4.31 1.15 0.43** 0.25** 0.15* 0.32** (0.69) EI (t2) 5.06 1.31 0.47** 0.13 0.25** 0.31** 0.28** (0.72) ATE (t2) 5.22 1.04 0.25** 0.32** 0.16* 0.17* 0.21* 0.57** (0.81) SN (t2) 4.07 7.07 0.24** 0.13 0.34** 0.17* 0.18* 0.43** 0.30** (0.86) PBC (t2) 4.68 1.27 0.38** 0.12 0.09 0.40** 0.21* 0.67** 0.47** 0.42** (0.78) OIP (t2) 4.38 0.954 0.29** 0.08 0.12 0.23** 0.35** 0.42** 0.34** 0.23** 0.41** (0.65) EI (t2-t1) 0.213 1.66 −0.57** −0.21* −0.13 −0.32** −0.18* 0.46** 0.28** 0.16* 0.24** 0.10 ATE (t2-t1) 0.083 1.31 −0.05 −0.54** 0.06 −0.02 −0.02 0.40** 0.64** 0.16* 0.32** 0.24** 0.42** SN (t2-t1) 1.82 7.86 −0.04 0.05 −0.44** −0.02 0.06 0.22** 0.16* 0.69** 0.33** 0.13 0.25** 0.10 PBC (t2-t1) 0.337 1.65 −0.22** −0.09 −0.14* −0.57** −0.12 0.32** 0.26** 0.22** 0.53** 0.16* 0.52** 0.35** 0.32** OIP (t2-t1) 0.074 1.41 −0.18* −0.18 −0.05 −0.12 −0.66** 0.07 0.07 0.01 13 0.47** 0.25** 0.21** 0.04 0.23** Age 22.08 1.72 0.15* 0.11 0.02 0.07 0.01 0.08 −0.03 0.05 0.06 −0.02 −0.07 −0.10 0.03 −0.02 −0.03 Gender 0.42 0.49 0.06 −0.22** −0.07 0.08 0.04 −0.09 −0.08 −0.04 −0.01 0.13 −0.12 0.10 0.02 −0.07 0.06 0.05 Selection 0.37 0.46 0.04 0.09 0.02 0.11 0.02 0.22** 0.07 0.08 0.13 0.09 0.14* −0.02 0.07 0.02 0.05 −0.30** −0.20* Ranking 2.14 0.92 −0.09 −0.03 −0.01 −0.06 −0.04 0.15* 0.03 0.11 0.24* 0.17* 0.10 0.04 0.11 0.10 0.12 −0.22** −0.06 0.22**

*p < .05. **p < .01. n = 205; Two-tailed tests of signiﬁcance were used. The square roots of AVE estimates are in bold on the diagonal. ATE, attitude toward entrepreneurship; AVE, average variance extracted; EI, entrepreneurial intention; OIP, opportunity identiﬁcation perception; PBC, perceived behavioral control; S.D., standard deviation; SN, subjective norms. Figure 1 The Proposed Research Model

SN H1a H4a

H4b H1b EEPs ATE EI

H4c H1c H2 H3a PBC H3b OIP

ATE, Attitude toward Entrepreneurship; EEP, Entrepreneurship Education Programs; EI, Entrepreneurial Intention; OIP, Opportunity Identiﬁcation Perception; PBC, Perceived Behavioral Control; SN, Subjective Norms.

Figure 2 The Proposed Research Model

SN H1a = 0.22/0.20 H4a R2 = 0.60/0.63 H1b = 0.28/0.300 EEPs ATE EI H4b H1c = 0.45/0.477

H4c H2 = 0.22/0.14 H3a = 0.20/0.22 PBC H3b = 0.31/0.34 OIP

R2 = 0.18/0.24 H5

Pretest/Post-Test; ATE, Attitude toward Entrepreneurship; EEP, Entrepreneurship Educa- tion Programs; EI, Entrepreneurial Intention; OIP, Opportunity Identification Perception; PBC, Perceived Behavioral Control; SN, Subjective Norms. Goodness-of-Fit Indices (Pretest): χ2 = 284.862; χ2/df = 1.319; GFI = 0.893; TLI = 0.963; CFI = 0.968; IFI = 0.969; RMSEA = 0.040. Goodness-of-Fit Indices (Post-Test): χ2 = 278.125; χ2/df = 1.282; GFI = 0.897; TLI = 0.973; CFI = 0.977; IFI = 0.977; RMSEA = 0.037. model explained respectively 60 and 63 percent and university ranking did not have significant of the variance in the EI in the pretest and correlations with the difference values of ATE, 2 2 post-test samples (R pretest = 0.60; R post-test = 0.63). SN, PBC, EI, and OIP (Table 2). The GLM results To test the relationships between the control also showed no significant differences in ATE, variables and the change in ATE, SN, PBC, EI SN, PBC, EI, and OIP, controlling for the cat- and OIP, a correlation and a general linear egorical variables (university and academic model (GLM) procedure were employed. The major), suggesting that the findings of this study results of correlation indicated that age, gender, were not affected by these control variables. In

KARIMI ET ET AL. 197 11 12 198

Table 3 Results of the Structural Equation Modeling ORA FSALBSNS MANAGEMENT BUSINESS SMALL OF JOURNAL ORA FSALBSNS MANAGEMENT BUSINESS SMALL OF JOURNAL Hypotheses Tested Estimate S.E.a CRb p (β Value) (t-Value)

Model at t1 H1a: Subjective Norm → Entrepreneurial Intention 0.22 0.014 3.299 0.000** H1b: Attitude toward Entrepreneurship → Entrepreneurial Intention 0.28 0.191 3.969 0.000** H1c: Perceived Behavioral Control → Entrepreneurial Intention 0.45 0.071 5.684 0.000** H2: Opportunity Identification → Entrepreneurial Intention 0.22 0.079 3.196 0.001** H3a: Attitude toward Entrepreneurship → Opportunity Identification 0.20 0.186 2.261 0.024* H3b: Perceived Behavioral Control → Opportunity Identification 0.31 0.066 3.636 0.000** Model at t2 H1a: Subjective Norm → Entrepreneurial Intention 0.20 0.012 3.056 0.002** H1b: Attitude toward Entrepreneurship → Entrepreneurial Intention 0.30 0.084 4.078 0.000** H1c: Perceived Behavioral Control → Entrepreneurial Intention 0.47 0.096 5.212 0.000** H2: Opportunity Identification → Entrepreneurial Intention 0.14 0.097 1.970 0.049* H3a: Attitude toward Entrepreneurship → Opportunity Identification 0.22 0.075 2.414 0.016* H3b: Perceived Behavioral Control → Opportunity Identification 0.34 0.074 3.481 0.000**

*p < .05. **p < .01. aS.E. is an estimate of the standard error of the covariance. bC.R. is the critical ratio obtained by dividing the covariance estimate by its standard error. Table 4 Results of Paired t-Test for the Program Impacts (n = 205)

Scale Pretest Post-Test Difference

M S.D. M S.D. t(204) p

EI 4.85 1.43 5.06 1.32 1.83 0.068 SN 2.25 5.67 4.08 7.07 3.28 0.001* ATE 5.13 0.95 5.22 1.04 0.90 0.367 PBC 4.35 1.32 4.68 1.28 2.92 0.004* OIP 4.31 1.15 4.38 0.97 0.75 0.453

*p < .01. ATE, attitude toward entrepreneurship; EI, entrepreneurial intention; OIP, opportunity identification perception; PBC, perceived behavioral control; S.D., standard deviation; SN, subjective norms. order to test H4e, we employed a correlation for EI, ATE, and OIP. The results therefore analysis, as summarized in Table 2. As expected, demonstrate that there are positive and signifi- a change in SN, ATE, PBC, and OIP was significant differences in pre and post-test values of cantly related to an increased intention to start SN and PBC, confirming H4a and H4c; one’s own business. Therefore, H5e was however, there are not significant differences in accepted. pre and post-test values of ATE, OIP, and EI, rejecting H4b, H4d, and H5. Impact of EEPs on Students. In order to assess the impacts of the entrepreneurship courses on Differences in EEP Impacts in Relation to the the students’ entrepreneurial attitudes, inten- Selection Mode. In order to examine whether tions, and OIP, we conducted the paired attitudes, intention, and opportunity identifica- samples t-test. Table 4 summarizes the results tion change are equally likely for the two types of this test. The results showed a positive and of EEPs (elective versus compulsory), we com- significant difference in the pretest (M = 2.25) pared the effects of these different programs by and post-test value (M = 4.08) of SN (t = 3.28, using the independent samples t-test. For each p = .001 < .01). The significant difference student, a gain score was calculated for each of between the pretest (M = 4.35) and post-test the five scales, which consisted of the student’s data (M = 4.68) was also evident for PBC score on the scale in the post-test survey minus (t = 2.92, p = .004 < .01). However, the mean his/her score on the same scale in the pretest score of ATE in the pretest sample (M = 5.13) survey. As can be seen in Table 5, in the pretest was not significantly different from the mean sample, the students in elective courses exhib- score in the post-test sample (M = 5.22) ited higher scores on all five scales compared (t = 0.904, p = .367 > .05). In addition, for OIP, with the students in compulsory courses, but the mean score in the pretest sample (M = 4.31) none of these differences is statistically signifi- was not significantly different from that in the cant. In the post-test sample, the two groups post-test sample (M = 4.38). The results also differed significantly in their EI, such that the revealed that the post-test value of EI students in the elective courses have greater EI (M = 5.06) was increased compared with than the students in the compulsory courses. the pretest value (M = 4.851), though this The elective courses had a significantly greater increase was not very significant (t = 1.83, positive impact on the students’ EI as the gain p = .068 > .05). The GLM procedure of ANOVA in EI was significantly higher for the students in also indicated significant differences between the elective courses than for the students in the the pre and post-test values for SN (F = 10.77, compulsory courses. The results of the paired p = .001) and PBC (F = 8.51, p = .004) but not samples t-test (Table 6) also showed significant

KARIMI ET ET AL. 199 13 14 200

Table 5 ORA FSALBSNS MANAGEMENT BUSINESS SMALL OF JOURNAL ORA FSALBSNS MANAGEMENT BUSINESS SMALL OF JOURNAL Differences in the EEP Impacts According to Selection Mode (Compulsory versus Elective)

Scale Pretest Post-Test Gain

Compulsory Elective Difference Compulsory Elective Difference Compulsory Elective Difference (n = 127) (n = 78) (n = 127) (n = 78) (n = 127) (n = 78)

M S.D. M S.D. t(203) p M S.D. M S.D. t(203) p M S.D. M S.D. t(203) p

EI 4.80 1.39 4.93 1.50 −0.59 0.550 4.84 1.33 5.44 1.22 −3.23 0.001* 0.03 1.67 0.51 1.59 −2.01 0.046* SN 2.19 5.77 2.35 5.53 −0.19 0.844 3.65 4.06 4.77 7.08 −1.10 0.272 1.46 8.21 2.42 7.54 −0.84 0.403 ATE 5.07 0.96 5.24 0.93 −1.25 0.212 5.16 1.04 5.31 1.04 −1.05 0.297 0.09 1.32 0.07 1.32 0.08 0.938 PBC 4.24 1.27 4.52 1.39 −1.52 0.131 4.55 1.28 4.89 1.25 −1.84 0.068 0.32 1.70 0.37 1.57 −0.20 0.839 OIP 4.30 1.16 4.33 1.15 −0.18 0.861 4.32 0.99 4.49 0.93 −1.28 0.203 0.02 1.41 0.17 1.40 −0.74 0.462

*p < .05. **p < .01. ATE, attitude toward entrepreneurship; EI, entrepreneurial intention; OIP, opportunity identiﬁcation perception; PBC, perceived behavioral control; S.D., standard deviation; SN, subjective norms. Table 6 Results of Paired t-Test for the Impacts of Elective and Compulsory Programs

Scale Compulsory (n = 127) Elective (n = 78)

Pretest Post-test Difference Pretest Post-test Difference

M S.D. M S.D. t p M S.D. M S.D. t p

EI 4.80 1.39 4.84 1.33 0.21 0.833 4.93 1.50 5.44 1.22 2.80 0.006** SN 2.19 5.78 3.65 7.06 2.00 0.047* 2.35 5.53 4.77 7.08 2.83 0.006** ATE 5.07 0.96 5.16 1.04 0.76 0.450 5.24 0.93 5.31 1.01 0.49 0.622 PBC 4.24 1.27 4.55 1.28 2.10 0.037* 4.52 1.39 4.89 1.25 2.06 0.043* OIP 4.30 1.16 4.32 0.99 0.14 0.892 4.33 1.15 4.49 0.93 1.05 0.298

*p < .05. **p < .01. ATE, attitude toward entrepreneurship; EI, entrepreneurial intention; OIP, opportunity identification perception; PBC, perceived behavioral control; S.D., standard deviation; SN, subjective norms. differences in pre and post-values of EI, SN, viduals’ intentions to start a new business and PBC for the elective courses, but for the (Souitaris, Zerbinati, and Al-Laham 2007). compulsory courses, they showed significant Notably, the comparison of elective and com- differences only in pre- and post-values of SN pulsory EEPs indicated that intention change is and PBC. not equally distributed across these programs. The elective EEPs had a significantly greater Discussion positive impact on students’ EI. Moreover, this The purpose of this study was to assess the study could not find a significant effect of impact of EEPs on students’ EI, drawing on the either elective or compulsory EEPs on ATE: TPB. Moreover, the proposed model incorpo- The programs failed in developing students’ rates the perception of opportunity identifica- ATE. This finding is in line with the results of tion into the TPB. To address this purpose, we Souitaris, Zerbinati, and Al-Laham (2007), and employed an ex ante and ex post survey, with Weber (2012), but it is not consistent with the 205 participants in elective and compulsory findings of Peterman and Kennedy (2003). EEPs at six Iranian universities. Contrary to our expectation, neither type of The findings were in line with earlier EEP led to a significant increase in OIP, which studies on the effects of EEPs but nevertheless contradicts the results of DeTienne and also present some differences. We found con- Chandler (2004). firmation for the impact of (both types of) EEPs The significant increase in the mean value of on SN (Souitaris, Zerbinati, and Al-Laham 2007; SN may reflect the emphasis within both pro- Weber 2012). For both voluntary and compul- grams on teamwork (e.g., working together in sory EEPs, the postprogram mean value of PBC teams of four to six to create business plans) was increased in relation to the preprogram and on giving students the opportunity to build value (Peterman and Kennedy 2003; Weber a network with entrepreneurially-minded 2012), something that Souitaris, Zerbinati, and friends and peers, and with entrepreneurs. A Al-Laham (2007) were not able to confirm. possible explanation for the increase in PBC However, this study did not provide evidence could be related to mastery experience and that EEPs have a significant effect on students’ vicarious experience (role modeling), which EI in the sample as a whole. This conflicts with might be gained by the students during the the idea that participating in EEPs fosters indi- programs. Most EEPs try to emphasize the

KARIMI ET ET AL. 201 15 “learning-by-doing” component (such as less emphasis during the courses. Neck and writing a business plan and field work) and to Greene (2011) point out that the majority of expose the students to the real world. In addi- entrepreneurship courses are focused on the tion, the teachers tell success stories about exploitation of opportunities and assume that entrepreneurs or invite guest entrepreneurs as the opportunity has already been identified. speakers who can serve as successful role Where this is the case, very little time and models for students. attention are given to creativity, the idea gen- The reason for the lack of a significant effect eration process, and how to identify new busi- of EEPs on ATE is not fully clear, and this ness opportunities. warrants future research. A few possibilities are explored here. The first plausible explanation is Implications that the students had relatively high scores for Theoretical Implications. This study has this variable at the beginning of the program, several theoretical implications. It provides so there was not much room left for improving further supporting evidence for the application their attitudes. It should be noted that small of the TPB in predicting and understanding EI differences in the mean do not imply that there in non-Western countries such as Iran. Further- is no change at all in these variables. Another more, this study contributes to the TPB by explanation could be related to the program examining the effect of entrepreneurship edu- design. EEPs may have not been designed suf- cation as an exogenous influence on EI and its ficiently well with regard to persuasion and antecedents, and it shows that the TPB can attitude change. provide a useful framework to assess the effec- The effects of compulsory EEPs on EI may tiveness of EEPs. In addition, this study devel- have been insignificant because participation ops and extends the TPB model by was compulsory, as the comparison analysis incorporating the OIP as a proximal cause of showed. A second possibility is that students EI, and it examines the relationship between may have gained a realistic picture of both this variable and EI and its antecedents. themselves and being an entrepreneur and decided, in this light, that they do not want to Practical Implications. In terms of practice, become an entrepreneur. In this sense, we the study provides valuable information and cannot say that the programs did not affect insight for those who formulate, deliver, and students’ EI; the programs may have enhanced evaluate educational programs aimed at the awareness of entrepreneurship among increasing the EI of students. The findings indi- these students and led them to assess their cate that PBC is the strongest predictor of EI, future as entrepreneurs. A similar explanation and as this study confirmed, PBC can be fos- was provided by Oosterbeek, van Praag, and tered through EEPs. Therefore, educators Ijsselstein (2010), who argue that the reason should focus more on the use of appropriate may have been that some participants had lost teaching methods in order to enhance students’ their excessive optimism about entrepreneur- PBC more effectively. According to Bandura ship and rejected the idea of becoming an (1997), an individual’s sense of self-efficacy can entrepreneur after the program had finished. be built and strengthened in four ways: mastery von Graevenitz, Harhoff, and Weber (2010) also experience or repeated performance accom- argue that EEPs provide individuals with plishments, vicarious experience or modeling, signals about their entrepreneurial ability and social persuasion, and judgments of one’s own aptitude. As a result, some students may physiological states, such as arousal and become aware that they are not well suited for anxiety. Entrepreneurship education can play a entrepreneurship. significant role in developing students’ entre- With respect to opportunity identification, preneurial self-efficacy in these ways by apply- one explanation for this result could be related ing the educational activities and teaching to the fact that despite the emphasis of EEPs on methods presented next (Segal, Schoenfeld, opportunity identification, most teachers did and Borgia 2007). Our findings strongly not pay the necessary attention to fostering this suggest that participation in both elective and competency in their classes. The results of compulsory EEPs can positively influence stu- interviews with some students and teachers dents’ PBC or self-efficacy, confirming that uni- after the post-test measurement indicated that versities can shape and foster entrepreneurial this competency was often ignored or received self-efficacy through EEPs.

16202 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Educational activities providing “real-world” appears to play a significant role. Therefore, it experience or “virtual reality” experiences in is suggested that teaching methods and con- the classroom, including the use of role tents specifically designed to improve SN playing, case methods, and business simula- should be included in EEPs. SN can be tions, facilitate the development of decision- improved by means of teamwork and by pro- making skills and strengthen entrepreneurial viding opportunities for students to build a self-confidence through mastery experiences network with entrepreneurial-minded friends or repeated performance accomplishments. and peers, and with role models and entrepre- Vicarious learning can be increased through neurs (Mueller 2011; Souitaris, Zerbinati, and educational activities such as successful entre- Al-Laham 2007; Weber 2012). It was concluded preneurs as guest speakers, video profiles of that EEPs did not influence ATE because the well-known entrepreneurs, case studies, mean score of this variable was high at the student internships, and participation in busi- beginning of EEPs. Therefore, we can suggest ness plan competitions. Encouraging com- that if an EEP has attendees who are already ments, positive feedback, and praise from—and highly motivated about entrepreneurship and persuasive discussions with—teachers and pro- have high attitudes and EI, the aim of such a fessionals in educational programs can increase program should be “education for start-up” self-efficacy through social persuasion. These rather than “entrepreneurial awareness educa- activities can also reduce stress levels and tion” (according to the classification by Linan anxiety. 2004). As discussed earlier, the objective of the In particular, the findings suggest that uni- latter program is to provide information for versities can develop students’ EI through elec- students about entrepreneurship so that they tive rather than compulsory EEPs. Therefore, consider entrepreneurship as a possible and educators should differentiate between com- alternative choice of career. The former pulsory courses offered to all students and program aims at the preparation of individuals courses offered as electives for students who for running conventional small businesses and are interested in entrepreneurship. According focuses on the practical aspects related to the to von Graevenitz, Harhoff, and Weber (2010) creation of a new business, such as how to and Oosterbeek, van Praag, and Ijsselstein obtain financing, legal regulations, and taxa- (2010), the primary aim for compulsory pro- tion (Curran and Stanworth 1989). Entrepre- grams, with a mix of participants interested in neurial awareness education can be offered as entrepreneurship and participants who are a compulsory or elective program, whereas uninterested, is a sorting effect: Students education for start-up can be offered only as an attending these programs become informed elective. about entrepreneurship as an alternative career Although opportunity identification is choice and gain more realistic perspectives, the core of entrepreneurship (Shane and regarding both themselves and what it takes to Venkataraman 2000), very little is done to train be an entrepreneur. Therefore, after complet- a student in how discover or create new busi- ing EEPs, some students will learn that they are ness opportunities (Neck and Greene 2011). well suited for entrepreneurship and be The findings of the present study also showed strengthened in their decision to become entre- that the EEPs did not significantly influence preneurs, whereas others will learn that they students’ perception of opportunity identifica- are not. In elective courses, on the other hand, tion. Therefore, in both compulsory and elec- self-selection will lead to a higher level of EI tive courses, enhancing this competency and increase the likelihood of participants should be a particularly important component becoming entrepreneurs. of entrepreneurship education and a funda- The findings also showed that SN influences mental design principle in EEPs. In addition, EI, and we can improve SN through EEPs. students should be equipped with tools Some previous studies (e.g., Linan and Chen enabling them to find opportunities and make 2009) found that SN also has a relevant effect opportunities (Neck and Greene 2011; on EI through ATE and PBC. In particular, Sarasvathy 2008). Lumpkin, Hills, and Shrader in a collectivistic culture such as Iran where (2004) argue that teaching creativity skills family life and relationships with close friends can enhance opportunity identification compe- and relatives are important (Javidan and tency. According to Sardeshmukh and Dastmalchian 2003; Karimi et al. 2013b), SN Smith-Nelson (2011), students’ ability to

KARIMI ET ET AL. 203 17 identify new business opportunities can be EEPs that accommodate these different cul- enhanced by means of a combination of class- tural values. room activities (e.g., activities associated with creativity such as divergent thinking and idea Limitations and Future Research generation exercises) and experiential activities The current study has several limitations that (e.g., internships and engaging students with provide future research opportunities. This community entrepreneurs by way of guest lec- study assessed only the effects of participating tures by entrepreneurs, mentoring by local in the EEPs on intention and opportunity iden- entrepreneurs, and live case studies). Another tification; future research should examine the factor which needs to be considered by edu- specific characteristics, design elements, con- cators is the role of networking in identifying tents, and teaching approaches of the EEPs, opportunities. Social networks have been and their relationships to these outcomes. found to be important in the opportunity iden- Future researchers may also address the ques- tification process (Ozgen and Baron 2007). tion of why the EEPs foster PBC and subjective Social networks are significant sources of norms but not attitude toward entrepreneur- knowledge (Johannison 1990) and of new ship or opportunity identification. ideas (Christensen and Peterson 1990) and As we did not have control groups to have been associated with the number of per- compare with our treatment groups, we are ceived new opportunities (Hills 1995; Ozgen unable to determine the exact impact of EEPs and Baron 2007; Singh et al. 1999). Therefore, on students’ EI. We can assume that these sig- educators should devote attention to develop- nificant pretest/post-test differences are the ing students’ network skills in EEPs and should results of participating in EEPs because the give them more opportunities to network with content of the EEPs is very specific and not peers and other entrepreneurs (Lumpkin, Hills, duplicated in other courses; however, the avail- and Shrader 2004). ability of a control group would have strength- As mentioned already, policymakers and ened our findings. It should be noted that we university faculties should recognize the dif- did not want to conduct an artificial random- ferential effects of various types of EEPs and ized trial; we preferred a study in a naturalistic that the effects will not be the same across all academic setting that would not deprive any of programs. Although we cannot recommend the undergraduate students in that department one type over the other in general terms, of the potential benefits of participating in policymakers and instructors who want to EEPs. produce more and better entrepreneurs while Finally, future research should focus on subject to cost constraints should keep in the intention–behavior relationship as this mind that elective programs may yield crucial link has been studied even less than better results than compulsory programs. the one between antecedent attitudes and EIs. Policymakers and educators should also be Consequently, a longitudinal study is recom- aware that cultural context and values play an mended for future research to be able to important part in EEPs. Studies show that the capture the changes in entrepreneurial atti- Iranian culture has changed over the last four tudes and intention over time and the subse- decades (Tajaddini and Mujtaba 2011). For quent formation of entrepreneurial behavior instance, the recent study by Karimi et al. from intention. (2013b) reported that although Iranian students are relational and show great affection Conclusions toward family members, close friends and This paper aimed to investigate the impact relatives (high family collectivism), they also of EEPs on students’ EI and opportunity iden- tend to embrace individualistic values (such as tification using the TPB. The data support both personal success and autonomy) to a greater the measurement and the structural model. degree than the older generations. Javidan Our study indicated that the EEPs significantly and Dastmalchian (2003) also reported that influenced subjective norms and PBC but that the Iranian culture is a mix of family ties and these programs did not have significant connections and a high degree of individual- impacts on students’ attitude toward entrepre- ism, and it has strong orientations toward neurship and their perceptions of opportunity achievement and performance. Therefore, identification. The study also showed that the policymakers and educators should develop elective EEPs significantly increased students’

18204 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT EI but that this increase was not significant for Bagozzi, R. P., D. J. Moore, and L. Leone the compulsory EEPs. We recommend that (2004). “Self-Control and The Self-Regulation others investigate if our findings can be repli- of Dieting Decisions: The Role of Prefactual cated in different educational institutions and Attitudes, Subjective Norms, and Resistance EEPs, perhaps using designs comparing the to Temptation,” Basic Applied Social Psy- outcomes of EEP participants with those of chology 26(2–3), 199–213. nonparticipant groups. As noted earlier, future Bandura, A. (1997). Self-efficacy: The Exercise research might also assess whether different of Control. New York: W. H. Freeman and teaching methods and learning environments Company. would have different effects on the outcomes Baron, R. A., and M. D. Ensley (2006). “Oppor- and whether course educator differences such tunity Recognition as the Detection of Mean- as skills or academic background would influ- ingful Patterns: Evidence from Comparisons ence the outcomes. In conclusion, this research of Novice and Experienced Entrepreneurs,” provides evidence that EEPs are effective, but Management Science 52(9), 1331–1344. the current form needs improvement. It is Bechard, J.-P., and D. Gregoire (2005). “Entre- imperative that we begin to understand how to preneurship Education Research Revisited: improve EEP learning outcomes, especially The Case of Higher Education,” Academy of regarding opportunity identification. If we do Management Learning & Education 4(1), not tackle these issues, we may end up with 22–43. graduates who lack the abilities and knowl- Bhave, M. P. (1994). “A Process Model of Entre- edge needed in order to identify new business preneurial Venture Creation,” Journal of opportunities and, as a result, fail in the first Business Venturing 9(3), 223–242. step of the entrepreneurship process. We hope Bird, B. (1988). “Implementing Entrepreneurial that our study will encourage further explora- Ideas: The Case for Intention,” Academy of tion of the results of EEPs and that it may Management Review 13(3), 442–453. guide and inspire policymakers and entrepre- Casson, M. C., and N. Wadeson (2007). “The neurship educators alike to design and deliver Discovery of Opportunities: Extending the successful EEPs. Economic Theory of the Entrepreneur,” Small Business Economics 28(4), 285–300. References Christensen, P. S., and R. Peterson (1990). Ajzen, I. (1991). “The Theory of Planned “Opportunity Identification: Mapping the Behaviour,” Organizational Behaviour and Sources of New Venture Ideas,” Paper Pre- Human Decision Processes 50, 179–211. sented at the 10th Annual Babson Entrepre- ——— (2002). “Perceived Behavioral Control, neurship Research Conference, April. Self-Efficacy, Locus of Control, and the Denmark, Aarhus University Institute of Theory of Planned Behavior,” Journal of Management. Applied Social Psychology 32(4), 665– Conner, M., and C. J. Armitage (1998). “Extend- 683. doi: 10.1111/j.1559-1816.2002.tb00236. ing the Theory of Planned Behavior: A Alberti, A., S. Sciascia, and A. Poli (2004). Review and Avenues for Further Research,” “Entrepreneurship Education: Notes on an Journal of Applied Social Psychology 28(15), Ongoing Debate,” Paper Presented at 14th 1429–1464. Annual IntEnt Conference. University of Curran, J., and J. Stanworth (1989). “Education Napoli Federico II, Italy. and Training for Enterprise: Some Problems Alvarez, S. A., and J. B. Barney (2007). “Discov- o Classification, Evaluation, Policy and ery and creation: Alternative theories of Research,” International Small Business entrepreneurial action,” Strategic Entrepre- Journal 7(2), 11–22. neurship Journal 1(1–2), 11–26. doi: DeTienne, D., and G. Chandler (2004). “Oppor- 10.1002/sej.4. tunity Identification and its role in the Entre- Ardichvilia, A., R. Cardozob, and S. Ray (2003). preneurial Classroom: A Pedagogical “A Theory of Entrepreneurial Opportunity Approach and Empirical Test,” Academy of Identification and Development,” Journal of Management Learning and Education 3(3), Business Venturing 18, 105–123. 242–257. Athayde, R. (2009). “Measuring Enterprise Dutta, D., J. Li, and M. Merenda (2011). “Fos- Potential in Young People,” Entrepreneur- tering Entrepreneurship: Impact of Special- ship Theory and Practice 3(2), 481–500. ization and Diversity in Education,”

KARIMI ET ET AL. 205 19 International Entrepreneurship and Man- Gonzalez-Alvarez, N., and V. Solis-Rodriguez agement Journal 7(2), 163–179. (2011). “Discovery of Entrepreneurial Dutton, J. (1993). “The Making of Organiza- Opportunities: A Gender Perspective,” tional Opportunities: Interpretive Pathway Industrial Management & Data Systems to Organizational Change,” in Research in 111(5), 755–775. Organizational Behavior. Eds. B. Staw and Gregoire, D., D. Shepherd, and L. Lambert L. Cummings. Greenwich, CT: JAI, 15. (2010). “Measuring Opportunity-Recognition Dutton, J., and S. Jackson (1987). “Categorizing Beliefs: Illustrating and Validating an Experi- Strategic Issues: Links to Organizational mental Approach,” Organizational Research Action,” Academy of Management Review Methods 13(1), 114–145. 12(1), 76–90. Gwynne, P. (2008). “More Schools Teaching Edelman, L., and H. Yli-Renko (2010). “The Entrepreneurship,” Research Technology Impact of Environment and Entrepreneurial Management 51(2), 6–8. Perceptions on Venture-Creation Efforts: Hair, J. F., R. E. Anderson, R. L. Tatham, and W. Bridging the Discovery and Creation Views C. Black (2006). Multivariate Data Analysis, of Entrepreneurship,” Entrepreneurship: 6th ed. Upper Saddle River, NJ: Prentice-Hall Theory and Practice 34(5), 833–856. International, Inc. Engle, R. L., N. Dimitriadi, J. V. Gavidia, C. Harris, M. L., and S. G. Gibson (2008). Schlaegel, S. Delanoe, I. Alvarado, X. He, S. “Examining the Entrepreneurial Attitudes Buame, and B. Wolff (2010). “Entrepreneur- of US Business Students,” Educa- ial Intent: A Twelve-Country Evaluation of tion + Training 50(7), 568–581. doi: 10.1108/ Ajzen’s Model of Planned Behaviour,” Inter- 00400910810909036. national Journal of Entrepreneurial Haustein, S., and M. Hunecke (2007). “Reduced Behaviour and Research 16(1), 36–58. Use of Environmentally Friendly Modes of Falkang, J., and F. Alberti (2000). “The Assess- Transportation Caused by Perceived Mobil- ment of Entrepreneurship Education,” ity Necessities: An Extension of the Theory Industry and Higher Education 14(2), 101– of Planned Behavior,” Journal of Applied 108. Social Psychology 37(8), 1856–1883. Fayolle, A., and B. Gailly (2008). “From Craft to Henry, C., F. Hill, and C. Leitch (2005). “Entre- Science: Teaching Models and Learning Pro- preneurship Education and Training: Can cesses in Entrepreneurship Education,” Entrepreneurship Be Taught? Part II,” Edu- Journal of European Industrial Training cation + Training 47(3), 158–169. 32(7), 569–593. Hills, G. E. (1995). “Opportunity Recognition Fayolle, A., B. Gailly, and N. Lassas-Clerc by Successful Entrepreneurs: A Pilot Study,” (2006). “Assessing The Impact of Entrepre- in Frontiers in Entrepreneurship Research neurship Education Programs: A New Meth- 1995. Eds. W. D. Bygrave, B. J. Bird, S. odology,” Journal of European Industrial Birley, N. C. Churchill, M. Hay, R. H. Keeley, Training 30(9), 701–720. and W. E. Wetzel Jr. Wellesley, MA: Babson Finkle, T., and D. Deeds (2001). “Trends in the College, 105–117. Market for Entrepreneurship Faculty During Hsu, M. H., C. H. Yen, C. M. Chiu, and C. M. the Period 1989–1998,” Journal of Business Chang (2006). “A Longitudinal Investiga- Venturing 16(6), 613–630. tion of Continued Online Shopping Fornell, C., and D. F. Larcker (1981). “Evaluat- Behavior: An Extension of The Theory of ing Structural Equation Models with Unob- Planned Behavior,” International Journal servable Variables and Measurement Error,” of Human-Computer Studies 64(9), 889– Journal of Marketing Research 48, 39–50. 904. Gaglio, C. M., and J. A. Katz (2001). “The Psy- Iakovleva, T., L. Kolvereid, and U. Stephan chological Basis of Opportunity Identifica- (2011). “Entrepreneurial Intentions in Devel- tion: Entrepreneurial Alertness,” Small oping and Developed Countries,” Education Business Economics 16(2), 95–111. and Training 53(5), 353–370. Gibbs, S. R. (2009). “Exploring the Influence of Jackson, S. E., and J. E. Dutton (1988). “Dis- Task-Specific Self-Efficacy on Opportunity cerning Threats and Opportunities,” Admin- Recognition Perceptions and Behaviors,” istrative Science Quarterly 33(3), 370–387. Frontiers of Entrepreneurship Research Javidan, M., and A. Dastmalchian (2003). 29(6), 1–15. “Culture and Leadership in Iran: The Land of

20206 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Individual Achievers, Strong Family Ties and preneurship Theory and Practice 18(1), Powerful Elite,” Academy of Management 5–21. Executive 17(4), 127–142. ——— (2000). “The Cognitive Infrastructure of Johannison, B. (1990). “Economics of Opportunity Emergence,” Entrepreneurship Overview—Guiding the External Growth of Theory and Practice 24(3), 5–24. Small Firms,” International Small Business ——— (2003). “The Cognitive Psychology of Journal 9, 32–44. Entrepreneurship,” in Handbook of Entre- Jones, P., A. Jones, G. Packham, and C. Miller preneurial Research. Eds. Z. A. Acs and D. (2008). “Student Attitudes Towards Enter- B. Audretsch. London: Kluwer Law Interna- prise Education in Poland: A Positive tional, 105–140. Impact,” Education + Training 50(7), 597– Krueger, N., and D. Brazeal (1994). “Entrepre- 614. neurial Potential & Potential Entrepreneurs,” Karimi, S., M. Chizari, H. J. A. Biemans, and M. Entrepreneurship Theory and Practice 18(3), Mulder (2010). “Entrepreneurship Education 91–104. in Iranian Higher Education: The Current Krueger, N., and A. L. Carsrud (1993). State and Challenges,” European Journal of “Entrepreneurial Intentions: Applying the Scientific Research 48(1), 35–50. Theory of Planned Behaviour,” Entrepre- Karimi, S., H. J. A. Biemans, T. Lans, M. Chizari, neurship & Regional Development 5, 330– M. Mulder, and K. Naderi Mahdei (2013a). 351. “Understanding Role Models and Gender Krueger, N., and P. Dickson (1994). “How Influences on Entrepreneurial Intentions Believing in Ourselves Increases Risk- Among College Students,” Procedia—Social Taking: Self-Efficacy and Perceptions of and Behavioral Sciences 94, 204–214. Opportunity and Threat,” Decision Sciences Karimi, S., H. J. A. Biemans, T. Lans, M. Chizari, 25(3), 385–400. and M. Mulder (2013b). “Using the Theory Krueger, N., M. Reilly, and A. Carsrud (2000). of Planned Behavior to Understand the “Competing Models of Entrepreneurial Influence of Cultural Values on Students’ Intentions,” Journal of Business Venturing Entrepreneurial Intention,” Manuscript sub- 15, 411–432. mitted for publication. KuratKo, D. F. (2005). “The Emergence of Karimi, S., H. J. A. Biemans, T. Lans, M. Chizari, Entrepreneurship Education: Development, and M. Mulder (2014). “Effects of Role Trends and Challenges,” Entrepreneurship Models and Gender on Students’ Entrepre- Theory and Practice 29(5), 577–597. neurial Intentions,” European Journal of Lans, T., V. Popov, K. Oganisjana, and M. Täks Training and Development 38(8), 1– (2013). “Learning for Entrepreneurship in 15. Heterogeneous Groups: Experiences from Katz, J. A. (2003). “The Chronology and Intel- an International, Interdisciplinary Higher lectual Trajectory of American Entrepreneur- Education Student Programme,” TRAMES ship Education 1876–1999,” Journal of 17(4), 383–399. Business Venturing 18(2), 283–300. Linan, F. (2004). “Intention-Based Models of Kolvereid, L. (1996). “Predictions of Employ- Entrepreneurship Education,” Piccola ment Status Choice Intention,” Entrepre- Impresa/Small Business 3, 11–35. neurship Theory and Practice 21(1), 47– Linan, F., and Y. W. Chen (2009). “Develop- 57. ment and Cross-Cultural Application of a Kolvereid, L., and E. Isaksen (2006). “New Specific Instrument to Measure Entrepre- Business Start-Up and Subsequent Entry Into neurial Intentions,” Entrepreneurship Theory Self-Employment,” Journal of Business Ven- and Practice 33(3), 593–617. turing 21, 866–885. Linan, F., J. C. Rodríguez-Cohard, and J. M. Kolvereid, L., and Ø. Moen (1997). “Entrepre- Rueda-Cantuche (2011). “Factors Affecting neurship Among Business Graduates: Does a Entrepreneurial Intention Levels: A Role Major in Entrepreneurship Make a Differ- for Education,” International Entrepreneur- ence?,” Journal of European Industrial ship and Management Journal 7(2), 195– Training 21(4), 154. 218. Krueger, N. (1993). “The Impact of Prior Entre- Lumpkin, G. T., and B. B. Lichtenstein (2005). preneurial Exposure on Perceptions of New “The Role of Organizational Learning in the Venture Feasibility and Desirability,” Entre- Opportunity-Recognition Process,” Entrepre-

KARIMI ET ET AL. 207 21 neurship Theory and Practice 29(4), 451– Bivariate Genetics Perspective,” Organiza- 472. tional Behavior and Human Decision Pro- Lumpkin, G. T., G. E. Hills, and R. C. Shrader cesses 110(2), 108–117. doi: 10.1016/ (2004). “Opportunity Recognition,” in Entre- j.obhdp.2009.08.005. preneurship the Way Ahead. Ed. H. P. Oosterbeek, H., M. van Praag, and A. Ijsselstein Welsch. London: Routledge, 73–90. (2010). “The Impact of Entrepreneurship Martin, B. C., J. J. McNally, and M. J. Kay (2013). Education on Entrepreneurship Skills and “Examining the Formation of Human Capital Motivation,” European Economic Review in Entrepreneurship: A Meta-Analysis of 54(3), 442–454. Entrepreneurship Education Outcomes,” Ozgen, E., and R. A. Baron (2007). “Social Journal of Business Venturing 28(2), 211– Sources of Information in Opportunity 224. Recognition: Effects of Mentors, Industry Net- Matlay, H. (2005). “Entrepreneurship Education works, and Professional Forums,” Journal of in UK Business Schools: Conceptual, Contex- Business Venturing 22(2), 174–192. tual and Policy Considerations,” Journal of Perugini, M., and R. P. Bagozzi (2001). “The Small Business and Enterprise Development Role of Desires and Anticipated Emotions in 12(4), 627–643. Goal-Directed Behaviours: Broadening and Mentoor, E. R., and C. Friedrich (2007). Deepening the Theory of Planned Behav- “Is Entrepreneurial Education at South ior,” The British Journal of Social Psychology African University Successful? An Empirical 40, 79–98. Example,” Industry & Higher Education Peterman, N. E., and J. Kennedy (2003). “Enter- 21(3), 221–232. prise Education: Influencing Students’ Menzies, T. V., and J. C. Paradi (2003). “Entre- Perceptions of Entrepreneurship,” Entrepre- preneurship Education and Engineering neurship: Theory and Practice 28(2), 129– Students: Career Path and Business Perfor- 144. mance,” International Journal of Entrepre- Pittaway, L., and J. Cope (2007). “Entrepreneur- neurship and Innovation 6(2), 85–96. ship Education: A Systematic Review of the Mitchell, J. R., and D. A. Shepherd (2010). “To Evidence,” International Small Business Thine Own Self Be True: Images of Self, Journal 25(5), 479–510. Images of Opportunity, and Entrepreneurial Reitan, B. (1997a). “Entrepreneurial Potential Action,” Journal of Business Venturing 25, and Its Antecedents: Theoretical and Empiri- 138–154. cal Evidence,” Unpublished manuscript, Mitra, J., and H. Matlay (2004). “Entrepreneurial The Norwegian University of Science and and Vocational Education and Training: Technology. Lessons from Eastern and Central Europe,” ——— (1997b). “Where Do We Learn That Industry and Higher Education 18(1), Entrepreneurship Is Feasible, Desirable 53–69. and/or Profitable?—A Look at the Processes Mueller, S. (2011). “Increasing Entrepreneurial Leading to Entrepreneurial Potential,” paper Intention: Effective Entrepreneurship Course presented at the 1997 ICSB World Confer- Characteristics,” International Journal of ence, San Francisco, CA. Entrepreneurship and Small Business 13(1), Saks, N. T., and C. M. Gaglio (2002). “Can 55–74. Opportunity Identification Be Taught?,” Munoz, A., S. Mosey, and M. Binks (2011). Journal of Enterprising Culture 10(4), 313– “Developing Opportunity Identification 347. Capabilities in the Classroom: Visual Evi- Sarasvathy, S. D. (2008). Effectuation: Elements dence for Changing Mental Frames,” of Entrepreneurial Expertise. Cheltenham, Academy of Management Learning and UK: Edward Elgar. Education 10(2), 277–295. Sardeshmukh, S. R., and R. Smith-Nelson Neck, H. M., and P. G. Greene (2011). “Entre- (2011). “Educating for an Entrepreneurial preneurship Education: Known Worlds and Career: Developing Opportunity Recogni- New Frontiers,” Journal of Small Business tion Ability,” Australian Journal of Career Management 49(1), 55–70. Development 20(3), 47–55. Nicolaou, N., S. Shane, L. Cherkas, and T. D. Segal, G., J. Schoenfeld, and D. Borgia (2007). Spector (2009). “Opportunity Recognition “Which Classroom-Related Activities and the Tendency to Be an Entrepreneur: A Enhance Students’ Entrepreneurial Interests

22208 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT and Goals?: A Social Cognitive Career Theory Raise Entrepreneurial Intention of Science Perspective,” Academy of Entrepreneurship and Engineering Students? The Effect of Journal 13(2), 79–98. Learning, Inspiration and Resources,” Shane, S. (2000). “Prior Knowledge and the Journal of Business Venturing 22(4), 566– Discovery of Entrepreneurial Opportuni- 591. doi: 10.1016/j.jbusvent.2006.05.002. ties,” Organization Science 11(4), 448–469. Tajaddini, R., and B. G. Mujtaba (2011). “Stress Shane, S., and S. Venkataraman (2000). “The and Leadership Tendencies of Respondents Promise of Entrepreneurship as a Field of from Iran: Exploring Similarities and Differ- Research,” Academy of Management Review ences based on Age and Gender,” Public 25, 217–226. Organization Review 11(3), 219–236. Shapero, A., and L. Sokol (1982). “The Social Ucbasaran, D., P. Westhead, M. Wright, and M. Dimensions of Entrepreneurship,” in The Binks (2003). “Does Entrepreneurial Experi- Encyclopedia of Entrepreneurship. Engle- ence Influence Opportunity Identification?,” wood Cliffs, NJ: Prentice Hall, 72–90. The Journal of Private Equity 7(1), 7–14. doi: Shepherd, D. A., and D. R. DeTienne (2005). 10.3905/jpe.2003.320059. “Prior Knowledge, Potential Financial Ucbasaran, D., P. Westhead, and M. Wright Reward, and Opportunity Identification,” (2009). “The Extent and Nature of Opportu- Entrepreneurship Theory and Practice 29(1), nity Identification by Experienced Entrepre- 91–112. neurs,” Journal of Business Venturing 24(2), Short, J. C., D. J. Ketchen, C. L. Shook, and R. 99–115. D. Ireland (2010). “The Concept of “Oppor- von Graevenitz, G., D. Harhoff, and R. Weber tunity” in Entrepreneurship Research: Past (2010). “The Effects of Entrepreneurship Accomplishments and Future Challenges,” Education,” Journal of Economic Behavior & Journal of Management 36(1), 40–65. Organization 76(1), 90–112. doi: 10.1016/ Singh, R. P., G. E. Hills, R. C. Hybels, and G. T. j.jebo.2010.02.015. Lumpkin (1999). “Opportunity Recognition Walter, S. G., and D. Dohse (2012). “Why Mode Through Social Network Characteristics of and Regional Context Matter for Entrepre- Entrepreneurs,” in Frontiers of Entrepre- neurship Education,” Entrepreneurship & neurship Research. Eds. P. D. Reynolds, W. Regional Development An International D. Bygrave, S. Manigart, C. M. Mason, G. D. Journal 24(9–10), 807–835. Meyer, H. J. Sapienza, and K. G. Shaver. Weber, R. (2012). Evaluating Entrepreneurship Wellesley, MA: Babson College, 228–256. Education—Innovation Und Entrepreneur- Souitaris, V., S. Zerbinati, and A. Al-Laham ship. GWV Fachverlage GmbH. (2007). “Do Entrepreneurship Programmes

KARIMI ET ET AL. 209 23 Journal of Small Business Business Management Management 2016 2014 54(1), ••(••), pp. pp. 210–228 ••–•• doi: 10.1111/jsbm.12138 How Does Race and Ethnicity Affect Persistence in Immature Ventures? by Robert E. Freeland and Lisa A. Keister

Does race/ethnicity affect persistence in an immature venture? Using data from the Panel Study of Entrepreneurial Dynamics II, we examine how race/ethnicity, access to supplier credit, and personal financial investment affect three entrepreneurial outcomes: continued engagement, new firm creation, and disengagement. We find that compared with whites, blacks were less likely to receive supplier credit and invest more of their own capital, whereas Hispanics did not significantly differ from whites. Blacks were more likely to persist and remain engaged in an immature venture if they did not achieve success after two years in operation, whereas Hispanics were more likely to disengage.

p. 59) studied reasons for business closures and Introduction found that true business failure is less frequent The decision to persist in an immature busi- than commonly perceived with only one-third ness has important implications for business (33 percent) of small businesses in the United success as survival and growth reflect both the States closing because owners considered them financial well-being of the new venture and the unsuccessful. He concluded that had owners ability of entrepreneurs to endure the unex- persisted, their prospects for survival and even- pected obstacles found in business ownership. tual success were reasonable. Although these Entrepreneurs may discover that operating a patterns suggest that persistence in new ven- business requires more time or money than tures is important, the factors that affect deci- expected or that profits are lower, competition sions to persist in an immature business have is more challenging, or government regulations attracted relatively little research attention. are more difficult to manage than anticipated Both individual and business characteristics (Giordano and Tabones 2013; Parker 2009; Van factor into decisions to persist or disengage, Gelderen 2012). For example, there is evidence and the race/ethnicity of the entrepreneur is that underperforming firms often outlast higher likely to be an important influence. There is performing ones because owners value nonfi- evidence that both traits of the start-up nancial benefits including autonomy in sched- (e.g., revenues, number of employees, avail- uling their work or personal satisfaction over ability of start-up capital) and of the entrepre- financial performance of the start-up (Gimeno neur (e.g., income, savings) affect business et al. 1997). In related work, Headd (2003, closure (Headd 2003). However, each of these

Robert E. Freeland is a PhD candidate in the Department of Sociology, Duke University. Lisa A. Keister is Gilhuly Family professor of Sociology, Duke University. Address correspondence to: R. Freeland, Department of Sociology, Duke University, Durham, NC 27708. E-mail: [email protected].

210 JOURNAL OFFREELAND SMALL BUSINESS AND KEISTER MANAGEMENT 1 traits varies with the race/ethnicity of the entre- processes have isolated them from important preneur. For example, there are important dif- mainstream social institutions. Applying his ferences in firm revenues by race/ethnicity: in perspective to the credit market, a financial 2002, total annual revenue averaged $74,018 institution central to entrepreneurial outcomes, for black-owned businesses, $141,044 for we explore how race and access to credit inter- Hispanic-owned businesses, and $439,579 for act to influence decisions to persist in an imma- white-owned businesses (Fairlie and Robb ture business venture. This exploratory study 2008, p. 34). Similarly, there are important dif- proceeds in three steps. We begin by examin- ferences in number of employees by the race/ ing the association between race/ethnicity ethnicity of the entrepreneur: the percentage of and supplier credit using binary logistic regres- minority-owned firms with at least one paid sion on data from the Panel Study of Entrepre- employee was 11 percent for blacks, 21 percent neurial Dynamics II (PSED II) (Reynolds and for Hispanics, and 21 percent for whites (Fairlie Curtin 2008). Next, we explore the relationship and Robb 2008, p. 41). Blacks and Hispanics between race/ethnicity and the amount of per- also tend to have less personal starting capital sonal financial investment made in the business on average relative to whites; and the average and how access to supplier credit conditions income of blacks and Hispanics is approxi- this relationship using a series of linear regres- mately two-thirds of whites with a net worth sion models. Last, we examine the association approximately one-tenth of whites (Fairlie and between race/ethnicity and continued engage- Robb 2008; Keister and Moller 2000; Shapiro ment after two years in operation using multi- 2004). In 2002, 1 percent of whites, 25 percent nomial logistic regression and examine how of Hispanics, and 32 percent of blacks had zero personal investment levels and supplier credit or negative net worth (Kochhar 2004, p. 2). In affect this relationship. This study makes both addition to differences in capital levels, borrow- theoretical and practical contributions to the ing patterns vary widely across racial/ethnic literature. First, we explicitly introduce race/ groups. In a comparison of financing sources ethnicity into studies of persistence in new for white-owned and black-owned businesses, businesses and, in the process, speculate about Fairlie and Robb (2008, p. 115) showed that the related macro-level factors that influence whites obtained loans from commercial lending individual decision-making in this context. institutions at nearly twice the rate of blacks Second, we introduce the notion of social dis- (12.1 percent versus 6.6 percent, respectively). location to the study of business persistence; Personal credit card debt was the only capital and finally, our results suggest important dif- source that black entrepreneurs used more ferences in business processes by race/ than whites (3.8 percent and 2.9 percent, ethnicity that might usefully inform policy respectively). Because black entrepreneurs face discussions related to business start-up and difficulties in obtaining external financing, they success. save more on average toward starting their business and provide more self-financing com- Social Dislocation and pared with whites (Parker and Belghitar 2006). Entrepreneurial Persistence These systematic racial/ethnic differences in Racial and ethnic differences in entrepre- the factors that influence decisions to persist neurial and related processes reflect a large suggest that persistence should also vary by number of conditions and actions that interact race/ethnicity, but to date, research has not in complex ways over large spans of time. explored how race/ethnicity affects persistence Early life conditions, geography, educational in an immature venture. attainment, occupational outcomes, saving, To fill this gap, we explore how race and borrowing, and home equity are some of ethnicity affects continued engagement in factors that can affect economic opportunities an immature venture. We draw on ideas and decision-making (Keister and Moller 2000; from Wilson’s (1987, 1991, 1997) theory of Pager, Western, and Bonikowski 2009; Shapiro social dislocation, a prominent structural theory 2004). Social dislocation (Wilson 1987, 1997) is of racial disadvantage. Social dislocation is a useful concept for understanding how racial/ the idea that migration patterns and ethnic differences affect persistence in entre- deindustrialization have left blacks particularly preneurship because it provides a broad disadvantaged even compared with other theoretical framework to integrate these inter- racial/ethnic groups because macro-level connected factors. Social dislocation explains

2 JOURNAL OFFREELAND SMALL BUSINESS AND KEISTER MANAGEMENT 211 how structural conditions create differential Stratified Access to Human access to mainstream institutions and social networks resulting in differences in economic Capital, Financial Capital, and social outcomes by race and ethnicity. In and Credit particular, Wilson posits that historical discrimi- Isolation from educational and economic nation, migration patterns, and structural institutions has led to significant disparities changes in the economy have left blacks iso- in human and financial capital that are likely lated from important mainstream social institu- to affect business persistence patterns by tions and local resources, resulting in a wide race/ethnicity. Lower educational attainment range of negative outcomes including jobless- reduces paid employment opportunities that ness, lower educational attainment, out-of- not only makes self-employment relatively wedlock births, and crime. Migration of more attractive but also limits reentry into the southern blacks to large northern metropolises paid labor market should a venture fail (Ahn followed by decentralization of manufacturing 2011). According to the National Center for jobs to the suburbs produced a “spatial mis- Education Statistics, high school dropout rates match” (Kain 1968) between employment were 5.2 percent for whites, 9.3 percent opportunities and black workers, resulting in for blacks, and 17.6 percent for Hispanics areas of concentrated joblessness and poverty (Chapman 2011, p. 9). The percentage of 25- to among urban blacks. Migration patterns includ- 29-year-olds who attained a bachelor’s degree ing the migration of working- and middle-class or higher was 40 percent for whites, 23 percent families from city centers to higher status for blacks, and 15 percent for Hispanics (Aud neighborhoods and suburbs combined with 2013). Disparities in the returns to education high rates of black migration back into high- exacerbated educational differences. In 2010, poverty neighborhoods exacerbated racial/ the median earnings for full-time, young adults ethnic segregation and poverty concentrations aged 25 to 34 with a bachelor’s degree were 10 (South, Crowder, and Chavez 2005). Indeed, by percent lower for blacks and Hispanics com- 1980, the racial/ethnic composition of neigh- pared with whites (Aud 2013, p. 228). Among borhoods categorized as extremely poor was 65 males who had not completed high school, percent black, 22 percent Hispanic, and 13 only 28 percent of black males were employed percent non-Hispanic and other races compared with 77 percent of Hispanic and 55 (Jargowsky and Bane 1991). percent of white males (Ross and Kena 2012, p. The social dislocation concept has been xv). At least partially a result of educational and credited with initiating a robust empirical occupational disadvantage, blacks and Hispan- literature on how neighborhood effects and ics have earned approximately two-thirds the poverty concentrations affect educational income of whites and own less than one-tenth attainment, unemployment, and female-headed their net worth (Fairlie and Robb 2008; Keister households (Eggers and Massey 1992; Sampson and Moller 2000; Shapiro 2004). In 2002, 13 2008). Spatial mismatch contributed to high percent of whites, 25 percent of Hispanics, and levels of black unemployment in areas that lost 32 percent of blacks had zero or negative net significant manufacturing jobs prior to the worth (Kochhar 2004, p. 2). 1980s (Ihlanfeldt and Sjoquist 1998; Johnson Although the effect of income and wealth 2006; Mouw 2000) and this process continued on entrepreneurial entry has been mixed into the 1990s, when further manufacturing (Davidsson 2006; Delmar and Davidsson 2000; declines contributed to unemployment rates in Kim, Aldrich, and Keister 2006), financial excess of 40 percent for working-age black resources have been shown to improve busi- men in low-income jobs (Quillian 2003). ness survival, profits, sales, and the likelihood Deindustrialization and the concentration of of hiring employees (Bates 1990, 1997; Fairlie poverty are particularly problematic for blacks, and Robb 2008; Taylor 1999). Not all busi- especially those with low income and educa- nesses require commercial loans, but supplier tion, because they are more likely to remain in credit is an important resource even among racially segregated urban centers, whereas His- small ventures. Supplier credit accounted for 31 panics are more likely to disperse to suburban percent of all small business debt in 1993 with and rural areas with more opportunities for 61 percent of firms holding some level of sup- paid employment (Kandel and Cromartie 2004; plier debt (Aaronson et al. 2004, p. 47), and is Lichter, Parisi, and Taquino 2012). a critical factor in the transition from nascent

212 JOURNAL OFFREELAND SMALL BUSINESS AND KEISTER MANAGEMENT 3 entrepreneurship into a new firm (Parker and trates this difference: between 2005 and 2010, Belghitar 2006). The lack of wealth and home value decreases reduced the net worth of resources among blacks and Hispanics can whites by 28 percent from $130,350 to affect perceptions of credit worthiness and lead $110,729, whereas the wealth of black and to differential patterns of credit and funding Hispanic households dropped 55 percent to that could ultimately affect persistence in an $4,955 and $7,424, creating a historic 20- and entrepreneurial effort (Blanchflower, Levine, 15-fold difference in black and Hispanic and Zimmerman 2003; Cavalluzzo and Wolken wealth, respectively, compared with whites 2005). In particular, there is evidence that (United States Census 2013). Access to financial white entrepreneurs use commercial lending resources from friends and family is similarly institutions at twice the rate of black entre- limited for blacks and Hispanics because social preneurs (12.1 percent versus 6.6 percent, networks tend to be homophilous. That is, the respectively) (Fairlie and Robb 2008, p. 115). friends and family of would-be black and His- Similarly blacks, especially those in inner-city panic entrepreneurs are likely to also have less neighborhoods, are more frequently discour- capital themselves to invest than the friends aged from applying for supplier credit, and and family of would-be white entrepreneurs when they do apply are significantly less likely (McPherson, Smith-Lovin, and Cook 2001; to be approved (Blanchflower, Levine, and Stack 1974). Consequently, whites are more Zimmerman 2003; Cole 2010; Coleman 2005). able use social ties for financial exchanges, Geographic segregation by race and ethnicity, whereas blacks more often use their network particularly in urban centers, has several impli- ties to provide practical or instrumental support cations for supplier credit access. Because sup- (Sarkisian and Gerstel 2004). plier credit arrangements are made with firms Because blacks are less likely to obtain rather than large financial institutions, eco- external credit and their social networks are nomically disadvantaged inner cities contain less able to provide financial backing, black fewer firms, limiting the number of opportuni- entrepreneurs are more likely to finance their ties for supplier credit. Racial segregation also operations from personal savings (Parker and alters the composition of social networks that Belghitar 2006). High interest credit card debt has been shown to affect supplier credit pat- is the only financing source blacks use more terns; Hispanic suppliers are able to provide than whites (3.8 percent for blacks versus 2.9 more credit to Hispanic business-owners, percent for whites) (Blanchflower 2004). The whereas no such relationship exists for black use of credit cards to fund entrepreneurial business-owners (Aaronson et al. 2004). Barri- activities increased among blacks but not ers to credit access resulting from the disadvan- whites in regions where interest rate ceilings taged structural location of blacks suggest that: were raised, illustrating the high demand and limited options black entrepreneurs have in H1: Blacks will have a lower likelihood of obtaining external capital (Chatterji and obtaining supplier credit than white or Seamans 2012). The barriers black entrepre- Hispanic net of demographic and business neurs face in obtaining external capital suggest controls. our second hypothesis:

In addition to supplier credit, nascent entre- H2: Blacks will invest more of their personal preneurs rely on personal financial resources funds into their venture than white or His- and loans and gifts from friends and family to panic net of demographic and business start their businesses (Reynolds 2007, p. 13), controls. and geographic concentration by race and ethnicity can adversely affect access to these. For Structural Disadvantage, most Americans, including most entrepreneurs, home equity is a significant source of capital. Investment, and Continued Given that blacks and Hispanics tend to be Engagement concentrated in neighborhoods where housing The decision to start or end a business values are relatively low and grow slowly even reflects the perceived costs and benefits of self- in times of economic prosperity (Shapiro 2004), employment relative to paid employment their access to this critical resource is more (Evans and Jovanovic 1989; Evans and Leighton limited than for whites. A recent pattern illus- 1989; Gatewood et al. 2002). Several factors can

4 JOURNAL OFFREELAND SMALL BUSINESS AND KEISTER MANAGEMENT 213 alter this calculus, resulting in racial and ethnic Åstebro, Jeffrey, and Adomdza (2007) found differences in persistence patterns. First, low that optimism and past expenditures resulted in wages and fewer paid employment opportuni- nearly one-third (29 percent) continuing to ties due to the spatial mismatch between loca- spend money and more than half continuing to tion and jobs increase the relative benefits to spend time on projects even after receiving self-employment, encouraging blacks to start a advice to cease. The combination of factors business and persist longer. Blacks are 57 resulting from the isolation of blacks into percent more likely to report participation in a racially segregated urban areas including start-up compared with whites, whereas His- reduced opportunities within the paid labor panic and white start-up rates did not statisti- market due to the spatial mismatch of educa- cally differ (Reynolds and Curtain 2008). When tion and jobs, selection effects, and higher self- jobs within the public sector became available, funding requirements due to difficulties in self-employment rates among blacks declined obtaining external funding suggests: (Boyd 1991). Highlighting the lack of alternative opportunities, black entrepreneurs who H3: Blacks will have a higher likelihood of quit their business were more likely to exit to remaining engaged in an immature venture nonemployment (Ahn 2011). net of demographic and business controls. Second, structural barriers create a selection effect filtering through the most motivated and Research Methods financially able black and Hispanic entrepre- Sample neurs. Research finds black entrepreneurs to be Data for our study come from the PSED II more confident and optimistic about their busi- (Reynolds and Curtin 2013). Designed to ness prospects, engaged in more start-up activi- examine the entire entrepreneurial process ties, and were more willing to remain in the from conception through new firm creation, nascent stage longer compared with whites the PSED II is a longitudinal, nationally repre- (Greene and Owen 2004; Kollinger and Minniti sentative survey of nascent entrepreneurs 2006; Parker and Belghitar 2006). Earnings managed by the Entrepreneurship Research from black-owned businesses generally lag far Consortium (Reynolds and Curtin 2013). The behind their white or Hispanic counterparts. PSED II contains a broad range of detailed Between 2005 and 2007, 60 percent of black- information regarding an entrepreneur’s work owned businesses were in the lowest business history, financial resources, business practices, earnings category (less than $30,000 annually) and other individual and background charac- compared with only 24 percent for Hispanic- teristics. For a detailed description of the owned and 10 percent for white-owned busi- PSED II’s development and methodology, see nesses (Valdez 2011, p. 103). In a comparison Reynolds and Curtin (2008). of black and Hispanic businesses, Valdez Beginning in 2005, data collection was con- (2011) found that blacks endured these eco- ducted in three phases. In the first phase, 1,214 nomic conditions because they were more nascent entrepreneurs were screened from likely to cite noneconomic reasons for starting 31,845 individuals in the United States using their business including autonomy or satisfac- random-digit telephone dialing. In phase two, tion. Blacks were also found to be more likely the nascent entrepreneurs were administered to take on additional jobs or rely on spousal detailed phone surveys lasting approximately income to continue their ventures. 60 minutes. The initial screening was con- Moreover, if the disadvantaged position of ducted by a commercial survey firm with sub- blacks increases self-funding, this can lead to a sequent surveys administered by the University psychological process known as an escalation of Michigan Institute for Social Research of commitment (Staw 1981) or tendency to (UMISR). A third phase consisted of annual continue investing more time, money, and follow-up phone interviews, lasting approxi- effort into a venture even after receiving advice mately 60 minutes, conducted 12 months after to quit (Arkes and Blumer 1985; Åstebro, completion of the initial survey. Our sample Jeffrey, and Adomdza 2007). As a result, consists of data from the initial survey and two increased personal investment has been shown follow-on waves. This two-year timeframe is to be a significant factor in continued engage- ideal for studying persistence because it is long ment (DeTienne, Shepherd, and De Castro enough for entrepreneurs to evaluate whether 2008). In a study of independent inventors, success is unlikely and disengage but short

214 JOURNAL OFFREELAND SMALL BUSINESS AND KEISTER MANAGEMENT 5 enough that some determined entrepreneurs persistence as “an individual’s continued active, will persist even if the venture has not matured behavioral involvement as a founder-owner in into a new firm (Fichman and Levinthal 1991). a business start-up attempt at a particular point Individual case weights were developed by the in time.” He argues that this is a theoretically UMISR to maintain representativeness based on complex concept connoting a range of factors age, education, race, and sex, and to correct for from positive qualities of optimism (Kollinger differences due to differential response rates. and Minniti 2006) to the negative psychological Response rates for follow-on interviews were trap of escalating commitment (Staw 1981). approximately 80 percent, reducing the two- Although Davidsson (2011) operationalizes year sample to 977 from the initial 1,214 par- persistence as a binary variable indicating that ticipants. Listwise deletion further reduced the the respondent remained actively engaged in analytic sample during modeling, and weights an immature venture and plans to continue for were recalculated for each wave to account the next six months, we operationalize persis- for the changing sample size resulting from tence similarly but disaggregate new firm cre- attrition. ation and disengagement into separate Although the PSED II is one of the most outcome states. The PSED II identifies three extensively used data sets in entrepreneurship separate outcome states: new firm creation, research in the United States, it does have its continued engagement, and disengaged. New limitations (for a full bibliography, see firm creation identified if the respondent Reynolds and Curtin 2013). One salient issue is received income in 6 of the past 12 months and that the recent recession has altered the credit if that income covered all expenses including market since the data were originally collected. owner’s wages and salaries. Actively engaged On one hand, economic and wealth disparities identified respondents that devoted more than between racial and ethnic groups have only 160 hours (4 weeks of full-time work) in the worsened (United States Census 2013), suggest- past 12 months and expected to spend 80 or ing that access to valuable resources remains a more hours in the next 6 months on their significant problem or has even grown in sig- business or expected it to be their major career nificance. On the other hand, greater regulation focus over the next 12 months but did not have and scrutiny of the credit process may reduce income in excess of expenses. This measure of lending discrimination. Because supplier credit persistence is similar to Davidsson’s (2011) is provided by firms rather than large financial with the added criterion that active engagement institutions, it is reasonable to assume that the could include entrepreneurs expecting their processes examined in this study remain venture to be their major career focus for the largely consistent although this suggests future next 12 months. Last, disengaged identifies research should directly examine how changes respondents that indicated that they were no in the credit market have altered these longer engaged in their business in either the processes. first or second follow-up surveys. Weighted descriptive statistics of all measures are shown Dependent Variables in Table 1. To study persistence and its causes, we use a series of regression models to investigate how Independent Variables race and ethnicity affect supplier credit, per- Race/ethnicity is our primary independent sonal financial investments, and entrepreneur- variable, and we use definitions that are con- ial outcomes. We operationalize supplier credit sistent with work on social dislocation. In par- as a binary variable coded 1 if the respondent ticular, we follow this literature in conceiving indicated that they had received any amount of of race and ethnicity as socially constructed supplier credit during the initial survey. Per- groupings of people (Berger, Ridgeway, and sonal financial investment is a continuous vari- Zelditch 2002; Ridgeway 1991). Ethnic catego- able measured in the natural log of U.S. dollars ries are based on a distinct sense of shared during the initial interview. national or regional heritage, language, or Outcome states including persistence are culture, whereas racial categories are based on more complex concepts warranting further dis- perceived physical differences or phenotypes cussion. In a comparison of end-state terminol- including skin color, facial features, or hair ogy (success, engagement, persistence, and texture (Brubaker 2004; Omi and Winant 1994). progress), Davidsson (2011, p. 310) defines Because they are abstract social constructions,

6 JOURNAL OFFREELAND SMALL BUSINESS AND KEISTER MANAGEMENT 215 Table 1 Descriptive Statisticsa,b

Variable n Mean S.D.

Outcome 977 New Firm 136 0.14 0.49 Active 385 0.39 0.50 Disengaged 455 0.47 0.71 Race/Ethnicity 968 White 567 0.59 0.49 Black 193 0.20 0.40 Hispanic 113 0.12 0.32 Other 95 0.10 0.30 Control Variables Personal Financial Investment ($U.S.) 977 $1,862 $144,803 Supplier Credit 973 0.18 0.39 Outside Funding Received 971 0.09 0.29 Business Characteristics Formal Business Plan 963 0.17 0.37 Formally Registered 956 0.29 0.46 Business Bank Account 974 0.29 0.45 Age 969 39.2 12.7 Female 977 0.41 0.49 Married 976 0.50 0.50 Education 976 No High School Diploma 51 0.05 0.22 High School Diploma 178 0.18 0.39 Vocational/Technical Degree 50 0.05 0.22 Some College 329 0.34 0.47 Bachelor’s Degree 255 0.26 0.44 Graduate/Professional Degree 113 0.12 0.32 Income ($U.S.) 910 $55,000 $29,045 Employment Status 971 Full-Time 584 0.60 0.49 Part-Time 137 0.14 0.35 Retired 58 0.06 0.24 Unemployed 192 0.20 0.40 Work Experience (Years) Years Full-Time Experience 972 17.2 11.9 Years Industry Experience 973 7.86 9.64 Prior Occupation 975 White Collar—Managerial 110 0.11 0.32 White Collar—Professional 348 0.36 0.48 White Collar—Sales/Clerical 124 0.13 0.33 Blue Collar—Craftsman/Foreman 117 0.12 0.33 Blue Collar—Semiskilled, Unskilled 65 0.07 0.25 Service 73 0.08 0.26 Other Occupation 138 0.14 0.35 aS.D., standard deviation. bOther races include Asian and multiracial; graduate/professional degrees include master’s, law, M.D., Ph.D., and Ed.D.; median personal investment shown (mean = $19,332); median income shown (mean = $88,000).

216 JOURNAL OFFREELAND SMALL BUSINESS AND KEISTER MANAGEMENT 7 race and ethnicity are not an intrinsic part of a professional degrees are more predictive of human being but an identity whose construc- business outcomes than bachelor’s degrees tion can vary considerably across nations (Marx (Bates 1985; Parker 2009). A bachelor’s degree 1998; Wade 1997). In both our theoretical and can therefore be viewed as an educational mid- subsequent empirical discussions, we use racial point, providing a contrast between the lower and ethnic categories from the PSED II. Con- attainment of those with a high school diploma sistent with the U.S. Census and other major or less and those possessing higher level population surveys, the PSED II uses a combi- graduate/professional degrees. Prior work nation of self-identified racial and ethnic cat- history, particularly industry experience, is egories (Bureau of Labor Statistics 2013; another critical factor affecting not only busi- National Opinion Research Center 2013). ness success but also the ability to transition Respondents were first asked if they are of back into paid employment should the business Hispanic or Latino origin followed by identifi- close (Ahn 2011). Controls for prior employ- cation of their race with options of white, ment history include the number of years of black, or African American, American Indian, previous paid employment, prior occupation, Asian, Pacific Islander, or of mixed racial back- and years of full-time and industry experience. ground (Reynolds and Curtin 2013). We use the We included four dummy variables to identify racial term black instead of the ethnic term employment status as part-time employment African American to reflect the centrality that affects entry into self-employment (Folta, race has played in the history of U.S. racial/ Delmar, and Wennberg 2010). We measure ethnic stratification. We coded four separate prior occupation with seven binary variables; binary race/ethnicity variables with 1 if the white-collar managerial is the reference respondent indicated that they were (1) white category. (reference category); (2) black; (3) Hispanic; or Financial capital is a critical factor in starting (4) other race/ethnicity. Although Fairlie and and growing a start-up. Following the method- Robb (2008) find significant heterogeneity ology employed by Kim, Aldrich, and Keister between Asians and other racial categories, (2004), we operationalize income as a continu- we combined Asians and mixed-race responses ous variable adjusting for cases using income into the other race/ethnicity category because brackets. In the majority of cases (1,105), we of the limited number of cases in our were able to code exact reports of the respon- sample. dent’s income as a continuous variable. In 34 cases, income was reported in categorical Control Variables income brackets. For these cases, we coded To isolate the effect of race and ethnicity on income at the midpoint of the categorical continued engagement, we control for impor- range. Cases were dropped if no income data tant demographic and business characteristics were provided or an exact range could not be that were measured during the initial survey. determined, including when there was no Background controls include age, gender, and upper categorical limit. This procedure yielded marital status. Human capital, the composite of income data on 93 percent of the cases. a person’s education, skills, and experience, is Controls for business characteristics include considered one of the most valuable resources whether the business was formally registered, for entrepreneurial success and is critical to had a formal business plan, had a separate new firm survival and growth (Bates 1985; business bank account, or received any outside Parker 2009; Taylor 1999). We use six binary funding. We coded all four characteristics variables indicating the highest level of educa- binary variables as 1 if the respondent indi- tion to measure education; having a bachelor’s cated that the business had these characteris- degree is selected as the reference category for tics. We performed collinearity diagnostics for several reasons. First, with 26 percent of all variables in the analyses. Measures of age respondents having a bachelor’s degree, it is and years of full-time experience had the one of the largest categories, second only to highest correlation at 0.84 with a variance infla- those with some college, and nearly equal to tion factor well within normal tolerances. the combined number of respondents with a high school diploma, vocational/technical Statistical Modeling degree, and those without a high school We use three sets of regression models diploma. Second, research finds that graduate/ to examine how race and ethnicity affect the

8 JOURNAL OFFREELAND SMALL BUSINESS AND KEISTER MANAGEMENT 217 likelihood of obtaining supplier credit, levels of Table 2 personal financial investment, and outcome Logistic Regression Model for states including persistence. The first model a,b uses binary logistic regression to examine how Supplier Credit race/ethnicity affects the likelihood of obtaining supplier credit. All models include the fol- Supplier Credit lowing set of demographic and business Odds Ratios controls: age, gender, marital status, education, (S.E.) income, work experience, employment status, prior occupation, and whether the business Race/Ethnicity was formally registered, had a formal business Black 0.20 (0.37)*** plan, or had a business bank account. Hispanic 1.15 (0.31) Next, we explore how race/ethnicity affects Other 0.76 (0.34) the amount of personal self-financing of the Control Variables business and how access to supplier credit Formal Business Plan 1.29 (0.26) affects this relationship using two generalized Formally Registered 1.71 (0.23) least squares regression models. The first Business Bank Account 2.55 (0.23) Age 1.02 (0.02) model includes the same set of controls used in Female 0.83 (0.22) the previous model, whereas the second model Married 1.22 (0.20) introduces controls for external funding includ- Education ing measures of supplier credit and outside No High School Diploma 2.68 (0.49)* funding. Our last model uses multinomial High School Diploma 1.42 (0.33) regression to examine how race/ethnicity Vocational/Technical Degree 0.84 (0.51) affects persistence and other outcome states. Some College 1.83 (0.27)* This model includes all measures of race/ Graduate/Professional Degree 1.26 (0.33) ethnicity, external funding, demographic, and Income (log $U.S.) 1.07 (0.14) business controls with disengaged as the refer- Work Experience (Years) ence category. Industry Experience 1.01 (0.01) Full-Time Experience 0.98 (0.02) Results Employment Status Blacks Less Likely to Obtain Part-Time 1.16 (0.29) Supplier Credit Retired 0.81 (0.45) The results of Model 1 in Table 2 show strong Not Employed 0.52 (0.32)* racial/ethnic effects and support our first Prior Occupation hypothesis that blacks are less likely to obtain White Collar—Professional 0.74 (0.31) supplier credit compared with either whites or White Collar—Sales/Clerical 0.77 (0.39) Hispanics. With an odds ratio of 0.20, blacks Blue Collar—Craftsman/ 1.24 (0.38) Foreman were five times less likely to have obtained Blue Collar—Semiskilled, 2.24 (0.44) supplier credit when compared with whites, Unskilled whereas Hispanics did not significantly differ Service 0.82 (0.50) from their white counterparts. Although these Other Occupation 0.71 (0.43) results are consistent with prior research that Constant 0.04 (1.64)* finds black business-owners, especially those in N = 841 urban areas, are particularly disadvantaged in -2Log 712.19 obtaining supplier credit, the strength of our finding warrants discussion of the underlying factors (Aaronson et al. 2004; Blanchflower, aS.E., standard error. b Levine, and Zimmerman 2003; Coleman 2005). Reference categories for independent variables First, part of this can be attributed to reduced include white (for race/ethnicity), college degree (for social capital in the high-poverty, racially segre- education), full-time (for employment status), and white collar-managerial (for prior occupation). gated urban neighborhoods blacks tend to live *p < .05 in as black business-owners are half as likely as **p < .01 Hispanic business-owners (20 percent and 39 ***p < .001 percent, respectively) to obtain supplier credit from firms located within their neighborhood

218 JOURNAL OFFREELAND SMALL BUSINESS AND KEISTER MANAGEMENT 9 (Aaronson et al. 2004, p. 53). The racial/ethnic financial needs or increase self-funding levels composition of social networks has also been if alternatives are unavailable. Although con- shown to affect supplier credit access as His- straints to capital may lead some entrepre- panic business-owners were 20 percent more neurs to close their business, prior research likely to be offered credit when dealing with that finds black entrepreneurs exhibit greater Hispanic suppliers, whereas no similar relation- optimism and willingness to take on additional ship was found when black business-owners jobs to support their venture suggests that dealt with black suppliers (Aaronson et al. they will likely increase self-financing rather 2004). Second, black-owned firms tend to be than disengage (Kollinger and Minniti 2006; concentrated in the less credit-dependent Valdez 2011). The differences in black and service sector with 47 percent of black-owned Hispanic investment patterns found in our firms in business/retail services, whereas study are also consistent with Valdez (2011) Hispanic-owned and white-owned business who found that blacks tend to rely more on were more evenly distributed across industries personal savings to fund their business com- with only 23 percent and 29 percent of their pared with whites and Hispanics, particularly businesses in the service sector, respectively middle-class Hispanics. (Aaronson et al. 2004, p. 52). Last, not only are Many entrepreneurs seek favorable supplier black entrepreneurs denied credit at twice the credit arrangements to manage costs when rate of comparable Hispanic business-owners external funds are unavailable; if this strategy is (Blanchflower, Levine, and Zimmerman 2003), not feasible for black entrepreneurs, they may but many do not even apply because they are be forced to invest relatively high levels of discouraged by lenders or believe they will be personal capital to make business start-up pos- turned down (Coleman 2005). In general, the sible (Winborg and Landström 2001). We results are consistent with prior research docu- examine this process in Model 3, which shows menting the difficulties blacks face in obtaining that supplier credit does significantly affect the supplier credit relative to whites and Hispanics. amount of personal finances invested where the coefficient for blacks increased to 0.40, Blacks Invest Significant Personal whereas Hispanics again remained nonsignifi- Capital in Start-Ups cant when compared with whites. The need to We proposed that blacks would be required increase self-financing levels in response to to invest more of their personal financial credit constraints does have implications as resources in a new business to compensate for new businesses reliant primarily on owner difficulties in obtaining external funding. financing are associated with lower profit Models 2 and 3 in Table 3 examine how race/ margins, problems obtaining long-term financ- ethnicity affects financial investment levels ing in the future, and higher transaction costs without and with external funding controls, (Winborg and Landström 2001). Research respectively. Consistent with prior research, showing blacks are more likely to resort to Model 2 shows that human and financial capi- high-interest credit cards to fund their business tals significantly affect investment levels. The illustrates the high transaction costs black predictors of larger self-funding levels include entrepreneurs are willing to accept when faced higher income, a graduate/professional with limited alternatives (Fairlie and Robb degree, and whether the business had a busi- 2008). ness bank account or formal business plan. Second, the coefficient of 0.37 in Model 2 also Blacks More Likely to Persist, Hispanics shows that blacks invested more of their own More Likely to Disengage personal capital, net of controls, compared Our third hypothesis is that blacks are more with whites, whereas Hispanics again did not likely to remain actively engaged in an imma- significantly differ from whites. These results ture venture after two years in operation com- are consistent with research exploring how pared with whites or Hispanics. Following a entrepreneurs dynamically alter their behavior methodology similar to Parker and Belghitar to meet the gap between available capital and (2006), we use multinomial logistic regression the financial demands of their business (Lam to examine the predictors of three outcome 2010; Winborg and Landström 2001). When states: successful new firm creation, continued external funding is limited, entrepreneurs must engagement, and disengagement. Our results either find alternative ways to manage their are similar to theirs on most measures, but

10 JOURNAL OFFREELAND SMALL BUSINESS AND KEISTER MANAGEMENT 219 Table 3 GLS Regression Models for Amount of Personal Finances Investeda,b

Without Supplier Credit With Supplier Credit Coeff. (S.E.) Coeff. (S.E.)

Race/Ethnicity Black 0.37 (0.14)** 0.43 (0.14)** Hispanic −0.19 (0.17) −0.19 (0.14) Other 0.13 (0.18) 0.14 (0.18) External Funding Supplier Credit 0.39 (0.14)** Outside Funding Received −0.06 (0.19) Control Variables Formal Business Plan 0.33 (0.14)* 0.32 (0.14)* Formally Registered 0.01 (0.13) −0.01 (0.13) Business Bank Account 0.43 (0.13)** 0.37 (0.13)** Age −0.02 (0.01)* −0.02 (0.01)* Female −0.03 (0.11) −0.03 (0.11) Married −0.04 (0.11) −0.05 (0.11) Education No High School Diploma −0.21 (0.27) −0.26 (0.27) High School Diploma 0.19 (0.17) 0.17 (0.17) Vocational/Technical Degree −0.24 (0.25) −0.24 (0.25) Some College 0.10 (0.14) 0.07 (0.14) Graduate/Professional Degree 0.39 (0.18)* 0.38 (0.18)* Income (log U.S.) 0.40 (0.07)*** 0.40 (0.07)*** Work Experience (Years) Industry Experience 0.01 (0.01) 0.01 (0.01) Full-Time Experience 0.01 (0.01) 0.01 (0.01) Employment Status Part-Time 0.11 (0.16) 0.10 (0.16) Retired 0.51 (0.26)* 0.54 (0.26)* Not Employed 0.12 (0.15) 0.14 (0.15) Prior Occupation White Collar—Professional −0.23 (0.18) −0.21 (0.18) White Collar—Sales/Clerical −0.05 (0.21) −0.04 (0.22) Blue Collar—Craftsman/Foreman −0.04 (0.22) −0.04 (0.22) Blue Collar—Semiskilled, Unskilled −0.43 (0.26) −0.47 (0.26) Service 0.02 (0.25) 0.04 (0.25) Other Occupation −0.25 (0.10) −0.22 (0.23) Constant −1.25 (0.85)*** −1.27 (0.84)*** N = 845 χ2 116.71 124.53 aGLS, generalized least squares; S.E., standard error. bDependent variable measured in log $U.S. Reference categories for independent variables include white (for race/ethnicity), college degree (for education), full-time (for employment status), and white collar-managerial (for prior occupation). *p < .05 **p < .01 ***p < .001

220 JOURNAL OFFREELAND SMALL BUSINESS AND KEISTER MANAGEMENT 11 because they aggregate all nonwhites into a in geographically isolated areas of racial segre- single category, separation of blacks and His- gation and poverty, they are less likely to panics in our study allows us to examine the equate success with profitability and instead heterogeneity that exists between these racial/ frame it in non-pecuniary terms from the outset ethnic groups. The results of our last model, compared with Hispanics or whites who tend shown in Table 4, support our hypothesis that to define success in financial terms (Valdez blacks would be more likely to persist com- 2011). Our findings are consistent with the idea pared with either white or Hispanic net of that black entrepreneurs give less weight to controls. Before exploring the determinants of profitability in deciding to persist or disengage. continued engagement, it is useful to examine Though we do not directly examine the causes the factors predicting new firm creation as underlying the tendency for Hispanics to dis- success is regularly the focus of entrepreneurial engage rather than persist, our results are con- studies. Consistent with prior research, the sistent with research that finds Hispanic-owned results of Model 4 show that human and finan- start-ups to be younger on average than black- cial capitals are the primary predictors of owned ventures (Valdez 2011). success, whereas race/ethnicity was not a sig- Similar to new firm creation, human capital nificant factor (Bates 1985; Parker 2009). The significantly affected the likelihood of contin- predictors of new firm creation include those ued engagement with those possessing lower who invested more capital into their business, educational credentials more likely to persist had a graduate/professional degree, had a busi- relative to those with a bachelor’s degree. ness bank account, and who had more years of Entrepreneurs with high school diplomas and industry experience. Higher income itself vocational degrees were twice as likely as reduced the likelihood of new firm creation, college graduates to remain engaged, and those but this is likely due to the control for personal with some college were similar with an odds investment, suggesting that higher income itself ratio of 1.75. These results are not surprising as does not predict new firm creation but whether lower educational attainment reduces paid that income was invested into the business. employment opportunities, which in turn Consistent with Parker and Belghitar (2006) makes self-employment more relatively attrac- who found nascent entrepreneurs were more tive (Le 1999). Increased industry experience likely to transition to start-up status if they had predicted both new firm creation and persis- established supplier credit lines, we find that tence, suggesting that its role in conditioning obtaining supplier credit more than doubled entrepreneurial outcomes is complex. On one the likelihood of new firm creation with an hand, industry experience not only makes odds ratio of 2.09. success more likely but also makes reentry into Examining only success might lead to the paid employment easier, suggesting that if conclusion that race/ethnicity does not affect success was unlikely, those with more paid entrepreneurial outcomes but by examining the employment opportunities would find it easier determinants of alternative outcomes including to disengage (Ahn 2011). On the other hand, continued engagement and disengagement, we greater specific human capital would enable find significant racial/ethnic difference not only them to more accurately judge the long-term between whites and blacks but also between viability of their venture and be more willing to blacks and Hispanics. Whereas blacks were undertake start-ups requiring longer gestation more than twice as likely to remain actively periods (Parker and Belghitar 2006). Because engaged compared with whites, Hispanics we find greater industry experience to increase were three times as likely to disengage if the likelihood of persisting, this suggests the success was not achieved after two years. These latter and supports Headd’s (2003) assertion findings are consistent with Parker and that many entrepreneurs are disengaging pre- Belghitar (2006) who found nonwhites are maturely because more experienced entrepre- more likely remain nascent entrepreneurs but neurs are choosing to remain engaged for elaborates by examining the heterogeneity longer periods. between black and Hispanic entrepreneurs. Last, we find that supplier credit does not One factor in this difference is that increased have a direct effect on persistence but instead personal investments were found to increase affects continued engagement indirectly by both persistence and new firm creation. In increasing the need for higher levels of per- addition, because blacks are more likely to live sonal investment. This finding differs from

12 JOURNAL OFFREELAND SMALL BUSINESS AND KEISTER MANAGEMENT 221 Table 4 Multinomial Logistic Regression Models of Entrepreneurial Outcomesa,b

New Firm Actively Engaged Odds Ratio (S.E.) Odds Ratio (S.E.)

Race/Ethnicity Black 1.12 (0.33) 2.04 (0.21)*** Hispanic 1.17 (0.34) 0.32 (0.32)*** Other 1.51 (0.38) 1.54 (0.28) Personal Financial Investment (log $U.S.) 1.18 (0.08)* 1.12 (0.05)* External Credit Supplier Credit 2.09 (0.28)** 1.32 (0.22) Outside Funding Received 0.69 (0.40) 1.34 (0.29) Control Variables Formal Business Plan 1.11 (0.29) 0.67 (0.23) Formally Registered 1.42 (0.28) 1.25 (0.21) Business Bank Account 2.32 (0.27)** 0.97 (0.21) Age 0.98 (0.02) 1.05 (0.01)*** Female 1.07 (0.24) 1.13 (0.18) Married 1.01 (0.24) 0.83 (0.17) Education No High School Diploma 0.13 (1.14) 1.49 (0.41) High School Diploma 1.46 (0.37) 2.13 (0.27)** Vocational/Technical Degree 1.09 (0.61) 2.25 (0.39)* Some College 1.51 (0.31) 1.75 (0.23)* Graduate/Professional Degree 2.13 (0.36)* 1.17 (0.29) Income (log $U.S.) 0.71 (0.15)* 1.22 (0.11) Work Experience (Years) Industry Experience 1.04 (0.01)** 1.04 (0.01)*** Full-Time Experience 1.01 (0.02) 0.96 (0.01)** Employment Status Part-Time 1.00 (0.35) 0.99 (0.25) Retired 0.34 (0.62) 0.63 (0.36) Not Employed 0.74 (0.32) 0.50 (0.25)** Prior Occupation White Collar—Professional 0.92 (0.36) 1.10 (0.29) White Collar—Sales/Clerical 0.68 (0.47) 1.25 (0.34) Blue Collar—Craftsman/Foreman 0.70 (0.46) 1.01 (0.36) Blue Collar—Semiskilled, Unskilled 0.42 (0.65) 0.83 (0.41) Service 0.37 (0.59) 0.50 (0.43) Other Occupation 1.11 (0.47) 1.44 (0.37) N = 847 χ2 176.6*** aS.E., standard error. bReference category for dependent variable is disengaged businesses. Reference categories for independent variables include white (for race/ethnicity), college degree (for education), full-time (for employment status), and white collar-managerial (for prior occupation). *p < .05 **p < .01 ***p < .001

222 JOURNAL OFFREELAND SMALL BUSINESS AND KEISTER MANAGEMENT 13 Parker and Belghitar’s (2006) study where race/ethnicity affects the likelihood of obtain- access to supplier credit lines directly increased ing external funding including supplier credit persistence and is likely accounted for by meth- and outside funding. We find that even after odological differences. First, because blacks controlling for business and demographic char- and Hispanics differ on measures of supplier acteristics, blacks are five times less likely to credit and levels of personal investment, aggre- obtain supplier credit compared with whites, gating blacks and Hispanics into a single, non- whereas Hispanics did not statistically differ white category may be conflating the dynamics from their white counterparts. These findings of supplier credit with racial/ethnic differences. are consistent with prior research documenting Second, though they use a binary measure of the difficulties blacks face in obtaining credit investing personal money, our continuous (Blanchflower, Levine, and Zimmerman 2003). measure allows greater granularity regarding We hypothesized that the lack of external how supplier credit and personal investments funding would result in additional self-funding interact to affect persistence. Despite differ- to compensate. The results from our second set ences in our results, it should be noted that of models support this hypothesis as blacks both sets of results show that supplier credit invested significantly more of their personal plays an important role in both success and finances into the business compared with either persistence albeit in slightly different ways. white or Hispanic net of controls. Similar to our analysis of supplier credit, whites and Hispan- Conclusion and Discussion ics did not significantly differ. The amount of The aim of this study was to examine how personal financial capital invested was modeled persistence and factors influencing decisions to with and without controls for external funding remain engaged in an immature business vary to examine how credit affects self-funding by race/ethnicity after two years in operation. levels. We then integrate this finding to explore The results of our study suggest that the effect how personal investments, supplier credit, and of race/ethnicity on entrepreneurial outcomes other demographic and business characteristics is complex. Focusing solely on success, our affect the likelihood of persisting, succeeding, results are consistent with Bates (1985) who or disengaging after two years in operation. argues that entrepreneurial outcomes are pri- Results find that although race/ethnicity did not marily a function of human and financial capital predict successful new firm creation, blacks rather than race/ethnicity, but by examining were twice as likely as whites to persist, alternative outcomes including persistence, we whereas Hispanics were three times as likely to find considerable racial and ethnic differences. disengage if success was not achieved in this In exploring the determinants of entrepreneur- timeframe. ial outcomes, race/ethnicity was not a signifi- By exploring how persistence varies by cant predictor of creating a new firm, but if race/ethnicity, we make several contributions success was not achieved after two years in to the literature on entrepreneurship. First, by operation, blacks were more likely to persist, showing that race/ethnicity affects persistence whereas Hispanics were more likely to disen- but not new firm creation, this study suggests gage when compared with white net of demo- that race/ethnicity has a greater influence on graphic and business controls. These results entrepreneurial outcomes than the success/ highlight not only the need to examine alterna- failure dichotomy would suggest, highlighting tive outcomes but also to look beyond the the need to explore not only racial and ethnic black/white dichotomy and examine the het- differences but also alternative outcome states. erogeneous experiences that exist between Second, by drawing on a structural theory of racial and ethnic groups. racial stratification, this study examines how Drawing on ideas from Wilson’s (1987, 1991, individual-level decisions to persist are influ- 1997) theory of social dislocation that theorizes enced by macro-level conditions. Fostering that macro-level processes including migration entrepreneurship has been offered as a policy patterns and deindustrialization have isolated strategy to help racial minorities overcome dis- blacks from mainstream social institutions, we advantage in the traditional labor market and examine how differences in one important achieve upward mobility (Alba and Nee 2003; institution, the credit market, affects self- Bates 1997; Butler and Kozmetsky 2004; financing levels, which in turn affects decisions Heilman and Chen 2003; Porter 1997), yet to persist or disengage. We first explore how research finds stark racial/ethnic differences in

14 JOURNAL OFFREELAND SMALL BUSINESS AND KEISTER MANAGEMENT 223 entrepreneurial outcomes (Fairlie and Meyer with low educational and financial entry 1996; Fairlie and Robb 2008; Parker 2009). requirements such as construction, retail sales, Southern (2011) argues that previous policies repair services, and landscaping. Construction of encouraging a culture of entrepreneurship (33 percent) and repair services (14 percent) have proven inadequate because structural dis- accounted for nearly half (47 percent) of all advantage excludes some motivated individuals Hispanic-owned businesses, whereas only 24 from full participation in business opportuni- percent of white-owned businesses were in ties. By exploring the structural features and these industries (Lofstrom and Wang 2007, p. processes that influence differential outcomes, 40). In contrast, the largest proportion of white- this study can help policy-makers develop owned businesses (17 percent) was in profes- more targeted programs. The approach of sional services compared with only 4 percent of examining and addressing structural barriers to Hispanic-owned businesses. These differences remedy inequality is similar to the European could affect exit rates in a number of ways. approach to social exclusion (Blackburn and First, lower entry costs for Hispanic businesses Ram 2006). Furthermore, similarities and differ- would make disengagement loss costly. ences between racial/ethnic entrepreneurs in Another possibility is that the human capital the United States and in other nations suggest gained through the process of business creation that cross-national comparisons would be a in industries with low entry costs may be more fruitful direction for future research. For transferrable to opportunities within paid example, structural differences may help employment. To test this hypothesis, future explain why Indian entrepreneurs in the United research should examine how race/ethnicity Kingdom more often list financial reasons for affects the types of jobs entrepreneurs obtain starting their business, whereas Ma¯ori entrepre- after exiting self-employment. neurs in New Zealand, who face similar issues of urban dislocation, crime, and poverty as blacks in America, view entrepreneurship as Acknowledgements We are grateful for helpful comments from more of an opportunity to extend kin and social Lynn Smith-Lovin, Scott Fitzgerald, G. Robin networks rather than as a means for pecuniary Gauthier, and S. Joshua Mendelsohn. gains (Haar and Delaney 2009; Wood, Davidson, and Fielden 2012; Zapalska, Perry, and Dabb 2003). References This study does have several limitations. Aaronson, D., R. W. Bostic, P. Huck, and R. Though we examined the effect of credit and Townsend (2004). “Supplier Relationships investment on persistence, these factors do not and Small Business Use of Trade Credit,” fully explain differential outcomes, suggesting Journal of Urban Economics 55(1), 46–67. that other processes are involved in differenti- Ahn, T. (2011). “Racial Differences in Self- ating outcomes. One particularly salient possi- Employment Exits,” Small Business Econom- bility is the growing research on perseverance ics 36(2), 169–186. that looks at the tendency of some entrepre- Alba, R., and V. Nee (2003). Remaking the neurs to continue despite adversity (Van American Mainstream: Assimilation and Gelderen 2012). Though we do not examine Contemporary Immigration. Cambridge, personality traits including perseverance, it is MA: Harvard University Press. possible that cultural differences in persever- Arkes, H. R., and C. Blumer (1985). “The Psy- ance could explain as much, if not more, of the chology of Sunk Costs,” Organizational variance in outcomes than structural condi- Behavior and Human Decision Processes tions. Also, though we document that outcomes 35(1), 124–140. vary significantly by race/ethnicity, we are not Aud, S. (2013). The Condition of Education able to explore why Hispanics are similar to 2013. Washington DC: National Center for whites in many ways including access to credit, Education Statistics, U.S. Department of yet are more likely to disengage if success is Education. not achieved. We speculate that this difference Åstebro, T., S. A. Jeffrey, and G. K. Adomdza may be due in part to the types of businesses (2007). “Inventor Perseverance after Being started by Hispanic entrepreneurs. Lofstrom Told to Quit: The Role of Cognitive Biases,” and Wang (2007) found that Hispanic-owned Journal of Behavioral Decision Making businesses were concentrated in industries 20(3), 253–272.

224 JOURNAL OFFREELAND SMALL BUSINESS AND KEISTER MANAGEMENT 15 Bates, T. (1985). “Entrepreneur Human Capital Race,” Journal of Financial Economics Endowments and Minority Business Viabil- 106(1), 182–195. ity,” The Journal of Human Resources 20(4), Cole, R. (2010). “Bank Credit, Trade Credit or 540–554. No Credit: Evidence from the Surveys of ——— (1990). “Entrepreneur Human Capital Small Business Finances, MPRA Paper Inputs and Small Business Longevity,” The No. 24689,” Munich Personal RePEc Review of Economics and Statistics 72(4), Archive. 551–559. Coleman, S. (2005). “Is There a Liquidity Crisis ——— (1997). Race, Self-Employment, and for Small, Black-Owned Firms?,” Journal of Upward Mobility: An Illusive American Developmental Entrepreneurship 10(1), Dream. Washington, DC: Woodrow Wilson 29–47. Center Press. Davidsson, P. (2006). “Nascent Entrepreneur- Berger, J., C. L. Ridgeway, and M. Zelditch ship: Empirical Studies and Developments,” (2002). “Construction of Status and Referen- Foundations and Trends in Entrepreneur- tial Structures,” Sociological Theory 20(2), ship 2(1), 1–76. 157–179. ——— (2011). “Engagement, Persistence, Prog- Blackburn, R., and M. Ram (2006). “Fix or Fixa- ress and Success as Theoretically Distinct tion? The Contributions and Limitations of Aspects of Business Creation Processes,” Entrepreneurship and Small Firms to Com- Frontiers of Entrepreneurship Research bating Social Exclusion,” Entrepreneurship & 31(9), 307–321. Regional Development 18(1), 73–89. Delmar, F., and P. Davidsson (2000). “Where Blanchﬂower, D. G. (2004). “Self-Employment: Do They Come From? Prevalence and Char- More May Not Be Better,” Swedish Economic acteristics of Nascent Entrepreneurs,” Entre- Policy Review 11(2), 15–74. preneurship and Regional Development Blanchﬂower, D. G., P. B. Levine, and D. J. 12(1), 1–23. Zimmerman (2003). “Discrimination in DeTienne, D. R., D. A. Shepherd, and J. O. the Small-Business Credit Market,” The De Castro (2008). “The Fallacy of ‘Only the Review of Economics and Statistics 85(4), Strong Survive’: The Effects of Extrinsic 930–943. Motivation on the Persistence Decisions for Boyd, R. L. (1991). “A Contextual Analysis of under-Performing Firms,” Journal of Busi- Black Self-Employment in Large Metropoli- ness Venturing 23(5), 528–546. tan Areas, 1970–1980,” Social Forces 70(2), Eggers, M. L., and D. S. Massey (1992). “A 409–429. Longitudinal Analysis of Urban Poverty: Brubaker, R. (2004). Ethnicity without Groups. Blacks in U.S. Metropolitan Areas between Cambridge: Harvard University Press. 1970 and 1980,” Social Science Research Bureau of Labor Statistics (2013). “Current 21(2), 175–203. Population Survey (CPS).” (http:// Evans, D. S., and B. Jovanovic (1989). “An www.census.gov/cps/). Accessed on June Estimated Model of Entrepreneurial Choice 20, 2013. under Liquidity Constraints,” The Journal of Butler, J. S., and G. Kozmetsky (2004). Immi- Political Economy 97(4), 808–827. grant and Minority Entrepreneurship: The Evans, D. S., and L. S. Leighton (1989). “Some Continuous Rebirth of American Communi- Empirical Aspects of Entrepreneurship,” The ties. Westport, CT: Praeger. American Economic Review 79(3), 519– Cavalluzzo, K., and J. Wolken (2005). “Small 535. Business Loan Turndowns, Personal Wealth, Fairlie, R. W., and B. D. Meyer (1996). “Ethnic and Discrimination,” Journal of Business and Racial Self-Employment Differences and 78(6), 2153–2177. Possible Explanations,” The Journal of Chapman, C. (2011). Trends in High School Human Resources 31(4), 757–793. Dropout and Completion Rates in the United Fairlie, R. W., and A. M. Robb (2008). Race and States: 1972–2009. Compendium Report. Entrepreneurial Success: Black-, Asian-, and NCES 2012–006. Washington DC: National White-Owned Businesses in the United Center for Education Statistics, U.S. Depart- States. Cambridge, MA: MIT Press. ment of Education. Fichman, M., and D. A. Levinthal (1991). “Hon- Chatterji, A. K., and R. C. Seamans (2012). eymoons and the Liability of Adolescence: A “Entrepreneurial Finance, Credit Cards, and New Perspective on Duration Dependence

16 JOURNAL OFFREELAND SMALL BUSINESS AND KEISTER MANAGEMENT 225 in Social and Organizational Relationships,” Kain, J. F. (1968). “Housing Segregation, Negro Academy of Management Review 16(2), 442– Employment, and Metropolitan Decentral- 468. ization,” The Quarterly Journal of Economics Folta, T. B., F. Delmar, and K. Wennberg 82(2), 175–197. (2010). “Hybrid Entrepreneurship,” Manage- Kandel, W., and J. Cromartie (2004). “New Pat- ment Science 52(2), 253–269. terns of Hispanic Settlement in Rural Gatewood, E. J., K. G. Shaver, J. B. Powers, America,” in Rural Development Research and W. B. Gartner (2002). “Entrepreneurial Report No. 99. Washington, DC: U.S. Depart- Expectancy, Task Effort, and Performance,” ment of Agriculture, Economic Research Entrepreneurship Theory and Practice 27(2), Service. 187–206. Keister, L. A., and S. Moller (2000). “Wealth Gimeno, J., T. B. Folta, A. C. Cooper, and Inequality in the United States,” Annual C. Y. Woo (1997). “Survival of the Fittest? Review of Sociology 26(1), 63–81. Entrepreneurial Human Capital and the Kim, P. H., H. E. Aldrich, and L. A. Keister Persistence of Underperforming Firms,” (2004). “Household Income and Net Worth,” Administrative Science Quarterly 42(4), in Handbook of Entrepreneurial Dynamics: 750–783. The Process of Business Creation. Eds. W. B. Giordano, J. A., and E. Tabones (2013). “Regu- Gartner, K. G. Shaver, N. M. Carter, and P. D. lations and Their Impact on Small Business Reynolds. Thousand Oaks, CA: Sage Publi- and Entrepreneurship,” Review of Business cations, 49–61. Research 13(1), 103. ——— (2006). “Access (Not) Denied: The Greene, P. G., and M. M. Owen (2004). “Race Impact of Financial, Human, and Cultural and Ethnicity,” in Handbook of Entrepre- Capital on Entrepreneurial Entry in the neurial Dynamics: The Process of Business United States,” Small Business Economics Creation. Eds. W. B. Gartner, K. G. Shaver, 27(1), 5–22. N. M. Carter, and P. D. Reynolds. Thousand Kochhar, R. (2004). The Wealth of Hispanic Oaks, CA: Sage Publications, 26–38. Households: 1996 to 2002. Washington, DC: Haar, J., and B. Delaney (2009). “Entrepreneur- Pew Research Center. ship and Maori Cultural Values: Using Kollinger, P., and M. Minniti (2006). “Not for ‘Whanaungatanga’ to Understanding Maori Lack of Trying: American Entrepreneurship Business,” New Zealand Journal of Applied in Black and White,” Small Business Eco- Business Research 7(1), 25–40. nomics 27(1), 59–79. Headd, B. (2003). “Redeﬁning Business Lam, W. (2010). “Funding Gap, What Funding Success: Distinguishing between Closure Gap? Financial Bootstrapping: Supply, and Failure,” Small Business Economics Demand and Creation of Entrepreneurial 21(1), 51–61. Finance,” International Journal of Entrepre- Heilman, M. E., and J. J. Chen (2003). “Entre- neurial Behaviour & Research 16(4), 268– preneurship as a Solution: The Allure of 295. Self-Employment for Women and Minori- Le, A. T. (1999). “Empirical Studies of Self- ties,” Human Resource Management Review Employment,” Journal of Economic Surveys 13(2), 347–364. 13(4), 381–416. Ihlanfeldt, K. R., and D. L. Sjoquist (1998). “The Lichter, D. T., D. Parisi, and M. C. Taquino Spatial Mismatch Hypothesis: A Review of (2012). “The Geography of Exclusion: Race, Recent Studies and Their Implications for Segregation, and Concentrated Poverty,” Welfare Reform,” Housing Policy Debate Social Problems 59(3), 364–388. 9(4), 849–892. Lofstrom, M., and C. Wang (2007). “Mexican- Jargowsky, P. A., and M. J. Bane (1991). Hispanic Self-Employment Entry: The Role “Ghetto Poverty in the United States, 1970– of Business Start-Up Constraints,” The 1980,” in The Urban Underclass. Eds. C. Annals of the American Academy of Jencks and P. E. Peterson. Washington, DC: Political and Social Science 613(1), 32– Brookings Institution. 46. Johnson, R. C. (2006). “Landing a Job in Urban McPherson, M., L. Smith-Lovin, and J. M. Cook Space: The Extent and Effects of Spatial Mis- (2001). “Birds of a Feather: Homophily in match,” Regional Science and Urban Eco- Social Networks,” Annual Review of Sociol- nomics 36(3), 331–372. ogy 27, 415–444.

226 JOURNAL OFFREELAND SMALL BUSINESS AND KEISTER MANAGEMENT 17 Marx, A. W. (1998). Making Race and Nation: A Nominal Characteristics,” Social Forces Comparison of South Africa, the United 70(2), 367–386. States, and Brazil. Cambridge: Cambridge Ross, T., and G. Kena (2012). Higher Educa- University Press. tion: Gaps in Access and Persistence Mouw, T. (2000). “Job Relocation and the Study. Washington, DC: National Center for Racial Gap in Unemployment in Detroit and Education Statistics, U. S. Department of Chicago, 1980 to 1990,” American Sociologi- Education. cal Review 65(5), 730–753. Sampson, R. J. (2008). “Moving to Inequality: National Opinion Research Center (2013). Neighborhood Effects and Experiments Meet “About GSS.” (http://www3.norc.org/ Social Structure,” The American Journal of GSS+Website/About+GSS). Accessed on Sociology 114(1), 189–231. June 20, 2013. Sarkisian, N., and N. Gerstel (2004). “Kin Omi, M., and H. Winant (1994). Racial Support among Blacks and Whites: Race and Formation in the United States: From the Family Organization,” American Sociological 1960s to the 1990s, 2nd ed. New York: Review 69(6), 812–837. Routledge. Shapiro, T. M. (2004). The Hidden Cost of Being Pager, D., B. Western, and B. Bonikowski African American: How Wealth Perpetuates (2009). “Discrimination in a Low-Wage Inequality. New York: Oxford University Labor Market: A Field Experiment,” Ameri- Press. can Sociological Review 74(5), 777–799. South, S. J., K. Crowder, and E. Chavez (2005). Parker, S. C. (2009). The Economics of Self- “Exiting and Entering High-Poverty Neigh- Employment and Entrepreneurship. New borhoods: Latinos, Blacks and Anglos Com- York: Cambridge University Press. pared,” Social Forces 81(2), 873–900. Parker, S. C., and Y. Belghitar (2006). “What Southern, A. (2011). Enterprise, Deprivation Happens to Nascent Entrepreneurs? An and Social Exclusion: The Role of Small Busi- Econometric Analysis of the PSED,” Small ness in Addressing Social and Economic Business Economics 27(1), 81–101. Inequalities (Routledge Studies in Entrepre- Porter, M. E. (1997). “New Strategies for neurship). New York: Routledge. Inner-City Economic Development,” Eco- Stack, C. (1974). All Our Kin: Strategies for nomic Development Quarterly 11(1), 11– Survival in a Black Community. New York: 27. Basic Books. Quillian, L. (2003). “The Decline of Male Staw, B. M. (1981). “The Escalation of Commit- Employment in Low-Income Black Neigh- ment to a Course of Action,” Academy of borhoods, 1950–1990,” Social Science Management Review 6(4), 577–587. Research 32(2), 220–250. Taylor, M. P. (1999). “Survival of the Fittest? An Reynolds, P. D. (2007). Entrepreneurship in the Analysis of Self-Employment Duration in United States: The Future Is Now. New York: Britain,” Economic Journal 109(454), C140– Springer Science and Business Media. C155. Reynolds, P. D., and R. T. Curtain (2008). “Busi- United States Census (2013). “Wealth and Asset ness Creation in the United States: Entry, Ownership.” (http://www.census.gov/ Start-up Activities and the Launch of New people/wealth/). Accessed on June 20, 2013. Ventures,” in The Small Business Economy: A Valdez, Z. (2011). The New Entrepreneurs: How Report to the President. Ed. C. Moutray. Race, Class, and Gender Shape American Washington, DC: United States Government Enterprise. Stanford, CA: Stanford University Printing Ofﬁce, 165–240. Press. Reynolds, P. D., and R. T. Curtin (2008). “Busi- Van Gelderen, M. W. (2012). “Perseverance ness Creation in the United States: Panel Strategies for Enterprising Individuals,” Study of Entrepreneurial Dynamics II Initial International Journal of Entrepreneurial Assessment,” Foundations and Trends in Behaviour & Research 18(6), 630–648. Entrepreneurship 4(3), 155–307. Wade, P. (1997). Race and Ethnicity in Latin ——— (2013). “Panel Study of Entrepreneurial America. Sterling, VA: Pluto Press. Dynamics.” (http://www.psed.isr.umich Wilson, W. J. (1987). The Truly Disadvantaged: .edu/psed/home). Accessed June 18, 2013. The Inner City, the Underclass, and Public Ridgeway, C. L. (1991). “The Social Construc- Policy. Chicago, IL: The University of tion of Status Value: Gender and Other Chicago Press.

18 JOURNAL OFFREELAND SMALL BUSINESS AND KEISTER MANAGEMENT 227 ——— (1991). “Studying Inner-City Social Dis- Resource Acquisition Behaviors,” Journal of locations: The Challenge of Public Agenda Business Venturing 16(3), 235–254. Research: 1990 Presidential Address,” Wood, G. J., M. J. Davidson, and S. L. Fielden American Sociological Review 56(1), 1– (2012). “Minorities in Entrepreneurship: An 14. International Review. Northhampton,” ——— (1997). When Work Disappears: The Edward Elgar Publishing. World of the New Urban Poor. New York: Zapalska, A., G. Perry, and H. Dabb (2003). Random House Publishing. “Maori Entrepreneurship in the Contempo- Winborg, J., and H. Landström (2001). “Finan- rary Business Environment,” Journal of cial Bootstrapping in Small Businesses: Developmental Entrepreneurship 8(3), 219– Examining Small Business Managers’ 236.

228 JOURNAL OFFREELAND SMALL BUSINESS AND KEISTER MANAGEMENT 19 Journal of Small Business Business Management Management 2016 2014 54(1), ••(••), pp. pp. 229–243 ••–•• doi: 10.1111/jsbm.12140 Costly Ties: Social Capital as a Retardant of Network-Level Entrepreneurial Orientation by Joakim Wincent, Sara Thorgren, and Sergey Anokhin

Challenging social capital research, we argue that network top management teams (TMTs) established to support entrepreneurship in strategic multipartner networks should be careful in developing ties to outside organizations and networks. We suggest that such ties limit the network TMT’s ability to engage in the strategy-making processes needed to facilitate entrepreneurial orientation on a network level. Based on ﬁve-year panel data from 53 formalized networks of small and medium-sized enterprises, we demonstrate that homogenous and highly educated network TMTs can compensate the negative effects of TMT ties to other organizations, but not the negative effects of interlocking directorates.

partner networks of small and medium-sized Introduction enterprises (SMEs)—extant research on social Social capital, or the resources embedded capital is too general to accurately predict out- within networks of relationships, has received comes. We suggest that network TMTs’ social extensive research attention in the past decade capital may act as a retardant for entrepreneurial (e.g., Arregle et al. 2007; Chen, Chang, and strategies, and we examine conditions under Hung 2008). Simultaneously, the value of ties for which the retardant mechanisms become more fostering entrepreneurship has come to the fore or less significant. More specifically, we show (Ariño, Ragozzino, and Reuer 2008; Hoang and how ties of the networks’ TMTs to outside Antoncic 2003; Rothaermel 2001; Teng 2007). constituents may restrict the entrepreneurial ori- The basic conclusion that this growing stream of entation (EO) of the networks they are assigned research provides is that social capital contrib- to govern (for a comprehensive examination of utes to entrepreneurship (see e.g., Aldrich and the concept of network-level EO, see Wincent, Martinez 2001; Davidsson and Honig 2003; De Thorgren, and Anokhin 2014). Because the Carolis, Litzky, and Eddleston 2009; Packalen number of studies investigating a darker side of 2007). The underlying theoretical arguments of social capital in entrepreneurship remains these studies are, however, very general; as limited (Klyver, Evald, and Hindle 2011; Weber such, conclusions may be inadequate for spe- and Weber 2011), we believe that our study cific conditions. provides novel insights on social capital that With this paper, we show that in certain deserve attention. contexts—namely, among network-level top We draw upon role theory and classical role management teams (TMTs) assigned to support strain models (Kahn et al. 1964; Rizzo, House, strategic processes in formalized strategic multi- and Lirtzman 1970) in relation to social capital

Joakim Wincent is professor of Entrepreneurship & Innovation at Luleå University of Technology and professor in Entrepreneurship at Hanken School of Economics. Sara Thorgren is assistant professor of Entrepreneurship & Innovation at Luleå University of Technology. Sergey Anokhin is associate professor of Marketing and Entrepreneurship at Kent State University. Address correspondence to: Joakim Wincent, Entrepreneurship & Innovation, Luleå University of Tech- nology, Luleå, Sweden. E-mail: [email protected].

WINCENT, THORGREN, AND ANOKHIN ANOKHIN 229 1 to suggest that adding ties comes with costs which captures top management efforts to that may outweigh their benefits in this specific build routines and structural arrangements sup- context. High tie involvement of network TMTs porting innovative ideas and creative projects creates significant resource constraints for cre- (Lyon, Lumpkin, and Dess 2000; Wiklund and ative role performance, thus causing problems Shepherd 2003). This, in turn, suggests that for the development of entrepreneurial strate- fulfilling such expectations is desirable and gies that the network TMT is assigned to would characterize role performance of develop for the networks. Specifically, we network TMTs positively. investigate the restricted ability of the network Importantly, and in contrast with most tra- TMTs to facilitate network-level EO when the ditional EO inquiry that takes place at the firm TMT members have links to other organizations level, these processes operate at the network and networks. Our conceptual development level. Nevertheless, similar to firm-level EO and and results provide a much-needed contrast to its salient dimensions of innovativeness, those arguing for a catalyst function of social proactiveness, and risk-taking, network-level capital in fostering entrepreneurship by dem- EO captures an orientation to develop products onstrating that in some contexts, social capital and service innovations, and to allow experi- could become a retardant. Yet we find certain ments and risk-taking by encouraging uncer- TMTs—homogeneous and highly educated tain projects (Covin and Slevin 1986, 1989; ones—mitigate the negative effects for reasons Lumpkin and Dess 1996). that we explain in the respective section of this manuscript. Social Capital, Network TMTs, and Network-Level EO Network TMTs and EO In the social capital literature, ties are consid- According to Lavie, Lechner, and Singh ered beneficial inasmuch as they contribute with (2007, p. 578), formalized multipartner net- knowledge transfer and compensate one’s own works can be understood as “a collective, shortcomings (Nahapiet and Ghoshal 1998; voluntary organizational association that inter- Powell, Koput, and Smith-Doerr 1996). Support actively engages multiple members in multilat- of these general statements in the social capital eral value chain activities, such as collaborative literature is substantial (see Parkhe, Wasserman, research, development, sourcing, production, and Ralston 2006). Much effort has also been or marketing of technologies, products, or ser- spent on examining contingencies when weak vices.” The interaction primarily occurs through ties or strong ties (referring to the closeness and pursuing joint project opportunities. Our study interaction frequency), as well as network focuses on a certain characteristic of such net- closure or structural holes, are to be preferred works that is rarely acknowledged: the fact that (Burt 1992, 1997; Coleman 1988; Granovetter the networks have a formal decision-making 1973). Whereas social capital (independent of body (the network TMT), which is responsible tie strength and network density) is generally for network strategies such as initiating new considered to have a wide range of positive joint projects and new joint product opportuni- consequences, the notion of the dark side of ties (Human and Provan 1997; Provan and social capital has also been mentioned in the Kenis 2007; Provan, Fish, and Sydow 2007). literature (Adler and Kwon 2002; Gargiulo and As a group of key decision-makers, the net- Benassi 2000). Almost two decades ago, Portes work’s TMT is established to work with and Sensenbrenner (1993) showed that the network strategy-making (Wincent, Thorgren, same ties that provided resources also and Anokhin 2013). Because those networks demanded resources that could outweigh the are created to initiate entrepreneurial network benefits. This has later been supported and projects among member firms, they often expanded upon. For example, some of the risks command government support. To ensure that associated with extensive social capital are the government spending is used in a proper opportunism and ineffective decision-making manner and in line with the expectations of (Villena, Revilla, and Choi 2011), structural and government agencies, the network TMT is personal lock-ins (Weber and Weber 2011), responsible for launching and initiating the time it consumes to maintain these relation- network member projects (Provan, Fish, and ships (Hansen, Podolny, and Pfeffer 2000; Sydow 2007). In other words, the network Molina-Morales and Martinez-Fernandez 2009), TMT’s tasks are consistent with the EO concept, and the vulnerability to competition (Burt 1997).

2230 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT In entrepreneurship, several studies highlighted different stakeholders to the network and the downside of social capital for creativity and member firms. From this also follows that other important outcomes of new venture cre- network TMTs differ from corporate TMTs in ation (Klyver, Evald, and Hindle 2011) or in the that the network TMT members belong to dif- context of corporate venture capital (CVC) ferent organizations, meaning that they have investments (Weber and Weber 2011). different “home bases.” Network TMTs may Whereas it is acknowledged that there is comprise individuals representing the network indeed a dark side to social capital, there are member firms as well as external actors such as still, however, calls for more research on pos- consultants, senior advisors, and industry sible trade-offs with social capital in different experts. This means that the individuals in the contexts and identification of mechanisms network TMT by definition have multiple roles. behind them (Payne et al. 2011; Zaheer, It does, however, also mean that the occur- Gözübüyük, and Milanov 2010). With our rence of multiple roles may be more evident in study, we add to this stream of research. some network TMTs than in others. There may Whereas much of the extant research on the be some network TMTs where the members dark sides of social capital has focused on the have extensive ties to organizations and firms costs associated with maintaining relationship beyond the network in question or hold TMT with others or the loss of objectivity and moni- positions in other formalized networks. toring, the present study takes a novel The conditions just described make network approach by highlighting how extensive ties TMTs’ work challenging enough in their own may translate into more role expectations to right. When combining it with what they are balance and could thus hinder, rather than expected to do for the network, we hypoth- facilitate, role performance. Although distinct, esize that ties to external constituents may be our approach builds on the contingency argu- negative for the development of network-level ments of social capital developed by Burt EO. In what follows, we elaborate on what is (1997) who suggested that social capital generally expected from the network TMT role. depended on the extent people were doing First, individuals assigned to network TMTs similar work, as peers mitigate opportunities to are expected to facilitate EO by building rou- act through mechanisms of competition and tines and structures for the support of innova- legitimacy. tive ideas and projects. This requires decisions In our conceptual development, we draw that promote proactive initiatives, experiment- upon role strain models in role theory (Kahn ing with frame-breaking renewal, and a willing- et al. 1964; Rizzo, House, and Lirtzman 1970) to ness to take risks by supporting uncertain theoretically expose the reasons for why TMT’s projects that use new and existing resources ties to external constituents are detrimental for effectively (Covin and Slevin 1986, 1989; the development of EO at the network level. Lumpkin and Dess 1996). To do this effectively, Role theory predicts that adding ties increases the network TMTs must keep in mind that they role expectations. Professional life involves are developing strategies for the entire acting upon organizational roles grounded in network, not any single firm. This means that role expectations; having many roles to navi- the network TMTs must not only work with the gate often entails being caught in the middle of challenging task of developing EO, but do so at various expectations. a higher level as it should attract and benefit all We reason that because of the specific con- the network member firms in projects, not just ditions surrounding network TMTs and expec- a select inner clique. Second, there are tations they are supposed to meet, such role unplanned, spontaneous discussions and idea navigation may be particularly challenging. exchange during TMT meetings. The TMT Though corporate TMTs and network TMTs meetings are rarely based on formal agendas both are responsible for implementing strategic exclusively, but rather unfold as issues are ran- decisions, they are different in how their domly brought to the table. To be able to make behavior is regulated. Whereas corporate TMTs a contribution to these discussions, it is critical are employed by the corporation and have job that the individuals assigned to the network descriptions that are formally regulated, this is TMT are updated on what is going on in the not the case for network TMTs. Instead, network firms and are aware of the relation- network TMTs’ behavior is primarily regulated ships among them. This means that TMT through role expectations in that they represent members must be alert and keep close to

WINCENT, THORGREN, AND AND ANOKHIN ANOKHIN 231 3 network operations. Third, acknowledging that ents restricts the sense of mutual reliance among network TMTs need fast decision-making to TMT members, which is problematic because for solve rather complicated tasks, they also need the development of EO, they must trust that the to avoid “interaction losses” and process diffi- initiatives rest on solid ground and that all culties to reach their highest effectiveness members will fulfill their obligations (Lumpkin (Forbes and Milliken 1999). As such, it is and Dess 1996). Thus, the restrictions that important that the individuals in the network follow adding external ties are likely to reduce TMT develop high-quality relationships in the quality of TMT actions and the overall which they communicate effectively and may network-level EO. To be effective and creative in trust each other. Finally, such networks are the network setting, network TMTs are encour- often formed when the industry is undergoing aged to concentrate on the strategy-making in structural changes, which creates particularly the networks they serve. harsh climates for network TMTs to work in Though professional involvements in other and makes it difficult to develop feasible and organizations can make network TMT members successful strategies. more effective by providing information and With this in mind, we posit that additional inspiration (Adler and Kwon 2002), especially engagements and involvement in ties to outside when they are nonredundant (Burt 1992), the constituents—ties to other organizations as extra commitments they add are likely to create well as interlocking directorates—of individu- information overloads. The competing expecta- als in the network TMT limit the capabilities of tions of fulfilling multiple role demands may the team to craft high-EO strategies. Network handicap TMT members’ ability to engage in the TMT members ought to be active in the original role (Singh 1998). Such additional ties strategy-making process in order to contribute can also expose them to conflicting assignments to network EO. It becomes important to reach and obligations where demands are incongruent consensus to be able to implement exciting, such that they are forced to make an explicit innovative, proactive, and risk-taking projects choice of what obligations to fulfill (Rizzo, that drive EO in the network. In this strategy- House, and Lirtzman 1970)—which automati- making process, the TMT member who initiates cally implies occasional anti-network choices if an idea must invest effort in actively convincing they are to keep the ties in their entirety. At the other members of the network TMT that this is aggregate level, a network TMT with members something worth pursuing. The implementa- sufficiently connected to outside constituents tion of any entrepreneurial project is preceded thus risks not being able to move forward inno- by one or several individuals on TMT identify- vative proposals, promote proactive initiatives, ing the project idea and then “selling” it to experiment with frame-breaking renewal, and others as a worthy project. Such a process take on uncertain projects that use new and consumes the initiator’s resources, but without existing resources effectively. these resources it would be difficult to facilitate Similar to the challenges of having ties to entrepreneurship at a network level. Having outside organizations, interlocking directorate ties to other organizations or different net- involvements to other networks may further works via interlocking directorates may sabo- restrict creative initiatives and the facilitation of tage the fulfillment of these important roles. EO strategies. There is a risk that such inter- When network TMT members engage exten- locks create distrust among the network TMT sively in such relationships beyond their net- members by spurring a belief that the networks, they effectively limit the resources (time work’s specific interests will not be cared for, and energy) they may employ to work on entre- prioritized, or protected. When network TMT preneurial proposals for the network TMT. members have loyalties to several strategic net- Thus, although external ties may provide certain works, the question naturally arises as to information benefits, we believe that they can which, if any, of these networks have the pri- significantly restrict the quality of TMT actions. ority, which leads to unexpected conse- Additional role expectations that accompany quences. For example, if an idea (such as a new external commitments interfere with TMT product or service) is proposed at a TMT members’ performance, suppress their creativ- meeting, the fact that some TMT members hold ity, and impede the innovativeness of their con- positions at other networks may instigate the tributions and actions (Kahn et al. 1964; Watson fear of the idea leak—especially if those net- 2007). High involvement with external constitu- works operate in the same industry. Similarly,

4232 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT the originality and intentions can be questioned negative impacts of being overly extended are when TMT members with multiple network less likely to manifest. Moreover, as TMT directorships propose ideas (i.e., was this idea members bring their own values and cognitions first proposed in another network? who really into crafting network strategy, their perceptions stands to gain if it is acted upon?). Obviously, and interpretations affect strategic choices they the uncertainties regarding the interlocking make (Hambrick and Mason 1984). This directorates can complicate the strategy-making includes those choices relating to the entrepre- process beyond manageable. neurial posture of their networks. Because of In short, both ties to external organizations limited time allocated to network TMT meetings, and TMT members’ interlocking directorates to making decisions that foster network EO is other networks can restrict the level of EO they particularly challenging. Decisions are more are able to instill in the networks they govern. easily reached when the network TMT members Hence, are homogeneous in terms of their expertise domains. Research on social categorization in H1: In strategic formalized multipartner net- organizational contexts has demonstrated that works, the network TMT members’ ties to people from the same expertise domain rely on external organizations are negatively similar biases and heuristics when making their related to network-level EO. decisions (Hogg and Terry 2000). In heterogeneous TMTs, such categorization can hamper H2: In strategic formalized multipartner net- the communication and shared understandings works, the network TMT members’ interlock- among the TMT members (Bunderson and ing directorates to other networks are Sutcliffe 2002), and thus prolong discussions negatively related to network-level EO. and the time it takes to reach a decision. Coupled with the lack of up-to-date information Contributions of Network TMT on network projects and intra-network relation- Homogeneity and Education Level ships that TMT members with ties to external As explained earlier, substantial amounts of constituents often exhibit, this makes the devel- ties to other organizations and networks may opment of EO in networks particularly trouble- hamper network-level EO by limiting the cre- some. On the contrary, we can expect that TMT ative initiatives among TMT members and by homogeneity smoothes the approval and imple- causing distrust within the group. Because of mentation of entrepreneurial initiatives and thus the limited time set off for network TMT meet- makes it easier to rapidly reach decisions, which ings, one of the greatest challenges for network becomes particularly important under time TMTs is to reach decisions that contribute to constraints. higher EO for the collective of multiple firms. Finally, individuals with identical back- In this study, we suggest two TMT attributes grounds tend to perceive themselves as sub- that may mitigate these problems: network groups and make stereotypical judgments of TMT homogeneity (which refers to the similar- those from other domains (Hogg and Terry ity of the functional domains of TMT members) 2000). Though not necessarily accurate, those and network TMT education level may lessen shared judgments facilitate trust within sub- the problem of extensive ties in fostering EO in groups, thus addressing another problem that networks. complicates the instilling of EO into networks Information overload of overly committed by TMT members with high social capital. That TMT members when they are engaged with is, entrepreneurial initiatives that otherwise outside constituents and lack of trust within the might be questioned are likely to be approved TMT that accompanies multiple allegiances of by think-alike members of homogenous TMTs. TMT members with external affiliations capture To state formally, the problematic sides of social capital in this context. TMT homogeneity in terms of func- H3: Network TMT homogeneity positively mod- tional background may help alleviate both these erates the relationship between network TMT problems. Because TMT members with identical members’ involvement ties to external orga- functional background tend to “speak the same nizations and network-level EO. language” (Hogg and Terry 2000), it takes less time to bring them up to speed when designing H4: Network TMT homogeneity positively mod- network entrepreneurial projects. Thus, the erates the relationship between network TMT

WINCENT, THORGREN, AND AND ANOKHIN ANOKHIN 233 5 members’ interlocking directorates to other Methods networks and network-level EO. Data We tested the hypotheses given in a five- Similarly, we also expect that a higher level year longitudinal sample consisting of 53 of education among TMT members will over- Swedish formalized strategic networks of SMEs. ride the negative influence of external ties on The networks tended to combine co-located the networks’ EO. The level of education firms in the same industry that shared the reflects the TMT’s cognitive ability and skills. common goal of fostering innovativeness and Research demonstrates that a higher level of entrepreneurship. The typical network had education of TMT members is associated with over 20 member firms and was about four years receptivity to innovation and strategic change, of age at time of data collection. Although some as they are better equipped to solve complex firms may have discontinued over the course of problems (Wiersema and Bantel 1992). That is, the years, and not all firms participated in all even if TMT members are time constrained network projects in all years (in fact, there were because of additional tie involvements, their instances when some networks did not pursue higher level of education makes it easier to entrepreneurial projects in certain years), all 53 develop and/or react to entrepreneurial initia- of the networks remained active during the tives that take member firms’ interests and specified period; there was no attrition in the capabilities into account. Additionally, higher network numbers. In fact, most networks grew levels of education make discussion of issues steadily and the average network size more less dysfunctional as it increases the TMT’s than doubled over the course of five years. tolerance for ambiguity, as well as their ability All networks had formal TMTs and kept to handle such ambiguity. various information registers available. The Though by itself higher education of TMT data contained complete five-year annual data members does not create trust to address the on various TMT characteristics and network remaining issue that may hinder the develop- processes in all 53 networks combined from ment of EO in networks, it makes it more both primary and secondary data sources. The likely that TMT members will analyze and see data used for testing the hypotheses had a the true merits of various entrepreneurial pro- panel nature, where each panel captured one- posals and will be able to see past the suspi- year observations for the variables studied. As cions of true motivations of those connected such, we obtained five usable panels based to outside constituents. Thus, even though it upon a total of 265 observations. Our panel does not by itself generate trust, it may alle- data covered a usable five-year period for all viate blatant distrust based on the fact that studied variables, with observations from each some TMT members may have multiple alle- network for each year. giances to balance, which is instrumental in In each network, we contacted the members instilling the EO in networks they govern. of the TMT listed as contact representatives. Besides, even if there is no time to discuss all They were all knowledgeable about the net- the pros and cons of ideas and each TMT work’s design, the TMT, its participating firms, member’s obligations, a network TMT with a and the activities pursued in the networks. Data higher level of education will generally be were collected through mailed-in question- more open to innovative projects because naires and follow-up telephone contacts. The they believe they have the ability to solve data included secondary information obtained problems that may arise along the way, no from the network information registers as well matter how complex. Stated formally, as primary information from the annualized survey data. H5: Network TMT education positively moderates the relationship between network TMT Measures members’ involvement ties to external orga- We measured network-level EO based on nizations and network-level EO. the modified Zahra’s (1996) operationalization. Though this measure has been subject of H6: Network TMT education positively moder- debate in the literature, our pretests with five ates the relationship between network TMT network representatives suggested that this members’ interlocking directorates to other scale was appropriate for this research setting networks and network-level EO. because it was not as abstract as the other

6234 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT options (such as the Miller and Friesen [1982] dent variables. As such, we recommend that scale) and thus provided greater face validity. these positive significant correlations are con- Because Zahra’s (1996; Table 1) scale was tar- sidered in their proper context and not be geting firms and not networks, we replaced the overgeneralized unnecessarily. recurrent phrase “our company” with “our Drawing upon previous literature on TMT network” and we omitted items 4, 8, and 11–14 and directorates, we operationalized network because our pretests indicated that those items TMT ties to other external organizations as the could not be used in the network context (i.e., total number of links the TMT members obviously, networks cannot acquire patents or reported to outside organizations and firms. companies nor can they develop operational Over the course of five years, network TMT ties changes for business units, which is what these grew steadily both in terms of the average items capture). In total, we used eight items to number of ties (which went up by a factor of capture EO (e.g., “The network has introduced two) and in terms of the extremes (the a large number of new products to the market,” maximum number of ties a network had “The network has maintained world-class increased by 25 percent during that time). We research and development facilities,” and “The operationalized the TMT interlocking director- network has pioneered the development of ates to other networks as the total number of breakthrough innovations”). We thus relied positions that the TMT members held on other upon a unified measure (α = 0.76) of key networks’ TMTs.1 These measurements have a aspects of EO at a network level, including long history in the TMT literature (see e.g., innovation, risk-taking, and proactiveness. Boyd 1990; Richardson 1987). This implies that To further establish the appropriateness of we follow the accepted operationalization to the operationalization, we correlated the scale count the relationships surrounding the TMTs with alternative measures available from infor- and considered interlocks to be higher the mation registers kept by the networks. For more the TMTs served multiple networks. Inter- instance, our measure was significantly corre- locking directorates fluctuated in intensity over lated to the number of new products for each the years with no clear pattern emerging. We network (p < .05), a possible proxy for operationalized network TMT homogeneity as innovativeness. The scale was significantly the degree of similarity of professional back- positively correlated (p < .001) with the amount grounds of individuals on the network TMT. of membership funding invested in the net- This is also in line with previous measures in works. We reasoned that this measure captures the TMT literature that generally builds upon the extent the network aimed to pursue bold Blau’s (1977) calculation of homogeneity. The endeavors and high-risk projects and thus variable was so scaled that low values referred could be thought of as a proxy for risk-taking. to the high number of professions represented Finally, the scale was significantly correlated among the TMT members whereas high values (p < .05) to the number of network projects, referred to TMTs that were comprised of indi- which we see as indicative of the overall viduals with similar professional backgrounds. proactiveness of the networks. Unfortunately, We operationalized network TMT education these three validation variables contained pro- level as the number of TMT members with at hibitively high levels of missing values (≈50 least a university degree. This is also based on percent), which prevented us from rerunning traditional measurements for education compo- our models using them as alternative depen- sition (Becker 1983; Coff 1999).

1We also considered an alternative operationalization where we looked at the proportion of TMT members with ties to external organizations and interlocking directorates. While doing so, as previously, we simultaneously controlled for network size. The results were consistent with the ones reported in Table 2 in terms of coefficient signs. Significance level was generally somewhat lower with the alternative operationalization of these constructs: the direct effect of ties to external organizations attained significance of p < .10 (p < .01 for the operationalization reported in Table 2); the direct effect of interlocking directorates previously significant at p < .10 has lost its significance although retained the sign; the interaction of homogeneity and ties to external organizations has actually gained in significance (p < .01 as opposed to p < .05); and the interaction of education and ties to external organizations has dropped its significance somewhat (p < .10 as opposed to p < .001).

WINCENT, THORGREN, AND AND ANOKHIN ANOKHIN 235 7 8 236

Table 1 Descriptive Statistics and Correlationsa

Variable 1 23456789101112

1. Network Size 1.00 2. Network Age 0.17** 1.00 ORA FSALBSNS MANAGEMENT BUSINESS SMALL OF JOURNAL ORA FSALBSNS MANAGEMENT BUSINESS SMALL OF JOURNAL 3. Network TMT Size 0.16** 0.31*** 1.00 4. TMT Meeting Frequency 0.23*** 0.35*** 0.50*** 1.00 5. TMT Homogeneity 0.21*** 0.23*** 0.40*** 0.67*** 1.00 6. TMT Education 0.12* 0.11 0.32*** 0.55*** 0.56*** 1.00 7. Wood Industry 0.05 0.13* −0.02 0.01 −0.00 0.03 1.00 8. IT Industry −0.00 −0.06 0.27*** −0.05 −0.16** −0.03 −0.17** 1.00 9. Metal Industry −0.05 −0.15* −0.17** −0.03 −0.01 0.11 −0.39** −0.23** 1.00 10. TMT Ties to Other −0.02 −0.04 0.24*** 0.21*** 0.21*** 0.35*** −0.05 0.16** −0.15* 1.00 Organizations 11. TMT Network −0.00 0.09 0.04 0.22*** −0.25*** 0.33*** 0.02 −0.02 0.06 0.17** 1.00 Interlocks 12. Network-Level EO −0.00 −0.04 0.19** 0.03 0.05 0.07 0.01 0.32*** 0.06 −0.12† −0.16** Mean 23.81 3.81 5.51 2.88 17.61 1.06 0.23 0.09 0.34 2.03 0.45 3.43 S.D. 73.85 6.37 7.44 3.13 3.08 2.02 0.42 0.29 0.47 8.01 1.65 1.09 Min 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.00 Max 180.00 43.00 54.00 15.00 20.00 9.00 1.00 1.00 1.00 50.00 12.00 6.01 Q1 0.00 0.00 0.00 0.00 16.00 0.00 0.00 0.00 0.00 0.00 0.00 2.50 Q3 25.00 6.00 8.00 5.00 20.00 1.00 0.00 0.00 1.00 0.00 0.00 4.26

aEO, entrepreneurial orientation; IT, information technology; S.D., standard deviation; TMT, top management team. N = 265. Two-tailed significance. †p < .10. *p < .05. **p < .01. ***p < .001. We also captured data on a set of control Results variables. We controlled for network size mea- Table 1 presents means, standard deviations, sured as the number of member firms partici- and the correlation matrix for the variables. At pating in each network and measured network the time of the study, the average network was age as the number of years since the network about four years of age, contained close to 25 started. We also controlled for network TMT member firms, and had a TMT team of five to size measured as the number of members on six individuals. We further observe that as the network TMT and frequency of TMT meet- expected, the hypothesized main effects of ings operationalized as the number of TMT ties and interlocking directorates to other meetings per year. In addition to the control networks are significantly correlated with variables given, we included three dummies to network-level EO. partial out industry variance (wood, informa- Table 2 presents the results of the hypoth- tion technology, and metal). eses testing and displays coefficient estimates and standard errors. In Model 1, where we Data Analysis entered the control variables, we noticed sig- We used hierarchical analysis to test the nificant influences of network size, network hypotheses. In the first step, we entered age, network TMT size, network TMT educa- control variables; in the second step, the main tion level, and the industry dummies on effects of ties and interlocking directorates network-level EO. (thus testing H1 and H2); and in the third In Model 2, where we entered the indepen- step, we added the interaction terms to test dent variables, network TMT members’ ties the moderation hypotheses (H3 through H6). to external organizations had a negative Because of the panel character of our data, we direct influence on network-level EO (β = used Prais–Winsten regression with panel- −0.21, p < .001), as did network TMT members’ corrected standard errors (PCSE). Although the interlocking directorates to other networks feasible generalized least squares (FGLS) (β = −0.16, p < .001), thus supporting H1 approach could also be used when error and H2. structures are characterized by panel In Model 3, where we added the interactions heteroskedasticity, panel autocorrelation, and of our independent variables with network TMT contemporaneous correlation, the PCSE homogeneity and network TMT education level approach is preferable because FGLS may to the set of predictors, we found the interaction produce standard errors that lead to extreme term between TMT homogeneity and network overconfidence. PCSE adequately accounts for TMT members’ ties to external organizations the panel structure and allows for significant (β = 0.32, p < .05), thus supporting heteroskedasticity, within-panel AR(1) serial H3. In other words, TMT homogeneity compen- correlation, and cross-sectional dependence, sates for the negative effect of TMT members’ and is more conservative than FGLS (Beck and ties to outside constituents on network-level EO. Katz 1995; Greene 2000).2 H4, which suggested a similar interaction We standardized the variables to minimize between TMT homogeneity and network TMT the risk of interaction-related multicollinearity members’ interlocking directorates, was not as suggested by Aiken and West (1991). Multi- supported. H5 found strong support: there is a collinearity diagnostics did not reveal serious positive interaction of network TMT’s ties to threats. The condition number of 9.98 is below external organizations and TMT education. Like both stringent (15.0) and lax (30.0) cutoff homogeneity, TMT education compensates for values (Cohen et al. 2003). That is, no evidence the negative impact of TMT ties to outside of multicollinearity was found. constituents on network-level EO (β = 0.34,

2As a robustness check, we retested our models using alternative estimation techniques including ordinary least squares (OLS) regression, OLS regression with standard errors corrected for cluster effects at the network level; feasible generalized least squares estimation, and the regression with Driscoll and Kraay’s (1998) heteroscedasticity-consistent standard errors that are robust to cross-sectional and temporal dependence as introduced in Hoechle (2007). The results remain consistent across different models. Results are available from the authors upon request.

WINCENT, THORGREN, AND AND ANOKHIN ANOKHIN 237 9 10 238 Table 2 Effects on Network-Level EOa

Model 1 Model 2 Model 3

Coefficient Standard Coefficient Standard Coefficient Standard Error Error Error ORA FSALBSNS MANAGEMENT BUSINESS SMALL OF JOURNAL ORA FSALBSNS MANAGEMENT BUSINESS SMALL OF JOURNAL Controls Variables Network Size 0.13† (0.07) 0.11† (0.07) 0.10† (0.06) Network Age −0.07† (0.06) −0.08† (0.05) −0.10† (0.06) Network TMT Size 0.16* (0.07) 0.16* (0.06) 0.20** (0.07) TMT Meeting Frequency 0.01 (0.08) 0.02 (0.08) 0.05 (0.07) TMT Homogeneity 0.12 (0.09) 0.09 (0.08) 0.15† (0.09) TMT Education 0.06† (0.08) 0.07† (0.09) 0.18† (0.09) Wood Industry Dummy 0.12* (0.06) 0.10† (0.06) 0.08 (0.06) IT Industry Dummy 0.31*** (0.08) 0.33*** (0.08) 0.25** (0.08) Metal Industry Dummy 0.14* (0.07) 0.11† (0.07) 0.09 (0.07) Independent Variables TMT Ties to Other Organizations −0.21*** (0.06) −0.38** (0.015) TMT Interlocking Directorates −0.16*** (0.05) −0.15* (0.08) Two-Way Interactions TMT Ties to Other Organizations × TMT Homogeneity 0.32* (0.17) TMT Interlocking Directorates × TMT Homogeneity −0.05 (0.12) TMT Ties to Other Organizations × TMT Education 0.34*** (0.08) TMT Interlocking Directorates × TMT Education −0.11 (0.07) Constant −0.01 (0.06) −0.06 (0.05) −0.03 (0.06) R2 0.16 0.23 0.27 χ2; df; significance 59.12; 9;*** 91.31; 11;*** 282.23; 15;***

adf, degree of freedom; EO, entrepreneurial orientation; IT, information technology; TMT, top management team. N = 265. Standard errors are in parentheses. Two-tailed tests. †p < .10. *p < .05. **p < .01. ***p < .001. Figure 1 Figure 2 Interaction Effects of Network Interaction Effects of Network Top Management Team (TMT) Top Management Team (TMT) Homogeneity on Network-Level Education on Network-Level Entrepreneurial Orientation (EO) Entrepreneurial Orientation (EO)

p < .001). Finally, H6 was not supported. The relationship between network TMT team’s inter- effect, whereas for heterogeneous TMTs and locking directorates and EO does not vary with those with lower education, the effect is strictly TMT education. negative. In order to probe the issue of causality and possible endogeneity, we reanalyzed our Discussion models using a temporal lag of one year In the present study, we build on research between the dependent and predictor vari- acknowledging that social capital may have a ables. The results were virtually identical to the “dark side” (Gargiulo and Benassi 2000; Portes ones reported in Table 2, but because of loss of and Sensenbrenner 1993) by presenting and one year’s worth of observations, the signifi- studying a context in which the conditions cance of the results diminished. make extensive ties a liability for role perfor- To illustrate the nature of the observed inter- mance. Whereas prior studies on the dark side action effects, we plotted significant interac- of social capital have primarily emphasized the tions at one standard deviation below and one cost associated with maintaining the relation- standard deviation above the mean as recom- ship (Hansen, Podolny, and Pfeffer 2000; mended by Aiken and West (1991). Figures 1 Molina-Morales and Martinez-Fernandez 2009), and 2 demonstrate that the relationship the present study takes on another dimension between TMT ties to external organizations and by shedding light on how extensive ties bring network-level EO was consistent with our challenges in terms of creating multiple roles predictions (H3 and H5). Specifically, for that are all associated with expectations and network TMTs with higher team homogeneity obligations, resulting in role strains that are and education level, the negative association negative for role performance. between such connections to external constitu- Utilizing the mechanisms proposed in role ents and network-level EO weakens substan- strain theory (Kahn et al. 1964; Rizzo, House, tially. Overall, however, TMTs seem to have and Lirtzman 1970) allowed us to theorize on difficulties converting their social capital into why it can be expected that social capital higher EO. TMTs with higher levels of homo- among strategic formalized networks’ TMTs— geneity and education merely find the negative captured by the network TMTs’ ties to other relationship eliminated, suggesting a null organization and interlocking directorates—

WINCENT, THORGREN, AND AND ANOKHIN ANOKHIN 239 11 may be negative for these TMTs role perfor- gests that the role of being a member on a mance in terms of EO. Constraints that network’s TMT is of such a character that it is accompany the development of new links to best not to engage those who are heavily external constituents include limited resources involved in other things. We believe this that network TMTs have available to govern insight has raised the awareness that the their networks and a delicate balancing act network TMT design is of great importance for between internal and external role expecta- how the network will perform. We encourage tions, in combination with distrust within the scholars to further this stream of research by network TMT. Together, this makes instilling of also examining possible interaction effects EO into the networks problematic. Overall, the between network TMT design and characteris- results indicate that problems with social tics of the network firms (e.g., density, com- capital can be found in this specific context. mitment, and interorganizational trust) on When it comes to the moderating effects of network performance. homogeneity and education, the results do, Further, although research on social capital however, reveal a distinction between ties to has acknowledged a “dark side” of social external organizations and interlocking direc- capital, it has primarily been either conceptual torates. Whereas both are significantly and or addressing the costs with maintaining rela- negatively related to network-level EO, only tionships. In this paper, we have made a con- ties to external organizations, not interlocking tribution by suggesting that social capital may network directorates, are significantly moder- result in diverging role expectations and loyalty ated by the TMTs’ homogeneity and level of issues, which may make it difficult to perform education. We can only speculate why these well. This implies that research on social capital differences emerge. Possibly, the problems would benefit from also taking role strains and with interlocking network directorates are loyalty issues into consideration when examin- greater than expected, meaning that the dis- ing the darker sides of social capital. trust that comes with being on several network These contributions should be viewed in light TMTs is difficult to solve with the kind of of the study’s limitations. First, the study is on a lubricant that homogeneity and education can specific kind of formalized networks. They create. Hence, we encourage future research to might not be directly generalized to other busi- look further into this issue and try to identify ness networks that may be informal or may have what, if anything, can mitigate the problems other governance units distinct from the with network TMT members serving several network TMTs studied here. Second, this study network TMTs at the same time. lack proper instrumental variables that would Our results have implications for the for- have enabled us to tease out the issue of causal- malized networks and social capital literatures. ity and endogeneity unequivocally. We did, With regard to the first, previous research has however, employ a lag dependent variable acknowledged that formalized strategic net- approach to address this issue. The results were works are an increasing phenomenon around very consistent with the ones reported in the world (Human and Provan 1997; Lipparini Table 2, but it is desirable to revisit this issue in and Sobrero 1994; Miller and Besser 2005), subsequent research. Third, we did not have the has provided descriptive studies on how they information on informal networks that could are governed (Provan and Kenis 2007; Provan, have affected some of tie- and interlocking Fish, and Sydow 2007), and presented some directorates-building activity by TMT members, empirical evidence on how the network such as their participation in alumni networks. members can best reach their objectives with Because we guaranteed our respondents per- cooperating (Fukugawa 2006; Sherer 2003). sonal anonymity, this information was off-limits. There is, however, very scant knowledge on Fourth, we used a modified version of Zahra’s the network TMTs’ effects on the network’s (1996) EO operationalization and we believe performance. To our knowledge, no studies more attention to scale development at the have examined how the networks’ perfor- network level is needed. Because some items mance is impacted by the network TMT’s ties made little sense to use at the network level, we outside the network in question. The negative needed to revise the scale. Many of these items relationship between external ties (to other concerned proactive elements, such that one external organizations and through interlock- could argue that the scale we used has become ing directorates) identified in this paper sug- somewhat unbalanced.

12240 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Conclusion Resource Dependence Model,” Strategic Our aim was to introduce a new way of Management Journal 11, 419–430. looking upon the dark side of social capital. We Bunderson, J. S., and K. M. Sutcliffe (2002). studied network TMTs and showed more tie “Comparing Alternative Conceptualizations involvements of such TMTs to be negatively of Functional Diversity in Management related to EO. Our most important contribution Teams: Process and Performance Effects,” is bringing an alternative explanation for the Academy of Management Journal 45, 875– dysfunctional side of tie formation. Although 893. research has emphasized shortcomings such as Burt, R. S. (1992). Structural Holes. New York: opportunism, lock-ins, and resource consump- Cambridge University Press. tion to maintain these relationships, we suggest ——— (1997). “The Continent Value of Social that higher tie involvements increases role Capital,” Administrative Science Quarterly expectations, which makes network perfor- 42(2), 339–365. mance and entrepreneurial outlook caught in Chen, M. H., Y. C. Chang, and S. C. Hung the middle. We believe that our ﬁndings (2008). “Social Capital and Creativity in R&D provide new insights into the understanding of Project Teams,” R&D Management 38, social capital as well as about the governance 21–34. of formalized strategic networks, and also open Coff, R. W. (1999). “How Buyers Cope with up for several new research questions to be Uncertainty When Acquiring Firms in addressed. Knowledge Intensive Industries: Caveat Emptor,” Organization Science 10, 144–161. References Cohen, J., P. Cohen, S. West, and L. Aiken Adler, P. S., and S.-W. Kwon (2002). “Social (2003). Applied Multiple Regression/ Capital: Prospects for a New Concept,” Correlation Analysis for the Behavioral Sci- Academy of Management Review 27, 17–40. ences. Mahwah, NJ: Lawrence Erlbaum Aiken, L. S., and S. G. West (1991). Multiple Associates Publishers. Regression: Testing and Interpreting Interac- Coleman, J. S. (1988). “Social Capital in the tions. Newbury Park, CA: Sage. Creation of Human Capital,” The American Aldrich, H. E., and M. A. Martinez (2001). Journal of Sociology 94, 95–120. “Many Are Called, but Few Are Chosen: An Covin, J. G., and D. P. Slevin (1986). Evolutionary Perspective for the Study of “The Development and Testing of Entrepreneurship,” Entrepreneurship Theory an Organizational-Level Entrepreneurship and Practice 25, 41–56. Scale,” in Frontiers of Entrepreneurship Ariño, A., R. Ragozzino, and J. J. Reuer (2008). Research. Eds. R. Ronstadt, J. A. Hornaday, R. “Alliance Dynamics for Entrepreneurial Peterson, and K. H. Vesper. Wellesley, MA: Firms,” Journal of Management Studies 45, Babson College, 628–639. 147–168. ——— (1989). “Strategic Management of Small Arregle, J.-L., M. A. Hitt, D. G. Sirmon, and P. Firms in Hostile and Benign Environments,” Very (2007). “The Development of Organi- Strategic Management Journal 10, 75–87. zational Social Capital: Attributes of Family Davidsson, P., and B. Honig (2003). “The Role Firms,” Journal of Management Studies 44, of Social and Human Capital Among Nascent 73–95. Entrepreneurs,” Journal of Business Ventur- Beck, N., and J. N. Katz (1995). “What to Do ing 18, 301–331. (and Not to Do) with Time-Series Cross- De Carolis, D. M., B. E. Litzky, and K. A. Section Data,” The American Political Eddleston (2009). “Why Networks Enhance Science Review 89, 634–647. the Progress of New Venture Creation: The Becker, G. (1983). Human Capital: A Theoreti- Inﬂuence of Social Capital and Cognition,” cal and Empirical Analysis with Special Ref- Entrepreneurship Theory and Practice 33, erence to Education. Chicago, IL: University 527–545. of Chicago Press. Driscoll, J., and A. C. Kraay (1998). “Consistent Blau, P. M. (1977). Inequality and Heterogene- Covariance Matrix Estimation with Spatially ity: A Primitive Theory of Social Structure. Dependent Data,” The Review of Economics New York: Free Press. and Statistics 80, 549–560. Boyd, B. (1990). “Corporate Linkages and Forbes, D. P., and F. J. Milliken (1999). “Cog- Organizational Environment: A Test of the nition and Corporate Governance: Under-

WINCENT, THORGREN, AND AND ANOKHIN ANOKHIN 241 13 standing Boards of Directors as Strategic and Involvement in Multipartner Alliances,” Decision Making Groups,” Academy of Man- Academy of Management Journal 50, 578– agement Review 24, 489–505. 604. Fukugawa, N. (2006). “Determining Factors in Lipparini, A., and M. Sobrero (1994). “The Glue Innovation of Small Firm Networks: A Case and the Pieces: Entrepreneurship and Inno- of Cross Industry Groups in Japan,” Small vation in Small-Firm Networks,” Journal of Business Economics 27, 181–193. Business Venturing 9, 125–140. Gargiulo, M., and M. Benassi (2000). “Trapped Lumpkin, G. T., and G. G. Dess (1996). “Clari- in Your Own Net? Network Cohesion, Struc- fying the Entrepreneurial Orientation Con- tural Holes, and the Adaptation of Social struct and Linking It to Performance,” Capital,” Organization Science 11, 183– Academy of Management Review 21, 135– 196. 172. Granovetter, M. S. (1973). “The Strength of Lyon, D. W., G. T. Lumpkin, and G. G. Dess Weak Ties,” The American Journal of Soci- (2000). “Enhancing Entrepreneurial Orienta- ology 78, 1360–1380. tion Research: Operationalizing and Measur- Greene, W. H. (2000). Econometric Analysis. ing a Key Strategic Decision Making Upper Saddle River, NJ: Prentice Hall. Process,” Journal of Management 26, 1055– Hambrick, D. C., and P. A. Mason (1984). 1085. “Upper Echelons: The Organization as a Miller, D., and P. H. Friesen (1982). “Innovation Reflection of Its Top Managers,” Academy of in Conservative and Entrepreneurial Firms: Management Review 9, 193–206. Two Models of Strategic Momentum,” Stra- Hansen, M. T., J. M. Podolny, and J. Pfeffer tegic Management Journal 3(1), 1–25. (2000). “So Many Ties, So Little Time: A Task Miller, N. J., and T. L. Besser (2005). “Exploring Contingency Perspective on the Value of Decision Strategies and Evaluations of Per- Social Capital in Organizations,” Working formance by Networked and Non- Paper No. 00–029, Harvard Business School, Networked Small U.S. Businesses,” Journal Boston. of Developmental Entrepreneurship 10, 167– Hoang, H., and B. Antoncic (2003). “Network- 186. Based Research in Entrepreneurship: A Criti- Molina-Morales, F. X., and M. T. cal Review,” Journal of Business Venturing Martinez-Fernandez (2009). “Too Much Love 18, 165–187. in the Neighborhood Can Hurt: How an Hoechle, D. (2007). “Robust Standard Errors for Excess of Intensity and Trust in Relation- Panel Regression with Cross-Sectional ships May Produce Negative Effects on Dependence,” Stata Journal 7, 281–312. Firms,” Strategic Management Journal 30, Hogg, M. A., and D. J. Terry (2000). “Social 1013–1023. Identity and Self-Categorization Processes in Nahapiet, J., and S. Ghoshal (1998). “Social Organizational Contexts,” Academy of Man- Capital, Intellectual Capital, and the Organi- agement Review 25, 121–140. zational Advantage,” Academy of Manage- Human, S. E., and K. G. Provan (1997). “An ment Review 23, 242–266. Emerging Theory of Structure and Outcomes Packalen, K. A. (2007). “Complementing in Small-Firm Strategic Manufacturing Net- Capital: The Role of Status, Demographic works,” Academy of Management Journal Features, and Social Capital in Founding 40, 368–403. Teams’ Abilities to Obtain Resources,” Entre- Kahn, R. L., D. Wolfe, A. Quinn, J. D. Snoek, preneurship Theory and Practice 31, 873– and R. Rosenthal (1964). Organizational 891. Stress: Studies in role Conflict and Role Parkhe, A., S. Wasserman, and D. Ralston Ambiguity. New York: Wiley. (2006). “New Frontiers in Network Theory Klyver, K., M. R. Evald, and K. Hindle (2011). Development. Introduction to special Topic “Social Networks and New Venture Creation: Forum,” Academy of Management Review The Dark Side of Networks,” in Handbook of 31, 560–568. Research on New Venture Creation. Eds. K. Payne, G. T., C. B. Moore, S. E. Griffis, and Hindle and K. Klyver. Cheltenham: Edward C. W. Autry (2011). “Multilevel Challenges Elgar, 145–159. and Opportunities in Social Capital Lavie, D., C. Lechner, and H. Singh (2007). “The Research,” Journal of Management 37(2), Performance Implications of Timing of Entry 491–520.

14242 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Portes, A., and J. Sensenbrenner (1993). Competitive Advantage,” Journal of Man- “Embeddedness and Immigration: Notes on agement Studies 44, 119–142. the Social Determinants of Economic Villena, V. H., E. Revilla, and T. Y. Choi (2011). Action,” American Journal of Sociology 98, “The Dark Side of Buyer-Supplier Relation- 1320–1350. ships: A Social Capital Perspective,” Journal Powell, W. W., K. W. Koput, and L. Smith-Doerr of Operations Management 29, 561–576. (1996). “Interorganizational Collaboration Watson, J. (2007). “Modeling the Relationship and the Locus of Innovation: Networks of between Networking and Firm Perfor- Learning in Biotechnology,” Administrative mance,” Journal of Business Venturing Science Quarterly 41, 116–145. 22(6), 852–874. Provan, K. G., and P. Kenis (2007). “Modes of Weber, C., and B. Weber (2011). “Exploring the Network Governance: Structure, Manage- Antecedents of Social Liabilities in CVC ment, and Effectiveness,” Journal of Public Triads: A Dynamic Social Network Perspec- Administration Research and Theory 19, tive,” Journal of Business Venturing 26, 255– 1–24. 272. Provan, K. G., A. Fish, and J. Sydow (2007). Wiersema, M. F., and K. A. Bantel (1992). “Top “Interorganizational Networks at the Management Team Demography and Corpo- Network Level: A Review of the Empirical rate Strategic Change,” Academy of Manage- Literature on Whole Networks,” Journal of ment Journal 35, 91–121. Management 33, 479–516. Wiklund, J., and D. Shepherd (2003). Richardson, R. J. (1987). “Directorship Inter- “Knowledge-Based Resources, Entrepre- locks and Corporate Profitability,” Adminis- neurial Orientation, and the Performance of trative Science Quarterly 32, 367–386. Small and Medium-Sized Businesses,” Stra- Rizzo, J. R., R. J. House, and S. I. Lirtzman tegic Management Journal 24, 1307–1314. (1970). “Role Conflict and Ambiguity in Wincent, J., S. Thorgren, and S. Anokhin Complex Organizations,” Administrative (2013). “Managing Maturing Government- Science Quarterly 15, 150–163. Supported Networks: The Shift from Moni- Rothaermel, F. T. (2001). “Incumbent’s Advan- toring to Embeddedness Controls,”British tage through Exploiting Complementary Journal of Management 24, 480–497. Assets via Interfirm Cooperation,” Strategic Wincent, J., S. Thorgren, and S. Anokhin Management Journal 22, 687–699. (2014). “Entrepreneurial Orientation and Sherer, S. A. (2003). “Critical Success Factors Network Board Diversity in Network Orga- for Manufacturing Networks as Perceived by nizations,” Journal of Business Venturing Network Coordinators,” Journal of Small 29, 327–344. Business Management 41, 325–345. Zaheer, A., R. Gözübüyük, and H. Milanov Singh, J. (1998). “Striking a Balance in (2010). “It’s the Connections: The Network Boundary-Spanning Positions: An Investiga- Perspective in Interorganizational Research,” tion of Some Unconventional Influences of Academy of Management Perspectives 24, Role Stressors and Job Characteristics on Job 62–77. Outcomes of Salespeople,” Journal of Mar- Zahra, S. A. (1996). “Governance, Ownership, keting 62, 69–86. and Corporate Entrepreneurship: The Mod- Teng, B.-S. (2007). “Corporate Entrepreneur- erating Impact of Industry Technological ship Activities through Strategic Alliances: Opportunities,” Academy of Management A Resource-Based Approach toward Journal 39, 1713–1735.

WINCENT, THORGREN, AND AND ANOKHIN ANOKHIN 243 15 Journal of Small Business Business Management Management 2016 2014 54(1), ••(••), pp. pp. 244–261 ••–•• doi: 10.1111/jsbm.12142 How Small Business Advisory Program Delivery Methods (Collective Learning, Tailored, and Practice-Based Approaches) Affect Learning and Innovation* by Sukanlaya Sawang, Rachel Parker, and Damian Hine

Past studies relate small business advisory program effectiveness to advisory characteristics such as advisory intensity and scope. We contribute to existing literature by seeking to identify the impact of different advisory program methods of delivery on learning and subsequent firm innovation behavior. Our research is based on a survey of 257 Australian firms completing small business advisory programs in the three years preceding the research. We explore the range of small business advisory program delivery methods in which our surveyed firms participated and, with reference to the literature on organizational learning and innovation, we analyze predictors of firms’ learning ability and innovativeness based on the identified delivery methods. First, we found that business advisory programs that involved high levels of collective learning and tailored approaches enhanced firms’ perceptions of their learning of critical skills or capabilities. We also found that small business advisory programs that were delivered by using practice-based approaches enhanced firms’ subsequent organizational innovation. We verified this finding by testing whether firms that have participated in small business advisory services subsequently demonstrate improved behavior in terms of organizational innovativeness, when compared with matched firms that have not participated in an advisory program.

petitive advantage by increasing strategic Introduction knowledge through these business advisory The use of publicly funded business advi- programs (Bennett and Robson 2004). Through sory support has gradually increased through- the resource-based theory lens, small firms out the manufacturing and service sectors seek advice in order to increase their competi- (Bennett and Robson 1999a, 1999b). This is tive capacity (Bennett and Robson 2005). This because there is increasing evidence that small is because firm’s growth is constrained by man- business advisory programs improve firms’ agement capability and a provision of business growth and performance (Bennett 2008; Mole advice will lift the key constrains to growth et al. 2009). Firms can thus increase their com- (Mole and Keogh 2009; Mole et al. 2009). Based

*This work was supported by the Australian Research Council (grant number LP0776386). Sukanlaya Sawang is a senior lecturer of Organisational Behaviour and Management in QUT Business School (Management) at Queensland University of Technology. Rachel Parker is a dean, Research Development and professor in Management in QUT Business School at Queensland University of Technology. Damian Hine is an associate professor in Strategy in UQ Business School at The University of Queensland. Address correspondence to: Sukanlaya Sawang, QUT Business School (Management), Queensland Uni- versity of Technology, GPO Box 2434, Brisbane, QLD 4001, Australia. E-mail: [email protected].

244 JOURNALSAWANG, OF SMALL PARKER, BUSINESS AND MANAGEMENT HINE 1 on empirical evidence, improving small and innovation, we analyze predictors of firms’ medium-sized enterprises’ (SMEs) performance learning ability and innovativeness based on leads to an increase in overall economic per- the identified delivery methods. We also formance, and therefore government pays spe- compare the level of firms’ innovativeness with cific attention to providing effective business a matched sample of assisted and unassisted advisory support (Porter and Ketels 2003). firms. Typically, government support to SMEs can be The remainder of this paper is organized as categorized in terms of either soft or hard follows. In the next section, we review the schemes (Rotger, Gørtz, and Storey 2012). The business advisory literature. In the following concept of soft scheme refers to guided prepa- two sections, we then identify the delivery ration including the provision of advice, edu- approaches for business advisory programs— cation, and business opportunity awareness to the subject of this research—and include the potential and existing business owners hypothesis development. Following this, we (Chrisman, McMullan, and Hall 2005). Hard focus on the data, method, analysis, and schemes typically involve the provision of results. The final sections conclude the paper direct financial assistance to the firm, such as with a discussion of the implications of the loans or grants. Our study focuses on soft results. support in the form of advice to SMEs. Past studies relate advisory program effec- Small Business Advisory Services tiveness to advisory characteristics such as The role and objective of government policy advisory intensity and scope of advice (e.g., in relation to small business and entrepreneur- Kosters and Obschonka 2011; Robson and ship can be described in terms of three broad Bennett 2000). For example, Kosters and approaches (Pearce 1998; Peterson 1988). Obschonka (2011) examined the effectiveness First, the laissez-faire approach is when govern- of business advice based on the intensity of the ment intervention involves a withdrawal from program, which was measured in terms of the economic sphere, such as telecommuni- provision of ongoing support and repeated, cations deregulation leading to new entrants regular contact between the advisors and the (Rakowski and Bejou 1992). Second, the firms. The authors concluded that programs limited environmental approach envisages gov- incorporating more intensive advice had a ernment’s role as active stewardship to elimi- greater impact on businesses than less inten- nate discrimination against small business. The sive advisory programs. Similar findings were type of intervention in this approach includes reported by Mole et al. (2011) when they exam- monitoring for price and wages inflation, inter- ined the effectiveness of advisory programs est rates, and taxation. Third, the strategic inter- based on breadth and depth of the advice. ventionist approach involves an activist role for Their results indicated that a more in-depth government in providing support for small approach, in which advisors spent more time business in the form of financial aid and busi- assisting a smaller proportion of SMEs, worked ness advisory programs. Research states that better in terms of maximizing the impact on governments used the resource-based theory as firm performance. a guide on how to intervene to promote small Our study contributes to existing literature business growth and development (Bennett on the impact of business advice delivery and Robson 2005; Chrisman, McMullan, and methods on firm outcomes. The effectiveness Hall 2005). As a result, governments in most of small business advisory programs is not only countries, including Australia, have attempted about “what” the advisors deliver (the content to address the resource gap among small busi- of advice), but also about “how” programs are nesses by supplying assistance either through delivered (the delivery method) (Bennett and subsidies or advisory advice. Robson 2004). Our contribution is distinguish- Through the resource-based lens, managers able from prior research, which has focused on improve management capabilities and increase the intensity of advice, in that it looks at a firm’s resources through business advisory ser- range of other delivery method characteristics. vices (Mole and Keogh 2009; Mole et al. 2009). We explore the range of small business advi- Because of individual differences and needs, sory program delivery methods in which our the resource-based theory suggests that the surveyed firms participated and, with reference level of interaction may differ from client to to the literature on organizational learning and client (Chrisman, McMullan, and Hall 2005).

2 JOURNALSAWANG, OF SMALL PARKER, BUSINESS AND MANAGEMENT HINE 245 There are various forms of small business advi- drive manufacturing performance to meet with sory programs depending upon the level of international standards through process and interaction between advisor and client (Bennett technology knowledge transfer. Lean Manufac- and Robson 2005; Chrisman, McMullan, and turing program is designed to train SMEs in Hall 2005; Mole et al. 2011). Chrisman and relation to inventory management, capacity McMullan (2000) propose two types of small building, effective communication, and opera- business advisory that are used in the Small tional efficiency. At a broad level, the common Business Development Center (SBDC) in U.S. goal of these programs is to promote entrepre- consulting and counseling advisory services. neurial and innovative capabilities. The authors suggest the use of counseling Government subsidies of between 25 service to facilitate and coach the performance percent and 100 percent to attend these pro- task by participating firms is more effective grams were available depending on various than consulting service that is to provide advice conditions. There were no specific eligibility and perform a task. Likewise, Mole et al. (2011) criteria to attend these programs. Firms compare and contrast between breadth and described their reasons for participating in depth of the Business Link (BL) in the United terms of being uncertain about the nature of Kingdom. Whereas breadth support refers to their firm’s problems, seeking solutions to “light-touch” strategy, which has little in-depth problems, or seeking to improve their perfor- interaction and follow-up sessions, depth mance. Firms heard about QMI programs from support focuses on intensive interaction and past participants, industry networks, or assistance. The authors suggest that participat- through attendance at QMI events. Prior to the ing firms received positive outcomes from program attendance, firms went through a depth of service over breadth of service. In needs analysis process using different tools Germany, Kosters and Obschonka (2011) ranging from basic discussion with senior man- describe the national EXIST1 program as opera- agement to a sophisticated questionnaire cov- tional and strategic advisory services. Opera- ering organization and culture, manufacturing tional services include business advice in terms cycle times, quality, plant and equipment, inno- of business setup procedure, writing a business vation, engineering operations and practice, plan, and finance, whereas strategic services product development process, and business include market and management intensive management. Then, tailored workshops analyses. The authors conclude that operational (ranging from one to five days) were delivered services are unlikely to impact on long-run on-site, addressing emerging issues with a business performance, when compared with follow-up session (up to 12 months after the strategic services. initial workshop). Our study examines a small business advi- The provision of business advisory services sory service that is provided by QMI Solutions, has become a major aspect of business an independent not-for-profit organization improvement (Bennett et al. 2000). Although partly funded by Australian government. The some studies found no positive, or only partial, programs that are offered through QMI are impact (Kosters and Obschonka 2011), there is intended to be a counseling approach and inte- some evidence of a positive effect from busi- grated with multiple follow-up sessions as a ness advisory services on SMEs’ growth and part of program coaching. Four major advisory performance (e.g., Bennett 2007; Bryson, programs through QMI were identified: “Ideas Keeble, and Wood 1997; Harrington, to Market,” “Microscope Action Plan” (MAP), MacPherson, and Lombard 1991; Mole and “Technology Access Programme” (TAP), and Keogh 2009; Mole et al. 2009; Ramsden and “Lean Manufacturing.” Ideas to Market program Bennett 2005). Bennett (2007) assessed the focuses on training SMEs to access new ideas, economic effect of government advisory ser- knowledge, and technologies, and to enable vices in the United Kingdom and found that businesses to become more innovative, effi- recipient SMEs demonstrated improved profit- cient, and competitive. MAP and TAP aim to ability and return on investment. Both

1EXIST is a support program of the Federal Ministry of Economics and Energy, Germany aimed at improving the entrepreneurial environment at universities and research institutions and at increasing the number of technology and knowledge based business start-ups.

246 JOURNALSAWANG, OF SMALL PARKER, BUSINESS AND MANAGEMENT HINE 3 Ramsden and Bennett (2005) and Bennett tional innovativeness from a delivery approach (2007) conclude there is a positive link perspective. We define organizational learning between business advisory services and SMEs’ as abilities and new knowledge that firms growth, measured by business turnover and acquire from participating in business advisory profitability. Another study by Mole et al. programs. Organizational innovativeness is (2008) also confirmed the positive effect of defined as changes in products, services, or business advisory services on sales and processes of the firm (Damanpour 1991). employee growth. How were the small business advisory pro- Although there is evidence of a positive link grams through QMI delivered? Through nine between business advisory services and firm interviews with program trainers and managers performance, there is increasing interest in how who participated in QMI programs (as different types of advisory programs differen- described previously), we identified that there tially impact on firm outcomes. Past research were a variety of delivery methods used by suggested that the type of interaction between different trainers within and across each of recipient firms and business advisors influ- these programs. We identified three major busi- enced the level of impact of the business advi- ness advisory program delivery characteristics: sory service (e.g., Bennett 2007; Mole and collective learning, tailoring of content, and Keogh 2009; Ramsden and Bennett 2005). The practice-based approaches. The details of each interaction was measured by the number of approach are discussed in the following contacts between advisors and the firm and the sections. program duration. Similar to previous studies, Mole et al. (2008) compared the economic Organizational Learning effect of intensive and non-intensive advisory A key objective of many small business advi- programs. The intensiveness of the program sory programs is for firms to learn new skills was classified by the degree of interaction— and capabilities. We draw on the literature on that is, non-intensive programs were “one-off” adult and organizational learning to develop an events whereas intensive programs involved understanding of how small business advisory the provision of a service over an extended methods are likely to impact on the learning of period of time. The finding demonstrated that skills and capabilities. Knowledge is generated advisory programs that incorporated an inten- through social processes when people are sive delivery approach had a better influence willing to work together and share their stories. on economic effect. Learning occurs when organizational members A recent study by Mole et al. (2011) looked observe and learn from other organizations beyond the degree of interaction and proposed (Levinson and Asahi 1995). Collective learning that the intervention strategy of the business thus enhances organizational learning of criti- advisory service also influenced business cal skills or abilities. Further, when a business growth. The author classified the strategy based advisory program is tailored to the organiza- on the level of financial spending per intervention’s needs, it engages the learner in the learn- tion and the proportion of assisted firms (Mole ing process and is therefore more likely to et al. 2011). The findings suggested that the enhance learning outcomes (Brown and more focused, high investment assistance Duguid 1998). Drawing on the learning litera- yielded a better impact on business growth. ture, the following discussion explains the way The weakest strategy was found to be when the in which collective learning experiences and business advisory service aimed to provide a tailoring in small business advisory programs relatively inexpensive service to a relatively might impact on learning. high proportion of firms. Although we have learned from previous studies that both the Collective Learning Approach. A potentially extent of interaction with advisors and the important mechanism through which small nature of the intervention strategy affect busi- business advisory programs facilitate learning ness growth, little is known about other varia- is through the creation of social exchange. The tions in delivery methods and their potential process of social exchange is the mechanism impact on firm outcomes. through which knowledge and meaning are Our study expands previous research by created and, as such, small business programs examining the effect of business advisory ser- potentially enable learning by creating relation- vices on organizational learning and organiza- ships and interactions that allow for collective

4 JOURNALSAWANG, OF SMALL PARKER, BUSINESS AND MANAGEMENT HINE 247 learning (Wenger 1998; Wenger and Snyder 2004). Thus, small business advisory programs 2000). Collective learning involves the develop- that are delivered by collective learning method ment of common visions, collective goals, would improve firms’ learning critical skills and social interaction, and a sharing of experiences capabilities. and group decision-making (Hoban 1999). Hoban explains that the key attribute of learn- Tailoring of Content Approach. Though a ing communities is the “social emphasis on number of small business training programs learning” (1999, p. 172). Learning communities use the traditional trainer-centered model, in involve action learning in which colleagues which knowledge is transmitted from the address relevant problems, reflect on their trainer to the learner(s), it is gradually being practice, share ideas, and provide feedback replaced by an alternative model that is (Hoban 1999, p. 175). Organizations create learner centered, in that it is focused on learn- knowledge through socialization (Nonaka and ers’ needs, rather than pure knowledge trans- Takeuchi 1995). Social interaction involves mission (Brookfield 1991; Duffy and individuals grouping together and sharing what Cunningham 1996). This distinction can also they experienced or what they know. By be understood in terms of whether the busi- sharing experiences and hearing stories, indi- ness advice is “operational” or “strategic” in viduals within a firm will collectively learn focus (Hjalmarsson and Johansson 2003). With common phenomena and unwritten rules about operational services, the objective of the how to respond to business problems (Järvinen program is determined prior to the beginning and Poikela 2001). of the service. This type of business advice is Further, social interactions play an essential static and the relationship between the role in fostering the ability of firms to think advisor and the business is that of expert- creatively and to minimize egocentric percep- client. In contrast, some business advisors tions of a business problem (Meyers 1991; deliver strategic services to SMEs, which are Roglio and Light 2009). Importantly, the collec- more dynamic and in which the issues tive learning experience enables learners to addressed in the service are generated by interact with other learners as well as advisors/ interplay between the advisor and the busi- facilitators to confront questions raised by ness over time. This latter type of service is others, to consider perspectives that had not “learner centered” and does not depend exclu- previously been considered, to share difficul- sively on the transmission of expert advice. ties and problems, and to become aware of Business advisory programs that employ problems that were previously unrecognized this tailoring of content method use facilita- (Meyers 1991; Roglio and Light 2009). tors (rather than trainers) as resource manag- The concept of collective learning is some- ers, encouraging firms to bring their what akin to entrepreneurial learning litera- experiences and beliefs into the learning tures. To start a new venture, people seek two process. Individuals learn best when they are types of knowledge: direct knowledge of a engrossed in the topic and they are motivated specific market and knowledge regarding how to seek out new knowledge and skills because to be entrepreneurial (Hillary 1995). They they need them in order to solve problems at acquire this knowledge through direct experi- hand (Norman and Spohrer 1996). To achieve ence, observing others, and codified sources the best learning outcome, business advisory such as books or articles (Hahn and Stavins services should be tailored to organizational 1992). Through collective learning, participat- needs. Business advisory services should ing firms are given an opportunity to learn by deliver programs that align with identified observing other firms sharing their direct expe- organizational problems (Lundström and rience of their success or failure, including Stevenson 2005). codified sources that they used for previous problem solving. The entrepreneurial learning H1: Business advisory programs that are deliv- literature has shown that through collective ered by focusing on a collective learning learning experience, firms acquire entrepre- approach will enhance organizational neurial skills of how to recognize an opportu- learning of critical skills or capabilities. nity and overcome traditional obstacles when organizing or managing newness such as a H2: Business advisory programs that are deliv- product or service (Pimenova and van der Vorst ered by focusing on a tailored approach will

248 JOURNALSAWANG, OF SMALL PARKER, BUSINESS AND MANAGEMENT HINE 5 enhance organizational learning of critical gies. Thus, participants learn to apply knowl- skills or capabilities. edge in practice and to innovate by creating new products and services, or to add value to Organizational Innovativeness. Innovation existing ones (Seagal and Horne 1997). involves firms making active changes in prod- To increase innovativeness, firms should be ucts, services, or processes (Damanpour 1991). able to better understand what they know and Though we would expect organizational learn- how to implement that knowledge in practice ing outcomes to depend on a collective learn- (Schön 1992). Through practice-based ing experience and the tailoring of program approaches, firms learn to critically evaluate content as described earlier, we argue that their business environment, make sense of it, organizational innovativeness involves changes and figure out implications for action (Teece in behavior and therefore depends on partici- 2007). Business advisory programs that adopt a pation in practice-based advisory programs, practice-based approach will focus on authentic which involve the adoption of immediate situations rather than theoretical scenarios. As a changes within the firm. That is, a change in result, program participants develop skills that behavior is best achieved through an advisory can detect latent demand, the structural evolu- program in which the learner is active, rather tion of industries and markets, and likely sup- than passive, and in which the learner is plier and competitor responses (Teece 2007), engaged in doing things in their own business. which then lead to firms’ improvement of their The following section explains how practice- innovativeness in various aspects, such as based learning approaches influence organiza- product, core business, and organizational pro- tional innovation. cesses (Damanpour 1991). Through a practice- based approach, individual firms develop skills Practice-Based Approach. The practice-based to coordinate and adapt in relation to changing approach focuses on integrating thought and environments, which results in changes in their action through reflection (Schön 1983). Facili- business practices (Augier and Teece 2008). tators encourage program participants to become researchers in the context of practice, H3: Business advisory programs that are to be free from established theory and tech- delivered by focusing on practice- niques, and to be able to construct a new based approaches will enhance firms’ theory to fit the firm’s situation. Business advi- innovativeness. sory programs that adopt a practice-based approach stimulate participants to evaluate Method their problem and context and determine if the In this section, we discuss the sources of our proposed solutions can actually be applied in data and our data collection procedure. We also their context (Marsick 1988). Participants can provide measures that were used in our survey learn theoretical knowledge (e.g., learning as well as validation information. about newly available technology), but without practical knowledge (e.g., how to apply such a Participants and Procedure technology in their business operation), they The population of interest included all small cannot effectively innovate or adopt new tech- to medium-sized firms that participated in one nology (Brown and Duguid 2001). Knowing or more business advisory programs over the that technology exists does not lead to effective past three years. We contacted senior manage- implementation or innovation. Brown and ment or business owners of every firm Duguid (2001) therefore argue that to achieve (N = 562) that participated in one of these pro- innovation involving the useful application of grams for the telephone survey, which knowledge, a practice-based learning approach employed the use of a computer-assisted tele- is required. phone interviewing system. This method allows Knowledge is a productive resource for interviewers to clarify answers from respon- innovation (Johannessen, Olsen, and Olaisen dents during the data collection and to obtain 1999). Business advisory programs that adopt a more substantive answers (Greenfield, practice-based approach require program par- Midanik, and Rogers 2000). A total of 257 firms ticipants to learn by using, doing, and experi- agreed to participate in the study (a response menting, through informal brainstorming rate of 45.73 percent). Majority of respondent meetings, using metaphors, stories, and analo- firms were in the manufacturing sector (60

6 JOURNALSAWANG, OF SMALL PARKER, BUSINESS AND MANAGEMENT HINE 249 percent) and their main activity was product changed behavior in the form of firms’ subse- development (81 percent) compared with quent organizational innovative behavior. For service development (19 percent). Our firms’ this phase of the analysis, we were able to main market is mostly situated locally and reach step 5 of Storey’s (2000) “six steps to nationally (87 percent), with 13 percent of firms heaven” by using matched samples. We did reported international market. The majority of not use a random panel approach (sixth step) firms attended at least one business advisory because it would have involved an ethical program and 26 firms reported that they had challenge to undertake this step in research participated in 10 or more business advisory on publicly funded SME programs, as it programs. An average duration of advisory would involve randomly excluding firms from program including follow-up session was four support, at least in the short term (Storey months. 2000). Storey (2000, 2003) commented that many We matched our data (assisted firms) with evaluations of business advisory service pro- unassisted firms, drawing on the Australian grams only go as far as measuring satisfaction Bureau of Statistics (ABS) survey, which was with the program. Our study thus attempted to collected on behalf of the Australian federal employ the most advanced stages of “six steps government during 2009–2010. This survey to heaven,” suggested by Storey (2000), in comprised self-administered, structured ques- assessing business advisory services’ impact. tionnaires containing essentially closed ques- The six steps are: tions. The ABS collected information on business activities and business characteristics (1) Measuring the number of participating that were relevant to organizational perfor- firms as an indicator of the program’s mance. All SMEs in the Australian economy impact. were included except for business units in (2) Analyzing assisted firms’ satisfaction with nonemploying businesses and government the program. enterprises. In the ABS sample, there were (3) Measuring assisted firms’ views regarding 1,690 firms that did not receive business advice the difference made by the assistance on in the form of government assistance. their business performance. Following Storey’s (2000) recommendation, (4) Undertaking a comparison of the perfor- we matched the firms in relation to key demo- mance of assisted firms with “typical” graphic characteristics such as industry sector, firms from the general population. location of main market, and firm size. By (5) Undertaking a comparison of assisted matching firms in this way, we were able to firms with a sample of firms from the “control” for potentially extraneous influences general population of firms that have been on firm innovation in the two samples. We “matched” with the assisted population on matched samples to supplement an examina- key demographic characteristics likely to tion of H3. As such, the level of firm innovation impact on business performance. in our sample could more confidently be (6) Using a random panel approach in which explained with reference to the fact that firms firms are randomly excluded from partici- had participated in a small business advisory pation in the program and their perfor- program. This similar approach has been mance compared with assisted firms over recently used in business advice literature time. (Mole and Keogh 2009; Mole et al. 2008, 2011).

The first phase of our analysis (H1 and H2) Measures was concerned with firms’ perceptions of the Programme Delivery Approach. We devel- skills and abilities they acquired from partici- oped the new measures for program delivery pation in the program—this reaches the third based on our nine interviews with program step of Storey’s (2000) “six steps to heaven” managers and trainers. The thematic analysis of involved in assessing SME policy impact. As the interviews revealed three major delivery this stage of the research was focused on approaches that were used across four advisory learning from the program, we necessarily had programs. These themes were collective learn- to rely on firms’ perceptions of the impact of ing, tailored, and practice-based approaches. the program. However, for the second phase During our pilot interviews with small business of our analysis (H3), we were interested in advisors who delivered these programs and

250 JOURNALSAWANG, OF SMALL PARKER, BUSINESS AND MANAGEMENT HINE 7 firms, we found that there was significant varia- knowledge to the firm. The Cronbach’s alpha for tion within the programs regarding the delivery collective learning was 0.78, indicating good method used such that each program was deliv- reliability. ered in a variety of different ways by a range of different trainers. The measures of each deliv- Practice-Based Approach. This construct mea- ery method were then developed based on the sures the degree of business advisory programs’ pilot interview and thematic analysis. use of practical methods such as reflection on Our interviews with program managers and business practice and the adoption of changes trainers revealed that the four advisory pro- within the business. The respondents were grams were delivered using a variety of deliv- asked “To what extent do you agree that this ery methods as different trainers would adopt program used ____ as a part of learning different delivery styles for the same program. process?” using a five-point Likert scale, 1 = not All four programs involved a range of different at all and 5 = a great deal. There were four items combinations of collective learning, tailored, describing core program activities that were and practice-based approaches, and program related to practice-based approach: (1) imple- facilitators varied in the approach they used. mentation of changes, (2) firm analysis, (3) best This means facilitators could deliver the practices, and (4) business practices reflection. content from the same program using more of The Cronbach’s alpha for collective learning was collective training for some clients and more of 0.82, indicating good reliability. practice-based approaches for other clients. There was no clear approach such that one Organizational Innovativeness. We adopted program would employ one particular method. questions from the Business Longitudinal Thus, the same advisory program varied in the Survey (Australian Bureau of Statistics 2009) delivery method that facilitators employed. measuring firms’ organizational innovativeness Therefore, asking the firm to choose one of in terms of changes in products, processes, and these delivery methods was not appropriate. core business. The respondents were asked, We thus asked firms to rate the intensity of during the last financial year, if the firm has each delivery method used in the program in made any changes in (1) organizational/ which they participated using a five-point managerial processes, (2) operational pro- Likert scale, 1 = not at all and 5 = a great deal. cesses, (3) management structure, (4) business A list of questions is reported in the appendix. structure, (5) core business, and (6) product or service. The respondents rated the level of Collective Learning Approach. This construct change using a five-point Likert scale, 1 = not at measures the degree that business advisory all and 5 = a great deal. The Cronbach’s alpha programs encourage firms to learn from each for organizational innovation was 0.81, indicat- other. The respondents were asked to rate the ing good reliability. nature of the program using a five-point Likert scale, 1 = not at all and 5 = a great deal. There Organizational Learning. This was a single were four items describing the collective learn- item and self-reported measuring a level of ing approach: (1) sharing business know-how, learning from the participated program. The (2) sharing similar needs among program par- respondents were asked to rate the following ticipants, (3) level of learning from facilitator, question using a five-point Likert scale, and (4) level of learning from other partici- 1 = strongly disagree and 5 = strongly agree: pants. The Cronbach’s alpha for collective “To what extent do you agree or disagree that learning was 0.71, indicating good reliability. your enterprise has been successful in learning critical skills or capabilities from participation Tailoring of Content Approach. This construct in this program?” measures to what extent the business advisory program was tailored to each firm’s context and Controls. Previous research shows that firm needs. The respondents were asked to rate the characteristics such as size, revenue, and indus- program using a five-point Likert scale, 1 = not try sector may influence program outcomes at all and 5 = a great deal. There were three (Kafouros et al. 2008). The number of partici- items describing the degree of tailoring: (1) pants from each firm may impact on the level of addressing the firm specific needs, (2) understat- collective learning, which is important for inno- ing specific firm context; and (3) applicable vation performance. Therefore, we included

8 JOURNALSAWANG, OF SMALL PARKER, BUSINESS AND MANAGEMENT HINE 251 these variables as controls in our analysis. program reported that they introduced changes We also controlled for firms’ motivation to to their products/services, operational systems, seek advice. Firms that seek business advice managerial processes, or market penetration are often more dynamic, growth-oriented, (see Table 1). Among the matched sample of and have more highly motivated owners/ 226 firms that did not participate in an advisory managers than firms that do not seek bus- program, only between 20 percent and 24 iness advice (Deschoolmeester, Schamp, and percent reported innovations in those same Vandenbroucke 1997; Mole et al. 2011). There- areas. fore, firms that participate in business advisory programs may be highly innovation-oriented The Impact of Business Advisory and that would result in high levels of innova- Programme Delivery Approaches: tive behavior even without attending a business Hypothesis Testing advisory program. To minimize this potential Table 2 provides the bivariate correlation bias, we controlled for motivation to seek coefficient for all variables in the current study. advice. Participants were asked to report their Organizational innovation was positively corre- motivation to attend the business advisory lated with collective learning (r = 0.27, p < .01), program. We categorized two types of motiva- tailored (r = 0.27, p < .01), and practice-based tions: firms that were driven by internal moti- approaches (r = 0.43, p < .01). Organizational vation (e.g., increasing competitiveness or learning was also positively associated with gaining more knowledge) and external motiva- collective learning (r = 0.37, p < .01), tailored tion (e.g., cost subsidy availability). (r = 0.46, p < .01), and practice-based approaches (r = 0.28, p < .01). None of the control Construct Validity variables significantly correlated with organiza- Construct validity was tested using factor tional innovation or learning outcomes, except analysis. All independent construct items industry, which was marginally significantly loaded uni-dimensionally on one factor and associated with learning outcomes (r = 0.13, had no significant cross-loadings with other p < .05). factors. Cross-loadings were all well below the To examine the role of program delivery cutoff of 0.40 suggested by Raubenheimer approaches on organizational learning and (2007) and our factor loadings were all above innovation outcomes specified by the research 0.50, which is considered a good loading (Hair hypotheses, two identical hierarchical multiple et al. 1998). Three factors (namely collective regression analyses were constructed. To learning, program modification, and program control possible confounding variables, firm practicality) were found to have Eigen values characteristics (industry, size, tenure, and over 1.0, explaining 64.45 percent of the total revenue), number of participants from each variance. firm, duration of program, and motivation to seek advice were entered into each equation at Results step one. To test the main effect of program In this section, we compare the level of delivery approaches, collective learning, tai- organizational innovativeness between assisted lored, and practice-based approaches were and unassisted firms. Given that our samples entered at step two. are matched, we have improved confidence The initial two steps in the regression equa- that differences in innovation performance tions explained 23 percent of the variance in between the two groups can be attributed to organizational innovativeness (F[11,206] = 5.29, the fact that one group has completed a busi- p < .001) and 29 percent of the variance in ness advisory program. Second, we examine organizational learning (F[11, 206] = 7.12, the impact of different program delivery p < .001). With the exception of motivation to approaches on learning and organizational seek advice, the control variables did not sig- innovativeness. nificantly influence our results as none of them obtained significant regression coefficients. The A Comparison of Organizational results show that internal (β = 0.17, t = 2.37, Innovativeness between Assisted and p < .05) and external motivations (β = 0.18, Unassisted Firms t = 2.47, p < .05) influenced innovativeness. We We found that between 45 percent and 69 also found that only internal motivation percent of firms that participated in an advisory (β = 0.20, t = 2.98, p < .05) significantly influ-

252 JOURNALSAWANG, OF SMALL PARKER, BUSINESS AND MANAGEMENT HINE 9 enced organizational learning, but not external motivation (β = 0.12, t = 1.77, not significant). By including motivation to seek advice as a control in our analysis, it demonstrated that the following results accounted for these confounding factors. Our results in Table 3 demonstrate that organizational innovativeness was significantly positively predicted by the practice-based approach (β = 0.41, t = 5.13, p < .001), thus H3 was supported. Organizational learning of critical skills or capabilities was significantly positively predicted by the collective learning approach (β = 0.13, t = 2.07, p < .05) and tailored approach (β = 0.41, t = 4.90, p < .001), thus H1 and H2 were supported. Because of significant beta coefficients between motivations and our dependent variables, we performed post hoc analysis of the possible interaction effect. We found no significant interaction between motivations and delivery methods. Although it was not hypothesized, we also examined a possible moderation effect among collective learning, tailored, and practice-based approaches; we did not find any significant results. Discussion This paper investigated the relationship Table 1 between delivery approaches used in business advisory programs and their impact on the learning of skills and abilities and organizational innovativeness. Our contribution is twofold. First, we demonstrate that firms that participate in small business advisory services report higher levels of subsequent innovation behavior than matched firms that do not participate in such services. Our study therefore proposes that business advisory services can be considered an antecedent of firms’ innovation behavior. Second, we provide insights into how

Product/Service Operational System Managerial Processthe Marketing delivery style of advisory programs affects organizational learning and subsequent innovation behavior. Though many studies have focused on the economic effect of business advisory services, the question of “What are effective ways to is used to test difference between assisted group and the unassisted group responses at the 5 percent conﬁdence level.

2 deliver the services?” has been largely ignored χ in previous evaluations (Mole et al. 2011). Because of the nature of business advisory services, which are described as intangible services with dependent relationships between advisors and participating ﬁrms, past research The Characteristics of Unassisted and Assisted Firms in General Population and Selected Cases

That Introduced . . . focused on interaction intensity as a key factor Assisted firms were found toin be relation significantly different to from key non-assisted demographic firms in characteristics relation such to their as innovation industry behavior. The sector, firms location were of matched main market, and firm size. Organizational Innovativeness Non-parametric Number of Firms Unassisted SampleAssisted Sample 45 (19.91 percent) 125 (55.30 percent)driving 55 (24.33 percent) impact 155 (68.58 percent) (Bennett 48 (21.23 151 percent) (66.81 percent) and Robson 45 102 (19.91 (45.13 percent) percent) 1999a,

10 JOURNALSAWANG, OF SMALL PARKER, BUSINESS AND MANAGEMENT HINE 253 254

Table 2 Descriptive and Correlation Coefﬁcients for Focal Variablesa

Variables 1 2 3 4 5 6 7 8 9 10 11 12 13 Mean S.D.

1. Collective Learning (0.71) 0.56** 0.48** 0.27** 0.37** 0.00 −0.01 0.00 0.03 0.08 0.02 0.28** 0.10 3.38 0.87 2. Tailoring for Content (0.78) 0.49** 0.26** 0.46** 0.09 0.10 −0.01 0.01 0.13* 0.18** 0.37** 0.09 3.70 0.91 ORA FSALBSNS MANAGEMENT BUSINESS SMALL OF JOURNAL 3. Practice-Based (0.82) 0.43** 0.28** 0.22** 0.01 −0.09 −0.13* 0.12* 0.05 0.37** 0.19** 3.43 0.95 4. Organizational (0.81) 0.13* 0.03 0.00 −0.01 −0.03 0.07 0.03 0.18** 0.18** 2.73 0.77 Innovativeness AAG AKR N IE11 HINE AND PARKER, SAWANG, 5. Organizational 0.13* −0.06 −0.04 −0.13 0.09 0.12* 0.31** 0.23** 3.91 0.99 Learning of Critical Skills or Capabilities 6. Industry 0.09 −0.17** −0.13 0.17** 0.12* 0.12 0.13* 0.60 0.49 7. Organization Tenure 0.46** 0.38** 0.12 −0.02 0.02 −0.11 23.82 24.74 8. Firm Size (Employee 0.84** 0.03 −0.11 −0.03 −0.17** See note Number) 9. Gross Revenue 0.16* −0.09 −0.12 −0.22** 3.65 120.2 (Million) 10. No. of Participants 0.15* 0.10 0.07 8.44 22.06 Each Firm 11. Program Duration 0.03 0.04 3.93 3.45 12. Internal Motivation 0.16* 4.12 0.89 13. External Motivation 2.73 1.10

aM, mean; S.D., standard deviation. *p < .05. **p < .01. Industry sectors are dummy coded (0 = non-manufacturing and 1 = manufacturing). Cronbach alphas (internal reliabilities) are in the diagonals. We have 245 firms (95.3 percent) that reported the number of employees as 500 or less. Our mean of employee was large (M = 179.19; S.D. = 762.62) because of three firms that reported the number of employees as 6,000. Our median of total sample was 20. We reran the descriptive statistics excluding these firms and our mean was 48.10 (S.D. = 82.86). 12

Table 3 Regression Analysis between Delivered Methods on Business Advisory Programs and a ORA FSALBSNS MANAGEMENT BUSINESS SMALL OF JOURNAL Firms’ Outcomes

Control Variables Organizational Innovativeness Organizational Learning of Critical Skills

AAG AKR N IE255 HINE AND PARKER, SAWANG, or Capabilities

Step 1 Step 2 Step 1 Step 2

Industry Sector β = −0.00, t = −0.05, (ns) β = −0.07, t = −1.06, (ns) β = 0.06, t = 0.94, (ns) β = 0.07, t = 1.06, (ns) Firm Tenure β = −0.02, t = −0.31, (ns) β = −0.02, t = −0.38, (ns) β = −0.08, t = −1.17, (ns) β = −0.10, t = −1.57, (ns) Firm Size β = −0.09, t = −0.67, (ns) β = −0.17, t = −1.38, (ns) β = 0.06, t = 0.49, (ns) β = 0.15, t = 1.24, (ns) Revenue β = 0.12, t = 0.91, (ns) β = 0.22, t = 1.71, (ns) β = −0.08, t = −0.64, (ns) β = −0.20, t = −1.62, (ns) No. of Participants Each Firm β = −0.01, t = −0.15, (ns) β = −0.03, t = −0.50, (ns) β = 0.04, t = 0.68, (ns) β = 0.05, t = 0.69, (ns) Program Duration β = 0.04, t = 0.66, (ns) β = 0.04, t = 0.64, (ns) β = 0.11, t = 1.69, (ns) β = 0.03, t = 0.50, (ns) Internal Motivation β = 0.16, t = 2.36, (p < .05) β = 0.01, t = 0.22, (ns) β = 0.20, t = 2.98, (p < .05) β = 0.01, t = 0.20, (ns) External Motivation β = 0.17, t = 2.47, (p < .05) β = 0.14, t = 2.14, (p < .05) β = 0.12, t = 1.77, (ns) β = 0.13, t = 2.06, (p < .05) Main Effects Collective Learning β = 0.08, t = 1.08, (ns) β = 0.16, t = 2.10, (p < .05) Tailoring for Content β = −0.01, t = −0.20, (ns) β = 0.41, t = 4.89, (p < .05) Practice-Based β = 0.41, t = 5.12, (p < .001) β = −0.10, t = −1.37, (ns)

ans, non-significant. 1999b). Interaction intensity is measured by the According to Storey (2000), most public poli- number of contacts, number of participants, cies in Organisation for Economic Cooperation and program duration. In our study, although and Development (OECD) countries are moni- the control variables such as the number of tored (step 1 to step 3) rather than evaluated contacts, number of participants, and program (step 4 to step 6) and in many cases do not pass duration, and firm characteristics were signifi- beyond step 1. Our study achieved step 3 to cant predictors in our regression, the correla- examine H1 and H2 and step 5 by matched tion coefficients were not equal to zero. This sample to supplement an examination of H3. information suggested that these control vari- However, we used firms’ subjective measures ables may minimally contribute to our depen- and did not evaluate the impact of these small dent variables, but the contribution is not business advisory programs based on objective strong enough to display a statistical significant measures (such as turnover, profitability, or prediction in our regression. liquidity). Thus, our findings should be inter- The recent study by Mole et al. (2011) sheds preted with respect to the limitations associated some light on this area. Mole et al. (2011) pro- with the use of qualitative measures. Although posed an alternative way to look at how busi- many past studies were focused on quantitative ness advisory services are delivered, proposing growth, Lambrecht and Pirnay’s (2005) research that advisory services are delivered in two showed that SMEs have qualitative targets such ways: broader or deeper, differentiated by the as continuity and stability. As a result, they level of financial resources spent per interven- argued that research on small business advisory tion and the proportion of assisted firms. Our programs should give more attention to qualita- study expands the broader versus deeper tive objectives. Further, in some fields of approach by “looking under the hood” to iden- inquiry, such as entrepreneurial learning, soft tify how the business advisory services are measures are more appropriate. It would be delivered, based on program managers’ and interesting for future research to capture addi- facilitators’ views. Instead of focusing on cost tional measures, such as the level of material per intervention or service, we argued that gain resulting from learning and innovation. other delivery elements are antecedents of Collective learning experience could also facili- firms’ learning and innovation. tate firms’ material gains through information We found that business advisory services that exchange to introduce a new product or were delivered by using collective learning and process, although the impact of learning on tailored approaches enhanced organizational material outcomes may take some time to be learning of critical skills or capabilities. This realized. finding is akin to previous literature, suggesting Our research clearly demonstrates that dif- that when firms group together, they learn col- ferent delivery methods of small business advi- lectively to find solutions for their problems by sory programs impact on different outcomes. It hearing stories and sharing their experiences would be interesting for future research to also (Järvinen and Poikela 2001). Further, each firm’s examine an impact of delivery methods as well context is different, thus it is important for as interaction among firms and advisors on program facilitators to ensure that the program small business advisory effectiveness. is tailored to suit organizational contexts and There are some limitations to our study that facilitate knowledge sharing for new knowledge should be acknowledged. Our research creation (Zahra and George 2002). We also involved a retrospective design in which found that the small business advisory services respondents were asked to recall characteristics that were delivered by focusing on practice- of small business advisory programs that they based approaches enhanced organizational attended in the previous three years. This innovativeness. To stimulate changes within approach may create response bias or a recall organizations, learning experiences should limitation (Miller, Cardinal, and Glick 1997). involve firms making changes in their business. Our study minimized this concern by using a When business advisory programs encouraged telephone survey, which enabled us to clarify participating firms to critically evaluate their questions for respondents and allowed us to business and adopt new approaches to solving exclude any respondents who could not business problems, they were more likely to describe the program(s) that they attended make innovative changes responding to their (Forgues and Vandangeon-Derumez 2001). dynamic environment (Augier and Teece 2008). Further, as our study involves a cross-sectional

256 JOURNALSAWANG, OF SMALL PARKER, BUSINESS AND MANAGEMENT HINE 13 design, future research may examine the longi- nosis process was less than the impact and tudinal impact of various program delivery satisfaction of more-intensive assistance (Mole methods on small businesses’ learning and and Keogh 2009). This could be explained by innovation, as we might expect the impact of the fact that diagnosis has not directly impacted programs to decline with time. Additionally, on changing ideas or action. BL has moved future research can extend our study by exam- toward offering more-intensive assistance and ining individual level phenomena and human offering intensive assistance to more companies. capital measures capturing the knowledge, A lesson learned from BL was that offering skills, and resources of the workforce and their intensive assistance to a high proportion of firms influence on organizational learning capability. was the least effective strategy (Mole et al. 2011). Conclusion and Implications The U.S.-based SBDC program provides management assistance to entrepreneurs and As there are limited resources available in small business owners. SBDC employed a the provision of small business advisory ser- counseling approach where all of the changes vices, such services should be delivered stra- are performed by participating firms with guid- tegically in order to maximize their impact. ance from a trained advisor. SBDC has a Our study has highlighted that the way in required condition for firms to receive services. which business advisory services are delivered All pre-ventures (those who have not started does matter. To enhance organizational learn- their business prior to their initial contact with ing, business advisory services should be the SBDC) are required to attend a full-day custom designed to suit each participant firm workshop to screen out firms that are not and the program should be facilitated in a serious or whose prospects are not attractive way that program participants can share and (Chrisman, McMullan, and Hall 2005). SBDC’s learn from each other, as supported by H1: approach is on the job learning, which is a collective learning approach significantly similar to practice based. This approach enhanced organizational learning of critical requires firms’ commitment to perform tasks skills or capabilities. To encourage firms to with SBDC advisor’s coaching. improve their innovation behavior, business The Swedish ALMI Företagspartner AB is a advisory services should deliver practical government-supported program to encourage experiences to program participants, which innovation among SMEs. ALMI programs target require them to make changes in their busi- young firms two to eight years. ALMI increases ness, as supported by H3: practice-based a number of participating firms by tailoring approaches significantly enhanced firms’ services and it has been well regarded by innovativeness. Swedish small firms (Storey and Greene 2010). Governments are often faced with a dilemma A majority of ALMI programs offer generic and in designing publicly funded business advisory standardized advices, which can reach a higher services in terms of allocating spending priori- number of SMEs. These advices include writing ties between generic or tailored service provi- business plans, venture setup procedures, and sion (Mole et al. 2011). The English BL services financing new business. According to ALMI are based on national resource allocation and program evaluation, we learn that offering stra- human resource availability. BL provides a one- tegic advice (such as market ad management stop shop in order to integrate locally frag- analysis) demonstrated better improvement mented government supports (Bennett, Robson, than operational advice (generic and standard- and Bratton 2001). In general, the BL program ized) (Kosters and Obschonka 2011). Similar to has been widely opened to all SMEs and has thus SBDC and BL, ALMI is also designed to be a widened support with less-intensive advice. one-stop shop to improve the visibility of busi- However, intensive advice will remain part of ness advice services. ALMI, akin to QMI, deliv- the portfolio of BL depending on a case by case ers its programs entirely through the public basis. BL services employ a brokerage model sectors whereas BL and SBDC deliver through where the advisor helps the process of manag- both public and private agencies. We learn, ing consultants (Mole and Keogh 2009). The BL from delivery through both public and private program often offers business diagnosis by bodies, that BL and SBDC have more flexibility highly experienced advisors. However, the to select advisors who have existing expertise impact and customer satisfaction with the diag- (Mole and Bramley 2006).

14 JOURNALSAWANG, OF SMALL PARKER, BUSINESS AND MANAGEMENT HINE 257 From international perspectives as well as (4) Business structure our research findings, it can be suggested that (5) Changes in core business government investment in tailored services (6) Changes in the range of products or ser- involving collective learning experiences is vices offered more likely to achieve learning outcome, and involving the introduction of changes in the Organizational Learning of Critical Skills business is more likely to achieve innovat- or Capabilities ion outcome. Therefore, these delivery To what extent do you agree or disagree that methods are a more effective mechanism for your enterprise has been successful in learning achieving the aims of small business advisory some critical skills or capabilities from partici- programs. pation in this program? Appendix References To what extent do you agree with the fol- Augier, M., and D. J. Teece (2008). “Strategy as lowing statements regarding this program? Evolution with Design: The Foundations of Dynamic Capabilities and the Role of Man- Collective Learning Approach agers in the Economic System,” Organiza- tion Studies 29(8–9), 1187–1208. (1) In this program, participants had similar Australian Bureau of Statistics (2009). Techni- needs to my enterprise. cal Manual: Business Longitudinal Data- (2) In this program, participants shared busi- base (cat.no.8168.0.55.002). Canberra: ness know-how with other participants in ABS. the program. Bennett, R. (2008). “SME Policy Support in (3) In this program, I learned a lot from the Britain Since the 1990s: What Have We program facilitator. Learnt?,” Environment and Planning C: Gov- (4) In this program, I learned a lot from other ernment & Policy 26(2), 375–397. program participants. Bennett, R., and P. Robson (2004). “Support Tailoring for Content Approach Services for SMEs: Does the Franchisee’ Make a Difference to the Business Link (1) This program addressed the specific needs Offer?,” Environment and Planning. C, Gov- of my enterprise. ernment and Policy 22, 859–880. (2) The program facilitators understood my Bennett, R., and P. J. A. Robson (2005). “The business context. Advisor-SMF Client Relationship: Impact, (3) The knowledge in this program was Satisfaction and Commitment,” Small Busi- immediately useful in my business. ness Economics 25(3), 255–271. Bennett, R. J. (2007). “Expectations-Based Practice-Based Approach Evaluation of SME Advice and Consultancy: An Example of Business Link Services,” (1) This program used “implementation of Journal of Small Business and Enterprise changes within your enterprise” as part of Development 14(3), 435–457. the learning. Bennett, R. J., and P. J. A. Robson (1999a). (2) This program used “analysis of your own “Intensity of Interaction in Supply of Busi- enterprise” as part of the learning. ness Advice and Client Impact: A Compari- (3) This program used “instruction on best son of Consultancy, Business Associations practice” as part of the learning. and Government Support Initiatives for (4) This program used “reflection on your SMEs,” British Journal of Management business practices” as part of the learning. 10(4), 351–369. ——— (1999b). “The Use of External Business Organizational Innovativeness Advice by SMEs in Britain,” Entrepreneur- To what extent did the business make ship & Regional Development 11(2), 155– changes to the following . . . 180. Bennett, R. J., W. A. Bratton, J. Paul, and A. (1) Organizational/managerial processes Robson (2000). “Business Advice: The Influ- (2) Operational processes ence of Distance,” Regional Studies 34(9), (3) Management structure 813–828.

258 JOURNALSAWANG, OF SMALL PARKER, BUSINESS AND MANAGEMENT HINE 15 Bennett, R. J., P. Robson, and W. Bratton Greenfield, T. K., L. T. Midanik, and J. D. (2001). “Government Advice Networks for Rogers (2000). “Effects of Telephone versus SMEs: An Assessment of the Influence of Face to Face Interview Modes on Reports of Local Context on Business Link Use, Impact Alcohol Consumption,” Addiction and Satisfaction,” Applied Economics 33(7), (Abingdon, England) 95(2), 277–284. 871–885. Hahn, R. W., and R. N. Stavins (1992). “Eco- Brookfield, S. D. (1991). Developing Critical nomic Incentives for Environmental Protec- Thinkers. San Francisco, CA: Jossey-Bass. tion: Integrating Theory and Practice,” The Brown, J. S., and P. Duguid (1998). “Organizing American Economic Review 82(2), 464–468. Knowledge,” California Management Hair, J. F., R. E. Anderson, R. L. Tatham, and W. Review 40(3), 91–111. C. Black (1998). Multivariate Data Analysis ——— (2001). “Knowledge and Organization: with Readings, 5th ed. Englewood Cliffs, NJ: A Social-Practice Perspective,” Organization Prentice-Hall. Science 12(2), 198–213. Harrington, J. W., A. D. MacPherson, and J. R. Bryson, J. R., D. Keeble, and P. Wood (1997). Lombard (1991). “Interregional Trade in Pro- “The Creation and Growth of Small Bus- ducer Services: Review and Synthesis,” iness Service Firms in Post-Industrial Growth and Change 22(4), 75–94. Britain,” Small Business Economics 9(4), Hillary, R. (1995). Small Firms and the 345–360. Environment: A Groundwork Status Report. Chrisman, J. J., and W. E. McMullan (2000). “A Birmingham, UK: The Groundwork Founda- Preliminary Assessment of Outsider Assis- tion, 3–10. tance as a Knowledge Resource: The Longer- Hjalmarsson, D., and A. W. Johansson (2003). Term Impact of New Venture Counseling,” “Public Advisory Services—Theory and Prac- Entrepreneurship Theory and Practice 24(3), tice,” Entrepreneurship & Regional Develop- 37–54. ment 15(1), 83–98. Chrisman, J. J., E. McMullan, and J. Hall (2005). Hoban, G. (1999). “The Role of Community in “The Influence of Guided Preparation on the Action Learning: A Deweyian Perspective,” Long-Term Performance of New Ventures,” in Learning Communities in Education. Eds. Journal of Business Venturing 20(6), 769– J. Retallick, B. Cocklin, and K. Coombe. 791. London: Routledge, 171–183. Damanpour, F. (1991). “Organizational Innova- Järvinen, A., and E. Poikela (2001). “Modelling tion: A Meta-Analysis of Effects of Determi- Reflective and Contextual Learning at Work,” nants and Moderators,” The Academy of Journal of Workplace Learning 13(7/8), 282– Management Journal 34(3), 555–590. 290. Deschoolmeester, D., T. Schamp, and A. Johannessen, J. A., B. Olsen, and J. Olaisen Vandenbroucke (1997). The Influence of (1999). “Aspects of Innovation Theory Based Management Training on the Entrepreneur- on Knowledge-Management,” International ial Attitudes and Managerial Techniques Journal of Information Management 19, of Small Business-Starters (SMEs): a Com- 121–140. parative Study of Business Planning Kafouros, M. I., P. J. Buckley, J. A. Sharp, and Business-owners. Paper presented at the C. Wang (2008). “The Role of International- Internationalizing Entrepreneurship Educa- ization in Explaining Innovation Perfor- tion and Training Monterey Bay, mance,” Technovation 28(1–2), 63–74. California. Kosters, S., and M. Obschonka (2011). “Public Duffy, T. M., and D. J. Cunningham (1996). Business Advice in the Founding Process: An “Constructivism: Implications for the Design Empirical Evaluation of Subjective and Eco- and Delivery of Instruction,” in Handbook of nomic Effects,” Environment and Planning. Research on Educational Communications C, Government and Policy 29(4), 577– and Technology. Ed. D. H. Jonassen. New 604. York: Scholastic, 170–198. Lambrecht, J., and F. Pirnay (2005). “An Evalu- Forgues, B., and I. Vandangeon-Derumez ation of Public Support Measures for Private (2001). “Longitudinal Analyses,” in Doing External Consultancies to SMEs in the Management Research: A Comprehensive Walloon Region of Belgium,” Entrepreneur- Guide. Ed. R.-A. Thietart. London: Sage Pub- ship & Regional Development 17(2), 89– lications, 332–351. 108.

16 JOURNALSAWANG, OF SMALL PARKER, BUSINESS AND MANAGEMENT HINE 259 Levinson, N. S., and M. Asahi (1995). “Cross- ally,” Journal of Small Business Manage- National Alliances and Interorganizational ment 26(2), 1–7. Learning,” Organizational Dynamics 24(4), Pimenova, P., and R. van der Vorst (2004). “The 50–63. Role of Support Programmes and Policies in Lundström, A., and L. Stevenson (2005). Entre- Improving SMEs Environmental Perfor- preneurship Policy: Theory and Practice. mance in Developed and Transition Econo- New York: Springer. mies,” Journal of Cleaner Production 12(6), Marsick, V. J. (1988). “Learning in the Work- 549–559. place: The Case for Reflectivity and Critical Porter, M. E., and H. M. Ketels (2003). “UK Reflectivity,” Adult Education Quarterly Competitiveness: Moving to the Next Stage,” 38(4), 187–198. DTI Economics Papers 3, Department of Meyers, C. (1991). Teaching Students to Think Trade and Industry, U.K. Critically. San Francisco, CA: Jossey Bass. Rakowski, J. P., and D. Bejou (1992). “Birth, Miller, C. C., L. B. Cardinal, and W. H. Glick Marriage, Life, and Death: A Life-Cycle (1997). “Retrospective Reports in Organiza- Approach for Examining the Deregulated US tional Research: A Reexamination of Recent Airline Industry,” Transportation Journal Evidence,” The Academy of Management 32(1), 15–29. Journal 40(1), 189–204. Ramsden, M., and R. Bennett (2005). “The Ben- Mole, K., and W. Keogh (2009). “The Implica- efits of External Support to SMEs: ‘Hard’ tions of Public Sector Small Business Advis- versus ‘Soft’ Outcomes and Satisfaction ers Becoming Strategic Sounding Boards: Levels,” Journal of Small Business and England and Scotland Compared,” Entrepre- Enterprise Development 12(2), 227–243. neurship and Regional Development 21(1), Raubenheimer, J. (2007). “An Item Selection 77–97. Procedure to Maximise Scale Reliability and Mole, K., M. Hart, S. Roper, and D. Saal (2008). Validity,” SA Journal of Industrial Psychol- “Differential Gains from Business Link ogy 30(4), 59–64. Support and Advice: A Treatment Effects Robson, P. J. A., and R. J. Bennett (2000). “SME Approach,” Environment and Planning C: Growth: The Relationship with Business Government & Policy 26(2), 315–334. Advice and External Collaboration,” Small Mole, K. F., and G. Bramley (2006). “Making Business Economics 15(3), 193–208. Policy Choices in Nonfinancial Business Roglio, K. D. D., and G. Light (2009). “Execu- Support: An International Comparison,” tive MBA Programs: The Development of the Environment and Planning C: Government Reflective Executive,” The Academy of Man- & Policy 24(6), 885–908. agement Learning and Education 8(2), 156– Mole, K. F., M. Hart, S. Roper, and D. S. Saal 173. (2009). “Assessing the Effectiveness of Busi- Rotger, G. P., M. Gørtz, and D. J. Storey (2012). ness Support Services in England,” Interna- “Assessing the Effectiveness of Guided tional Small Business Journal 27(5), 557– Preparation for New Venture Creation and 582. Performance: Theory and Practice,” Journal ——— (2011). “Broader or Deeper? Exploring of Business Venturing 27(4), 506–521. the Most Effective Intervention Profile for Schön, D. A. (1983). The Reflective Practitioner: Public Small Business Support,” Environ- How Professionals Think in Action. New ment and Planning A 43(1), 87–105. York: Basic Books. Nonaka, I., and H. Takeuchi (1995). The Knowl- ——— (1992). The Reflective Turn: Case Studies edge Creating Company: How Japanese Com- in and on Educational Practice. New York: panies Create the Dynamics of Innovation. Teachers College Press. New York: Oxford University Press. Seagal, S., and D. Horne (1997). Human Norman, D. A., and J. C. Spohrer (1996). Dynamics: A New Framework for Under- “Learner-Centered Education,” Communica- standing People and Realizing the Potential tions of the ACM 39(4), 24–27. in Our Organizations. Cambridge, MA: Pearce, D. (1998). “Cost Benefit Analysis and Pergasus Communications. Environmental Policy,” Oxford Review of Storey, D. (2000). “Six Steps to Heaven: Evalu- Economic Policy 14(4), 84–100. ating the Impact of Public Policies to Peterson, R. (1988). “Understanding and Support Small Business in Developing Encouraging Entrepreneurship Internation- Economies,” in The Blackwell Handbook of

260 JOURNALSAWANG, OF SMALL PARKER, BUSINESS AND MANAGEMENT HINE 17 Entrepreneurship. Eds. D. Sexton and H. Strategic Management Journal 28(13), 1319– Landstrom. Oxford: Blackwell, 176–192. 1350. ——— (2003). “Public Policies to Assist Small Wenger, E. (1998). Communities of Practice: and Medium Sized Enterprises,” in Hand- Learning, Meaning, and Identity. Cam- book of Entrepreneurship Research. Ed. Z. A. bridge: Cambridge University Press. A. D. Audretsch. London: Klewer, 473– Wenger, E., and W. Snyder (2000). “Communi- 511. ties of Practice: The Organizational Fron- Storey, D. J., and F. Greene (2010). Small Busi- tier,” Harvard Business Review January– ness and Entrepreneurship. Rotolito February, 139–145. Lombarda, Italy: Pearson. Zahra, S. A., and G. George (2002). “Absorptive Teece, D. J. (2007). “Explicating Dynamic Capa- Capacity: A Review, Reconceptualization, bilities: The Nature and Microfoundations and Extension,” Academy of Management of (Sustainable) Enterprise Performance,” Review 27(2), 185–203.

18 JOURNALSAWANG, OF SMALL PARKER, BUSINESS AND MANAGEMENT HINE 261 Journal of Small Business Business Management Management 2016 2014 54(1), ••(••), pp. pp. 262–278 ••–•• doi: 10.1111/jsbm.12144 Small and Medium-Sized Enterprise’s Entrepreneurial versus Market Orientation and the Creation of Tacit Knowledge* by Ana Pérez-Luño, Patrick Saparito, and Shanti Gopalakrishnan

The ability to create tacit knowledge is important to the competitive advantage of ﬁrms in general but is critical to the survival and growth of small and medium-sized enterprises (SMEs). Conse- quently, SME strategic orientations that facilitate tacit knowledge creation, especially in hostile environments, are important factors that can enhance SME competitiveness. This paper shows that while an entrepreneurial orientation (EO) and environmental hostility are positively related to an SME’s cultivation of tacit knowledge, market orientation (MO) is negatively related to SME’s tacit knowledge. Additionally, we ﬁnd that in benign environments, the relationship between an SME’s MO and tacit knowledge becomes more strongly negative than in hostile environments.

Tacit knowledge is knowledge that is not easily Introduction described or transcribed, and that must be con- The ability to transfer and accumulate knowl- textually grounded to be understood and make edge across a firm’s boundaries is a central sense (Polanyi 1966). Tacit knowledge forms element to a firm’s growth and performance the underlying foundation for idea generation, (Liebeskind 1996; Nahapiet and Ghoshal 1998). adaptation, and by extension, innovation Indeed, the competitive landscape typified by (Nonaka and Takeuchi 1995). Because of the intense competition, compressed product life unique qualities of tacit knowledge, the gen- cycles, and disruptive change exerts increasing eration, accumulation, and retention of this pressure in particular on small and medium- type of knowledge are considered particularly sized enterprises (SMEs) to more effectively important for building competitive advantage acquire, manage, and make sense of knowledge (Barney 1991). Similarly, organizational orien- in order to survive and be competitive tations and practices that facilitate acquisition (Audretsch and Acs 1994; Dean, Brown, and and institutionalization of tacit knowledge are Bamford 1998; Turner and Makhija 2006). important factors that enhance SME competi- Of particular value to SMEs is the ability to tiveness (Grant 1996; Zhao et al. 2011). create and retain tacit knowledge (Audretsch SMEs can have various effective strategic and Acs 1994; Dean, Brown, and Bamford orientations that determine their overall actions 1998; Grant 1996; Lane and Lubatkin 1998). (Hakala 2011). Strategic orientations are social

*Financial support from the Consejería de Economía, Innovación y Ciencia de la Junta de Andalucía, grant SEJ-6392, is gratefully acknowledged. Ana Pérez-Luño is an associate professor in Organization Management and Marketing at Pablo de Olavide University. Patrick Saparito is an associate professor of Strategic Management and Entrepreneurship and director of the Family Business and Entrepreneurship Program at Saint Joseph’s University, Philadelphia, PA. Shanti Gopalakrishnan is a professor at the School of Management at NJIT, New Jersey. Address correspondence to: A. Pérez-Luño, Organization Management and Marketing, Pablo de Olavide University, Carretera de Utrera, Km 1, Seville, Seville 41013, Spain. E-mail: [email protected].

262 PÉREZ-LUÑO,JOURNAL OF SAPARITO, SMALL BUSINESS AND GOPALAKRISHNAN MANAGEMENT 1 learning mechanisms that affect a firm’s search ability to exploit organizational and environ- routines and internal communications, and that mental resources (Cavusgil, Calantone, and influence the way problems are interpreted and Zhao 2003; Grant 1996; Lane and Lubatkin 1998; solved (Atuahene-Gima and Ko 2001; Gatignon Lee and Choi 2003; Nonaka and Takeuchi 1995). and Xuereb 1997; George and Zahra 2002; The primary purpose of this paper is to take Hult, Hurley, and Knight 2004). Two such some initial steps to fill this research gap by orientations are entrepreneurial orientation attempting to understand the role of an SME’s and market orientation. These orientations entrepreneurial and market orientation on its are correlated but remain distinct constructs cultivation and retention of tacit knowledge. (Baker and Sinkula 2009). An entrepreneurial Additionally, Miller (2011) calls for further atten- orientation is one in which top managers are tion to be paid to the influence of environmental proactive, inclined to take risks, and engage in context when examining SME strategic orienta- exploratory behavior that favors innovation tion. Thus, as a second step, and based on the and frequent and dramatic technological assumption that SMEs increasingly compete in change (Atuahene-Gima and Ko 2001; Covin hostile environments (Zahra and Garvis 2000), and Lumpkin 2011; Lumpkin and Dess 1996). we explore the role of environmental hostility in Thus, SMEs with an entrepreneurial orientation moderating the relationship between strategic often introduce pioneering new products to orientation and tacit knowledge. gain competitive advantage and are willing to We pursue three specific research questions: compete aggressively (Covin and Slevin 1989; (1) Is an SME’s entrepreneurial and market Lumpkin and Dess 1996; Miller 1983). Alterna- orientation related to the extent to which the tively, SMEs with a market orientation often firm cultivates tacit knowledge? (2) Does envi- focus on satisfying novel customer needs ronmental hostility influence the appropriate through more incremental change and refine- extent of tacitness of an SME’s knowledge base? ment of products to create superior value and (3) Does environmental hostility impact the build long-term customer relationships relationship between strategic orientation and (Maatoofi and Tajeddini 2011; Narver and Slater tacit knowledge? 1990). The paper proceeds as follows. First, using Research within the entrepreneurship and the knowledge-based view (KBV) of the firm, marketing literatures suggests that a strategic we describe the concepts of tacit knowledge orientation also influences the primary way that and the knowledge management process. organizations learn (e.g., explorative versus Second, we hypothesize both direct and inter- exploitative learning) and posits how the use of action effects of entrepreneurial orientation, these predominant modes of organizational market orientation, and environmental hostility learning within particular firms influences the on the creation and use of tacit knowledge. We type and level of innovation within the firm present our methods and test our hypotheses (Atuahene-Gima and Ko 2001; Covin and Slevin using a sample of 374 SMEs in Spain. We con- 1989; Narver and Slater 1990; Wang 2008; Zhao clude by discussing our results and implications et al. 2011). Past work on entrepreneurial and for future research. market orientation has focused primarily on their influence on SME financial performance Tacit Knowledge and (Baker and Sinkula 2009; Li et al. 2008). Addi- Organizational Learning tionally, some attention has been paid to their The KBV of the firm focuses on the nature influence on nonfinancial factors, such as firm and role of knowledge in a firm’s efforts to owner satisfaction (Rauch et al. 2009) or inno- build competitive advantage (Grant 1996; vation (Renko, Carsrud, and Brännback 2009). Turner and Makhija 2006; Zander and Kogut While there has been occasional discussion 1995). Knowledge is deep contextual under- about the role of tacit knowledge in this litera- standing that surrounds facts, and it is contin- ture (Zhao et al. 2011), its role has not been fully gent on the task at hand (Nonaka 1994). discussed, and it has not been empirically tested. Business situations and circumstances are often This is despite the fact that within the innovation ambiguous and subject to multiple and con- management literature, tacit knowledge has flicting interpretations. In order to make long been considered an outcome variable pro- sense about potential outcomes, ambiguous viding substantive importance to firm situations require a higher understanding of the innovativeness, competitive advantage, and the context and relationships surrounding the situ-

2 PEREZ-LUJOURNALNO,~ OF SAPARITO, SMALL BUSINESS AND GOPALAKRISHNAN MANAGEMENT 263 ation (Daft and Lengel 1986; Nonaka 1994). Slater and Narver 1995). These learning pro- Thus, special capabilities for accumulating and cesses can help create new knowledge inter- interpreting knowledge can translate into an nally within an SME (experimentation, R&D, important source of competitive advantage knowledge recombination, etc.), and they may (Grant 1996; Nahapiet and Ghoshal 1998). also influence the search for knowledge While researchers identify numerous attri- acquired from external sources (Turner and butes associated with the nature of knowledge, Makhija 2006). Clearly, learning is path depen- perhaps the most widely discussed attribute is dent in that an SME’s ability to use tacit knowl- the continuum of knowledge codifiability tacit- edge will depend upon the firm’s current ness (Grant 1996; Kogut and Zander 1993; knowledge base and learning processes Polanyi 1966). Codifiability refers to the extent (Nonaka 1994; Turner and Makhija 2006) in that knowledge is unambiguous, observable, that they provide a framework both for the and indisputable (Turner and Makhija 2006). search for new knowledge and a context for Highly codifiable knowledge is also known as interpreting existing tacit knowledge (Nonaka explicit knowledge that can be transcribed, 1994; Turner and Makhija 2006). removed from context, and transferred without Generally, the literature of organizational loss of meaning (Polanyi 1966). At the other learning identifies two types of learning. First, end of the continuum is tacit knowledge. Tacit exploratory or acquisitive learning refers to the knowledge is not easily codifiable and needs to external search, acquisition, and subsequent be contextually grounded in order to be under- internalization of knowledge (Dess et al. 2003; stood (Polanyi 1966). As knowledge becomes March 1991; Sapienza et al. 2005; Turner and more tacit, greater interaction and socialization Makhija 2006). It involves the pursuit of new are required among individuals for successful knowledge that does not currently exist in the knowledge transfer. Socialization and shared firm to create new customer value or that experiences create common contexts, values, replaces a firm’s existing knowledge to enrich understandings, and even emotions that allow current customer value. individuals to share each other’s thinking pro- Alternatively, exploitative or experimental cesses. These common ways of looking at the learning occurs internally to the SME and gen- world become increasingly important in order erates knowledge that is unique and distinct to for individuals to make sense of tacit knowl- the SME itself (Dess et al., 2003; March 1991; edge (Grant 1996; Nonaka 1994; Polanyi 1966). Sapienza et al. 2005; Turner and Makhija 2006). Indeed, tacit knowledge will often make little It involves refining and deepening existing sense if it is separated from embedded emo- knowledge that is unique to the SME and tions and contexts associated within a shared results in expanding or enriching current cus- experience. tomer value (Kang, Morris, and Snell 2007). It is argued that a firm’s tacit knowledge is While the entrepreneurship literature has positively associated with the ability to inno- explored how strategic orientation influences vate, create value, and identify opportunities learning processes (e.g., Dess et al. 2003; (Conner and Prahalad 1996; Grant 1996; Sapienza et al. 2005; Zhao et al. 2011), there Liebeskind 1996; Turner and Makhija 2006; has not been sufficient research exploring the Zander and Kogut 1995). Thus, we submit that impact of strategic orientation on the type of an SME’s ability to accumulate and interpret knowledge acquired and accumulated by SMEs. tacit knowledge is a critical element to the SME’s future success and competitive advan- Entrepreneurial Orientation tage. Given the importance of an SME’s tacit Both entrepreneurial and market orienta- knowledge, an examination of the SME’s pro- tions are alternative strategic orientations that a cesses for accumulating and managing tacit firm can take. Strategic orientations are social knowledge becomes fundamentally important. learning and selection mechanisms that aim to Organizational learning is centrally impor- maintain coherence between strategic intent tant to the acquisition and development of tacit and operational activities (Burns and Stalker knowledge by SMEs (Grant 1996; Nonaka and 1961; Miller and Friesen 1978; Mintzberg 1973). Takeuchi 1995). Organizational learning is a They create internal environments where process that includes knowledge acquisition, desired behaviors are encouraged and sup- dissemination, shared interpretation, and orga- ported (Atuahene-Gima and Ko 2001). The stra- nizational memory (Cohen and Levinthal 1990; tegic management literature presents evidence

264 PÉREZ-LUÑO,JOURNAL OF SAPARITO, SMALL BUSINESS AND GOPALAKRISHNAN MANAGEMENT 3 that an organization’s strategic orientation On the surface, it might appear that in order affects the firm’s search routines and internal for an SME to recognize potential customer communications processes, influences the way needs and to pursue these market opportuni- problems are interpreted and addressed, and ties, they must act in an entrepreneurial provides a significant indicator of the firm’s fashion. Thus, one could argue that to pursue a ability to gain sustainable competitive advan- market orientation, an SME must in part have tage (Atuahene-Gima and Ko 2001; Gatignon an entrepreneurial orientation. However, the and Xuereb 1997; George and Zahra 2002; two concepts are distinct (Baker and Sinkula Hult, Hurley, and Knight 2004). Thus, it 2009) in that the core of an entrepreneurial appears that an SME’s strategic orientation may orientation is to focus on risky innovations that influence the organization’s ability to accumu- may exceed customers’ needs or that may late, transfer, and interpret tacit knowledge. create new and emerging product markets, An entrepreneurial orientation is a strategic pushing the frontier of technology (technology orientation characterized by innovativeness, push), while the core of a market orientation is risk taking, and proactivity (Covin and to create superior customer value through Lumpkin 2011; Miller 1983). It reflects the understanding of and responding to immediate degree to which firm’s growth objectives are customer needs (market pull) (Hamel and driven by the identification and exploitation of Prahalad 1994; Slater and Narver 1995). Indeed, untapped market opportunities (Baker and it is suggested that because customers can often Sinkula 2009). Thus, SMEs with an entrepre- be shortsighted, SMEs with strong market ori- neurial orientation would emphasize aggressive entations and a tight focus on serving customer product innovation and higher risk projects and needs may risk missing potential benefits would have a proclivity for pioneering innova- derived from more creative innovation efforts tions that preempt the competition (Covin and (Hamel and Prahalad 1994). Lumpkin 2011; Covin and Slevin 1989; Miller SMEs can use their accumulated knowledge 1983; Slater and Narver 1993, 1995). Innovation to identify and pursue various opportunities within SMEs with an entrepreneurial orienta- (Grant 1996). Accumulated knowledge pro- tion tends to be more radical with regard to vides a basis from which new knowledge can creating new products/markets, moving a be understood and interpreted, can foster inno- company towards new challenges, and compet- vation, and can facilitate an organization’s ing aggressively against other companies ability to adapt to change (Cavusgil, Calantone, (Covin and Lumpkin 2011; Covin and Slevin and Zhao 2003; Gold, Malhotra, and Segars 1989; Hamel and Prahalad 1994; Miller 2011; 2001; Turner and Makhija 2006). However, the Slater and Narver 1995). association between strategic orientation and an SME’s tacit knowledge has yet to be fully Market Orientation explored. Therefore, in the next section, we Alternatively, a market orientation is a stra- posit how an SME’s entrepreneurial or market tegic orientation characterized by a set of orientation can influence the tacitness of the behaviors and activities related to an SME’s SME’s accumulated knowledge. strong customer focus, coordinated marketing across the organization, and profitability Entrepreneurial Orientation (George and Zahra 2002; Hunt and Morgan 1995; Kohli, Jaworski, and Kumar 1993). Kohli, as a Source of Tacit Jaworski, and Kumar (1993, p. 468) identify Knowledge three sets of activities associated with a market Entrepreneurs are a creative force that breed orientation including: (1) intelligence genera- radical innovations, obsolesce the rules of the tion, which includes “the collection and assess- game by which industries or economies ment of both customer needs/preferences and operate, and destroy old ways of doing things the forces that influence the development and (Damanpour 1991; Schumpeter 1961). This refinement of those needs”; (2) intelligence dis- process referred to Schumpeter (1961) as cre- semination, which is “the process and extent of ative destruction describes how entrepreneurial market exchange within a given organization”; rent-seeking activities are the central source of and (3) responsiveness, which is “action taken innovation and an endogenous source of eco- in response to intelligence that has been gen- nomic change. In terms of results, Schumpe- erated and disseminated.” terian innovation is principally radical—it

4 PEREZ-LUJOURNALNO,~ OF SAPARITO, SMALL BUSINESS AND GOPALAKRISHNAN MANAGEMENT 265 destroys existing industries and creates new that these unique pools of knowledge that are ones (Damanpour 1991; Schumpeter 1961). created are largely tacit in nature. SMEs that Indeed, SMEs with a high entrepreneurial ori- produce radical innovations are more likely to entation value the creation of new businesses create new knowledge through exploiting and and the rejuvenation and transformation refining existing knowledge that already exists of current businesses that have stagnated or within the firm, and the recombination and declined (Covin and Lumpkin 2011; Miller refinement of these existing stocks of knowl- 2011; Slater and Narver 1995). Thus, an entre- edge are likely to create new knowledge that is preneurial orientation suggests a propensity also tacit in nature (Zhao et al. 2011). Thus, the toward the creation of new products, auto- ability to be entrepreneurial is based on knowl- nomy, the assumption of greater risks, edge and insights drawn from past experience proactivity, and competitive aggressiveness and from aspects of decision-making that are (George and Zahra 2002; Lumpkin and Dess largely tacit (Klein 1999). Ultimately, the ability 1996; Wiklund and Shepherd 2005). of a firm to be entrepreneurial is based on the In terms of its entrepreneurial source, collective intuition and experience of its Schumpeterian radical innovation is the result members and tends to be generated from knowl- of either the creation of totally new knowledge edge within the firm. Based upon this, we or the recombination of existing knowledge in suggest: new ways (Galunic and Rodan 1998). Both knowledge creation and recombination require H1: There is a positive relationship between an the firm to be able to manage complex and SME’s entrepreneurial orientation and tacit difficult-to-transfer technological knowledge knowledge. (Lane and Lubatkin 1998; Tolstoy 2009), which is better developed internally than externally. The recombination of knowledge at the orga- Market Orientation as a nizational level is accomplished through a syn- Source of Explicit Knowledge thesis of the combined specialized knowledge As previously mentioned, a market orienta- and experience of individuals within the firm tion focuses on creation and maintenance of (Grant 1996; Kogut and Zander 1992). customer needs and responsiveness to market Based on these ideas, researchers have sug- information (Slater and Narver 1995). Market- gested that the entrepreneurial orientation of oriented firms generate information on changes SMEs will be positively associated with exploit- in customer needs and their markets (Golann ative learning that builds upon stocks of accu- 2006). While a positive outcome of a market mulated knowledge and that extends this orientation is the creation of customer value learning in ways that can be a source of com- through refinement and adaptation of products petitive advantage (Dess et al. 2003; Sapienza and services, the downside of a market orienta- et al, 2005; Slater and Narver 1995; Wang 2008; tion is that a company’s focus may be limited to Zhao et al. 2009). In such a process, an SME’s only the expressed needs of customers knowledge base is “refined, transformed, (Atuahene-Gima and Ko 2001; Slater and Narver extended, and updated so that innovations can 1995). Researchers suggest that because custom- be achieved . . .” (Zhao et al. 2011, p. 296). This ers are generally focused on their immediate refining process takes place by sharing special- needs, market-oriented SMEs risk becoming ized and private knowledge among various myopic and lose the vision for creative innova- organizational members within the SME, bring- tion (Hamel and Prahalad 1994; Slater and ing together knowledge that is otherwise dis- Narver 1995). Thus, market-oriented SMEs may persed within the firm. Such an approach to focus on the refinement and adaption of current creating new knowledge is generated internally innovations to meet current customer needs from common experience (Dess et al. 2003; rather than the development of new products Zhao et al. 2011). targeted at potential or emerging new needs While past research suggests that an entre- (Atuahene-Gima and Ko 2001; March 1991). preneurial orientation will foster such an In support of this strategic orientation, exploitative learning process (Dess et al. 2003; market-oriented SMEs are more likely to Sapienza et al., 2005; Slater and Narver 1995; engage in learning that is explorative in nature Wang 2008; Zhao et al. 2011), we extend this (Slater and Narver 1995). Thus, in support of a work by suggesting that the ultimate result is market orientation, SMEs focus on identifying

266 PÉREZ-LUÑO,JOURNAL OF SAPARITO, SMALL BUSINESS AND GOPALAKRISHNAN MANAGEMENT 5 changes in the environment, assimilating this available resources. Firms operating in less knowledge internally to the organization, and hostile environments that are less resource con- responding to this new knowledge. The trans- strained could make do with a lesser degree of fer and accumulation of this new knowledge tacit knowledge and experience. about market trends are often combined with the SME’s current knowledge in a way that H3: There is a positive relationship bet- previously held assumptions are not challenged ween environmental hostility and tacit and the dominant existing logic of the firm is knowledge. left unchanged (Hamel and Prahalad 1994; Slater and Narver 1995). We suggest that the Environmental Hostility’s more market oriented the SME, the greater its Moderating Effects between propensity to acquire explicit knowledge by either exploring customer needs or competitor Strategic Orientation and actions. Thus, the market orientation necessi- Tacit Knowledge tates increasing sensitivity to the multiple Prior research has shown that environmental parties that constitute the external environ- hostility plays a key role in moderating ment. Based upon this, we propose: between different types of firm behavior and performance (Calantone, Schmidt, and di H2: There is a negative relationship between Benedetto 1997; Lindelhof and Lofsten 2006; market orientation and tacit knowledge. Zahra and Garvis 2000). For example, Calantone, Schmidt, and di Benedetto (1997) Environmental Hostility, found that in hostile environments, improving the proficiency of new product development Strategic Orientation, and activities greatly increased the likelihood of Tacit Knowledge product success. Firms in hostile international Environmental hostility refers to the extent environments that pursued higher levels of or lack of resources available to the firm internal corporate venturing had higher levels (Mintzberg 1979). When the firm has fewer of profitability as measured by return on assets resources available, the environment is consid- (Zahra and Garvis 2000). In other words, ered hostile since the firm is constrained and hostile environments necessitate a firm making may not have the requisite resources to adapt. more efficient use of existing resources since Environments may be hostile for various there is fewer rooms for experimentation and reasons. For example, radical changes in the error (Lindelöf and Löfsten 2006). These ideas industry, intense regulatory burdens placed on lead us to propose that environmental hostility the industry, or fierce rivalry among competi- will reinforce the positive relationship between tors (Werner, Brouthers, and Brouthers 1996) entrepreneurial orientation and tacit knowl- can cause environmental hostility. Hostility can edge. Thus, we suggest: also result from perceived product- and market- related uncertainties (Dess and Beard 1984). H4: The positive relationship between entrepre- Firms in such situations must devote scarce neurial orientation and tacit knowledge will resources to managing within environmental strengthen as the hostility of the environ- uncertainty and adversity in order to achieve ment increases. their organizational and performance goals (Zahra 1993). We had hypothesized that there is a negative Hostile environments require that firms rely relationship between market orientation and more on past learning and draw from their tacit knowledge. We believe that in a hostile previous experience to navigate the current environment with scarce resources, the uncertainties that they face. Tacit knowledge is negative effects of market orientation on tacit the repository of knowledge that firms have knowledge will be reduced. The reason is that based on their prior actions and experience. even when market-oriented companies focus Firms need to use this knowledge even more in on the refinement and adaption of current inno- a hostile environment, where resources are in vations to meet current customer needs using short supply. Thus, hostile environmental con- explicit knowledge (Atuahene-Gima and Ko ditions require that firms be more innovative 2001), due to the lack of resources faced by and creative and make more efficient use of firms in hostile environments, the initial

6 PEREZ-LUJOURNALNO,~ OF SAPARITO, SMALL BUSINESS AND GOPALAKRISHNAN MANAGEMENT 267 negative relationship between MO and tacit Measures knowledge will be reduced. With the exception of our three control variables—SME size, R&D expenditures, and H5: The negative relationship between market industry—all constructs were measured with orientation and tacit knowledge will multi-item scales. We took several steps to weaken as the hostility of the environment ensure data validity and reliability. First, we increases. pretested all measures in 25 interviews with R&D managers and asked them to closely review the survey, to ensure the clarity of the Research Method questions, and to ascertain whether or not the Research Design and Sample scales captured the desired information. We Between March and November 2006, data then revised any potentially confusing items for the study’s dependent, independent, and before submitting the questionnaire. The indi- control variables were collected. We used the vidual measurement items for the study’s SABI1 database to identify all companies in the dependent, independent, and control variables most innovative industries.2 There were a total are listed in Table 1. of 2,845 SMEs having more than 10 workers in Tacit knowledge, entrepreneurial orienta- our target industries. tion, market orientation, and environmental First, the 2,845 firms were contacted by hostility loaded onto four separate factors. Each telephone, and shortly thereafter, all firms factor with eigenvalues greater than 1.0 interviewed were sent a mail survey. We (ranging from 1.99 to 8.91) accounted for 54.19 spoke to the R&D manager. If the firm did percent of the total variance. Additionally, each not have an R&D manager, we instead spoke item loaded strongly on its respective factor to the CEO. In total, 2,685 SMEs responded to and no cross-loading was above the 0.40 level, our phone calls (response rate of 94.38 thus providing evidence that the measures are percent). During the interview, we first distinct (Cohen and Cohen 1983). ensured that the firm indeed belonged to the sample frame (i.e., that it operated within one Knowledge Tacitness (Dependent Variable). of the target sectors and that it had more than We measured knowledge tacitness using a 10 employees). Those firms with fewer than Subramaniam and Venkatraman (2001) six-item 10 employees (19), that did not belong to our scale. These authors adapted Zander and target sectors (539), or that were duplicated Kogut’s (1995) measure. Zander and Kogut or without real activity (443) were excluded (1995) considered that knowledge could from our sample. We sent questionnaires to be analyzed as a function of its codifiability the remaining 1,684 SMEs. In total, 376 SMEs and teachability. Then, Subramaniam and responded to this questionnaire, of which 374 Venkatraman (2001) used some of their items responses were considered valid. This corre- to develop a measure of what they call tacit sponds to a response rate of 22.33 percent of knowledge (see Table 1). SME managers rated the firms in our target population. In order to the six items using a seven-point scale. check for nonresponse bias, we compared mean differences between respondents and Entrepreneurial Orientation (Independent nonrespondents for industry membership, Variable). We measured entrepreneurial ori- number of employees, and revenue. No sig- entation using Covin and Slevin’s (1991) nine- nificant differences were found, suggesting item scale (see Table 1). SME managers rated that nonresponse bias was not present. the nine items using a seven-point scale.

1Spanish regulations oblige the companies located in Spain to disclose their yearly accounts in an official and public register overseen by Spanish economic authorities. The SABI database compiles this data, including location, financial information, industrial sector, ownership, and other miscellaneous information. Therefore, it is the most comprehensive database of company information in Spain. 2The National Statistical Institute (INE) of Spain identified five industries of the economy as containing the most “innovative” firms. These industries are: NACE 24, chemical companies; NACE 32, radio, TV, and communication equipment; NACE 33, medical, precision, and optical instruments; NACE 34, manufacture of motor vehicles, trailers, and semitrailers; and NACE 35, manufacture of other transport equipment.

268 PÉREZ-LUÑO,JOURNAL OF SAPARITO, SMALL BUSINESS AND GOPALAKRISHNAN MANAGEMENT 7 Table 1 Results of Factor Analysis for Multi-Item Measures

MO EH KT EO

Market Orientation (Narver and Slater 1990) (Cronbach’s alpha. 0.934) MO1. Our firm has a customer commitment 0.775 0.017 0.038 −0.086 MO2. Our firm creates customer value 0.773 0.087 0.047 −0.109 MO3. Our firm understands customer needs 0.796 −0.011 0.048 −0.094 MO4. Our firm has customer satisfaction objectives 0.794 −0.037 0.024 −0.098 MO5. Our firm measures customer satisfaction 0.737 −0.037 0.084 −0.056 MO6. Our firm has alter-sales service 0.607 0.099 0.016 0.061 MO7. Salespeople share competitor information 0.676 0.157 0.134 0.144 MO8. Our firm responds rapidly to competitors’ actions 0.659 0.110 0.030 0.109 MO9. Top managers discuss competitors’ strategies 0.691 0.149 0.082 0.156 MO10. Our firm targets opportunities for competitive 0.758 0.194 0.017 0.068 advantage MO11. Interfunctional customers calls 0.730 0.070 0.008 −0.067 MO12. Information shared among functions 0.784 0.121 0.084 0.044 MO13. Functional integration in strategy 0.737 0.171 0.062 0.012 MO14. All functions contribute to customer value 0.754 0.127 0.101 −0.131 MO15. Share resources with other business units 0.614 0.136 0.034 0.067 Environment Hostility (Khandwalla 1977) (Cronbach’s alpha. 0.724) How would you characterize the external environment within which your firm operates? H1. Very safe with little threat to the survival and 0.003 0.101 −0.103 0.774 well-being of my firm (versus very risky. a false step can be my firm’s undoing) H2. Rich in investment and marketing opportunities −0.068 −0.104 −0.150 0.746 (versus very stressful. exacting. hostile; very hard to keep afloat) H3. An environment that my firm controls and 0.039 −0.107 0.057 0.777 manipulates to its own advantage. (versus a dominating environment in which my firm’s initiatives count for very little against tremendous competition) Knowledge Tacitness (Adapted from Subramaniam and Venkatraman 2001) (Cronbach’s alpha. 0.820) How would you characterize the knowledge used in the innovation development? K1. Was easy to comprehensively document in manuals 0.096 0.071 0.820 0.026 and reports K2. Was easy to comprehensively understand from written 0.156 0.080 0.844 0.018 documents K3. Was easy to precisely communicate through written 0.117 0.088 0.832 0.005 documents K4. Was obvious to all competitors 0.103 −0.138 0.640 −0.035 K5. Was easy to identify without personal experience in −0.052 −0.156 0.622 −0.121 overseas locations K6. Was complex (1) versus simple (7) 0.044 −0.108 0.581 −0.084 Entrepreneurial Orientation (Covin and Slevin 1989) (Cronbach’s alpha. 0.851) In general. the top managers of my firm favor . . . OE1. A strong emphasis on the marketing of tried and 0.240 0.578 −0.022 0.115 true products (versus a strong emphasis on R&D. technological leadership. and innovations)

8 PEREZ-LUJOURNALNO,~ OF SAPARITO, SMALL BUSINESS AND GOPALAKRISHNAN MANAGEMENT 269 Table 1 Continued MO EH KT EO

OE2. We have not introduced any new product (versus we 0.133 0.542 −0.072 0.255 have introduced many new product) OE3. Changes in products have been mostly of a minor 0.193 0.682 −0.087 0.050 nature (versus changes in products have usually been quite dramatic) In dealing with its competitors. my firm . . . OE4. Typically responds to actions which competitors 0.085 0.738 −0.018 0.007 initiate (versus typically initiates actions which competitors then respond to) OE5. Is very seldom the first business to introduce new −0.087 0.578 −0.080 −0.081 products or technologies (versus is very often the first business to introduce new products or technologies) OE6. Typically seeks to avoid competitive clashes. 0.047 0.701 0.022 0.013 preferring a “live-and-let-live” posture (versus typically adopts a very competitive. “undo-the-competitors” posture) In general. the top managers of my firm . . . OE7. Have a strong proclivity for low-risk projects (with 0.071 0.694 0.016 −0.087 normal and certain rates of return) (versus have a strong proclivity for high-risk projects, with chances of very high return) OE8. Believe that owing to the nature of the environment. 0.153 0.727 −0.002 −0.105 it is best to explore it gradually via timid, incremental behavior (versus believe that owing to the nature of the environment, bold, wide-ranging acts are necessary to achieve the firm’s objectives) When confronted with decision-making situations involving uncertainty. my firm . . . OE9. Typically adopts a cautious “wait-and-see” posture in 0.129 0.723 −0.011 −0.168 order to minimize the probability of making costly decisions (versus typically adopts a bold, aggressive posture in order to maximize the probability of exploiting potential opportunities Eingenvalues 8.913 3.968 3.012 1.989 Explained Variance 24.843 13.039 10.094 6.214 Cumulative Explained variance 24.843 37.881 47.975 54.189

Note: MO = market orientation; EH = environmental hostility; KT = knowledge tacitness; EO = entrepreneurial orientation.

Market Orientation (Independent Variable). using Khandwalla’s (1977) three-item scale. We measured market orientation using Narver SME managers rated the 15 items using a seven- and Slater’s (1990) ﬁfteen-item scale (see point scale. Table 1). SME managers rated the 15 items using a seven-point scale. Factor Analysis of Multi-Item Measures As is shown in Table 1, tacit knowledge, Environmental Hostility (Independent Vari- entrepreneurial orientation, market orientation, able). We measured environmental hostility and environmental hostility were loaded onto

270 PÉREZ-LUÑO,JOURNAL OF SAPARITO, SMALL BUSINESS AND GOPALAKRISHNAN MANAGEMENT 9 four separate factors each with eigenvalues inclusion of the interaction terms, the variance greater than 1.0 (ranging from 1.99 to 8.91) inflation factors ranged from 1.06 to 1.93, accounting for 54.19 percent of the total vari- which are below the suggested 10 cutoff ance. Additionally, each item loaded strongly level. on its respective factor and no cross-loading The hierarchical regression results are dis- was above the 0.40 level, thus providing evi- played in Table 3. The base models displayed dence that the measures are distinct (Cohen in the first column explain a nonstatistically and Cohen 1983). significant share of the variance. The main effects model is in the next column and makes Control Variables. R&D Expenditures. We a significant contribution over and above the controlled for internal and external R&D base models ( R² = 0.049, p < .001). H1 sug- expenditures, size, and the company’s industry. gested that there was a positive relation- The R&D expenditures data were obtained ship between entrepreneurial orientation and from the questionnaire, while we took the size knowledge tacitness. As is shown in Table 3 and industry data from the SABI database. (independent model), the relationship between Internal R&D expenditures are measured as an SME’s entrepreneurial orientation and internal average percentages of the sales turn- knowledge tacitness is positive and modera- over of the company for the last five years. tely significant (β = 0.105, p < .10). Thus, H1 External R&D expenditures are measured as received moderate support. external average percentages of the sales turn- H2 suggested there was a negative relation- over of the company for the last five years. Size ship between an SME’s market orientation and variable was measured by the number of knowledge tacitness. As is shown in Table 3 employees. (independent model), the relationship between market orientation and knowledge tacitness is Industry. Industry effects were captured by negative and significant (β = −0.201, p < .001). dummy variables for each firm’s main sector as Thus, H2 is strongly supported. indicated by their industry code (NACE code) H3 suggested there was a positive relation- taken from the sample frame. Dummy variables ship between environmental hostility and envi- were created for industries NACE 24, Chemical ronmental tacitness. As is shown in Table 3 companies; NACE 32, radio, TV, and communi- (independent model), the relationship between cations equipment; NACE 33, medical, preci- environmental hostility and knowledge tacit- sion, and optical instruments; NACE 34, ness is positive and significant (β = 0.104, manufacture of motor vehicles, trailers, and p < .05). Thus, H3 is supported. semitrailers; and NACE 35, manufacture of To test H4 and H5 we created interaction other transport equipment. terms by multiplying environmental hostility with both entrepreneurial orientation and Results market orientation (Cohen and Cohen 1983). Table 2 provides means, standard devia- The moderating relationships proposed within tions, and correlations for all continuous vari- the hypotheses would be supported if the inter- ables. Skewness (ranged from −3.09 to 0.01) action terms are statistically significant, and and kurtosis (ranged from −0.08 to 12.11) they made a significant explanatory contribu- statistics of the dependent variable fall well tion over and above the direct effects (Cohen within the boundaries for normality (Shapiro and Cohen 1983). We also plotted graphs to and Wilk 1965), allowing parametric tests of further understand these relationships. significances. The two interactions were entered sepa- We tested hypotheses using hierarchical rately for each dependent variable as recom- multiple regression. In order to reduce multi- mended in the literature (Cohen and Cohen collinearity, we centered entrepreneurial orien- 1983). H4 suggested that the positive relation- tation, market orientation, and environmental ship between entrepreneurial orientation and hostility (the variables are in deviation score knowledge tacitness would strengthen (i.e., form so that their means are zero) before cre- become more positive) as environmental hos- ating interaction terms that were produced by tility increased. The interaction term for entre- multiplying together the centered predictors. preneurial orientation and environmental There were no problems with multicollinearity hostility is reported in column 3 of Table 3 tests for all regressions (Table 3). Even with the and does not make a significant contribution

10 PEREZ-LUJOURNALNO,~ OF SAPARITO, SMALL BUSINESS AND GOPALAKRISHNAN MANAGEMENT 271 272 ÉE-UO AAIO N OAARSNN11 GOPALAKRISHNAN AND SAPARITO, PÉREZ-LUÑO, ORA FSALBSNS MANAGEMENT BUSINESS SMALL OF JOURNAL Table 2 Means, Standard Deviations, and Correlations

Mean S.D. 1 2 3 4 5 6 7 8

1. Tacit Knowledge 3.612 1.045 1 2. EO 4.204 0.975 0.060 1 3. MO 5.718 0.929 −0.187** 0.271** 1 4. Hostility 4.206 1.073 0.118* −0.008 −0.037 1 5. Internal R&D 9.834 14.453 0.024 0.269** 0.105* 0.035 1 6. External R&D 2.541 6.880 0.012 0.100* 0.055 −0.096 0.285** 1 7. Age 31.323 23.526 0.036 0.018 −0.021 0.086 −0.075 −0.052 1 8. Workers 156.093 395.471 0.020 0.093 −0.002 0.118* −0.076 −0.023 0.209** 1

*p < .05 **p < .01 S.D., standard deviation. 12

Table 3 Regression Results

Dependent Variables Base Model Independent Model Contingent Model 1 Contingent Model 2

β t Statistic β t Statistic β t Statistic β t Statistic P EREZ-LU ORA FSALBSNS MANAGEMENT BUSINESS SMALL OF JOURNAL Control Variables Industry 24 −0.073 −0.496 −0.036 −0.253 −0.035 −0.246 0.020 0.140 Industry 32 0.047 0.246 0.063 0.339 0.064 0.345 0.106 0.574 O AAIO N OAARSNN273 GOPALAKRISHNAN AND SAPARITO, NO, ~ Industry 33 0.044 0.220 −0.010 −0.051 −0.003 −0.014 0.027 0.140 Industry 34 −0.133 −0.783 −0.136 −0.817 −0.130 −0.774 −0.129 −0.781 Workers 0.083 1.548 0.062 1.154 0.064 1.176 0.072 1.348 Internal R&D 0.016 0.294 0.001 0.024 0.002 0.031 0.016 0.228 External R&D −0.003 −0.059 0.015 0.271 0.015 0.272 0.013 0.246 Main Effect Variables EO 0.105**** 1.870 0.101**** 1.788 0.072 1.265 MO −0.201*** −3.764 −0.207*** −3.822 −0.207*** −3.904 Hostility 0.104* 1.972 0.105* 1.990 0.099**** 1.903 Interactions OE × Hostility 0.035 0.684 OM × Hostility 0.137** 2.938 Model R2 0.010 0.059 0.060 0.081 Adjusted R2 −0.009 0.033** 0.032** 0.053** F statistic 0.536 2.276 2.109 2.898 Change in R2 0.049*** 0.001 0.022** Change in F 6.283 0.467 8.632

*p < .05 **p < .01 ***p < .00. ****p < .1 Figure 1 in one time period may also influence its stra- Interaction of Market tegic orientation in a subsequent time period, which in turn reinforces greater accumulation Orientation and Environmental of a certain type of knowledge (Zhao et al. Hostility 2011). The type of environment in which the firm operates moderates the relationship 1 between strategic orientation and tacit knowledge accumulation. We proposed five hypotheses suggesting various relationships between an SME’s entrepreneurial orientation, an SME’s market orien- 0 Low Hostility High Hostility tation and environmental hostility with the extent of the knowledge tacitness created and Knowledge Tacitness used by the SME. Four of our five hypotheses are supported. As proposed, our results show –1 that while an entrepreneurial orientation is Low Market Orientation High Market Orientation positively related to a firm’s tacit knowledge, market orientation is negatively related to an SME’s tacit knowledge. Additionally, we find that environmental hostility is positively related over and above the main effects ( R² = 0.001, to an SME’s tacit knowledge. With regard to p > .100), and the interaction term is not sig- moderating effects, we find that the relation- nificant (β = −0.035, p > .100). Thus, H4 is not ship between an SME’s market orientation and supported. tacit knowledge becomes more strongly nega- H5 suggested that the negative relationship tive in benign than in hostile environments. between market orientation and knowledge However, environmental hostility does not tacitness weaken (i.e., become less negative) as seem to have a moderating effect on the rela- environmental hostility increased. The interac- tionship between an SME’s entrepreneurial ori- tion of market orientation and environmental entation and tacit knowledge. hostility reported in column 4 of Table 3 makes It is interesting that our results show that a significant contribution over and above the while an SME’s entrepreneurial orientation is main effects ( R² = 0.022, p < .01) and is statis- positively related to the SME’s creation and use tically significant (β = 0.137, p < .01). To deter- of tacit knowledge, market orientation is nega- mine the nature of the significant interaction, tively related to an SME’s tacit knowledge. The we plotted the effect of market orientation on positive relationship between an SME’s strategic the dependent variable for values of the envi- entrepreneurial orientation and tacit knowledge ronment hostility set at the mean and one stan- has important implications for both theory and dard deviation above and below the mean, as practice. Cultivating tacit knowledge is a skill suggested by Cohen and Cohen (1983). that is related to firm’s creativity and competitive As is shown in Figure 1, when environmen- advantage (Lane and Lubatkin 1998). Leiponen tal hostility is high, the negative relationship (2006) also found that the collective tacit knowl- between market orientation and knowledge edge is associated with radical new service tacitness becomes weakened. Thus, we con- introductions. It seems that taking risks and clude that H5 is supported. being proactive are the result of internally created mindsets often associated with tacit Discussion knowledge. A culture of risk taking and innova- In this paper, we take the view that there are tion is typically institutionalized in the organi- a number of different strategic orientation alter- zation through common stories, experience with natives. According to Hakala (2011) firms may common crises, and the familiarity with norms choose a market or an entrepreneurial orienta- that evolve from common experience. All of tion depending on the industry, sector, or these are key to building tacit knowledge. market. We feel that the relationship between Alternatively, our results suggest that strategic orientation and knowledge may be market-oriented SMEs generally develop more recursive: The extent and type of knowledge explicit knowledge systems since they need to (tacit versus explicit) that the firm accumulates deconstruct customer need patterns and be

274 PÉREZ-LUÑO,JOURNAL OF SAPARITO, SMALL BUSINESS AND GOPALAKRISHNAN MANAGEMENT 13 more responsive to them. However, in the long Finally, we find that environmental hostility run, if they are only focused on explicit cus- has a positive relationship with an SME’s tacit tomer needs, they run the risk of not internally knowledge. This finding makes a contribution developing new knowledge that pushes the to the literature because it shows that hostile frontier of existing knowledge (Levinthal and environments necessitate companies to try to March 1993; March 1991). Additionally, these protect their knowledge by leaving it tacit, customer-oriented innovations improve the making it is less appropriable by competitors. performance of established products along The accumulation of tacit knowledge through dimensions of performance that mainstream shared experience and shared norms provides customers currently value. Many of these inno- a better adaptive mechanism that helps firms vation decisions may also sow the seeds for cope with the resource constraints that hostility failure in the future for market-oriented firms engenders. With regard to moderating effects, since listening to customer gives firms a sense we find that in benign environments, the rela- of comfort which over time makes them more tionship between an SME’s market orientation complacent and less able to take big risks and tacit knowledge becomes more strongly (Christensen 1997). negative than in hostile environments. The The entrepreneurially oriented SMEs, on the implication of this finding is important both to contrary, create new knowledge associated theory and to practice because it shows that with radical innovation and new products that benign environments allow market-oriented could make what the market-oriented firms companies to feel relatively safe, and further have been selling for years obsolescent without allows these firms to have most of their knowl- major modifications. As Immelt, Govindarajan, edge bases be explicit without the fear of being and Trimble (2009) argue, market-oriented outperformed. The benign resource-rich envi- firms may be managing the present while entre- ronments do not pressure these firms to draw preneurial firms were creating the future. Long- as much on their tacit knowledge base. term competitiveness may favor the later rather than the former. Limitations and Future Perhaps firms need to consider their strate- Research Directions gic orientation as a part of a developmental Like all studies, ours has some limitations process where an initial entrepreneurial orien- that provide opportunity for future research. tation in course of time would cause firms to First, care should be taken in generalizing these turn to a more market-oriented focus (Hakala findings to other contexts. We relied on data 2011). As Baker and Sinkula (2009) found, EO from 374 Spanish SMEs in the most innovative could complement MO by instilling an oppor- industries. While these data were ideal for tunistic culture that creates the appropriate testing our hypotheses, future studies are context where firms can increase both the needed to establish if the general tendencies quality and quantity of innovations. that we observed hold up for firms in other Overall, this study’s results suggest that an industries and countries. Some support for this SME’s strategic orientation and the type of is provided by a meta-analysis showing that environment that the SME competes contrib- entrepreneurial orientation–performance rela- ute to the creation and accumulation of tionships are robust across different industries knowledge. The type of knowledge that gets and countries (Rauch et al. 2009). It is therefore accumulated and predominates depends on not unreasonable to assume that the relation- the preferred strategic orientation of the firm. ships between an entrepreneurial orientation It could also be interpreted that the relation- and knowledge are also valid in other contexts. ship between knowledge creation and strate- Second, ideally, we should have controlled for gic orientation is to some extent recursive. additional variables which could influence the Entrepreneurial orientation leads to the cre- relationships in our model. This is related with ation of tacit knowledge systems, and the tacit our third limitation, the R2 of our regression knowledge systems in turn gets further rein- analyses are low. In part, we believe that this is forced by the entrepreneurial orientation and because the study describes a longitudinal gets dissipated by the firm’s market orienta- process of how the various attributes of an tion. Firms may need to be ambidextrous and entrepreneurial or market orientation influence create ways to alternate their predominant ori- an SME’s creation and retention of tacit knowl- entations over time. edge. While this is a longitudinal process, our

14 PEREZ-LUJOURNALNO,~ OF SAPARITO, SMALL BUSINESS AND GOPALAKRISHNAN MANAGEMENT 275 data are cross sectional. Finally, we measured Administrative Science Quarterly 35(1), the firm’s overall entrepreneurial or market ori- 128–152. entation as opposed to specific organization- Covin, J. G., and G. T. Lumpkin (2011). “Entre- level behaviors and actions and how these preneurial Orientation Theory and Research: influenced the SME’s accumulation and reten- Reflections on a Needed Construct,” Entre- tion of tacit knowledge. Future research could preneurship Theory and Practice 35(3), 855– use lagged measures for the accumulation and 872. retention of tacit knowledge and also examine Covin, J. G., and D. P. Slevin (1989). “Strategic more specific behaviors, and actions associated Management of Small Firms in Hostile and with an SME’s entrepreneurial or market orien- Benign Environments,” Strategic Manage- tation create firm-specific tacit knowledge over ment Journal 10(1), 75–87. time. ——— (1991). “A Conceptual Model of Entre- preneurship as Firm Behavior,” Entrepre- References neurship Theory and Practice 16(1), 7– Atuahene-Gima, K., and A. Ko (2001). “An 25. Empirical Investigation of the Effect of Damanpour, F. (1991). “Organizational Innova- Market Orientation and Entrepreneurship tion: A Meta-Analysis of Effects of Deter- Orientation Alignment on Product Innova- minants and Moderators,” Academy of Man- tion,” Organization Science 12(1), 54–74. agement Journal 34(3), 555–590. Audretsch, D., and Z. Acs (1994). “New Firm Dean, T., R. Brown, and C. Bamford (1998). Start-ups, Technology, and Macroeconomic “Differences in Large and Small Firm Fluctuations,” Small Business Economics 6, Responses to Environmental Context: 439–449. Strategic Implications from a Comparative Baker, W. E., and J. M. Sinkula (2009). “The Analysis of Business Formations,” Strategic Complementary Effects of Market Orienta- Management Journal 19, 709–728. tion and Entrepreneurial Orientation on Dess, G. G., and D. Beard (1984). “Dimensions Profitability in Small Business,” Journal of of Organizational Task Environments,” Small Business Management 47(4), 443– Administrative Science Quarterly 29, 52– 464. 73. Barney, J. B. (1991). “Firm Resources and Sus- Galunic, C., and S. Rodan (1998). “Resource tained Competitive Advantage,” Journal of Recombinations in the Firm: Knowledge Management 17(1), 99–120. Structures and the Potential for Schumpe- Burns, T., and G. M. Stalker (1961). The Man- terian Innovation,” Strategic Management agement of Innovation. London: Tavistock Journal 19(12), 1193–1201. Publication. Gatignon, H., and J. M. Xuereb (1997). “Strate- Calantone, R. J., J. B. Schmidt, and C. A. di gic Orientation of the Firm and New Product Benedetto (1997). “New Product Activities Performance,” Journal of Marketing Re- and Performance: The Moderating Role of search 34(1), 77–90. Environmental Hostility,” Journal of Product George, G., and S. A. Zahra (2002). “Being Innovation Management 14(3), 179–189. Entrepreneurial and Being Market-Driven: Cavusgil, S. T., R. J. Calantone, and Y. Zhao Exploring the Interaction Effect of Entrepre- (2003). “Tacit Knowledge Transfer and neurial and Market Orientation on Firm Per- Firm Innovation Capability,” The Journal of formance,” in Babson College. Frontiers in Business and Industrial Marketing 18(1), Entrepreneurship Research. Ed. N. Bygrave. 6–21. Babson Park, MA: Center for Entrepreneurial Christensen, C. M. (1997). The Innovator’s Studies, Babson College, 453–462. Dilemma: When New Technologies Cause Golann, B. (2006). “Achieving Growth and Great Firms to Fail. Harvard Business School Responsiveness: Process Management and Press. Market Orientation in Small Firms,” Journal Cohen, J., and P. Cohen (Eds.) (1983). Applied of Small Business Management 44(3), 369– Multiple Regression/Correlation Analysis for 385. the Behavioral Sciences. Mahwah, NJ: Law- Gold, A. H., A. Malhotra, and A. H. Segars rence Erlbaum Associates, Publishers. (2001). “Knowledge Management: An Orga- Cohen, W. M., and D. A. Levinthal (1990). “New nizational Capabilities Perspective,” Journal Perspective on Learning and Innovation,” of Management Systems 18(1), 185–214.

276 PÉREZ-LUÑO,JOURNAL OF SAPARITO, SMALL BUSINESS AND GOPALAKRISHNAN MANAGEMENT 15 Grant, R. M. (1996). “Toward a Knowledge- Leiponen, A. (2006). “Managing Knowledge for Based Theory of the Firm,” Strategic Man- Innovation: The Case of Business-to- agement Journal 17(Winter Special Issue: Business Services,” Journal of Product Inno- Knowledge and the Firm), 109–122. vation Management 23(3), 238–258. Hakala, H. (2011). “Strategic Orientations in Levinthal, D. A., and J. G. March (1993). “The Management Literature: Three Approaches Myopia of Learning,” Strategic Management to Understanding the Interaction between Journal, 14 (Special Issue: Organizations, Market, Technology, Entrepreneurial and Decision Making and Strategy), 95–112. Learning Orientation,” International Journal Li, Y., Y. Zhao, J. Tan, and Y. Liu (2008). of Management Reviews 13, 199–217. “Moderating Effects of Entrepreneurial Hult, G. T. M., R. F. Hurley, and G. A. Knight Orientation and Market Orientation- (2004). “Innovativeness: Its Antecedents and Performance Linkage: Evidence from Impact on Business Performance,” Indus- Chinese Small Firms,” Journal of Small Busi- trial Marketing Management 33(5), 429– ness Management 46(1), 113–133. 442. Liebeskind, J. P. (1996). “Knowledge, Strategy, Hunt, S. D., and R. M. Morgan (1995). “The and the Theory of the Firm,” Strategic Comparative Advantage Theory of Competi- Management Journal 17(Winter), 93– tion,” Journal of Marketing 59(2), 1–14. 107. Immelt, J. R., V. Govindarajan, and C. Trimble Lindelöf, P., and H. Löfsten (2006). “Environ- (2009). “How GE Is Disrupting Itself,” mental Hostility and Firm Behavior—An Harvard Business Review 87(10), 56–65. Empirical Examination of New Technology- Kang, S. C., S. S. Morris, and S. A. Snell (2007). Based Firms on Science Parks,” Journal “Relational Archetypes, Organizational of Small Business Management 44(3), 386– Learning, and Value creation: Extending the 406. Human Resource Architecture,” Academy of Lumpkin, G. T., and G. G. Dess (1996). “Clas- Management Review 32(1), 236–256. sifying the Entrepreneurial Orientation Con- Khandwalla, P. N. (1977). The Design of Orga- struct and Liking It to Performance,” nizations. New York: McGill University Academy of Management Review 21(1), 135– Press. 172. Klein, G. (1999). Sources of Power: How People Maatooﬁ, R. A., and K. Tajeddini (2011). Make Decisions. Cambridge, MA: The MIT “Effect of Entrepreneurial Orientation and Press. Market Orientation on Innovation: Evidence Kogut, B., and U. Zander (1992). “Knowledge from Auto Parts Manufacturing in Iran,” of the Firm, Combinative Capabilities and Journal of Management Research 11(1), the Replication of Technology,” Organiza- 20–30. tion Science 3, 383–397. March, J. G. (1991). “Exploration and Exploita- Kogut, B., and U. Zander (1993). “Knowledge tion in Organizational Learning,” Organiza- of the Firm and the Evolutionary Theory of tion Science 1 (Special Issue: Organizational the Multinational Corporation,” Journal of Learning: Papers in Honor of (and by) James International Business Studies 24(4), 625– G. March), 71–87. 645. Miller, D. (1983). “The Correlates of Entrepre- Kohli, A. K., B. J. Jaworski, and A. Kumar neurship in Three Types of Firms,” Manage- (1993). “Markor: A Measure of Market Ori- ment Science 29(7), 770–791. entation,” Journal of Marketing Research ——— (2011). “Miller (1983) Revisited: A 30(4), 11–37. Reﬂection on EO Research and Some Sug- Lane, P. J., and M. Lubatkin (1998). “Relative gestions for the Future,” Entrepreneurship Absorptive Capacity and Interorganizational Theory and Practice 35(3), 873–894. Learning,” Strategic Management Journal Miller, D., and P. H. Friesen (1978). “Arche- 19(5), 461–477. types of Strategy Formulation,” Management Lee, H., and B. Choi (2003). “Knowledge Man- Science 24(9), 921–933. agement Enablers, Processes, and Organi- Mintzberg, H. (1973). “Strategy-Making in sational Performance: An Integrative View Three Modes,” California Management and Empirical Examination,” Journal of Review 16(2), 44–53. Management Information Systems 20(1), ——— (1979). The Structuring of Organiza- 179–228. tions. Englewood Cliffs, NJ: Prentice-Hall.

16 PEREZ-LUJOURNALNO,~ OF SAPARITO, SMALL BUSINESS AND GOPALAKRISHNAN MANAGEMENT 277 Nahapiet, J., and S. Ghoshal (1998). “Social Product Development Capability: Testing the Capital, Intellectual Capital, and the Organi- Influence of Transferring and Deploying zational Advantage,” The Academy of Man- Tacit Overseas Knowledge,” Strategic Man- agement Review 23(2), 242–266. agement Journal 22(4), 359–378. Narver, J. C., and S. F. Slater (1990). “The Effect Tolstoy, D. (2009). “Knowledge Combination of a Market Orientation on Business Profit- and Knowledge Creation in Foreign-Market ability,” Journal of Marketing 54(4), 20–42. Network,” Journal of Small Business Man- Nonaka, I. (1994). “A Dynamic Theory of Orga- agement 47(2), 202–220. nizational Knowledge Creation,” Organiza- Turner, K. L., and M. V. Makhija (2006). “The tion Science 5(1), 14–37. Role of Organizational Controls in Managing Nonaka, I., and H. Takeuchi (1995). The Knowl- Knowledge,” The Academy of Management edge Creating Company. Oxford: Oxford Review 31(1), 197–217. University Press. Wang, C. (2008). “Entrepreneurial Orientation, Polanyi, M. (1966). The Tacit Dimension. New Learning Orientation, and Firm Perfor- York: Anchor Day. mance,” Entrepreneurship Theory and Prac- Rauch, A., J. Wiklund, M. Frese, and G. T. tice 33(2), 635–657. Lumpkin (2009). “Entrepreneurial Orienta- Werner, S., L. E. Brouthers, and K. D. Brouthers tion and Business Performance: Cumula- (1996). “International Risk and Perceived tive Empirical Evidence,” Entrepreneurship Environmental Uncertainity: The Dimension- Theory and Practice 33(3), 761–779. ality and Internal Consistency of Miller’s Renko, M., A. Carsrud, and M. Brännback Measure,” Journal of International Business (2009). “The Effect of A Market Orientation, Studies 27, 571–587. Entrepreneurial Orientation, and Techno- Wiklund, J., and D. Shepherd (2005). “Entre- logical Capability on Innovativeness: A preneurial Orientation and Small Business Study of Young Biotechnology Ventures in Performance: A Configurational Approach,” the United States and in Scandinavia,” Journal of Business Venturing 20(1), 71– Journal of Small Business Management 91. 47(3), 331–369. Zahra, S. (1993). “Environment, Corporate Schumpeter, J. A. (1961). The Theory of Entrepreneurship and Financial Perfor- Economic Development: An Inquiry into mance: A Taxonomic Approach,” Journal of Profits, Capital, Credit, Interest, and the Business Venturing 8, 319–340. Business Cycle. New Brunswick, NJ: Trans- Zahra, S. A., and D. M. Garvis (2000). “Interna- action Publishers. tional Corporate Entrepreneurship and Firm Shapiro, S. S., and M. B. Wilk (1965). “An Performance: The Moderating Effect of Inter- Analysis of Variance for Normality,” national Environmental Hostility,” Journal of Biometrika 52, 591–611. Business Venturing 15(5–6), 469–492. Slater, S. F., and J. C. Narver (1993). “Product- Zander, U., and B. Kogut (1995). “Knowledge Market Strategy and Performance: An Analy- and the Speed of the Transfer and Imitation sis of the Miles and Snow Strategy Types,” of Organizational Capabilities: An Empirical European Journal of Marketing 27(10), Test,” Organization Science 6(1), 76– 33–51. 92. ——— (1995). “Market Orientation and the Zhao, Y., Y. Lee, S. Li, and L. Chen (2011). Learning Organization,” Journal of Market- “Entrepreneurial Orientation, Organizational ing 59(3), 63–74. Learning, and Performance: Evidence from Subramaniam, M., and N. Venkatraman (2001). China,” Entrepreneurship Theory and “Determinants of Transnational New Practice 35, 1, 293–317.

278 PÉREZ-LUÑO,JOURNAL OF SAPARITO, SMALL BUSINESS AND GOPALAKRISHNAN MANAGEMENT 17 Journal of Small Business Business Management Management 2016 2015 54(1), ••(••), pp. pp. 279–298 ••–•• doi: 10.1111/jsbm.12145 The Franchise Dilemma: Entrepreneurial Characteristics, Relational Contracting, and Opportunism in Hybrid Governance by Heiner Evanschitzky, Barbara Caemmerer, and Christof Backhaus

Using franchise data, we identify that entrepreneurial characteristics of franchisees partially impact on their opportunistic tendencies. Further, relational contracting increases franchisee opportunism by strengthening the opportunism-enhancing impact of entrepreneurial characteristics. These ﬁndings point to a key dilemma franchisors need to be aware of: Entrepreneurially minded franchisees who might be better at exploiting market opportunities for their units may also behave more opportunistically, if given the chance through a more relational contracting regime. At the same time, if they perceive the contractual framework as being too rigid, they may be less able to leverage their capabilities, become dissatisﬁed, and exit the system.

importance and research attention over the Introduction last three decades (e.g., Blut et al. 2011; The choice of organizational governance Castrogiovanni, Combs, and Justis 2006; Hunt form is a key strategic management decision as 1977; Kaufmann and Dant 1999; Kaufmann it has long-term implications on firm perfor- and Rangan 1990; Kidwell and Nygaard 2011; mance (Mitsuhashi, Shane, and Sine 2008; Nair, Tikoo, and Liu 2009; Tracey and Jarvis Sorenson and Sorensen 2001). A leading per- 2007; Windsperger and Dant 2006). Franchis- spective in governance form choice is transac- ing is particularly common in sectors where tion cost economics (TCE), which holds that organizations need to operate through dif- organizations will handle transactions to mini- fused distribution networks with outlets that mize the costs involved in carrying them out are located close to customers and managed (Williamson 1985, 1991, 2010). Increasingly, by individuals with local knowledge (Berg organizations turn to hybrid forms of gover- and Friedman 1980; Carney and Gedajlovic nance to have greater control than in market 1991). exchange and to avoid disadvantages associ- The benefits of franchising in comparison ated with vertical integration (Norton 1988; with other channel relationships are rooted Powell 1990; Yin and Zajac 2004). in the unique combination of two functional In particular, franchising as a hybrid gov- principles: On the one hand, integrating a ernance model has gained increased economic network of outlets under one brand and unified

Heiner Evanschitzky is Professor of Marketing at Aston University, Birmingham, UK. Barbara Caemmerer is Professor of Marketing at ESSCA School of Management, Boulogne-Billancourt, France. Christof Backhaus is Professor of Marketing at Newcastle University Business School, UK. Address correspondence to: Heiner Evanschitzky, University of Aston, Marketing, Birmingham B4 7ET, UK. E-mail: [email protected].

EVANSCHITZKY, CAEMMERER, AND BACKHAUS BACKHAUS 279 1 operational policies allows for capitalizing on relational forms of contracting (Carson, economies of scale (Shane 1996). On the other Madhok, and Wu 2006). Relational contracting hand, successful franchising depends on incorporates the use of “unwritten codes of autonomously acting franchisees able to adapt conduct” (Davies et al. 2011, p. 325) to safe- to local market conditions and capitalize on the guard against negative relationship outcomes benefits of entrepreneurial engagement at a (Cochet, Dormann, and Ehrmann 2008). In this microsystem level (Sorenson and Sorensen type of contracting implicit, “relational norms 2001). Certainly, other hybrid forms of organi- that govern acceptable behavior between zation such as cooperative purchasing groups exchange partners” (Lusch and Brown 1996, p. or licensing agreements might possess similar 19) are developed over time and are based on advantages; yet franchising is probably the social consensus or mutual understanding. most visible hybrid form of governance and Thus, relational contracting is long-term ori- hence the focus of this paper. ented, reciprocal, and extends beyond mere Given the apparent tension between the two buying and selling (Li and Dant 1997) as it aforementioned functional principles, franchise forms the basis for the development of shared, relationships are fragile and in need of balanc- long-term relational norms and behaviors ing standardization and adaptation (Sorenson between partners. Within a given franchise and Sorensen 2001) as well as autonomy and system, franchisors generally exercise the same dependence (Dant and Gundlach 1999). To find level of relational contracting. Differences in that balance, franchising crucially depends on levels of relational contracting are based more both the proper selection (ex-ante) as well on perceptions of franchisees than on franchi- as the management of franchisees (ex-post; sor conduct. Jambulingam and Nevin 1999; Michael and The key relational norms and behaviors that Combs 2008). have been widely acknowledged in the litera- However, two key assumptions of TCE, ture are flexibility, information exchange, and bounded rationality and opportunism, advocate collegiate problem solving (Heide and John a focus on the contracting of partners. As a 1992; Kaufmann and Stern 1988; Poppo and consequence, research has in the main focused Zenger 2002). As these norms develop in fran- on appropriate contracting so as to minimize chising contexts between franchisees and their levels of opportunism and related transaction franchisor over time, relational contracting can costs (e.g., Mitsuhashi, Shane, and Sine 2008; be regarded as a system-level behavioral vari- Sorenson and Sorensen 2001; Yin and Zajac able, which is likely to differ across systems, 2004). rather than a channel design variable (Hoetker Though the limitations of formal contracting and Mellewigt 2009). As such, in relational con- have been well established in the literature tracting, social forces may create barriers to (e.g., Aulakh and Gençtürk 2007; Poppo and opportunistic behavior and foster cooperation Zenger 2002), less is known about the condi- (Carson, Madhok, and Wu 2006). For these tions under which relational contracting is an reasons, it has been argued that relational con- effective way of safeguarding against negative tracting may create economic value (Brown, outcomes such as opportunism. This is impor- Dev, and Lee 2000). tant, though, as there is evidence that this form It has been acknowledged that the two of contracting is being increasingly used in described types of contracting, formal and rela- franchise systems to confront problems with tional, often coexist in organizational environ- agents (Cochet, Dormann, and Ehrmann 2008). ments, that is, they are not mutually exclusive Neoclassical contract theory suggests that (Lusch and Brown 1996). The extent to which formal and more extensive contracting efforts one or the other is used depends on the situa- help to safeguard against opportunism (Luo tion and the assessment as to what type of 2002). The reason for this is that formal contracts contracting is more efficient, that is, which one explicitly state how members should behave, can better safeguard against opportunism in a which leaves little room for interpretation of the given business context (Carson, Madhok, and rules outlined in the contract (Lusch and Brown Wu 2006). 1996). As a result, sanctioning against miscon- Although some studies suggest that rela- duct is relatively straightforward. tional contracting is efficient (Birkeland 2002; In contrast, sociological approaches to Lado, Dant, and Tekleab 2008), there is also contracting emphasize to a greater extent of evidence that its efficiency is dependent on

2280 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT organizational governance form and, in Here exactly lies the franchising dilemma particular, the level of ambiguity pertinent which is the focus of our study: On the one within organizations. Ambiguity (e.g., Sillince, hand, entrepreneurial characteristics and moti- Jarzabkowski, and Shaw 2010) refers to the vations are desirable as they are related to the possibility that an interpretation of a situation, identification and exploitation of market oppor- information, or role is doubtful or uncertain. tunities for the system. To fully leverage the The pertinence of ambiguity in principal–agent entrepreneurial capital franchisees bring into relationships, as can be found in hybrid gover- the system, franchisors might need to grant a nance forms such as franchising, has been well certain level of autonomy (Kaufmann and acknowledged in the literature (Kashyap, Antia, Eroglu 1999; Schul, Little, and Pride 1985; and Frazier 2012). In the main, ambiguity can Strutton, Pelton, and Lumpkin 1995). On the be observed in the form of franchisee role other hand, there is a reason to suggest that ambiguity, for example, in relation to what entrepreneurial characteristics in combination functions or responsibilities the franchisee with greater autonomy may be related to higher needs to fulfill in a franchise system. levels of opportunism and thus increased trans- Agency theory highlights the issues that action costs for the franchisor (e.g., Cochet, arise with the principal (franchisor) employing Dormann, and Ehrmann 2008). an agent (franchisee). According to agency As little research exists concerning the theory, complete formal contracts can be for- dilemma we described, we will first investigate mulated in order to control agent actions. how key entrepreneurial characteristics— However, TCE assumes that such contracts are innovativeness, risk taking, and proactiveness necessarily incomplete (e.g., Williamson 1975, (Covin and Slevin 1986, 1988; Miller 1983)— 1985), which in turn gives room for interpreta- influence opportunistic tendencies of franchi- tion, for example, in relation to the functions sees. Thus, in this study, we consider these and responsibilities of the franchisee. This, by entrepreneurial characteristics at an individual definition, means that ambiguity is higher in franchisee level. franchising contexts with principal and agent Secondly, in order to understand the extent relationships than in hierarchy, as company- to which relational contracting, a franchise owned units are under stronger, explicit control system-level variable, may or may not safe- and have fewer possibilities for own decision- guard franchisors against agency problems in making (Yin and Zajac 2004). these industries, we assess its moderating role Further, based on the extant literature, there on the link between franchisees’ propensity to is also a reason to suggest that agent motivation favor innovation, take risks, be proactive, can impact on the safeguarding effects of rela- and be opportunistic in franchisee–franchisor tional contracting in franchising: From an dyads. As just mentioned, this is particularly agency theory perspective (Eisenhardt 1989; important as there is evidence that relational Gillis et al. 2011), franchisors (principals) have contracting is increasingly adopted as a man- to consider the motivations of the individual agement strategy in franchising systems franchisee (agent) before determining contrac- (Cochet, Dormann, and Ehrmann 2008). tual arrangements (Michael and Combs 2008). Addressing this research agenda, we add to However, little research has been carried out to the extant literature by highlighting some investigate the combined effects of contracting important issues that need to be considered by and agent motivation on opportunism. franchisors ex-ante in the selection, as well as Franchisees are frequently referred to as ex-post in the management of franchisees. legally independent, small-scale entrepreneurs (Cochet and Garg 2008). Though there is some Theoretical Background debate as to whether franchisees actually are and Hypotheses entrepreneurs (Ketchen, Short, and Combs The Franchisee Entrepreneur 2011) and possess entrepreneurial characteris- Performance Dilemma tics (Birkeland 2002), there is growing support One of the main arguments for the adoption for the notion (Cochet, Dormann, and Ehrmann of franchising as a hybrid governance form 2008; Kaufmann and Dant 1999), whether they advanced in the literature is the problem of are in single-unit or sequential multiunit agree- agent motivation (Price 1997). Due to the ments (Grünhagen and Mittelstaedt 2005; geographically dispersed nature of franchise Ketchen, Short, and Combs 2011). systems, it is difficult to oversee the actions of

EVANSCHITZKY, CAEMMERER, AND AND BACKHAUS BACKHAUS 281 3 outlet managers. Thus, in order to reduce trans- tically (George and Zhou 2001; Runco 2004). action costs associated with safeguarding Risk taking has been defined as a “tendency to against negative business outcomes, franchis- take bold actions such as venturing into ing is viewed as a favorable option. The com- unknown new markets, committing a large monly held view is that franchisees as owners portion of resources to ventures with uncertain of their outlets are more motivated to act in the outcomes, and/or borrowing heavily” (Lumpkin interest of the business than employed manag- and Dess 2001, p. 431). If someone is more ers (Castrogiovanni, Combs, and Justis 2006; prone to risk taking, then they are more likely to Michael and Combs 2008). take a gamble that they are not going to get Contrary to the opinion of some, franchising caught when disregarding rules or norms. Thus, is often described as an entrepreneurial activity the risk of acting opportunistically increases (Cochet and Garg 2008). Birkeland (2002) (Jambulingam and Nevin 1999). Proactiveness highlights that franchise systems do not represents an “opportunity-seeking, forward- hamper entrepreneurial motivations of their looking perspective involving introducing franchisees and that most franchise operations new products or services ahead of the competi- are small and run like family businesses. In tion and acting in anticipation of future line with this, many authors suggest that fran- demand to create change and shape the envi- chisees are entrepreneurs and possess entre- ronment” Lumpkin and Dess 2001, p. 431). As preneurial characteristics (Grünhagen and proactiveness deals with competitiveness and Mittelstaedt 2005; Kaufmann and Dant 1999). getting ahead of competition, opportunistic Similarly, Birkeland (2002) states that franchi- behavior is more likely as “cutting corners” sees think of themselves as entrepreneurs as seems acceptable (Kickul and Gundry 2002). they own their franchise unit. Given that the mentioned characteristics are Specifically, in a recent paper, Ketchen, not necessarily strongly related to each other Short, and Combs (2011) provide an in-depth (Lumpkin and Dess 2001), we look at each of discussion of the question of whether them separately instead of integrating them franchisees can be regarded as entrepreneurs. into a higher order construct, “entrepreneurial The authors provide compelling empirical orientation” (e.g., Covin and Wales 2012; evidence that franchisees possess entrepre- Wiklund 1999). This is in response to the sug- neurial motivations (Ketchen, Short, and gestion from the literature stating that future Combs 2011) which are linked to positive busi- research in this domain may benefit from such ness outcomes. In sum, the idea that franchi- an approach, instead of looking at entrepre- sees’ entrepreneurial capacities are desirable neurial orientation as an aggregate concept and may contribute to system success is rela- (Kreiser, Marino, and Weaver 2002; Lumpkin tively uncontested. and Dess 1996). The current state of research provides strong Literature investigating consequences of support for three key entrepreneurial charac- entrepreneurial motivations has linked the con- teristics: innovativeness, risk taking, and struct to performance outcomes (Li, Huang, and proactiveness (Caliendo, Fossen, and Kritikos Tsai 2009; Wiklund 1999). Consequently, also 2009; Carland et al. 1984; Covin and Slevin in the context of franchising, one could assume 1986, 1988; Delmar and Davidsson 2000; that entrepreneurial characteristics of franchi- George and Marino 2011; Kaufmann and Dant sees drive their individual franchisee perfor- 1999; Korunka et al. 2003; Miller 1983). These mance. However, there are also arguments for have been defined by Lumpkin and Dess (2001, the notion that entrepreneurial characteristics p. 431) as follows: of franchisees may, in fact, have a negative Innovativeness refers to “a willingness to impact on franchise system performance: All support creativity and experimentation in intro- three entrepreneurial characteristics under ducing new products/services, and novelty, investigation—namely, innovativeness, busi- technological leadership and R&D in developing ness risk taking, and proactiveness (Morris and new processes.” Though innovativeness is posi- Sexton 1996; Wiklund and Shepherd 2005)— tively linked to openness to new experience, its are related to lower levels of agreeableness in downside is its negative association with agree- work settings, higher tolerance to friction, and ableness and conscientiousness. It is more likely a stronger urge for autonomy (Birkeland 2002; that such individuals engage in “breaking rules” Caliendo, Fossen, and Kritikos 2009; Eysenck and have a greater propensity to act opportunis- 1976; George and Zhou 2001; Jambulingam and

4282 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Nevin 1999; Kickul and Gundry 2002; Lumpkin, Thus, from a franchisor perspective, a Cogliser, and Schneider 2008; Runco 2004). dilemma evolves in which the trade-off Concerning innovativeness, it can be argued between potential benefits and hazards of fran- that this characteristic should enhance the desire chisees’ entrepreneurial characteristics needs to for work settings that allow for personal fulfill- be considered. Given the managerial impor- ment and the freedom to move forward with tance of these considerations, as they concern one’s own ideas without any constraints ex-ante selection as well as ex-post manage- imposed by the franchisor. As an entrepreneur- ment practices, it is important to assess empiri- ial characteristic, innovativeness should thus cally the link between individual franchisees’ correspond with an increased interest to imple- entrepreneurial characteristics and their levels ment own ideas in order to positively influence of opportunism. Based on the previously dis- own store performance (Combs, Michael, and cussed literature, we hypothesize: Castrogiovanni 2009; Kaufmann and Eroglu 1999). Given that, at least to some extent, fran- H1a: Franchisees’ levels of innovativeness are chise systems naturally restrict personal initia- positively related to levels of opportunism. tives and freedom to realize innovative ideas, it can be assumed that innovativeness does not H1b: Franchisees’ levels of proactiveness are necessarily correspond with cooperative behav- positively related to levels of opportunism. ior but may instead foster opportunistic tendencies. Franchisees with higher levels of risk H1c: Franchisees’ levels of risk taking are posi- taking are more prone to “challenge the way tively related to levels of opportunism. business is conducted” (Jambulingam and Nevin 1999) by their franchisor. In such cases, there The Moderating Effects of may be a greater propensity to engage in diver- Relational Contracting gent activities in order to enhance the perfor- The few studies that have explored the link mance of own units. This behavior could be between franchisees’ characteristics and fran- further encouraged by the fact that high risk- chise system outcomes have not taken contex- taking franchisees feel less threatened by the tual moderators, such as governance form or risk that their deviance may be uncovered. type of contracting, into account (Jambulingam Proactiveness, similarly, may increase opportu- and Nevin 1999). However, research suggests nistic tendencies as this entrepreneurial charac- that the link between personal characteristics teristic is related to searching for opportunities and behavior is moderated by a person’s beliefs to transform and develop one’s own business as to whether it is easy or difficult to perform (Kickul and Gundry 2002). The interest of the an act (Ajzen 1991; Krueger and Carsrud 1993). entire system may not necessarily be at the These perceptions of whether it is easy or dif- forefront of business activity considerations of ficult to behave in a particular way depend on franchisees that are highly proactive. past experiences as well as current obstacles to, On this basis, individuals who possess high or opportunities for, the performance of a levels of innovativeness, risk taking, and behavior. proactiveness may be successful entrepreneurs The mentioned observations provide a par- but may simultaneously be less suited for fran- allel to TCE and agency theory which both chise system environments. In particular, the highlight the need for contracting, as otherwise bilateral dependency on which such hybrid partners or agents will behave in their own forms of governance are built (David and Han interest (Eisenhardt 1989; Williamson 1975). 2004; Williamson 1991) is in opposition to the Applied to the franchise context, we argue that very nature of these entrepreneurial character- ex-post contracting moderates the link between istics. Therefore, we propose that such indi- the previously discussed entrepreneurial char- viduals are more likely to cope with internal acteristics of franchisees and their levels of conflicts and therefore pursue their own per- opportunism. sonal objectives in an opportunistic manner, Research in TCE and agency theory has tra- which can negatively impact the overall success ditionally focused on organizational manage- of the system (Cochet, Dormann, and Ehrmann ment through formal contracting. This is 2008; Mitsuhashi, Shane, and Sine 2008; because in the presence of opportunism and Sorenson and Sorensen 2001; Yin and Zajac self-interest, organizations have to deal with 2004). goal divergence between contracting parties. It

EVANSCHITZKY, CAEMMERER, AND AND BACKHAUS BACKHAUS 283 5 is important to note, though, that TCE and tional contracting may foster autonomy and agency theory have differing underlying thus entrepreneurial initiative. Thus, it might assumptions: TCE assumes bounded rationality, aid in the realization of decentralized decision- suggesting that organizations possess only making, whereas a formal contracting approach limited foresight and cannot protect themselves could be perceived as providing too narrow a against all eventualities, whereas agency theory framework for franchisees. suggests that all contingencies can be specified However, herein may also exactly lie a limi- ex-ante (David and Han 2004; Eisenhardt 1989; tation of relational contracting. Referring back Kim and Mahoney 2005; Mitsuhashi, Shane, and to the franchisee–entrepreneur performance Sine 2008). dilemma, we point to evidence which suggests Understanding the efficiency of contracts that relational contracting may not be robust and different contracting mechanisms (i.e., the across all governance forms. This may be par- extent to which potential opportunistic behav- ticularly the case in governance forms that are ior can be suppressed) is particularly important higher in ambiguity than hierarchy, such as in franchise system contexts due to the nature franchising (Kashyap, Antia, and Frazier 2012). of franchising as a business model. Features In such contexts, where the principal needs to such as a dominant entrepreneurial ideology clearly establish the functions and responsibili- and geographical dispersion lead to opportunities of the agent, relational contracting may be ties for franchisees to operate units in a self- less likely to safeguard against opportunism interested manner that may damage the system (Carson, Madhok, and Wu 2006). The reason is in the long run (Birkeland 2002). It is for this that relational contracting further increases reason that formal contracting as a control ambiguity due to the fact that it gives franchi- mechanism has been widely discussed in the sees more autonomy within a framework of franchising literature. rules that are not explicitly stated, as they A different stream of the literature has would be in formal contracting. Therefore, rela- advanced the relational contracting approach tional contracting through its ambiguity- (Koza and Dant 2007; Lado, Dant, and Tekleab enhancing properties may moderate the links 2008; Macneil 1980), a form of contracting between entrepreneurial characteristics and which is increasingly used in practice (Cochet, opportunistic behavior.1 Dormann, and Ehrmann 2008). In contrast to Autonomy is “the extent to which a party, TCE and agency theory, relationalism embeds here a franchisee, is unconstrained to indepen- exchange into a social context. Some of the key dently make decisions and to take action” exchange characteristics between partners in (Cochet, Dormann, and Ehrmann 2008, p. 52). relational contracting involve flexibility, infor- In contrast to formal contracting, relational con- mation exchange, and collegiate problem tracting relies on a set of unwritten rules (e.g., solving (Heide and John 1992; Kaufmann and Lado, Dant, and Tekleab 2008). This means that Stern 1988). in comparison with formal contracting, under Proponents of relational contracting suggest relational contracting, there is greater ambigu- that it creates economic value through the ity, a term which refers to the possibility that development of unwritten shared norms, various interpretations of a situation, informa- values, and behaviors that safeguard against tion, or role are possible which makes them conflict (Brown, Dev, and Lee 2000). A further thus doubtful or uncertain. The reason for this advantage may be that this approach is more is that with unwritten rules, standards are more flexible than formal contracting and can thus open to interpretation, more difficult to be rein- be adjusted for the management of different forced (Carson, Madhok, and Wu 2006), and franchisees that come from a wide range of give thus more leeway in decision-making, backgrounds and experiences, and as such which in turn means greater autonomy. have varying aspirations and expectations for Though relational contracting mechanisms, their units (Birkeland 2002). Moreover, rela- such as information exchange, may be rein-

1It is worth noting that it is beyond the scope of this paper to assess in more detail the mediating role of ambiguity; here, we are concerned with the moderating effect of relational contracting. However, future research should investigate the role of ambiguity to gain further insights into the mechanisms of how relational contracting impacts on the entrepreneurial characteristics–opportunism link.

6284 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT forced to ensure franchisee collaboration, these opportunism in that it strengthens the posi- mechanisms are by definition less formalized tive link between the two constructs. than those deployed under formal contracting and thus provide greater room for autonomy. H2c: Relational contracting moderates the link Based on the previous discussion, it can be between franchisees’ risk taking and oppor- argued that the link between ambiguity and tunism in that it strengthens the positive link opportunism has been relatively well estab- between the two constructs. lished. However, though it has been suggested that relational contracting cannot perfectly Conceptual Model enforce cooperative behavior and may thus In summary, the mentioned arguments only provide a partial safeguard to agency costs advanced leading to H1a–c suggest that oppor- ex-post (such as costs incurred through opportunism (OPP) is increased at the individual tunistic behavior) (e.g., Jap and Ganesan 2000), franchisee level (level 1) by three entrepreneur- as sanctioning is more difficult, if not impos- ial characteristics. More precisely, we assume sible, only very few studies have in fact inves- that innovativeness (INNO), proactiveness tigated to what extent the level of relational (PRO), and business risk taking (RISK) are posi- contracting may moderate the link between tively related to opportunism. In addition to the franchisee characteristics (here entrepreneurial three entrepreneurial characteristics, we characteristics) and franchisee behavior (here control for the effect of the time a franchisee opportunistic behavior). It is for this reason has been with the system (FR_AGE)asithas that our study focuses on the constructs rela- been suggested that over time, more agency tional contracting, franchisee characteristics, conflicts may evolve in a relationship (Cochet, and opportunism. This is particularly important Dormann, and Ehrmann 2008). as franchisees, who possess high levels of Turning to system-level effects (level 2), entrepreneurial motivations, might more and in line with H2a–c, we assume moderating actively engage in opportunistic behaviors effects on the franchise system level (level 2) under stronger relational contracting as it is such that relational contracting (REC) difficult for the franchisor to observe, control, strengthens the opportunism-enhancing impact or sanction such behaviors. Especially, we of the three entrepreneurial characteristics— argue that innovativeness, proactiveness, and innovativeness, proactiveness, and business risk risk taking help to explain franchisees’ oppor- taking. In addition to investigating the moderat- tunistic behaviors in case of highly relational ing impact of relational contracting as our key governance regimes. construct of interest and in order to avoid model If a franchisee possesses high levels of these misspecification, we control for the age of the three characteristics, they might show a higher franchise system (AGE_SYS) and the number of probability of engaging in self-interest-seeking franchisees (NO_FR). The rational is that fran- behaviors which—from the perspective of the chise system age and the number of franchisees franchise system as a whole—are not benefi- are proxies for franchisor experience which cial, if they are given the opportunity to do so could have an impact on the average opportun- without being observed or sanctioned. On the ism level in a given system (Dant and Gundlach other hand, franchisees with lower levels of the 1999; Lilis, Narayana, and Gilman 1976). We also three entrepreneurial characteristics will possi- include average success of the system (SUCC)as bly be less inclined to opportunistically exploit there is evidence that performance plays a pre- their freedom. Therefore, we suggest that rela- dominant role in determining partner opportun- tional contracting strengthens the opportunism- ism (Deeds and Hill 1999). enhancing effects of innovativeness, proactive- A simplified conceptual model is illustrated ness, and risk taking. Hence, in Figure 1. As we have matched data, namely H2a: Relational contracting moderates the link franchisee-level data and franchise system-level between franchisees’ innovativeness and data, whereby the franchisee data are nested in opportunism in that it strengthens the posi- the franchise system data, we use hierarchical tive link between the two constructs. linear modeling (HLM) for the analysis (Hox 1995; Muthen and Satorra 1995; Raudenbush H2b: Relational contracting moderates the link and Bryk 2002; Snijders and Bosker 1999). between franchisees’ proactiveness and HLM is well suited for our research objectives

EVANSCHITZKY, CAEMMERER, AND AND BACKHAUS BACKHAUS 285 7 Figure 1 German Franchise Association and that have Conceptual Model had at least ﬁve franchisees for the last two years. We asked the franchisors for their support by providing a list of their franchisees Ex-Ante Ex-Post for us to survey. Twenty-two systems offered

Relational Contracting their support. The ﬁnal sampling list contains a • Information Exchange total of 2,530 franchisees across the 22 systems. • Flexibility • Problem Solving It is also worth mentioning that all cases from

Entrepreneurial Characteristics the franchisee data can be matched to a par- • Innovativeness Franchisee ticular franchise system, enabling a multilevel • Risk Taking Opportunism • Proactiveness analysis.

Franchise System Data for our study were collected through Characteristics and Franchisee Outcomes standardized questionnaires. Three hundred Practices twenty-one usable questionnaires were returned which equates a response rate of 12.3 percent. With regard to system size, the majority (64 percent) of systems consist of a maximum of 100 as it allows for a simultaneous investigation of franchisees, 36 percent are between 101, and level 1 drivers of opportunism and level 2 mod- 500 franchisees (mean = 115) with a system- erators in a single model. The ﬁnal model looks wide turnover of 187 Mio Euros (standard devia- as follows: tion [S.D.] = 57 Mio). On average, franchisees Level 1 opportunism model belong to systems that have been on the market for 25 years (S.D. = 9.06). Moreover, the major-

OPP=+ββ01()()() INNO + β 2 RISK + β 3 PRO + r ity of franchisees (67.3 percent) are from the (1) service and retail sector and can be considered small business operations. Table 1 provides a detailed breakdown of the sample.2 Level 2 opportunism model Apart from the primary data collected from (1) Intercept effects the franchisees, we also gathered secondary data on the franchise systems using the fran- βγ=+ γ + γ 00001(_)(_)AGE SYS 02 NO FR chise systems’ websites, objective information −+γ (2) 03()SUCC u 0 provided by the franchisor, various franchise handbooks, annual reports, reports from the (2) Slope effects national franchise associations, and other publicly available sources. βγ=+ γ + 11011()REC u 1 (3) Measures Level 1 Measures. In the survey, we measure each franchisee’s entrepreneurial characteris- βγ=+ γ + 22021()REC u 2 (4) tics (innovativeness [INNO], proactiveness [PRO], and business risk taking [RISK]) (Covin and Slevin 1986, 1988; Jambulingam and Nevin βγ=+ γ()REC + u (5) 33031 3 1999; Lumpkin, Cogliser, and Schneider 2008; Miller 1983). The analysis suggests that all mea- Methods sures are valid and reliable. Particularly, dis- Data Collection criminant validity is given as correlations are We ﬁrst compiled a list of 153 franchise generally rather low and none of the squared systems operating in Germany. This list only correlations between pairs of constructs exceed contains systems that are organized in the the average variance extracted (AVE) of either

2Though not claiming representativeness, our sample seems to be a good representation of the overall franchise market in Germany: 52.3 percent services industries (41.2 percent in Germany); 26.8 percent retailing (22.5 percent); 7.5 percent hotel and fast-food industries (13.5 percent); 7.8 percent handcraft and construction services (7.3 percent); 0 percent ﬁtness and health industries (5.4 percent); 5.6 percent tourism industries (4.1 percent); and 0 percent other industries (6 percent).

8286 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Table 1 than 0.7, and the mean Rwg(j) is larger than 0.8 Sample Characteristics (Bliese 2000). We thereby calculate a system- wide measure of the degree of relational contracting (REC; CR = 0.897; AVE = 721), which is Sector Number of Percentage assessed using the key dimensions flexibility, Franchisees information exchange, and collegiate problem in Sample solving (Brown, Dev, and Lee 2000; Heide and John 1992; Jambulingam and Nevin 1999; DIY/Home 69 21.4 Kaufmann and Stern 1988). Moreover, we Improvement measure perceptions of success of the franchise Telecommunication/ 55 17.1 system (SUCC; CR = 901; AVE = 0.698). Dis- IT criminant validity is not an issue in our level 2 Computer 51 15.9 data as the variables are not significantly corre- Hard/Software lated. As before, we create scale means for all (Specialty) Food 24 7.5 level 2 variables. Home Electronics 17 5.4 Insurance/Finance 62 19.3 Covariables. At level 1, we include the age Handcraft/Small 25 7.8 of the franchise system and the number of Trade franchisees as measures of franchisor experi- Travel/Tourism 18 5.6 ence. Literature finds support for the notion Sum 321 100.0 that more experienced franchisors should be more advanced in terms of franchisee selection, incentive design, and monitoring (Stump and Heide 1996). We also include system construct (Fornell and Larcker 1981). We also performance corresponding to the study of measure opportunism (OPP; critical ratio Deeds and Hill (1999), who found that alli- [CR] = 0.901; AVE = 0.787) with seven items ance performance plays a predominant role in suggested by Jambulingam and Nevin (1999).3 determining partner opportunism. The theo- We also assess common method bias as level retical rationale for this effect is that under- 1 items are from the same source. Using the performing partnerships usually go along procedure suggested by Podsakoff et al. (2003), with disappointment and friction in the rela- we note that common method variance does tionship. Thereby, blame for failure is regu- not seem to be a severe issue in our data.4 larly allocated to the franchisor. Due to the We calculate the scale means of the three high level of dependency of the franchisee, independent variables as well as the dependent the probability of behaving opportunistically variable for our HLM analysis. Apart from these increases. level 1 measures, we also assess the years a Table 2 provides key psychometric proper- franchisee has been a part of the system ties of the level 1 and level 2 measures. For a (AGE_FR). complete overview of all the items used in the survey, please refer to the Appendix. Level 2 Measures. On the franchise system level, we use objective company data to assess Analysis Overview the age of the franchise systems (AGE_SYS) and As suggested by Raudenbush and Bryk the number of franchisees in the system (2002), we employ HLM to analyze our data as (NO_FR) as a measure for system size. Moreover, we have two levels of data, namely, franchisee we use the relational contracting measures from level data and franchise system level data, the franchisee survey to aggregate them to the whereby the franchisee data is nested in the system level (as suggested by Kozlowski and franchise system data. Hence, ordinary regres- Klein 2000). It is deemed suitable to aggregate sion analysis would not be appropriate as such level 1 data to level 2 as the intraclass correlation data violates the assumption of independence (ICC) (1) is larger than 0.2, the ICC (2) is larger of observations (error term correlation), a key

3To further assess the quality of these measures, we run EFA and CFA and ﬁnd no cross-loadings, but a good overall ﬁt of the measurement model (chi-squared/df = 1.976; AGFI = 0.932; TLI = 0.913; RMSEA = 0.042). 4Details of these analyses are available upon request.

EVANSCHITZKY, CAEMMERER, AND AND BACKHAUS BACKHAUS 287 9 Table 2 suggest that whenever DEFF is larger than 2, Psychometric Propertiesa there is sufficient reason to assume a severe violation of the assumption of independent observations across all classes and propose that Level 1 Measures in such a case, multilevel analysis should be employed. Our results show that the design 1. 2. 3. 4. effect for opportunism is DEFF = 8.44. There- fore, HLM is an appropriate technique to 1. Opportunism 1 analyze our data. 2. Innovativeness −0.078 1 3. Proactiveness 0.010 0.088 1 Results 4. Risk Taking 0.618 −0.101 −0.006 1 Based on our conceptual model, we CR 0.978 0.819 0.833 0.814 estimate a random intercept/random slope AVE 0.867 0.548 0.626 0.525 model. Table 3 provides estimations of the unstandardized regression coefficients for all level 1 and level 2 effects. Level 2 Measures We note that the three independent variables are very weakly correlated (Table 2), and 1. 2. 3. when including all in a regression analysis, variance inflation factors are between 1.009 1. Information 1 and 1.308, sufficiently below the threshold Exchange suggested in the literature (Hair et al. 2006). 2. Flexibility 0.363** 1 Hence, multicollinearity is not an issue in 3. Problem 0.176** 0.363** 1 our data. The results of our HLM model show Solving that the independent variables, except CR 0.766 0.868 0.833 innovativeness, significantly predict opportun- AVE 0.526 0.688 0.555 ism. More precisely, we find that a one-unit change in risk taking leads to an increase in opportunism of 0.676 points (on the five-point a AVE, average variance extracted; CR, critical opportunism scale); similarly, a one-unit ratio. increase in proactiveness results in a 0.180- **: Coefficient is significant at .01-level. point increase in opportunism. We also note that the covariable franchisee age is negatively related to opportunism such premise of regression analysis (Hox 1995; that a one-unit increase in age leads to a 0.591 − Raudenbush et al. 2004; Snijders and Bosker *103 decrease in opportunism. 1999). Clearly, our franchisee data points are Turning to the level 2 intercept effects of the not independent but nested in higher-level three covariables, we find that age and success groups, in our case the franchise system an of the franchise do not seem to be related to the individual franchisee belongs to. average opportunism level in the system. We Along suggestions by Raudenbush and Bryk further note that the size of a franchise system (2002) as well as Muthen and Satorra (1995), (measured by the number of franchisees) is we first of all assess if there is sufficient varia- significantly, systematically, and positively tion between the 22 franchise systems with related to average opportunism levels in a par- respect to opportunism, our dependent vari- ticular franchise system, possibly suggesting able. In order to do so, we calculate the design that larger systems are more prone to opportu- effect (DEFF) as an average class size adjusted nistic behavior of their franchisees. ICC as follows: As for the key level 2 slope effects, we find some support for the assumption that rela- DEFF=+11()([ c − ×σσ22 + σ 2 ]),(6) tional contracting increases the positive (oppor- BB W tunism increasing) effects of innovativeness, proactiveness, and business risk taking on 2 with c being the average class size, σ B the opportunism. However, only the moderating 2 between-class correlation, and σW the within- effect on the link between business risk taking class correlation. Muthen and Satorra (1995) and opportunism is significant at the 0.05 level,

10288 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Table 3 HLM Results for Opportunism

Predictor Unstandardized t-Ratio Signiﬁcance Coefﬁcient Level

Level 1 (Dependent Variable OPP) Innovativeness 0.032 0.435 p > .1 Proactiveness 0.180 2.449 p < .01 Risk Taking 0.676 4.460 p < .001 Franchisee Age −0.000591 −4.917 p < .001 Level 2 Intercept Effects Age of Franchise System 0.006 0.0166 p > .1 Number of Franchisees in the System 0.158 2.930 p < .01 Average System Success 0.218 0.585 p > .1 Level 2 Slope Effects Innovativeness Relational Contracting 0.241 1.882 p < .1 Proactiveness Relational Contracting 0.027 0.926 p > .1 Risk Taking Relational Contracting 0.465 2.144 p < .05 Franchisee Age Relational Contracting −0.001 −0.708 p > .1

whereas the innovativeness–opportunism link characteristics impact partially on opportunism is only significant at the 0.1 level.5 The (Castrogiovanni, Combs, and Justis 2006). This proactiveness–opportunism link is moderated carries implications for the assessment and in the expected direction, yet the effect is sta- reduction of transaction costs ex-ante, for tistically nonsignificant (t = 0.926; p > .1). example, through social mechanisms such as Taken together, these findings provide only selective franchisee recruitment (Birkeland partial support for H2.6 2002). Specifically, looking at the results of our H1b Discussion and Implications and c, we note that proactiveness and risk Contribution to Theory taking are important antecedents of franchisee Research grounded in TCE focuses on the opportunism (Jambulingam and Nevin 1999). efficiency of governance form and contracting The dimension franchisee innovativeness, to safeguard against opportunism (e.g., however, did not show the positive relationship Mitsuhashi, Shane, and Sine 2008; Sorenson assumed in H1a. Therefore, it needs to be said and Sorensen 2001). Based on agency theoreti- that the relationship between innovativeness cal considerations (Eisenhardt 1989), we argue and opportunism may be less straightforward that also individual agent motivation needs to compared with the other two entrepreneurial be taken into account in order to explain vari- characteristics under investigation here. Given ance in opportunism. This research contributes that franchise systems vary to a large degree in to existing franchising research first of all by terms of innovation management practices, showing that franchisees’ entrepreneurial the existence and direction of the relationship

5We are aware that the normal cutoff value for significance of effects is 0.05. However, considering the sample size at level 2 (n = 22) and the t-value of 1.882, we argue that reporting the effect as nonsignificant might be unadvisable. 6The inclusion of an industry-dummy as a covariable yielded a nonsignificant result for the dummy.

EVANSCHITZKY, CAEMMERER, AND AND BACKHAUS BACKHAUS 289 11 between innovativeness and opportunism individual agent motivation and important out- might depend on the experience franchisees comes. More precisely, our data show that rela- have made when trying to market innovative tional contracting strengthens the link between ideas at the system level. Franchisors who keep risk taking and opportunism, and (to a lesser responsibilities for innovation development in degree) between innovativeness and opportun- their own realm of activities might in fact ism. Put differently, we find that the encounter problems such as intrasystem inertia opportunism-increasing effect of risk taking is (Kaufmann and Eroglu 1999), less cooperation, further increased by relational contracting, ren- or an increased probability of opportunism dering this form of governance in a franchise with highly innovative franchisees. In franchise context questionable. In addition, though systems in which franchisees are actively innovativeness has not been identified as a encouraged to participate in innovation pro- direct antecedent of opportunism, it is important cesses, however, the assumed relationship to be aware of its opportunism-increasing might not necessarily exist. Given that innova- effects when following a relational contracting tion activities are traditionally seen as the duty approach. It should be mentioned that this of the franchisor (Falbe, Dandridge, and Kumar finding does not suggest that the use of rela- 1999), there is a point to be made here that the tional contracting is unsuitable for safeguarding potential of innovative franchisees may be real- against opportunism; however, it suggests that ized without the negative aspects of an there exists at least one boundary condition for increased level of opportunism. relational contracting to protect against oppor- Empirically, the reason for the nonsignificant tunistic behavior. impact might be the fact that innovativeness This result contributes to the understanding has a rather low S.D. This suggests that of a key dilemma in franchising: Though there is innovativeness does not differ a great deal evidence that higher levels of autonomy—which among franchisees. Though this does not imply can be fostered through relational contracting— that innovativeness is not an important predictor are important to leverage the entrepreneurial of opportunism, it suggests that variation in capacities of franchisees (Kaufmann and Eroglu innovativeness is unlikely to explain the varia- 1999; Schul, Little, and Pride 1985; Strutton, tion in the opportunism construct. Pelton, and Lumpkin 1995), we demonstrate Taken together, the results of H1a–c show that at the same time, this form of contracting that it is useful to look at the three entrepreneur- does not only have positive consequences. The ial characteristics separately, instead of aggre- potential reason for this observation is that gating them to one higher-order construct, relational contracting—compared with hierar- entrepreneurial orientation, when investigating chical forms of governance—increases the outcomes such as opportunism (Kreiser, already existing ambiguity in hybrid governance Marino, and Weaver 2002; Lumpkin and Dess forms, such as franchise systems. The conse- 1996). Thereby, our findings complement the quence is that it is not only more difficult to results of a meta-analysis investigating the rela- evaluate the actions of franchisees in the first tionship between personality characteristics and place but also to correctly interpret franchisees’ entrepreneurial status (Zhao and Seibert 2006), behaviors as compliant or opportunistic which reveals that one of the Big Five person- (Carson, Madhok, and Wu 2006). Therefore, the ality traits, agreeableness, is negatively related boundaries for the judgment of and sanctioning to entrepreneurial status. Thus, according to the against opportunism may become more blurred. definition of agreeableness, entrepreneurs seem We have to note, however, that this is our to be more strongly determined by self-interest interpretation of the effect, and we do not have while being less altruistic, forgiving, or gullible the necessary data to formally test the mediating (Zhao and Seibert 2006). However, our results role of ambiguity in the moderating effect of indicate that this may only hold for the entre- relational contracting. Further research should preneurial characteristics proactiveness and risk possibly consider doing this as a fruitful avenue taking, but not for innovativeness. for advancing our understanding of the func- In the case of franchising as a hybrid gover- tioning of relational contracting. nance form, and using franchisees’ entrepre- As expected, franchisee age reduces oppor- neurial characteristics and relational contracting tunism, possibly through the fact that franchi- as examples, we demonstrate how elements of see trust increases over time (after an initial ex-post contracting moderate the link between drop; see Blut et al. 2011), which likely reduces

12290 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT opportunism. Moreover, it can be argued that Considering these, a combination of methods, the longer a franchisee is part of a franchise including the assessment of objective informa- system, the more the norms affect their tion, such as prior business experience, as well behavior—thus leading to the inverse relation- as a candidate’s personality and motivations, ship between franchisee age and opportunism. including risk taking and proactiveness (Jambulingam and Nevin 1999), may enable Managerial Implications franchisors to identify the most suited future The results of our study have several mana- franchisees. For example, one characteristic gerial implications. Given that proactiveness that franchisors may be looking for is franchi- and risk taking are at least partially linked to sees’ willingness to cooperate. This in combi- opportunism, franchisors should more fully con- nation with the entrepreneurial capacity of sider how to select potential franchisees and innovativeness could be a good foundation in assess future transaction costs ex-ante. For order to implement a system strategy properly, example, it is possible that individuals with augmented through the ideas and initiatives of higher levels of proactiveness and risk taking the franchisee. Looking at the results related to may not necessarily drive franchise system per- H2c, it is suggested that in the case of a highly formance. Instead, they may cause more ex-post relational contract regime, franchisors should transaction costs due to opportunistic behavior. be especially aware of the danger of allowing Looking at the relative magnitude of the franchisees with a high propensity to risk effects, risk taking (H1c) can be identified as taking to enter their system. the variable that is—by far—most strongly It is also possible that individuals with stron- related to opportunism. Thus, it might be useful ger entrepreneurial motivations are more likely for franchisors to incorporate assessments of to exit the system sooner as they become frus- risk taking into their selection policies. trated and dissatisfied with the boundaries set However, innovativeness (H1a) does not seem within franchise systems in general (Ping to be a “warning sign” of future opportunistic 1993). Evidence for this could be our finding behavior under all circumstances. Given that that there is a negative link between the ideas generated by individual franchisees are an number of years of belonging to a franchise increasingly important source to secure and system and levels of opportunism. It is quite develop a franchise system’s business model (in possible that franchisees who have, for the case of McDonald’s, the invention of the Big example, a stronger entrepreneurial motivation Mac and McCafé provide famous examples of and/or more substantial business experience innovations that originated from franchisee ini- will want to leverage their capacities more fully tiatives), there is, in general, less danger in and in their own terms. For this reason, they offering franchise contracts to candidates who may strive for more autonomy and, if this is not have a proven track record of innovative ideas. being granted within the system, may decide to Selecting franchisees on the basis of their inno- open their own venture. vative capabilities, thus, might be generally The key managerial implication for franchi- useful as such candidates might probably sors is to be aware of a “dark side” of entre- engage in advancing the franchise system as a preneurial motivation: opportunistic behavior. whole without primarily seeking their self- It seems that franchisors cannot have one interest and exhibiting free-riding tendencies. In without the other and thus have to find a franchise systems that follow a highly relational balance between the two that best suits their contracting approach, however, innovativeness system. It is necessary to understand where the may foster opportunism. Therefore, franchisors optimal point lies that gives entrepreneurial who intend to capitalize on their franchisee’s and successful franchisees enough flexibility to innovation capabilities should take care of exploit their capacities and thereby keep them having implemented suitable (formalized) satisfied and in the system while at the same follow-up processes to effectively handle fran- time not running into the danger of creating chisee initiatives at the franchise system level. In opportunities for self-interested actions that addition, if innovative franchisees are prevented may damage the system. from implementing their ideas in their own The identification of how this balance can be outlets by their franchisor, this may similarly achieved is beyond the scope of our article. result in negative consequences such as reac- However, there is one suggestion we would tance, less cooperation, and opportunism. like to be considered: Related to the discussion

EVANSCHITZKY, CAEMMERER, AND AND BACKHAUS BACKHAUS 291 13 of franchisors selecting franchisees is the the system for longer. Again, in order to notion of franchisees’ self-selection. If the fran- achieve this, franchisors need to be aware of chise system establishes through external com- the trade-off they have to make between grant- munications, such as recruitment ads and ing enough autonomy to keep franchisees and promotional material, a realistic picture of what setting enough boundaries in order to prevent is expected of franchisees, it is more likely that opportunism. In order to achieve this success- candidates with a better fit with the system fully, franchisors may have to consider the apply as potential franchisees. As a result, it experience and characteristics of individual might be easier for management to identify the franchisees as well as their position in the fran- right balance between flexibility and control as chisee lifecycle (Blut et al. 2011). the gap between franchisees’ and franchisors’ Therefore, though it sounds plausible that expectations in relation to behavioral norms franchisees need to be treated the same may be smaller (Bretz and Judge 1998). throughout the system in order to avoid inter- There is also an alternative explanation for nal quarrels, there is an argument to be made the negative link between the length of fran- whether the management of franchisees can be chisees belonging to a system and levels of really standardized. Given that these individu- opportunism: With increased years of member- als have made over the years different experi- ship, any contractual arrangements, formal or ences in the franchise system, and have very relational, may become better defined, which different backgrounds, capacities, and aspira- decreases the propensity to behave opportunis- tions, it is possible that an idiosyncratic tically. Due to reasons of bounded rationality, approach to franchisee management might be franchisors cannot anticipate all problems that more efficient (Birkeland 2002). This can also may arise throughout a partnership with a fran- be achieved by the franchisors’ right to com- chisee. Therefore, initial formal contracts may plete existing individual contracts after franchi- be suboptimal at the start. However, with the sees have entered the system (Arruñada, evolution of the system, franchisors’ increasing Garicano, and Vázquez 2001). experience, and knowledge of specific franchi- An idiosyncratic approach would not only sees, new contracts for franchisees entering the consider contracting methods, either formal or system in the future may become more efficient informal, but also include other practices such over time (Cochet and Garg 2008). At the same as the identification of individual training and time, in order to safeguard against opportunism development needs and opportunities. For and to ensure performance over time of exist- example, highly driven and successful individu- ing franchisees, franchisors can complement als could be more quickly given the option to the content of contracts by specifying perfor- develop entire regions through the ownership mance terms and control mechanisms post hoc of multiple units. Through such initiatives, a (Arruñada, Garicano, and Vázquez 2001). Thus, stronger sense of achievement may be instilled, formalization, which is regarded as a cost of which in turn could lead to higher levels of contracting, may in fact be beneficial within a satisfaction and longer loyalty to the system franchise system as it serves as a frame of (Ping 1993). reference to identify opportunistic behavior (Combs and Ketchen 1999). Alongside formal- Conclusion and Directions for ization, dimensions of relational contracting Future Research can also develop over time, which may lead to To conclude, we suggest that, in general, stronger social barriers to opportunism. In managers should take into account that rela- cases where relational contracting does not tional contracting might not be efficient in safe- establish appropriate safeguards, franchisors guarding against all risks of opportunism, in can still retain the right to complete existing particular as the level of franchisees’ entrepre- formal contracts with additional terms. neurial motivation impacts on opportunistic These arguments lend support for the notion tendencies. It is for these reasons that franchi- that through learning and experience with spe- sors need to be aware of the dilemma they are cific franchisees, franchisors become better at facing in order to make an efficient ex-ante efficient contracting, which in turn should posi- selection, as well as ex-post contracting deci- tively impact on a system’s success. The mana- sions: Entrepreneurially minded franchisees gerial implication for franchisors is that it might who might be better at exploiting market be desirable to ensure that franchisees stay in opportunities for their units may at the same

14292 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT time behave more opportunistically, if given Rights and Incentives: The Case of Automo- the chance. However, if they perceive the con- bile Distribution,” Journal of Law, Econom- tractual framework as being too narrow, they ics and Organization 17(1), 257–284. may exit the system faster in order to start their Aulakh, P. S., and E. F. Gençtürk (2007). “Con- own venture. As we observed, a negative link tract Formalization and Governance of between a franchisees’ time in the system and Exporter–Importer Relationships,” Journal the tendency to act opportunistically, it seems a of Management Studies 45(3), 457–479. worthwhile pursuit to keep franchisees satis- Berg, S. V., and P. Friedman (1980). “Causes fied and in the system. and Effects of Joint Venture Activity: Following this thought, there is great merit Knowledge Acquisition versus Parent to explore further whether initial transaction Horizontality,” Antitrust Bulletin 25(1), 143– costs caused by opportunism of entrepreneur- 168. ially oriented franchisees outweigh the later Birkeland, P. M. (2002). Franchising Dreams: gains through gradually more efficient contract- The Lure of Entrepreneurship in America. ing and revenue streams through franchisees’ Chicago, IL: University of Chicago Press. entrepreneurial motivation. Bliese, P. D. (2000). “Within-Group Agreement, Finally, we would like to point out that we Non-Independence, and Reliability: Implica- only focused on the extent to which relational tions for Data Aggregation and Analysis,” in contracting may or may not safeguard against Multilevel Theory, Research, and Methods in negative effects stemming from entrepreneurial Organizations. Eds. K. J. Klein and S. W. J. characteristics. In order to gain a better insight Kozlowski. San Francisco, CA: Jossey-Bass, into the advantages and disadvantages of dif- 349–381. ferent types of contracting in hybrid gover- Blut, M., C. Backhaus, T. Heussler, D. nance forms, studies should compare the Woisetschlager, H. Evanschitzky, and D. effects of formal and relational contracting Ahlert (2011). “What to Expect after the Hon- (Carson, Madhok, and Wu 2006). It may be eymoon: Testing a Lifecycle Theory of Fran- particularly interesting to analyze how these chise Relationships,” Journal of Retailing two types of contracts develop and define the 87(3), 306–319. relationship between franchisors and franchi- Bretz, R., and T. Judge (1998). “Realistic Job sees over time (Cochet and Garg 2008). As in Previews: A Test of the Adverse Self- organizational reality the two types of contract- Selection Hypothesis,” Journal of Applied ing may complement each other (Poppo and Psychology 83(2), 330–337. Zenger 2002), there is merit to investigate in Brown, J. R., C. S. Dev, and D. Lee (2000). more depth how the two are interrelated and “Managing Marketing Channel Opportun- impact on franchisors’ transaction costs. ism: The Efficacy of Alternative Governance In conclusion and despite the limitations of Mechanisms,” Journal of Marketing our study, our results point to an important 64(April), 51–65. dilemma franchisors need to be aware of: Caliendo, M., F. Fossen, and A. Kritikos (2009). Entrepreneurially minded franchisees who “Risk Attitudes of Nascent Entrepreneurs— might be better at exploiting market opportu- New Evidence from an Experimentally Vali- nities may also behave more opportunistically, dated Survey,” Small Business Economics if given the chance through a more relational 32(2), 153–167. contracting regime. At the same time, if they Carland, J. W., F. Hoy, W. R. Boulton, and J. A. perceive the contractual framework as being C. Carland (1984). “Differentiating Entrepre- too rigid, they may be less able to leverage their neurs from Small Business Owners: A Con- capabilities, become dissatisfied, and exit the ceptualization,” Academy of Management system. That is arguably the key challenge in Review 9(2), 354–359. franchising. Carney, M., and E. Gedajlovic (1991). “Vertical Integration in Franchise Systems: Agency References Theory and Resource Explanations,” Strate- Ajzen, I. (1991). “The Theory of Planned gic Management Journal 12(8), 607–629. Behaviour,” Organizational Behavior and Carson, S. J., A. Madhok, and T. Wu (2006). Human Decision Processes 50(2), 179–211. “Uncertainty, Opportunism and Governance: Arruñada, B., L. Garicano, and L. Vázquez The Effects of Volatility and Ambiguity (2001). “Contractual Allocation of Decision on Formal and Relational Contracting,”

EVANSCHITZKY, CAEMMERER, AND AND BACKHAUS BACKHAUS 293 15 Academy of Management Journal 49(5), Deeds, D. L., and C. W. L. Hill (1999). “An 1058–1077. Examination of Opportunistic Action Within Castrogiovanni, G. J., J. G. Combs, and R. T. Research Alliances: Evidence from the Bio- Justis (2006). “Resource Scarcity and Agency technology Industry,” Journal of Business Theory Predictions Concerning the Contin- Venturing 14(2), 141–163. ued Use of Franchising in Multi-Outlet Net- Delmar, F., and P. Davidsson (2000). “Where works,” Journal of Small Business Do They Come From? Prevalence and Char- Management 44(1), 27–44. acteristics of Nascent Entrepreneurs,” Entre- Cochet, O., and V. Garg (2008). “How Do Fran- preneurship and Regional Development chise Contracts Evolve? A Study of Three 12(1), 1–23. German SMEs,” Journal of Small Business Eisenhardt, K. M. (1989). “Agency Theory: An Management 46(1), 134–151. Assessment and Review,” Academy of Man- Cochet, O., J. Dormann, and T. Ehrmann agement Review 14(1), 57–74. (2008). “Capitalizing on Franchisee Auto- Eysenck, H. J. (1976). The Measurement of nomy: Relational Forms of Governance as Personality. Baltimore, MD: University Park Controls in Idiosyncratic Franchise Dyads,” Press. Journal of Small Business Management Falbe, C. M., T. C. Dandridge, and A. Kumar 46(1), 50–72. (1999). “The Effect of Organizational Combs, J. G., and D. J. Ketchen (1999). “Can Context on Entrepreneurial Strategies in Capital Scarcity Help Agency Theory Explain Franchising,” Journal of Business Venturing Franchising? Revisiting the Capital Scarcity 14, 125–140. Hypothesis,” Academy of Management Fornell, C., and D. F. Larcker (1981). “Evaluat- Journal 42(2), 196–207. ing Structural Equation Models with Unob- Combs, J. G., S. C. Michael, and G. J. servable Variables and Measurement Error,” Castrogiovanni (2009). “Institutional Inﬂu- Journal of Marketing Research 18(February), ences on the Choice of Organizational Form: 39–50. The Case of Franchising,” Journal of Man- George, B., and L. Marino (2011). “The agement 35(5), 1268–1290. Epistemology of Entrepreneurial Orientation: Covin, J. G., and D. P. Slevin (1986). “The Conceptual Formation, Modeling, and Opera- Development and Testing of an tionalization,” Entrepreneurship Theory and Organization-Level Entrepreneurship Scale,” Practice 35(5), 989–1024. in Frontiers of Entrepreneurship Research. George, J. M., and J. Zhou (2001). “When Open- Eds. R. Ronstadt, J. A. Hornaday, and K. H. ness to Experience and Conscientiousness Vesper. Wellesley, MA: Babson College, are Related to Creative Behavior: An Inter- 626–639. actional Approach,” Journal of Applied Psy- ——— (1988). “The Inﬂuence of Organization chology 86(3), 513–524. Structure on the Utility of an Entrepreneurial Gillis, W. E., E. McEwan, T. R. Crook, and S. C. Top Management Style,” Journal of Manage- Michael (2011). “Using Tournaments to ment Studies 25(3), 217–259. Reduce Agency Problems: The Case of Fran- Covin, J. G., and W. J. Wales (2012). “The chising,” Entrepreneurship Theory and Prac- Measurement of Entrepreneurial Orienta- tice 35(3), 427–447. tion,” Entrepreneurship Theory and Practice Grünhagen, M., and R. A. Mittelstaedt (2005). 36(4), 677–702. “Entrepreneurs or Investors: Do Multi-Unit Dant, R. P., and G. T. Gundlach (1999). “The Franchisees Have Different Philosophical Challenge of Autonomy and Dependence in Orientations,” Journal of Small Business Franchised Channels of Distribution,” Management 43(3), 207–225. Journal of Business Venturing 14(1), 35–67. Hair, J., W. C. Black, B. Babin, R. E. Anderson, David, R. J., and S. Han (2004). “A Systematic and L. Tatham (2006). Multivariate Data Assessment of the Empirical Support for Analysis. Englewood Cliffs, NJ: Prentice Transaction Cost Economics,” Strategic Man- Hall. agement Journal 25(1), 39–58. Heide, J. B., and G. John (1992). “Do Norms Davies, W. L., C. Manolis, M. Prince, and R. D. Matter in Marketing Relationships?,” Journal Winsor (2011). “A Model of Trust and Com- of Marketing 56(2), 32–44. pliance in Franchise Relationships,” Journal Hoetker, G., and T. Mellewigt (2009). “Choice of Business Venturing 26(3), 321–340. and Performance of Governance Mecha-

16294 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT nisms: Matching Alliance Governance to and Agency Theory: An Organizational Eco- Asset Type,” Strategic Management Journal nomics Approach to Strategic Management,” 30(10), 1025–1044. Managerial and Decision Economics 26(4), Hox, J. J. (1995). Applied Multilevel Analysis. 223–242. Amsterdam: TT-Publikaties. Korunka, C., H. Frank, M. Lueger, and J. Mugler Hunt, S. D. (1977). “Franchising: Promises, (2003). “The Entrepreneurial Personality in Problems, Prospects,” Journal of Retailing the Context of Resources, Environment, and 53(3), 71–84. the Startup Process—A Configurational Jambulingam, T., and J. R. Nevin (1999). “Influ- Approach,” Entrepreneurship Theory and ence of Franchisee Selection Criteria on Out- Practice 28(1), 23–42. comes Desired by the Franchisor,” Journal Koza, K. L., and R. P. Dant (2007). “Effects of of Business Venturing 14(4), 363–395. Relationship Climate, Control Mechanism, Jap, S. D., and S. Ganesan (2000). “Control and Communications on Conflict Resolution Mechanisms and the Relationship Life Cycle: Behavior and Performance Outcomes,” Implications for Safeguarding Specific Journal of Retailing 83(3), 279–296. Investments and Developing Commitment,” Kozlowski, S. W., and K. J. Klein (2000). “A Journal of Marketing Research 37(2), 227– Multilevel Approach to Theory and Research 245. in Organizations,” in Multilevel Theory, Kashyap, V., K. D. Antia, and G. L. Frazier Research, and Methods in Organizations: (2012). “Contracts, Extracontractual Incen- Foundations, Extensions and New Direc- tives, and Ex-Post Behavior in Franchise tions. Eds. K. Klein and S. Kozlowski. San Channel Relationships,” Journal of Market- Francisco: Jossey-Bass, 3–90. ing Research 49(2), 260–276. Kreiser, P. M., L. Marino, and K. M. Weaver Kaufmann, P. J., and R. P. Dant (1999). “Fran- (2002). “Assessing the Psychometric Proper- chising and the Domain of Entrepreneurship ties of the Entrepreneurial Orientation Scale: Research,” Journal of Business Venturing A Multi-Country Analysis,” Entrepreneurship 14(1), 69–85. Theory and Practice 26(4), 71–94. Kaufmann, P. J., and S. Eroglu (1999). “Stan- Krueger, N. F., and A. L. Carsrud (1993). “Entre- dardization and Adaptation in Business preneurial Intentions: Applying the Theory Format Franchising,” Journal of Business of Planned Behavior,” Entrepreneurship and Venturing 14(1), 69–85. Regional Development 5, 315–330. Kaufmann, P. J., and V. K. Rangan (1990). “A Lado, A. A., R. Dant, and A. G. Tekleab Model for Managing System Conflict During (2008). “Trust-Opportunism Paradox, Franchise Expansion,” Journal of Retailing Relationalism, and Performance in Interfirm 66(2), 155–173. Relationships: Evidence from the Retail Kaufmann, P. J., and L. W. Stern (1988). “Rela- Industry,” Strategic Management Journal tional Exchange Norms, Perceptions of 29(4), 401–423. Unfairness, and Retained Hostility in Com- Li, Y. H., J. W. Huang, and M. Tsai (2009). mercial Litigation,” Journal of Conflict Reso- “Entrepreneurial Orientation and Firm Per- lution 32(2), 534–552. formance: The Role of Knowledge Creation Ketchen, D. J., J. C. Short, and J. G. Combs Process,” Industrial Marketing Management (2011). “Is Franchising Entrepreneurship? 38(4), 440–449. Yes, No, and Maybe So,” Entrepreneurship Li, Z. G., and R. P. Dant (1997). “An Exploratory Theory and Practice 35(3), 583–593. Study of Exclusive Dealing in Channel Rela- Kickul, J., and L. Gundry (2002). “Prospecting tionships,” Journal of the Academy of Mar- for Strategic Advantage: The Proactive Entre- keting Science 25(3), 201–213. preneurial Personality and Small Firm Inno- Lilis, C. M., C. L. Narayana, and J. L. Gilman vation,” Journal of Small Business (1976). “Competitive Advantage Variation Management 40(2), 85–97. over the Life Cycle of a Franchise,” Journal Kidwell, R., and A. Nygaard (2011). “A Strategic of Marketing 40(4), 77–80. Deviance Perspective on the Franchise Form Lumpkin, G. T., C. Cogliser, and D. R. of Organizing,” Entrepreneurship Theory Schneider (2008). “Understanding and Mea- and Practice 35(3), 467–482. suring Autonomy: An Entrepreneurial Orien- Kim, J., and J. Mahoney (2005). “Property tation Perspective,” Entrepreneurship Theory Rights Theory, Transaction Costs Theory, and Practice 33(1), 47–69.

EVANSCHITZKY, CAEMMERER, AND AND BACKHAUS BACKHAUS 295 17 Lumpkin, G. T., and G. G. Dess (1996). “Clari- mended Remedies,” Journal of Applied Psy- fying the Entrepreneurial Orientation Con- chology 88(5), 879–903. struct and Linking It to Performance,” Poppo, L., and T. Zenger (2002). “Do Formal Academy of Management Review 21(1), 135– Contracts and Relational Governance Func- 172. tion as Substitutes or Complements?,” Stra- ——— (2001). “Linking Two Dimensions of tegic Management Journal 23(8), 707–725. Entrepreneurial Orientation to Firm Perfor- Powell, W. W. (1990). “Neither Market nor Hier- mance: The Moderating Role of Environment archy: Network Forms of Organization,” in and Industry Life Cycle,” Journal of Business Research in Organizational Behavior. Eds. Venturing 16(5), 429–451. B. M. Staw and L. L. Cummings. Greenwich, Luo, Y. (2002). “Contract, Cooperation and Per- CT: JAI Press, 295–336. formance in International Joint Ventures,” Price, S. (1997). The Franchise Paradox—New Strategic Management Journal 23, 903– Directions, Different Strategies. London, UK: 919. Cassell. Lusch, R. F., and J. R. Brown (1996). “Interde- Raudenbush, S. W., and A. S. Bryk (2002). pendency, Contracting, and Relational Hierarchical Linear Models: Applications Behavior in Marketing Channels,” Journal of and Data Analysis Methods. Newbury Park, Marketing 60(4), 19–38. CA: Sage. Macneil, I. R. (1980). The New Social Contract: Raudenbush, S. W., A. S. Bryk, Y. F. Cheong, R. An Inquiry into Modern Contractual Rela- Congdon, and M. du Toit (2004). HLM 6: tions. New Haven, CT: Yale University Press. Hierarchical Linear and Nonlinear Model- Michael, S. C., and J. G. Combs (2008). “Entre- ing. Lincolnwood, IL: Scientiﬁc Software preneurial Failure: The Case of Franchisees,” International. Journal of Small Business Management Runco, M. (2004). “Creativity,” Annual Review 46(1), 73–90. of Psychology 55(1), 657–687. Miller, D. (1983). “The Correlates of Entrepre- Schul, P. L., T. E. Little, and W. Pride (1985). neurship in Three Types of Firms,” Manage- “Channel Climate: It’s Impact on Channel ment Science 29(7), 770–791. Members’ Satisfaction,” Journal of Retailing Mitsuhashi, H., S. Shane, and W. D. Sine 61(2), 9–38. (2008). “Organization Governance Form in Shane, S. (1996). “Hybrid Organizational Franchising: Efﬁcient Contracting or Organi- Arrangements and their Implications for zational Momentum?,” Strategic Manage- Firm Growth and Survival: A Study of New ment Journal 29(10), 1127–1136. Franchisors,” Academy of Management Morris, M. H., and D. L. Sexton (1996). “The Journal 39(1), 216–234. Concept of Entrepreneurial Intensity: Impli- Sillince, J., P. Jarzabkowski, and D. Shaw cations for Company Performance,” Journal (2010). “Shaping Strategic Action through of Business Research 36(1), 5–13. the Rhetorical Construction and Exploitation Muthen, B. O., and A. Satorra (1995). “Complex of Ambiguity,” Organization Science 23(3), Sample Data in Structural Equation Model- 630–650. ling,” Sociological Methodology 25, 267–316. Snijders, T. A. B., and R. J. Bosker (1999). Nair, S. K., S. Tikoo, and S. Liu (2009). “Valuing Multilevel Analysis: An Introduction to Basic Exclusivity from Encroachment in Franchis- and Advanced Multilevel Modeling. London, ing,” Journal of Retailing 85(2), 206–210. UK: Sage. Norton, S. W. (1988). “An Empirical Look at Sorenson, O., and J. B. Sorensen (2001). Franchising as an Organizational Form,” “Finding the Right Mix: Franchising, Organi- Journal of Business 61(2), 197–218. zational Learning, and Chain Performance,” Ping, R. A. (1993). “The Effects of Satisfaction Strategic Management Journal 22(6/7), 713– and Structural Constraints on Retailer 724. Exiting, Voice, Loyalty, Opportunism, and Strutton, D., L. E. Pelton, and J. R. Lumpkin Neglect,” Journal of Retailing 69(3), 320– (1995). “Psychological Climate in Franchis- 352. ing System Channels and Franchisor- Podsakoff, P. M., S. B. MacKenzie, J. Y. Lee, Franchisee Solidarity,” Journal of Business and N. P. Podsakoff (2003). “Common Research 34(2), 81–91. Method Biases in Behavioral Research: A Stump, R. L., and J. B. Heide (1996). “Control- Critical Review of the Literature and Recom- ling Supplier Opportunism in Industrial

18296 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Relationships,” Journal of Marketing Proactiveness Research 33(November), 431–441. (Three items, five-point Likert scale, strongly Tracey, P., and O. Jarvis (2007). “Toward a disagree to strongly agree) Theory of Social Venture Franchising,” Entrepreneurship Theory and Practice 31(5), (1) I look for challenges at work. 667–685. (2) I think a lot about how to improve my Wiklund, J. (1999). “The Sustainability of the chances for getting ahead. Entrepreneurial Orientation-Performance (3) I get a thrill out of confronting challenges Relationship,” Entrepreneurship Theory and at work. Practice 24(1), 37–48. Wiklund, J., and D. Shepherd (2005). “Entre- Business Risk Taking preneurial Orientation and Small Business (Four items, five-point Likert scale, strongly Performance: A Configurational Approach,” disagree to strongly agree) Journal of Business Venturing 20(1), 71–89. Williamson, O. E. (1975). Markets and Hierar- (1) My business strategy is characterized by a chies. New York: Free Press. strong tendency to undertake high-risk ——— (1985). The Economic Institution of projects. Capitalism. New York: Free Press. (2) A business should only take risks in areas ——— (1991). “Strategizing, Economizing, and that it knows well (R). Economic Organization,” Strategic Manage- (3) Research is important before making a ment Journal 12(Special Issue), 75–94. risky decision (R). ——— (2010). “Transaction Cost Economics: (4) I am a gambler—it is impossible to plan The Natural Progression,” Journal of Retail- for the future. ing 86(3), 215–226. Windsperger, J., and R. Dant (2006). “Contract- Franchisee Opportunism ibility and Ownership Redirection in Fran- (Seven items, five-point Likert scale, strongly chising: A Property Rights View,” Journal of disagree to strongly agree) Retailing 82(3), 259–272. Yin, X., and E. J. Zajac (2004). “The Strategy/ (1) I sometimes alter facts slightly in order to Governance Structure Fit Relationship: gain the cooperation of my franchisor. Theory and Evidence in Franchising (2) I sometimes explicitly promise to do Arrangements,” Strategic Management things requested by my franchisor without Journal 25(4), 365–383. actually doing them later. Zhao, H., and S. E. Seibert (2006). “The Big Five (3) I sometimes withhold information that Personality Dimensions and Entrepreneurial would help my franchisor to run his Status: A Meta-Analytical Review,” Journal of business. Applied Psychology 91(2), 259–271. (4) I do not always share information in a timely manner with my franchisor. Appendix: Questionnaire Items (5) I sometimes make vague promises to my Entrepreneurial Characteristics franchisor that I later ignore. (Measures Adapted from Jambulingam (6) I sometimes purposely withhold informa- and Nevin 1999) tion that would put me in a bad light. Innovativeness (7) I sometimes tell my franchisor what I (Four items, five-point Likert scale, strongly think he wants to hear instead of telling disagree to strongly agree) him the truth.

(1) Franchisees should provide leadership in Relational Contracting (Jambulingam and new product development. Nevin 1999) (2) The success of my franchise depends (10 items, ﬁve-point Likert scale, strongly on constantly improving our products/ disagree to strongly agree) services. (3) Franchisees should provide leadership in Information Exchange new operating procedure development. (4) In my franchise, we emphasize the promo- (1) Exchange of information in this relation- tion of new innovative products/services. ship takes place frequently and informally

EVANSCHITZKY, CAEMMERER, AND AND BACKHAUS BACKHAUS 297 19 and not only according to prespeciﬁed (2) My franchisor always makes an honest agreement. attempt to work out problems between us. (2) In our relationship, it is expected that any (3) My franchisor is very unwilling to settle information that might help the other disputes with me (R). party will be provided to them. (4) Disagreements with my franchisor are (3) In this relationship, it is expected that usually solved by one-sided, “take-it-or- parties will provide conﬁdential informa- leave-it” solutions proposed by the fran- tion if it can help the other party. chisor (R).

Flexibility Success of the Franchise System (Three items, ﬁve-point Likert scale, strongly (1) Flexibility in response to requests disagree to strongly agree) for changes is a characteristic of our relationship. (1) In all, the franchise system has been very (2) The ability to make adjustments in order to successful in the past three years. cope with changing circumstances is a (2) Compared with our competition, we have characteristic of our relationship. been very successful. (3) My franchisor and I are open to modifying (3) We have achieved our goals in terms of our agreements if unexpected events ﬁnancial success. occur.

Problem Solving

(1) When problems occur, I can easily ﬁnd a mutually agreeable solution with my franchisor.

20298 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Journal of Small Business Business Management Management 2016 2015 54(1), ••(••), pp. pp. 299–318 ••–•• doi: 10.1111/jsbm.12146 The When and Why: Student Entrepreneurial Aspirations by Caleb Kwong and Piers Thompson

Although connections between university enterprise courses and entrepreneurial activity have been examined, less work has investigated the intended timing of future entrepreneurial activities. Using data from a survey of U.K. business students, it is found that those intending to enter entrepreneurship right away place less emphasis on avoiding stress and responsibility, seeing themselves as natural leaders. They were also more conﬁdent of succeeding, but not because of superior knowledge. A greater emphasis on entrepreneurial activities in all institutional environments, including the corporate, may help balance the need to harness enthusiasm while it lasts with the need to acquire relevant experience.

the entrepreneurial career option (Donckels Introduction 1991; Kantor 1988), and for those who have The potential for greater small business already developed interest in entrepreneurship, ownership to increase the level of entrepre- to increase their start-up and small enterprise neurial activity, innovation, and creativity management capabilities (Johannisson 1991; within an economy has made policies and Kantor 1988). Traditionally, university entre- programs aiming to promote new venture cre- preneurship education pays most attention to ation extremely attractive to policymakers (Acs the latter, with the action-orientated “go-out- and Audretsch 2003; Audretsch, Keilbach, and and-do-it-now” philosophy remaining the most Lehmann 2006; Gilbert, Audretsch, and prominent approach (Ronstadt 1985). Within McDougall 2004; van Stel, Carree, and Thurik this philosophy, the role of enterprise educa- 2005). Along with a greater dissatisfaction with tion through the acquisition of skills and traditional corporate careers, such a shift has network connections is to increase students’ led more business students into contemplating self-conﬁdence in relation to the process of careers as business owners in their own right starting a business and thereby creating a (Brockhaus and Horwitz 1986). This has in linkage from vision to action (Johannisson recent years led to an explosion in the number 1991). of entrepreneurship modules and courses However, whereas a minority of business taught within business schools (Katz 2003; school graduates immediately embark on an Kuratko 2005; Vesper and Gartner 1997). The entrepreneurial career upon graduation, a aims of such programs are to increase aware- majority prefer working for others ﬁrst before ness among those who have little knowledge of taking the plunge (Brockhaus and Horwitz

Caleb Kwong is senior lecturer at Essex Business School, University of Essex. Piers Thompson is Lecturer in Economics at Nottingham Business School, Nottingham Trent University. Address correspondence to: Piers Thompson, Nottingham Business School, Nottingham Trent University, Burton Street, Nottingham NG1 4BU, UK. E-mail: [email protected].

KWONG AND THOMPSON 299 1 1986; Brown 1990; Collins, Hannon, and Smith also the way in which the nature of the subse- 2004; Galloway and Brown 2002; Ronstadt quent growth and development of these ven- 1985). The desire to wait can partly be attrib- tures can be best supported. On one hand, uted to a lack of emphasis on practical start-up those starting immediately may require greater skills, knowledge, and network connections in support and assistance with practical skills and university courses (Carter and Collinson 1999; network creation (Carter and Collinson 1999). Volery, Doss, and Mazzarol 1997), with many On the other hand, for the “wait-and-see” entre- students taking both technical or business sub- preneurs, assistance in developing a long-term jects found to prefer developing greater expe- yet imaginable path to obtain essential skills, rience and knowledge prior to business experience, and finance to enter entrepreneur- ownership (Bird and Schjoedt 2009; Collins, ship may need to be provided, otherwise their Hannon, and Smith 2004; Ronstadt 1985). Thus, entrepreneurial intention may tend to dissipate the decision to wait is essentially to decrease (Carsrud and Brannback 2011; Galloway et al. the risk of failure, which coincidentally is at the 2006; van Geldren et al. 2006). This time lag is highest at the initial stage of a new venture rarely factored into the development of univer- (Choi, Levesque, and Shepherd 2008; Das sity enterprise education curriculum, with most 1987). Although learning from failures can existing programs confined to the period of also be beneficial where serial entrepreneur- university attendance and any support beyond ship occurs (March 1991; Shepherd 2003), this graduation, outside of the limited capacity of process of learning can be emotionally difficult incubator units, is deemed beyond the univer- to handle (Shepherd 2004). Some argue that sity’s remit (Galloway and Brown 2002). only experiential human capital has any value This study explores the issue of entrepre- (Politis 2005); this would suggest that delaying neurial timing using data from a survey of U.K. initiation to gain greater occupational experi- business and enterprise students within the ence would have little effect on the probability context of a number of well-known intention of success. Others, however, have found that and behavior models. The study concentrates on experience in work helps develop routines that attitudes expressed by those considering the will be used to guide the management of busi- next stage in their careers after completing their nesses in the same industry as well as building formal education at university. These expecta- social capital through professional networks, tions will not necessarily come to realization. although further learning occurs after new For example, studies have found around a third venture creation, which cannot be undertaken of undergraduate students display positive atti- beforehand, such as managing relationships tudes to entrepreneurship (Henley et al. 2009), with employees (Rae 2005). Waiting may allow but generally less than 1 in 10 will become the correct opportunities to be identified and self-employed within the first five years after the relevant resources put in place, so those graduation (Rosa 2003). Discrepancies can that wait may be more innovative and able to relate to entrepreneurial aspiration questions achieve greater growth in the future (Capelleras capturing desires with no or little commitment et al. 2010; West and Meyer 1997). to action rather than firm intentions (van Though there has been some interest in the Geldren et al. 2006). This means that studies of temporal issues within existing ventures (Bird alumni may provide a more accurate picture of 1992; Bird and West 1997; Capelleras et al. the actual drivers and inhibitors of entrepre- 2010) and the common observation that there is neurial activities. However, the study takes a often a significant time lag between the occur- forward-looking approach rather than consider- rence of entrepreneurial intention and the ing actual behavior of alumni, as a considerable actual start-up behavior (Bird and Schjoedt variety of outside events and influences beyond 2009; Carsrud and Brannback 2011; Katz 1994; graduation are likely to come into play, which Krueger, Reilly, and Carsrud 2000; Reynolds educators have little or no control over. In 1994), existing literature on entrepreneurship addition, there is also likely to be some hind- education has made little effort to distinguish sight and retrospective bias in responses from between the two groups mentioned earlier. alumni, which may lead to merging of actual Understanding the timing of entrepreneurial behaviors and their original preferences (Chell activities would not only enrich our under- and Allman 2003). For example, theoretical standing regarding the role of education in the models considering the entrepreneurial choice emergence of graduate entrepreneurship but have also suggested that those considering

2300 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT entering entrepreneurship are not fully aware of toward the activity, whether entrepreneurship their true ability and only become aware of their is perceived as “desirable,” and the probabili- true ability over time (Evans and Jovanovic 1989; ties of succeeding. The latter is described by Jovanovic 1982). Given that expectations of Shapero and Sokol (1982) as the perceived unobserved adjustment appear relatively slowly, feasibility of the behavior and perceived behav- although more quickly in the case of younger ioral control (PBC) by Ajzen (1991). Such entrepreneurs (Parker 2006), this process could perceived ability to control the event is take a relatively long time. Those choosing to extremely important for the concept of entre- not start ventures because of uncertainty about preneurship, because entrepreneurial activities ability may cite other practical reasons rather operate in an environment where barriers, than admit the truth. This makes it unlikely that resource obstacles, and uncertainty make the alumni will recall with complete clarity their success of entrepreneurial activity impossible confidence of success on entering business own- to predict beforehand, that is, where complete ership and even their motivations at the time, volitional control over accomplishment of the but rather their recall will be colored by their behavior is absent (Ajzen and Fishbein 1980). experiences upon engagement. The study there- Such unpredictability has considerable impor- fore splits current students into those who tance when one attempts to understand the gap expect to start businesses within the next 3 years between the positive attitudes and intentions and those intending to wait between 3 and 10 expressed by students toward entrepreneurship years. The groups are compared in terms of while undertaking their studies and the rela- what they consider to be entrepreneurial activi- tively low level of realization of these aspira- ties, their preferences of occupational character- tions. In addition to PBC, the models also point istics, and the attitudes they possess, which to the role of the individual’s attitudes toward relate to the intentions of becoming entrepre- the behavior as well as social norms in the neurs. This is not to say that alumni experiences development of intention. In the entrepreneur- are not important, and as such, studies examin- ship context, becoming an entrepreneur should ing these where appropriate are used to inform be attractive to not only the individual them- the following hypotheses developed. As such, selves but also to their “important others,” that this study tries to concentrate on the expected is, friends, family, and those who had previous choices of students and their reasoning behind entrepreneurial experiences. These influences these choices in order to examine where differ- are what determine the entrepreneurial inten- ences exist and how the universities may best tions of the individuals (Lüthje and Franke design entrepreneurship education to cater for 2003). In turn, intentions have been found to both groups and where necessary, develop be a strong predictor of actual behavior in a support beyond the end of university careers. variety of contexts (Armitage and Conner 2001; Lüthje and Franke 2003; Phan, Wong, and The Temporal Dimension of Wang 2002). Planned Entrepreneurial Though possible displacement events between intention and behavior have been dis- Behavior of Potential cussed in the existing literature as an explana- Entrepreneurs tion for the discrepancy between intention and In a vast majority of cases, the decision to behavior (Bird and Schjoedt 2009; Carsrud and start a new venture is a clearly planned behav- Brannback 2011; Shapero 2002), few studies ior, and as such, models developed to explore attribute such a discrepancy to the matter of this decision are based around influences that timing. Studies have found that when a tempo- make the behavior more attractive and increase ral dimension is specified, situational and per- the probability of success. The two models that ceptual factors are better at explaining long- dominate the literature are Ajzen’s (1991) term rather than short-term intentions (Audet Theory of Planned Behavior (TPB) and the 2004; Reitan 1996). This is consistent with Shapero model of the entrepreneurial event Armitage and Conner’s (2001) suggestion that (Shapero and Sokol 1982). Despite some where intention measures require less commit- notable differences, both models suggest the ment and are closer to desires, these factors decision to start a business is driven by the will play a smaller role. In the context of entre- attitudes that individuals have toward entrepre- preneurship, the heavy commitment required neurship, whether they are favorably disposed to start a business often means that, even after

KWONG AND THOMPSON THOMPSON 301 3 visualizing the entrepreneurial process for the completion of specific technical events, such themselves and making a realistic assessment, as the often formidable task of writing a busi- some potential entrepreneurs may still decide ness plan (Bird 1988) and dealing with the not to enter entrepreneurship immediately, ambiguous and chaotic nature of early-stage because of a perceived lack of some essential business development (Boussouara and skills, knowledge, and experiences that often Deakins 1999; Soloman 2007). In combination, can be best gained outside the context of entre- these may mean the feasibility of entrepreneur- preneurship, such as through employment ship is quite low or unclear at this stage of (Collins, Hannon, and Smith 2004; Katz 2007). business development (Bird 1988). Presumably In fact, Carter and Collinson (1999) found 20 then, those who are willing to commit to more percent of graduates were considering entering explicit behavior expectations when questioned entrepreneurship immediately upon gradua- are those whose PBC is greater. Conversely, tion, and Rosa (2003) found only 1 in 10 gradu- would-be entrepreneurs who do not possess the ates had become an entrepreneur 5 years after required start-up skills are likely to perceive finishing their studies. Therefore, it would be more difficulties the closer it is to the launch of logical to assume that those who are intending a business, and are more likely to delay the a rapid business start are likely to possess very start-up process (Volery, Doss, and Mazzarol different personal qualities to those who opt for 1997). Evidence from studies of alumni has the “wait-and-see” approach. found that a lack of confidence in possessing the Based upon these findings, it might be rea- relevant skills, particularly those relating to sonable to assume that students fall into a practical competencies (Matlay 2008), can act as number of groups based on quite different a deterrent to immediate entrance to self- desired and expected career paths. In this employment upon graduation (Carter and section, we develop our hypotheses in order to Collinson 1999). A lack of confidence has also explain these differences. Figure 1 depicts our been identified as a reason for not attempting to operational model. follow up an entrepreneurial aspiration (Rae and Woodier 2006). Based on the discussion PBC earlier, the following hypotheses are proposed. The importance of PBC for entrepreneurship, as demonstrated earlier, suggests that a person’s H1a: Those looking to immediately move into PBC may have a role to play in determining the entrepreneurial activities are more likely to time lag between the occurrence of entrepre- feel that they have the skills and knowledge neurial intention and the actual behavior in required to start a business than those who starting a business. The start-up process requires wait for longer.

Figure 1 Framework of Entrepreneurial Types, Behaviors, and Activities

4302 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT H1b: Those looking to immediately move into tantly, enable them to get down to the nitty- entrepreneurial activities are more likely to gritty of the events leading to the actual be confident in their own ability to start a business start-up, including exploration, exami- business than those who wait for longer. nation, categorization, and organization of opportunities (Vecchio 2003). Therefore, indi- Attitude toward Entrepreneurship viduals who perceived their leadership capa- In addition to skills and resources acquisi- bilities more positively will have a shorter time tion, entrepreneurship research has empha- lag between the occurrence of entrepreneurial sized the role played by personality traits in intention and the actual behavior. Likewise, contributing to entrepreneurial behaviors perceived desirability for having authority over (Ashworth, Johnson, and Conway 1998; Boyd others, for example, wishing to lead or to avoid and Vozikis 1994; Brockhaus and Horwitz being led by others, and to gain a non- 1986; Cooper, Woo, and Dunkelberg 1988; De pecuniary return, can also affect the behavior Noble, Jung, and Ehrlich 1999; Douglas and of the individual upon graduation and make Fitzsimmons 2008; Ismail et al. 2009; entrepreneurship more likely (Blanchflower Kristiansen and Indarti 2004). Little difference and Oswald 1998). It is therefore only reason- should be expected in terms of the penchant able to expect that these individuals would for entrepreneurship between those who place greater importance on achieving leader- intend to start a business rapidly and those who ship or autonomy in a shorter period of time. intend to start a business at a later date, as both Furthermore, some studies have found the groups display an intention to start a business. temporal dimension of entrepreneurial activity A fact reflected in those alumni of entrepre- and life stress are related (Bluedorn and Martin neurship and small business management 2008). Those who can withstand greater stress course (Donckels 1991). Carter and Collinson or those who are more capable of coping with (1999) also found that such positive attitudes to stress through better time management behav- entrepreneurial activities did not dissipate ior involving goal and priority setting are more quickly on leaving university even where stu- likely to become entrepreneurs within a shorter dents chose to enter employment for others. timeframe (Bluedorn and Martin 2008; Macan Differences in timing, however, may arise in 1994). The literature on personality traits there- terms of personal attitudes toward the alterna- fore leads to the development of the following tive to entrepreneurship, which is working for hypotheses. others. Those who dislike working for others are more likely to concentrate more on their H2a: Those looking to enter entrepreneurship setting up of a business immediately after immediately will display similar levels of graduation, whereas those who do not mind desire for entrepreneurship as those looking working for others maybe more inclined to use to work for others first before becoming employment as an opportunity to learn the entrepreneurs at some point in the future. trade (Carter and Collinson 1999). Trait studies have mainly focused on identi- H2b: Those looking to enter entrepreneurship fying specific personality variables that would immediately are more likely to display a distinguish entrepreneurs from other groups dislike of employment than those looking to and that were presumed to lead to the founding work for others first before becoming entre- of new organizations (Cogliser and Brigham preneurs at some point in the future. 2004). Unsurprisingly, studies have found considerable overlap between entrepreneurship H2c: Those looking to immediately move into and leadership (Cogliser and Brigham 2004; entrepreneurial activities will be more Vecchio 2003). Leadership qualities such as likely to feel that they have stronger leader- extraversion, sensing, and good judgment are ship capabilities than those who wait for most likely to be prominent among first- longer. generation entrepreneurial leaders (Stavrou, Kleanthous, and Anastasiou 2005). Such lead- H2d: Those looking to immediately move into ership qualities are not only essential in pro- entrepreneurial activities will enjoy being a viding inspiration, vision, and value (Kelly, leader more than those looking to work for Athanassiou, and Crittenden 2000; Kets de others first before becoming entrepreneurs at Vries 1993; Ling et al. 2008), but more impor- some point in the future.

KWONG AND THOMPSON THOMPSON 303 5 H2e: Those looking to immediately move into on social norms. Those only displaying entre- entrepreneurial activities are more likely to preneurial desires, on the other hand, may display a greater willingness to take on either lack support in terms of social norms additional responsibilities and stress than and/or will value these opinions of important those who wait for longer. others more. The following hypotheses are developed. Social Norms toward Entrepreneurship Networks and external support are vital in H3a: Those looking to immediately move into determining the speed of venture creation, entrepreneurial activities will be more likely not only through shaping one’s PBC (Shane to feel that they have the support from impor- 2003) but also through the social norms expe- tant others than those who wait for longer. rienced by the students. Studies have found that it is those who possessed network ties with H3b: Those looking to immediately move into executives and bankers who are most likely entrepreneurial activities will value the to start their businesses in a speedy manner opinion of others less than those who wait (Capelleras et al. 2010). However, for a majority for longer. of students with little or no employment history and little associated human, social, and finan- Activities That Are Considered to cial capital, they will be reliant on the support Be Entrepreneurial of their family and friends to internalize risk The difference in timing may also have (Das 1987; Katz 2007). This may manifest itself important implications regarding the types of through PBC when students consider the activities that these potential entrepreneurs resources and emotional support that others may regard as entrepreneurial (Quinn 1985). are likely to make available for them (Allen Carter, Gartner, and Reynolds (1996) found 2000; Cromie, Birley, and Callaghan 1993). In considerable differences in terms of activities addition, the support of friends, family, and undertaken during new venture creation for important others is also crucial in shaping the nascent entrepreneurs who engage in start-up social norms experienced by the students activities and those who prefer to wait and see. (Henderson and Robertson 1999; Matlay 2008). Less work has examined what activities stu- For example, family commitments may put a dents who intend to start a business at some pressure on individuals to fulfill certain roles, point in the future regard as constituting entre- such as providing a secure and stable income preneurship. However, given the different for the family, or alternatively, to follow a careers and skills that students will possess career seen as desirable by parents. Both of when entering entrepreneurship if they follow which may influence the timing of entrepre- the different paths, it might be expected that neurial events. Although Trafimow and Finlay those looking to enter entrepreneurship imme- (1996) suggest that only a minority of individu- diately may regard entrepreneurship in a dif- als are strongly influenced by societal pressure, ferent light to those who wish to acquire such pressure is likely to be more influential resources and an in-depth knowledge of an among those who expect to start a business industry before making the leap. Our study early in their career. Therefore, opinions from proposes that those with a future orientation the group of important others received at the are more likely to consider innovative activities point of undertaking university studies in rela- such as research and development as entrepre- tion to possible career choices are likely to neurial because of their intention to explore have less impact at later dates. Henley et al. their entrepreneurial idea through careful long- (2009) found most students did not feel parents term planning (Das 1987; West and Meyer felt strongly about their future careers. 1997). Fleming (1996) finds that for alumni, the Although not expressing a desire to see their lack of an appearance of a business opportu- children avoid entrepreneurial activities, given nity is seen as the strongest reason for not the uncertainty present in a new venture cre- starting a business. On the other hand, those ation, a lack of vocalized support could have a who rush into the market rapidly may be more similar effect. This means that those expressing inclined to accept cost reduction practices as stronger behavioral expectations may poten- entrepreneurial, as the quick capture of oppor- tially have either greater support in terms of tunities is likely to be the essence of their social norms and/or will place less importance entrepreneurial strategy (Das 1987; Eisenhardt

6304 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT 1989; Eisenhardt and Bourgeois 1988). Given The final usable sample of responses was 151– the literature on attitudes of students and 156 first-year, 38 second-year, 24 third-year actions of alumni, it may be that potential rapid undergraduate respondents, and 33 postgradu- entrepreneurs have a wider conception of ate respondents. The main division of students is entrepreneurship and seek to start a business, based upon the timeframe within which they which may or may not innovate, whereas expect to become an entrepreneur. Our study entrepreneurs-in-waiting feel a new innovative divided the students into three groups: potential niche must be identified and then business rapid entrepreneurs (intending to start in less ownership follows. These predictions are cap- than 3 years); entrepreneurs-in-waiting (those tured within the following hypotheses. wishing to have a career working for others before starting a business in between 3 and 10 H4a: Rapid entrepreneurs are more likely to years time); and finally those only looking to emphasize the importance of cost reduction become entrepreneurs in the distant future (10 compared with those who intend to wait for years or more), or not at all, that can be longer. described as doubtful entrepreneurs. Those already entrepreneurially active are excluded H4b: Rapid entrepreneurs are less likely to from the sample as this group although small are emphasize the importance of innovation likely to be outliers in terms of their responses compared with those who intend to wait for compared with even the potential rapid entre- longer. preneur group. This study concentrates on those items relat- Data and Methodology ing to the preference and intention for entre- In order to examine the hypotheses devel- preneurial activities and what students felt oped in the preceding section, quantitative data these activities included. In order to examine from a survey of U.K. students were utilized. whether the choice of timing could be The sample of students was drawn from those explained by the planned behavior models, studying business and enterprise courses at a items relating to attitudes toward entrepreneur- U.K. higher education institution, who had ial activity were compared for the different attended at least one module on entrepreneur- groups of students. A majority of the items used ship. The cross-sectional data were obtained in in the survey are based on seven-point Likert the form of a questionnaire containing items scales, requiring the extent of agreement with a designed to examine personality traits, entre- statement to be indicated (1 strongly disagree preneurial intentions and preferences, and to 7 strongly agree). Alternatively, where pref- career intentions and preferences in general. erence style items are included, the scales are These items were developed from prior studies bipolar, so for example, students are asked to of entrepreneurial attitudes and traits of those what extent they would prefer working as self- in higher education to ensure that the items employed or working for someone else (1 were contextually suitable. Initially, a pilot was would definitely prefer to be employed by conducted with a group of nine volunteer post- someone else whereas 7 would definitely graduate students in order to ensure that the prefer to be self-employed). Given the ordinal wording of items were suitable and identify any nature of these measures and the relatively problems associated with the completion of the small subsample sizes, comparisons are made questionnaire. After completing the question- using Mann–Whitney nonparametric tests, naire, the students provided feedback to one which are the equivalent of the parametric of the project team members, and a number of t-tests used with continuous data. Where com- minor changes were made to the wording of parisons are made between the scores given by some items to provide clarity. Some additional the same individuals on different items, items representing other aspects of the course Wilcoxon rank sum tests are applied. that students felt were important were added As well as using items associated with atti- where previously absent. tudes toward entrepreneurship for consistency The questionnaire was administered to with the TPB, those capturing social norms and all students studying business and enter- PBC are also examined. However, as there is no prise courses across all years including both consensus of what constitutes entrepreneur- undergraduates and postgraduates. Identically ship, a selection of items is included to deter- worded online or paper versions were available. mine the extent to which the students agree that

KWONG AND THOMPSON THOMPSON 305 7 Table 1 Characteristics of Sample

Male Female All

Gender 57.6 percent 42.4 percent 139 Under 21 Years of Age 52.5 percent 54.2 percent 74 Chi-Square 0.041 [1] (.839) First Year of Undergraduate Study 38.8 percent 39.0 percent 54 Second Year of Undergraduate Study 22.5 percent 28.8 percent 35 Third Year of Undergraduate Study 12.5 percent 16.9 percent 20 Masters or Other Postgraduate Studies 26.3 percent 15.3 percent 30 Chi-Square 2.907 [3] (.406) Parents Started a Business 53.8 percent 61.0 percent 139 Chi-Square 0.731 [1] (.393) n 80 59 139

Degrees of freedom are shown in squared brackets and p-values in parentheses. these activities constitute entrepreneurship. Once identified, the different groups of students Figure 2 are also compared in terms of their preferences Time Scale of Becoming an for different work roles and characteristics relating to work such as perceptions of leadership Entrepreneur abilities. In the case of those variables related to the TPB, these may be interrelated (Ajzen 1991). In order to accommodate this, a multiple analysis of variance (MANOVA) approach is used to supplement the bivariate Mann–Whitney analysis. Entrepreneurial Time Scales of Potential Entrepreneurs Given the nature of the courses studied by the students, it is of no surprise that a majority are male (57.6 percent); however, there are no significant differences between the male and female students in terms of their ages and stage of study. Half the sample is aged between 18 and 21 years, with a further 40 percent in the 21 to 25 years category. As might be expected, preneurs rapidly after graduation (within the for students taking business and enterprise next three years), consistent with other studies courses, many have a strong entrepreneurial (Galloway and Brown 2002; Henley et al. background with three-fifths of the students 2009), a majority of those who see themselves claiming that their parents had at some point becoming entrepreneurs have a much longer started a business of their own (Table 1). time span in mind. Nearly a quarter expected to Figure 2 shows the distribution of students become entrepreneurs only after at least 10 indicating their expected time scale for entre- years. This shows why questions in studies that preneurial activities. As with the other items have asked students to specify which career discussed earlier, no significant difference was path they expect to follow on graduation find a found between the genders. much lower preference for entrepreneurship The results clearly show that although 40 than items just capturing interest (Armitage and percent of the students intend to become entre- Conner 2001).

8306 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Table 2 Agreement that Activities Represent Entrepreneurship by Intended Date of Involvement

Doubtful Entrepreneurs- Potential Rapid Entrepreneurs in-Waiting Entrepreneurs

Innovation Exploration Activities Inventors Bring New Products to Market 5.8 6.4b 5.7c R&D Activity in Large Firmsa 4.6 4.7 4.7 University Research 4.1 4.1 4.5 Market Exploitation Activities Providing Goods to Those in Deprived 4.9 4.6 4.9 Areas Opening a Shop 4.3 4.3 3.8 Cost Reduction Activity Managers Reducing Costs in Large Firms 3.7 4.1 4.4b Organizational Restructuring Activity Corporate Mergers and Takeovers 4.6 4.9 4.6 aR&D, research and development. bMann–Whitney tests indicate a signiﬁcant difference at the 5 percent level with those expecting to take over 10 years to become an entrepreneur (doubtful entrepreneurs). cSigniﬁcant difference with those expecting to take between 3 and 10 years to start a business (entrepreneurs-in-waiting).

The Mann–Whitney tests indicate that there of their ability to make their entrepreneurial are few significant differences between the experience a success, thus confirming H1b. groups in terms of what they class as an entre- However, there is no evidence of H1a that such preneurial activity (Table 2). Entrepreneurs-in- confidence came from the skills and knowledge waiting are found to have the strongest feelings that students felt they possessed, with no sig- regarding inventors bringing new products to nificant difference found between the groups. market, consistent with H4b, and may therefore Although this group may have possessed develop more innovative ventures (Capelleras slightly greater entrepreneurial experience as et al. 2010). At the same time, potential rapid 35.4 percent of the potential rapid entrepre- entrepreneurs are more likely to perceive cost- neurs were postgraduates compared with only cutting measures as entrepreneurial activities 15.9 percent of the entrepreneurs-in-waiting than doubtful entrepreneurs. Potential rapid (chi-square 4.046, p-value .033). Potential rapid entrepreneurs, however, still indicate greater entrepreneurs were also more likely to have agreement that new product commercialization parents who started businesses than the was an entrepreneurial activity than cost reduc- entrepreneurs-in-waiting (77.1 percent com- tion (Wilcoxon = 3.970, p-value = .000). pared with 54.5 percent, chi-square 5.219, The remainder of the analysis largely con- p-value .022). This means that although these centrates on the two groups intending to start individuals are by and large relatively inexpe- within the next 10 years, as the final group, as rienced themselves, they have potentially is shown earlier, is the group which exhibits strong role models from their parents, allied the least preference for an entrepreneurial with a higher level of formal education. career. As Ajzen (1991) suggests the different constructs in the TPB are likely to be in part PBC interdependent, it is reasonable to examine The results presented in Table 3 suggest that these variables together using MANOVA analy- potential rapid entrepreneurs are more certain sis, to allow for any correlation (Table 4). Given

KWONG AND THOMPSON THOMPSON 307 9 Table 3 Perceived Behavioral Control, Attitude toward Entrepreneurship, and Social Norms, by Intended Data of Involvement

Entrepreneurs- Potential Rapid U-Test p-Value in-Waiting Entrepreneurs

Perceived Behavioral Control Perceived Certainty of Success of 4.6 5.1 −1.980 (.048) Start-Up Perception of Knowledge Required for 2.9 3.3 −1.162 (.245) Start-Up Perceived Difﬁculty of Starting a 2.7 3.0 −0.923 (.356) Business Attitude toward Entrepreneurship Preference for Self-Employment 5.5 5.7 −0.542 (.588) Social Norms Important Others Support 4.7 5.2 −1.461 (.144) Self-Employment Importance of Others’ Opinions 4.3 4.0 −0.679 (.497)

that these variables may also be influenced by greater inclination for acquiring skills working the level of study, this is included as an addi- for others first before becoming entrepreneurs tional factor alongside the type of latent entre- (Table 5). Once an entrepreneurial career has preneur. No significant relationship was found been started, neither group shows a greater between level of study and type of latent entre- preference than the other for serial or portfo- preneur, so it was possible for both to enter as lio entrepreneurship. In order to establish the independent factors. Interestingly, only the reason(s) for such differences, the analysis type of entrepreneur was found to have a sig- now turns to the factors behind career choice nificant influence on the TPB variables. The decisions. largest differences are clearly between the One explanation for the timing difference is doubtful entrepreneurs and the others, but the the desire to control expressed by the individu- contrasts do weakly confirm the findings of the als (Stavrou, Kleanthous, and Anastasiou 2005). bivariate analysis in Table 3. A significant inter- The results in Table 6 suggest that rapid entre- action is found with rapid entrepreneurs in preneurs are more likely to perceive them- their last year of undergraduate study more selves as leaders (H2c), although there is no likely to feel they have the knowledge evidence that they have substantially greater required. This is not found for rapid entrepre- confidence in their leadership skills or enjoy neurs undertaking postgraduate study. being in such a position. This means that there is little evidence for H2d that potential rapid Attitude toward Entrepreneurship entrepreneurs seek out responsibility and It is found that both potential rapid entre- control of others. These results may reflect a preneurs and entrepreneurs-in-waiting display degree of modesty as they do indicate that they a strong preference for self-employment naturally tend to be selected by others or fate to (Table 3), with no significant difference be in these positions. Another explanation for found between the two groups (confirming the difference in timing is because of the value H2a). Although the MANOVA results indicate individuals placed on stress and responsibili- that the entrepreneurs-in-waiting and potential ties. Though both groups do not place a great rapid entrepreneurs do show a greater deal of importance on avoiding responsibility preference for entrepreneurship than doubtful and only moderate importance on avoiding entrepreneurs (Table 4). Understandably, how- stress (Bluedorn and Martin 2008), the desires ever, entrepreneurs-in-waiting do show a to avoid responsibility and stress are more

10308 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Table 4 Multiple Analysis of Variance (MANOVA) Analysis of Variables Relating to the Theory of Planned Behaviora

Level of Contrasts Type of Latent Contrasts Interactiona Levene’s/Box Studya Entrepreneurb Testc WN N HMSN309 THOMPSON AND KWONG WN N HMSN11 THOMPSON AND KWONG Level 1 versus Level 2 versus Level 3 versus Doubtful versus In Waiting Above Above Masters Others versus Rapid

Perceived Certainty of 0.860 (.464) −0.256 (.310) 0.206 (.535) 0.432 (.331) 5.122 (.007) −0.781 (.010) −0.595 (.063) 0.622 (.712) 1.500 (.140) Success of Start-Up Perception of Knowledge 0.286 (.835) 0.015 (.956) −0.018 (.960) −0.434 (.365) 0.868 (.422) −0.167 (.603) −0.417 (.225) 1.899 (.086) 0.863 (.578) Required for Start-Up Perceived Difﬁculty of 1.133 (.338) 0.369 (.112) −0.108 (.722) 0.229 (.573) 2.505 (.086) −0.518 (.060) −0.355 (.224) 0.930 (.476) 0.906 (.537) Starting a Business Preference for 0.843 (.473) −0.314 (.345) −0.565 (.196) −0.146 (.803) 9.978 (.000) −1.731 (.000) −0.276 (.509) 0.624 (.711) 1.172 (.313) Self-Employment Important Others Support 0.017 (.997) −0.017 (.954) −0.019 (.961) −0.107 (.840) 7.576 (.001) −1.247 (.001) −0.646 (.089) 0.620 (.714) 0.715 (.723) Self-Employment Importance of Others’ 0.839 (.475) −0.377 (.262) −0.384 (.383) 0.624 (.290) 0.033 (.967) −0.083 (.834) 0.062 (.884) 1.577 (.159) 0.751 (.688) Opinions Overall—Multivariate 0.853 (.636) 2.754 (.002) 1.056 (.383) 0.961 (.625)

ap-values are in parentheses. bF-tests based on the Pillai-Bartlett trace. cTests of group variance homogeneity. Table 5 Personal Estimation of Likelihood toward Entrepreneurship and Employment

Entrepreneurs- Potential Rapid U-Test p-Value in-Waiting Entrepreneurs

Likelihood of Pursuing a Career 5.3 6.1 −2.977 (.003) as Self-Employed Likelihood of Pursuing a Career 5.0 4.3 −2.001 (.045) as an Employee Likelihood of Working for 5.3 4.5 −2.239 (.025) Others First before Becoming Self-Employed Likelihood of Spending the 4.2 4.8 −1.530 (.126) Entire Career on Different Entrepreneurial Ventures

important for the entrepreneurs-in-waiting than to innovative activities. These findings match the potential rapid entrepreneurs (thus con- with studies of barriers to entrepreneurship firming H2e). Clearly, the potential rapid entre- experienced by alumni. For example, a lack of preneurs have a strong image or vision they viable ideas (Carter and Collinson 1999) and a wish to complete and are more willing to lack of security (Rae and Woodier 2006) have accept some stress to accomplish it. been identified as reasons for delaying start-up activity. It is understandable that those not Social Norms intending to break new ground and less worried The results provide little evidence to support about the stress associated with business own- either H3a or H3b, with no significant differ- ership will be those that are more likely to take ences in the extent that two groups of entre- the plunge relatively rapidly. Entrepreneurship preneurs feel they have the support of others courses in the United Kingdom are focused and the degree they care about this support more on business students rather than engineers (Table 3). Table 6 also finds that the two and scientists who are perhaps more likely to groups display minimal differences in their create innovative products (Levie 2009). Bring- preferences for participating in a social envi- ing nonbusiness students into the courses may ronment. The contrasts in Table 4 did, be of great value to potential rapid entrepre- however, provide weak evidence that potential neurs as their desire for business ownership can rapid entrepreneurs did feel they had more be linked to those who are perhaps less com- support than entrepreneurs-in-waiting (H3a). mercially minded but have the potential to generate innovations with commercial potential Discussion (Thursby 2005). Consistent with previous literature (Carter For those that choose to delay entry into and Collinson 1999; Donckels 1991), our study entrepreneurship, the results suggest that there found no considerable attitudinal differences is still a distinct preference over working for between rapid entrepreneurs and entrepreneur- others, but entrepreneurship is part of a career in-waiting in terms of their desire to start a planned over a longer period. However, for business. Although the motivations for and the entrepreneurs-in-waiting, fulfilling these ambi- form that these start-ups will take do appear to tions of starting a business requires entrepre- differ. For potential rapid entrepreneurs, they neurial aspirations to be sustained beyond are less worried about avoiding stress and university. Much of this choice to delay seems responsibility and entrepreneurship is poten- to be associated with a third barrier found in tially more closely associated with business studies of entrepreneurial activities of alumni— ownership in general and less strongly restricted that of a desire to acquire more skills (Matlay

12310 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Table 6 Factors’ Importance in Choosing Future Career Path by Intended Date of Involvement

Entrepreneurs- Potential Rapid U-Test p-Value in-Waiting Entrepreneurs

Leadership I Enjoy Having Authority over Other 4.9 5.2 −1.286 (.198) People Perceive Themselves to Usually Be 4.2 4.9 −2.093 (.036) Leaders Perception of Capability to Be a Good 5.7 6.0 −1.425 (.154) Leader Stress and Responsibilities I Prefer Having a Non-Stressful Job 4.5 3.6 −2.398 (.016) I Prefer Not Taking on Too Much 3.2 2.7 −1.808 (.071) Responsibility I Prefer Not Working Long Hours 4.5 3.7 −1.972 (.049) Financial I Prefer Job Security and Stability 5.4 5.1 −0.721 (.471) I Prefer Pay Based on Performance 5.6 5.7 −0.210 (.833) and Effort I Prefer to Keep a Large Portion of 5.4 5.1 −0.905 (.366) Proﬁts I Prefer Being Able to Build Great 5.9 6.2 −1.383 (.167) Wealth Need for Achievement I Prefer Having Opportunities for 6.3 6.4 −0.009 (.993) Career Progression and Promotion I Prefer Having a Challenging and 6.4 6.4 −0.149 (.882) Interesting Job I Prefer Being Able to Achieve 6.2 6.1 −0.810 (.418) Something and Get Recognition Social I Prefer to Participate in a Social 5.4 5.1 −1.142 (.254) Environment

2008). Though our study found no difference in increase the probability of success. As such, terms of the skills and knowledge possessed by perceived deﬁciencies can be due to the actual both rapid entrepreneurs and entrepreneurs-in- lack of technical skills, but equally it can be due waiting, the potential rapid entrepreneurs were to lack of practical know-how (Matlay 2008; more conﬁdent of succeeding if they were to Rae and Woodier 2006). Studies have criticized start a new venture. It is impossible to deter- the impractical, “bums-on-seats” approach of mine which group of aspiring entrepreneurs many of the more traditional management edu- was incorrect as the skills required will vary by cation program, which do not enable students type of start-up instigated. In addition, though to connect the different competencies required some studies suggest that the most relevant to start a business together in a meaningful skills and knowledge are only likely to come manner (Matlay 2008). Concerns of entrepre- from experience of business ownership (Politis neurial alumni in relation to a lack of skills can 2005), there are likely to be some resources, be attributed frequently to a lack of actual work which can be acquired in preparation to or entrepreneurship experience (Carter and

KWONG AND THOMPSON THOMPSON 311 13 Collinson 1999). Working for others helps accu- introduction of innovative methods to develop mulate this missing experience and allows skills and experience including the use of long- them to place the knowledge they gained from term apprenticeships has been strongly encour- their formal management and entrepreneurship aged (Aronsson 2004). Indeed, there is training at university. Without such experience, evidence of changes in delivery of training, alumni lacked the context of immediacy sur- from the more traditional approach between rounding these issues (Matlay 2008). According 1995 and 1999 to a more mixed approach to these entrepreneurial alumni, this perceived between 2000 and 2004 with increased variety lack of skills is also related to a lack of access of courses and an increasing use of information to specialist support, guidance, and advice (Rae and communications technology and electronic and Woodier 2006; Smith and Beasley 2011). platforms within curriculum delivery (Matlay An alternative view is that in trying to generate and Carey 2007). New innovative and experi- creative enterprising individuals, a mythical mental programs have been developed, which image of the superhuman entrepreneur is aim to increase efficiency, relevance, and prac- created; those lacking “the next big idea” may tical value of entrepreneurship education on worry they cannot live up to this and delay offer (Kwong et al. 2012; Matlay and Carey involvement in pursuit of this unicorn 2007; Smith, Collins, and Hannon 2006). Many (Laukkanen 2000). The further constraint that of these programs aim to encourage interaction might be overcome with time spent working for with entrepreneurs with greater emphasis on others is a lack of finance (Carter and Collinson participation, responsibility, and decision- 1999; Smith and Beasley 2011). making. There is also increasing used of syn- On average, both groups suggested that ergistic learning, which focuses on learning important others were mildly supportive of through cooperation, co-learning, consultation, careers in self-employment, but not greatly so. and collective action. Although important others are likely to relate The practical difficulty, however, is to largely to family members rather than univer- sustain the interest of entrepreneurial alumni sity staff (Henderson and Robertson 1999), tra- and to carry their interest from university to a ditionally the lack of emphasis placed on work environment. After working for a few networking opportunities in many courses, years, such entrepreneurial aspirations may be which restricts this group of important others lost as a result of a change in circumstances to individuals with less direct knowledge and (Kwong et al. 2012). Though many universities information on entrepreneurship, may have provide some form of enterprise training for been a cause (Garavan and O’Cinneide 1994). their student population, relatively few have Though opportunities to network with active considered extending this provision to their entrepreneurs are now generally incorporated alumni community (Carter and Collinson within courses, perhaps there are still opportu- 1999). Such an approach often neglected nities to increase this further. End-of-year dis- the “entrepreneurs-in-waiting” type graduates semination events with invitations to parents whose aspirations need to be continuingly and other family members to join the audience updated and regenerated beyond university. could increase these social norms. Studies have thus argued for the introduction of The findings mentioned earlier highlight the continuing post-experience education for challenges faced by universities in preparing alumni (Donckels 1991), most notably on pro- graduates for an entrepreneurial career. Studies viding a more practical grounding for gradu- found that alumni would like to receive more ates, including financial management and vocational-orientated, technical-based training business communications skills, to help cope while studying at university (Carter and with the transition and the often hazy division Collinson 1999; Donckels 1991). These include between employment and self-employment a “portfolio of entrepreneurial skills” to help (Carter and Collinson 1999; Matlay 2008). It is manage a long-term entrepreneurial career, found that nearly two-thirds of entrepreneurial including financial management, particularly an alumni would like to attend short courses on understanding of the balance sheet, business financing business start-ups and also on busi- communication, and other business start-up ness planning, whereas half would like to skills such as evaluating a business idea and to receive special training (Carter and Collinson draw up a business plan (Carter and Collinson 1999). However, universities should also take 1999; Donckels 1991). At the same time, the note of the very different requirements of these

14312 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT alumni compared with the traditional student provide social support for entrepreneurs-in- catchments. Most of these “entrepreneurs-in- waiting, which may be just as important in waiting” are working for others in order to encouraging entrepreneurial activity among accumulate experience, develop professional alumni (Carter and Collinson 1999) to create a networks, and raise finance. This means they community that includes access to support in require a very different educational provision, more practical terms, including free access to most notably their preference for courses libraries, specialized scientific equipment, and outside the normal business hours including staff consultation (Carter and Collinson 1999). the evenings and weekends (Carter and Such support could be just as important for Collinson 1999). Alternatively, these courses potential rapid entrepreneurs, because as noted can be delivered in blocks, mimicking other earlier, although they are more confident of executive education programs such as the success, it is not clear that such confidence is Master of Business Administration (Nixon, any more justified and without the professional Helms, and Fletcher 1997). Studies have also networks and support the entrepreneurs-in- found that these entrepreneurial alumni also waiting may have built up, the university may prefer the more flexible multimedia delivery have a key role to provide in fulfilling these approach, including the extensive use of online needs rather than the very basic services often delivery (Carter and Collinson 1999). One pos- provided in incubators such as photocopying sible way to connect these experiences with and conference suites (Chell and Allman 2003). continuous study is through a degree in work- based learning. Such a degree would require Conclusions alumni to create their personal development This paper has examined the entrepreneurial plan. This fits with calls for employees, employ- intentions of business students at a U.K. higher ers, and educational establishments to engage education establishment, with particular regard with such activities to help individuals to iden- paid to the timeframe within which students tify knowledge and experience deficiencies, intend to become entrepreneurs. As found in and to attend courses and events organized previous studies, the students displayed strong both by the universities and elsewhere in order desires and considerable preferences for entre- to address such deficiencies (Rodrigues 2006). preneurial careers, but when the issue of timing Though work-based learning degrees are was considered, most students were not becoming increasingly popular (Raelin 1997), looking for rapid involvement. In fact, the most an entrepreneurship stream of such a degree favored path to entrepreneurship was to work can be created to allow alumni to follow a for others first and then become an entrepre- specific pattern that is likely to enable them to neur at a later stage. However, a considerable accumulate the knowledge and experience group of students did intend to become entre- required to start a business. The development preneurs within the next three years. Two main of a personal development plan would enable groups of students positively disposed to entre- alumni to keep track of the additional training preneurial careers were identified. The first required and their current level of such train- preferred fairly immediate engagement upon ing, and help them identify what relevant train- graduation, so were “potential rapid entrepre- ing is provided. It is hoped that through such a neurs,” the others wanted to work for others course of study, the alumni would continuously for 3 to 10 years being rather “entrepreneurs- refresh their entrepreneurial aspirations, and at in-waiting.” Though both groups are equally the same time, develop a portfolio of skills that enthusiastic about starting a business, there are would enable them to start their own business some notable differences between them. Our in the long run. As studies found that some study found that the main difference is not in alumni would prefer such training to be accred- terms of perceived capability, but attitudinal. ited (Carter and Collinson 1999), a degree in Despite being slightly more experienced in work-based learning would also enable them to terms of parental role models and level of achieve such qualifications while working qualifications being studied for, our study toward starting a business. found minimal evidence that potential rapid The study also found that there was some entrepreneurs are more skillful or perceive evidence that potential rapid entrepreneurs felt fewer problems relating to start-up than those that they had greater support. Once in the who “wait and see.” Despite this, potential workplace, it is possible that universities can rapid entrepreneurs are more certain of their

KWONG AND THOMPSON THOMPSON 313 15 ability to succeed than those who prefer to For potential rapid entrepreneurs, the availabil- wait. When examining their attitude toward ity of incubators attached to universities could starting a business, it is found that potential provide access to trusted advisors in the form of rapid entrepreneurs are more likely to feel that their university tutors, which will help to over- they are naturally selected as leaders. There are come some of the problems of inexperience also signs that potential rapid entrepreneurs (Chell and Allman 2003; Rodrigues 2006). For were driven to entrepreneurship by their the entrepreneurs-in-waiting, refresher courses dislike of employment. Such a desire to avoid may help reignite entrepreneurial aspirations employment is so strong that they feel they are (Carter and Collinson 1999), but perhaps there is obliged to take on more stress and responsi- no need for entrepreneurs to leave their employ- bilities in order to start up a business within a ers, with intrapreneurship and eventually relatively short timeframe. However, it is spinout companies offering a method of tapping unclear whether this greater confidence of into their entrepreneurial potential. It is there- potential rapid entrepreneurs can be justified. fore important that an entrepreneurial environ- One potential danger of plunging straightaway ment is created at the workplace that would into entrepreneurship is the focus of short-term allow for creative and innovative practices to be cost reduction practices rather than boundary undertaken, but also that enterprise educators spanning innovative activities, which many ensure their courses are relevant (and seen argued would hinder the growth potential of to be by students) for both corporate and small the business in the long run (Capelleras et al. and medium-sized enterprise environments 2010). On the other hand, our findings suggest (Heinonen 2007), for example, providing stu- that entrepreneurs-in-waiting place more value dents with an understanding of the nature and on acquiring the skills and resources they need skills required by internal spinoff structures of under others, perhaps reflecting the different existing enterprises. understanding of what constitutes entrepre- With regard to those who prefer to “wait and neurship where innovation and the develop- see,” our study also argues that, though the ment of new products are emphasized to a number of entrepreneurship courses being run greater extent. in universities has increased greatly in the past The results of the paper show the difficult 20 years (Kuratko 2005), the embracement of a balancing act that those providing enterprise “go-out-and-do-it-now” approach in most of education face. In order to create more graduate these courses alienates those who prefer to entrepreneurs, it is essential that positive take a more cautious approach toward entre- student attitudes are created. Those looking to preneurship. Instead, we urge those designing become entrepreneurs, as compared with those and running enterprise courses to adapt their for whom an entrepreneurial career is doubtful, courses to also cater for their needs by provid- show a greater preference for working for them- ing continuous support until they feel ready to selves. There is also greater confidence that an start a business. As those who “wait and see” entrepreneurial career will be pursued at some are less confident of their probability of suc- point where potential start-up initiation is ceeding, enterprise education needs to be expected in the near future. This means even taught in a way that does not scare students before the end of their studies, students them- away from the pursuit of entrepreneurial activi- selves are aware that if not moving into entre- ties to avoid diminishing students’ intentions of preneurship almost immediately, the probability becoming entrepreneurs (Shepherd 2004). that any entrepreneurial ambitions will be The study is limited by the depth to which fulfilled declines substantially (Carter and the decisions of students with regard to the Collinson 1999). At the same time, it is essential career paths can be examined. Qualitative that overconfidence is not generated, otherwise follow-up studies will help provide a greater those less prepared may enter entrepreneurship understanding of why students favor differ before they have the full set of skills that they approaches or why they have greater PBC will require. Though it is possible that as serial when at first it appears they may not have entrepreneurs a new venture failure will provide the required experience. The findings are of a good learning experience, this is by no means course based around a single group of students certain (March 1991; Shepherd 2003). studying on courses in a single higher educa- Resources already available in many univer- tion establishment in the United Kingdom. sities may be the answer to some of these issues. Comparative studies in other institutions and

16314 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT cultures would be required to confirm the find- Bird, B. (1988). “Implementing Entrepreneurial ings or determine whether factors such as the Ideas: The Case for Intentions,” Academy of courses studied or the backgrounds of the stu- Management Review 13(3), 442–453. dents generate the results found here. Like ——— (1992). “The Operation of Intentions in most studies of entrepreneurship, the biggest Time: The Emergence of the New Venture,” limitation of the study is the cross-sectional Entrepreneurship: Theory and Practice nature of the data. As noted in the introduction 17(1), 11–20. to this paper, recalled alumni experiences may Bird, B., and L. Schjoedt (2009). “Entrepreneur- not accurately reflect their choices made at ial Behavior: Its Nature, Scope, Recent earlier stages, but actual outcomes and their Research, and Agenda for Future Research,” reasons are just as important element to study. in Understanding the Entrepreneurial Mind. However, as noted by others, such as Chell and Eds. A. L. Carsrud and M. Brannback. New Allman (2003), to best understand the choices York, NY: Springer, 327–358. made and the outcomes of these choices, a Bird, B., and P. West (1997). “Time and Entre- longitudinal approach is more appropriate, and preneurship,” Entrepreneurship: Theory and it is only with such studies that a real under- Practice 22(2), 5–10. standing of the impact of entrepreneurship Blanchflower, D. G., and A. J. Oswald (1998). education can be truly established. “What Makes an Entrepreneur?,” Journal of Labor Economics 16(1), 26–60. References Bluedorn, A., and G. Martin (2008). “The Time Acs, Z. J., and D. B. Audretsch (2003). “Inno- Frames of Entrepreneurs,” Journal of Busi- vation and Technological Change,” in Hand- ness Venturing 23(1), 1–20. book of Entrepreneurship Research. Eds. Z. J. Boussouara, M., and D. Deakins (1999). Acs and D. B. Audretsch. Boston, MA: “Market-Based Learning, Entrepreneurship Kluwer Academic Publishers, 55–79. and the High Technology Small Firm,” Inter- Ajzen, I. (1991). “The Theory of Planned national Journal of Entrepreneurial Behavior,” Organizational Behavior and Behaviour and Research 5(4), 204–223. Human Decision Processes 50(2), 179–211. Boyd, N. G., and G. S. Vozikis (1994). “The Ajzen, I., and M. Fishbein (1980). Understand- Influence of Self Efficacy on the Develop- ing Attitudes and Predicting Social Behav- ment of Entrepreneurial Intentions and ior. Englewood Cliffs, NJ: Prentice-Hall. Actions,” Entrepreneurship: Theory and Allen, W. D. (2000). “Social Networks and Self- Practice 18(4), 63–77. Employment,” Journal of Socio-Economics Brockhaus, R. H., and P. S. Horwitz (1986). 29(5), 487–501. “The Psychology of the Entrepreneur,” in Armitage, C. J., and M. Conner (2001). “Efficacy The Art and Science of Entrepreneurship. of the Theory of Planned Behaviour: A Meta- Eds. D. L. Sexton and R. W. Smilor. Cam- Analytic Review,” British Journal of Social bridge, MA: Ballinger Publishing Company, Psychology 40(4), 471–499. 25–48. Aronsson, M. (2004). “Education Matters—But Brown, R. (1990). “Encouraging Enterprise: Does Entrepreneurship Education? An Inter- Britain’s Graduate Enterprise Program,” view with David Birch,” Academy of Man- Journal of Small Business Management agement Learning and Education 3(3), 289– 28(4), 71–77. 292. Capelleras, J., F. Greene, H. Kantis, and R. Ashworth, J., P. Johnson, and C. Conway Rabetino (2010). “Venture Growth Speed (1998). “How Good Are Small Firms at Pre- and Subsequent Growth: Evidence from dicting Employment?,” Small Business Eco- South America,” Journal of Small Business nomics 10(4), 379–387. Management 48(3), 302–324. Audet, J. (2004). “A Longitudinal Study of Carsrud, A., and M. Brannback (2011). “Entre- Entrepreneurial Intention of University Stu- preneurial Motivation: What Do We Still dents,” Academy of Entrepreneurship Need to Know?,” Journal of Small Business Journal 10(1), 3–15. Management 49(1), 9–26. Audretsch, D. B., M. C. Keilbach, and E. E. Carter, N. M., W. B. Gartner, and P. D. Reynolds Lehmann (2006). Entrepreneurship and Eco- (1996). “Exploring Start-Up Event nomic Growth. New York, NY: Oxford Uni- Sequences,” Journal of Business Venturing versity Press. 11(3), 151–166.

KWONG AND THOMPSON THOMPSON 315 17 Carter, S., and E. Collinson (1999). “Entrepre- Eisenhardt, K., and L. J. Bourgeois (1988). neurship Education: Alumni Perceptions of “Politics of Strategic Decision Making in the Role of Higher Education Institutions,” High Velocity Environments: Toward a Mid- Journal of Small Business and Enterprise range Theory,” Academy of Management Development 6(3), 229–239. Journal 31(4), 737–770. Chell, E., and K. Allman (2003). “Mapping the Evans, D. S., and B. Jovanovic (1989). “An Motivations and Intentions of Technology Estimated Model of Entrepreneurial Choice Orientated Entrepreneurs,” R&D Manage- Under Liquidity Constraints,” Journal of ment 33(2), 117–134. Political Economy 97(4), 808–827. Choi, R., M. Levesque, and D. Shepherd (2008). Fleming, P. (1996). “Entrepreneurship Educa- “When Should Entrepreneurs Expedite or tion in Ireland: A Longitudinal Study,” Delay Opportunity Exploitation?,” Journal of Academy of Entrepreneurship Journal 2(1), Business Venturing 23(3), 333–355. 94–118. Cogliser, C., and K. Brigham (2004). “The Inter- Galloway, L., and W. Brown (2002). “Entrepre- section of Leadership and Entrepreneurship: neurship Education at University: A Driver in Mutual Lessons to be Learned,” Leadership the Creation of High Growth Firms?,” Edu- Quarterly 15(6), 771–799. cation and Training 44(8/9), 398–405. Collins, L., P. D. Hannon, and A. Smith (2004). Galloway, L., W. Brown, M. Anderson, and L. “Enacting Entrepreneurial Intent: The Gaps Wilson (2006). “Investigating the Potentials Between Student Needs and Higher Educa- of Entrepreneurship Education,” Interna- tion Capability,” Education and Training tional Journal of Management Education 46(8/9), 454–463. 5(1), 57–65. Cooper, A. C., C. Y. Woo, and W. C. Dunkelberg Garavan, T. N., and B. O’Cinneide (1994). (1988). “Entrepreneurs’ Chances for “Entrepreneurship Education and Training Success,” Journal of Business Venturing Programmes: A Review and Evaluation— 3(2), 97–108. Part 1,” Journal of European Industrial Cromie, S., S. Birley, and I. Callaghan (1993). Training 18(8), 3–12. “Community Brokers: Their Role in the For- Gilbert, B. A., D. B. Audretsch, and P. P. mation and Development of Business Ven- McDougall (2004). “The Emergence of Entre- tures,” Entrepreneurship and Regional preneurship Policy,” Small Business Eco- Development 5(3), 247–264. nomics 22(3/4), 313–323. Das, T. (1987). “Strategic Planning and Indi- Heinonen, J. (2007). “An Entrepreneurial- vidual Temporal Orientation,” Strategic Directed Approach to Teaching Corporate Management Journal 8(2), 203–209. Entrepreneurship at University Level,” Edu- De Noble, A., D. Jung, and S. Ehrlich (1999). cation and Training 49(4), 310–324. “Entrepreneurial Self-Efﬁcacy: The Develop- Henderson, R., and M. Robertson (1999). “Who ment of a Measure and its Relationship to Wants to Be an Entrepreneur? Young Adult Entrepreneurial Action,” in Frontiers of Attitudes to Entrepreneurship as a Career,” Entrepreneurial Research. Eds. P. Reynolds, Education and Training 41(5), 236–245. W. Bygrave, S. Manigart, C. Mason, G. Henley, A., C. DeCock, P. Latreille, C. Dawson, Meyer, H. Sapienza, and K. Shaver. and I. Humphreys (2009). Entrepreneurial Waltham, MA: P&R Publications, 73–87. Aspirations and Activity Amongst Students: Donckels, R. (1991). “Education and Entrepre- A Comparative Study for Wales, Final Report neurship Experiences from Secondary and for the Welsh Assembly Government. University Education in Belgium,” Journal of Swansea: Swansea University. Small Business and Entrepreneurship 9(1), Ismail, M., S. A. Khalid, M. Othman, K. Jusoff, 35–42. N. Abdul Rahman, K. M. Mohammed, and R. Douglas, E., and J. Fitzsimmons (2008). “Indi- Z. Shekh (2009). “Entrepreneurial Intention vidual Intentions Towards Entrepreneurship among Malaysian Undergraduates,” Interna- vs. Intrapreneurship,” paper presented at tional Journal of Business and Management 5th ASGE International Entrepreneurship 4(10), 54–60. Research Exchange, Melbourne, February. Johannisson, B. (1991). “University Training for Eisenhardt, K. (1989). “Building Theories from Entrepreneurship: Swedish Approaches,” Case Study Research,” The Academy of Man- Entrepreneurship and Regional Develop- agement Review 14(4), 532–550. ment 3(1), 67–82.

18316 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Jovanovic, B. (1982). “Selection and the London: Department for Education and Evolution of Industry,” Econometrica: Employment. Journal of the Econometric Society 50(3), Ling, Y., Z. Simsek, M. H. Lubatkin, and J. F. 649–670. Veiga (2008). “Transformational Leader- Kantor, J. (1988). “Can Entrepreneurship Be ship’s Role in Promoting Corporate Entre- Taught?—A Canadian Experiment,” Journal preneurship: Examining the CEO–TMT of Small Business and Entrepreneurship Interface,” Academy of Management Journal 5(4), 12–19. 51(3), 557–576. Katz, J. (1994). “Modelling Entrepreneurial Lüthje, C., and N. Franke (2003). “The ‘Making’ Career Progressions: Concepts and Consid- of an Entrepreneur: Testing a Model of erations,” Entrepreneurship: Theory and Entrepreneurial Intent among Engineering Practice 19(2), 23–39. Students at MIT,” R&D Management 33(2), Katz, J. A. (2003). “The Chronology and Intel- 135–147. lectual Trajectory of American Entrepreneur- Macan, T. (1994). “Time Management: Test of a ship Education,” Journal of Business Process Model,” Journal of Applied Psychol- Venturing 18(2), 283–300. ogy 79(3), 381–391. ——— (2007). “Education and Training in March, J. G. (1991). “Exploration and Exploita- Entrepreneurship,” in The Psychology of tion in Organizational Learning,” Organiza- Entrepreneurship. Eds. J. R. Baum, M. Frese, tion Science 2(1), 71–87. and R. Baron. Mahwah, NJ: Lawrence Matlay, H. (2008). “The Impact of Entrepre- Erlbaum Associates, 209–235. neurship Education on Entrepreneurial Out- Kelly, L. M., N. Athanassiou, and W. F. comes,” Journal of Small Business and Crittenden (2000). “Founder Centrality and Enterprise Development 15(2), 382–396. Strategic Behavior in the Family-Owned Matlay, H., and C. Carey (2007). “Entrepreneur- Firm,” Entrepreneurship: Theory and Prac- ship Education in the UK: A Longitudinal tice 25(2), 27–42. Perspective,” Journal of Small Business Kets de Vries, M. (1993). “The Dynamics of and Enterprise Development 14(2), 252– Family Controlled Firms: The Good and the 263. Bad News,” Organizational Dynamics Nixon, J., M. Helms, and L. Fletcher (1997). 21(3), 59–72. “Integrating Team-Teaching, Technology Kristiansen, S., and N. Indarti (2004). “Entre- and Distance Learning in MBA Programmes: preneurial Intention among Indonesian and A Case Study,” Industrial and Commercial Norwegian Students,” Journal of Enterpris- Training 29(7), 218–225. ing Culture 12(1), 55–78. Parker, S. C. (2006). “Learning about the Krueger, N. F., M. D. Reilly, and A. L. Carsrud Unknown: How Fast Do Entrepreneurs (2000). “Competing Models of Entrepreneur- Adjust Their Beliefs?,” Journal of Business ial Intentions,” Journal of Business Ventur- Venturing 21(1), 1–26. ing 15(5/6), 411–432. Phan, P. H., P. K. Wong, and C. L. Wang (2002). Kuratko, D. F. (2005). “The Emergence of “Antecedents to Entrepreneurship among Entrepreneurship Education: Development, University Students in Singapore: Beliefs, Trends, and Challenges,” Entrepreneurship: Attitudes and Background,” Journal of Theory and Practice 29(5), 577–597. Enterprising Culture 10(2), 151–174. Kwong, C., P. Thompson, C. W.-M. Cheung, Politis, D. (2005). “The Process of Entrepre- and H. Manzoor (2012). “The Role of Envi- neurial Learning, A Conceptual Framework,” ronment in Fostering Conductive Entrepre- Entrepreneurship: Theory and Practice neurial Learning,” Journal of General 29(4), 399–424. Management 38(1), 45–71. Quinn, J. B. (1985). “Managing Innovation: Laukkanen, M. (2000). “Exploring Alternative Controlled Chaos,” Harvard Business Review Approaches in High-level Entrepreneurship May–June, 73–84. Education: Creating Micro-Mechanisms for Rae, D. (2005). “Entrepreneurial Learning: A Endogenous Regional Growth,” Entrepre- Narrative-Based Conceptual Model,” Journal neurship and Regional Development 12(1), of Small Business and Enterprise Develop- 25–47. ment 12(3), 323–335. Levie, J. (2009). Entrepreneurship Education in Rae, D., and N. Woodier (2006). “Graduate Higher Education in England: A Survey. Career Choices and Entrepreneurship,”

KWONG AND THOMPSON THOMPSON 317 19 Birmingham, Report for the National Council Smith, K., and M. Beasley (2011). “Graduate for Graduate Entrepreneurship. Entrepreneurs: Intentions, Barriers and Solu- Raelin, J. (1997). “A Model of Work-Based Learn- tions,” Education and Training 53(8/9), ing,” Organization Science 8(6), 563–578. 722–740. Reitan, B. (1996). “Entrepreneurial Intentions— Soloman, G. (2007). “An Examination of Entre- A Combined Models Approach,” paper pre- preneurship Education in the US,” Journal of sented at the 9th Nordic Small Business Small Business and Enterprise Development Research Conference, Lillehammer, May. 14(2), 168–182. Reynolds, P. (1994). “The Entrepreneurial Stavrou, E., T. Kleanthous, and T. Anastasiou Process: Preliminary Explorations in the (2005). “Leadership Personality and Firm U.S.,” paper presented at 1st Eurostate Inter- Culture during Hereditary Transitions in national Workshop on Techniques of Enter- Family Firms: Model Development and prise Panels, Luxembourg, February. Empirical Investigation,” Journal of Small Rodrigues, M. J. (2006). “The Lisbon Strategy Business Management 43(2), 187–206. after the Mid-term Review: Implications for Thursby, M. C. (2005). “Introducing Technol- Innovation and Life-Long Learning,” Corpo- ogy Entrepreneurship to Graduate Educa- rate Governance 6(4), 349–357. tion: An Integrative Approach,” Advances in Ronstadt, R. (1985). “The Educated Entrepre- the Study of Entrepreneurship, Innovation neurs: A New Era of Entrepreneurial Educa- and Economic Growth, 16, 211–240. tion Is Beginning,” American Journal of Traﬁmow, D., and K. A. Finlay (1996). “The Small Business 10, 7–23. Importance of Subjective Norms for a Rosa, P. (2003). “‘Hardly Likely to Make the Minority of People: Between Subjects Japanese Tremble’: The Businesses of and Within-Subjects Analyses,” Personality Recently Graduated University and College and Social Psychology Bulletin 22(8), 820– ‘Entrepreneurs’,” International Small Busi- 828. ness Journal 21(4), 435–459. van Geldren, M., M. Brand, M. van Praag, W. Shane, S. (2003). A General Theory of Entrepre- Bodewes, E. Poutsma, and A. van Gils neurship: The Individual-Opportunity (2006). “Explaining Entrepreneurial Inten- Nexus. Cheltenham, UK: Edward Elgar. tions by Means of the Theory of Planned Shapero, A. (2002). “The Displaced, Uncomfort- Behavior,” Career Development Interna- able Entrepreneur,” in Entrepreneurship: tional 13(6), 538–559. Critical Perspectives on Business and Man- van Stel, A., M. Carree, and R. Thurik (2005). agement. Ed. N. Krueger. London: “The Effect of Entrepreneurial Activity on Routledge, 251–259. National Economic Growth,” Small Business Shapero, A., and L. Sokol (1982). “Social Economics 24(3), 311–321. Dimensions of Entrepreneurship,” in The Vecchio, R. P. (2003). “Entrepreneurship and Encyclopedia of Entrepreneurship. Eds. C. Leadership: Common Trends and Common Kent, D. Sexton, and K. Vesper. Englewood Threads,” Human Resource Management Cliffs, NJ: Prentice-Hall, 72–90. Review 13(2), 303–328. Shepherd, D. A. (2003). “Learning from Busi- Vesper, K. H., and W. B. Gartner (1997). “Mea- ness Failure: Propositions of Grief Recovery suring Progress in Entrepreneurship Educa- for the Self-Employed,” Academy of Manage- tion,” Journal of Business Venturing 12(5), ment Review 28(2), 318–328. 403–421. ——— (2004). “Educating Entrepreneurship Volery, T., N. Doss, and T. Mazzarol (1997). Students about Emotion and Learning from “Triggers and Barriers Affecting Entrepre- Failure,” Academy of Management Learning neurial Intentionality: The Case of Western and Education 3(3), 274–287. Australian Nascent Entrepreneurs,” Journal Smith, A., L. Collins, and P. Hannon (2006). of Enterprising Culture 5(3), 273–291. “Embedding New Entrepreneurship Pro- West, P., and D. Meyer (1997). “Temporal grammes in UK Higher Education Institu- Dimensions of Opportunistic Change in tions: Challenges and Considerations,” Technology-Based Ventures,” Entrepreneur- Education and Training 48(8/9), 555–567. ship: Theory and Practice 22(2), 31–52.

20318 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Journal of Small Business Business Management Management 2016 2015 54(1), ••(••), pp. pp. 319–340 ••–•• doi: 10.1111/jsbm.12147 Entrepreneurs as Sophisticated Iconoclasts: Rational Rule-Breaking in an Experimental Game* by Richard J. Arend

We analyzed the theoretical relationship between rule-breaking by entrepreneurs and realized advantage. We tested the tenets of that relationship using an experiment where subjects, including entrepreneurs, compete in a business strategy game where rule-breaking is possible (but does not directly correlate with beneﬁts). We found that entrepreneurs break the rules more often and realize greater beneﬁts because they break rules in a smarter way. The ability to be sophisticated in rule-breaking behavior stems from breaking the rules earlier and responding more aggressively to positive feedback. Such actions can create new net value in the economy—reinforcing the idea of entrepreneurship as a social good.

Venkataraman 1997). In this article, we analyze Introduction whether those connections are true by using a Entrepreneurial innovation drives new value human subjects experiment that focuses on creation in modern economies. Larger gains rule-breaking in competitive contexts where through productivity and product variety come “entrepreneurs may break rules . . . and yet from discovery-focused, game-changing inno- their actions are accepted as what entrepre- vations rather than from incremental innova- neurs do” (Brenkert 2009, p. 448). tions, and these game-changing innovations are We summarize the relationships between connected to entrepreneurship because (1) the entrepreneurs and rule-breaking innovation as efforts of entrepreneurs are associated with follows. The entrepreneur breaks the existing radical change whereas incumbents are associ- rules in order to enter a market and to create a ated with incremental change; (2) entrepre- window of opportunity in which to succeed. neurial efforts must focus on breaking existing The entrepreneur breaks the rules shrewdly, rules to gain an advantage over incumbents; for example, by creating new value for custom- and (3) entrepreneurs must make such efforts ers in order to lure them away from the incum- rationally, that is, with the expectation of real- bent by putting a differentiated technological izable beneﬁts compensating for the risks (e.g., discovery to use in a new product offering. For Henricks 2005; Lim, Garnsey, and Gregory all this effort and risk—that is, in making a 2006; Schumpeter 1950; Stock, Greis, and discovery and commercializing it—the entre- Fischer 2002; Tushman and Anderson 1986; preneur expects a compensating reward. We

*We are grateful to the associate editor and the anonymous reviewers for their valued input in the revisions of this paper. All errors remain our own. Richard J. Arend is Professor of Strategy and Entrepreneurship at the Henry W. Bloch School of Management, University of Missouri—Kansas City. Address correspondence to: Richard J. Arend, Bloch School of Management, University of Missouri— Kansas CIty, 5100 Rockhill Road—303 Bloch School, Kansas City, MO 64110-2499. E-mail: [email protected].

AREND 319 1 explore these relationships in an experiment We chose to examine our research question where business students play a well-defined, using a laboratory experiment for several well-controlled game; a game where rule- reasons. First, we consider rule-breaking to be breaking is possible and where the subjects a human activity, that is, located at the level of include entrepreneurs. In so doing, we address the individual entrepreneur; so, observing how the following research question: Do entrepre- real subjects play a game is likely to provide neurs discover, learn, and use rule-breaking additional insights into rule-breaking beyond opportunities better than non-entrepreneur what economic and behavioral models can peers to reap rewards in inter-firm competition? predict. Second, the approach allows us to When we use the term rule-breaking here, we better measure and control actor-related vari- are referring to a combination of opportunism ables, such as gender, as well as situational defined by Williamson (1975) and organiza- variables, such as perceived injustice. Third, tional expedience defined by Parks, Ma, and computer-based experiments provide a lot of Gallagher (2010), that is, rule-breaking behavior data from few subjects, allowing meaningful involves actions taken in self-interest, with statistical analysis, for example, of learning guile, in the pursuit of firm goals, knowing that tactics.2 norms are being broken. These are managerial Our analysis provides several contributions. behaviors that are intended to fulfill a firm’s We explain what differentiates entrepreneurs in competitive performance goals and are known their use of rule-breaking from managerial peers to be violations of inter-firm rules and of the and we explain how that leads to higher perfor- sanctioned norms of competition. This concept mance. These are new insights to the entrepre- is consistent with the idea that higher perfor- neurship literature. These insights complement mance requires aggressive behavior by agents the entrepreneurial discovery concept (Hayek that can cross the normal margins of what is 1978; Kirzner 1985)—where entrepreneurs acceptable (Balch and Armstrong 2010); in other acquire experience-based knowledge to exploit words, the core motive for rule-breaking here is in a particular context—as we can detail which to achieve advantage (i.e., it is not how you experiences affect our entrepreneurs’ actions play the game, but whether you win). We and outcomes. These insights also complement believe that this type of behavior is a defining the subjective entrepreneurship view—a view trait of the entrepreneur, as previous studies that also considers the process of discovery have suggested—for example, in linking the (Kor, Mahoney, and Michael 2007; Penrose entrepreneur to opportunism (Maitland, Bryson, 1959) with its focus on how entrepreneurs and Van de Ven 1985) and in linking the entre- experiment and adapt to changing contexts—as preneur to rule-breaking. “To be an entrepre- we can delineate different levels of the adapta- neur, it is often said, one must break the rules so tion skill. Additionally, we present and test a as to take advantage of opportunities one iden- repeatable, controlled experimental decision tifies or can create” (Brenkert 2009, p. 448). game, where clear rule-breaking play is possible Entrepreneurs not only break rules to create and where different types of rule-breaking can advantage directly but also indirectly through be delineated. This type of experimental game is their creativity (Baucus et al. 2008), and often do also new to the field. so out of necessity because entrepreneurs deal The rest of the article is organized as with more decisions at higher levels of uncer- follows: The second section develops the basic tainty and complexity than managers (Busenitz hypotheses regarding rule-breaking and entre- and Barney 1997).1 preneurship. The third section describes the

1Fortunately, these are also the conditions in which rule-breaking is more tolerated (Brenkert 2009). 2Note that this approach is also supported by precedent. Related past research inquiries have also used this approach: in experiments with rule-breaking, some with formal games (e.g., Baltag 2002; Mikulay and Gofﬁn 1998; Morrison 2006; Russano et al. 2005); in experiments on entrepreneurial discovery (e.g., Demmert and Klein 2003; Kitzman and Schierbeck 2005); and in game-theoretic cases of rule-following (e.g., Faysse 2005). Additionally, the method parallels the experimental work done in related phenomena, like in the study of cooperation, for example, with the formal Prisoners’ Dilemma game analysis, which was complemented by experimental and simulation research (see Axelrod 1984).

2320 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT experimental methods, including the formal of differentiation (e.g., in risk tolerance, social description of the experimental game, the data, ability, and so on) of entrepreneurs (Hartog, the empirical analysis, and the results. The Van Praag, and Van Der Sluis 2010; Zahra and fourth section discusses the results, the impli- Wright 2011). We instead follow in the wake of cations, the limitations, the conclusions, and research on cognitive sources of differentiation some areas for future research. (e.g., on decision-making processes) that appear as more supported candidates for dif- Generation of Hypotheses ferentiation of entrepreneurs, especially those Defining the Focus sources related to opportunity recognition Our focus on rule-breaking involves entre- (Mayo and Nohria 2005; Short et al. 2010). In preneurial discovery and creativity. The focus is other words, we assume entrepreneurs do on a “subjective entrepreneurship” (Kirzner differ from non-entrepreneurs and that this is 1973) process of discovery and exploitation of more likely due to “nurture”—due to experi- rule-breaking opportunities. The rule-breaking ence in a differentiated context—than to behavior is proactive,3 self-oriented, and “nature.” involves finding opportunities by violating We expect entrepreneurs to be more com- formal explicit rules and accepted norms. It is petitively opportunistic, that is, to engage in self-oriented in that the goal is to benefit greater rule-breaking than non-entrepreneurs. oneself primarily, although other parties may This is because entrepreneurs have character- be affected positively. It involves breaking rules istics that differentiate them from managerial and norms, but not laws. We believe that there peers because of differences in what makes is value in understanding how an entrepreneur- each successful. Managers oversee inherited ial discovery of the holes in the rules of the processes to maximize efficiency, whereas competitive game is exploited by decision- entrepreneurs find success through discovery makers who evaluate, assimilate, and learn and risk-taking in the exploitation of that dis- through experience about the benefits and covery. The literature describes the functions of methods of creatively using that discovery to the entrepreneur and his or her associated generate a competitive advantage. Though we characteristics. The connection of the entrepre- believe that there is also value in understanding neur to discovery and to creative application is opportunistic behavior in other contexts,4 we spoken to in the concept of subjective entre- restrict our focus on opportunism to the exter- preneurship in Kor et al. (2007, p. 1187): “the nal competitive context in this study of entre- process of discovery and creativity, which con- preneurial activity. stitute the heart of entrepreneurship.” The connection of the entrepreneur with discovery has Formalizing the Hypotheses earlier roots, of course (e.g., in Hayek 1948; Our hypotheses capture how entrepreneurs Penrose 1959; Schumpeter 1934; and others). It differ from non-entrepreneurs in their behav- is often explained in terms of the entrepreneur iors and the outcomes from those behaviors. being better at discovering a new opportunity, We focus mainly on rule-breaking behaviors evaluating it, and using it (North 1990; Penrose and what we expect drives such behaviors. 1959). More generally, entrepreneurship is Thus, this study differs from the three-plus associated with rule-breaking to win against decades of traits-based research that has failed the overwhelming odds of failure for new ven- to consistently determine noncognitive sources tures (Dunne, Roberts, and Samuelson 1989;

3Proactive rule-breaking is where the focal subject initiates it, rather than being a cooperating co-conspirator with a partner who initiated the activity. 4At its broadest, the related literature covers research on any behaviors considered as variants of opportunism, that is, behaviors that exhibit “self-interest seeking with guile” (Williamson 1975, p. 6). This set of literature covers a wide range of unusual behavior, including actions taken inside the firm and within cooperative arrangements between firms; however, these are actions that are not of interest in the current paper, which is focused on opportunism in competitive (inter-firm) contexts only. Our focus allows us to avoid the interesting, but often value-destroying opportunism studied in alternative research (e.g., in workplace deviance in organizational behavior, where employees engage in rule-breaking that is costly to the firm, for example, Robinson and Bennett 1995).

AREND 321 3 Romanelli 1989; Sharma and Kesner 1996), neurs break rules rationally to find a competi- proactivity as a personal disposition of self- tive advantage and to compensate for the interested action, prospecting, risk-taking added risk of doing something new and differ- (Stevenson and Jarillo 1990), performance ori- ent (even if they differ over certain assumptions entation, opportunity awareness, and so on. underlying their rational choices). Entrepre- The idea of entrepreneurs as self-interested neurial rents are typically the short-lived eco- rule-breakers implies proactivity; entrepre- nomic gains arising from opportunity neurs direct efforts to change their current exploitation (Mahoney and Pandian 1992); the situation (Bateman and Crant 1993), as pros- gain that occurs before that opportunity is pectors (Miles and Snow 1978) and as path- either copied or completed. In other words, finders do (Leavitt 1988). Moreover, the entrepreneurs, when given an opportunity, will calculus of the risk–reward trade-off is skewed outperform peers, doing so in an economically toward taking risky actions when the potential measurable fashion. perpetrator overestimates success (Brockhaus Entrepreneurs are expected to outperform 1980; Sitkin and Pablo 1992) and underesti- non-entrepreneurs on average, in economic mates failure (Morrison 2006). (Such observed terms, although the variance in performance is greater taking of “risks” is explained by a dif- also expectedly higher. This is an economically ference in the way entrepreneurs see those rational expectation, given most entrepreneurs risks relative to non-entrepreneurs; they do not have a choice whether or not to start a new consider their actions taken to actually be risky venture, and given they have similar nonmon- because they differ in their assumptions etary utility functions as non-entrepreneurs. because of certain biases and heuristics in their When one attributes greater creativity and thinking—Busenitz and Barney 1997). Now, control to those in entrepreneurial positions, the way to capture the proactivity and risk- then an additional argument can be made that taking relevant to discovery, the performance they face fewer constraints for addressing a orientation, the ability to evaluate and to cre- problem and thus have expectations of a larger atively apply new discoveries, and the leader- “optimization space” and, hence, greater ship to do so as described earlier is to returns. When one attributes greater drive, generalize these into a type of person—the competitiveness, hubris, and so on to the entre- entrepreneur (Foss et al. 2000). The first preneurs, there is a further “internal” push to hypothesis follows. outperform non-entrepreneurs in any competitive situation. When one attributes greater H1: Entrepreneurs will display relatively grea- search, alertness, and open-mindedness to ter rule-breaking behavior in inter-firm com- entrepreneurs who normally look for new petitive contexts than non-entrepreneurs. rather than existing solutions to problems, then again there is an expectation of relatively Entrepreneurs discover opportunities and higher performance. Thus, there is a general exploit them for gain. Here, we focus on the expectation of entrepreneurs as higher per- most common measure of that gain—economic formers, whether based on rule-breaking activ- rewards. Even when other outcomes are ity or not, in competitive contexts relative to weighed heavily—for example, success, non-entrepreneurs. The second hypothesis power, independence, social good, notoriety— follows. economic success remains not only a proxy for these other goals but also a necessary condition H2: Entrepreneurs will obtain relatively higher for buying the resources to obtain and retain economic rewards in competitive situations them. Rule-breaking precedes value creation, than non-entrepreneurs. as Penrose (1959, p. 34) notes: “the decision to search for opportunities is an enterprising deci- Although it may appear that there is an sion requiring entrepreneurial intuition and implied direct relationship between H1 and H2, imagination and must precede the ‘economic’ there is not; in fact, we propose that only when decision . . .” Entrepreneurs combine knowl- rule-breaking is done in a rational, sophisti- edge, including newly discovered rule-breaking cated manner that net benefits will accrue. In opportunities, to identify activities that create other words, there is a difference between the economic value (Bull and Willard 1993; blind use of a discovery of a rule-breaking Hagedoorn 1996; Schumpeter 1934). Entrepre- opportunity and the skilled exploitation of it,

4322 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT especially in a competition. When the ways to first, the entrepreneur must discover the oppor- use rule-breaking and the specific opportuni- tunity early, in order to maximize the length of ties to use it are many, then there are better and the learning window, and to give an opportu- worse ways and opportunities. We expect the nity to learn faster than rivals. Second, the skill that optimizes the impact of rule- entrepreneur must use that window effectively, breaking—the skill to use it correctly in the by a combination of experimenting through use right situations—will vary across the popula- and aggressive reaction to positive feedback, tion; further, we expect entrepreneurs to be and by predicting rival moves so that the ben- more likely to possess that skill. efits can be maximized in any interdependent We consider the parallel of the skills payoff settings. We summarize this hypothesis required for the successful exploitation of dis- testing approach in Figure 1. coveries of technology to the skills required for In what follows, we consider each of the the successful exploitation of discoveries of foundations of rational rule-breaking to com- breakable rules. Success is based on smart plete the set of hypotheses we test in this article. exploitation, through the use of specific skills. We expect the entrepreneur to discover an Entrepreneurs are more often attributed the opportunity for rule-breaking earlier, not only to skills and assets that correlate with successful exploit it earlier but also to provide a head start exploitation of discoveries, including network- on learning about its benefits. As outlined in the ing (Ferriani, Cattani, and Baden-Fuller 2009), concept of subjective entrepreneurship, Kor social capital (Haber and Reichel 2007), and et al. (2007, p. 1189) note: “The ‘surprise’ speed (Capelleras and Greene 2008). element is at the heart of the life of an entrepre- Here, we expect the entrepreneurs to use neur, and the detection of errors and learning rule-breaking more effectively, that is, exploit- are an integral component of ‘entrepreneurial ing their opportunities so that they not only have a greater number of “successful” rule-breaks but also a lower level of “unsuccessful” rule-breaks. Figure 1 It is possible that a discovery and the simple use Analytical Approach to of that discovery do not lead directly to benefits. When that is the case, it is likely that something Entrepreneurial Iconoclasty extra is involved in creating the desired benefits, such as the skills to exploit the discovery in a sophisticated rather than simple manner. Rule-breaking Performance We term that skill rational rule-breaking behavior. When entrepreneurs use more rational rule-breaking then an interesting connection Rational Performance between the previous two hypotheses arises: to Rule-breaking gain more benefits, the entrepreneurs must use rule-breaking more often in a rational way. They can do so by discovering the opportunity earlier 1. Discover Rule-breaking (early) 2. Learn to use it well (predict rival moves and then learning to use it more effectively, for accurately and aggressively respond to example, by experimenting with its use. Those positive feedback) actions imply more rule-breaking overall (H1) and a faster path to the rational type of rule- breaking that generates more benefits (H2). The Early Performance third hypothesis follows. Rule-Breaking H4 H2

H3: Entrepreneurs will pursue more Rule-Breaking “rational” rule-breaking behavior than Accurate Rival Entrepreneur Predictions H1 non-entrepreneurs. H5

H3 H6 Rational Using a discovery effectively—that is, par- Rule-Breaking taking in rational rule-breaking—is an identiﬁ- Aggressive Reaction to able skill, a skill that is based on learning from Feedback experience with rule-breaking. A superior path to learning in such a manner is observable:

AREND 323 5 discovery’ . . .” In that vein, the sooner the “rule control and superior data generation volume error” is detected, then the sooner the learning per subject. The challenge is to find an appro- about how to use it can occur. Of course, priate game for subjects to play against each entrepreneurs have been identified with moving other in the lab. In what follows, we outline the first to gain an advantage over incumbents necessary characteristics for such a game; we before (Mahoney and Pandian 1992; Michael describe the game and its analysis in the sec- 2003). Our fourth hypothesis follows. tions thereafter. An appropriate game must fulfill several H4: Entrepreneurs will display earlier rule- conceptual and pragmatic requirements; we breaking than non-entrepreneurs. consider the conceptual here and the pragmatic ones in the procedures section. Conceptually, We expect the entrepreneur to better predict the game must allow subjects to discover the a rival’s moves as part of the learning done to opportunity to rule-break, allow the subjects to optimize the use of a rule-breaking opportu- learn how to use that discovery effectively, and nity. When the entrepreneur can accurately allow those subjects that do learn to be predict what the rival will do then an optimal rewarded. That means the game must allow countermove—that is, one that maximizes rule-breaking, but that such rule-breaking be expected payoffs and minimizes the possibility non-obvious, so it is discovered rather than of losses—can be selected. Entrepreneurs have simply imitated. It also means that the rule- been defined in the literature partly by their breaking is conditionally beneficial—there ability to find opportunities by being able to must be some learned skill to its effective use predict the future more accurately (Bull and (e.g., based on calculated expectations)—and Willard 1993; Bygrave 1998; Kirzner 1985; that entails some risk to its use. Finally, for Knight 1921; McMullen and Shepherd 2006; learning to occur, the game must provide Ogbor 2000). The fifth hypothesis follows. meaningful and timely feedback. When all of these conditions are met, then the data gener- H5: Entrepreneurs will more accurately predict ated by subjects playing the game can be used their rivals’ moves than non-entrepreneurs. to capture the “sophisticated” opportunism we are interested in studying. In other words, such The superior ability to learn how to ratio- games can be used to differentiate entrepre- nally break the rules is also based on the neurs from others in their abilities to be rational aggressive response to positive feedback from iconoclasts in competitive situations. In what experimentation with the use of rule-breaking. follows, we describe how the game we chose Feedback is essential to learning, and those fulfills these requirements. who respond to it more aggressively—that is, by more often using what works—learn the Formalizing the Game and right lessons faster. Entrepreneurs are associ- Rule-Breaking in the Game ated with superior experience-based learning We chose to have subjects play a game that through experimentation and feedback (Hayek had a well-known base; we then modified it 1945; Kor, Mahoney, and Michael 2007). Entre- with new, unfamiliar yet understandable twists. preneurs are also associated with more aggres- The well-known base is the rock-paper-scissors sion to learning, especially in response to (RPS) game; it is a competitive, often repeated, positive feedback (Minniti and Bygrave 2001; head-to-head game. It is a familiar tie-breaking Stringer 2000). Our sixth hypothesis follows. game from childhood. The twists we used provided a way to make the game strategic rather H6: Entrepreneurs will display more aggressive than a game of pure chance. The twists, of reaction to positive feedback from rule- course, also allowed for a rule-breaking type of breaking than non-entrepreneurs. behavior. The modified game is not easily solv- able, in a Nash equilibrium sense, and that Experimental Methodology provides the not uncommon sense of ambiguity and Results about an opportunity that entrepreneurs face The most appropriate means of testing our in a competitive context. Such ambiguity has hypotheses is through a lab experiment; as we been connected to rule-breaking (Parks, Ma, outlined in the introduction, this empirical and Gallagher 2010). The modified game also approach has many benefits, including superior makes it risky to use rule-breaking; if all

6324 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT players rule-break all the time, the expected twice in any three-trial game. In other words, payoffs for each player actually decrease from rule-breaking is the breaking of the formal rules the case where all players follow the rules all of play (please see the Appendix for example the time (see Analysis of the Game section). instructions). Our base case RPS game is structured so that The second modification provides cover for only asymmetric outcomes provide rewards rule-breaking. The noise created by flipping (e.g., rock beats paper), and so that any game hides a possible rule-break. For example, against a rival lasts three trials of one simulta- seeing a rival play RR does not necessarily neous move each. In what follows, see the trial mean a rule-break has occurred as RP could game in normal form: have been chosen with the second move flipped by the computer. Such cover is useful for several reasons: (1) it keeps the discovery Your Your Partner’s Decision that rule-breaking is possible hidden to those Decision not yet rule-breaking; (2) it increases the pos- RPSsibility that an observed rule-break may not invite automatic retaliation;6 and (3) it reflects R0,0 0,2 3,0 reality better, given rule-breaking does not P2,0 0,0 0,1 often occur when it is blindingly obvious to the S0,3 1,0 0,0 victim. Formally, each subject plays 15 games, lasting 3 trials each, with each game played We then add two significant modifications to against a different anonymous partner. Subjects this base case. The first modification is a restric- are informed when they change partners and tion: that once a particular strategy is chosen in which trial they are in. In each trial, subjects any one trial against a set opponent then that have a simple choice among three strategies: R, strategy becomes unavailable in any remaining P, and S (rock, paper, and scissors). Subjects trials in that game. So, if the player chooses R are motivated to play the game seriously and in the first trial and P in the second, only S can competitively as they are rewarded course be chosen in the third trial. The second modi- grades based on their combined scores across fication is that noise is introduced into the all games they play. There is also an ego-related game in the following manner: any player’s reward for besting their peers to consider, as chosen strategy can be flipped to another strat- scores are made public to the class after the egy by the computer with a small positive prob- experiment is completed. ability, and all players know that this noise Now let’s consider how to differentiate exists when playing the game. For example, if between rational and irrational rule-breaking. the choice was P, it may well end up as R when The most straightforward context to do this in the move is revealed to the rival and the is the third trial of any game, because by then payoffs are computed.5 the partner will have played two strategies and The first modification provides the opportu- has only one left when that partner does not nity for rule-breaking here, that is, rule-breaking rule-break. When we look only at cases where occurs when a player uses the same strategy the partner has so far not broken the rules, we

5This second modification reflects the idea that the strategic position chosen is not always the one that gets implemented (e.g., Mintzberg 1990; Mintzberg and Waters 1985). Often, a firm will have trouble getting its employees to switch strategies, or may end up with what the employees have interpreted the plans for a strategy to be rather than what the plans actually were (Martin 1993; Nutt 1999). In other words, there is often noise in the system that occurs upon attempted implementation of a chosen strategy. One may argue that a third modification exists, that is, the introduction of asymmetric payoffs to winning under different matches (e.g., winning with rock has the highest payoff). There were two primary reasons for this payoff structure: (1) to make the solution to the game more difficult; and (2) to provide clear preferences for rule-breaking choices, which makes identification of tactics that set up superior rule-breaking more clear (e.g., playing R early so that it is more likely that the rival will play S late). 6That cover for rule-breaking activity is likely to increase the probability of such behavior by individuals, which is important to generating the numbers we need here for statistical power.

AREND 325 7 Figure 2 choice of S with one’s own choice of P, which Graph of Expected Welfare to provides one with a payoff of 0 and the partner with a payoff of 1).7 Player Rule-Breaking Analysis of the Game 4.4 There are several approaches for analyzing this game, and the approaches get more com- 4.2 plicated as more factors are modeled (e.g., the noise, player beliefs, etc.). Each approach uses the concept of backward induction, that

e 4.0 r is, starting from the end of a game (the third trial) and working back by identifying the 3.8 conditionally optimal choices of strategy per l welfa

ta trial until the start of a game (the ﬁrst trial) is 3.6 reached, at which point the choices for all of a game’s trials are optimized. Each approach

ected to assumes that subjects play rationally, that is, p 3.4

Ex that subjects choose the move, or mix of moves, that maximizes his or her own 3.2 expected payoffs. Through such an analysis, we can prove that (1) the game is nontrivial 3.0 to solve; and (2) the “optimal” level of rule- 020406080100breaking is not at either extreme of such activity.8 For our purposes, we do not need to α – percentage of rule-breaking by each rival solve this game in a theoretical super-rational player sense. This is a paper that looks at how and why one type of subject outperforms another type of subject, where that relative performance is measured in observed payoff can deﬁne a “rational” rule-break as a move levels and not theoretical possibilities. Using that is expected to result in a positive payoff some simple anchor cases, we can show how (e.g., given both players have chosen R and P rule-breaking activity affects payoffs. For an α in the previous two trials, addressing an (0 < α < 1) that indicates the rule-breaking expected rival choice of S with one’s own probability for the three-trial game, we can choice of R is rational and provides one with a show that the α* that “maximizes” the payoff of 3 and the partner with a payoff of 0) expected welfare lies at the 29 percent level and an “irrational” rule-break as a bad move approximately9 (see Figure 2). that is expected to result in a positive payoff for Thus, our modiﬁed and structured game the partner (e.g., addressing an expected rival includes several important features: (1) the

7This may seem obvious, but rational rule-breaking was not common. For example, consider that in our experiment, if such rationality was used in every third trial where the partner had not yet broke the rules, the average total payoff per player would have increased 26 percent. 8A full analysis approach would model each subject, for each trial, as having a set of beliefs (e.g., set by BR and BP) that deﬁne what the probabilities of a rival playing any of the three moves would be

(e.g., BR,BP, 1-BR-BP), where those beliefs are based on priors, updated for past game moves, and corrected for the possibility of ﬂipping. Given those beliefs, the model would consider the subject then choose an optimal mixed strategy of moves (e.g., randomizing between R, P and S) to maximize expected payoffs, accounting for how that strategy would inform the rival’s beliefs for any further trials. The full equilibrium would occur when those beliefs aligned with actions and signals, and when players could not choose better beliefs, actions, and signals given rivals were also playing such optimized strategies. 9It is interesting to note that the observed average level of rule-breaking closely approximated this optimal level (it was about 27 percent). This is also approximately the level of entrepreneurial discovery reported in previous experiments (Demmert and Klein 2003; Kitzman and Schierbeck 2005).

8326 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT game is easy to play but is not trivial to solve; rule-breaking. In what follows, the details of (2) an observed breaking of rules may not be a Study 2 are provided. “real” breaking of rules by an opponent; (3) individual rule-breaking by itself will not guar- Subjects. Study 2 involved 60 subjects—20 antee benefits over rule-following, but it can do entrepreneurs and 40 other businesspeople so if used smartly; and (4) too much rule- attending a strategy course. The entrepreneurs breaking in the population provides lower were self-identified as individuals with experi- payoffs than rule-following. The analysis and ence starting and running a business for over the description of this game indicate that this two years. The non-entrepreneurs were self- game does fulfill the requirements for our identified as businesspeople with varying experiment, and so is suitable for testing the degrees of work and managerial and entrepre- hypotheses. neurial experience. This experience-based dichotomy separating out the entrepreneurs has precedence (e.g., Busenitz and Barney 1997) The Studies of Subjects Playing and is based on the idea that people who have the Game faced the “unique” decision-making context in a We tested our hypotheses with two studies. start-up venture—for example, the uncertainty, The first study was an initial program to deter- the fast pace of short windows of opportunity, mine whether a sufficient level of rule-breaking the high-powered incentives, the need to would occur during play; the second study was provide leadership, etc.—will think differently. the source of the data used in the present They may be more creative and exploratory, analysis. they may use different heuristics, and they may Study 1 involved undergraduate business stu- be more prone to certain biases. The subjects dents. This was our pilot study that was used to were randomly assigned to one of three experi- investigate our concern over rule-breaking activ- mental conditions described as follows. Subjects ity level and whether the “entrepreneurs” would were motivated to make thoughtful decisions by break the rules more. We used an exit survey to basing a portion of their course grade, nominal measure a small array of entrepreneurial char- cash prizes, and on how many points they acteristics of the subjects, including their aspi- earned during the course of the experiment. The rations and abilities to recognize opportunities. subjects voluntarily submitted to that condition, Regardless, the study provided promising having been introduced to game theory and the results: first, there was a statistically different- concept of game-changing actions in general in from-zero level amount of rule-breaking behav- the normal course of the contents covered in the ior; second, the students reporting higher strategy course content. Subjects were debriefed entrepreneurial characteristics broke the rules after all sessions were completed. Subjects were more often and enjoyed higher payoffs, on between the ages of 23 and 40, with an approxi- average; and third, those students broke the mate balance between males and females rules more effectively than peers. In other overall. words, the results lent support to our first three hypotheses. Thus, the pilot study addressed our Procedure. We conducted the experiment concerns. It also provided an impetus to alter in a computer lab at a major southwestern some survey questions, to use more stringent university. The lab contained over 20 well- entrepreneurship classification requirements, spaced networked computer stations. As sub- and to set up a treatment on the “acceptability of jects entered the lab, they were randomly rule-breaking” in the game, all for the main assigned to a workstation that was preset to play study. one of three experimental conditions. Subjects Given those results and that experience, it were required to complete a short survey prior was decided to conduct a second study. Study to starting the experiment. They were told to 2 involved a more appropriate group, that of reread the instructions (see the Appendix for a young business managers, that is, subjects copy of example instructions) that were placed with at least two years of professional work beside each station; they had gone over the experience; the mean over nine years of expe- instructions in detail prior to the lab experiment rience. We used an entry rather than an exit in a session where they were given examples survey, controlled for more factors, and applied and had questions addressed. The subjects were treatments concerning the acceptability of asked not to communicate with one another

AREND 327 9 during the experiment. The instructions indi- Dependent Variables. We measured the per- cated that they would play a simple two-person centage of rule-breaking out of all possible game over a computer network. The game, opportunities to do so in the game through which was designed as a repeated RPS (rock- the variable amount_rule_breaking. We mea- paper-scissors game), was introduced as a stra- sured the subject’s total rewards in the game tegic choice exercise. The instructions also (i.e., across the 45 trials) through the variable indicated that during the course of the experi- total_payoff. We measured the amount of ment, subjects would be randomly paired with rational rule-breaking through a proxy that we several different players for a number of term rational_rule_breaking. It was calculated repeated trials. At the beginning of each trial, based on the third-trial choices made in the subjects were required to make one of three game, because in that trial, a non-rule- decisions: R, P, or S. Their score depended on breaking rival’s move is predictable, making the decision of the person they were paired the rational rule-breaking choice clearly iden- with. For example, if both players chose R, each tifiable. It was calculated as the fraction of earned 0 for the trial. The instructions contained third-trial rule-breaking activity that did result a version of the payoff table to aid subject in a positive payoff out of all the subject’s decision-making (see the Appendix for the third-trial rule-breaking activity that could table). have resulted in a positive payoff, less the At the end of each trial, the computer fraction of third-trial rule-breaking activity informed subjects of their partner’s imple- that resulted in a positive payoff for the rival mented decision—that is, the partner’s out of all the subject’s third-trial rule-breaking intended decision then possibly flipped by the activity that could have resulted in a positive computer—and reported both the cumulative payoff for the subject. We measured the trial and current earnings for the trial. The computer number of the subject’s first rule-breaking informed subjects every time that they changed through the variable first_rule_break.We partners. All partnering was anonymous. At the measured the subject’s ability to accurately end of the 15th pairing, the computer informed predict rival moves—that is, the variable subjects that the experiment had ended and good_rival_prediction—through a proxy, cal- provided a recap of their cumulative earnings. culated as the fraction of third-trial predictions After the experiment had ended, subjects were of the rival’s move that were correct. We debriefed and dismissed from the lab. measured the reaction to positive feedback Players were told that their grade for this from rule-breaking behavior—that is, the vari- part of the course and any bonus grade able positive_feedback_from_rule_breaking— depended on their score in that lab session through a proxy, calculated as a count of the relative to their peers, with top scores earning number of times the subject chose to rule- the highest grades and extra credit. break after an immediately previous rule- As described earlier, the modified RPS game break that resulted in a positive payoff. We met our practical requirements: it was easy for measured the subject type through the the subjects to understand, which reduced the dichotomous variable entrepreneur, equaling learning curve and probability of error. It was a one if the subject had experience starting and relatively quick game to play, which allowed us running a business for over two years, and to generate a sufficient quantity of data and to equaling zero otherwise, consistent with pre- retain the attention of subjects repeatedly vious studies identifying entrepreneurs in the playing it. It was a nontrivial game to solve. population (e.g., Ardichvili and Gasparishvili Also, it allowed rule-breaking activity that was 2003; Busenitz and Barney 1997; Manolova non-obvious, especially to rivals who are et al. 2007). victims of rule-breaking, because we provided “cover” for it through noise. Explanatory Variables. We used two types of non-focal independent variables—one based on subject type and one based on the experi- Variables mental context. Both trait-based and social We tracked several items of data for each exchange-related controls have been con- subject relevant to the analysis of the hypoth- nected to rule-breaking behaviors in the lit- eses. We calculated the variable values as erature (Parks, Ma, and Gallagher 2010). follows (see Table 1 for details). Indeed, both individual and situational factors

10328 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Table 1 Variables and Their Deﬁnitions

Variable Name Description entrepreneur (0,1); =1 if the subject had experience starting and running a business for over two years male (0,1); =1 if subject was male work_experience Integer; number of years of subject’s professional work experience entrepreneurial_orientation Averaged score on seven five-point Likert-scaled questions: (where 1 = strongly disagree, 2 = disagree, 3 = neutral, 4 = agree, 5 = strongly agree): Iamor intend to become self-employed in the near future; I have the ability to recognize ideas for self-employment (Singh and DeNoble 2003). Wherever I have been, I have been a powerful force for constructive change; when I have a problem, I tackle it head-on; I can spot a good opportunity long before others can (Bateman and Crant 1993). I am not willing to take risks when choosing a job or firm to work for; I prefer a low risk/high security job with a steady salary over a job that offers high risks and high rewards (Gomez-Mejia and Balkin 1989). Note that the responses to these last two statements are reverse-coded. session2_minimal_rule (0,1); =1 if subject was in session 2 of the main study where no explicit rule instruction was provided session3_pop_up (0,1); =1 if subject was in session 3 of the main study where both an explicit rule and a pop-up reminder were provided amount_flipped (0–1); fraction of trials in which subject’s initial choice was flipped by the program by chance amount_rule_broken_on (0–1); fraction of trials in which subject’s rival rule-broke of trails where rule-breaking was possible first_rule_break Integer; trial number when subject first broke the rule good_rival_prediction (0–1); fraction of third-trial predictions of rival moves that were correct rule_break_positive_feedback_ Integer; the number of times the subject chose to reaction rule-break after an immediately previous rule-break that resulted in a positive payoff amount_rule_breaking (0–100); the percentage of the subject’s rule-breaking out of all possible opportunities to do so in the game total_payoff Integer; the subject’s total rewards in the game rational_rule_breaking Score (−1 to 1); the fraction of third-trial rule-breaking activity that did result in a positive payoff out of all the subject’s third-trial rule-breaking activity that could have resulted in a positive payoff less the fraction of third-trial rule-breaking activity that resulted in a positive payoff for the rival out of all the subject’s third-trial rule-breaking activity that could have resulted in a positive payoff for the subject

AREND 329 11 have been used to explain rule-breaking injustice with two variables—amount_flipped behavior (e.g., Zimbardo 2007).10 and amount_rule_broken_on. We expected that individuals who experienced a flip would Subject’s Individual Characteristics as Variables. choose any strategy, regardless of the rules, that The variable entrepreneurial_orientation con- would help them reestablish control and reduce sisted of seven items measuring a subject’s anxiety (Cherns 1976; Plaks, Grant, and Dweck interest in entrepreneurship (these seven items 2005), so we controlled for the amount of flip- were normalized, weighted, and combined into ping experienced by the subject. We also the final score, with a Cronbach’s α = 0.723). expected being a victim of a rival’s rule-breaking We controlled for gender with the would increase the emotional responses that variable male; this is because gender had drive defiance (Greenberg and Alge, 1998; Vardi been shown in previous studies to be a factor and Weiner, 1996).12 in entrepreneurial discovery (e.g., Kitzman and Schierbeck 2005). We controlled for the Empirical Methods subject’s business-related knowledge with the We tested all six hypotheses in three ways: (1) variable work_experience. with a simple two-sided means difference t-test based on the cardinal number data underlying Experimental Conditions as Variables. We the variables of interest; (2) with a logit analysis controlled for the experimental conditions to determine whether the variable of interest (i.e., treatment) with two variables—session2_ correlated significantly with the identification of minimal_rule and session3_pop_up; session 1 the entrepreneurs; and (3) with a suitable was considered the base condition. Differing regression on the variable of interest (i.e., levels of the acceptability of the rule-breaking double-censored tobit analysis for H1, H3, and were captured in the treatments (where rule- H5; ordinary least squares for H2; survival analy- breaking was more acceptable in session 2 and sis for H4; and count model regression for H6). less acceptable in session 3).11 Results Combined Experimental Individual Variables. Table 2 depicts the descriptive statistics We controlled for possible reaction to perceived and correlations. Table 3 depicts the various

10From behavioral-related literatures, we know that individual characteristics are important in rule-breaking; for examples, from research on escalating commitment (Whyte, Saks, and Hook 1997), on employee deviance (Lee, Ashton, and Shin 2005, and on betrayal of trust (Elangovan and Shapiro 1999). We also know from behavioral research that the situational factors explain some variance in rule-breaking behavior, for example, from the fraud triangle model (Cressey 1973; Howe and Malgwi 2006). That model predicts that rule-breaking is more likely to occur in the presence of three situational factors: an opportunity, sufficient motivation, and accompanying rationalization. 11The first experimental condition (i.e., session 1) in Study 2 consisted of explicit rule instruction, that is, where the rule of playing a strategy only once per three-trial game was written and explained. The second experimental condition (i.e., session 2) consisted of no explicit rule instruction, that is, where the example of play provided was consistent with rule-following but no rule-restricting strategy choices were explicitly stated or written in the instructions. The third experimental condition (i.e., session 3) consisted of an explicit rule instruction like in the first condition, but with a pop-up message that was presented any time a subject tried to play a previously played strategy in any three-trial game. The message warned the subject that they were about to play a previously chosen strategy and that such an action was against the rules; it allowed them then to either continue with the choice or choose another strategy. Rule-breaking behavior is more difficult to rationalize if one is warned and reminded explicitly of what that behavior is (McCabe and Trevino 1993; Nyborg and Telle 2004). Warnings reduce mistakes in rule-breaking (Peters 1985). Warnings create added anxiety, guilt, and risk to rule-breaking, essentially lowering the reward of, and the commitment to, rule-breaking (Collins 1977). 12According to Blau (1964), trust is reinforced when a rival follows the rules but a break in the rules is seen as a violation of the relationship. As a result, there may be a breach of trust with a sense of interactional injustice experienced and that is likely to provoke the most intentional emotional response of all types of injustice (Bies 2001)—anger and retaliation-in-kind (Ambrose, Seabright, and Schminke 2002; Rudolph et al. 2004; Skarlicki and Folger 1997); thus, we controlled for the amount of being rule-broken on that the subject experiences.

12330 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Table 2 Descriptive Statistics and Correlationsa

Descriptive Statistics

Non-Entrepreneur Entrepreneur

Min Max Mean S.D. Min Max Mean S.D. RN 331 AREND RN 13 AREND amount_rule_breaking 3.333 43.333 22.276 14.670 20.000 56.667 36.491 13.582 total_payoff 16 46 28.000 8.903 16 45 34.263 7.779 rational_rule_breaking −1.000 1.000 0.099 0.533 −0.333 1.000 0.358 0.355 good_rival_prediction 0.133 0.867 0.566 0.160 0.267 0.800 0.533 0.139 ﬁrst_rule_break 2 39 6.049 7.190 2 6 3.000 2.414 rule_break_positive_feedback_reaction 0 5 2.195 1.585 3 10 5.105 2.792 male 0 1 0.390 0.494 0 1 0.368 0.496 work_experience 2.0 22.0 9.098 4.587 1.5 25.0 9.263 5.287 amount_ﬂipped 0.000 0.333 0.130 0.097 0.000 0.333 0.139 0.098 amount_rule_broken_on 0.133 0.500 0.256 0.084 0.133 0.400 0.289 0.082 entrepreneurial_orientation 2.714 5.000 3.676 0.536 2.857 5.000 3.842 0.587 n 40 20 14 332 Table 2 Continued

Correlations

Non-Entrepreneur

12345678910 ORA FSALBSNS MANAGEMENT BUSINESS SMALL OF JOURNAL ORA FSALBSNS MANAGEMENT BUSINESS SMALL OF JOURNAL 1 amount_rule_breaking 2 total_payoff 0.014 3 rational_rule_breaking −0.060 0.225 4 good_rival_prediction −0.033 0.000 −0.030 5 ﬁrst_rule_break −0.543 −0.213 −0.081 0.223 6 rule_break_positive_feedback_reaction 0.695 0.133 0.431 −0.072 −0.427 7 male 0.096 −0.183 −0.073 0.067 0.213 0.060 8 work_experience −0.114 0.107 0.019 −0.036 −0.045 −0.058 0.192 9 amount_ﬂipped 0.195 0.089 0.133 0.140 −0.120 0.049 0.090 −0.346 10 amount_rule_broken_on −0.003 0.020 −0.147 0.084 −0.017 −0.165 −0.138 0.119 0.263 11 entrepreneurial_orientation −0.048 −0.145 −0.091 0.009 0.057 −0.020 0.031 0.063 −0.248 −0.127

Entrepreneur

12345678910

1 amount_rule_breaking 2 total_payoff −0.275 3 rational_rule_breaking −0.601 0.244 4 good_rival_prediction −0.289 0.286 0.234 5 ﬁrst_rule_break −0.449 0.167 0.374 −0.151 6 rule_break_positive_feedback_reaction 0.644 0.147 −0.139 0.030 −0.241 7 male −0.055 0.280 −0.301 0.377 0.000 0.016 8 work_experience 0.075 −0.243 0.118 −0.061 −0.368 0.193 −0.272 9 amount_ﬂipped 0.328 0.015 −0.524 0.128 −0.349 0.300 0.208 0.065 10 amount_rule_broken_on −0.185 −0.298 0.130 −0.388 0.143 −0.356 −0.262 0.236 0.085 11 entrepreneurial_orientation 0.097 0.042 −0.339 0.221 0.163 0.183 0.620 −0.350 0.066 −0.315

aS.D., standard deviation. Italicized numbers are signiﬁcant at p < .05 in two-tailed test. Table 3 Empirical Analyses of the Six Hypothesesa

Analysis Logit Analyses on “Entrepreneur” Double-Censored OLS on Double-Censored Cox Regression Double-Censored Poisson Analysis Tobit on amount_ total_payoff Tobit on rational_ on ﬁrst_rule_ Tobit on good_ on rule_break_ rule_breaking rule_breaking break rival_prediction positive_feedback_ reaction

Hypothesis H1 H2 H3 H4 H5 H6 H1 H2 H3 H4 H5 H6 Variable Coefficient Coefficient Coefficient Coefficient Coefficient Coefficient Coefficient Coefficient Coefficient Coefficient Coefficient Coefficient (S.D.) (S.D.) (S.D.) (S.D.) (S.D.) (S.D.) (S.D.) (S.D.) (S.D.) (S.D.) (S.D.) (S.D.)

(Constant) −8.980* −8.528* −6.345† −4.137 −5.075 −8.511† 21.674 26.220** 0.852 0.463* 0.770 (4.248) (3.880) (3.548) (3.488) (3.420) (5.200) (14.690) (9.736) (0.692) (0.193) (0.787) male −0.206 −0.226 −0.058 −0.030 −0.239 −0.062 −0.794 −0.903 −0.141 −0.029 0.044 −0.072 (0.777) (0.753) (0.699) (0.710) (0.677) (1.009) (3.189) (2.077) (0.148) (0.296) (0.041) (0.170) work_experience 0.000 0.002 0.002 −0.019 0.005 −0.106 0.031 0.066 0.007 −0.016 0.000 0.014 (0.073) (0.065) (0.065) (0.069) (0.063) (0.109) (0.323) (0.211) (0.015) (0.033) (0.004) (0.016) entrepreneurial_ 0.807 0.631 0.760 0.717 0.684 0.201 −0.280 −0.533 −0.088 0.364 0.024 0.038 orientation (0.691) (0.700) (0.650) (0.637) (0.598) (0.940) (2.828) (1.841) (0.131) (0.251) (0.037) (0.150) session2_minimal_rule 0.542 2.163† 2.062* 1.552† 1.469 0.367 7.637† −2.865 −0.237 0.147 −0.097† 0.086 (1.055) (1.151) (1.015) (0.919) (0.942) (1.405) (4.431) (2.955) (0.209) (0.372) (0.059) (0.224) RN 333 AREND RN 15 AREND session3_pop_up 0.876 0.656 1.856 1.524 1.602 0.606 6.972 4.711 −0.225 −0.662 −0.055 0.160 (1.766) (1.742) (1.593) (1.562) (1.572) (2.023) (7.036) (4.618) (0.327) (0.434) (0.092) (0.359) amount_ﬂipped −0.199 −1.979 5.627 2.386 5.056 −5.831 49.790 34.444 −0.295 −0.216 0.178 1.684 (8.492) (8.777) (7.633) (7.679) (7.373) (10.181) (34.411) (22.789) (1.611) (0.182)b (0.452) (1.684) amount_rule_broken_on 6.195 0.100 0.552 1.698 1.564 11.597 −35.199 −2.887 −0.361 −0.919** 0.171 −2.044† (5.440) (5.283) (4.871) (4.494) (4.467) (7.546) (22.271) (14.798) (1.054) (0.251)b (0.294) (1.155) amount_rule_breaking 0.103** −0.050 −0.007 −0.001 (0.034) (0.084) (0.006) (0.002) total_payoff 0.149** (0.052) rational_rule_breaking 1.809* (0.812) ﬁrst_rule_break −0.271 (0.175) good_rival_prediction −0.770 (2.207) rule_break_positive_ (0.449) feedback_reaction entrepreneur 12.947** 7.711** 0.466** −0.070 0.002 0.861** (3.283) (2.398) (0.171) (0.328) (0.048) (0.158) χ2-Stat 21.02** 18.25* 13.94† 13.10 7.59 37.86** 15.21† 22.68** 37.83** Log Likelihood Function −229.59 −48.90 31.32 Standard or Pseudo-R2 0.281 0.244 0.186 0.175 0.101 0.505 0.224 0.146

aS.D., standard deviation; OLS, Ordinary Least Squares. bThese are computed pre-first cheat for this analysis only. †p < .10. *p < .05. **p < .01. empirical analyses. We find support for most of the hypotheses. Discussion and Conclusions Entrepreneurs are more opportunistic than Discussing the Results their peers, as the proportion of rule-breaking The results from hypothesis testing have in the game is significantly greater for entre- theoretical, practical, and policy implications preneurs than non-entrepreneurs. In other to discuss. The empirical results suggest that words, H1 is supported: based on the underly- entrepreneurs rule-break more and with ing means t-test (p < .01); based on the logit higher benefits because these subjects break regression coefficient significance and sign the rules in a more rational way, a way that is (p < .01); and based on the double-censored correlated with breaking the rules sooner and tobit coefficient significance and sign (p < .01). with stronger reaction to positive feedback. In Entrepreneurs also score higher in the other words, this analysis provided support games than peers, as the score is significantly for the introductory explanation of the set of higher for entrepreneurs than for non- relationships, that is, that entrepreneurs break entrepreneurs. In other words, H2 is sup- the rules more for gain, which is consistent ported: based on the underlying means t-test with empirical work like Zhang and Arvey (p < .01); based on the logit regression coeffi- (2009). cient significance and sign (p < .01); and based There are several theoretical implications for on the ordinary least squares (OLS) coefficient the entrepreneurship literature. First, the analy- significance and sign (p < .01). sis complements the entrepreneurship discov- The higher scores occur because the entre- ery concept. Though discovery may be preneurs are more rational (and less irrational) necessary for successful entrepreneurship, in their rule-breaking, as that measure is sig- there are other necessary factors that need to nificantly higher for entrepreneurs than for follow, including the effective use of that dis- non-entrepreneurs. In other words, H3 is sup- covery, that is, where the new knowledge from ported: based on the underlying means t-test the discovery is evaluated, assimilated, and (p < .01 for rational rule-breaks); based on the then applied in a value-adding way. This new logit regression coefficient significance and behavior links the proactive front-end of entre- sign (p < .05); and based on the double- preneurial discovery of hidden information censored tobit coefficient significance and sign (Hayek 1978; Kirzner 1985) with the creative (p < .01). economic value-oriented back end of assimila- The tests to determine the entrepreneurs’ tion and application. Besides that proactive learning characteristics had mixed results. front-end, our study also highlighted the impor- There is only weak evidence that entrepreneurs tance of the skills in application of the discov- are more likely to rule-break earlier. In other ery, that is, in the rational rule-breaking words, H4 is only partially supported: based on concept. Failures of discovery exploitation due the underlying means t-test (p < .05 for the trial to a lack of such skills have been exemplified number of the first rule-break), where the logit by the blocked or lack of access to complemen- regression coefficient was not significant, nor tary resources (Teece 1986) at Xerox PARC, was the coefficient on the survival (Cox regres- incorrect market timing for the Iridium phone, sion) analysis. There is no evidence that entre- the miscalculation of rival reaction at Netscape, preneurs are better at predicting rival moves. In and the improper financial planning for hyper- other words, H5 is not supported: the underly- growth at the one-time Fortune 500 firm, 180s. ing means t-test was not significant, nor was Entrepreneurs appear to differentiate them- the logit regression coefficient, nor was the selves on both ends of the process—by being double-censored tobit coefficient. However, more aggressive and experimental in gathering there is strong evidence to suggest that entre- new information and by being more effective in preneurs do respond more aggressively to posi- applying that new and unusual (rule-breaking) tive feedback on rule-breaking. In other words, information. H6 is supported: based on the underlying Second, the analysis complements the means t-test (p < .01 for the conditional number subjective entrepreneurship view by: (1) proof rule-breaks); based on the logit regression viding one example of how entrepreneurs coefficient significance and sign (p < .01); and learn differently; and (2) providing a way to test based on the double-censored tobit coefficient some of the tenets of that view in a controlled significance and sign (p < .01). context. Entrepreneurs learn differently by

16334 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT starting experimentation earlier and then Limitations responding more aggressively to positive feed- The contributions are, of course, tempered back. Our lab setting allowed testing where by the limitations of the underlying study. First, several observable individual behavioral differ- the modified RPS model does not apply to all ences could lead to discovery of a rule-breaking rule-breaking situations, only those that involve opportunity and its effective application. competitive, symmetric, two-party games, with Third, the results also suggest an alternative clear strategic choices and payoffs, with explicit approach to detecting the entrepreneurial per- rules, etc. Second, several aspects of the experi- sonality (e.g., Mitton 1989). It may be useful to mental design may be questioned, including complement the measures that only pick up the the effects of other possibly relevant subject static issues—for example, gender, education, characteristics;13 a focus on only specific vari- work experience, etc.—with measures that pick able levels and only symmetric payoff matrices; up dynamic issues—for example, the path and the limitations that required us to reward dependency of starting early and responding subjects with primarily grades as motivators. aggressively to positive feedback—in order to We readily acknowledge the common limita- provide a fuller picture of the candidate. tion of most lab studies meeting efficacy The results also have implications for requirements that there are “severe” restrictions entrepreneur–managers. In competitive con- on the stakes of motivators and risks possible texts, efforts should be directed toward early to apply. We have, however, no evidence to discovery of loopholes in the system, as well suspect that age or other demographic charac- as toward capabilities to exploit such discov- teristics, perturbed payoff matrices, or even eries, that is, in getting clear and direct feed- paying subjects in an alternative manner would back from rule-breaking and for continuing have changed the main results. In fact, even the exploitation should positive feedback be despite the relatively small sample size and experienced. limitations, relatively strong results were Moreover, the results have implications for observed. policy, given we found that certain levels of rule-breaking were welfare enhancing. Clear Conclusions legislation and accessible legal advice would We ran an experiment on young business- help entrepreneurs discover the loopholes people, some of whom were entrepreneurs, to earlier. Funding and other support systems may test the link between entrepreneurs and a type decrease the costs of experimentation for learn- of rule-breaking behavior that generates a coming about the effects of rule-breaking, in the petitive advantage. Our main contribution lies competitive context, to help determine the ones in supporting and explaining a set of relation- that create value. Policies that would make ships that links entrepreneurship to rule- technology-related discovery exploitation more breaking and subsequent performance in inter- likely, such as better access to technology firm competitive settings. In such a setting, we transfer and licensing opportunities, and that found that entrepreneurs break rules more and would mitigate the miscommunication of with higher benefits, but only because they market signals, such as those that would reduce break the rules in a smarter way—where the lag and noise in market reactions to new smarter rule-breaking was correlated with products, would also increase the likelihood breaking the rules sooner and having more that smart rule-breaking would occur and be aggressive reaction to positive feedback.14 Our net beneficial. secondary contribution was the introduction of

13n.b., we considered and preliminarily tested for age, and several rule-breaking and risk-related characteristics in each study, with no significant impacts noted. 14To summarize, entrepreneurs differ in rule-breaking behavior for several reasons. They rule-break more than managers because the incentives differ, for example, large firms are harmed by rule-breaking whereas entrepreneur–managers may benefit from it. Although there appears to be a “contradiction” for the net effects at firms that desire both compliance-based efficiency and rebellion-based innovation, there is a way to explain how entrepreneurs enjoy the latter while mitigating the harm of the former in their rule-breaking activity. Entrepreneurs rule-break in a qualitatively different way (than non-entrepreneurs)—doing so more rationally, by learning faster where opportunities for benefits exist and how to exploit them.

AREND 335 17 a novel repeatable game and method for the makers, and academics allocate resources more experimental study of rule-breaking behavior. efficiently to improve a targeted economy. Our experimental results complement We also hope that our specific initial inquiry several theoretical constructs. We complement can be extended. For example, on the empirical the concept of entrepreneurial discovery; we side, one extension would involve field studies show that it is not simply discovery that leads that parallel the game, like testing real-world to benefits, nor is it discovery and its blind decisions where there are opportunities to break exploitation, but it is the discovery with its rules, where there is a better and worse way to smart use that matters—where that smart use do so, and where the better way can be learned. comes from risky experimentation on the dis- One could also assess whether managers who covery’s use and aggressive action on feedback. wish to encourage rule-breaking in the context We also complement the conceptualization of of entrepreneurial discovery do so by better entrepreneurship as, at its best, creating new identifying and supporting individuals or do so value, that is, generating net welfare benefits. by providing better situational cues or both. Here, entrepreneurs, as sophisticated icono- As part of the current interest in mitigating clasts, increased the overall benefits generated rule-breaking (e.g., in sports through anti- in the game by acting in their own-self-interest, steroid policies), it may be useful to consider behavior that involved extra risk and effort. the analysis of the ethics of rule-breaking Our analysis leaves open several and varied behavior (e.g., Brenkert 2009) in order to paths for future work. For example, we could appreciate what is ultimately acceptable to the consider alternative experimental designs to economy and to society. And that question of investigate rule-breaking contagion rates among ethics may be even more important to assess differing proportional levels of entrepreneurs in with entrepreneurs, who cannot offer the same industries. We could consider alternative traits, level of attention to what rule-breaking could behaviors, and mediation effects in possible mean in terms of its importance for being the process models of rule-breaking behavior. Also, “right” decision (Robin, Reidenbach, and we could continue the testing of the subjective Forrest 1996). Regardless of the specific goal, entrepreneur view that Kor et al. (2007, p. 1188) however, gaining a deeper appreciation for the describe as “. . . the non-deterministic, evolu- who, how, what, and why of rule-breaking tionary nature of dynamic capabilities and entre- should lead to better economic outcomes for preneurial activities . . .” where there are several both entrepreneurs and society. further issues to consider regarding the ability of the entrepreneur to learn differently. For References example, one could test an entrepreneur’s per- Ambrose, M., M. Seabright, and M. Schminke sistence along a strategy path in the face of (2002). “Sabotage in the Workplace: The negative feedback. Or, we could test an entre- Role of Organizational Injustice,” Organiza- preneur’s proclivity to prod at, and experiment tional Behavior and Human Decision Pro- with, potential opportunities in a competitive cesses 89(1), 947–965. context over time, and how the rule features Ardichvili, A., and A. Gasparishvili (2003). interact with that context to stimulate rule- “Russian and Georgian Entrepreneurs and breaking (Lehman and Ramanujam 2009). Then Non-Entrepreneurs: A Study of Value we could try to define where and when such Differences,” Organization Studies 24(1), differentiated approaches pay off (e.g., only in 29–46. dynamic or ambiguous contexts?). At a more Axelrod, R. (1984). The Evolution of Coopera- basic behavioral level, it may be useful to deter- tion. New York: Basic Books Inc. mine the differences in belief-updating of entre- Balch, D. R., and R. W. Armstrong (2010). preneurs, for example, is it close to Bayesian “Ethical Marginality: The Icarus Syndrome updating? Is it asymmetrical regarding confirm- and Banality of Wrongdoing,” Journal of ing versus disconfirming feedback (like Business Ethics 92, 291–303. Kahneman and Tversky’s [1979] prospect Baltag, A. (2002). “A Logic for Suspicious theory)? Is it based on representativeness Players: Epistemic Actions and Belief- (Busenitz and Barney 1997)? What does it take to Updates in Games,” Bulletin of Economic make an entrepreneur change their opinion Research 54, 1–48. about an opportunity? Answers to such ques- Bateman, T. S., and J. M. Crant (1993). “The tions may help entrepreneurs, investors, policy- Proactive Component of Organizational

18336 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Behavior: A Measure and Correlates,” Academy of Management Review 23(3), 547– Journal of Organizational Behavior 14, 566. 103–118. Faysse, N. (2005). “Coping with the Tragedy of Baucus, M. S., W. I. Norton Jr., D. A. Baucus, the Commons: Game Structure and Design and S. E. Human (2008). “Fostering Creativ- of Rules,” Journal of Economic Surveys ity and Innovation without Encouraging 19(2), 239–261. Unethical Behavior,” Journal of Business Ferriani, S., G. Cattani, and C. Baden-Fuller Ethics 81, 97–115. (2009). “The Relational Antecedents of Bies, R. J. (2001). “Interactional (In)Justice: The Project-Entrepreneurship: Network Central- Sacred and the Profane,” in Advances in ity, Team Composition and Project Perfor- Organizational Justice. Eds. J. Greenberg mance,” Research Policy 38(10), 1545– and R. Cropanzano. Stanford, CA: Stanford 1558. University Press, 89–118. Foss, N., V. Mahnke, P. Cohendet, P. Llerena, Blau, P. M. (1964). Exchange and Power in and L. Marengo (2000). “Is There a Pilot in Social Life. New York: Wiley. the Evolutionary Firm?,” in Competence, Gov- Brenkert, G. G. (2009). “Innovation, Rule- ernance & Entrepreneurship. Eds. N. Foss Breaking and the Ethics of Entrepreneur- and V. Mahnke. New York: Oxford Univer- ship,” Journal of Business Venturing 24, sity Press, 95–117. 448–464. Gomez-Mejia, L. R., and D. B. Balkin (1989). Brockhaus, R. H. (1980). “Risk-Taking Propen- “Effectiveness of Individual and Aggregate sity of Entrepreneurs,” Academy of Manage- Compensation Strategies,” Industrial Rela- ment Journal 23, 509–520. tions 28(3), 431–445. Bull, I., and G. Willard (1993). “Towards a Greenberg, J., and B. Alge (1998). “Aggressive Theory of Entrepreneurship,” Journal of Reactions to Workplace Injustice,” in Dys- Business Venturing 8, 183–196. functional Behavior in Organizations: Busenitz, L. W., and J. B. Barney (1997). “Dif- Violent and Deviant Behavior. Eds. R. W. ferences between Entrepreneurs and Manag- Grifﬁn, A. O’Leary-Kelly, and J. Collins. ers in Large Organizations: Biases and Greenwich, CT: JAI, 83–117. Heuristics in Strategic Decision-Making,” Haber, S., and A. Reichel (2007). “The Cumu- Journal of Business Venturing 12, 9–30. lative Nature of the Entrepreneurial Process: Bygrave, W. B. (1998). “Building an Entrepre- The Contribution of Human Capital, Plan- neurial Economy,” Business Strategy Review ning and Environment Resources to Small 9, 11–18. Venture Performance,” Journal of Business Capelleras, J.-L., and F. J. Greene (2008). “The Venturing 22(1), 119–145. Determinants and Growth Implications of Hagedoorn, J. (1996). “Innovation and Entre- Venture Creation Speed,” Entrepreneurship preneurship: Schumpeter Revisited,” Indus- and Regional Development 20(4), 317–343. trial and Corporate Change 5, 883–896. Cherns, A. (1976). “Behavioral Science Engage- Hartog, J., M. Van Praag, and J. Van Der Sluis ments: Taxonomy and Dynamics,” Human (2010). “If You Are So Smart, Why Aren’t Relations 29(10), 905–910. You an Entrepreneur? Returns to Cognitive Collins, S. (1977). “Three-Level Fraud Control,” and Social Ability: Entrepreneurs versus Journal of Accountancy 143(6), 94–95. Employees,” Journal of Economics and Cressey, D. R. (1973). Other People’s Money. Management Strategy 19(4), 947–989. Montclair, NJ: Patterson Smith. Hayek, F. A. (1945). “The Use of Knowledge in Demmert, H., and D. B. Klein (2003). “Experi- Society,” The American Economic Review 35, ment on Entrepreneurial Discovery: An 519–530. Attempt to Demonstrate the Conjecture of ——— (1948). Individualism and Economic Hayek and Kirzner,” Journal of Economic Order. Chicago: Chicago University Press. Behavior and Organization 50, 295–310. ——— (1978). “Competition as a Discovery Dunne, T., M. Roberts, and L. Samuelson Process,” in New Studies in Philosophy, Poli- (1989). “The Growth and Failure of U.S. tics, Economics and the History of Ideas. Manufacturing Plants,” Quarterly Journal of Chicago: University of Chicago Press, 179– Economics 104, 671–698. 190. Elangovan, A. R., and D. L. Shapiro (1999). Henricks, M. (2005). “The Fast Lane,” Entrepre- “Betrayal of Trust in Organizations,” neur April, 65–67.

AREND 337 19 Howe, M. A., and C. A. Malgwi (2006). “Playing Opportunism,” Academy of Management the Ponies: A $5 Million Embezzlement Review 10(1), 59–65. Case,” Journal of Education for Business Manolova, T. S., N. M. Carter, I. M. Manev, and 82(1), 27–33. B. S. Gyoshev (2007). “The Differential Kahneman, D., and A. Tversky (1979). “Pros- Effect of Men and Women Entrepreneur’s pect Theory: An Analysis of Decision under Human Capital and Networking on Growth Risk,” Econometrica: Journal of the Econo- Expectancies in Bulgaria,” Entrepreneurship metric Society 57, 263–291. Theory and Practice 31(3), 407–426. Kirzner, I. (1973). Competition and Entrepre- Martin, R. (1993). “Changing the Mind of the neurship. Chicago: University of Chicago Corporation,” Harvard Business Review Press. 71(6), 81–94. ——— (1985). Discovery and the Capitalist Mayo, A., and N. Nohria (2005). “Zeitgeist Lead- System. Chicago: University of Chicago Press. ership,” Harvard Business Review 83(10), Kitzman, J., and D. Schierbeck (2005). “Entre- 45–60. preneurial Discovery and the Demmert/ Michael, S. C. (2003). “First Mover Advantage Klein Experiment: Another Attempt at through Franchising,” Journal of Business Creating the Proper Context,” The Review of Venturing 18(1), 61–80. Austrian Economics 18(2), 169–178. Mikulay, S. M., and R. D. Goffin (1998). “Mea- Knight, F. H. (1921). Risk, Uncertainty and suring and Predicting Counterproductivity in Profit. Chicago: University of Chicago Press. the Laboratory Using Integrity and Personal- Kor, Y. Y., J. T. Mahoney, and S. C. Michael ity Testing,” Educational and Psychological (2007). “Resource, Capabilities and Entrepre- Measurement 58(5), 768–790. neurial Perceptions,” Journal of Manage- Miles, R., and C. Snow (1978). Organizational ment Studies 47(7), 1187–1212. Strategy, Structure, and Process. New York: Leavitt, H. (1988). Managerial Psychology: McGraw-Hill. Managing Behavior in Organizations. Minniti, M., and W. Bygrave (2001). “A Chicago: Dorsey Press. Dynamic Model of Entrepreneurial Learn- Lee, K., M. C. Ashton, and K. H. Shin (2005). ing,” Entrepreneurship Theory and Practice “Personality Correlates of Workplace Anti- 25(3), 5–16. Social Behavior,” Applied Psychology 54, Mintzberg, H. (1990). “The Design School: 81–98. Reconsidering the Basic Premises of Strate- Lehman, D. W., and R. Ramanujam (2009). gic Management,” Strategic Management “Selectivity in Organizational Rule Viola- Journal 11, 171–195. tions,” Academy of Management Review Mintzberg, H., and J. A. Waters (1985). “Of 34(4), 643–657. Strategies, Deliberate and Emergent,” Strate- Lim, L. P., E. Garnsey, and M. Gregory (2006). gic Management Journal 6, 257–272. “Product and Process Innovation in Mitton, D. G. (1989). “The Complete Entrepre- Biopharmaceuticals: A New Perspective on neur,” Entrepreneurship: Theory and Prac- Development,” R&D Management 36(1), tice 13, 9–19. 27–36. Morrison, E. W. (2006). “Doing the Job Well: An McCabe, D., and L. Trevino (1993). “Academic Investigation of Pro-Social Rule Breaking,” Dishonesty: Honor Codes and Other Contex- Journal of Management 32(1), 5–28. tual Influences,” Journal of Higher Educa- North, D. C. (1990). Institutions, Institutional tion 64(5), 522–538. Change and Economic Performance. Cam- McMullen, J. S., and D. A. Shepherd (2006). bridge: Cambridge University Press. “Entrepreneurial Action and the Role of Nutt, P. (1999). “Surprising But True: Half the Uncertainty in the Theory of the Entrepre- Decisions in Organizations Fail,” Academy of neur,” Academy of Management Review Management Executive 13(4), 75–90. 31(1), 132–152. Nyborg, K., and K. Telle (2004). “The Role of Mahoney, J. T., and J. R. Pandian (1992). “The Warnings in Regulation: Keeping Control Resource-Based View within the Conversa- with Less Punishment,” Journal of Public tion of Strategic Management,” Strategic Economics 88(12), 2801–2816. Management Journal 13, 363–380. Ogbor, J. O. (2000). “Mythicizing and Reification Maitland, I., J. Bryson, and A. Van de Ven in Entrepreneurial Discourse: Ideology- (1985). “Sociologists, Economists, and Critique of Entrepreneurial Studies,” Journal

20338 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT of Management Studies 37(5), 605– Singh, G., and A. DeNoble (2003). “Views on 635. Self-Employment and Personality: An Parks, J. M., L. Ma, and D. G. Gallagher (2010). Exploratory Study,” Journal of Developmen- “Elasticity in the “Rules” of the Game: Explor- tal Entrepreneurship 8(3), 265–281. ing Organizational Expedience,” Human Sitkin, S. B., and A. L. Pablo (1992). Relations 63(5), 701–730. “Reconceptualizing the Determinants of Risk Penrose, E. T. (1959). The Theory of the Growth Behavior,” Academy of Management Review of the Firm. New York: John Wiley & Sons. 17, 9–38. Peters, G. (1985). “Warning Signs and Safety Skarlicki, D., and R. Folger (1997). “Retaliation Instructions: Covering All the Bases,” Risk in the Workplace: The Roles of Distributive, Management 32(9), 64–66. Procedural, and Interactional Justice,” Plaks, J., H. Grant, and C. Dweck (2005). “Vio- Journal of Applied Psychology 82(3), 434– lations of Implicit Theories and the Sense of 443. Prediction and Control: Implications for Stevenson, H., and J. C. Jarillo (1990). “A Para- Motivated Person Perception,” Journal of digm of Entrepreneurship: Entrepreneurial Personality and Social Psychology 88(2), Management,” Strategic Management 245–262. Journal 11, 17–27. Robin, D. P., R. E. Reidenbach, and P. J. Forrest Stock, G. N., N. P. Greis, and W. A. Fischer (1996). “The Perceived Importance of an (2002). “Firm Size and Dynamic Technologi- Ethical Issue as an Influence on the Ethical cal Innovation,” Technovation 22(9), 537– Decision-Making of Ad Managers,” Journal 549. of Business Research 35, 17–28. Stringer, R. (2000). “How to Manage Radical Robinson, S. L., and R. J. Bennett (1995). “A Innovation,” California Management Review Typology of Deviant Workplace Behaviors: 42(4), 70–88. A Multidimensional Scaling Study,” Academy Teece, D. J. (1986). “Profiting from Technologi- of Management Journal 38(2), 555–572. cal Innovation: Implications for Integration, Romanelli, E. (1989). “Environments and Strat- Collaboration, Licensing and Public Policy,” egies at Start-Up: Effects on Early Survival,” Research Policy 15, 285–305. Administrative Science Quarterly 34, 369– Tushman, M. L., and P. Anderson (1986). 387. “Technological Discontinuities and Organi- Rudolph, U., S. Roesch, T. Greitemeyer, and B. zational Environments.,” Administrative Weiner (2004). “A Meta-Analytic Review of Science Quarterly 31, 439–465. Help Giving and Aggression from an Vardi, Y., and Y. Weiner (1996). “Misbehavior Attributional Perspective: Contributions to a in Organizations: A Motivational Frame- General Theory of Motivation,” Cognition work,” Organization Science 7(2), 151– and Emotion 18(6), 815–848. 165. Russano, M. B., C. A. Meissner, F. M. Narchet, Venkataraman, S. (1997). “The Distinctive and S. M. Kassin (2005). “Investigating True Domain of Entrepreneurship Research: An and False Confessions Within a Novel Editor’s Perspective,” in Advances in Entre- Experimental Paradigm,” Psychological preneurship, Firm Emergence, and Growth, Science 16(6), 481–486. Vol. 3. Eds. J. Katz and R. Brockhaus. Green- Schumpeter, J. A. (1934). The Theory of Eco- wich, CT: JAI Press, 119–138. nomic Development. Cambridge: Harvard Whyte, G., A. M. Saks, and S. Hook (1997). University Press. “When Success Breeds Failure: The Role of ——— (1950). Capitalism, Socialism and Self-Efficacy in Escalating Commitment to a Democracy, 3rd ed. New York, NY: Harper. Losing Course of Action,” Journal of Orga- Sharma, A., and I. Kesner (1996). “Diversifying nizational Behavior 18(5), 415–432. Entry: Some Ex Ante Explanations for Post- Williamson, O. E. (1975). Markets and Hierar- Entry Survival and Growth,” Academy of chies: Analysis and Antitrust Implications. Management Journal 39, 635–677. New York: Free Press. Short, J. C., D. J. Ketchen, C. L. Shook, and R. Zahra, S. A., and M. Wright (2011). “Entrepre- D. Ireland (2010). “The Concept of ‘Oppor- neurship’s Next Act,” Academy of Manage- tunity’ in Entrepreneurship Research: Past ment Perspectives 25(4), 67–83. Accomplishments and Future Challenges,” Zhang, A., and R. D. Arvey (2009). “Rule Break- Journal of Management 36, 40–65. ing in Adolescence and Entrepreneurial

AREND 339 21 Status: An Empirical Investigation,” Journal competitive battle. That is the problem that you of Business Venturing 24(5), 436–447. are trying to solve, creatively. Zimbardo, P. G. (2007). The Lucifer Effect— You will be provided screen information for: Understanding How Good People Turn Evil. New York: Random House. (1) Choosing your strategic position for every period in every game. Appendix (2) Indicating when you are playing a NEW University of X Strategy Choice rival. (3) Indicating your last period’s score and the Experiment strategic positions implemented by you Welcome to the experiment on strategic and your rival*. choice. The number of grade points you earn (4) Indicating your running total score across will depend on your relative performance in all your games played. this experiment. If you have any questions (5) Commenting on your play when before or during the experiment, please raise appropriate. your hand. The Competitive Battle Background Per-Period Choices Businesses often compete against each Here, you have a simple choice between other, and do so by taking specific strategic THREE strategies per round: R, P, and S. These positions. Any specific strategic position could: represent three different competitive positions. (1) be advantageous relative to the rival’s As you can see from the table that follows, chosen position; (2) be disadvantageous rela- some combinations give advantage, some give tive to the rival’s chosen position; or (3) be at disadvantage, and some provide parity in parity with (e.g., cancelled out by) the rival’s payoffs to any one firm. chosen position. Also, research has suggested that taking any Your Your Partner’s Decision specific position can affect the choices of future Decision positions, perhaps reducing the options avail- RPS able. Thus, there is a history dimension to strategic choice. R0,0 0,2 3,0 Finally, research has also shown that the P2,0 0,0 0,1 strategic position chosen is not always the one S0,3 1,0 0,0 that gets implemented. Often, a firm will have trouble getting its employees to switch strategies, or may end up with what the employees The Impact of History have interpreted the plans for a strategy to be For each three-period game against any one rather than what the plans actually were. In rival, you do not choose any one strategy more other words, there is often “noise” in the than once. For example, if you chose R in the system of competitive strategies. first play, then you do choose P or S in the We offer a simplified model—a second period. simulation—of these issues for you to play. The Impact of Noise Description of the Task Note that there will be a low but positive You will play multiple competitive games probability of any player’s chosen strategy against 15 different anonymous partners. being “flipped” to another strategy by the com- Each set of games with the same partner puter. Thus, if you chose P in the example lasts three periods. mentioned earlier for the second period of The combined score across games is what play, it may well end up as R. *Note that ONLY determines the points you score for the course the final, noise-manipulated strategy choice grade (and extra credit). The scores will be will show up on the results screen (e.g., only R calculated compared with peers. will show up in the example) for your rival to The premise of the game is that you repre- see, and, similarly, you will only see your rival’s sent one of two firms that is engaged in a noise-manipulated strategy choice.

22340 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Journal of Small Business Business Management Management 2016 2015 54(1), ••(••), pp. pp. 341–355 ••–•• doi: 10.1111/jsbm.12148 Linking Small-Scale Commercial Activities and Women’s Health: The Jamu System in Urban Areas of Java, Indonesia by Maria Costanza Torri

Traditional medicine practices are widely documented and analyzed in Asia. Despite this, the interlinkages existing between small-scale commercial activities based on ethnomedicine and local health have been ignored. This paper attempts to overcome this shortcoming by analyzing the possible synergies existing between small-scale commercial activities centered on traditional herbal medicine in Indonesia (jamu) and health. The paper shows how the existence of these links in the city of Yogyakarta, Java, could represent a valuable basis to increment primary health care and enhance local livelihoods of rural women through commercial activities in the herbal sector.

wild plant gatherers and managers, home gar- Introduction deners and plant domesticators, herbalists and The relationship between poverty and poor healers, as well as seed custodians (Howard health conditions has been characterized as 2003; Howard-Borjas and Cuijpers 2002). synergistic and bidirectional (Judge and Despite the wide recognition of the rel- Patterson 2001; Wagstaff 2001). Poverty limits evance of ethnomedicine and medicinal plants the capacity to maintain good health and thus in the local health system and livelihoods of leads to further impoverishment, diminishing local communities, the interlinkages existing the potential of individuals and households to between these elements and the resulting syn- improve their economic well-being (Judge and ergies between them have been ignored Patterson 2001; Krishnan 1999; Wagstaff 2001). (Adams et al. 2005; Berkes, Colding, and Folke Despite the recognition of the connection 2003; Langwick 2007). between health and poverty, microcredit and Considering this gap, the objective of this small-scale commercial activities programs paper is to explore the associations between have, up to the present, been mainly focused health, active agency, and women’s commercial on the economic aspect, which remains the activities by examining small-scale commercial priority (Ronsmans and Graham 2006). activities in Indonesia which has health of rural Gender plays a key role in people’s access to women as its cornerstone. These small-scale use of and control over medicinal plants commercial activities are composed of medici- (Howard 2003) as in numerous societies, nal herb gatherers which transform and com- women have primary responsibility for health mercialize medicinal plants and produce care at the household level. Literature shows phytomedicines on the basis of traditional that, worldwide, it is mainly women who are knowledge in local ethnomedicine (jamu).

Maria Costanza Torri is an Assistant Professor in the Department of Sociology at the University of New Brunswick. Address correspondence to: Maria Costanza Torri, University of New Brunswick, Sociology, 9 Macaulay Lane, Fredericton, Canada E3B 5A3. E-mail: [email protected].

TORRI 341 1 These business initiatives carried out an infor- and Ashcroft (2002) characterize traditional mal scale in the urban settings of Java (city of medical systems as being “open” systems; they Yogyakarta) and presented an opportunity to accept inputs from, and are thus capable of, critically assess the transformative potential to functioning in and contributing to economic, improve both women’s livelihoods and local familial, ritual, moral, and other institutional health while promoting active agency among sectors. This openness makes such systems the women concerned. more adaptive to local needs (Richerzhagen The study of the possible links existing and Holm-Mueller 2004). between traditional medicine (TM) and small- In this article, I shall refer to TM or scale commercial activities could provide inter- ethnomedicine as a practice dealing with esting insights into innovative and integrated healing; they have it has the general character- approaches for the promotion of traditional istic of being integrated into daily life, and they health knowledge systems through gender are generally not documented or standardized commercial activities initiatives. Before pro- (Appadurai 1996; Ellen and Harris 2000). “Local” ceeding to elaborate on the case study case, and “indigenous knowledge” refer to the cumu- this paper focuses on the relationship between lative and complex bodies of knowledge, know- women’s health and poverty reduction, provid- how, practices, and representations that are ing a critical review of the current debates on maintained and developed by societies with these issues in development literature. The extended histories of interactions with their second part will analyze the case study in the surrounding natural environment. These knowl- city of Yogyakarta, Java, Indonesia, proving a edge systems, which include practices and per- discussion of the main results of the adoption ceptions, are integral to a cultural complex that of an integrated approach to traditional medici- also encompasses language, systems of classifi- nal practices. cation, resource use practices, social interactions, rituals, spirituality, and worldview. For Enterprises, Health, and TM: An rural and indigenous peoples and everyone else Underexplored Domain (in the sense that knowledge is always TM and medicinal plants have also a poten- indigenized or localized—see Appadurai 1996), tially important role in enhancing the socioeco- local knowledge informs decision-making about nomic development of local communities. fundamental aspects of day-to-day life. These Numerous studies have left little doubt that unique ways of knowing are important compo- herbal products based on local plants contribute nents of the world’s cultural diversity (Langwick to the well-being and, sometimes, the very 2007). survival of millions of poor rural households The interlinkages existing between gender (Arnold and Ruiz-Perez 2002; Belcher, commercial activities and health have been ana- Ruiz-Perez, and Achdiawan 2005; Fisher 2004; lyzed in the literature (Fisher, Holland, and Godoy et al. 2000; Narendran et al. 2001; Scherr, James 2001; Mushtaque, Chowdhury, and White, and Kaimowitz 2004). Natural product Bhuiya 2001). A positive relationship has been markets have been shown to be significant in identified between these factors. Key mecha- assisting rural households to realize some, if not nisms through which gender commercial activi- all, of their cash requirements (Marshall and ties can enhance health include increased Newton 2003; Marshall, Schreckenberg, and access to economic resources and overall Newton 2006; Narendran et al. 2001) and are improvements in material conditions (Lynch particularly crucial for the most marginalized et al. 2000). It has been demonstrated that and vulnerable segments of society (Beck and these mechanisms may improve health via the Nesmith 2001; Cavendish 2000), women production channel through two main actions: included. maintaining and protecting health, and restor- In numerous developing countries, plants ing health (Fisher, Holland, and James 2001). have long played an important role in both The increased experiences and roles as eco- preventive and curative health (German nomic actors may empower women through Technical Cooperation 2001; Hamilton 2004; greater control over income which, in turn, may Hirt and Pia 1995; Kartal 2006; Lawrence et al. increase their power in decision-making about 2005). This is partly due to poverty but also health care and their ability to access and pay because traditional systems are more culturally for the services. Research on women’s employ- acceptable than allopathic medicine. Schuklenk ment indicates that increased income may act

2342 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT as an enabling factor for the use of health (1) To analyze the cultural values and medicinal services (Chakraborty et al. 2003). Increased uses of traditional jamu pharmacy among income is associated with lower mortality and jamu sellers in the urban environment. morbidity and higher utilization of health ser- (2) To ascertain the ways in which traditional vices, even after controlling for other key practices can improve the health and well- factors (Anson 2004; Kuate 1997; Taguchi et al. being of entrepreneurs active in traditional 2003). Economic opportunity in terms of par- medicinal sector and their households. ticipation in credit programs has also been con- (3) To analyze up to which extent these sidered beneficial: A study in Bangladesh found commercial activities have allowed these that women who participated in credit pro- women to gain active agency in their health grams were more likely to use health care, decisions and to improve their access to controlling for other factors (Nanda 1999). health. Despite the availability of these studies, the possible links between health and commercial The research questions that will be explored activities have been explored in a general way, in this study are the following: without making reference to commercial activities focused on TM. Thus, this field of enquiry (1) What are the different uses of medicinal remains unexplored. plants among the sellers of herbal medicine This article also aims to bridge the gap exist- (jamu) in the city of Yogyakarta (Indone- ing in the literature on jamu sector in Indone- sia), and what is the relevance of traditional sia. The literature available on traditional jamu herbal medicine for the health for these mainly consists of studies in the domain of women as well as for their household? ethnopharmacology which document the use (2) How do these women perceive traditional and active ingredients of plant substances herbal medicine, and what are the main used for therapeutic purpose (Vera-Toscano, health beliefs associated with this health Phimister, and Weersink 2004). Several of these system? studies adopt an ecological approach and (3) In which ways has the participation in investigate the uses of medicinal plants in dif- these commercial activities improved the ferent regions (Anderson and Jack 2002). For capacity of these women in making deci- the greatest part, these ethnobotanical invento- sions regarding their health and their ries present botanical and pharmacologic data access to health care? without an analysis of the socio-cultural and economic contexts (Schippmann and Leaman Traditional medicinal system in urban areas 2003). Thus, they overlook the specific circum- has been previously analyzed in the literature stances and contexts in which the uses of (Anderson and Jack 2002; Schippmann and medicinal plants occur. Leaman 2003), but no study is available on the In contrast to general inventories are interlinkages existing between small-scale few studies that attempt more broadly to enterprise focusing on jamu and their impact contextualize perceptions about plants and their on health of women in Indonesia. The intent of therapeutic uses by linking pharmacological this study is to show how an ethnographic and socio-cultural aspects of jamu (Mafra and approach on herbal medicines can provide Stadler 2007; Taguchi et al. 2003). These studies insights on both the socioeconomic and thera- explore both biological and behavioral and cul- peutic values of traditional ethnomedicine tural parameters to formulate questions within among urban women in Java. the broad outlines of a human ecology that seeks to understand human–plant interactions Methods in the most comprehensive sense and to assess The study was carried out in the city of the impact of such behaviors on health. Yogyakarta, the capital of Yogyakarta Special This article will analyze the possible Region in Java, Indonesia, between June and interlinkages existing between women small- July 2010. scale commercial activities in the jamu sector A sample of 35 entrepreneurs was selected on operating in the city of Yogyakarta (Java) and the basis of parameters such as age, marital the health of its members. status, and economic conditions; this last In particular, the objectives of this study are element was measured on the basis of the the following: average budget available at household level.

TORRI 343 3 The age of the women ranged between 26 prised the use of key informants. These con- and 64. sisted in the local leaders of community and The sample has been selected among the members of Jamu Associations in Yogyakarta jamu sellers in the streets or in the local and knowledgeable members of the commu- markets of the city. The majority of the jamu nity in relation with ethnomedicine, such as sellers interviewed were living in the periurban elderly. The second method involved partici- or rural areas around Yogyakarta. They also pant observation with the jamu sellers during belonged to a low social class and generally the preparations of their products of herbal had a low level of education, usually elemen- medicine. The author lived with several jamu tary school, with a very small percentage who sellers during her stay in Yogyakarta while attained a secondary school diploma. The inter- observing the selection methods of the plants views had a duration of 45 minutes/one hour and the preparation techniques of jamu. The and consisted of structured and semistructured third method consisted of in-depth interviews questions. On the one hand, structured ques- with a selected number of jamu sellers tions aimed to collect quantitative data, such as recruited through the snow-ball technique. To income generation and household budgets. On reduce the risk of producing a biased sample, the other hand, semistructured questions aimed I have matched it with a sample of randomly to gather information about the different uses selected jamu sellers among the participants of of jamu within the household, with a special the Jamu Associations present in the city of focus on the health of women and children. Yogyakarta. Other questions aimed at gaining insights on In terms of data analysis, primary thematic the perceptions of jamu sellers on the thera- categories, such as the different therapeutic peutic effectiveness and safety of jamu. uses of jamu within the household, perceptions The data obtained from the interviews have on the therapeutic effectiveness of jamu of been complemented, and validity has been women and children, and perceptions of safety checked by two group interviews. These two of jamu, have been developed, followed by group interviews consisting of 10 participants codes for subthemes. Independent coders each were also organized on occasion of a established reliability in the use of the coding meeting of a Jamu Association active in the system before all excerpts had been formally neighborhood of Karanggayan. The selected analyzed. Overlaps or distortions of data and women were jamu sellers who belonged to the thematic categories have been refined, merged, Jamu Association but who had not been previ- or subdivided in an iterative process. Data have ously interviewed. been summarized thematically and with illus- Two local translators have been hired with trative quotes to capture the range of perspec- the support of the Jamu Associations. Being a tives represented in the interviews. woman has allowed me to have access to the interviewees and to be able to carry out the Results interviews at their households without experi- The Jamu Gendong in Indonesia and encing major problems. The complexity of the Household Health role of gender in the fieldwork and its possible Jamu gendong is liquid jamu, which is influence on the data has been carefully taken stored in bottles. It is usually sold by women into account, and to avoid the gender difficul- who carry the bottles in bamboo baskets slung ties that may arise when a male interpreter on their backs. Jamu gendong is cultivated approaches women, both of the translators extensively and may be seen growing in Java, who carried out interviews were female. Sumatra, and Kalimantan island in the cities, Upon consent of research participants, the towns, or remote areas. Most jamu gendong interviews have been audio-recorded. The sellers come from the Central Java province names of the interviewees have been omitted to (especially Wonogiri) and make their living respect their privacy. All the interviews have wandering about the country. They come from been transcribed in Javanese and subsequently a low socioeconomic stratum and have a low translated in English. level of education, usually elementary level. Three methods of data collection were According to Javanese custom, it is tradition- used, each of which views the interaction from ally the mother who is responsible for looking differing perspectives and requires different after the health of the family members, sampling strategies. The first method com- especially the children. This practice is

4344 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT well entrenched, especially both in rural and aurantifolia) and onion are also used. Fresh urban areas (Beers 2001; Hull 1985; Mangestuti materials are extracted from certain parts of 2007). plants (leaves, roots, flowers, fruits, trunk, The interviews with the jamu sellers show seeds) or from the entire plant. The jamu pro- that within a family, the tradition of using ducers usually utilize a number of plant jamu is introduced at a young age. These species; rarely is just a single plant used. The women generally learn and practice making number of plant species used for a particular jamu by observing their mothers or grand- kind of jamu usually ranges from five to 10. mothers, as well as by gathering information Certain jamus contain 30–40 species. from neighbors. Among the wild plants that are commonly Some training sessions are also organized in used are sambiloto (Andrographis paniculata), Yogyakarta in the local Jamu associations that tapak liman (Elephantopus scaber), patikan regroup jamu sellers active in the informal kebo (Euphorbia hirta), tapak dara (Vinca sector and that live in the same neighborhood. rosea), ciplukan (Physalis minima), kudu According to the jamu sellers, especially the (Morinda citrifolia), gempur batu (Ruellia younger ones, the training has been very naifera), sembung (Blumea balsamifera), important as it has allowed them to gain expe- pares (Momordica charantia), daun ungu rience on how to prepare the different types of (Pseuderthemum andersoniil), and pecut kuda jamu. Some jamu sellers, both the young and (Stachytarpheta jamaicensis). the elderly ones, have emphasized how these To make jamu drinks or external medicines, meetings are important to exchange jamu these herbs or spices are simply processed. The recipes among the participants and to exchange process of making jamu requires boiling, and their knowledge. kitchen utensils such as ceramic pots may be Apart from restoring health, jamu is also used. A traditional tool called lumpang or used for common illnesses and health problems pipisan, made of stone, can be used for pound- such as cold, flue, diarrhea, lack of appetite, ing. Jamu drinks can be made in the form of stomach ache, and indigestion. Fever resulting percolates (water-extracted) or concentrates of from infection is commonly treated with jamu jamu materials. To obtain the essence of the made from turmeric juice or water extracted raw materials, they are often cleaned and from papaya leaves. Crushed onion mixed chewed. Homemade jamu is usually made with kajuputi (Leucodedron leacodendra) oil when the need arises, so it is always fresh. The or coconut oil is commonly applied to children unused jamu is disposed of. with fever. Febrile convulsion is prevented by a Different names are devised for types of compound containing sangket (Moschosma herbal products containing different mixture polystachyum) leaves. Children with a poor of plants and roots. Kudu laos is believed to appetite are forcibly given temu ireng be good for hypertension and for improving (Curcuma aeruginosa), which is very bitter. blood circulation; Jamu pahitan relieves the Applying a paste of heated castor leaves on the itching caused by allergies, parasitic infections, abdomen relieves abdominal disorders (colic and diabetes mellitus; Jamu cabe puyang is and dyspepsia), whereas chewing young guava good for muscle stiffness and fatigue, whereas leaves is effective in treating diarrhea. jamu sinom refreshes and cools the body; A jamu drink like beras kencur, which is Jamu beras kencur improves the appetite and made from finely pounded rice and kencur relieves fatigue; Jamu kunci suruh is helpful (Kaempferia galanga), tamarind flesh, and red in cases of leucorrhoea or excessive vaginal sugar, is used to reduce the edema and pain discharge, and jamu uyup-uyup stimulates caused by a sprain. Rheumatic pain can be lactation. treated by rubbing the painful area with finely pounded ginger. Rubbing castor sap or papaya Therapeutic Uses of Jamu among leaves, kamboja (Plumeria acuminata)or Jamu Sellers patah tulang (Pedilanthus pringlai Robins) From a health perspective within the house- sap, on a fresh cut/wound can provide imme- hold, taking part into jamu commercial activi- diate relief. ties rose confidence of herb use in a primary The materials that jamu sellers and produc- health care context among the jamu sellers. The ers use to make jamu include turmeric, ginger, interviews showed that jamu can provide an and garlic. Some fruits like nipis lemon (Citrus effective treatment for the most common health

TORRI 345 5 problems within the family. From the inter- energy of the future mother and will allow her views, it resulted that the members of the to go more easily through the bodily and hor- family that most benefited medically from the monal changes. consumption of jamu were the women and The findings of this investigation are also children, who accounted for nearly two-thirds confirmed by Handayani, Suparto, and Suprapto of the cases treated with home remedies and (2001) in a study among the Maduranese women jamu. in north-east Java. This study noted that women Jamu is generally consumed by all the use both in the early stages of pregnancy and in members of the family, children included. The the later stages of pregnancy to relieve fatigue interviews highlighted how the jamu seller also and pain. Two-thirds of the jamu sellers also plays an important role within the community, declared that they used jamu during the period being consulted by the neighbors when they that followed their pregnancy. According to the have a health problem. A woman in her late 20s interviewees, drinking jamu in this phase is affirms: useful to remove blood residue and mucus from the uterus and to restore the uterus and I drink jamu when I am not feeling well, abdomen to their former size. A third of the for example when I have pains because interviewees (i.e., 12) also stressed that there of my period or when I have a cold .... was also a type of jamu that was very useful to I can also cure my family member if they fight possible infections immediately after the are sick. It is not only I who drink the delivery. A middle-aged woman explains: herbal drink but also my husband and my children. But they drink if they are It is important to drink jamu after the tired or sick. Sometimes my neighbors delivery . . . my aunt used to say that she also ask what kind of herbal drink that never had any problems or infections they can drink if they are sick. because she was drinking jamu after having given birth to her children . . . The interviews show that jamu is used for its she had 8 children all together . . . a therapeutic effect not only to cure common woman should drink jamu for forty days complaints, such as cold, cough, fever, pain, after the delivery. Before I started to sell stomach-ache, etc., but also to enhance the jamu, when I had my first child, I did not immune system and to energize the body. Jamu drink jamu very often . . . It was not easy for this preventive purpose is generally con- to go to buy it as I had to stay with my sumed early in the morning once a day or every baby . . . I spent all the time at home two days. during the first month after the delivery A jamu seller affirms: . . . when I had my second and third child I started drinking jamu after the My whole family drink jamu like beras delivery on daily basis . . . I could really kencur and hangat jahe (types of jamu). notice the difference . . . I was feeling I make it by myself. It can prevent with energy and strength and I have certain sickness such as cold and flu. never had a fever. Jamu can also warm our body and this is good especially during the first months Numerous jamu sellers (i.e., 21 out of 35) of pregnancy. We don’t drink jamu also emphasized that they often brought jamu everyday but three times a week. We to their neighbors or the close-by family after a drink it in the morning. woman had a delivery. An old jamu seller affirms: The majority of jamu sellers affirmed that they took jamu during pregnancy (around 82 [T]here are many women in their percent of the interviewees) and during lacta- neighbourhood who asked me to bring tion (88 percent) to improve breast milk quality them jamu in their houses on daily basis and quantity. Indeed, there is the common . . . my relatives live close to my house belief that if the future mother drinks jamu, the too and I used to bring the jamu to my fetus will develop more easily and that the baby cousins when they had their babies . . . will not be underweight. It is also believed that this helped them a lot as after the deliv- consuming jamu will increase the strength and ery the women normally stay at home for

6346 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT 30–40 days and cannot expose them- 2005). There may be some positive aspects of selves too much to the light of the sun. self-medication when performed according to rational criteria (Borg and Scicluna 2002). Nev- A few jamu sellers (i.e., four) affirmed that ertheless, self-medication with TM can cause they used a type of jamu whose main ingre- problems when it involves specific diseases or dients are turmeric and curcuma. This type of serious health problems that would need a spe- jamu is believed to sooth the uterus inflam- cific medical intervention (e.g., infections). mation after a miscarriage and to restore Having emphasized the harmful effects of such stamina. practices, it is important to highlight how the The results of this research emphasize also use of jamu to remedy these “so-called ills” can that a few jamu sellers (five out of 35), espe- be detrimental for health as it can encourage cially the women in their 20s and 30s, believed people toward practices of self-medication that it was not appropriate to drink hot jamu when a medical advice is urgently needed. (i.e., the one containing “hot” plants such as The interviews also show that the consump- jhai [Zingiber officinale]) during the pregnancy. tion of jamu is believed to be effective also to This group of jamu sellers declared that they prevent the emergence of infectious diseases in were not drinking jamu during the pregnancy young children, such as cold and flu by rein- as they believed that its use could have reduced forcing their immune system. The jamu sellers the contractions. interviewed affirmed that after they started pro- The questions related to efficacy and safety ducing and selling jamu on daily basis, it was of jamu and its interaction with other allopathic much easier for them to give their children drugs are currently largely unanswered as there jamu once a day, especially during the cold are insufficient data to show that TM is safe season, when it is easier to get ill. Eighteen (Chard, Lilford, and Gardiner 1999; Ernst 2000; women out of 35 affirmed that this cure had Finger and Schuler 2004; Ghimire, McKey, and been very effective to reduce the insurgence of Aumeeruddy-Thomas 2005). flu and cold in their children or at least to The jamu sellers also emphasized how they reduce the intensity of the systems and the used the jamu products that they sold to cure duration of the disease. some of their gynecological problems, in par- Another pediatric use of the jamu is to ticular the leucorrhea. This condition is char- increase the appetite of children and babies. acterized by a whitish vaginal discharge, and it The interviews highlight that the jamu sellers is mostly common during pregnancy. Leucor- often used this type of jamu with their own rhoea can be accompanied by a strong and children, if the occasion arose. On this respect, offensive odor and/or vaginal itching. Accord- a jamu seller explains: ing to 11 out of 35 jamu sellers interviewed, drinking jamu can reduce the problem of [W]hen my children lose their appetite, I vaginal discharge. These women confirmed prepare a jamu called endak-endak that their health condition improved since cacing . . . when children lose their when they have been drinking jamu. The use appetite, most probably there is a worm of jamu for gynecological problems has par- in their stomach . . . This can easily ticularly been emphasized by the interviewees. happen to children because they usually Being a Muslin country, the women feel play outside and forget to wash their intimidated and not at ease with dealing with hands. To cure this illness, we use to gynecologists, especially male doctors. Thus, prepare jamu with a local plant temu jamu products can offer an easy solution to ireng. First of all, we clean it, and then tackle with these problems. Despite this, the we grate it. After that, we take the safety and efficacy of jamu used against essence of temu ireng. After drinking it, vaginal discharges are questionable, as some- the worm will die and the illness can be times, vaginal discharges are due to infections cured. and may need antibiotic treatment to obtain sufficient cure. From the interviews with old jamu sellers, it The nature and extent of self-medication emerged that they were also using jamu for vary in different cultural contexts, and social geriatric purposes, especially for rheumatisms and educational influences may be greater than and join pains. A 75-year-old jamu seller the influence of medical practice (Sharma et al. explains:

TORRI 347 7 I prepare a type of jamu when I have to Allah . . . I am afraid to consume pains in my bones and I cannot walk medicine from drug store because there properly. To cure this illness, we use are side effects and besides I have no bengle (a kind of herbal plant). The idea how they are produced. Jamu is length of this plant is about 7 cm. First of safer since it is made from herbs and all, we clean it with water. After that we there are no chemical ingredients which cut it and grate it. Then, we dry it. We are added. The process of making jamu only take 0.5 mg of bengle. After that, is also safe and natural . . . I prepare we pound it and ﬁlter it. After that we jamu every day and I know the ingredi- pour hot water and it is ready to be ents and the preparation methods . . . consumed. This herbal can cure rheuma- that is why I prefer jamu to synthetic tisms effectively. drugs.

The positive effects of jamu to relieve joint Finally, by providing an alternative to bio- pains are also emphasized by younger jamu medical clinical care, women’s knowledge of sellers. The latter affirmed to use different medicinal plants affords them a degree of inde- decoctions to reduce the pain of their old pendence from the public health care system. parents or relatives. Within the household, women reported that their experience in the area of medicinal plants Before I become a jamu seller and I took is a source of pride and authority. Women’s some training courses at the Jamu asso- control over medicinal plants stands in stark ciation, I was unable to cure my parents contrast to their position in the public health who, being old, are suffering from joint care system. pains . . . We used to spend quite a lot of A middle-aged woman declared: money in allopathic medicines and in medical consultations but without good Here the health system is quite expen- results . . . now, after drinking this type sive . . . if you don’t have a health insur- of jamu every day, they are feeling better ance you end up paying high fees for and the intensity of the pain has medical care and treatments . . . before decreased. when I was not very involved in jamu preparation, I used to buy chemical Enhanced Autonomy in Health Care. One drugs in the chemist shops . . . they are theme that repeatedly emerged in the conver- quite expensive and sometimes I was sations with jamu sellers about medicinal plants feeling even worse after having taken and jamu was the issue of control over the them . . . I was suffering from headaches health care process. The jamu sellers empha- and started taking some pills . . . after sized that since when they used jamu, they felt some time my headache disappeared but in control of the health care process in their I started having pain in my stomach family. Jamu sellers who became experts in because of all the pills I took . . . Now medicinal plants, thanks to their commercial that I sell jamu I am very happy . . . I use activity, declared that they utilized their knowl- the products I sell in the market also edge and expertise to decide which plant(s) to with my family and I feel I can make a use and prepare and administer a home difference to my children and my remedy. Women’s therapeutic knowledge husband. about medicinal plants gives them confidence in their ability to care for ill family members. In The interviews also show that selling jamu fact, jamu sellers often boasted of their medici- products had enabled the jamu sellers and nal plant knowledge. their families to improve their diet through The use of jamu also gives women a sense of the consumption of herbs, such as the control because they know the medicine they sambiloto (Andrographis paniculata), tapak give their family members is safe since they liman (Elephantopus scaber), and the patikan participate in the preparation process. kebo (Euphorbia hirta), that are used both for their therapeutic values and for their nutri- If I feel tired or ill I can made my own tious values, being rich in iron and vitamins. herbal drink and feel better . . . All praise One-fourth of the jamu sellers affirmed that

8348 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT the use of these plants in the jamu was ben- Indirect Impact of Jamu Commercial Activities eficial as a diet supplement too. Plant medi- on Health. The participation of women in the cines administered for the mentioned jamu activities has also had some indirect conditions would work far better if accompa- effects on the health of the family through an nied by behavioral and diet changes. This is increased household budget due to revenue an important contrast with allopathic medi- increase for the sale of jamu and the reduction cine, which often ignores such dietary regimes of expenditures for health. or underestimates their importance. It is The factor of the reduction of expenditures important to promote the interlinkages exist- for health has been especially emphasized by ing between the different uses of plants younger women who received their training in through their possible cultivation in home jamu preparations organized by the local Jamu gardens and to favor this trend through edu- association. These women highlighted how cation and increase of awareness of the before the training they only knew the prepa- importance of the combined values of medici- ration of simple home remedies and they were nal and edible plants as they can enhance not feeling confident about using plants and both nutrition and health among both lower roots to cure their family. and middle classes. This is confirmed by a statement of a woman Table 1 summarizes the different forms of in her late 20s who affirms jamu consumed by the sellers, the plants most commonly used, as well as the most common Before receiving my training and becom- therapeutic uses of local plants. ing a jamu seller I used to buy drugs in

Table 1 Different Forms of Jamu

Forms of jamu consumed Jamu kunci suruh: type of jamu believed to reduce excessive vaginal discharge Kudu laos: type of jamu believed to be good for hypertension and blood circulation Jamu pahitan: type of jamu believed to relieve allergies, parasitic infections, and diabetes Jamu sinom: type of jamu believed to refresh the body Jamu uyup-uyup: type of jamu believed to stimulate lactation Number of plants most commonly used It generally ranges from five to 10 of plants. Certain types of jamu may contain between 30 and 40 species. Plants most commonly used by the sellers Sambiloto (Andrographis paniculata), tapak liman (Elephantopus scaber), patikan kebo (Euphorbia hirta), tapak dara (Vinca rosea), ciplukan (Physalis minima), kudu (Morinda citrifolia), gempur batu (Ruellia naifera), sembung (Blumea balsamifera), pares (Momordica charantia), daun ungu (Pseuderthemum andersoniil), and pecut kuda (Stachytarpheta jamaicensis) Members of the family that consume jamu All the members of the family, in particular children and women Uses of jamu for reproductive health Jamu is believed to remove blood residue and mucus from the uterus and to restore the uterus and abdomen to their former size; to improve breast milk quality and quantity; to relieve fatigue and pain during and after the pregnancy; to fight possible infections immediately after the delivery; and to reduce vaginal discharge. Pediatric uses of jamu Jamu is believed to increase the appetite of children and babies and to boost the immune system to reduce the insurgence of flu and cold.

TORRI 349 9 a chemist shops for expensive price. It to enhance or undermine women’s sense of was about 25.000 rupiah. Those stores autonomy, especially within their capacity to only want to get money from the cus- decide on the allocation of money which tomers . . . being a jamu producer I can directly or indirectly impact health (Thachuk now use my products to cure myself and 2007). Kabeer (2003) reinforces the concept of my family so I don’t need to buy relational autonomy by differentiating between anything. “passive agency” (actions taken when there is little choice) versus “active agency” (purposeful In terms of indirect effects of jamu activity behavior). According to Kabeer (2003), agency on health, 25 out of 35 interviewees afﬁrmed encompasses not only active choice (decision- that the money gained with the sales of jamu making and negotiation capacity) but also the was used to buy clothing and more nutritious meaning, motivation, and purpose that indi- food for the children, which can allow enhanc- viduals bring to their actions. That is, an indi- ing the well-being of the family. Studies vidual with active agency should be able to show that there is a clear link between malnu- deﬁne self-interest and choice, and consider trition and diseases, especially among young themselves as not only able but entitled to people (Chaudhury and Rafei 2001; Hartini make choices (Nussbaum 2000). Resources are et al. 2005). seen as the means to improve women’s active The interviewees also stressed that the agency. savings obtained through the jamu activity This aspect of active agency, power relation- could be used as a health insurance in case of ships, and health within the household is par- health problems. A woman in her late 30s ticularly delicate and needs to be explored explains: more in detail in other studies.

[S]ince when I am a jamu seller, I can The Issue of Therapeutic Effectiveness save some money for the family. Nor- and Safety of Jamu mally we are able to save 5000 rupiah Regarding the perceptions on the effective- per month in case we need it someday. ness of using jamu, more than half (i.e., 17) of This is very important indeed . . . the jamu sellers interviewed, especially the imagine that my husband and I we get young mothers with several children, affirmed sick and we cannot work . . . we have that it was better to use herbal medicines small children and we need to provide instead of allopathic drugs for minor health for them . . . but we can use these complaints as jamu has no side effects and is savings if we cannot earn money when freshly made and thus was more natural and we are sick. effective. Many interviewees also felt that jamu works gently and slowly in comparison with Despite this, some conflicts could arise inside conventional medicines and that they are less the household, especially regarding the decision powerful. of how to spend the money. Research casts some A few jamu sellers interestingly attributed doubt on assumptions about automatic positive the therapeutic effectiveness of jamu to the effects on gender equality and gender commer- ancient preparation methods and technical pro- cial activities, especially in traditional societies cedures they followed: (Bangser 2002; Fisher, Holland, and James 2001; Mushtaque, Chowdhury, and Bhuiya 2001). I produce jamu as my grandmother used Much of the emphasis on social status and active to do . . . There is no difference. I agency as a process is found in the conceptual prepare jamu based on the recipe from literature of commercial activities, but this our ancestor. For example I don’t use understanding is only beginning to be incorpo- anymetallic tools. I only use tools that rated into the frameworks of studies on health. are made of wood and stone. I boil the The literature on gender and health recog- ingredients by using coal and wood, not nizes that many inequalities in health are a gaz. My grandmother used to say that result of power relations that have an effect on there are four elements: air, land, fire, the distribution of resources (Armstrong 2009). and water. So, if we use tools from In particular, at household level, power rela- metals, we are using an element that is tions within the household have the potential not present among the natural elements

10350 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT . . . This can reduce the capability to cure exact ingredients. If the proper ingredi- the illness. My jamu is effective as I don’t ents are not mixed in an appropriate use any metallic utensil. way, the jamu can make people sick.

In the present study, two-thirds of the jamu Though the existence of jamu medicine is sellers interviewed believe that jamu has wide potentially of great benefit to people living in acceptance in their family and that it was pref- Indonesia, further research on how traditional erable to allopathic medicine. The main reason and allopathic systems of care interact is behind the preference of jamu to modern medi- needed so as to investigate not only the poten- cine was that the former is considered to be tial beneficial and synergistic effects but also more efficacious than the later, especially in potential toxicities and drug interactions. Pre- terms of preventive cure. This finding is in vious studies have shown that some natural agreement with previous studies (Balick et al. medicines such as botanicals and herbal prod- 2000; Kerr and Saryan 1986; Mason, Tovey, and ucts can be potentially harmful to patients, and Long 2002). thus, this is a research area of crucial impor- One-third of the interviewees affirmed that tance that requires further investigation (Abebe jamu was not only effective to cure minor dis- 2002; Huang, Wen, and Hsiao 1997; Ikawati, eases such as cough and cold or to energize the Wahyuono, and Maeyama 2001; Pramono body but that could also be effectively used to 2002). cure more serious health conditions, such as liver and heart diseases. In this respect, a Conclusions middle-aged jamu seller affirms: The case study of small-scale commercial activities selling jamu represents an interesting There is a tendency that they underesti- experience which combines the livelihoods mate jamu. Well, actually jamu can cure with local health enhancement and the conser- serious illness such as liver problems vation of ethnomedicine through commercial and heart disease. I have tried to cure my activity. Reduced dependence vis-à-vis health uncle who had cancer with a type of services from outside, improvement in jamu . . . at the beginning he was feeling women’s access to health and market resources better but unfortunately, as the cancer at household and community level, and was very advanced I was unable to help increased active agency in health can be iden- him and he died. If the cancer was in the tified as the main positive outcomes of this first or second stage, I believe I would commercial initiative. have been able to cure him. Although the therapeutic values of plants for reproductive health have not always been It can be noticed here how the allegedly documented in studies nor proven through sci- positive therapeutic effectiveness of jamu in entific evidence, numerous plants are widely tackling diseases such as cancer could possibly used by urban women to meet their reproduc- produce misleading therapeutic choices, such tive health needs and have maintained their as the total reliance of herbal treatment also in popularity. To date, scientific evidence from the case of grave illnesses. randomized clinical trials is only strong for a From the interviews, it also emerged that the limited number of herbal remedies (Anderson majority of the jamu sellers overall are not 2005). Although further research is needed to totally aware of the risks of jamu and herbal ascertain the efficacy and safety of medicinal medicines. When asked about the risks of jamu plants, the continued valuing and validation of and herbal medicines, the most common local health knowledge are an ongoing neces- response was “Don’t know.” sity in international development campaigns A minority of jamu sellers (i.e., seven out of (Hollenberg et al. 2008). 35) emphasized how the uses of the plants can It is important to promote the understanding be dangerous if the doses and the combination of the different uses of plants both for gyneco- of the herbs are not precisely established. logic and pediatric uses and to support this local knowledge through education, and It is important to know how to use increased awareness of the importance of plants before making jamu . . . It will be the combined value of this form of TM for dangerous if people don’t know the reproductive health. There is also the need to

TORRI 351 11 understand through statistical ad hoc studies Arnold, J. E. M., and M. Ruiz-Perez (2002). “Can what are the effects of jamu on the morbidity Non-Timber Forest Products Match Tropical related to reproductive health. Forest Conservation and Development The patterns of taking jamu deserve atten- Objectives?,” Ecological Economics 3(9), tion as many types of jamu are consumed regu- 437–447. larly as preventive as well as curative medicine. Balick, M., F. Kronenberg, A. Ososki, M. Reiff, Thus, by understanding the cultural beliefs and M. Fugh-Berman, A. O’Connor, B. Roble, M. practices of the use of jamu, commercial activi- Lohr, and D. Atha (2000). “Medicinal Plants ties could be generated which are not just able Used by Latino Healers for Women’s Health to improve the livelihoods of local women but Conditions in New York City,” Economic also to have positive repercussions on the Botany 5(4), 344–357. health and well-being of the family. Bangser, M. (2002). “Policy Environments: Mac- Despite the challenges pertaining to efﬁcacy roeconomics, Programming, and Participa- and safety of these herbal products, the small- tion,” in Engendering International Health: scale businesses centered on jamu should be The Challenge of Equity. Eds. G. Sen, A. encouraged and sustained through policies that George, and P. Östlin. Cambridge: MIT entice change and institutional innovation in Press, 256–280. order to revitalize TM and reinforce the links Beck, T., and C. Nesmith (2001). “Building on between culture, health, and socioeconomic Poor People’s Capacities: The Case of development of local sellers. Common Property Resources in India and West Africa,” World Development 29(1), 119– References 133. Abebe, W. (2002). “Herbal Medication: Poten- Beers, S. J. (2001). Jamu, the Ancient Indone- tial for Adverse Interactions with Analgesic sian art of Herbal Healing. Jakarta: Periplus Drugs,” Journal of Clinical Pharmacy and Editions (HK) Ltd. Therapeutics 2(7), 391–401. Belcher, B., M. Ruiz-Perez, and R. Achdiawan Adams, V., S. Miller, S. Craig, S. Nyima, L. (2005). “Global Patterns and Trends in the Droyoung, and M. Varner (2005). “The Chal- Use and Management of Commercial NTFPs: lenge of Cross-Cultural Clinical Trials Implications for Livelihoods and Conserva- Research: Case Report from the Tibetan tion,” World Development 33(9), 1435– Autonomous Region, People’s Republic of 1452. China,” Medical Anthropology Quarterly Berkes, F., J. Colding, and C. Folke (2003). 19(3), 267–289. Navigating Social-Ecological Systems: Anderson, A. R., and S. L. Jack (2002). “The Building Resilience for Complexity and Articulation of Social Capital in Entrepre- Change. Cambridge: Cambridge University neurial Networks: A Glue or Lubricant?,” Press. Entrepreneurship and Regional Develop- Borg, M. A., and E. A. Scicluna (2002). “Over- ment 1(4), 193–210. the-Counter Acquisition of Antimicrobial Anderson, E. N. (2005). Everyone Eats: Under- Drugs in the Maltese General Population,” standing Food and Culture. New York: New International Journal of Antimicrobial York University Press. Agents 2(2), 253–257. Anson, O. (2004). “Utilization of Maternal Cavendish, W. (2000). “Empirical Regularities Care in rural HeBei Province, the People’s in the Poverty–Environment Relationship in Republic of China: Individual and Structural Rural Households: Evidence from Zimba- Characteristics,” Health Policy 70(2), 197– bwe,” World Development 28(11), 1979– 206. 2003. Appadurai, A. (1996). Modernity at Large. Cul- Chakraborty, N., M. A. Islam, R. I. Chowdhury, tural Dimensions of Globalization. London W. Bari, and H. H. Akhter (2003). “Determi- and Minneapolis: Minneapolis and London nants of the Use of Maternal Health Services University of Minnesota Press. in Rural Bangladesh,” Health Promotion Armstrong, P. (2009). “Public Policy, Gender, International 18(4), 327–337. and Health,” in Social Determinants of Chard, J., R. Lilford, and D. Gardiner (1999). Health: Canadian Perspectives, 2nd ed. Ed. “Looking Beyond the Next Patient: Sociol- D. Raphael. Toronto: Canadian Scholars’ ogy and Modern Health Care,” Lancet 3(3), Press, Inc., 350–361. 486–489.

12352 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Chaudhury, R. R., and U. M. Rafei (2001). Tra- Hirt, M., and M. Pia (1995). Natural Medicine ditional Medicine in Asia. Geneva: World in the Tropics: Tropical Plants as a Source Health Organization. 318p. of Health Care, Production of Medicines Ellen, R., and H. Harris (2000). “Introduction,” and Cosmetics. Kisubi, Uganda: Marianum in Indigenous Environmental Knowledge Press. and its Transformations. Critical Anthropo- Hollenberg, D., D. Zakus, T. Cook, and X. Xu logical Perspectives. Eds. R. Ellen, P. Parkes, (2008). “Re-positioning the Role of Tradi- and A. Bicker. Sydney: Harwood Academic tional, Complementary and Alternative Medi- Publishers, 1–34. cine as Essential Health Knowledge in Ernst, E. (2000). “Risks Associated with Global Health: Do They Still Have a Role to Complementary Therapies,” in Meyler’s Side Play?,” World Health and Population 10(4), Effects of Drugs, 14th ed. Eds. M. N. G. 62–75. Dukes and J. K. Aronson. Amsterdam: Howard, P. (Ed.) (2003). Women and Plants: Elsevier, 1649–1681. Gender Relations in Biodiversity Manage- Finger, J. M., and P. Schuler (2004). Poor Peo- ment and Conservation. London and New ple’s Knowledge. Promoting Intellectual York: Zed Books. Property in Developing Countries. Oxford: Howard-Borjas, P., and W. Cuijpers (2002). Oxford University Press. “Gender and the Management and Conser- Fisher, E., J. Holland, and S. James (2001). vation of Plant BiodiversIty,” in Biotechnol- “Becoming Poverty Focused: Implications ogy. Eds. H. W. Doelle and E. Da Silva. for Health Actors,” Development (Cam- Oxford, UK: Encyclopedia of Life Support bridge, England) 44(1), 22–30. Systems (EOLSS). Fisher, M. (2004). “Household Welfare and Huang, W. F., K. C. Wen, and M. L. Hsiao Forest Dependence in Southern Malawi,” (1997). “Adulteration by Synthetic Therapeu- Environment and Development Economics tic Substances of Traditional Chinese Medi- 9(2), 135–154. cines in Taiwan,” Journal of Clinical German Technical Cooperation (2001). Medici- Pharmacology 3(7), 344–350. nal Plants: Biodiversity for Healthcare. Hull, V. J. (1985). “Breastfeeding, Birth Spacing Germany: Eschborn. and Social Change in Rural Java,” in 1985. Ghimire, S. K., D. McKey, and Y. Aumeeruddy- Breastfeeding, Child Health & Child Spacing: Thomas (2005). “Heterogeneity in Ethnoe- Cross-Cultural Perspectives. Eds. V. J. Hull cological Knowledge and Management of and M. Simpson. London: Croom Helm, Medicinal Plants in the Himalayas of Nepal: 78–109. Implication for Conservation,” Ecology and Ikawati, Z., S. Wahyuono, and K. Maeyama Society 9(6), 123–135. (2001). “Screening of Several Indonesian Godoy, R., D. Wilkie, H. Overman, A. Cubas, G. Medicinal Plants for Their Inhibitory Effect Cubus, and J. Demmer (2000). “Valuation of on Histamine Release from RBL-2H3 Cells,” the Consumption and Sale of Forest Goods Journal of Ethnopharmacology 7(5), 249– from a Central American Rainforest,” Nature 256. 4(6), 62–73. Judge, K., and I. Patterson (2001). “Poverty, Hamilton, A. C. (2004). “Medicinal Plants, Con- Income Inequality and Health,” Treasury servation and Livelihoods,” Biodiversity and Working Paper 01/29, Wellington. Conservation 1(3), 1477–1517. Kabeer, N. (2003). Gender Mainstreaming in Handayani, L., H. Suparto, and A. Suprapto Poverty Eradication and the Millennium (2001). “Traditional System of Medicine in Development Goals. A Handbook for Policy- Indonesia,” in Traditional Medicine in Asia. Makers and Other Stakeholders. Ottawa: Eds. R. R. Chaudhury and U. M. Rafei. Commonwealth Secretariat/IDRC/CIDA. Geneva: World Health Organization, 47– Kartal, M. (2006). “Intellectual Property Protec- 69. tion in the Natural Product Drug Discovery, Hartini, T. N. S., R. S. Padmawati, L. Lindhlom, Traditional Herbal Medicine and Herbal A. Surjono, and A. Winkvist (2005). “The Medicinal Products,” Phytotherapy Research Importance of Eating Rice: Changing Food 21(2), 113–119. Habits among Pregnant Indonesian Women Kerr, H. D., and L. A. Saryan (1986). “Arsenic during the Economic Crisis,” Social Science Content of Homeopathic Medicines,” Clini- and Medicine 61(1), 199–210. cal Toxicology 2(4), 451–459.

TORRI 353 13 Krishnan, T. N. (1999). “Access to Health Care Mushtaque, A., R. Chowdhury, and A. Bhuiya and the Burden Of Treatment in India: An (2001). “Do Poverty Alleviation Programmes Interstate Comparison,” in Disinvesting in Reduce Inequities in Health? The Bangla- Health: The World Bank’s Prescriptions for desh Experience,” in Poverty, Inequality and Health. Ed. M. Rao. New Delhi: Sage Publi- Health. An International Perspective. Eds. D. cations, 122–146. Leon and G. Walt. Oxford: Oxford University Kuate, D. B. (1997). “Effects of Socioeconomic Press, 312–332. Disadvantage and Women’s Status on Nanda, P. (1999). “Women’s Participation in Women’s Health in Cameroon,” Social Rural Credit Programmes in Bangladesh and Science and Medicine 44(7), 1023–1042. Their Demand for Formal Health Care: Is Langwick, S. (2007). “Devils, Parasites, and There a Positive Impact?,” Health Economics Fierce Needles: Healing and the Politics of 8(5), 415–428. Translation in Southern Tanzania,” Science, Narendran, K., I. K. Murthy, H. S. Suresh, H. S. Technology and Human Values 32(3), Dattaraja, N. H. Ravindranath, and R. 88–117. Sukumar (2001). “Nontimber Forest Product Lawrence, A., O. Phillips, A. Reategui, M. Extraction, Utilization and Valuation: A Case Lopez, S. Rose, D. Wood, and A. José (2005). Study from the Nilgiri Biosphere Reserve, “Local Values for Harvested Forest Plants in Southern India,” Economic Botany 55(4), Madre de Dios, Peru: Towards a More 528–538. Contextualised Interpretation of Quantitative Nussbaum, M. (2000). Women and Human Ethnobotanical Data,” Biodiversity and Con- Development: The Capabilities Approach. servation 1(4), 45–79. New York: Cambridge University Press. Lynch, J., G. D. Smith, G. Kaplan, and J. House Pramono, E. (2002). The Traditional use of (2000). “Income Inequality and Mortality: Traditional Knowledge and Medicinal Importance to Health Of Individual Income, Plants in Indonesia. Multi-Stakeholder Dia- Psychosocial Environment, or Material Con- logue on Trade, Intellectual Property and ditions,” British Medical Journal 3(2), 1200– Biological Resources in Asia. Rajendrapur, 1214. Bangladesh: BRAC Centre for Development Mafra, M. S. H., and H. H. C. Stadler Management. (2007). “Etnoconhecimento e Conservaçao da Richerzhagen, C., and K. Holm-Mueller (2004). Biodiversidade em Áreas Naturais e Agricolas “Biodiversity Conservation, Access and no Planalto Sul Catarinense, International Benefit Sharing and Traditional Knowledge,” Society for the Sistems Sciences ISSSBrasi- Ecological Economics 53(4), 445–460. Usp,” 13p. Ronsmans, C., and W. J. Graham (2006). Mangestuti, S. (2007). “Traditional Medicine of “Maternal Mortality: Who, When, Where, Madura Island in Indonesia,” Journal of Tra- And Why,” The Lancet 3(2), 1189–2000. ditional Medicines 2(94), 90–104. Scherr, S. J., A. White, and D. Kaimowitz Marshall, E., and A. C. Newton (2003). “Non- (2004). A New Agenda for Forest Conserva- Timber Forest Products in the Community tion and Poverty Reduction. Making Markets of El Terrero, Sierra de Manantlan Bio- Work for Low Income Producers. Washing- sphere Reserve, Mexico: Is Their Use Sus- ton, DC: Forest Trends and Centre for Inter- tainable?,” Economic Botany 5(2), 262– national Forestry Research (CIFOR). 278. Schippmann, U., and D. Leaman (2003). Impact Marshall, E., K. Schreckenberg, and A. C. of Cultivation and Collection on the Conser- Newton (2006). Commercialization of Non- vation of Medicinal Plants: Global Trends Timber Forest Product: Factors Influencing and Issues. The Third World Congress o Success. Lessons Learned from Mexico and Medicinal and Aromatic Plants, Chiang Bolivia and Policy Implications for Decision- Mai, Thailand, Acta Horticulturae. Leuven, Makers. Cambridge: UNEP World Conserva- Belgium: International Society for Horticul- tion Monitoring Centre. tural Science. Mason, S., P. Tovey, and A. F. Long (2002). Schuklenk, U., and R. Ashcroft (2002). “Afford- “Evaluating Complementary Medicine: Meth- able Access to Essential Medication in Devel- odological Challenges of Randomised Con- oping Countries: Conflicts between Ethical trolled Trials,” British Medical Journal 3(5), and Economic Imperatives,” Journal of 832–834. Medicine and Philosophy 2(7), 179–195.

14354 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Sharma, R., U. Verma, C. L. Sharma, and Thachuk, A. (2007). “Midwifery, Informed B. Kapoor (2005). “Self-Medication among Choice, and Reproductive Autonomy: A Urban Population of Jammu City,” Indian Relational Approach,” Feminism and Psy- Journal of Pharmacology 3(7), 37– chology 17(1), 39–56. 45. Vera-Toscano, E., E. Phimister, and A. Weersink Taguchi, N., M. Kawabata, M. Maekawa, T. (2004). “Panel Estimates of the Canadian Maruo, M. Aditiawarman, and L. Dewata Rural/Urban Women’s Wage Gap,” Ameri- (2003). “Inﬂuence of Socio-Economic Back- can Journal of Agricultural Economics ground and Antenatal care Programmes on 86(4), 1138–1152. Maternal Mortality in Surabaya, Indonesia,” Wagstaff, A. (2001). “Poverty and Health,” CMH Tropical Medicine and International Health Working Paper Series, Paper No. WG1 5, 8(9), 847–852. March, Cambridge and Geneva.

TORRI 355 15 Journal of Small Business Business Management Management 2016 2015 54(1), ••(••), pp. pp. 356–372 ••–•• doi: 10.1111/jsbm.12150 Valuation of Angel-Backed Companies: The Role of Investor Human Capital by Veroniek Collewaert and Sophie Manigart

This article examines how angel investors’ human capital affects the valuation of their portfolio companies, based on the pre-money valuation of 123 investment rounds in 58 Belgian companies. We argue that angel investors with higher levels of human capital will perceive a higher value- creating potential in entrepreneurial opportunities through their ability to see more value-creating options, a higher value-adding potential post-investment, and an enhanced legitimacy provided to the venture. Economic theories suggest they appropriate these rents through lower valuations, whereas stewardship theory suggests they share value creation with entrepreneurs. Consistent with stewardship theory, we show angel investors negotiate higher valuations when they have higher levels of human capital, more speciﬁcally if they studied longer, have a business degree, more entrepreneurial experience, or previous professional law experience. As such, our results contrast with the behavior of venture capital investors who negotiate lower valuations when they have more experience.

Introduction creation (e.g., Collewaert, Manigart, and Aernoudt 2010; Da Rin, Nicodano, and “Clearly, there will be elements with Sembenelli 2006). Research on the processes regards to price. If you don’t feel you are and decision models of venture capital and getting a fair deal, then it is not going to angel investors has expanded signiﬁcantly. be conducive to a partnership.” Insights have been generated on how these investors screen and select their investment Alex MacPherson, CEO of Octopus Ven- targets from a pool of opportunities presented tures, in http://FT.com, 4 April 2009. to them (e.g., Knockaert, Clarysse, and Wright 2010; Maxwell, Jeffrey, and Lévesque 2011; The importance of angel and venture capital Zacharakis and Shepherd 2001), on how con- investors as a ﬁnancing source for entrepre- tracts between investors and entrepreneurs are neurial ventures is well documented (e.g., designed (e.g., Kaplan and Strömberg 2004; Mason and Harrison 1995, 1996; Sohl 2003). Kelly and Hay 2003), and on how investors and Companies backed by risk capital providers entrepreneurs work together once the invest- contribute disproportionately to regional eco- ment has taken place (e.g., Mason and Harrison nomic development, innovation, and job 1996). Interestingly, one central element in the

Veroniek Collewaert is an Associate Professor at the Entrepreneurship, Governance and Strategy Area, Vlerick Business School. Sophie Manigart, is a Full Professor at the Accounting and Corporate Finance Department, Ghent University and at the Accounting and Finance Area, Vlerick Business School. Address correspondence to: Veroniek Collewaert, Vlerick Business School, Entrepreneurship, Governance and Strategy Area, Reep 1, Gent 9000, Belgium. E-mail: [email protected].

356 JOURNALCOLLEWAERT OF SMALL BUSINESS AND MANIGART MANAGEMENT 1 investment process has remained largely unex- Early research on valuations in venture plored in the academic literature, namely the capital investment rounds suggests that valuation of the investment opportunity—or valuations are driven by firm characteristics the “pricing of the deal.” This gap in the litera- (Armstrong, Davila, and Foster 2006; Hand ture is mainly due to the practical difficulties, 2005), market factors (Gompers and Lerner such as the lack of reliable data, associated 2000), and competition (Fairchild 2004; with researching investment agreements and Heughebaert and Manigart 2012). Characteris- valuations. tics of the venture capital investor also explain Despite its neglect in the academic literature, a considerable part of firm valuations firm valuation and deal pricing are of crucial (Cumming and Dai 2011; Hsu 2004), with more importance for both entrepreneurs and inves- experienced investors typically investing at tors. The equity stake that investors get in lower valuations (Hsu 2004). To our knowl- return for their investment in the company is edge, no study to date has examined the rela- equal to the proportion of their investment tionship between angel investor characteristics compared with the company value. Conversely, and the valuations of their investments. Given the equity stake of entrepreneurs is diluted the differences between angel investors and after the investment, as they retain a smaller venture capitalists (Mason and Stark 2004; Van stake of the company compared with the pre- Osnabrugge 2000), insights on venture capital investment situation (Manigart and Meuleman investment behavior cannot always be trans- 2004). For entrepreneurs, valuation is hence ferred to the angel investor context. Because important as it determines the equity stake they angel investors vary in their human capital are able to retain in return for an investment by endowment, the goal of the present paper is to a risk capital investor (Vance 2005). This, in understand how angel investors’ human capital turn, will impact the control they are able to impacts valuations when investing in entrepre- secure and hence their feeling of ownership neurial ventures. over their company. For investors, a lower valu- It is well documented that venture capital ation at investment for a given investment investors with higher levels of human or social amount implies a higher equity stake. This capital create more value in their portfolio com- enhances their return potential, as the fraction panies through better monitoring, stronger of the exit value that accrues to investors will value adding post-investment, and enhanced be higher for higher equity stakes (Manigart legitimacy (Bottazzi and Da Rin 2002; Colombo and Meuleman 2004). A higher equity stake for and Grilli 2010), leading to higher investment investors also increases the control they can performance (Dimov and Shepherd 2005). Hsu gain over the venture (Cumming and Dai 2011; (2004) showed that experienced venture capital Hsu 2007; Mason and Harrison 2002a). investors negotiate lower valuations, enabling Given these conflicting incentives, it is not them to appropriate a large part of the value surprising that investment proposals are often they are expected to create in their portfolio turned down by entrepreneurs who feel the companies. Consistent with this economic view valuation offered by an investor—or, the equity of investors’ behavior, business angels with stake they can retain in their venture—is too high levels of human capital might negotiate low. Alternatively, entrepreneurs without other lower valuations. Alternatively, following a options may accept valuations that they feel are stewardship perspective, business angels might too low but may therefore end up feeling more be more concerned about the post-investment negatively toward their investors. It is thus not relationship with the entrepreneur. From this surprising that agreeing on the price of the deal perspective, business angels might be prepared has been shown to be one of the main hurdles to share the value created through their human for angel and venture capital investments to take capital with the entrepreneurs, hence negotiat- place (De Clercq et al. 2006; Mason and Harrison ing higher valuations. 2002b). As such, gaining a better understanding Hypotheses following these competing theo- of what determines entrepreneurial firm valua- ries are empirically examined using a unique, tions is relevant and valuable because it may hand-collected, and unbiased sample of 123 help participating parties to better understand angel investment rounds in 58 Belgian investee the underlying rationale of this price-setting firms between 1997 and 2008, controlling for process, thereby helping to smooth the risk company and market conditions and taking capital investment process as a whole. potential selection biases into account. Our

2 JOURNALCOLLEWAERT OF SMALL BUSINESS AND MANIGART MANAGEMENT 357 findings strongly support a stewardship view of venture development, and ultimately future the business angel investment process, as busi- value, compared with investors who expect to ness angels with higher levels of human capital be of less assistance. As expected, future negotiate higher valuations at investment. This venture development depends upon investors’ finding contrasts strongly with venture capital human capital; ex ante, venture valuation will investment behavior. As such, we provide also differ. We hence posit that opportunity further insight into how angel investor hetero- valuation is subjective and impacted by angel geneity impacts their investment behavior by investors’ human capital. Specifically, we focusing on a highly important but rarely expect angel investors’ experience, education researched phenomenon, namely the valuation level, and education type to affect expected of their investments. outcome at investment. We elaborate on these relationships hereafter. Theoretical Development We first elaborate on how angel investors’ Experience, Education, and the human capital is positively related with value Perceived Value of an Opportunity creation in their investments. Thereafter, we According to experiential learning theory, expand on how the expected value creation learning is grounded in an individual’s expe- may be shared between business angels and rience and occurs when an individual acquires entrepreneurs in the investment negotiation information and transforms it with prior process. We hereby distinguish between two knowledge in memory (Kolb 1984). The more different views of angel investor behavior, an individual’s stock of knowledge specifically namely an economic view and a stewardship relates to the task at hand, the more efficient view. We expand on both views, which will an individual becomes in accumulating and lead to opposing hypotheses. interpreting new knowledge related to that task and the more refined mental schemata The Impact of Human Capital on become (Dimov and Shepherd 2005; Matlin Expected Firm Value 2005). Experienced individuals thus possess Driven by heterogeneous knowledge more complete and detailed schemata than endowments and experiences, individuals hold inexperienced individuals (Holcomb et al. different perceptions on the characteristics of 2009), integrate domain-specific knowledge in an opportunity and on their potential role in a more meaningful ways, draw on clearer con- venture’s value creation process; hence, they cepts, create richer connections between con- may assess a venture’s outcome differently cepts, and are able to apply domain-specific (Shane 2000). This partially explains why some problem-solving procedures they have devel- individuals act upon an opportunity and start a oped over time (Adelson 1981; Gobbo and business, whereas others do not. In the same Chi 1986; Gruber, Kim, and Brinckmann vein, heterogeneity in angel investor human 2010). capital will affect how angel investors assess Given their more refined and richer sche- the value of an investment opportunity pre- mata, angel investors with more entrepreneur- sented to them. ial and investment experience should be able to First, individuals’ human capital endow- better understand which course of action to ments, developed through education and expe- take to shape the opportunity into a value- rience, lead to the possession of idiosyncratic creating firm and how to adapt to a changing information and knowledge before investing environment and to feedback from stakehold- (Shane 2000). This affects their decision ers or to unexpected events. In all, and condi- whether or not to invest in a venture as well as tional on investing in an opportunity, their their perspective on how value can be created refined insights should make them see more in an opportunity, conditional upon investing value-creating options in a given opportunity (Gruber, Kim, and Brinckmann 2010). Second, compared with inexperienced investors investors with idiosyncratic levels of human (Rotefoss and Kolvereid 2005), having a posi- capital are likely to differ in their expected tive effect on the anticipated value of the value creation potential after the investment opportunity. (Dimov and Shepherd 2005). Investors who In addition to experience affecting investors’ expect to add a lot of value to their venture will ability to see value-creating options, the active have a more positive expectation about future post-investment involvement of experienced

358 JOURNALCOLLEWAERT OF SMALL BUSINESS AND MANIGART MANAGEMENT 3 angel investors in their portfolio companies is Consistent with these arguments, it has been expected to be more valuable than the involve- shown that firms founded by entrepreneurs ment of less experienced investors (Sorensen with higher levels of or more specific education 2007). Like venture capitalists, angel investors have a higher probability of survival and of are considered value-adding investors or “smart achieving higher performance levels (Cooper, money” (Mason 2006; Mason and Harrison Gimeno-Gascon, and Woo 1994), suggesting 1996; Sapienza, Manigart, and Vermeir 1996). that their educational background provides Investors will have different perspectives on them with a broader stock of relevant the value they can create in the opportunity competences and capabilities (Colombo and depending on their experience (Gruber, Kim, Grilli 2005). Higher levels of education and and Brinckmann 2010). Moreover, previous more specific education also benefit venture research suggests that more experienced angel capital investors, leading to more highly suc- investors also have a better reputation and a cessful companies and less failures in their stronger network in this market (Kelly and Hay investment portfolio (Dimov and Shepherd 2000). They may therefore serve as stronger 2005). In the same vein, it is expected that certifiers and legitimizers of the ventures’ value companies backed by highly educated angel to the outside world compared with less expe- investors should have a higher probability of rienced investors (Hsu 2004). This may, for survival and, upon success, a higher perfor- example, enable ventures, backed by experi- mance. This increases the ex ante value of an enced angel investors, to hire better managers opportunity. The question now is whether the and employees, to contract with high-quality angel investor or the entrepreneur will appro- suppliers and customers, or to raise higher priate most of the extra value created through amounts of further financing or financing at the angel’s higher human capital. better conditions (Politis 2008; Sorheim 2005). In all, because experienced angel investors are An Economic View on Value Distribution expected to add more value to their portfolio Empirical research in the venture capital companies through their actions and their cer- industry has shown that more experienced tification, their expected outcome and value of venture capital investors negotiate lower valu- an opportunity will be higher. ations (Hsu 2004). Entrepreneurs are willing to An individual’s human capital is not only accept these lower valuations as they should developed through professional experience but ultimately also benefit from associating with also through formal education, as an individu- experienced investors (Hsu 2004). This also al’s stock of knowledge acquired through edu- implies that venture capital investors appropri- cation positively impacts his or her learning ate the additional value created by their expe- process (Cohen and Levinthal 1990; Shane rience in the negotiation process. This is 2000). Highly educated individuals have a consistent with an economic view on the broader knowledge base and are generally investment process, which assumes that inves- better informed (Colombo and Grilli 2005; tors are rational. Rational individuals are nar- Mudd, Pashev, and Valev 2010; Shane 2000). rowly self-interested: they pursue one This allows them to assimilate new knowledge overarching goal, which is to maximize their more quickly and in a more meaningful way. own utility (Persky 1995). They seek to attain Less educated individuals appear to not go their predetermined goals to the greatest extent through a process of assessing changing con- with the least possible cost (Persky 1995). In an ditions (Mudd, Pashev, and Valev 2010), investment context, rational investors’ utility is thereby decelerating their learning curve. maximized if their wealth is maximized (Cohen Developing a new venture requires a lot of and Kudryavtsev 2012). Venture capital inves- flexibility as environmental conditions continu- tors that negotiate lower valuations hence act ously change. Highly educated investors exhib- as rational investors, appropriating as much iting stronger learning dynamics and value in the negotiation process as possible. In adaptation skills should be able to more flex- an economic view of the angel investment ibly adapt their behavior and the venture’s process, more experienced angels will also act strategy. Highly educated angel investors are rationally, appropriate the value they expect to therefore expected to be able to help develop create, and thus negotiate lower valuations. the opportunity more strongly compared with In addition, angel investors who have expert their less educated colleagues. experience, such as finance or law experience,

4 JOURNALCOLLEWAERT OF SMALL BUSINESS AND MANIGART MANAGEMENT 359 will also have a better understanding of the 1993; Mason and Harrison 2002b). Hence, just process of negotiating and structuring deals like entrepreneurs themselves, angels tend (Dimov and Shepherd 2005; Hsu 2007). In con- to be intrinsically motivated (Politis 2008). trast, negotiating with investors is typically a Third, research has indicated that angels once-in-a-lifetime event for entrepreneurs, who emphasize post-investment involvement rather therefore lack a fine-grained understanding of than control (Mason and Harrison 1996; Van these complex valuation and negotiation Osnabrugge 2000). This is also reflected in issues. Hence, if business angels are rational, angel contracts, which tend to be more they will further exploit their expert advantage entrepreneur-friendly, have weaker control and negotiate lower valuations. This leads to rights, and use less contractual provisions than our first hypothesis. their venture capital counterparts (Goldfarb et al. 2008; Kelly and Hay 2003; Wong 2010). H1a: Angel investors with higher levels of Taken together, it is not surprising that the human capital, including (1) education angel investor–entrepreneur relationship has and (2) experience, will negotiate lower been argued to be the primary means through valuations. which angel investors add value to their portfolio companies (Freear, Sohl, and Wetzel 1995; A Stewardship View on Politis 2008). Angel investors’ willingness to Value Distribution share the rents of their higher value-creating Whereas the economic view depicts angels potential, as represented by higher levels of as self-serving individuals only out to maximize human capital, is hence in line with angel their own personal gain, the stewardship view investors viewing entrepreneurs as partners depicts them as partners to the entrepreneur and the importance they attach to creating and out to serve the collective interest rather than sustaining a cooperative relationship with their serving their own interest (Davis, Schoorman, entrepreneurs (Landström 1992; Mason and and Donaldson 1997). In contrast with the eco- Harrison 1996; Politis 2008). This leads to our nomic view on value appropriation, and based alternative hypothesis. upon stewardship theory, we argue that angel investors may at least partially relinquish their H1b: Angel investors with higher levels of superior (expected) value creation to the entre- human capital, including (1) education preneur, in line with the view that angels tend and (2) experience, will negotiate higher to treat entrepreneurs as true partners. Accord- valuations. ing to this theory, angels will be more inclined to act like stewards when their motives are Data and Methods closely aligned with the entrepreneurs’, when Methodology they are intrinsically rather than extrinsically An ideal empirical strategy to test the motivated, and when their focus is on involve- hypotheses would be to compare the valuations ment rather than control (Davis, Schoorman, offered by different business angels (with dif- and Donaldson 1997). All of these closely align ferent human capital endowments) for the with insights from the angel literature on same entrepreneurial company (Hsu 2004). angels in general. Two reasons explain why this method was not First, several studies have revealed that used in the present research. First, few entre- angels tend to have similar motives as entrepreneurial companies receive offers from mul- preneurs (Duxbury, Haines, and Riding 1996; tiple angel investors (Maxwell, Jeffrey, and Politis 2008). Most angels see themselves as Lévesque 2011)—they are mostly very happy partners to the entrepreneur rather than as when receiving a single offer. Second, the angel pure financiers, with some even referring to investor scene is very secretive in nature, even them as “co-entrepreneurs” (Politis 2008, p. more so than the venture capital industry, and 137). Second, angel investors are known to be it is therefore extremely challenging to get more altruistic investors than venture capital- these data, should they exist. We therefore ists, pursuing other goals such as personal sat- follow the approach of Armstrong, Davila, and isfaction, opportunities to influence the Foster (2006) and Hand (2003) and focus on development of a new venture, and job cre- the valuation of actual deals observed in the ation in addition to economic value maximiza- market. Whereas Hsu (2004) mitigates the tion (e.g., Harrison and Mason 1992; Landström potential issue of unobserved heterogeneity

360 JOURNALCOLLEWAERT OF SMALL BUSINESS AND MANIGART MANAGEMENT 5 affecting firm valuations by examining valua- capital gains realized by private individuals are tion differences across multiple offers and by not taxed) and numbers of active business keeping firm effects constant (allowing him to angel networks and deals made are similar leave out firm-level control variables), we do so (Collewaert, Manigart, and Aernoudt 2010). by including an extensive list of firm-level To further reduce sample selection bias and control variables, which prior research has obtain the most representative sample possible, shown to matter for firm valuations. Moreover, 20 different data sources were used to identify robustness checks are specifically targeted angel-backed companies, including a random toward dealing with the issue of potential directory of start-ups, deal lists of angel net- unobserved heterogeneity. We hereafter works, Global Entrepreneurship Monitor data explain our sample selection strategy and the (from the Belgian chapter), directories of high- variables used in the multivariate analyses. technology companies, media articles, incubators, and snowballing. After having contacted Data all companies by phone to ensure they had Hypotheses are tested on a data set of indeed received financing from angel investors, Belgian angel-backed companies. The Belgian this resulted in a sample of 102 angel-backed research setting provides us with the advantage companies. Data for this study were gathered of all firms (even unquoted ones) having a legal through four instruments. More specifically, (1) obligation to publish information on all capital valuation and investment information was increases. This information, validated and retrieved from the Belgian Law Gazette; (2) deposited by an official notary with the human capital variables from face-to-face inter- National Bank of Belgium and published in the views conducted with (or, alternatively, ques- Belgian Law Gazette, allows accurately calcu- tionnaires sent to) the angel investors of these lating implied valuations. Given the obligatory companies, (3) information on patent applica- character of this information, the reliability of tions from the European Patent Office, and (4) our data is excellent. The unique institutional the remaining variables from the companies’ setting allows access to information that is financial accounts, which all Belgian compa- typically only available in commercial data- nies are obliged to file with the National Bank bases when companies voluntarily disclose this of Belgium. The latter were retrieved from the information. However, relying on voluntarily Bel-First database by Bureau Van Dijk. Data released information entails problems of reli- unavailability reduced the final sample size to ability and of sample selection bias as firms 123 investment rounds in 58 companies, repre- self-select to voluntarily disclose information senting 46 angel investors. on valuation in investment rounds. Few companies that raise equity from angel investors do Dependent Variable so, however, as this is typically considered to The analyses focus on the pre-money valu- be private and sensitive information. Our data ation of angel-backed companies. Following set, on the other hand, suffers neither from Hsu (2004), pre-money valuation is defined as self-selection problems nor from unreliable the number of shares outstanding prior to the information, enhancing the validity of the initial angel investment multiplied with the results. offered per-unit share price. As such, any External validity is also warranted as changes in the value of the venture introduced Belgium has traditionally been rather similar to by the angel investment itself are excluded. All other European countries (e.g., Germany, the numbers are inflation-adjusted. The mean pre- Netherlands, France, and Austria) in terms of money valuation of the firms in the sample is socioeconomic indicators such as entrepreneur- 1.59 million EUR, ranging from a minimum ship rate, unemployment rate, private con- valuation of 19,318 EUR up to a maximum sumption expenditures, and social security fees valuation of 12 million EUR in a venture active (Eurostat 2011; Stroobandt, Lievens, and Waege in the communications industry (see Table 1). 2005). With regard to the Belgian angel market This shows that the angels in our sample invest in specific, the scarce information available in companies with a wide variation in pre- suggests that differences with regard to conti- investment value. Table 1 further shows that nental European countries should be limited; the median pre-money valuation is highest for angel investments are taxed in a comparable early-stage communications investments made manner (with the main difference being that in 2007.

6 JOURNALCOLLEWAERT OF SMALL BUSINESS AND MANIGART MANAGEMENT 361 Table 1 and Mason 2010; Wiltbank 2005). Hence, the Pre-Money Valuations (N = 123)a number of years of entrepreneurial experience as a founder and/or entrepreneur was measured (mean value of 11 years). Second, we Variables n Pre-Money asked for the number of companies founded by Valuation the investor before his or her investment in the (000 EUR) sample company. The majority (62.6 percent) of angel investors in our sample had founded Mean Median more than one company with the mean equal- ling 10 companies founded. We also included All N = 123 1,589.00 692.87 investment experience, measured by the Investment Year number of angel investments made prior to the 1997 1 31.71 31.71 first investment in the sample company. On 1998 2 183.79 183.79 average, angel investors had made one invest- 1999 5 174.21 22.92 ment prior to the current one. Finally, we mea- 2000 9 2,244.88 1,352.75 sured whether angel investors possessed 2001 8 492.39 260.43 professional experience especially relevant to 2002 13 935.31 453.67 the angel investment process, which we labeled 2003 19 1,025.87 562.01 expert experience. Expert experience was mea- 2004 17 1,322.48 568.71 sured as the number of years investors had 2005 14 1,464.40 720.16 professionally worked in a finance (years of 2006 17 1,764.39 1,605.07 finance experience) or in a legal department 2007 12 4,642.15 1,735.39 (years of law experience). On average, investors 2008 6 1,616.37 1,174.30 had 3.5 years of finance experience and no Investment Stage legal experience. Taken together, the angel Seed 43 876.00 348.26 investors’ investment behavior and characteris- Early Stage 53 2,232.74 1,212.72 tics are consistent with those of angel investors Later Stage 27 1,460.88 815.72 in other countries (Mason 2006), supporting Industry the external validity of this study. IT 46 1,133.60 565.36 Communications 17 3,989.09 2,020.21 Control Variables Life Sciences 2 645.88 645.88 Previous research has shown valuations of Non-Tech 58 1,279.22 548.14 venture capital-backed companies to be significantly affected by company characteristics (Armstrong, Davila, and Foster 2006; Cumming a IT, information technology. and Dai 2011; Hand 2005; Heughebaert and Manigart 2012; Hsu 2004, 2007). Building on these studies, a first set of controls were added, Independent Variables including company age at time of investment Seven variables were included in the model (measured in years), patents, investment round to measure the angel investors’ human capital number, and total capital increase in the (Colombo and Grilli 2005). investment round. On average, a company was Education was measured as the number of 4 years old at the time of investment with some years of postsecondary education, with angel companies raising angel money at start-up and investors having an average of 5.5 years of the oldest one being 26 years old. As the postsecondary education (as shown in Table 2). number of patent applications was generally A dummy variable measured whether angel low, a dummy variable was included taking on investors had received a business degree (1 if he the value of 1 if the company had applied for or she had, 0 if he or she had not), with 45 patents prior to investment and 0 if it had not. percent of all angels having received one. Only 16 percent of all investment rounds in our Entrepreneurial experience of the angel inves- sample occurred in companies that had applied tor was measured with two variables. One type for at least one patent. The majority of the of experience specifically relevant for evaluat- angel investments in the sample are first-round ing new venture opportunities is experience in investments (54 rounds), with a maximum of 4 new venture formation (e.g., Smith, Harrison, investment rounds. The total capital increase

362 JOURNALCOLLEWAERT OF SMALL BUSINESS AND MANIGART MANAGEMENT 7 8

Table 2 Descriptive Statistics and Correlations (N = 123)a

Variables Mean S.D. 1 23456789101112131415

ORA FSALBSNS MANAGEMENT BUSINESS SMALL OF JOURNAL Control Variables 1. Company Age 4.31 5.02 — 2. Patent Applications 0.16 0.37 −0.13 — 3. Investment Round Number 1.72 0.73 0.12 0.05 — OLWETADMNGR 363 MANIGART AND COLLEWAERT 4. Total Capital Increase 460.85 557.77 −0.17 0.02 −0.12 — 5. Intangible Fixed Assets 178.99 278.48 −0.04 0.27 0.16 0.29 — 6. Cash Assets 117.83 293.98 −0.15 0.03 −0.12 0.65 0.09 — 7. Non-Cash Assets 450.87 600.99 0.03 0.07 −0.03 0.41 0.48 0.29 — 8. Proﬁt/Loss −431.52 633.31 0.02 −0.19 −0.28 −0.42 −0.50 −0.25 −0.33 — Angel Investor Human Capital 9. Postsecondary Education 5.50 2.11 −0.09 0.02 −0.06 0.11 0.14 0.01 0.14 −0.10 — Years 10. Business Degree 0.45 0.50 −0.13 0.22 0.14 0.06 −0.02 −0.08 −0.02 0.12 −0.33 — 11. Years of Entrepreneurial 11.45 7.94 0.00 −0.02 0.08 0.02 −0.03 0.03 0.15 −0.05 −0.04 −0.15 — Experience 12. Number of Companies 9.59 23.11 0.05 −0.08 −0.06 0.09 −0.16 0.08 0.03 0.01 −0.17 0.28 0.46 — Founded 13. Number of Prior Angel 0.85 1.19 −0.01 0.15 −0.11 0.19 0.19 0.07 0.06 −0.45 0.08 −0.08 0.25 0.35 — Investments Made 14. Years of Finance Experience 3.59 6.94 −0.15 0.23 0.14 −0.13 −0.01 −0.12 −0.15 0.20 −0.20 0.52 −0.18 −0.05 −0.16 — 15. Years of Law Experience 0.05 0.46 −0.08 −0.05 −0.08 −0.07 −0.02 0.03 −0.06 0.02 0.05 −0.10 −0.13 −0.04 −0.06 −0.05 — Dependent Variable 16. Pre-Money Valuation 1,589.00 2,326.10 0.02 0.11 0.21 0.41 0.43 0.09 0.46 −0.57 0.22 0.09 0.09 −0.06 0.17 −0.09 −0.06

aS.D., standard deviation. This table provides a Pearson correlation matrix of the variables used in the empirical analysis. Variables 2 and 10 are dummy variables; these correlations hence need to be interpreted with caution. Total capital increase, intangible fixed assets, cash assets, non-cash assets, profit/loss, and pre-money valuation are in euros. Bold correlations significant at p ≤ .05. represents the total amount invested in the 2003; Hsu 2007), with cluster-robust standard investment round. Total capital increase was, errors (clustered at the company level) on average, 460,854 EUR, with a minimum of (Cumming and Dai 2011). To this end, the 7,640 EUR and a maximum of 3.6 million EUR.1 natural logarithm was taken of all continuous A second set of control variables pertains to variables, including pre-money valuation. This information from the company’s financial transformation was deemed appropriate accounts in the year prior to investment or the because of the skewed distribution of the valu- year of investment in case of start-ups ation numbers and the aptness of this tech- (Armstrong, Davila, and Foster 2006; Hand nique for dealing with nonlinearities in the 2005). We included the value of intangible relationship between the dependent and inde- fixed assets, cash, and non-cash assets as well pendent variables (Armstrong, Davila, and as the profit or loss of the company. On Foster 2006; Hand 2003). Furthermore, it also average, intangible fixed assets at the time of lessens the impact of outliers. Cluster-robust investment amounted to 178,995 EUR, whereas standard errors were used to control for mul- cash assets were on average 117,831 EUR at the tiple investment rounds in the same company time of investment and non-cash assets were on (Wooldridge 2002). Table 3 includes the results average 450,868 EUR. Finally, the mean value of the hypotheses tests. of the control variable representing profit (loss) Model 1 includes the control variables only; was negative (−431,516 EUR), indicating that the model is significant and explains 58 percent on average investee firms were not making a of the variation in angel-backed companies’ profit at time of investment. The maximum loss valuation in this sample. The results are consis- was 3.8 million EUR, whereas the maximum tent with previous valuation research (such as profit was 59,186 EUR. Cumming and Dai 2011; Hsu 2004, 2007) in that company age, having applied for patents prior Results to investment, and total capital increase at Correlation Analysis investment are statistically significant and posi- Table 2 shows the correlation matrix of all tive indicators of angel-backed companies’ valu- variables included in the full models. Correla- ation. Older companies are worth more, as they tion coefficients are generally low and variance have had more time to mature and develop. A inflation factors for all models are well below positive association with patent applications is the value of five (maximum is 2.36), suggesting consistent with the notion that patents are con- no multicollinearity problems (Gujarati 2003). sidered a signal of venture quality and should The strongest correlations appear between the hence be positively reflected in the venture’s financial statement variables on the one hand valuation. This effect, however, disappears in and total capital increase as well as pre-money the full model. The total capital increase, as an valuation on the other. Consistent with previ- indicator of the amount invested in a portfolio ous studies, pre-money valuation is positively company, is positively related to valuation. This correlated with investment round number, total is consistent with higher amounts being capital increase, amount of intangible fixed invested in companies with higher growth assets, non-cash assets, and losses. All these opportunities, which in turn lead to higher correlations are statistically significant. Only valuations. Consistent with Hand (2005) and one angel investor human capital variable is Armstrong, Davila, and Foster (2006), a venture positively correlated with pre-money valuation, with higher levels of cash or non-cash assets is namely the number of postsecondary education valued higher, although the effect of non-cash years (p < .05). assets is only marginally significant. The hypothesized effects of angel investors’ Hypotheses Testing human capital are tested in model 2. Adding Hypotheses are tested using log-linear the angel investor human capital variables to the ordinary least squares (OLS) regression (Hand model is a significant improvement of the

1Note that company age measured at the time of investment serves as a proxy for investment stage. Additional analyses were run with industry controls included in the model (results available from the authors upon request). Results remain qualitatively the same and industry controls are not statistically signiﬁcant. To ensure a parsimonious model, these controls were therefore dropped from the subsequent analyses.

364 JOURNALCOLLEWAERT OF SMALL BUSINESS AND MANIGART MANAGEMENT 9 Table 3 Regression Analysis: The Effect of Angel Investor Human Capital on Valuation

Dependent Variable: L Pre-Money Valuation Model 1 Model 2 Model 3

Control Variables Non-Financial Statement Controls L Portfolio Company Age 0.32* 0.35* 0.41** Patent Applications 0.50† 0.33 0.24 Investment Round Number 0.27 0.22 0.16 L Total Capital Increase 0.64*** 0.53*** 0.35*** Financial Statement Controls L Intangible Fixed Assets 0.04 0.05 0.09*** L Cash Assets 0.10*** 0.10*** 0.10*** L Non-Cash Assets 0.15† 0.15† 0.11 L Proﬁt/Loss −0.03 −0.03 −0.02 Angel Investor Human Capital Variables Education L Postsecondary Education Years 0.31† 0.38* Business Degree 0.67** 0.50** Entrepreneurial Experience L Years of Entrepreneurial Experience 0.28* 0.26* L Number of Companies Founded −0.24* −0.18* Investment Experience L Number of Prior Angel Investments 0.31† 0.18† Expert Experience L Years of Finance Experience −0.03 −0.08 L Years of Law Experience 0.46* 0.22 F-value 22.01*** 22.66*** R2 0.58 0.63 Wald Test (F-value) 2.14* LR Test Chi2 346.17 Prob > Chi2 0.000

The regression models consist of 123 rounds of angel investments between 1997 and 2008 for which we were able to gather valuation, financial account, and investor human capital data. The dependent variable is the natural logarithm of pre-money valuation. All independent variables of which the natural logarithm was taken are denoted with L as a prefix. For models 1 and 2, the p-values are based on standard errors clustered at the company level (one-tailed tests for hypothesized effects). To control for unobserved company-level drivers of valuation, we use hierarchical linear modeling for model 3 (Raudenbusch and Bryk 2002). The p-values are based on robust standard errors. †p ≤ .10. *p ≤ .05. **p ≤ .01. ***p ≤ .001. control model as indicated by the Wald test priation), whereas they provide strong support (p < .05), suggesting that investor characteristics for H1b (the stewardship view of value distribu- partially drive valuations. The results broadly tion). There is mostly a positive and significant reject H1a (the economic view of value appro- association between angel investors’ human

10 JOURNALCOLLEWAERT OF SMALL BUSINESS AND MANIGART MANAGEMENT 365 capital, acquired through education or previous Valuations of unquoted ventures may also be experience (except the number of companies affected by valuations in the stock markets and founded), and ventures’ pre-money valuation. changes therein (Gompers and Lerner 2000; More specifically, and supporting the stew- Hand 2005). Therefore, we reran the earlier ardship hypothesis, the number of years of analyses including the relative (from time of entrepreneurial experience is positively associ- investment to three months earlier) Belgian ated with valuation (p < .05). In contrast, the Morgan Stanley Capital International (MSCI) number of companies founded is negatively index. On average, the relative return was 2 associated with pre-money valuation (p < .05): percent with a wide variation from a minimum angel investors who have founded numerous of −37 percent to a maximum of 23 percent. As companies act more like more rational venture an alternative, models were also rerun, adding a capital investors and appropriate value. Consis- dummy variable taking the value of 1 if the tent with the stewardship view and supporting investment took place between 1999 and 2001 H1b, angel investors with more years of because one could argue that the MSCI index postsecondary education and with a business inadequately captures the specific effect of the degree value their companies significantly bubble period during which valuations were higher (the former at p < .10, the latter at skyrocketing, as also illustrated in our sample p < .01). Prior investment experience has a mar- (see Table 1). In both cases, added variables did ginally significant effect in the hypothesized not have a statistically significant effect and positive direction, providing some support for previous results remained robust. H1b. Number of years of finance experience has no statistically significant effect. Finally, Conclusions and Discussion angel investors with more expert experience in Research Findings law value their investments higher (p < .05), Despite valuation playing a crucial role for consistent with H1b. both entrepreneurs and investors, our understanding of the relationship between investor Robustness Checks characteristics and the valuation of the ventures Additional analyses were performed to check they invest in remains limited (Cumming and the robustness of the results. First, in order to Dai 2011; Hsu 2004). The goal of this article control for unobserved company-level drivers of was to study the effect of angel investors’ valuation, we follow the approach adopted by human capital on their portfolio companies’ Cumming and Dai (2011) by estimating a hier- valuations. We started theorizing that angel archical linear model using maximum likelihood investors with high levels of human capital, (Raudenbusch and Bryk 2002). Allowing for a more specifically with high levels of experience company-level random effect should at least or education, would perceive higher value- partially remove the effect of any remaining creating potential in entrepreneurial opportuni- unobserved company-level factors. Table 3 ties conditional on investing in them, driven by (model 3) shows that the results are consistent a higher perceived value in an opportunity, a with those of the main analysis (model 2) in that higher value-adding potential post-investment, most relationships are in the same direction and a stronger legitimacy provided to the and have similar significance. The likelihood venture. The hypotheses were tested on a ratio-test indicates the statistical significance of highly reliable hand-collected sample of the company-level random effect (residual Belgian angel investor deals. The findings company-level variance), thereby supporting suggest that company valuations are signifi- the relevance of this robustness check. There is cantly and positively associated with angel a statistically significant and positive relation- investors’ human capital. As such, this article ship between years of entrepreneurial experi- contributes to the entrepreneurship literature in ence, years of postsecondary education, and multiple ways. having a business degree on the one hand and First, it contributes to this literature by focus- pre-money valuation on the other hand, though ing on angel investors who despite their impor- the latter is still negatively affected by the tance to entrepreneurial ventures are number of companies founded. The number of still largely neglected by entrepreneurship years of law experience, however, does not researchers. This is important, as our study remain significant. The latter result should stresses that angel investors behave differently hence be interpreted with care. compared with venture capital investors, who

366 JOURNALCOLLEWAERT OF SMALL BUSINESS AND MANIGART MANAGEMENT 11 are more extensively researched. Second, it results hence stress that differentiating between also contributes to the entrepreneurship litera- different types of expert experience, such as ture by studying the effect of investor human finance and law experience, is important in capital on company valuation, as valuation generating a finer grained understanding of the studies so far have generally exclusively impact of experience (Dimov and Shepherd focused on the effect of company characteris- 2005; Gruber, Kim, and Brinckmann 2010). tics (with Cumming and Dai 2011 and Hsu 2004 being notable exceptions). More specifically, Discussion the positive effect on venture valuation of angel Our findings are far from trivial as most are investors’ years of entrepreneurial and law in contrast with findings from the venture experience and education, as well as the posi- capital industry. Hsu (2004) has shown that tive effect of having a business degree, support reputed venture capital investors negotiate the argument that angel investors largely follow lower valuations. Entrepreneurs accept these a stewardship logic in their relationship with lower valuations from reputed venture capital the entrepreneur in that they do not appropri- investors, as portfolio companies can benefit ate the excess value resulting from their more from the stronger certification of their increased value-creating options and post- venture by highly reputed investors and the investment value-adding potential. higher value-adding potential of experienced The strong and positive effect of education is investors (Hsu 2004; Kelly and Hay 2000; interesting, as previous entrepreneurship Sorensen 2007). Both reputation and value- research has hinted that education might be adding services provide venture capital inves- less important than experience in predicting tors with leverage when negotiating with entrepreneurial success (Politis 2005). Research entrepreneurs, and entrepreneurs are willing on the performance of venture capital teams, to pay for affiliation with higher quality on the other hand, has stressed the positive venture capital investors. Venture capital relationship between the education of venture investors hence appropriate the extra value capital managers and the probability of having they create through their experience and highly successful portfolio companies (Dimov social capital. and Shepherd 2005). Finding that education In contrast, our findings suggest that high- has a significant impact on the angel invest- quality angel investors do not fully appropriate ment process is hence broadly consistent with this extra value and hence do not maximize the results from the venture capital setting, economic value of their investments. In con- although its effect is opposite. trast to venture capital investors, angel inves- In contrast with other human capital vari- tors are willing to—at least partially—share this ables, the effect of the number of companies value with the entrepreneurs they fund. This founded by an angel investor on company valu- fits with the view of angel investors not solely ation is negative. This suggests that angels who investing for financial reasons, but also for have founded numerous companies act more more personal reasons. Important goals for as professional venture capital investors and angel investors are, among others, personal take a more portfolio-oriented approach, hence satisfaction, opportunities to influence the appropriating the expected value creation development of a new venture, and job cre- during the negotiation process rather than ation (Harrison and Mason 1992; Landström sharing it with the entrepreneur. 1993). This more altruistic side of the relation- The lack of a significant effect of expert ship between angel investors and entrepre- finance experience, whereas expert law expe- neurs is also illustrated by the fact that rience is significantly positively related with compared with venture capital contracts, angel valuations, is interesting. The irrelevance of contracts are more entrepreneur-friendly, have finance experience is in line with Smith, weaker control rights, use less contractual pro- Harrison, and Mason (2010), who suggest that a visions, and are used more from a transactional banking background does not provide a rel- than from a control point of view (Ibrahim evant toolkit to become an angel investor and 2008; Kelly and Hay 2003). Our findings hence that “angels may have to unlearn some of their provide another illustration of the fact that previous [banking] experience before they can angel investor behavior may differ significantly start to learn how to be an effective investor” from venture capital investor behavior (Bruton (Smith, Harrison, and Mason 2010, p. 9). Our et al. 2010).

12 JOURNALCOLLEWAERT OF SMALL BUSINESS AND MANIGART MANAGEMENT 367 In addition to contributing to the academic hold in Continental Europe. It would be inter- literature, our research also has important prac- esting to expand this research to countries with tical implications for entrepreneurs and angel other institutional settings, however. investors. Entrepreneurs should carefully con- Second, hypotheses were tested based on a sider whose money they accept when searching rather small sample. Relative to the number of for finance, as not all money is the same. predictors used in the model, this is however Whereas previous research has shown that similar to other valuation studies of risk experienced and highly educated investors capital-backed companies (such as Hsu 2007). have a positive impact on the development of Regardless, this does preclude the possibility their portfolio companies, our findings show of testing a more complete model. Alterna- that angel investors with high levels of human tively, the research design could be changed capital will also negotiate higher valuations. by comparing the valuations offered by differ- This will enable entrepreneurs to capture a ent angel investors to the same entrepreneur more important part of the value created in (Hsu 2004), which would address issues of their venture. Entrepreneurs should, however, unobserved firm heterogeneity. Understanding be cautious, as better valuations might come the negotiation dynamics between entrepre- hand in hand with stricter contract terms. From neurs and investors, and also between angel a broader perspective, given the fact that agree- investors investing together in an investment ing on the price of the deal has been one of the syndicate as is often done when deals are pre- main hurdles for angel investments to be con- sented to angel networks, would further summated (Mason and Harrison 2002b), our improve our understanding of the relationship findings are important as they may help both between investor characteristics and deal entrepreneurs and investors to get more insight terms, including valuation. into what drives valuation. Gaining a better Finally, though we focused on the impact of understanding of valuation’s underlying drivers human capital, including experience and edu- may help to avoid conflicts between angel cation, it is well known that other sources of investors and entrepreneurs about this point, learning may be important. In the present thereby potentially facilitating negotiations and study, we ignored, for example, vicarious learn- the angel investment process as a whole. ing, that is, learning by observing others’ behavior and its consequences. Smith, Limitations and Avenues for Harrison, and Mason (2010) have shown that Future Research vicarious learning is important as angel inves- This study is not without its limitations. First, tors may co-invest and learn from other angels. all data are collected from Belgian angel- More specifically, vicarious learning impacts backed companies, which may limit the the initial screening process (Smith, Harrison, generalizability of the findings. This might be and Mason 2010). Understanding how vicarious particularly true for the United Kingdom and learning impacts valuations is left for future the United States as the angel financing market studies. in those countries is more developed than it is Based on the results and limitations of the in Continental Europe (EBAN [European study, there are several avenues for future Business Angel Network] 2005). Moreover, dif- research. First, it would be interesting to under- ferences in institutional settings, stronger cor- stand which underlying mechanisms explain porate governance regimes, or stronger the effect of angel investor human capital on minority shareholder rights may lead to differ- company valuation. More specifically, we ences in behavior of angel investors (Bruton offered several potential explanations such as et al. 2010). However, the Belgian angel setting value adding and negotiation skills. To test is quite similar to that in other Continental which of these mechanisms actually occur or European countries. Further, Belgian socioeco- which is strongest, one would need finer nomic indicators as income distribution, grained data on the investment decision and employment rate, social security fees, and trade negotiation process, or on the post-investment balance are also similar to indicators in other performance of these companies. Second, con- Continental European countries such as sistent with previous studies on drivers of valu- Germany, the Netherlands, France, Austria, ations in venture capital deals (e.g., Cumming Spain, and Italy (Stroobandt, Lievens, and and Dai 2011; Hand 2005; Heughebaert and Waege 2005). Hence, we expect the results to Manigart 2012), the data we observed were

368 JOURNALCOLLEWAERT OF SMALL BUSINESS AND MANIGART MANAGEMENT 13 limited to valuations. There might be a trade- Boeker 2007), this study is the first to look into off, however, between higher valuations and valuations of angel-backed companies and, as stricter contract terms. This would require such, contributes to the entrepreneurship litera- further research to investigate the relationship ture. We hope that our study will stimulate between valuations and contract terms. Inves- future research in this area. tigating the link between both may hence make an important contribution to academic Acknowledgments research. Differentiating hereby between angel We thank the Policy Research Center on investors and venture capital investors seems Entrepreneurship and International Business highly relevant. (STOIO) and the Hercules Fund of Ghent Uni- Considering the different effects that were versity for financial help and Pieterjan found for angel investors’ human capital com- Behaeghe for excellent research assistance. pared with what venture capital studies have This paper benefited from presentations at the found so far, it would also be interesting to 2011 Academy of Management Meeting (San understand what happens when angel inves- Antonio, Texas) and the 2009 RENT Conference tors and venture capitalists co-invest and hence (Budapest). participate in valuation negotiations. As venture capital investors typically bring more References money to the table, it may be expected that Adelson, B. (1981). “Problem Solving and the they will steer negotiations and valuations. Development of Abstract Categories in Pro- Angel investors, however, might influence the gramming Languages,” Memory and Cogni- negotiations to some extent, bringing in their tion 9, 422–433. different view to the investment process Armstrong, C., A. Davila, and G. Foster (2006). (Bruton et al. 2010). “Venture-Backed Private Equity Valuation Finally, as scholars have suggested similarity and Financial Statement Information,” between investors and entrepreneurs to play a Review of Accounting Studies 11, 119–154. significant role in the investment process (e.g., Bottazzi, L., and M. Da Rin (2002). “Venture Bruns et al. 2008; Franke et al. 2006), another Capital in Europe and the Financing of Inno- interesting avenue for future research would be vative Companies,” Economic Policy 17, to look into the effects of fit between investors’ 231–269. and entrepreneurs’ human capital. In other Bruns, V., D. Holland, D. Shepherd, and J. words, would investors value opportunities dif- Wiklund (2008). “The Role of Human Capital ferently when the entrepreneurs’ human capital in Loan Officers’ Decision Policies,” Entre- fits better with their human capital? Do they preneurship Theory and Practice 32, 485– seek out entrepreneurs with similar human 506. capital or with complementary human capital? Bruton, G. D., I. Filatotchev, S. Chahine, and M. It might be that not only investor characteristics Wright (2010). “Governance, Ownership matter for the valuation of the deal but also the Structure and Performance of IPO Firms: concordance between investor and entrepre- The Impact of Different Types of Private neur characteristics. Equity Investors and Institutional Environ- ments,” Strategic Management Journal 31, Conclusion 491–509. To conclude, this study shows that investor Cohen, G., and A. Kudryavtsev (2012). “Inves- heterogeneity matters and provides evidence tor Rationality and Financial Decisions,” for the widely accepted notion that angel inves- Journal of Behavioral Finance 13(1), 11–16. tors are indeed different from venture capital Cohen, W. M., and D. A. Levinthal (1990). investors. For the latter, more human capital is “Absorptive Capacity: A New Perspective on seen as an economic good, which can be traded Learning and Innovation,” Administrative against a higher price. For angel investors on Science Quarterly 35, 128–152. the other hand, more human capital allows Collewaert, V., S. Manigart, and A. Aernoudt them to meet entrepreneurs halfway and relin- (2010). “Assessment of Government Funding quish some of the venture’s value to the entre- of Business Angel Networks in Flanders,” preneurs they back. Whereas previous studies Regional Studies 44, 119–130. have tried estimating returns to angel investors Colombo, M., and L. Grilli (2005). “Founders’ (e.g., Mason and Harrison 2002a; Wiltbank and Human Capital and the Growth of New-

14 JOURNALCOLLEWAERT OF SMALL BUSINESS AND MANIGART MANAGEMENT 369 Technology Based Firms: A Competence- Freear, J., J. E. Sohl, and W. E. Wetzel (1995). Based View,” Research Policy 34, 795– “Angels: Personal Investors in the Venture 816. Capital Market,” Entrepreneurship and ——— (2010). “On Growth Drivers of High- Regional Development 7, 85–94. Tech Start-Ups: Exploring the Role of Found- Gobbo, C., and M. Chi (1986). “How Knowl- ers’ Human Capital and Venture Capital,” edge Is Structured and Used by Expert and Journal of Business Venturing 25, 610– Novice Children,” Cognitive Development 1, 626. 221–237. Cooper, A. C., F. J. Gimeno-Gascon, and C. Y. Goldfarb, B., G. Hoberg, D. Kirsch, and A. Woo (1994). “Initial Human and Financial Triantis. (2008). “Does Angel Participation Capital as Predictors of New Venture Perfor- Matter? An Analysis of Early Venture Financ- mance,” Journal of Business Venturing 9, ing,” Working Paper 06–072, Robert H. 371–395. Smith School of Business, University of Cumming, D. C., and N. Dai (2011). “Fund Size, Maryland, USA. Limited Attention and Valuation of Venture Gompers, P. A., and J. Lerner (2000). “Money Capital Backed Firms,” Journal of Empirical Chasing Deals? The Impact of Fund Inflows Finance 18, 2–15. on Private Equity Valuation,” Journal of Da Rin, M., G. Nicodano, and A. Sembenelli Financial Economics 55, 281–325. (2006). “Public Policy and the Creation of Gruber, M., S. M. Kim, and J. Brinckmann Active Venture Capital Markets,” Journal of (2010). “How Experience Shapes The Sub- Public Economics 90, 1699–1723. jective Evaluation of Business Opportuni- Davis, J., D. Schoorman, and L. Donaldson ties,” Frontiers of Entrepreneurship Research (1997). “Toward a Stewardship Theory of 30. Article 3. Management,” Academy of Management Gujarati, D. (2003). Basic Econometrics. New Review 22, 20–47. York: McGraw-Hill. De Clercq, D., V. Fried, O. Lehtonen, and H. Hand, J. (2003). “Profits, Losses and the Non- Sapienza (2006). “An Entrepreneur’s Guide Linear Pricing of Internet Stocks,” in Intan- to The Venture Capital Galaxy,” Academy of gible Assets: Values, Measures and Risks. Management Perspectives August, 90–112. Eds. J. R. Hand and B. Lev. New York: Dimov, D., and D. Shepherd (2005). “Human Oxford University Press, 248–268. Capital Theory and Venture Capital Firms: ——— (2005). “The Value Relevance of Finan- Exploring ‘Home Runs’ and ‘Strike Outs’,” cial Statements in the Venture Capital Journal of Business Venturing 20, 1–21. Market,” The Accounting Review 80, 613–648. Duxbury, L., G. Haines, and A. Riding (1996). Harrison, R., and C. Mason (1992). “Interna- “A Personality Profile of Canadian Informal tional Perspectives on the Supply of Infor- Investors,” Journal of Small Business Man- mal Venture Capital,” Journal of Business agement 24, 44–55. Venturing 7, 459–475. EBAN (European Business Angel Network) Heughebaert, A., and S. Manigart (2012). “Firm (2005). European Directory of Business Valuation in Venture Capital Financing Angel Networks in Europe. Brussels: EBAN. Rounds: The Role of Bargaining Power,” Eurostat. (2011). “Key Figures on European Journal of Business Finance and Accounting Business,” http://epp.eurostat.ec.europa.eu/ 39(3–4), 500–530. cache/ITY_OFFPUB/KS-ET-11-001/EN/KS Holcomb, T. R., R. D. Ireland, R. M. Holmes, -ET-11-001-EN.PDF. and M. A. Hitt (2009). “Architecture of Entre- Fairchild, R. (2004). “Financial Contracting preneurial Learning: Exploring the Link Between Managers and Venture Capitalists: among Heuristics, Knowledge, and Action,” The Role of Value-Added Services, Reputa- Entrepreneurship Theory and Practice 33, tion Seeking, and Bargaining Power,” 167–192. Journal of Financial Research 27, 481– Hsu, D. H. (2004). “What Do Entrepreneurs Pay 495. for Venture Capital Affiliation?,” Journal of Franke, N., M. Gruber, D. Harhoff, and J. Finance 59, 1805–1844. Henkel (2006). “What You Are Is What You ——— (2007). “Experienced Entrepreneurial Like—Similarity Biases in Venture Capital- Founders, Organizational Capital, and ists’ Evaluations of Start-Up Teams,” Journal Venture Capital Funding,” Research Policy of Business Venturing 21, 802–826. 36, 722–741.

370 JOURNALCOLLEWAERT OF SMALL BUSINESS AND MANIGART MANAGEMENT 15 Ibrahim, D. (2008). “The (Not So) Puzzling Mason, C., and M. Stark (2004). “What Do Behavior of Angel Investors,” Vanderbilt Investors Look for in a Business Plan?,” Law Review 61, 1405–1452. International Small Business Journal 22, Kaplan, S. N., and P. Strömberg (2004). “Char- 227–248. acteristics, Contracts, and Actions: Evidence Matlin, M. W. (2005). Cognition, 6th interna- from Venture Capitalist Analyses,” Journal of tional ed. New York: John Wiley. Finance 59, 2177–2210. Maxwell, A., S. A. Jeffrey, and M. Lévesque Kelly, P., and M. Hay (2000). “Deal-Makers”: (2011). “Business Angel Early-Stage Decision Reputation Attracts Quality,” Venture Making,” Journal of Business Venturing 26, Capital 2, 183–202. 212–225. ——— (2003). “Business Angel Contracts: the Mudd, S., K. Pashev, and N. Valev (2010). “The Inﬂuence of Context,” Venture Capital 5, Effect of Loss Experiences in a Banking 287–312. Crisis on Future Expectations and Behavior,” Knockaert, M., B. Clarysse, and M. Wright The B.E. Journal of Macroeconomics 10. (2010). “The Extent and Nature of Heteroge- Article 32. neity of Venture Capital Selection Behavior Persky, J. (1995). “Retrospectives: The Ethology in New Technology-Based Firms,” R&D of Homo Economicus,” The Journal of Eco- Management 40, 357–371. nomic Perspectives 9(2), 221–231. Kolb, D. A. (1984). Experiential Learning: Politis, D. (2005). “The Process of Entrepre- Experience as the Source of Learning and neurial Learning: A Conceptual Framework,” Development. Upper Saddle River, NJ: Entrepreneurship Theory and Practice 29, Prentice-Hall. 399–424. Landström, H. (1992). “The Relationship ——— (2008). “Business Angels and Value between Private Investors and Small Firms: Added: What Do We Know and Where Do An Agency Theory Approach,” Entrepre- We Go?,” Venture Capital 10, 127–147. neurship and Regional Development 4, 199– Raudenbusch, S., and A. Bryk (2002). Hierar- 223. chical Linear Models: Applications and Data ——— (1993). “Informal Risk Capital In Sweden Analysis Methods. Thousand Oaks, CA: Sage and Some International Comparisons,” Publications. Journal of Business Venturing 8, 525–540. Rotefoss, B., and L. Kolvereid (2005). “Aspiring, Manigart, S., and M. Meuleman. (2004). Financ- Nascent and Fledgling Entrepreneurs: An ing Entrepreneurial Companies: How to Investigation of the Business Start-Up Raise Equity as A High-Growth Company. Process,” Entrepreneurship and Regional Brussels: De Boeck & Larcier. Development 17, 109–127. Mason, C. (2006). “Informal Sources of Venture Sapienza, H., S. Manigart, and W. Vermeir Finance,” in The Life Cycle of Entrepreneur- (1996). “Venture Capitalist Governance and ial Ventures. Ed. S. Parker. New York: Value Added in Four Countries,” Journal of Springer, 259–299. Business Venturing 11, 439–469. Mason, C., and R. Harrison (1995). “Closing the Shane, S. (2000). “Prior Knowledge and the Regional Equity Gap: The Role of Informal Discovery of Entrepreneurial Opportuni- Venture Capital,” Small Business Economics ties,” Organization Science 11, 448–469. 7, 153–172. Smith, D., R. Harrison, and C. Mason (2010). ——— (1996). “Informal Venture Capital: A “Experience, Heuristics and Learning: The Study of the Investment Process, the Post- Angel Investment Process,” Frontiers of Investment Experience and Investment Per- Entrepreneurship Research 30. Article 3. formance,” Entrepreneurship and Regional Sohl, J. E. (2003). “The U.S. Angel and Venture Development 8, 105–126. Capital Market: Recent Trends and Develop- ——— (2002a). “Is It Worth It? The Rates of ments,” The Journal of Private Equity 6, Return from Informal Venture Capital Invest- 7–17. ments,” Journal of Business Venturing 17, Sorensen, M. (2007). “How Smart Is Smart 211–236. Money? A Two-Sided Matching Model of ——— (2002b). “Barriers to Investment in the Venture Capital,” Journal of Finance 62, Informal Venture Capital Sector,” Entrepre- 2725–2762. neurship and Regional Development 14, Sorheim, R. (2005). “Business Angels as Facili- 271–287. tators for Further Finance: An Exploratory

16 JOURNALCOLLEWAERT OF SMALL BUSINESS AND MANIGART MANAGEMENT 371 Study,” Journal of Small Business and Enter- Wiltbank, R., and W. Boeker. (2007). “Returns prise Development 12, 178–191. to Angel Investors in Groups,” Kauffmann Stroobandt, F., J. Lievens, and H. Waege (2005). Foundation Research Report. “Cultuurparticipatie in Het Europa Van de Wong, A. (2010). “Angel Finance: The Other 15. Vlaanderen ‘Best in Class’?,” in Venture Capital,” in Companion to Venture Cultuurparticipatie in Breedbeeld. Eds. J. Capital. Ed. D. J. Cumming. Hoboken, NJ: Lievens and H. Waege. Antwerp: Uitgeverij Wiley Press, 71–110. De Boeck, 233–264. Wooldridge, J. (2002). Econometric Analysis of Van Osnabrugge, M. (2000). “A Comparison of Cross Section and Panel Data. Cambridge, Business Angel and Venture Capitalist MA: MIT Press. Investment Procedures: An Agency Theory- Zacharakis, A. L., and D. A. Shepherd (2001). Based Analysis,” Venture Capital 2, 91–109. “The Nature of Information and Overconﬁ- Vance, D. E. (2005). Raising Capital. New dence on Venture Capitalists’ Decision York: Springer. Making,” Journal of Business Venturing 16, Wiltbank, R. (2005). “Investment Practices and 311–332. Outcomes of Informal Venture Investors,” Venture Capital 7, 343–357.

372 JOURNALCOLLEWAERT OF SMALL BUSINESS AND MANIGART MANAGEMENT 17 Journal of Small Business Business Management Management 2016 2015 54(1), ••(••), pp. pp. 373–391 ••–•• doi: 10.1111/jsbm.12151 Linking Unlearning with Work–Life Balance: An Initial Empirical Investigation into SMEs by Juan-Gabriel Cegarra-Navarro, Maria-Eugenia Sánchez-Vidal, and David Cegarra-Leiva

The use of work–life balance (WLB) supporting cultures to provide and support the welfare of both the individuals and their families is expected to improve organizational outcomes by enabling company members to improve levels of job satisfaction and commitment to work. However, creating or introducing another culture in the company frequently generates internal problems when it conflicts with the current culture in terms of the beliefs, habits, and things managers take for granted. In this paper, we develop a framework for modeling how the unsuitable knowledge may be mitigated in order to strengthen a WLB culture. It is based on the idea that the presence of an unlearning context that fosters the updating of knowledge is likely to be essential for small and medium-sized enterprises (SMEs) that are trying to implement a WLB culture. The research model and hypothesized relationships are empirically tested using the structural equation modeling approach, validated by factor analysis of 229 SMEs in the Spanish metal industry. Our findings show that in order to strengthen a WLB culture and thus positively influence innovation- related outcomes, SMEs meet the challenge of developing an unlearning context to counteract the negative effects of the outdated knowledge in relevant areas and to facilitate the replacement of out-of-date or obsolete knowledge.

their use. Research has revealed that organiza- Introduction tional support for WLB (operationalized in this A work–life balance (WLB) is a concept refer- paper as WLB culture) has a greater impact on ring to the equal prioritization of career and employees’ and organizational outcomes than personal life development. In recent years, WLB the implementation of formal WLB initiatives. practices have ranked among the most popular The question of a WLB culture is an aspect beneﬁts for employees as they can help foster that becomes key, particularly in small and employees’ quality of life, and as a consequence, medium-sized enterprises (SMEs), where the workers will be more satisﬁed, motivated, and adoption and use of WLB practices by the committed to the company (Allen 2001). employees depend more on the existence of However, scholars have highlighted that the positive values toward WLB than on the avail- existence of WLB practices does not guarantee ability of formalized WLB practices (De Kok

Juan-Gabriel Cegarra-Navarro is associate professor of Business Administration in the Departamento de Economía de la Empresa at the Universidad Politécnica de Cartagena. Maria-Eugenia Sánchez-Vidal is associate professor of Business Administration in the Economía de la Empresa at the Universidad Politécnica de Cartagena. David Cegarra-Leiva is assistant professor of Business Administration in the Economía de la Empresa at the Universidad Politécnica de Cartagena. Address correspondence to: Juan Gabriel Cegarra-Navarro, Economía de la Empresa, c/ Real 3, Cartagena, Murcia 30201, Spain. E-mail: [email protected].

CEGARRA-NAVARRO, SSÁNCHEZ-VIDAL,ANCHEZ-VIDAL, AND CEGARRA-LEIVA CEGARRA-LEIVA 373 1 and Uhlaner 2001; Dex and Scheibl 2001; will look more closely at what can be done to Harrington and Ladge 2009). In this regard, counteract the negative effects of unsupportive SMEs often do not have the resources (finan- attitudes and behavior of managers toward cial, human, equipment, etc.) to develop flex- WLB practices. We also develop a framework ible working structures or to facilitate WLB for modeling how unsuitable knowledge may opportunities (Hughes and Bozionelos 2007). be mitigated in order to strengthen a WLB As a result, WLB culture within SMEs is often culture. It is based on the idea that the presence more insular than in larger organizations. of an internal context that fosters the updating Employees in SMEs often multitask and lack of knowledge is likely to be essential for SMEs the time, training, or inclination to examine the that are trying to implement a WLB culture. opportunities that a WLB can provide (Martin From this perspective, unlearning is the process and Chapman 2006). In such situations, a of replacing unsuitable knowledge with new WLB culture becomes extremely important for knowledge where appropriate. This process the welfare of both the individuals and their will help those managers seeking to enhance families. effectiveness via a WLB culture. The key benefits of encouraging a WLB There is no empirical evidence, particularly culture in SMEs are clear; it enables them to in relation to SMEs, to support the concept of retain skilled workers, reduces costs through an unlearning context (UC) and how it relates better productivity and efficiency (improving to a WLB culture, or to the improvement of services to customers), and increases profitabil- organizational innovation outcomes (e.g., ity (Hughes and Bozionelos 2007). However, Akgün et al. 2007; Hughes and Bozionelos despite the opportunities that a WLB culture 2007; Lavoie 2004). In this paper, we test our will offer as a result of its influence on produc- postulated hypotheses about such relation- tivity, turnover, and absenteeism, the fact is ships, the UC, the WLB culture, and the that SMEs have been slow to adopt WLB cul- improvement of innovation-related outcomes. tures (Kirby and Krone 2002). This is largely These relationships are examined through an because the creation of a WLB culture sup- empirical investigation of 229 SMEs in the poses, in each case, the reactivation and devel- Spanish metal sector. The subprocesses that opment of new values which, in turn, foster characterize the UC and the WLB culture are learning and the integration of new knowledge discussed in detail in the following section. The in members of the organization (Poelmans, third section investigates the development of Chinchilla, and Cardona 2003). hypotheses as to how the UC and the WLB It should be noted, however, that creating or culture contribute to innovation-related out- introducing another culture in the company comes. Details of the survey which was used to frequently generates internal problems when it collect appropriate data to test the models is conflicts with the current culture in terms of the presented in the fourth section, whereas the beliefs, habits, and things managers take for results of testing the models are presented in granted (Argyris and Schön 1978). Regarding the fifth section, followed by a discussion in this, previous studies in business management sixth section. confirm that the numerous failures in the implementation of WLB cultures are a conse- Conceptual Framework quence of incompatible organizational cultures, Research has shown that lack of balance in such as an entrenched culture of long working one’s life is related to higher stress, less life hours and the unsupportive attitudes and satisfaction, and lower work effectiveness behavior of senior managers, line managers, (Kofodimos 1995). WLB initiatives are consid- and colleagues (Poelmans, Chinchilla, and ered among the human resource practices that Cardona 2003). are believed to be more relevant today The ideas just outlined provide an illustra- (Fleetwood 2007). Although there is no clear tion that in order to strengthen a WLB culture definition of WLB, it could be described as the and thus positively influence innovative “individuals’ ability of finding a life rhythm that performance, an SME must be flexible when allows them to combine their work with other configuring (combining) knowledge in a responsibilities, activities or aspirations, irre- way that is appropriate for delivering value to spective of age and gender” (Felstead et al. the company and be effective in updating the 2002). In relevant literature, authors have con- actual knowledge of its staff. In this paper, we sidered two different domains regarding WLB:

2374 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT “WLB practice” and “WLB culture.” Several aca- organization’s values and practices) or because demics stress, for example, that formal WLB of their personal preferences (Carlson, Upton, organizational policies have less impact on and Seaman 2006). This means that in SMEs, organizations and employees than informal managerial support for WLB is the factor to be organizational support for WLB (Behson 2005). considered as managers are the key decision- Other authors also affirm that WLB efforts have makers who have the last word in deciding historically focused on implementing WLB who enjoys the benefits of WLB. If this argu- practices rather than changing the WLB culture ment is correct, understanding the nature and within organizations (Harrington and Ladge role of prior knowledge for sustainable devel- 2009). opment is a necessary step toward understand- WLB culture is defined as the “shared ing if new WLB practices will be used and how assumptions, beliefs and values regarding the they might be used. It is also worth noting that extent to which an organisation supports and the creation of this WLB culture may be a values the integration of employees’ work and challenging task for small organizations, par- family lives” (Thompson, Beauvais, and Lyness ticularly while their stakeholders are still 1999). Allen (2001) calls companies with actively involved with the business (Seiling high levels of WLB values “Family-Supportive 2008). It should be noted that many managers/ Organisations,” and Cappelli (2000) further owners encourage their employees to test the emphasizes that a WLB culture is a key element validity of their beliefs about the cause and of an organization’s employee retention strate- effect relationships that guide their behavior. In gies. There are several studies in relevant many cases, learned cultures are so organiza- literature examining which aspects characterize tionally inculcated that only at a time of crisis a WLB culture. One of the predominant will managers begin to question them approaches in this literature is the one devel- (Kimberley and Hartel 2008). oped by Thompson, Beauvais, and Lyness In an attempt to counteract the negative (1999) that divides WLB culture into three effects of inappropriate beliefs and combine subdimensions: managerial support for WLB, depreciated knowledge with new knowledge perceived negative career consequences of updated and acquired by managers and using work–family benefits, and organizational employees at the company, “unlearning” and time expectations. its different subdimensions are frequently cited This approach has been used by other as antecedents for the elimination of old logic researchers, such as O’Neill et al. (2009) who at the individual level and making room for found that managerial support for WLB was new logic at the organizational level (e.g., the most important factor of a WLB culture and Akgün et al. 2007; De Holan and Phillips 2004). the main determinant of lower levels of turn- At its heart, this unlearning process attempts over and employee commitment. Managerial to reorientate organizational values, norms, support for WLB indicates how managers and/or behavior by changing cognitive struc- empathize with employees’ efforts to balance tures (Nystrom and Starbuck 1984), mental their work and life and how they are sensitive models (Day and Nedungadi 1994), dominant to their familial responsibilities (O’Neill et al. logics (Bettis and Prahalad 1995), and core 2009). In this regard, based on their reading of assumptions which guide behavior (Shaw and Kofodimos (1995) and Burke (2004), Burke, Perkins 1991). Burgess, and Fallon (2006), explain that when In recent organizational learning literature, workers negotiate the use of WLB practices the term unlearning has been analyzed in two with their managers (supervisors), the positive related streams (i.e., international unlearning attitude of managers toward “WLB” becomes a and accidental unlearning). Whereas accidental key factor that supports and encourages a WLB unlearning results from the loss of knowledge culture in large and small organizations. Hence, as a result of such events as the loss or acci- in this paper, the term WLB culture is used to dental deletion of computerized data, the describe companies in which WLB values are failure to preserve adequate documentation in supported and encouraged by management. order to interpret reports, the failure of sensor It is worth noting that in SMEs, managers readings, and so on, intentional unlearning is implement WLB initiatives based on personal an intentional process that explicitly addresses concerns that are expressed by employees or the discrepancy between the knowledge relat- other stakeholders (groups affected by the ing to the current reality and the existing body

CEGARRA-NAVARRO, SSÁNCHEZ-VIDAL,ANCHEZ-VIDAL, AND AND CEGARRA-LEIVA CEGARRA-LEIVA 375 3 of knowledge (Fernandez and Sune 2009). being, and it is through such a framework that Intentional unlearning typically takes place individuals of an organization will have access when there is some indication that extant to new perceptions; (2) the framework for knowledge is obsolete or inappropriate requir- changing individual habits (CIH): This refers to ing that “old” knowledge be replaced with new the challenge of inhibiting wrong habits when organizational knowledge (Lei, Slocum, and an individual has not only understood the new Pitts 1999). If intentional unlearning (unlearn- idea but is quite motivated to make the change; ing) involves intentionally abandoning ways of and (3) the framework for consolidating emer- behaving, habits, assumptions, and knowledge gent understandings (CEU): This refers to the and knowledge structures, then, as a number of organizational process that can free employees researchers have argued, intentional unlearning up to apply their talents by implementing new has to be initiated by the individual (e.g., De mental models based on adaptation to new Holan and Phillips 2004; Lei, Slocum, and Pitts knowledge structures. 1999). The UC just shown is not only a way to Along with other authors (e.g., Carlson, unlearn outdated knowledge but also the way Upton, and Seaman 2006; Chell 1993), we that companies are able to relearn and develop argue that while in situations where organiza- updated knowledge. As Sinkula, Baker, and tions and their members face rapidly changing Noordewier (1997) noted, as workers relearn, environments, it is necessary for the knowledge they also unlearn, and new knowledge replaces provided by groups affected by the organiza- old routines as the workers forget. As intention’s values and practices, both explicitly and tional unlearning involves both individual and tacitly in the form of processes and routines, to organizational change (Becker 2010), it is be challenged prior to its adoption as new appropriate to say that it is operationalized knowledge by the organization (i.e., intentional though three different mechanisms (i.e., aware- unlearning). Regarding this, Becker (2010) ness, relinquishing, and relearning). “Aware- has examined the antecedents of intentional ness” is operationalized through the ELF unlearning with specific focus on the organiza- framework (e.g., noticing failures, mistakes, tional context. Just as with learning, intentional problems, and paying attention to customers). unlearning takes place within an organizational “Relinquishing” is operationalized through context. Therefore, as Azmi (2008) noted, a UC tactics such as breaking habitual stimuli– can be ingrained in the organizational structure response connections (i.e., the CIH frame- by creating and supporting a culture where work). The CEU framework is associated with people consciously acquire new skills and the concept of “relearning,” through which knowledge in tandem with creating both the management supports an adaptation process time and opportunity to examine and explore where new information structures have to existing and new knowledge. Through a trans- replace old ones (Akgün et al. 2007). formation of the organizational context, we may cause individuals to pay more attention to Hypotheses environmental cues, mistakes, and wrong The previous discussion provides an illustra- assumptions. Thus, the contribution of the tion that a WLB culture requires organizational “UC” relates to its ability to pave the way for members to learn new ways of acting and new knowledge. absorbing new thought patterns. However, cre- In order to reduce the likelihood of an insuf- ating new knowledge frequently generates ficient amount of knowledge being captured or internal problems when it conflicts with current uncritical knowledge being captured, we would knowledge in terms of the beliefs, habits, and propose that it is necessary for SMEs to imple- things individuals take for granted (Argyris and ment a UC to update the relevant knowledge of Schön 1978). Without care, managers can fall its staff. In order to do this, we argue in this into a “competence trap” (Leonard-Barton paper that SMEs have to actively develop a UC. 1992; Levinthal and March 1993), increasingly In this paper, we have followed the suggestion exploiting prior experience, relations, and of Cegarra, Eldridge, and Martinez (2010) by knowledge. As Macinnes (2005) stated, one of considering that a UC requires the presence of the factors preventing new knowledge struc- three preconditions: (1) the examination of lens tures is managers’ negative attitude toward fitting (ELF): This refers to an interruption of them due to perceived ambiguous results (e.g., the employees’ habitual, comfortable state of the perceived cost of implementing some

4376 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT initiatives and doubts with respect to the sub- The question to consider now is whether a sequent benefits). It should be also noted that WLB culture has a direct effect on innovation- inertia (i.e., the utilization of outdated knowl- related outcomes (IROs). Regarding this, rel- edge) potentially lead to degradation in some evant literature stresses that the existence of a managerial functions (e.g., commanding, orga- WLB culture helps both employees and organizing, coordinating, and controlling) and, nizations obtain beneficial results. Among the hence, potentially serious negative outcomes benefits of implementing a WLB culture, aca- (Hannan, Burton, and Baron 1996). demics highlight improvement in public image The ideas just outlined provide an illustra- as the company is seen to be socially respon- tion that knowledge and knowledge structures sible (Harrington and Ladge 2009). There are may become outdated or less relevant over also other more tangible benefits, such as time (Akgün et al. 2007), and thus, top man- reduction of absenteeism, lower stress levels, agement teams may bring to their small firms higher levels of productivity and performance, their prior outdated experience and knowledge and greater quality of life, satisfaction, and structures (Sinkula, Baker, and Noordewier commitment among employees (Hughes and 1997). Therefore, in order to strengthen a WLB Bozionelos 2007; Nelson et al. 1990; Scandura culture, all of the individual manager’s knowl- and Lankau 1997). In addition, a WLB culture edge needs to be reviewed and updated where is considered a key for retaining and attract- appropriate when they are going through a ing highly qualified professionals (Harrington transition from a culture based on traditional and Ladge 2009). Moreover, as O’Neill et al. managerial functions (e.g., commanding, orga- (2009) argue, companies should take care of nizing, coordinating, and controlling) to a WLB their leaders’ WLB, implementing and negoti- culture. Otherwise, old logic and routines ating WLB practices with them, because the about how management must perform and crossover effects of their work-to-family rela- what is possible and what is not possible that tionship spill over to the intentions of other have not only become outdated but also con- employees in the company with regard to tinue to surface unexpectedly can hinder the commitment and turnover. O’Neill et al. implementation of WLB initiatives. (2009) call this phenomenon the “contagion As just indicated, to get an updated view of effect.” work–life conflict and to understand its effects, In this paper, we focus on the importance of managers have to examine the phenomena developing a WLB culture within SMEs that from a number of different angles. If they come foster employee outcomes. This way, we to rely uncritically on implicit knowledge, they provide an illustration that it is the positive are likely to become less creative (Sinkula, employee outcomes stemming from a WLB Baker, and Noordewier 1997). Under these cir- culture that in turn improve organizational out- cumstances, the atmosphere most likely to comes (Burke, Burgess, and Fallon 2006). In induce self-renewal and essentially correct this terms of organizational outcomes, a “WLB state of affairs is one that promotes openness to culture” facilitates the participation of all stake- new ideas and the critical evaluation of signals holders and acknowledges the interdepen- that are contrary to established beliefs and dence of the problems within an organization. values (Akgün et al. 2007). Thus, we argue that It proceeds from a treatment of the whole to it is through the implementation of processes the interaction of the parts and then finally to that result in the establishment of a “UC” that the parts themselves. For example, a WLB members of an organization are able to coun- culture is a source of higher levels of produc- teract the negative effects of outdated knowl- tivity and performance and greater quality of edge. Such a context may also lead to new life, satisfaction, and commitment among interpretations of existing values, behaviors, employees (Burke 2004; Burke, Burgess, and attitudes, and habits or the elimination of what Fallon 2006; Kofodimos 1995), which allows was formerly considered to be knowledge or the incorporation of authentic value into plan- accepted wisdom (Huber 1991; Nonaka and ning, generates consensus and commitment Takeuchi 1995). The hypothesis we propose and mobilizes participants, releases suppressed under this framework is creativity, and harnesses it to individual and organizational development (Thompson 1998). H1: The UC is positively associated with the As the impact of a WLB culture on organi- culture supporting a WLB. zational outcomes may differ depending on

CEGARRA-NAVARRO, SSÁNCHEZ-VIDAL,ANCHEZ-VIDAL, AND AND CEGARRA-LEIVA CEGARRA-LEIVA 377 5 the outcome under analysis (e.g., internal, employees and their IROs. Under this frame- innovation-related and external organizational work, the hypotheses that we propose is outcomes), in this study, we have considered that a WLB culture should impact on their inno- H2: The degree of a culture supporting WLB is vation outcomes, such as new product/service positively related to the achievement of IROs. quality and development capacity (Bowonder et al. 2010; Nambisan 2009; Prahalad and Methods Krishnan 2008; Ramaswamy and Gouillart Data Collection 2010). We also adopted a question focusing on In order to test the hypotheses mentioned, the satisfaction of customers as a side effect of we use a sample of manufacturing companies innovation outcomes (Nambisan and Baron from the same industry. The metal industry in 2007; Nambisan and Nambisan 2008). This is south east Spain was the subject of our data justified due to the reason that happy employ- collection as it is one of the most important ees lead to happy customers (e.g., Yang and industrial sectors of the Spanish economy. Peterson 2004; Zeithaml, Parasuraman, and According to the State Industry Survey (INE Berry 1990). As Reichheld (1996) has noted, an 2007a), the metal industry produced an output organization with difficulty will achieve to of EUR 229,022 million, and it employed satisfy its customers if previously it has not 625,014 workers in 2007 (24.22 percent of been made with the workers. Therefore, in this Spanish industrial workers). Annual business paper, we focus on the need for creating a WLB turnover represents 36.59 percent of total culture that enables a starting point to bring industrial turnover. Metal transformation activ- together employees and managers in the imple- ity is composed of approximately 52,250 estab- mentation of a knowledge creation network lishments, in which 96.8 percent have fewer from which appropriate measures for custom- than 50 workers, and among them, 84.18 ers can derive. percent have fewer than 10 employees (INE Based on the previous discussion, this 2007a). paper adopts an innovative approach by Relevant literature affirms that traditional examining the effect of a WLB culture on manufacturing organizations are changing their “innovation-related outcomes.” It is a well- strategies, and it is difficult to find pure manu- known fact that people working in organiza- facturing companies that do not offer other tions are responsible for a large amount of services to clients. As Levitt (1972) pointed out, IROs (Andrews, 1975; Andrews & Smith, “all industries are effectively service industries.” 1996). In fact in order to be innovative, In this regard, Johnstone, Dainty, and someone needs to come with creative ideas. Wilkinson (2008) note that today, manufactur- Creativity seems to be a previous step to inno- ing companies do not consider services as a vation (Andrews, 1975). Organizational stress peripheral activity but a key aspect to guaran- and work load, which are some aspects of an tee business survival. In fact, in a recent study unbalanced WLB culture, inhibit creative by Martínez-Fernández (2010), the author finds thinking and therefore also might inhibit inno- that innovation in services in the mineral sector vative behavior (Andrews & Smith, 1996). With is fostering the transformation of the industry a culture supporting WLB, organizations may into the knowledge economy. Considering this, create a “family-friendly” environment to work the metal industry in south-east Spain was (Scandura and Lankau 1997), which in turn chosen for two main reasons, namely, its eco- may stimulate employees to the achievement nomic importance and the increasing environ- of IROs (Shipton et al., 2006). Regarding this, mental challenges it faces. The metal industry individuals belonging to a WLB culture might usually absorbs 29.05 percent of the total exchange more information with managers employment in industry in the south-east of and other employees and might therefore Spain (INE 2007b); 91.38 percent of the com- likely to be more involved in their work panies employ fewer than 50 workers, and (Scandura and Lankau 1997). As Shipton et al. among them, 83.46 percent of the companies (2006) noted, a culture supporting WLB do not have more than 10 workers (INE 2007b). involves the generation of new ideas through Furthermore, the metal industry in south-east searching for alternative viewpoints and per- Spain is an appropriate setting for an investi- spectives. Put another way, leaving the WLB gation into a UC and its impact on a WLB- out of consideration might be detrimental for supporting culture (WBC) because the sector

6378 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT has suffered increased competition in the last European Union classification as SMEs,1 with a year, which creates a very dynamic environ- total number of employees under 250 (average ment with high levels of competition, requiring workforce in 2008: 27.59 workers). Most of the the intensive use and updating of knowledge companies employ between 10 and 49 indi- (Confemetal 2013). viduals: 88.2 percent of them have fewer than According to the Establishments Industrial 50 workers and 23.58 percent fewer than 10. Of Register (Carm 2009), the metal industry is com- the firm sample, 71.7 percent are family busi- posed of 832 SMEs. We used a list of small nesses, and the mean tenure of the companies companies (832 companies) provided by the is 20 years. Regarding the gender of the Establishments Industrial Register (http:// employees, 90.5 percent of the firms have less www.carm.es) as an initial sampling frame. All than 25 percent of women in their workforce companies were invited to participate in the as they belong to a male-dominated sector. study, and 263 agreed. In a first visit, managers Average annual turnover of the companies is were given a questionnaire to be filled in and EUR 9,107,386. were given an explanation of every part of The information was collected by personal the questionnaire. They were assured of its visits to all companies that agreed. These data strictly scientific and confidential character, as were collected by personal visits to the chief well as the global and anonymous treatment of executive officer (CEO) or general director of the data. They were also told that in case of each of the SMEs accompanied by a question- doubt, they could contact the researchers who naire. This means that only subjective informa- would answer their questions. In a second tion relating to the adoption of organizational visit, they were asked to give the questionnaire unlearning and WLB culture at the managerial back. In total, 610 visits were carried out, and level was solicited. As just noted, the majority 230 surveys were collected. A comparison of companies involved in the study were small between companies who had answered and enterprises where the CEO and the HR companies who had not answered yielded no manager were the same person. Put in another significant differences relevant to total assets way, the CEO or general director addressed and number of employees, which suggested that was expected to have a broad overview of the nonresponse bias was not a problem (Armstrong innovative issues studied in this paper. and Overton 1977). This information is available from the “SABI” Database (based on the statis- Measures tical year 2007), and the independent sample Churchill’s (1979) approach to questionnaire t-test revealed no significant difference between development was used. Scales were combined the two groups (p = .427 and .396, respectively). from several other relevant empirical studies Therefore, nonresponse bias should not be a with new items to make an initial list of 20 problem in this study (Armstrong and Overton items (four measuring the range of the ELF; 1977). four measuring the existence of conditions Surveys were carried out over a period of facilitating the changing of individual habits, eight months, from March to October 2009. four measuring the framework for CEU, five Only those companies which provided com- measuring the range of the WBC, and three plete answers to the survey questions were relating to IROs). Several items were modified included in our analysis. As a result, one through interviews with colleagues, and a first company was excluded for failing to provide draft of the questionnaire was tested with three complete answers. Consequently, we had 229 SMEs. Appendix provides an overview of the complete surveys giving a response rate of final 20 questions used in the questionnaire. 27.52 percent of the total number of companies The questionnaire was developed as follows: invited to participate (87.07 percent response rate from the companies who agreed), with a • Three dimensions form the UC: “consolida- factor of error of 5.5 percent for p =q=50 tion of emergent understandings,” “the percent and a reliability level of 95.5 percent. ELF,” and “the framework for CIH.” The All companies were classified according to the initial scale comprises 10 items taken from a

1According to the European Commission (2003), SMEs comprise fewer than 250 employees, with an annual turnover not exceeding EUR 50 million and an annual balance sheet total not exceeding EUR 43 million.

CEGARRA-NAVARRO, SSÁNCHEZ-VIDAL,ANCHEZ-VIDAL, AND AND CEGARRA-LEIVA CEGARRA-LEIVA 379 7 scale designed by Cegarra, Eldridge, and of each one. Initially, each construct was Martinez (2010) to quantify the construct of assessed using the item-to-total correlation and a UC. Consequently, the framework for exploratory factor analysis. The application of examining the lens fitting was measured this technique requires an initial model to be using four items which recognize the proposed so that if data do not fit well, then the support of policies, rules, reporting, struc- model can be modified until a good fit is tures, and decision-making protocols that achieved. One model is transitioned to the encourage the identification of problems, other with the removal of those items which mistakes, and new ways of doing things. To did not converge substantially with their measure the process for “CIH,” four items respective latent variable. The Satorra-Bentler were used. This scale focuses on self- χ2 difference test was employed using available awareness issues about our own mistakes, software (Crawford 2007) to provide a signifi- ways of thinking, and wrong behavior that cance test of the relative goodness of fit (GFI) guide our everyday attitudes. The measures between nested models. As a result of the relating to the existence of a process for exploratory analysis, several items were “consolidating the emergent understand- dropped. To perform this, analysis loadings ings” describe the way organizational and regression weights were examined members faced up to change, introduced it (Anderson and Gerbing 1988). After the refine- actively into the company through projects, ment process, 15 items were used (three for collaborated with other members of the each construct: ELF, CIH, CEU, WBC, and IRO). organization, and recognized the value of new information or taking risks. Assessment of the Measures • WBC was measured using a scale of five In order to obtain a more robust evaluation items taken from previous studies (Burke of the quality of the resulting 15 items, a con- 2004; Burke, Burgess, and Fallon 2006; firmatory factor analysis (CFA) was carried out Kofodimos 1995). The final measures relat- using the covariance matrix as input via the ing to the existence of “the WBC” scale EQS 6.1 (Structural Equation Modeling Soft- consisted of five items adapted from a scale ware, Multivariate Software, Inc, Ventura Bou- by Kofodimos (1995) to measure the con- levard, Encino, CA 91436, United States) struct of the WBC. These items recognize (Bentler 1988) robust maximum likelihood support of policies, rules, reporting struc- method. As our model uses reflective indica- tures, and decision-making protocols that tors of a principal factor latent construct and encourage the introduction of work–family our data are non-normal, other techniques of benefit measures and approaches to structural equation modeling (SEM; e.g., improve workers’ family lives. LISREL or AMOS) cannot be applied in these • In this study, the measurement of IROs is circumstances (e.g., Diamantopoulos and based on the stakeholder approach, which Winklhofer 2001). The CFA produced a good considers the different types of perceived fit with an incremental fit index (IFI) of benefits that are gained by managers. 0.98 and a comparative fit index (CFI) of 0.98 2 Therefore, IROs were measured by asking (furthermore, Satorra-Bentler χ (80) = 103.87; the managers to evaluate different aspects χ2/df = 1.29; GFI = 0.93; root mean square of their business results over the last three error of approximation [RMSEA] = 0.037). It years compared with competitors, as sug- should be noted here that when applying mul- gested by Delaney and Huselid (1996). This tivariate analysis in management disciplines, construct is composed of three items: (1) the assumption that the empirical data origi- quality of products, services, and programs; nate from single homogeneous population (2) development of new products and ser- (i.e., data drawn from a single population) is vices; and (3), satisfaction of customers or often unrealistic. It exists what is denominated clients. Answers were measured on seven- such unobserved heterogeneity. Therefore, we point scale where 1 = worse than competi- show that taking into account unobserved het- tors and 7 = better than competitors. erogeneity is important both in terms of obtaining information of the UC and also in An exploratory factor analysis of the five terms of obtaining unbiased parameters of the constructs (i.e., ELF, CIH, CEU, WBC, and IRO) observed independent variable of the model. was carried out to assess the unidimensionality This heterogeneity can affect both the

8380 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Table 1 Factor Loadings of the Resulting Items and Scale Reliability

Construct Standardized t-Value Reliability Loading (SCRa, AVEb)

The Consolidation of Emergent Understandings (CEU) CEU_1 0.85 12.52 AVE = 0.63 CEU_2 0.83 13.17 SCR = 0.83 CEU_3 0.81 13.27 The Framework for the Examination of Lens Fitting (ELF) ELF_2 0.70 10.16 AVE = 0.67 ELF_3 0.68 11.98 SCR = 0.85 ELF_4 0.80 15.35 The Framework for Changing the Individual Habits (CIH) CIH_2 0.93 17.97 AVE = 0.57 CIH_3 0.94 16.72 SCR = 0.80 CIH_4 0.76 11.26 Work life Balance Culture (WBC) WBC_3 0.73 10.75 AVE = 0.65 WBC_4 0.85 13.54 SCR = 0.85 WBC_5 0.69 10.14 Innovation-Related Outcomes (IRO) IRO_1 0.78 11.77 AVE = 0.67 IRO_2 0.75 11.36 SCR = 0.85 IRO_3 0.63 9.30

2 2 The fit statistics for the measurement model were: Satorra-Bentler χ (80) = 103.87; χ /df = 1.29; GFI = 0.93; CFI = 0.98; IFI = 0.98; RMSEA = 0.037. a 2 2 Scale Composite Reliability (SCR) of pc =(Σλi) var (ξ)/[(Σλi) var (ξ)+Σθii] (Bagozzi and Yi 1988). b 2 Average variance extracted (AVE) of pc =(Σλi2 var (ξ))/[Σλi var (ξ)+Σθii] (Fornell and Larcker 1981). The asymptotic covariance matrices were generated to obtain the scaled chi-squared (Satorra and Bentler 1988) and robust estimation of standard errors. measurement part (e.g., different latent vari- Yi’s (1988) composite reliability index and able means in each segment) and the structural Fornell and Larcker’s (1981) average variance part (e.g., different relationships between the extracted index are higher than the evaluation latent variables in each segment) of our causal criteria of 0.7 for composite reliability and 0.5 model. In order to address that unobserved for the average variance extracted. These heterogeneity, we performed a finite mixture results suggest the use of three scales to algorithm to probe the existence or nonexis- measure the framework for examining the lens tence of unobserved heterogeneity (Jedidi et al. fitting (AVE = 0.67; SCR = 0.85), three to 1996). We ran the algorithm until five seg- measure the framework for changing the indi- ments. All the entropy statistics are below 0.5 vidual habits (AVE = 0.57, SCR = 0.80), three that means there is no unobserved heteroge- items to measure the consolidation of emergent neity for these data and this model. understandings (AVE = 0.63, SCR = 0.83), three From an examination of the results shown in to measure the WBC (AVE = 0.65, SCR = 0.85), Table 1, we can state that all of the constructs and finally, another three to measure the IROs are reliable. For all the measures, Bagozzi and (AVE = 0.67, SCR = 0.85).

CEGARRA-NAVARRO, SSÁNCHEZ-VIDAL,ANCHEZ-VIDAL, AND AND CEGARRA-LEIVA CEGARRA-LEIVA 381 9 Table 2 Construct Correlation Matrixa

Correlation matrix

Mean S.D. CA 12345

Consolidation of Understandings 5.42 1.27 0.86 0.79 (Range 1.33–7) Examination of Lens Fitting 4.97 1.19 0.77 0.47 0.81 (Range 1.67–7) Changing the Individual Habits 5.33 1.23 0.89 0.60 0.49 0.75 (Range 1–7) Work–Life Balance Culture (Range 1–7) 4.44 1.29 0.79 0.47 0.38 0.42 0.80 Innovation-Related Outcomes 5.32 0.89 0.76 0.47 0.34 0.38 0.32 0.81 (Range 3–7) aMean, the average score for all of the items included in this measure; S.D., standard deviation; CA, Cronbach’s alpha; Intercorrelations are presented in the lower and shady triangle of the matrix. The bold numbers on the diagonal are the square root of the average variance extracted.

Discriminant validity was determined by second-order factor model demonstrated a com- comparing the square root of the AVE (i.e., the posite UC in this study. diagonals in Table 2) with the correlations among constructs (i.e., the lower triangle of the Results matrix in Table 2). On average, each construct Once the psychometric properties of the related more strongly to its own measures than measures had been checked, the next step was to others (Fornell and Larcker 1981). In the to evaluate the hypothesized relationships interest of thorough discriminant validity, an developed from consideration of the relevant additional test was examined, which supports literature (see Figure 1), discussed in the text as this assumption as the confidence interval (±2 H1–H2. The fit of the model is satisfactory standard errors) around the correlation esti- (Satorra-Bentler χ2(84) = 125.18; χ2/df = 1.49; mated between any two latent indicators never GFI = 0.91; CFI = 0.96; IFI = 0.96; RMSEA = includes 1.0 (Anderson and Gerbing 1988). The 0.048), thereby suggesting that the nomological constructs correlation matrix, shared variances, network of relationships fits our data – another means, and standard deviations are shown in indicator of support for the validity of these Table 2. scales (Churchill 1979). Figure 1 shows that the The UC was operationalized as a second- UC had a positive influence on the existence of order construct with three dimensions (i.e., the a WBC at a level of (p < .01). We also estimate three facets of the UC). A second-order CFA of a the Sobel’s critical ratio suggested by Sobel model depicting the consolidation of emergent (1982) to test the indirect effect of independent understandings, the ELF, and the framework for variables on the dependent variable by way of CIH was conducted. From an examination of the the mediator. If the effect of the independent results shown in Table 3, this model yielded variables differs significantly from zero, then 2 acceptable fit indices (Satorra-Bentler χ (24) = indirect mediation is assumed (Preacher and 39.22; χ2/df = 1.63; GFI = 0.96; CFI = 0.98; Leonardelli 2003). Figure 1 illustrates the sig- IFI = 0.98; RMSEA = 0.054). In addition, all first- nificant results of Sobel’s test for the indirect order and second-order factor loadings were effects of the UC on the IROs through the WBC; significant, thereby providing evidence that UC thus, the UC had a positive and significant is a multifaceted construct, construed from indirect impact on the IROs (κ = 0.271, p < .01). consolidation of emergent understandings, the Figure 1 also shows that the WBC at a level of ELF, and the framework for CIH. Hence, the (p < .01) had a significant effect on IROs.

10382 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Table 3 Second-Order Conﬁrmatory Factor Analysis of the Unlearning Context

First-Order Construct Indicator First Order Second Order

Loading t-value Loading t-value

Framework for Consolidating the CEU_1 0.853 —a Emergent Understandings (CEU) CEU_2 0.835 14.101 0.786 9.809 CEU_3 0.813 13.670 Framework for Examining the Lens ELF_2 0.710 —a Fitting (ELF) ELF_3 0.690 8.692 0.717 7.734 ELF_4 0.810 9.447 Framework for Changing the Individual CIH_2 0.940 —a Habits (CIH) CIH_3 0.940 23.409 0.835 11.225 CIH_4 0.721 13.715

Fit statistics for measurement model of nine indicators for three constructs: Satorra-Bentler 2 2 χ (24) = 39.228; χ /df = 1.63; GFI = 0.961; CFI = 0. 987; IFI = 0.987; RMSEA = 0.054. aFixed parameter.

Figure 1 of organizational performance have appeared Model Statistics in literature, and we adopt the growth-based measures proposed by McDougall et al. (1994), Roth and Ricks (1994), and Bontis, Chua, and R2=0.463 R2=0.159 Richardson (2000) for sales on total employees 0.681*** 0.399*** UC WBC IRO (productivity) and sales on total assets. It should be noted here that at the current moment, this information is not available for the 235 respondents from the database (SABI) The fit statistics for the measurement (based on the statistical year 2007). As a result, model were as follows: Satorra-Bentler 95 (from 230 to 135) respondents were 2 2 χ (84) = 125.18; χ /df = 1.49; GFI = 0.918; excluded because they did not provide com- CFI = 0.968; IFI = 0.969; RMSEA = 0.048. plete information to these questions. Conse- ***p < .01. The unlearning context → quently, in order to triangulate our perceptual organisational outcomes (indirect path coef- performance measure with both sets of data ficient: 0.271***). aSobel Test Statistic: 3.97*** (sample respondents and objective data), we computed via http://www.danielsoper.com/ run again the SEM using 135 complete surveys. statcalc/calc31.aspx (accessed September 26, The results indicated consistent correlation 2012) ratios. The effect of WBC on organizational performance had a standardized coefficient of β = 0.352, thus suggesting a positive influence Together, these results provided full support on the business performance at a level of for H1, UC → WBC, and also for H2, WBC → p < .01. Therefore, this offers a control for IROs. single respondent bias. Moreover, this rein- The fact that data on both dependent and forces the literature which also claims that self independent variables in our model come from report is a valid measure in the social sciences the same key informant is a weakness of the (e.g., Cepeda, Cegarra, and Jimenez 2012). study; therefore, objective measures have been Anyway, future research might profitably used to triangulate our perceptual performance sample multiple holders of knowledge within a measure with objective data. Several measures company and will be helpful in testing for

CEGARRA-NAVARRO, SSÁNCHEZ-VIDAL,ANCHEZ-VIDAL, AND AND CEGARRA-LEIVA CEGARRA-LEIVA 383 11 interrater reliability and improving the internal new WLB culture that is not compatible with validity of knowledge management studies. the current culture, especially for those companies urged to retain skills and knowledge for Discussion longer periods of time (Harrington and Ladge The use of WBC to provide and support the 2009). As Chirico (2008) noted, a bad knowl- welfare of both the individuals and their fami- edge creation network proves a source of dif- lies is expected to improve organizational out- ficulties and misalignments to the operational comes by enabling company members to environment of family companies, mostly due improve levels of job satisfaction and commit- to the peculiar nature of the decision-making ment to work (Hughes and Bozionelos 2007; processes in SMEs, which are generally not Kirby and Krone 2002). A review of relevant very formalized and are very centralized literature, however, questions this purported (Chirico and Salvato 2008). In this regard, this link between WLB practices and organizational paper provides a starting point to bring effectiveness. The majority of studies investigat- together managers and employees in the imple- ing the outcomes of work–life practices do not mentation of a WLB culture from which WLB measure negative perceptions and thus cannot practices for the diverse needs of the workforce support this proposed mediated relationship can derive. From our framework, we suggest (Eby et al. 2005). Regarding this, there is an that any SME wishing to implement a WLB increasing amount of research supporting the culture should initially make efforts to update notion that workers who make use of work–life the relevant knowledge of its workforce. This is practices suffer negative perceptions from col- vital to the processes that are needed to meet leagues and superiors (e.g., Allen and Russell the changing needs of building or introducing a 1999; Frye and Breaugh 2004). A possible new culture. explanation for these negative perceptions may This research’s second contribution is pro- relate to the fact that the majority of companies vided by the results of empirically testing the have only introduced WLB practices (e.g., flex- proposed hypotheses. This paper analyzes the ible hours, telework, and informational assis- relationships between a UC and WLB culture tance) without modifying the existing prevalent and tries to identify whether WLB culture culture in the company. impacts on the IROs through an empirical study Therefore, the first contribution of this of 229 SMEs in the Spanish metal industry. The research is to question the existing studies managerial implications of the relationships which relate to WLB practices and organiza- observed between the factors that constitute tional outcomes (e.g., Beauregard and Henry the conceptual framework shown in Figure 1 2009). Our findings show that in order to are discussed next in more detail. support IROs, managers need to provide and With respect to the test of H1, the results support a positive WLB culture. We think that support the position that in order to implement this is an important finding as potential for any a WBC and hence foster the adoption of WLB small enterprise to develop will depend sub- practices, companies need to provide and stantially on its ability to maintain a balance support a UC, which is customized and based between responsibilities at work and at home; on three frameworks: (1) the framework for thus, managers may be trapped in a suboptimal examining the lens through which individuals stable equilibrium. As many overloaded man- view situations; (2) the framework for CIH; and agers are paring their resources, they are not (3) the framework for CEU. One interpretation actively listening to their employees, and man- of this relationship is that through the UC, a agers may be overinvesting in the development company can allow individuals to adjust their of WLB practices preserving and following mental models and the nature of the assump- inappropriate traditions (e.g., anticipating tions shared to break current workplace career penalties upon workers who make use culture. Because old, outdated knowledge can of work–life practices). impede adaptation to new configurations, Adapting to a different culture is not an easy senior managers need to create a context of task, particularly while founders or owners are continuous unlearning. still actively involved with the business. For The considerations mentioned also imply example, not everyone has the time or inclina- that an UC can encourage individuals to question to build a WBC. Under this framework, tion not only the information they own but also unlearning could be a necessary task to adopt a whether their particular approach to adopt a

12384 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT new WLB culture is applicable or not (Sinkula, to set their people clear expectations and goals, Baker, and Noordewier 1997). Such question- understand their individual motivations, or ing may also lead to new interpretations of acknowledge their efforts; otherwise, they will existing knowledge or the elimination of what undermine any formal WLB practices. In doing was formerly considered to be knowledge or so, the concept of unlearning was explored by accepted wisdom. This accomplishes what capturing the processes behind the UC and authors, such as Yeandle et al. (2002), express testing its impact on the WBC. Our results also when they highlight that the workforce may support that the UC has an indirect effect on need to be aware of the WLB practices offered IROs through the WBC. What this could mean by their organizations and managers may need for a small company is that a WLB culture is to be aware of the barriers to the use of WLB useful for the discovery of workforce needs in practices and focus on identifying and under- real time (live); the WLB culture provides infor- standing the nature of those barriers during the mation about the wishes and necessities of the assessment. Moreover, results obtained from workforce, and by using the WBC, companies H1 show that the government and official insti- may reduce the development time for new tutions should focus their efforts to foster WLB and/or redesigned WLB practices and gain a initiatives not only on external factors (e.g., competitive advantage (e.g., redesigning the laws and regulations) but also on modifying practice at the same time as employee demands managers’ psychological aspects such as cre- and being the first to pay attention to work- ative thinking and open mindedness. As force requirements). Marcati, Guido, and Peluso (2008) affirm, this Organizations providing a WBC may be could be achieved through specific tools such able to generate cost savings by offering as communication campaigns or specific train- appropriately balanced practices that do not ing courses that should aim to train creative require an extensive initial outlay of resources. problem solving, critical thinking, lateral think- This idea is recognized by Delaney and ing, etc. Huselid (1996), Thompson, Beauvais, and With regard to H2, the results support the Lyness (1999) and Zhu, Chew, and Spangler position that IROs are likely to suffer if an SME (2005), who suggest that a WBC helps employ- does not have a WBC. This confirms the position ees respond to the ever-changing nature of the adopted by Malik and Wilson (1995) when they workplace and reduce stress, which in turn emphasize the role of organizational climate could lead to improved IROs. Put another in IROs and effectiveness. Therefore, for an way, through a WLB culture, individuals will SME to grow and prosper in a dynamic and achieve higher levels of job satisfaction and turbulent environment, such as the Spanish commitment by focusing WLB practices on metal industry during the period we have exam- problems that are more important for the orga- ined (Confemetal 2013), it is necessary for the nization, which in turn facilitates the easy management to foster a WBC (Allen 2001). This retrieval of relevant information, and is a is in accordance to the social exchange theory factor in superior performance. that emphasizes the importance of employee This study has some limitations. Firstly, only perceptions relating to managerial support. subjective information relating to the measure- Hence, this research highlights the importance ment of company performance was solicited. of support from the managerial team for WLB Therefore, external measures should be used to initiatives in organizations. This also confirms supplement subjective information (e.g., sales, the views of authors, such as Scandura and profits, or customer satisfaction). We would Lankau (1997) and Hughes and Bozionelos further observe that this study relies on the (2007), who argue that WLB is much more than assumption that the manager or general direc- flexible working hours—it is about employee tor of each of the SMEs has key knowledge of fulfillment and engagement resulting in the company, and therefore, the emphasis here increased productivity and competitive advan- is on finding commonalities among companies tage. Therefore, a WBC not only does benefit the in terms of the critical knowledge areas they workforce but it also increases company pro- considered important and had available. Sec- ductivity (Harrington and Ladge 2009). ondly, this research was performed in a specific This study provides managers with a better country and sector of activity, which might understanding of work–life experiences of the prevent the generalization of the results to workforce and highlights that managers need other sectors or countries. In addition, the

CEGARRA-NAVARRO, SSÁNCHEZ-VIDAL,ANCHEZ-VIDAL, AND AND CEGARRA-LEIVA CEGARRA-LEIVA 385 13 transverse nature of this research impedes analysis of the effect of time on organizational References results, the WBC, and level of implementation Akgün, A. E., J. C. Byrne, G. S. Lynn, and H. of WLB practices. Finally, in the model we have Keskin (2007). “Organizational Unlearning been using, there is very likely to be a reverse as Changes in Beliefs and Routines in Orga- causality problem—performance drives adop- nizations,” Journal of Organizational tion of a WLB culture and favors unlearning. A Change Management 20(6), 794–812. possible explanation for that causality problem Allen, T. D. (2001). “Family-Supportive Work may relate to the advantages and disadvantages Environments: The Role of Organizational of the process of unlearning. It is important to Perceptions,” Journal of Vocational note that a company’s failure to reach the Behaviour 58(3), 414–435. expected performances leads to changes in Allen, T. D., and J. E. Russell (1999). “Parental knowledge structures (e.g., adoption of a WLB Leave of Absence: Some Not So Family culture) which may stimulate the process of Friendly Implications,” Journal of Applied unlearning (Akgün et al. 2007). Thus, the Social Psychology 29(1), 166–191. process of unlearning can be both a result Anderson, J. C., and D. W. Gerbing (1988). and a trigger of a WLB culture. Additional “Structural Equation Modelling in Practice: A research is, however, necessary to delineate the Review and Recommended Two-Step extent to which particular WLB cultures are Approach,” Psychological Bulletin 103(3), drivers rather than outcomes of the process of 411–423. unlearning. Andrews, F. M. (1975). “Social and Psychologi- The limitations of this research could be cal Factors that Influence the Creative addressed in future studies. Future research Process,” in Perspectives in Creativity. Eds. I. sampling multiple holders of knowledge within A. Taylor and W. Getzels. Chicago: Aldine. a company (e.g., employees) will be helpful to Andrews, J. and D. C. Smith (1996). “In Search test for interrater reliability and to improve the of Marketing Imagination: Factors Affecting internal validity of these constructs. As the Creativity of Marketing Programs for Santangelo and Pini (2011) noted, no previous Mature Products,” Journal of Marketing studies have explicitly accounted for innova- Research 33, 174–187. tion heterogeneity in relation to the adoption of Argyris, C., and D. A. Schön (1978). Organiza- new WLB practices at the shop floor level. tional Learning: A Theory of Action Perspec- Thus, it is still an open question whether and to tive. San Francisco: Jossey-Bass. what extent the findings at the managerial level Armstrong, J. S., and T. S. Overton (1977). apply to the shop floor level. Future research “Estimating Nonresponse Bias in mail might profitably sample multiple holders of Surveys,” Journal of Marketing Research knowledge within a company and will be 14(3), 396–402. helpful in testing for interrater reliability and Azmi, F. T. (2008). “Mapping the Learn- improving the internal validity of other man- Unlearn-Relearn Model: Imperatives for Stra- agement studies. Another line of research could tegic Management,” European Business be to develop a longitudinal study that exam- Review 20(3), 240–259. ines the changes in the variables of the study Bagozzi, R. P., and Y. Yi (1988). “On the Evalu- over time. On the other hand, we encourage ation of Structural Equation Models,” conducting research in other sectors of activity Journal of the Academy of Marketing Science and types of companies (large organizations) 16(1), 74–94. and examining the effect of the existence of a Beauregard, T. A., and L. C. Henry (2009). UC and a WLB culture on the implementation “Making the Link between Work-Life of other organizational policies or practices. Balance Practices and Organizational Perfor- Therefore, it would also be interesting to mance,” Human Resource Management extend the survey to other sectors as national Review 19(1), 9–22. cultural issues might influence the way organi- Becker, K. (2010). “Facilitating Unlearning zations unlearn. Finally, we suggest future during Implementation of New Technology,” researchers to investigate the factors that dis- Journal of Organizational Change Manage- suade managers from adopting a UC in their ment 23(3), 251–268. companies in order to develop recommenda- Behson, S. J. (2005). “The Relative Contribution tions for improvements in the future. of Formal and Informal Organizational

14386 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Work-Family Support,” Journal of Voca- Chirico, F. (2008). “Knowledge Accumulation tional Behaviour 66(3), 57–70. in Family Firms. Evidence from Four Case Bentler, P. (1988). “Practical Issues in Structural Studies,” International Small Business Modeling,” in Common Problems/Proper Journal 26(4), 433–462. Solutions: Avoiding Error in Quantitative Chirico, F., and C. Salvato (2008). “Knowledge Research. Ed. J. S. Long. Newbury Park, CA: Integration and Dynamic Organizational Sage Publications, 161–192. Adaptation in Family Firms,” Family Busi- Bettis, R. A., and C. K. Prahalad (1995). ness Review 9(2), 199–208. “The Dominant Logic: Retrospective and Churchill, G. A. (1979). “A paradigm for Devel- Extension,” Strategic Management Journal oping Better Measures for Marketing Con- 16(1), 5–14. structs,” Journal of Marketing Research Bontis, N., W. Chua, and S. Richardson (2000). 16(1), 64–73. “Intellectual Capital and the Nature of Busi- Confemetal (2013), “Confederación española ness in Malaysia,” Journal of Intellectual de asociaciones empresariales del metal,” Capital 1(1), 85–100. http://www.confemetal.es (latest access date Bowonder, B., A. Dambal, S. Kumar, and A. 30/03/2013). Shirodkar (2010). “Innovation Strategies for Crawford, J. R. (2007), “sbdiff.exe [Computer Creating Competitive Advantage,” Research- software],” from http://www.abdn.ac Technology Management 53(3), 19–32. .uk/psy086/dept/psychom.htm (Retrieved Burke, R. J. (2004). “It’s not How Hard You January 29, 2010). Work but How You Work Hard: Evaluating Day, G. S., and P. Nedungadi (1994). “Manage- Workaholism Components,” International rial Representations of Competitive Advan- Journal of Stress Management 6(4), 225–239. tage,” Journal of Marketing Vo. 58(2), Burke, R. J., Z. Burgess, and F. Fallon (2006). 31–44. “Organizational Practices Supporting De Holan, P. M., and N. Phillips (2004). “The Women and Their Satisfaction and Well- Remembrance of Things Past? The Dynamics Being,” Women in Management Review of Organizational Forgetting,” Management 21(5), 416–425. Science 50(11), 1603–1613. Cappelli, P. (2000). “Making the Most of De Kok, J., and L. M. Uhlaner (2001). “Organi- On-Line Recruiting,” Harvard Business zation Context and Human Resource Man- Review 78(1), 103–111. agement in Small Firm,” Small Business Carlson, D. S., N. Upton, and S. Seaman (2006). Economics 17(4), 273–291. “The Impact of Human Resource Practices Delaney, J. T., and M. A. Huselid (1996). “The and Compensation Design on Performance: Impact of Human Resource Management An Analysis of Family-Owned SMEs,” Practices on Perceptions of Organizational Journal of Small Business Management Performance,” Academy of Management 44(4), 531–543. Journal 39(4), 949–969. Carm (2009), “Registro de Establecimientos Dex, S., and F. Scheibl (2001). “Flexible and Industriales, Sector del Metal. Conserjería de Family-Friendly Working Arrangements in Universidades, Empresa e Investigación,” UK- Based SMEs: Business Cases,” British http://www.carm.es/ (latest access date11/ Journal of Industrial Relations 39(3), 411– 25/2009). 431. Cegarra, J. G., S. Eldridge, and A. Martinez Diamantopoulos, A., and H. Winklhofer (2001). (2010). “Managing Environmental Knowl- “Index Construction with Formative Indica- edge Through Unlearning in Spanish Hospi- tors: An Alternative to Scale Development,” tality Companies,” Journal of Environmental Journal of Marketing Research 38(2), 269– Psychology 30(1), 249–257. 277. Cepeda, G., J. G. Cegarra, and D. Jimenez Eby, L. T., W. J. Casper, A. Lockwood, C. (2012). “The Effect of Absorptive Capacity Bordeaux, and A. Brinley (2005). “Work and on Innovativeness: Context and Information Family Research in IO/OB: Content Analysis Systems Capability as Catalysts,” British and Review of the Literature (1980–2002),” Journal of Management 23, 110–129. Journal of Vocational Behavior 66(1), 124– Chell, E. (1993). The Psychology of Behaviour 197. in Organizations. London: The Macmillan European Commission (2003). “SMEs in Europe Press Ltd. 2003,” No 7, http://www.europa.eu.int/

CEGARRA-NAVARRO, SSÁNCHEZ-VIDAL,ANCHEZ-VIDAL, AND AND CEGARRA-LEIVA CEGARRA-LEIVA 387 15 comm/enterprise/enterprise_policy/ and Simultaneous Equation Models,” analysis/doc/smes_observatory_2003 Structural Equation Modeling 3(3), 266– _report7_en.pdf (accessed October 29, 289. 2012). Johnstone, S., A. Dainty, and A. Wilkinson Felstead, A., N. Jewson, A. Phizacklea, and S. (2008). “In Search of Product Service: Evi- Walters (2002). “Opportunities to Work at dence from Aerospace, Construction, and Home in the Context of Work-Life Balance,” Engineering,” The Services Industries Human Resource Management Journal Journal 28(6), 861–875. 12(1), 54–76. Kimberley, N., and C. E. Hartel (2008). Fernandez, F., and A. Sune (2009). “Organiza- “Employee/Customer Interface in a Service tional Forgetting and Its Causes: An Empiri- Crisis: Impact of Senior Management Attri- cal Research,” Journal of Organizational butes and Practices on Customer Evalua- Change Management 22(6), 620–634. tion,” Journal of Management and Fleetwood, S. (2007). “Re-Thinking Work Life Organization 14(2), 207–218. Balance: Editor’s Introduction,” The Interna- Kirby, E. L., and K. J. Krone (2002). “The Policy tional Journal of Human Resource Manage- Exists, But You Can’t Use It”: Negotiating ment 18(3), 351–359. Tensions in Work-Family Policy,” Journal of Fornell, C., and D. F. Larcker (1981). “Evaluat- Applied Communication Research 30(1), ing Structural Equation Models with Unob- 50–77. servable Variables and Measurement Error,” Kofodimos, J. (1995). Balancing Act. How Man- Journal of Marketing Research 18(1), 39– agers Can Integrate Successful Careers and 50. Fulfilling Personal Lives, 1st ed. San Fran- Frye, N. K., and J. A. Breaugh (2004). “Family- cisco: Jossey-Bass Publishers. Friendly Policies, Supervisor Support, Work- Lavoie, A. (2004). “Work-Life Balance and Family Conflict, Family-Work Conflict, and SMEs: Avoiding the “One-Size-Fits-All” Satisfaction: A Test of a Conceptual Model,” Trap,” CFIB Research July, 1–13. Journal of Business and Psychology 19(2), Lei, D., J. W. Slocum, and R. A. Pitts (1999). 197–220. “Designing Organizations for Competitive Hannan, M. T., M. D. Burton, and J. N. Baron Advantage: The Power of Unlearning and (1996). “Inertia and Change in the Early Learning,” Organizational Dynamics 27(3), Years: Employment Relations in Young, 24–38. High Technology Firms,” Industrial and Leonard-Barton, D. (1992). “Core Capabilities Corporate Change 5(2), 503–536. and Core Rigidities: A Paradox in Managing Harrington, B., and J. Ladge (2009). “Present New Product Development,” Strategic Man- Dynamics and Future Directions for Organi- agement Journal 13(Summer Special Issue), zations,” Organizational Dynamics 38(2), 111–125. 148–157. Levinthal, D., and J. G. March (1993). “The Huber, G. P. (1991). “Organizational Learning Myopia of Learning,” Strategic Management the Contributing Processes and the Litera- Journal 14(8), 95–112. tures,” Organization Science 2(1), 88– Levitt, T. (1972). “A Product-Line Approach to 115. Service,” Harvard Business Review 50(5), Hughes, J., and N. Bozionelos (2007). “Work 41–52. life Balance as Source of Job Dissatisfaction McDougall, P. P., J. G. Covin, R. B. Robinson, and Withdrawal Attitudes. An Exploratory and L. Herron (1994). “The Effects of Indus- Study on the Views of Male Workers,” Per- try Growth and Strategic Breadth on New sonnel Review 36(1), 145–154. Venture Performance and Strategy Content,” INE (2007a), “Encuesta Industrial de Empresas,” Strategic Management Journal 15(7), 537– http://www.ine.es/ (accessed November 25, 554. 2009). Macinnes, J. (2005). “Diez Mitos Sobre la ——— (2007b), “Directorio Central de Empresas Conciliación de la vida Laboral y Familiar,” DIRCE,” http://www.ine.es/ (accessed Cuadernos de Relaciones Laborales 23(1), November 25, 2009). 35–71. Jedidi, K., V. Ramaswamy, W. S. DeSarbo, Malik, S. D., and D. O. Wilson (1995). “Factors and M. Wedel (1996). “On Estimating Influencing Engineers’ Perceptions of Orga- Finite Mixtures of Multivariate Regression nizational Support for Innovation,” Journal

16388 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT of Engineering and Technology Manage- Prahalad, C. K., and M. S. Krishnan (2008). The ment 12(3), 201–218. New Age of Innovation. New York: McGraw Marcati, A., G. Guido, and A. M. Peluso Hill. (2008). “The Role of SME Entrepreneurs’ Preacher, K. J., and G. J. Leonardelli (2003), Innovativeness and Personality in the Adop- “Calculation of the Sobel Test,” Available at: tion of Innovations,” Research Policy 37(9), www.psych.ku.edu/preacher/sobel/ 1579–1590. sobel.htm (accessed April 15, 2010). Martínez-Fernández, C. (2010). “Knowledge Ramaswamy, V., and F. Gouillart (2010). Intensive Service Activities in the Success of “Building the Co-Creative Enterprise,” the Australian Mining Industry,” The Service Harvard Business Review 88(10), 100–109. Industries Journal 30(1), 55–70. Reichheld, F. (1996). The Loyalty Effect: The Martin, P., and D. Chapman (2006). “An Explo- Hidden Force Behind Growth, Profits, and ration of Factors that Contribute to the Lasting Value, Vol. 323. Boston: Harvard Reluctance of SME Owner-Managers to Business School Press, 24–95. Employ First Destination Marketing Gradu- Roth, K., and D. A. Ricks (1994). “Goal Configu- ates,” Marketing Intelligence and Planning ration in a Global Industry Context,” Strategic 24(2), 158–173. Management Journal 15(2), 103–120. Nambisan, S. (2009). “Virtual Customer Santangelo, G. D., and P. Pini (2011). “New Environments: IT-Enabled Customer HRM Practices And Exploitative Innovation: Coinnovation and Value Co-Creation,” in A Shopfloor Level Analysis,” Industry and Nambisan, S., (Ed.), “Information Technol- Innovation 18(6), 611–630. ogy and Product Development,” Annals of Satorra, A., and P. M. Bentler (1988). Scaling Information Systems 5(2), 109–127. Corrections for Chi-Square Statistics in Cova- Nambisan, S., and R. Baron (2007). “Interactions riance Structure Analysis. ASA 1988 Pro- in Virtual Customer Environments: Implica- ceedings of the Business and Economic tions for Products Support and Virtual Cus- Statistics Section. Alexandria, VA: American tomer Relationship Management,” Journal of Statistical Association, 308–313. Interactive Marketing 21(2), 42–62. Scandura, T. A., and M. J. Lankau (1997). “Rela- Nambisan, S., and P. Nambisan (2008). “How to tionships of Gender, Family Responsibility Profit from a Better ‘Virtual Customer Envi- and Flexible Work Hours to Organizational ronment’,” MIT Sloan Management Review Commitment and Job Satisfaction,” Journal 49(3), 53–61. of Organizational Behaviour 18(4), 377– Nelson, D. L., J. C. Quick, M. A. Hitt, and D. 391. Moesel (1990). “Politics, Lack of Career Seiling, J. G. (2008). “The Role of the Customer Progress, and Work/Home Conflict: Stress Advocate: Contextual and Task Performance and Strain For Working Women,” Sex Roles as Advocacy Participation,” Journal of Man- 23(3/4), 169–185. agement and Organization 14(2), 127–140. Nonaka, I., and H. Takeuchi (1995). The Shaw, R. B., and D. N. Perkins (1991). “Teach- Knowledge-Creating Company. New York: ing Organizations to Learn,” Organization Oxford University Press. Development Journal 9(4), 1–12. Nystrom, P. C., and W. H. Starbuck (1984). Shipton, H., M. West, J. Dawson, K. Birdi, and “To Avoid Organizational Crises, Unlearn,” M. Patterson (2006). “HRM as a Predictor of Organizational Dynamics 12(4), 53– Innovation,” Human Resource Management 65. Journal 16(1), 3–27. O’Neill, J. W., M. M. Harrison, J. Cleveland, D. Sinkula, J. M., W. E. Baker, and T. Noordewier Almeida, R. Stawski, and A. C. Crouter (1997). “A Framework for Market-Based (2009). “Work-Family Climate, Organiza- Organizational Learning: Linking Values, tional Commitment, and Turnover: Multi- Knowledge and Behaviour,” Journal of the level Contagion Effects of Leaders,” Journal Academy of Marketing Science 25(4), 305– of Vocational Behaviour 74(1), 19–28. 318. Poelmans, S. A., N. Chinchilla, and P. Cardona Sobel, M. E. (1982). “Asymptotic Intervals for (2003). “The Adoption of Family Friendly Indirect Effects in Structural Equations HRM Policies: Competing for Scarce Models,” in Sociological Methodology. Ed. S. Resources in the Labour Market,” Interna- Leinhart. San Francisco, CA: Jossey-Bass, tional Journal of Manpower 24(2), 128–147. 290–312.

CEGARRA-NAVARRO, SSÁNCHEZ-VIDAL,ANCHEZ-VIDAL, AND AND CEGARRA-LEIVA CEGARRA-LEIVA 389 17 Thompson, C. A., L. L. Beauvais, and K. Lyness Yeandle, S., A. Wigfield, R. Crompton, and J. (1999). “When Work-Family Benefits Are Not Dennett (2002). Employed Carers and Enough: The Influence of Work-Family Family Friendly Employment Policies. Culture on Benefit Utilization, Organiza- Bristol: Policy Press. tional Attachment, and Work-Family Con- Zeithaml, A., A. Parasuraman, and L. L. Berry flict,” Journal of Vocational Behavior 54(3), (1990). Delivering Quality Service: Balanc- 392–415. ing Customer Perceptions and Expectations. Thompson, J. L. (1998). “Competency and Mea- New York: The Free Press. sured Performance Outcomes,” Journal of Zhu, W., I. K. Chew, and W. D. Spangler (2005). Workplace Learning 10(5), 219–231. “CEO Transformational Leadership and orga- Yang, Z., and R. T. Peterson (2004). “Customer nizational Outcomes: The Mediating Role of Perceived Value, Satisfaction, and Loyalty: Human-Capital-Enhancing Human Resource Of Switching Costs,” Psychology and Market- Management,” The Leadership Quarterly ing 21(10), 799–822. 16(1), 39–52.

Appendix: Questionnaire Items

The consolidation of emergent understandings: With respect to your organization, indicate the degree to which you agree or disagree (1 = strongly disagree and 7 = strongly agree): CEU_1: Managers seem to be open to new ideas and new ways of doing things. CEU_2: Managers adopt suggestions from staff in the form of new routines and processes. CEU_3: Management has tried to initiate projects and introduce innovations. CEU_4: Managers are prone to collaborate with members of the organization and to solve problems together. Source: Cegarra, Eldridge, and Martinez (2010) The examination of lens ﬁtting: With respect to your current position, indicate the degree to which you agree or disagree (1 = strongly disagree and 7 = strongly agree): ELF_1: Employees try to reﬂect and learn from their own mistakes. ELF_2: Employees are able to see mistakes from their colleagues. ELF_3: Employees are able to identify problems (new ways of doing things) easily. ELF_4: Employees are able to listen to my customers (e.g., complaints, suggestions). Source: Cegarra, Eldridge, and Martinez (2010) The framework for changing individual habits: With respect to your personal skills, indicate the degree to which you agree or disagree (1 = strongly disagree and 7 = strongly agree): CIH_1: New situations have helped individuals identify their own mistakes. CIH_2: New situations have helped individuals recognize unwanted attitudes. CIH_3: Individuals recognize forms of reasoning or arriving to solutions as inadequate. CIH_4: New situations have helped individuals identify improper behavior. Source: Cegarra, Eldridge, and Martinez (2010)

WLB supporting culture: With respect to the following questions, indicate how the management values the following: (1 = very negatively, 7 = very positively):

WBC_1: Talking about personal life at work WBC_2: Refusing promotions requiring geographical moves WBC_3: Starting a family, becoming pregnant, or adopting a child WBC_4: Going home during the workday to attend personal responsibilities such as a sick child WBC_5: Setting limits on hours spent at work Source: Kofodimos (1995)

18390 JOURNALJOURNAL OF OF SMALL BUSINESS MANAGEMENT MANAGEMENT Appendix: Continued Innovation-related outcomes: With respect to their competitors, indicate the degree in which your company reached the following objectives in the last three years (1 = did not reach at all and 7 = reached easily): IRO_1: Quality of products, services, and programs IRO_2: Development of new products and services IRO_3: Satisfaction of customers or clients Source: Delaney and Huselid (1996)

CEGARRA-NAVARRO, SSÁNCHEZ-VIDAL,ANCHEZ-VIDAL, AND AND CEGARRA-LEIVA CEGARRA-LEIVA 391 19 THE INTERNATIONAL COUNCIL FOR SMALL BUSINESS

WHAT IS ICSB? Founded in 1955, ICSB is the fi rst international, non-profi t, membership driven organization to promote the growth and development of small, entrepreneurial, and family businesses worldwide. ICSB brings together educators, researchers, policy makers and practitioners to share knowledge and experience in their respective fi elds with a focus on small business development around the world. The Council was founded on the belief that enlightened small business management is necessary for successful profi table small business, that successful small businesses are essential to each nation’s economy, and that entrepreneurship needs to be fostered to stimulate a dynamic and growing economic system.

WHO SHOULD BELONG TO ICSB? Membership in the International Council for Small Business is open to all persons interested in improving small, entrepreneurial, and family business knowledge and skills around the world. Council membership is represented by individuals, organizations, and agencies from: • RESEARCH: Researchers in universities, colleges, research agencies and research companies. • EDUCATION: Secondary, college, graduate, distributive and adult education • PROFESSIONS: Management consultants, accountants, lawyers • TRADE ASSOCIATIONS: Directors, membership • GOVERNMENT: Ofﬁ cials in international, federal, state, regional, local, or special agencies that are concerned with small business. • ECONOMIC DEVELOPMENT CENTERS: Minority, community, and other concerned ofﬁ cers and employees • BUSINESS: Independent entrepreneurs

MEMBERSHIP CLASSES There are four classes of membership available in ICSB: individual, organizational, contributing, and student. Many colleges, universities, and businesses select the organizational category. Student member- ships are also welcome at a special annual rate. For speciﬁ c information regarding fees, see the application for membership in this edition.

ICSB AND ITS AFFILIATES ICSB is organized into regionally oriented affi liates which provide activities such as annual regional conferences, workshops, newsletters, website, etc. See www.icsb.org for a list of ICSB affi liates. The typical ICSB member is a member of an affi liate and thus automatically an ICSB member, allowing access to all ICSB and affi liate benefi ts: member registration for annual world conference, free copy of JSBM, newsletter, and access to members-only ICSB website *Serving the World of Entrepreneurship and Small Business for over 50 Years*

www.icsb.org

JOURNAL OF SMALL BUSINESS MANAGEMENT

Join the JSBM Editorial Review Board

The Journal of Small Business Management JSBM is looNing for TualiÀeG reYieZers ,f \ou are interesteG in Moining the JSBM (Gitorial 5eYieZ BoarG as a reYieZer FontaFt us toGa\ SenG \our Furrent &urriFulum 9itae &9 to email MsEm#gZueGu (fax) +1-202-994-4930 (Sost) 2201 * St 1: Ste 31 :ashington '& 2002

5eYieZers are seleFteG EaseG on SuEliFation reForG exSerienFe anG oYerall FontriEu- tion to the researFh Fommunit\ 5eYieZers Fan exSeFt to reYieZ an aYerage of ÀYe () manusFriSts Ser \ear

Mission Statement. $ll manusFriSts shoulG aGGress researFh issues in a rigorous Za\ using TualitatiYe TuantitatiYe or a FomEination of these teFhniTues hoZeYer JSBM enFourages the inForSoration of SragmatiF aGYiFe for SraFtitioners EaseG on the researFh results