REGIONAL DIMENSIONS OF INNOVATIVE ACTIVITY IN OUTER WESTERN SYDNEY
Samantha A. Sharpe
A thesis submitted to the College of Health and Science In fulfilment of the requirements for the Degree of Doctor of Philosophy August 2007
Urban Research Centre University of Western Sydney
Samantha A Sharpe © 2007
Statement of Authentication
The work presented in this thesis is, to the best of my knowledge and belief, original except as acknowledged in the text. I hereby declare that I have not submitted this material, either in full or in part, for a degree at this or any other institution.
Samantha A Sharpe
Abstract
The aim of this research is to understand the socio-economic development of a metropolitan region in Sydney through an analysis of regional innovative activity. South West Sydney, a major growth region within Sydney, includes the Local Government Areas (LGAs) of Liverpool, Campbelltown, Camden and Wollondilly. This region has absorbed 25% of Sydney’s population growth in the period from 1991-2001. Although South West Sydney has experienced rapid population growth, this has not been matched by associated employment growth. In some sectors such as business services employment growth has been minimal in the previous decade, this is particularly the case in Liverpool, the regional centre of South West Sydney. Population growth is estimated to continue at the current rate (in excess of 5% per annum) for at least the next fifteen years. In this environment, local government authorities in the region are seeking ways in which to develop the regional economy of South West Sydney and increase the amount of sustainable employment commensurably with current population and labour force increases.
The role of innovative activity has a central place in economic development. This thesis uses a ‘systems of innovation’ (SI) approach to examine innovative activity in the South West Sydney region. SI understands innovation as a socially embedded process of transforming ideas and knowledge into novel products, processes and services through the processes of learning and searching. The approach recognises that innovative activity is determined by various actors (firms and institutions) and the interactivity between these actors and the cumulative base of knowledge in which they operate. The Regional Innovations Systems (RIS) framework develops from an acknowledgement that innovation is primarily a geographically bounded phenomenon. The RIS approach sees that specific local resources are important in determining and encouraging the innovative activities carried out by local firms and hence, the competitiveness of these areas.
The RIS literature provides two fields of understanding of what constitutes a regional innovation system. The first takes the global examples of highly innovative regions such as Silicon Valley and Route 128 in the United States of America (Saxenian 1994), South West England (Cooke and Morgan 1998), Baden Wurttemberg in Germany (Cooke 2001; Braczyk, Cooke et al. 2004), Northern Italy (Piore and Sabel 1984) and in Australia, the North Ryde corridor (Searle and Pritchard 2005). These regions represent ‘ideal’ or ‘star’ RIS, with highly specialised and networked clusters of firms, many
Abstract iii
forms of supporting regional infrastructure, and high levels of interactivity. The second and emerging field understands RIS to be in existence in all regions and individual RIS are identified on a scale from weak to strong (Wiig and Wood 1995; Cooke and Morgan 1998; Cooke 2001). This second stream includes the analysis of regions seeking to encourage innovative activity by using the RIS approach to examine their local resources and connectedness. It seeks to determine how not only local resources but also their connectedness could be enhanced to increase firm competitiveness.
The innovation systems represented in the ‘ideal’ regions are largely a world away from what is available and what is necessary in the encouragement of RIS in most other regions. However, the conceptual framework for examining and interpreting RIS is derived from the analysis of these ‘ideal’ regions. This framework does not provide for measurement and effective interpretation of a range of activities that may be present in less exceptional regions.
This research contributes to this endeavour by providing a method that allows for interpretation of a wider range of innovation activities through the analysis of knowledge intensive services activities (KISA). The focus on knowledge gathering, particularly through the KISA analysis, provides an examination of the relationship between innovation, learning and knowledge, much more so than more traditional measures of innovative activity e.g. patents and research and development (R&D) expenditure. KISA analysis is an emerging field of innovation research. KISA are closely linked to firm innovative activity (OECD 2006) and through an analysis of regional KISA usage, an understanding of innovation and knowledge activities within the region can be constructed. This analysis applies equally across various regions and provides an opportunity to guide regional economic development policy intervention at the local government level in South West Sydney.
Abstract iv
Acknowledgements
There are many people I would like to thank for their help and encouragement over the past three and a half years. Although my name is on the title page, the journey for me to get here has not been a solo one.
First, I would like to thank my principal supervisor, Dr. Cristina Martinez- Fernandez for all of her support, guidance and encouragement. I have learnt a great deal from working with her and I know I would not have got to this point without her. I would also like to thank my co-supervisor, Prof Russel Cooper. His advice and comments have been invaluable to my progress particularly in the final few months. I would like to acknowledge the assistance of Prof. Tong Wu in providing advice and guidance in the final preparation of this thesis.
As my research is a linkage project with Liverpool City Council I have been fortunate to work with two industry partners, Graham Larcombe and Phil Tolhurst. I would like to thank Graham for encouraging me to join the project in the first place and for his ongoing support since. I would also like to express my appreciation for Phil Tolhurst, who has given the project his every assistance despite his busy schedule.
I would like to thank my colleagues at Liverpool City Council, their continued dedication to the South West Sydney community, sometimes in the face a great odds, is inspiring. It has been a pleasure working with them over the past six years. I would like to particularly mention Colleen, Bozena, Mira, Martin, Carl, Pierre, Debbie, Grace, Aidan and the rest of the Strategic Planning Team.
At the University of Western Sydney I have been fortunate to work alongside a number of talented and exciting researchers, Merete, Marc, Tavis, Monty, Navin and Tamara and from the University of Sydney, Santosh, Asif, Richard and
Acknowledgements v Tony. These people have contributed to my intellectual development and passion for research as well as becoming firm friends.
Finally, I would like to thank my family; my parents Sharon and Russell, my grandparents Aub and Gladys, my siblings Dale, Miles and Cassandra and my husband Christian. The two most influential people in my life are my mother and my husband and I thank them especially. My mother for always believing that there was nothing I could not achieve if I wanted to, and for also teaching me that if I did do something, to make sure I did it well. I thank my husband, Christian, for supporting me in so many ways; through all the late nights, the stressful computer episodes, the house covered in papers and reading and learning much more about regional innovation than I am sure he ever intended. I thank them both and it is to them that this thesis is dedicated.
Acknowledgements vi
for Sharon and Christian
Acknowledgements vii Table of Contents
Abstract ...... iii Acknowledgements...... v Table of Contents...... viii Index and Tables, Charts and Graphs...... x Glossary of abbreviations...... xiii
Chapter 1 - Introduction ...... 1 1.1 The research context - Innovation and regional economic development...... 2 1.2 The research problem ...... 8 1.3 Research design...... 12 1.4 Scope of the study ...... 14 1.5 Research significance ...... 20 1.6 Structure of the thesis ...... 21
Chapter 2 - Knowledge & innovation activities in regions ...... 23 2.1 Innovation and knowledge activities ...... 24 2.1.1 Extent of innovative activities ...... 27 2.1.2 Knowledge and learning ...... 27 2.1.3 Knowledge intensive activities...... 32 2.1.4 Defining regions ...... 34 2.1.5 Innovation – the systems approach...... 35 2.2 Regional Innovation Systems...... 37 2.2.1 Iterations of the concept...... 40 2.2.2 Identifying actors, interactions and their attributes...... 42 2.5 Summary...... 45
Chapter 3 - Research design and methods...... 47 3.1. Selection of Research Design...... 48 3.1.1 Limitations of research design...... 51 3.2 Selection of Research cases ...... 54 3.3 Field Work ...... 57 3.3.1 Statistical audit ...... 58 3.3.2 Questionnaire design...... 58 3.3.3 Questionnaire field procedures...... 62 3.3.4 Interview field procedures...... 67 3.3.5 Rationale for selection of firms for in-depth interviews ...... 69 3.4. Data Analysis ...... 70 3.4.1 Statistical Analysis...... 71 3.4.2 Qualitative Analysis...... 72 3.5. Summary...... 72
Chapter 4 – Economic and Industrial Audit of Outer Western Sydney ...... 74 4.1 Suburban Sydney...... 76 4.2 Growth in Outer Western Sydney ...... 79 4.2.1 Population growth...... 79 4.2.2. Employment growth ...... 80 4.2.3 Population growth implications...... 81 4.3 Labour force...... 83 4.3.1 Employment ...... 83 4.3.2 Employment by industry...... 85
Table of Contents viii 4.3.3 Occupation ...... 86 4.3.4 Income ...... 87 4.3.5 Labour force implications ...... 89 4.4 Regional industry base...... 89 4.4.1 Industry concentration and specialisations...... 94 4.4.2 Firm demographics ...... 100 4.5 Regional knowledge base ...... 104 4.5.1 Knowledge based occupational analysis...... 107 4.6 Summary...... 109
Chapter 5 – Industrial Innovation in Outer Western Sydney...... 112 5.1 Innovation counts ...... 114 5.1.1 Novelty of innovation ...... 122 5.1.2. Innovation implications...... 123 5.2 Knowledge sources...... 124 5.2.1 Knowledge sources and region, firm size and industry ...... 126 5.2.2Factor analysis of knowledge sources...... 131 5.2.3 Knowledge source implications...... 135 5.3 Knowledge Intensive Service Activities...... 136 5.3.1 Region, firm size and industry size usage of KISA...... 137 5.3.2 KISA factor analysis...... 142 5.4.3 KISA implications ...... 145 5.4 KISA Location ...... 146 5.4.1 KISA location implications...... 152 5.5 Summary...... 153
Chapter 6 – Regional orientation of firms...... 161 6.1. Locations of key knowledge organisations...... 162 6.1.1 Knowledge organisation implications...... 169 6.2 Factor analysis of knowledge organisation networks...... 172 6.3 Supply chain geography...... 176 6.5 Summary...... 182
Chapter 7 – Firm strategies for innovation: the process of mixing and matching ...... 189 7.1 Cluster analysis of innovative activities in firms ...... 190 7.1.1 Cluster membership cross-tabulations ...... 195 7.1.2 Cluster implications ...... 200 7.2 Identifying firm strategies...... 201 7.2.1 Knowledge sources and KISA...... 201 7.2.2 Firms’ knowledge networks ...... 213 7.3 Regional experience and its effect...... 217 7.3.1 It is where the boss lives ...... 217 7.4 Summary...... 220
Chapter 8 - Conclusions and recommendations ………………………………………225 8.1 Research Conclusions...... 225 8.1.1 Regional innovative activity in Outer Western Sydney...... 226 8.1.2 Policy recommendations for South West Sydney...... 230 8.1.3 RIS: an integrated approach ...... 232 8.1.4 KISA analysis...... 234 8.2 Limitations of the research...... 235 8.3 Suggestions for future research...... 236 References ...... 237 Appendices ...... 251
Table of Contents ix Appendix 1 Survey Information sheet for companies ...... 252 Appendix 2 Survey Questionnaire ...... 255 Appendix 3 Firm Interview Protocol...... 267 Appendix 4 Firm Interview Questions ...... 273 Appendix 5 Criteria Table for interview selection...... 275 Appendix 6 Social profile of regions...... 276 Appendix 7 Knowledge Sources Factor analysis (PCA) ...... 284 Appendix 8 KISA Factor analysis (PCA)...... 290 Appendix 9 Network Factor Analysis (PCA) ...... 295 Appendix 10 Hierarchical Cluster Analysis...... 299 Appendix 11 Cluster membership correlations...... 300 Appendix 12 Cluster membership cross-tabulation...... 301 Index and Tables, Charts and Graphs
Tables
Table 2.1 Defining elements of RIS...... 42 Table 2.2 Key elements of integrated RIS investigation...... 45
Table 3.1 Research questions and corresponding data collection methods...... 57 Table 3.2 Survey sample by list...... 63 Table 3.3 Survey sample by region and broad industry grouping ...... 65 Table 3.4 Research questions and corresponding data collection methods...... 71
Table 4.1 Labour force composition in the regions 2001...... 83 Table 4.2 Labour force composition in regions in 1991, and comparisons with 2001 ...84 Table 4.3 Labour force industrial employment in Outer Western Sydney 2001 ...... 86 Table 4.4 Labour force occupation in Outer Western Sydney 2001...... 87 Table 4.5 Income levels in Outer Western Sydney...... 88 Table 4.6 Regional industrial base (from employment) 2001 ...... 90 Table 4.7 Industrial change (based on employment) in Outer Western Sydney ...... 92 Table 4.8 Detailed breakdown of manufacturing employment in the regions, 2001 ...... 95 Table 4.9 Location Quotients for manufacturing industries in the regions, 2001...... 98 Table 4.10 Detailed breakdowns of Financial & Business Services Employment ...... 98 Table 4.11 Location Quotients for Finance & Business services in the regions, 2001...99 Table 4.12 Industrial Composition of Firms in Regions, 2001...... 102 Table 4.13 Estimated Firm turnover (Ranges) % in regions, 2001...... 103 Table 4.14 New ABN Registrations 2001-2004 ...... 103 Table 4.15 RIS analysis elements ...... 111
Table 5.2 Rotated Component Matrix for Knowledge Sources Component Analysis.. 134 Table 5.4 KISA Factor Analysis...... 144
Table 6.1 Knowledge networks: Factor analysis (Rotated matrix)...... 174 Table 6.2 RIS analysis elements ...... 187
Table 7.1 Factors and factors characteristics of three clusters (of variables) ...... 192 Table 7.2 Mix of internal and external KISA in interview firms...... 203
Table 8.1 RIS analysis elements ...... 227 Table 8.2 South West Sydney RIS analysis...... 231
Table of Contents x Figures and diagrams
Figure 1.1 May of Sydney Metropolitan area by regions...... 15 Figure 1.2 Three regions under investigation ...... 16 Figure 1.3 Central West Sydney in detail...... 17 Figure 1.4 North West Sydney in detail...... 17 Figure 1.5 South West Sydney in detail...... 18
Diagram 3.1 Research methodology ...... 50
Diagram 6.1 Extent of regional knowledge networks...... 172
Diagram 6.2 Regional supply chain locations ...... 182 Diagram 6.3 Orientation and location of key knowledge organisations in the regions. 186
Diagram 7.1 Dendrogram using Ward Method ...... 192
Charts and Graphs
Chart 3.1 Respondent firms by region...... 66 Chart 3.2 Industry of respondent firms by region...... 66 Chart 3.3 Industry of survey sample and survey response firms ...... 67
Graph 4.1 Regional Population projections 2001-2021 ...... 80 Chart 4.1 – Share of Sydney Metropolitan Population and Employment Growth...... 81 Chart 4.2 Knowledge based occupations as a percentage of total workforces ...... 107 Chart 4.3 Types of Knowledge occupations by region ...... 108
Chart 5.1 Innovations by type and by region...... 115 Chart 5.2 Business cycle and firm innovation activity...... 116 Chart 5.3 Innovation by type and by business size ...... 118 Chart 5.4 Innovation by type and by broad industry category...... 119 Chart 5.6 Novelty of innovation by type and region...... 123 Chart 5.7 Knowledge sources in firms by level of importance...... 125 Chart 5.8 Knowledge sources (rated high and medium importance) by region...... 127 Chart 5.9 Knowledge sources (rated high and medium importance) by business size.. 128 Chart 5.10 Knowledge sources (rated high and medium importance) by industry ...... 130 Chart 5.11 KISA usages by type...... 137 Chart 5.12 KISA usages by type by region ...... 139 Chart 5.13 KISA usages by type by firm size...... 140 Chart 5.14 KISA usages by type by industry...... 142 Chart 5.15 Overall location of KISA ...... 147 Chart 5.16 Location of KISA Central West Sydney firms ...... 149 Chart 5.17 Location of KISA, North West Sydney...... 150 Chart 5.18 Location of KISA, South West Sydney...... 151
Chart 6.1 Government organisations included in firm knowledge networks ...... 163 Chart 6.2 Industrial based organisations included in firm knowledge networks ...... 165 Chart 6.3 Educational organisations included in firm knowledge networks ...... 166 Chart 6.4 Regional organisations included in firm knowledge networks...... 169 Chart 6.2 Government organisations included in firm knowledge networks ...... 163 Chart 6.2 Industrial based organisations included in firm knowledge networks ...... 165 Chart 6.3 Educational organisations included in firm knowledge networks ...... 166 Chart 6.4 Regional organisations included in firm knowledge networks...... 169
Index of Tables, Charts and Graphs xi Chart 6.5 Central West Sydney – Customers by geography (level of importance) ...... 177 Chart 6.6 Central West Sydney – Suppliers by geography (level of importance) ...... 177 Chart 6.7 North West Sydney – Customers by geography (level of importance)...... 178 Chart 6.8 North West Sydney – Suppliers by geography (level of importance)...... 179 Chart 6.9 South West Sydney – Customers by geography (level of importance)...... 180 Chart 6.10 South West Sydney – Suppliers by geography (level of importance)...... 181
Chart 7.1 Cluster membership and innovation performance ...... 196 Chart 7.2 Cluster memberships by region ...... 197 Chart 7.3 Industry sector and cluster membership...... 198 Chart 7.4 Firm size and cluster membership ...... 199
Index of Tables, Charts and Graphs xii Glossary of abbreviations
ABL Australian Business Limited AIG Australian Industry Group ARC Australian Research Council BEC Business Enterprise Centres CWS Central West Sydney DOTARS Department of Transport and Regional Services (Federal) DSRD Department of State and Regional Development (State) GWS Greater Western Sydney GWSEDB Greater Western Sydney Economic Development Board KIBS Knowledge Intensive Business Services KISA Knowledge Intensive Service Activities LGA Local Government Area LQ Location Quotients MACROC Macarthur Area Regional Organisation of Councils NIS National Innovation Systems NSW New South Wales (State) NWS North West Sydney OECD Organisation for Economic Co-operation and Development OMWS Office of the Minister of Western Sydney PCA Principal Components Analysis R&D Research and Development REDP Regional Economic Development Policy RIS Regional Innovation Systems SIS Sectoral Innovation Systems SMA Sydney Metropolitan Area SWS South West Sydney TAFE Technical and Further Education Colleges WSROC Western Sydney Regional Organisation of Councils UNSW University of New South Wales UOW University of Wollongong USYD University of Sydney UTS University of Technology Sydney UWS University of Western Sydney
Glossary of Abbreviations xiii
Chapter 1 - Introduction
Encouraging regional industrial development and employment growth is by no means a new pursuit. Many scholars and policy makers have worked towards the same goal with varying levels of success over many decades. A recent focus has been on the industrial innovation capacity of regions, which is the cumulative result of two factors. Firstly, there has been an increasing realisation of the importance of knowledge based activities to economic development and growth. The link between innovation and increased economic activity, especially industrial activity, is well established in the innovation literature (Porter 1990; OECD 1996a; OECD 1999b; Porter 2001; OECD 2001a; OECD 2001b; Acs 2002; Lever 2002). This focus on innovation and knowledge is also by no means new, but the increasing globalisation of world economies in recent decades has meant the necessary “stock of knowledge on which economic activity is based” (Brenner 2007 p.121) has increased exponentially. As a result, the success of economies to create wealth and maintain increasing standards of living is said to now rely on creating, transferring, and commercialising new knowledge (2004; Simmie 2005).
The second, and closely linked factor, is the increasing realisation that many knowledge based activities have a spatial context (Storper 1995; Storper 1997; Breschi 2000; Clark, Feldman et al. 2000; Glaeser 2000; Krugman 2000; Acs,
Chapter 1 - Introduction 1 Anselin et al. 2002; Scott and Storper 2003; Howells 2005; Lorenzen 2005), and that the regional level provides a ‘nexus’ (Storper 1995) for competitive advantage and growth. These advantages are determined by a region’s unique combination of location based assets including institutional composition, resources, and firms, and how this combination of assets shapes the flow of knowledge and learning into, out of, and within the region. Therefore, in every region there are different levels of advantage and industrial innovative capacity.
Spatial effects on economic activities are well known and these spatial impacts have also been translated to knowledge based activities. The arguments of space and location not mattering in the knowledge based economy have well and truly been refuted (Florida 2000; Chapple 2001; Acs 2002). Instead the spatial aspects of innovation and knowledge processes have brought regional inequalities into sharper relief, even beyond historically recognised socio-economic differences.
1.1 The research context - Innovation and regional economic development
Innovation is defined as any new or significantly improved product, service and process with economic value (Lundvall 1992; Edquist 1997; Braczyk, Cooke et al. 2004). In the past, the focus of innovation studies was primarily on innovation outputs such as new products and patents, or innovation inputs defined in the narrowest sense, such as expenditure on research and development (R&D) and number of employed research scientists. These measures correspond with a linear model of innovation that emphasises the process of innovation in a straight line from scientific discovery (usually in R&D laboratories and universities) to commercialisation and market. Subsequent research showed this linear model only applied to a very small amount of innovative activities (Freeman 1988; Lundvall and Johnson 1994) and the experience of innovation in the vast majority of cases appeared to follow a more interactive model, with various sources of innovation, rather than just research and development laboratories and universities. In his landmark study of the success of the Japanese economy, and
Chapter 1 - Introduction 2 the success of Japanese production methods, Freeman (1988) provided evidence that a greater sphere of innovative sources surrounded firms.
This sphere identified clients and customers as the main sources of information and knowledge on product improvements and innovations. The concept of feedback loops was also introduced in this context. This is the perception that innovation occurs when information and knowledge loop backwards from the consumer, to provide feedback and thus ideas for further progress, thereby creating an interactive model. This concept of a broader sphere of participants in innovative activity, and the interactive aspects of knowledge gathering and access, led to the analysis of innovation activity through a much wider lens (OECD 1996a).
The interactive innovation model quickly gained significant theoretical ground, because although innovation was acknowledged as the driver of economic growth, many traditional and established mainstream economic theories did not adequately conceptualise how innovative activity in firms operated. These conceptual difficulties and the continuing demands from policymakers rapidly needing to come to grips with a ‘knowledge based’ economy encouraged the development of alternate models of explanation.
The System of Innovation Approach was one such alternative model of explanation. The approach, taking as its foundation the interactive model (Lundvall 1992; Lundvall, Johnson et al. 2002), examines innovation as a systemic process. The theory suggested that an innovation system is made up of actors, and the interactivity between these actors is the source of innovation. Therefore, the key to understanding innovation is through the analysis of these actors and their interactions, and how learning occurs between them. Learning, also seen as an interactive process, is the ways and means by which firms transform collected information into usable and distinctive knowledge. The activities of innovation and the processes of learning are critically linked (Lundvall 1992; Capello 1999). The analysis of knowledge gathering by firms provides a link between innovation activities and learning processes.
Chapter 1 - Introduction 3
The approach gained prominence quickly1, and scholarship developed in three distinct paths. The first of these paths is at the national level with the National Systems of Innovation (NSI) (Lundvall 1992; Freeman 1995; Gregersen and Johnson 1997; OECD 1997; OECD 1999b; OECD 2001b). NSI focuses on actors and interactions at the national level including: the role of national institutions and peak universities; and the national policy framework for education, science and technology and industrial development. The second path is the analysis of sectoral and technological systems of innovation (Breschi 2000; Carlsson, Jacobsson et al. 2002; Malerba 2005). The focus here is on technological regimes of different industrial sectors and their levels of technological intensity.
The third path, and the one providing the theoretical framework for this research, is the Regional Innovation Systems approach (RIS). This approach acknowledges the central role of regional drivers of innovation processes through the interaction and learning undertaken between regional actors (Asheim and Isaksen 1996; de La Mothe 1998; Cooke 2001; Andersson and Karlsson 2004; Braczyk, Cooke et al. 2004; Asheim and Coenen 2005; Doloreux and Parto 2005; Asheim and Coenen 2006). Each of these three paths shows a preference for particular actors and institutions in their analysis, but all work from the same principles of SI: that a range of actors and their activities is important for innovation, as are the processes of learning and how these are arranged.
RIS brings together the systematic analysis of innovation with the long-standing territorial models of economic development and regional science. The territorial development theories of regional economic geography draw upon a number of tools and factors to explain growth disparities. The RIS approach specifically draws out the spatial dimensions of knowledge activities as a factor to explain growth. This operates from the base assumptions of the innovation systems literature, that innovation is the driver of growth in economic activity and
1 For an overview see Carlsson, B., S. Jacobsson, et al. (2002). "Innovation systems: analytical and methodological issues." Research Policy 31: 233-245.
Chapter 1 - Introduction 4 therefore regional variations of knowledge (the process of innovating) provides the basis for the analytical framework.
The approach notes the spatial dimensions of knowledge processes. It recognises that specific local resources are important in determining and encouraging innovative activities carried out by local firms. However what the approach as yet does not provide is a mechanism through which to examine how these local resources characterise the innovative activities taking place in firms. Knowledge activities are the key link. They encompass the conversion of external and local resources into innovative activities in firms.
Building a critical analysis of regional knowledge activities into the RIS approach provides a stepping stone into further illuminating the relationship between local resources and innovative capacity. This in turn provides valuable insight for understanding the capability of policy, addressing on one side the development of local resources and on the other side the expected effect this may have on regional innovative capacity and capability of firms.
The usage of the knowledge intensive service activity methodology within the RIS approach provides a clear opportunity for this research to contribute first to the understanding of regional innovative and economic activity in three regions of Outer Western Sydney. Secondly it contributes to the wider expansion of the capability of RIS to analyse regions in a more comprehensive and dynamic way. RIS authors have acknowledged that providing means by which to identify the key indicators of innovation capability is necessary for RIS to continue to be a policy relevant approach (Cooke and Memodovic 2003; Doloreux and Parto 2005).
The research informing this regional policy agenda is characterised by two fields of understanding of what constitutes a regional innovation system. The first field encompasses the globally recognised examples of highly innovative regions such as Silicon Valley and Route 128 in the United States of America (Saxenian 1994; Best, Paquin et al. 2004), South West England (Cooke and Morgan 1998), Baden
Chapter 1 - Introduction 5 Wurttemberg in Germany (Cooke 2001; Braczyk, Cooke et al. 2004), and the North Ryde corridor in Australia (Searle and Pritchard 2005). These regions represent ‘ideal’ RIS, with highly specialised and networked clusters of firms, many forms of supporting regional infrastructure, and high levels of interactivity.
The second and emerging field understands RIS to be in existence in all regions, and individual RIS are identified on a scale from weak to strong (Wiig and Wood 1995; Cooke and Morgan 1998; Cooke 2001). This second stream includes the analysis of regions seeking to encourage innovative activity by using the RIS approach to examine their local knowledge resources and connectedness. It also seeks to determine how not only local resources but also their connectedness could be enhanced to increase firm competitiveness. In this sense, RIS is concerned with the identification of how actors interact and how knowledge flows within a region. It is aimed at determining how a region is performing on “a selection of key indicators on various aspects of organisational and infrastructural capacity, competence and capability … with regard to innovative capability” (Doloreux and Parto 2004 p. 17), and how regional economic development policy may assist and encourage these capabilities.
This division in the RIS approach in many ways is due to its own success and ready acceptance in policy circles (De Bruijn & Lagendijk, 2005). The first field or iteration of RIS was the analysis and interpretation of clearly anomalous regions. These regions were exceptional in their capacity for technological development and growth to such an extent that they ‘stood out’ in terms of the levels of economic development that was being witnessed in many other regions. They were analysed and discussed, rightly so, as the new way forward for economic and industrial organisation, and the analysis of these success stories provided much of the conceptual development of the RIS approach.
However, analysis of these regions also showed that their success and associated development patterns were the consequence of specific historical and locational factors. Although these regions had a wide range of actors involved in innovation, and high levels of interactivity and learning that were driving high
Chapter 1 - Introduction 6 levels of innovation, these elements (actors, interactivity, and learning) only provided a specific outcome (regional success) in the context of certain historical and locational factors.
The second and more recent iteration has emerged from many other regions that have seen the growing importance of innovation and knowledge processes to economic development, usually through the harsh experience of industrial re- organisation. These regions do not possess the same historical and locational assets as the star regions but are seeking similar outcomes in terms of economic and employment development.
The speed with which the RIS theoretical and empirical analysis has progressed (the term has only really been in existence for the last 15 or so years), has not allowed an adequate development of the approach for application to this broader range of regions, as noted by de Bruijn & Lagendijk (2005) in their analysis of the application of RIS in Europe:
“Within Europe, there is substantial variety in the way the RIS concept is understood and applied, notably at the regional level. Yet, within the realm of European innovation policy and regional policy, one can identify a dominant line of thinking bearing on the conceptualisation and application of RIS” (de Bruijn & Lagendijk 2005, p.1156).
The confusion is further compounded by the flexible definition of a ‘region’ and, in the case of this research, by the particular tradition of regional economic development policy in the Australian context. The term ‘regional’ encompasses many meanings, from various sub-national levels to supra-national clusters of nations. In the case of RIS, the literature refers to sub-national geographical areas (Braczyk, Cooke et al. 2004), which still allows for significant variations in the geographies of regions under analysis.
In Australia, the adoption of innovation-led regional economic development has not gained wide acceptance. This is due to two factors: the connotation of
Chapter 1 - Introduction 7 ‘region’ in Australia; and the spread of industrial innovation policy throughout the three levels of governance in the Australian federal system. The term ‘region’ in Australia is still largely associated with rural, and usually declining, areas (Beer, Maude et al. 2003; Beer and Kearins 2004) and despite the majority of the population of Australia residing in three major cities – Sydney, Melbourne and Brisbane - non-metropolitan areas are the common focus of any regional economic development policies. Also, in the Australian federalist system, industrial development policy is largely developed and deployed at the national level, with minimal opportunities for customisation to smaller geographical levels. Even at the state level, where these is some attention paid to industrial innovation (NSW Government 2005), there is again minimal room for customisation, even though some Australian states are geographically as large as many other small countries. The third level of governance, local government, is rarely involved in this sphere of policy-making, despite the regional basis of much of this activity.
1.2 The research problem
In the RIS framework, innovation is acknowledged as the driver of economic growth, and regions are recognised as the locus for innovative activities. The RIS approach suggests that the actors (firms, institutions, and knowledge base) and their interactivity within the region drive knowledge flows and innovative activity, and that encouraging these activities drives economic development. However, the RIS has two iterations of empirically understood regions: ‘ideal’, or ‘star’ regions; and other regions, including peripheral and lagging regions looking to increase economic development through a focus on industrial innovation. It is this second iteration that has become the focus of most regions, although it has not benefited from the associated development of suitable methodological and policy instruments to support these activities, particularly given the multiple levels of government in Australia.
Chapter 1 - Introduction 8 Regional Economic Development Policy (REDP) is still built around either traditional theories of economic development (the same theories that have limitations in explaining innovative activity as the driver of economic growth), or the first iteration of RIS based on ‘ideal’ successful regions, which are largely a world away from what is available and what is necessary and possible in the encouragement of RIS in most other regions (Thomi and Bohn 2003). This means that in Australian policy circles, the available knowledge of innovation- led economic development policy models is largely based on variations of the ‘ideals’ regions, therefore rarely applicable to current regional geographies in Australia.
We have then the situation where to a large extent, as Markusen notes (2003), we still do not yet ‘know’ what an RIS looks like in reality, because these archetypes (in the first iteration) have shaped the conception and measurement of regional innovation activity. Each individual RIS is shaped by the unique local dynamics and characteristics that encompass them. Common elements and characteristics across systems can be identified utilising the basic focus of SI in understanding innovation: focus on actors, interactions and learning; as the works of Cooke (Cooke and Morgan 1998; Cooke 2001; Cooke, Heidenreich et al. 2004), and Asheim and Coenen (2005) demonstrate. These elements and their characteristics, however, have developed out of the first iteration of RIS and the success story regions, and do not easily allow for a range of other activities, or a lesser number of ‘ideal’ activities, which are observable in less exceptional regions, to be measured and interpreted within the RIS approach.
The identification and assessment of more applicable characteristics or “key indicators…in regions with regard to innovative capability” (Doloreux and Parto 2004 p.17) needs to become the focus of the RIS approach if it is to successfully offer explanation and, direction to the increasingly broad range of regions seeking advice on innovation-led economic development. Although RIS is attractive to regional policymakers in that it offers a way to examine systematic innovation processes within regions, the differences between the two fields of
Chapter 1 - Introduction 9 understanding of RIS must be bridged in order to secure the continued progress of the approach.
This represents a research opportunity, because so much theoretical and policy work has been undertaken examining and understanding the ‘ideal’, exceptional examples of regions, whereas most of the current empirical and policy need is with less exceptional and less (empirically) understood regions. In the Australian context, the opportunity is even wider given the connotation of region, and the industrial policy output of the Australian federal system.
The research opportunity entails providing a more widely applicable characteristics and key indicators of innovative activity suitable to the analysis of an array of regions. This research develops such indicators through the analysis of knowledge gathering activities, particularly Knowledge Intensive Service Activities (KISA) (Martinez-Fernandez, Soosay et al. 2005; Martinez-Fernandez 2006; Martinez-Fernandez and Miles 2006; OECD 2006) By focusing on knowledge gathering activities, the research is able to examine a more diverse range of activities than is traditionally associated with innovation analysis. Knowledge gathering activities also offer an opportunity for detailed analysis of learning, as they provide a link between learning processes and innovative activity.
Secondly, through the use of the RIS framework, there is an opportunity to examine an ‘ordinary’2 region (and, in the context of Australia, an urban region), and establish the role of local institutions in the current application of economic and industrial development policy in Australia. This should provide policy direction for innovation-led regional economic development policy at the local government level. To do this, the research project investigates an urban metropolitan region of South West Sydney and assesses its economic development potential through an analysis of industrial innovative activity. The aim is to understand the economic
2 Ordinary in comparison to the ‘ideal’ regions and therefore representative of the broader range of regions focusing on RIS
Chapter 1 - Introduction 10 development of South West Sydney in comparison with two other Sydney metropolitan regions, Central West and North West Sydney, by examining the innovative activities undertaken by firms in these regions. The focus is on ‘localized capabilities’ (Malmberg and Maskell 2002) particularly knowledge gathering activities, the relationship between these activities and each region’s ‘assets’, and how these relationships may contribute to regional development.
In doing this, the research also aims to further develop a relatively new methodological approach to innovation studies through the analysis of Knowledge Intensive Service Activities (KISA) and its application in regional analysis. KISA are closely linked to firm innovative activity (OECD 2006), and through analysis of firms’ use of KISA, an understanding of innovation and knowledge activities undertaken by firms is constructed. The advantage of using KISA for this analysis is that the approach should work equally well for weak and strong RIS. In weak RIS, KISA would still be produced, just in small amounts and more locally. Stronger RIS would exhibit more sophisticated use of KISA and a greater variety of types of KISA. Thus a measure applicable to multiple RIS is provided.
Within the analysis of regional innovative activity, attention is focused on three aspects of knowledge gathering: knowledge sources; KISA usage; and firm knowledge networks. These three aspects provide an opportunity to identify actors, their interactivity and their processes of learning. The spread of these actors and activities is also analysed to identify the regional dimension of RIS. The research analyses the extent to which actors and activities are bounded and influenced specifically by the unique regional location. In addition, two other factors affecting firm innovative activity: firm size (Storey 1985; Acs and Audretsch 1990; Jensen and Webster 2004; de Jong and Marsili 2006; Macpherson and Holt 2007) and industry sector (Breschi 2000; Capello 2001; Malerba 2005; Marsili and Salter 2005), are also considered.
The research will specifically provide answers to the following research questions and test the following hypotheses:
Chapter 1 - Introduction 11
Research question 1- How does innovative activity manifest in the three regions? Research question 2 - What kind of knowledge do firms access and from where? Research question 3 – How much knowledge is external and how much is internal? Hypothesis 1 – Firm size affects the mix of internal and external knowledge that firms access. Hypothesis 2 – Industry sector affects the kind of knowledge that firms access. Research question 4 - What are the key organisations for external knowledge sourcing? Research question 5 - What is the regional component of knowledge flows? Hypothesis 3 – The orientation of the regional supply chain is reflected in the location of key knowledge organisations. Research Question 6 - What determines the mix of knowledge that firms obtain for their innovation activities, and why? Who drives this process? Research question 7 - How and why does ‘regional experience’ affect this process? Hypothesis 4 – Regional experience affects the mix of knowledge firms access by shaping the type and location (internal versus external) of accessible knowledge.
1.3 Research design
The design for this research is cross-sectional. Although the focus of the research output in terms of policy is South West Sydney, in order to ascertain if specific regional factors are at work, a comparative methodology is necessary. The cross- sectional research design is a somewhat under-utilised research design in the regional innovation analysis field, however much research into regional development and economic geography utilises aspects of the design in an implicit sense for comparative work. Cross-sectional design allows for the analysis of variables based on established group differences. Data is collected at one point in time and used to examine the extent to which variation in different
Chapter 1 - Introduction 12 categories of the independent variable differ in relation to the dependent variable (De Vaus 2001). The main emphasis on data collection in the cross-sectional research design is to provide a structured data set over the groups to be analysed. The design does not lean towards any particular data collection method to be used. Data collection in this research takes place in three phases: firstly an economic and industrial audit is undertaken; this is followed by a business innovation survey; and finally there is a round of in-depth firm interviews.
The two regions selected for comparison are neighbouring outer Sydney regions. The selection of these regions provides for the ‘same but different’ elements necessary for comparative research. The regions share a somewhat similar geography both physically and economically to South West Sydney. All of them are located on the outskirts of the Sydney metropolitan area. All have experienced population growth pressures, with population growth well in excess of the metropolitan average in the past decade, and projected to grow at above average levels in the future, as all three regions have future ‘greenfields’ residential development sites identified within their boundaries. Also, all of the regions are part of the Sydney Metropolitan Area (SMA) and as such experience the same overall state governance, the same metropolitan labour market and similar access to available infrastructure in terms of roads, rail, water, airports and ports.
Points of difference include: employment growth levels; industrial differences revealed through finer aggregations of industrial and occupational data; and the policy responses of the local government authorities to regional economic development.
The research analysis and discussion is presented over three chapters. The three chapters have an inherent structure, in that each one focuses on a different research question and associated hypotheses. Statistical analyses, particularly measures of association, principal components analysis, and hierarchical cluster analysis, are used in the analysis of survey data, and qualitative analysis is used in interpreting the interview data.
Chapter 1 - Introduction 13
This research forms the central data collection component of a broader Australian Research Council (ARC) linkage project3 with one of the local government authorities located within the South West Sydney region; Liverpool City Council. The ARC Linkage project provides broad research and policy outcomes for this local government authority, and elements of the data collection are targeted specifically for this purpose.
1.4 Scope of the study
South West Sydney is an urban metropolitan region on the outskirts of the metropolitan area of Sydney. The region comprises four Local Government Areas4(LGA): Liverpool; Campbelltown; Camden; and Wollondilly (see maps in Figure 1.1-1.3). The research examines South West Sydney against the background of two comparative regions: (i) Central West Sydney, which includes the LGAs of Blacktown and Penrith; and (ii) North West Sydney, which contains the LGAs of Baulkham Hills and Hawkesbury. These regions broadly5 match the Sydney growth corridors mapped out in the 1950s metropolitan planning instrument, the Cumberland Plan (NSW Government 1951). The Cumberland Plan represented Sydney and Australia’s first concentrated metropolitan planning instrument (Meyer 2005). These regional boundaries have therefore been present in some way or other in all of Sydney’s metropolitan and regional planning, including in the most recent policy framework for metropolitan planning, the Sydney Metropolitan Strategy, (NSW Government 2005).
3 LP0347917 Chief Investigator Dr. Cristina Martinez-Fernandez APA(I) Samantha Sharpe 4 Local government is the lowest level of the three tiers of government within Australia. Local Government Areas are the boundaries of these jurisdictions. 5 There have been a number of minor boundary changes at the local government level over previous decades in line with the growth patterns of the metropolitan area.
Chapter 1 - Introduction 14 Figure 1.1 May of Sydney Metropolitan area by regions
Chapter 1 - Introduction 15 Figure 1.2 Three regions under investigation
Chapter 1 - Introduction 16 Figure 1.3 Central West Sydney in detail
Figure 1.4 North West Sydney in detail
Chapter 1 - Introduction 17 Figure 1.5 South West Sydney in detail
Naturally, as a designated residential growth corridor, South West Sydney has experienced rapid population growth, particularly in the last decade, increasing by 32% in the period from 1991-2001. This population growth constituted almost 20% of the Sydney metropolitan areas’ total population increase. Population projections for the next twenty years predict that South West Sydney will continue to grow at a similar rate, increasing by more than 200,000 new residents in the next twenty years.
However, during the same time period, employment growth in the region has only been slightly above the metropolitan average: 3.5% per annum in South West Sydney, and 3.1% per annum in the Sydney Metropolitan area (ABS 2001, 1991). South West Sydney’s employment growth only accounted for a 5% share of Sydney’s total employment growth. In some sectors and locations, for example, business and finance services employment in the Liverpool area, there has been no growth, or even negative growth in the five years between 1996 and 2001.
Chapter 1 - Introduction 18
South West Sydney is in the position of experiencing rapid population growth, with only moderate levels of employment growth, and with a relatively young age profile and high labour force participation rates, all of which has led to many residents travelling outside of the region for employment. Local government are concerned that if the same ratio of population/employment growth is achieved in this coming two decades as was present in the previous one, there will be a significant decline in regional employment self-sufficiency. In some regions such a decline may not necessarily be a cause for concern provided adequate transport infrastructure and resources are present to allow the effective movement of large numbers of people around the wider metropolitan area, so that access to jobs, facilities and markets is not affected.
In the case of Sydney, however, the reverse is true. Sydney is moving to a multi- centred metropolitan area platform, with an emphasis on regional containment of employment, recreation, and retail facilities, to reduce the need for cross-city commuting in the uniquely low density and therefore geographically large metropolis of Sydney. Within the current metropolitan planning framework the focus is on building up defined regional centres, as opposed to investing in inter- regional accessibility infrastructure. The current metropolitan strategy, subtitled ‘Sydney, a City of Cities’, has as its first listed strategy:
“1. STRONGER CITIES WITHIN THE METROPOLITAN AREA Sydney City and North Sydney will continue to be the harbour cities at the heart of Global Sydney. They will be the focus for world class business, tourism, cultural, health, and education and entertainment activities. The river cities of Parramatta, Liverpool (South West Sydney) and Penrith (Central West Sydney) will provide a focus for innovative business environments, jobs and more lifestyle and work opportunities closer to growing parts of Sydney. These centres will attract new shopping, health, education, business and cultural facilities.” (NSW Government, 2005, p8)
In addition, in comparing the types of employment which are increasing (retail) with the types that are not increasing highlights that South West Sydney is missing key industrial components. These are principally in financial and
Chapter 1 - Introduction 19 business services. These components allow adequate city and industrial development through the provision of Knowledge Intensive Business Services (KIBS) (Sharpe and Martinez-Fernandez 2006).
Therefore, the situation facing South West Sydney is that it needs significant employment generation in the coming decades, across a variety of occupations, to ensure current and future population growth is sustained in conjunction with associated employment growth. This phenomenon is also occurring within a metropolitan planning environment promoting multi-centred development, and a corresponding de-emphasis on inter-regional accessibility, thereby providing further negative consequences for sustainability (i.e. less accessibility to gain employment in other regions). Therein lays the empirical problem behind this research.
1.5 Research significance
The significance of this research is principally methodological and policy related. Methodologically its significance is in terms of the application of KISA in the analysis of regional innovative activity; its policy related significance in respect of drawing policy conclusions for innovation-led regional economic development policy in a specifically urban Australian context. However, in the application of the methodology within this particular study, the research also aims the make theoretical contributions to the RIS approach in terms of highlighting the integration of elements of RIS.
The research is also significant to the region of South West Sydney in its provision of previously unavailable data on the innovative capacity of firms within the region and advice on policy arising from this analysis. This research, as earlier stated, is part of a larger Australian Research Council (ARC) linkage project with a prominent local government authority in South West Sydney, Liverpool City Council. The policy advice to the Council will also be in the form of other reports.
Chapter 1 - Introduction 20 1.6 Structure of the thesis
This thesis consists of eight chapters. Chapter Two further explores the theoretical and empirical literature of the RIS approach, detailing the established elements of the RIS approach, and current debates and contested areas in terms of RIS applicability. The chapter also provides theoretical reasoning for the focus on KISA and the firm knowledge networking analysis as proxies for the principle elements of the SI approach; actors, interactions, and learning.
Chapter Three outlines the research methodology, including the selection of the cross-sectional research design and the limitations of this research design, and how these are mediated. The chapter also provides details of data collection methods, samples and forms of data analysis.
Chapter Four provides an analysis of the economic and industrial environment of the three regions, detailing socio-economic factors of the labour force, industrial composition, and growth areas and proxies for the regional knowledge base. This chapter gives an understanding of the socio-economic profile of each of these regions and how this may affect the firm innovation results discussed in the following chapters.
Chapter Five is the first of the three chapters to report empirical research results and discussions. Chapter Five focuses on the innovation activities and KISA usages of firms. The chapter presents results of the business innovation survey on innovation counts, knowledge sources, and KISA usage and location to establish levels of innovation activity and knowledge sourcing across the three regions. The first three research questions, and research hypotheses one and two, are the focus of this chapter.
Chapter Six also draws on the business innovation survey results to explore external organisational networks of firms. The chapter examines the reach of firms’ knowledge networks and how these may shape the learning processes of
Chapter 1 - Introduction 21 firms. The fourth and fifth research questions, and hypothesis three, are addressed in this chapter.
Chapter Seven is the final chapter of research results and discussions. This chapter draws together results presented in Chapters Five and Six, and discusses the mix of knowledge gathering activities of firms and the drivers of these activities. The chapter concludes by examining how the ‘regional experience’ of firms and the knowledge professionals within these firms affect their knowledge gathering mechanisms. The final two research questions (six and seven), and the fourth hypothesis, are addressed in this chapter.
The final chapter, Chapter Eight, offers conclusions on the methodological and empirical problems established in this thesis; namely the re-conception of RIS for a broader range of regions, including less favoured regions, and through the application of the KISA analysis and a focus on regional organisational networking. Some theoretical conclusions for the RIS approach in its second iteration are also proposed. This in turn provides a platform from which to offer policy recommendations in the specific case of South West Sydney.
Appendices and a reference list conclude the thesis.
Chapter 1 - Introduction 22
Chapter 2 - Knowledge & innovation activities in regions
Innovation scholarship has a long history, despite the current focus on knowledge based economies. The concept of a knowledge based economy is met with a variety of opinions within the literature, ranging from being considered a ‘buzzword’ (Godin 2006) to being considered the result of significant structural change in the forces of production within the economy and, as a result, the basis for understanding the economy and its development (Lundvall 1992; OECD 1996a). Despite this diversity in opinion, there is broad acknowledgement that knowledge and innovation are central drivers of economic growth (Brokel 2007), and that innovation and knowledge and their associated activities need to be a concern for all firms and not just those situated in highly technological or emerging industries (Simmie 2003).
This chapter outlines some of the existing literature on regional industrial innovation, and the processes and types of knowledge and learning associated with these activities. The RIS approach is referred to as an approach and not a theory because it does not fully explain the relationships between geography and innovation; it only provides a framework through which both can be examined. The reason for this is RIS is essentially aspatial. It does not take account of
Chapter 2 – Innovation and knowledge activities in regions 23 factors such as distance, transport costs and rent differentials, which are at the centre of location theories.
Rather it emerges from growth theories (with innovation being the driver of economic growth), specifically those dealing with the cumulative nature of economic activities. It is in the cumulative or embedded aspects of RIS that a sense of territory and place emerges. This concept is not related to actual geographical form but rather to embedded economic behaviours (work/residential patterns, market relationships, political boundaries). Space in the RIS sense is a socio-economic and political construction of territory rather than a sum of location factors. Whilst this may be defined as ‘aspatial’ in strict geographical terms, these socio-economic boundaries have as much an effect on activity within regions as locational factors.
A true theoretical synthesis of these two elements is not available, and whilst developing such a synthesis must be the continuing focus of current and future generations of regional scientists (Capello 2007) research into the two aspects; the drivers of economic growth and the spatial geographical forms these growth pattern, must continue. This research aims to contribute to the former. The contribution this research makes to this field is to increase the sophistication and sensitiveness of regional knowledge analysis within the RIS approach through the application of the Knowledge Intensive Service Activity methodology; recently developed through OECD research.
2.1 Innovation and knowledge activities
“It is impossible to overstate the importance of innovation. It is one of the key drivers of productivity. It can help businesses improve the way products and services are made and deliver or introduce entirely new ones. It can reduce costs by increasing efficiency. Evidence shows that innovating companies sustain a higher performance and grow faster than non-innovators. Innovation is the source of real competitive advantage for industrial businesses. For the economy as a whole, it is central to our prospects for sustained prosperity. Innovation is about anything that enables business to improve the products and services they can offer.
Chapter 2 – Innovation and knowledge activities in regions 24 Exploiting new technology may be one way of doing so but it is equally likely to come from adopting a new business process, using new management techniques or increasing the skills of your workforce” (Sainsbury 2003).
As the previous quote states, innovation is a broad concept encompassing the inception and creation of new products, reducing costs of producing existing products, and adopting new business practices or management techniques. Innovation is defined by the OECD as follows:
“An innovation is the implementation of a new kind or significantly improved product (good or service), or process, a new marketing method, or a new organisational method in business practices, workplace organisations or external relations” (OECD 2005)
The two definitions of innovation quoted above highlight three primary elements. Firstly, that innovation is something new, whether completely new or, as is more likely the case, a recombination of previous knowledge to create significant improvements to existing products and processes. Secondly, innovation can occur in a number of spheres, not just in relation to new products, but also new processes, both operational and organisational. Thirdly, that innovation includes activities that are external to the actual innovating agent.
The OECD definition of innovation is the foundation for most western countries’ innovation data collection methods, including the European Union Community Innovation Survey, and innovation surveys by national statistical agencies in both Australia and New Zealand. More broadly, the elements of this definition are used in other commercial innovation studies, such as the IMB Global CEO study (IBM 2006).
As innovation is such a broad and all-encompassing term, numerous categorisations have emerged to refine the concept. Product innovation can be further classified into goods and services, and process innovations into technological processes and organisational processes (Edquist, Hommen et al. 2001). Further categorisations can also be made regarding the source of demand for individual innovations in the production cycle, these being: investment
Chapter 2 – Innovation and knowledge activities in regions 25 products; intermediate products; and consumer products. Sectoral factors offer further classifications in manufacturing regarding the sectoral use of technology, with divisions being made between high, medium or low usage sectors, and service sectors being categorised according to their knowledge intensity (Breschi 2000; Edquist, Hommen et al. 2001).
The distinction between product and process innovation has a long history, originating with Schumpeter, who defined a new product as “the introduction of a new good…or new quality of good”, and process innovation as “the introduction of a new method of production…(or) a new way of handling a commodity commercially” (Schumpeter 1937 p.66). The role of innovation in the design of new products and services, and productivity gains achieved through technological process innovations, is well established. A more recent realisation is the role of organisational and managerial process innovations. These processes relate to the managerial or organisational “strategies, structures or routines of a business, which aim to improve the performance of the business” (ABS 2003 p.58). Examples include changed corporate directions, implementation of advanced management techniques such as Total Quality Management, improved business performance measures, significant workplace re-organisations, and important changes to communication and information networks. Organisational innovations are themselves an important source of productivity growth, but have also been shown to be closely related to the successful implementation of other innovations, both product and technological processes (Edquist, Hommen et al. 2001).
The distinction between types of innovation is important because of the breadth of activity and behaviour that encompasses innovative activity. Different innovation types are associated with differing behaviours and actors. For example, product innovation may be more associated with research and development activities whereas organisational innovations may result from management training and techniques.
Chapter 2 – Innovation and knowledge activities in regions 26 2.1.1 Extent of innovative activities
Innovative activities are defined as:
“…all the scientific, organisational, financial and commercial steps which actually, or are intended to, lead to the implementation of innovation” (OECD, 2005)
Whereas the previous definitions discussed innovation, this definition refers to the activities associated with innovation. These activities can include novel and exceptional activities that firms may be undertaking, but also more commonplace business functions that are equally essential for the implementation of innovations (OECD, 2005).
It can be seen therefore that innovative activities are much more than research and development (R&D) programs, and intellectual property and patent advice - inputs that are usually associated with more traditional measures of innovative activity such as R&D expenditure and employment, and patent and trademark counts and applications. The ability to measure and interpret the usage of this wide range of activities is behind the empirical and theoretical focus on firm knowledge gathering, particularly developments of Knowledge Intensive Business Services (KIBS) and, more recently, Knowledge Intensive Service Activities (KISA). These two items are explored in the following section, but first a more detailed discussion of knowledge and the process of acquiring knowledge and learning are necessary.
2.1.2 Knowledge and learning
Knowledge is the resource driving innovation. It is understood as the “dynamic framework or structure from which information can be stored, processed and understood” (Howells 2002:p.872). The processes of innovation and knowledge gathering within a firm are critically linked to their processes of learning. Learning itself is dependent on interaction, and is inherently a socially embedded process (Lundvall 1992; Capello 1999). This context includes behaviours and
Chapter 2 – Innovation and knowledge activities in regions 27 practices, the ‘rules of the game’, in a specific setting or location. The recognition of the importance of these institutional and cultural factors, and the interaction between them within the learning process, is one of the key reasons innovation is now examined within a systems approach (Lundvall 1992; Smith 1996; Edquist 1997; OECD 1997; de La Mothe 1998; OECD 1999b; Cooke 2001; Braczyk, Cooke et al. 2004).
The value of knowledge has increased within the current economy because the quantity of information currently available to firms and individuals, and the ability to make sense of it all and identify relevant pieces of information, has become a valuable task. Knowledge within a firm can be analysed along two dimensions: its mode of expression; and where the locus of knowledge is located (Lam 2002). The first dimension gives rise to the distinction between tacit and codified knowledge, and the second between individual and collective levels of knowledge.
The first dimension emanates from Polanyi’s argument that knowledge exists in two forms: tacit and codified (Polanyi 1962; Polanyi 1966). The distinction between tacit and codified knowledge is based on the accessibility of knowledge and how the knowledge is transferred and understood and the degree to which it is formalised. Codified knowledge is formalised within a set code or language. It takes the form of documents, publications and knowledge embodied within machines or artefacts (Howell 2002). Codified knowledge can be easily transferred and acquired through formal study because as long as the code is known, the knowledge can be accessed. This form of knowledge can also be generated through logical deduction (Lam 2002).
Tacit knowledge, on the other hand, relies on a context of shared experience for knowledge transfer; it requires ‘learning by doing’. It is argued that because of this, tacit knowledge is limited in its ability to be transferred, as direct experience usually requires close or face-to-face contact (Koschatzky 1999; Lawson and Lorenz 1999; Lissoni 2001; Howells 2002).
Chapter 2 – Innovation and knowledge activities in regions 28 The differences between tacit and codified knowledge are noted as a distinction rather than a division. Knowledge is conceived on a spectrum, with tacit and codified knowledge at either end of the spectrum rather than in two distinct forms. This is because the generation of new knowledge requires the combination and interaction of both types of knowledge (Nonaka and Takeuchi 1995).
Nevertheless, the consequence of this distinction is a division in knowledge accessibility between what is known and accessible to firms, and what is not (Brenner 2007). It is argued that new and highly innovative knowledge usually exists firstly in the tacit form. As new knowledge is primarily transferred through face-to-face contact, knowledge transfer will occur through these personal contact channels, as Lam summarises
“…tacit knowledge, which is difficult to create and transfer in the absence of social interaction and labour mobility, constitutes a most important source of learning and sustainable competitive advantage in an increasingly globalised knowledge-based economy” (Lam 2002:81)
Therefore tacit knowledge naturally tends to be associated with geographical proximity (Lawson and Lorenz 1999; Malmberg and Maskell 2002; Simmie 2003; Lorenzen 2005; Ronde and Hussler 2005). As a result, it is argued, certain spaces and locations can access tacit knowledge that is not readily available to all firms. Access to this tacit knowledge may not be on a permanent basis, because as knowledge transfers repeatedly, it also codifies in some ways, curtailing any spatial advantage.
This notion is not entirely settled. There are contested issues around both the tacit and codified knowledge transfer arguments. Firstly, Breschi and Lissoni suggest that the idea that highly innovative, new knowledge would only be available in a tacit and fuzzy way, seems improbable (Breschi and Lissoni 2001). Although it must be pointed out that their examples were of highly specified scientific material, while the idea may still hold true for other knowledge types. Breschi
Chapter 2 – Innovation and knowledge activities in regions 29 and Lissoni also highlight the fact that modern telecommunication methods provide the ability for tacit messages to be sent over long distances, thus negating the supposition that tacit knowledge transfer is a geographically bounded phenomenon. This is supported by the work of Maskell, Bathelt et al (2006) who talk of trade fairs, exhibitions and conferences as providing “hotspots of intense knowledge exchange, network building and idea generation” (2006:p 997), and thereby also providing another form of proximate knowledge transfer in addition to regional transfers.
In terms of codified knowledge, the argument that this type of knowledge is equally and evenly available to all firms is also not borne out in the empirical literature (Carter 1994; Andersson and Karlsson 2004; Fagan and Dowling 2005; Sharpe and Martinez-Fernandez 2006). As Brenner (2007) notes, the two forms of knowledge (tacit and codified) are complimentary, in that ‘knowing’ that codified knowledge exists, and where it is reliably located, requires a degree of tacit knowledge. This is particularly true with the current explosion of information available to firms, especially through the internet.
This discussion highlights the complexity of the tacit/codified knowledge interplay. However, the distinction provides the first point of entry for spatial analysis of innovation, with the proposition that uneven levels of tacit knowledge transfer are behind differential economic performance of locations or, as Brokel (2007) comments, “…tacit knowledge locks economic performance and innovative activities into a tight geographical prison” (2007:153). The importance of tacit knowledge also highlights the crucial role that networks (both organisational and individual) play in the transfer of this type of knowledge. It is seen then that knowledge networks and their structure are important in understanding innovative activities and hence economic development. However, the importance of tacit knowledge cannot be over-emphasised at the expense of codified knowledge because, as Brokel (2007) says, tacit and codified knowledge are complementary.
Chapter 2 – Innovation and knowledge activities in regions 30 These dynamic issues of knowledge form and its limits will not be resolved in this research. What however can be drawn from this literature is the realisation of the complex interplay between the two types of knowledge (tacit and codified) and that the presence of both is critical in the innovation processes of firms. Therefore in the empirical investigations of this research, data on both tacit and codified sources of knowledge must be collected and analysed, as well as the information on how these knowledges are used, transmitted and stored.
There are two levels where knowledge can reside and learning can occur: both on an individual level; and on a collective level, between groups of individuals in some kind of organisation. Individual knowledge is owned by the individual. It can be applied independently to specific tasks and transferred through the movement of the individual. Collective knowledge, however, resides with a number of people within an organisation, and manifests within the organisation’s accumulated history of routines, procedures and processes. Collective knowledge can be referred to as: ‘stock’, a collection of stored knowledge; or a ‘flow’, knowledge generated through organisational interactions (Lam 2002).
The process of collective learning, defined as “a process of cumulative knowledge, based on a set of shared rules and procedures which allow for individuals to co-ordinate their action in search for problem solutions” (Capello 1999), is at the heart of many territorial innovation theories (Lawson and Lorenz 1999; Ronde and Hussler 2005; Hussler and Ronde 2007). The theories however have their roots within organisational theory and have not directly emerged from the Innovation Systems approach. The link between collective learning and territorial innovation systems has been made because whilst collective learning can occur within formal organisations such as firms, the same processes are also evident in regions.
These advantages of collective learning in regions have variously been defined as knowledge spillovers (Feldman 1999; Breschi and Lissoni 2001; Simmie 2002; van Stel 2004; Audretsch and Lehmann 2005) and relational capital (Camagni 1995; Capello 2001; Camagni 2003; Capello and Faggian 2005). They are an
Chapter 2 – Innovation and knowledge activities in regions 31 advantage based on a firm belonging to a particular region that enables a different flow of knowledge between other firms in the same region which can allow for better and more customised knowledge transfer and generation. This advantage is not available or easily transferable to firms outside the region. Therefore the analysis of knowledge gathering activities is essential in determining the flow of knowledge in and between firms in a region. The next section refers to how this can be achieved.
2.1.3 Knowledge intensive activities
Knowledge intensive activities have been of interest to innovation scholars, as they offer a way to capture and analyse knowledge and knowledge gathering activities in firms and geographies. Initially this interest focused on Knowledge Intensive Business Services (KIBS), but more recently attention has turned to Knowledge Intensive Service Activities (KISA) (Miles, Kastrinos et al. 1994; Miles 2003; Martinez-Fernandez and Krishna 2006; Martinez-Fernandez and Miles 2006).
KIBS are defined as privately owned firms whose main function is to create and provide knowledge-generating services for other firms (Thomi and Bohn 2003). The focus on KIBS grew as their role in the innovative activity of firms emerged (Hales 2001; Miles 2003). They contribute to firms’ innovation processes directly through the provision of new knowledge, but also indirectly as a conduit through which other relevant knowledge from external sources can flow via the KIBS and into the client firm. KIBS also act as ‘linking agents’ between the service and manufacturing sectors of the economy, and offer the opportunity for manufacturing firms to extend their service provision (Martinez-Fernandez and Miles 2006). KIBS are now seen as being of critical importance to innovative activities in the service sector, and hence the innovation systems of modern economies (Martinez-Fernandez and Miles 2006).
Chapter 2 – Innovation and knowledge activities in regions 32 Recent literature on KIBS has focused on their important role as the link between these knowledge intensive services, and innovation. This is certainly the case for the number of studies that have analysed KIBS within the SI and RIS framework (Den Hertog 2000; Martinez-Fernandez, Potts et al. 2005). KIBS’ role as a conduit means that knowledge from spatially distant regions, previously unavailable to the client firms through their own resources and contacts, is available through the KIBS firms. This means that the transferral limitations outlined earlier of highly specialised and new knowledge, which were previously only accessible in the form of tacit knowledge, can now be transferred over long distances, but only within the framework of the KIBS and its codified structure. Therefore, the availability and use of KIBS can overcome some of the potential spatial limitations of knowledge accessibility for firms in particular regions.
The analysis of KISA developed from KIBS. Critical questions about the how, when and why firms accessed specific types and forms of knowledge intensive services, led to a shift in focus away from the firms providing the services (KIBS) and towards the types and forms of activities accessed (KISA). The new focus was on the “activities that firms demand and use themselves rather than those they supply to others” (Martinez-Fernandez 2006 p.110), as research showed that knowledge intensive services are provided by firms that are not necessarily defined as KIBS (AEGIS 2003). The analysis of KISA instead of KIBS can therefore provide a more detailed picture of the knowledge intensive service activity of regions and, through KISA’s connection with firm innovation, innovative activity as well.
KISA is defined as:
“…the production or integration of service activities, undertaken by firms and public sector actors – in the context of manufacturing or services, in combination with manufactured outputs or as a stand- alone service” (OECD, 2006, p31)
Its role in the innovation activities of firms is also outlined below.
Chapter 2 – Innovation and knowledge activities in regions 33 “The role of KISA in innovation is pervasive and important across the economy in developed nations. Increasing supply and demand for specialised KISA signifies the evolving divisions of labour in the economy. Specialised experts and integrator services help organisations manage increasingly complex technologies, rapidly changing operational environments and evolving business concepts.” (OECD, 2006 p.47)
It is the role that KISA plays in the innovative activities of firms, as outlined in the previous quote, and the interactive nature of KISA, that makes its analysis suitable within the RIS framework. KISA analysis includes analysis of firm knowledge creation and transfer and the interface of firms’ internal knowledge resources with external knowledge sources, all occurring within an interactive space. KISA analysis also offers the advantage of covering a large range of activities, some of which are business function activities, and others of which are more exceptional and sophisticated activities (Albors, Hervas et al. 2007). This in turn allows the analysis of a broader range of innovative activities and regions within the RIS framework, and the ability to identify distinctions between the levels of complexity within different systems.
2.1.4 Defining regions
The term ‘regional’ encompasses many meanings, from various sub-national levels up to supra-national clusters of nations. For the purposes of this research, ‘regional’ will refer to a specific sub-national geo-political area. To be considered as a ‘region’, the main criterion required of a selected geography is some sense of connection and continuity that is recognised by the population. This connection and continuity usually relates to the functionality of the region in terms of economic and institutional purpose, hence the term ‘functional regions’ (OECD 2002; Andersson and Karlsson 2004).
The selection of regional boundaries is a contested issue. The selection process needs to consist of a compromise between two things: administrative functions and efficiencies on the one side; and more traditional regional identities based on
Chapter 2 – Innovation and knowledge activities in regions 34 a shared history of people and place on the other side. As Lagendijk (2004) summarises,
“Different stories about the region suggest different meanings. Economic stories draw a line around core assets of economic strengths, territorially embedded and mutually networked. Spatial planning and sustainability stories point at the importance of a region as an area of a certain size, with certain functional characteristics. Governance stories rely on the notion of the region as an area of civic identification and institutional capacity. From an institutional perspective, boundaries are a compromise between two sets of conditions. First, the purposes and limitations associated with the process of chopping up a national territory into smaller units. Second, the (im)possibilities of scripting regions given the wider policy objectives, and given the pre-existing regional endowments (identity, institutional capacity, economic potential, etc)” (2004, p. 15)
In many cases regional boundaries will be congruent with administrative functional boundaries and traditional regional identity and be the result of geo- political boundaries or aggregations of them.
Commuting patterns are a common way of identifying functional regions, particularly with the growing recognition of the importance of labour markets to regional innovative capacity (De Laurentis 2006). Such commuting patterns were used in the establishment of the regional boundaries of the three regions under investigation in this research. These patterns emerged from the detailed background work undertaken for the Sydney Metropolitan Strategy (NSW Government 2006), and demonstrate that Sydney is not a homogenous metropolitan area but rather that distinct intra-regional labour markets exist because of commuting times and access to rail and road infrastructure.
2.1.5 Innovation – the systems approach
A system is defined as a “set of arrangements of things so related and connected as to form a unity or organic whole” (Carlsson, Jacobsson et al. 2002 p.233). A system is made up of components, relationships and attributes. Therefore, in an
Chapter 2 – Innovation and knowledge activities in regions 35 innovation system, innovative activity is determined by: its components (various actors such as individuals, firms and institutions); their relationships (through various networks - personal, informational etc); and their attributes (what they do, who they are and the cumulative base of knowledge, culture and routines in which they operate).
The previous discussion around knowledge and learning, and the inherently interactive nature of the associated processes demonstrate why the SI approach is suitable to innovation analysis, because the ‘systems of innovation’ approach positions interactivity and learning as critical players within the system. The approach understands innovation to be a socially embedded process. It interprets innovation as being the transformation of ideas and knowledge into novel technologies, products and services, through the process of learning and searching (Asheim and Coenen 2005). Andersson and Karlsson (2004) distinguish four key foundations of the SI approach. Firstly, that firms must be seen as part of a network of other actors both public and private, and through their activities and interactions generate, transmit and diffuse new technologies and knowledge. Secondly, the approach recognises both formal and informal linkages between organisations. Thirdly, it identifies the analysis of the flow of resources and knowledge between the organisations as being critical. Finally, the approach recognises the elements of learning as a key economic process.
The SI approach provides a framework through which to examine interactivity and learning, with the addition in regional innovation systems of a regional geographical setting. Some authors consider the SI approach to be simply a framework, rather than a theory, because the approach does not provide convincing propositions regarding relationships between variables (Acs, Anselin et al. 2002), which further highlights some of the complications with the RIS, as noted in Chapter One.
Comprehending the SI approach as a framework for analysis rather than a theory may have emerged from the atmosphere of its development,
Chapter 2 – Innovation and knowledge activities in regions 36 “…instead of looking for clear-cut intellectual origins of the innovation systems concept its main background should rather be found in the needs of policy makers and students of innovation” (Lundvall, Johnson et al. 2002 p.215).
The approach developed out of policy demands rather than theoretical progress (Lundvall, Johnson et al. 2002). SI scholarship flourished at the time when many academics and policy practitioners alike were attempting to come to terms with the effects on economic activity of knowledge, information, and learning in a rapidly globalising world. The SI approach draws theoretical insights from Schumpeter’s concept of innovation: “the understanding of innovation as a new combination” of things (Schumpeter 1937 p.66). This highlights an “important contradiction in the elements of innovation; its continuity (existing elements) and radical change (the new combination)” (Lundvall, Johnson et al. 2002 p.216). The approach also draws from work by Perroux (1969) on growth poles, Myrdal (1957) on positive and negative feedback, cumulative causation and the virtuous and vicious circles of development, and also the evolutionary economics of Nelson and Winter (1982).
2.2 Regional Innovation Systems
RIS developed within this theoretical environment, and from the acknowledgement that innovation is a somewhat geographically bounded phenomenon. As stated in Chapter One, within the systems framework there are three levels of analysis that provide an understanding of innovation: (i) national innovation systems (Lundvall 1994): (ii) sectoral and technological innovation systems (Carlsson and Stankiewicz 1991; Breschi 2000; Breschi, Malerba et al. 2000; Breschi and Lissoni 2001) and (iii) the regional innovation systems approach (Cooke 1999; Cooke 2001; Braczyk, Cooke et al. 2004; Doloreux and Parto 2005). This research is located within the third approach. Despite the differences, all three of these approaches share a common philosophy shaped around three elements of innovation systems. These are: examining innovation through the interactivity between a set of actors; providing a framework that
Chapter 2 – Innovation and knowledge activities in regions 37 focuses on learning and knowledge processes; and a close and dynamic relationship with policy development.
The RIS approach was developed by scholars working from a regional science and economic geography background (Cooke 2001). The approach differentiates itself from the other two levels of analysis through its focus on the importance of proximity, particularly geographical proximity, in directing the complexity of knowledge and learning processes. The limited spatial range for transfer of tacit knowledge, and the association of tacit knowledge with new and emerging knowledge, is the basis for the RIS approach. The idea that tacit knowledge appears to be best transferred through face-to-face contact means that geographical proximity is highly significant, particularly within a region, in which it is assumed that, through trust and social relationships, higher levels of face-to-face interactivity will provide optimal mechanisms for knowledge transmission (Capello 1999; Simmie 2003; Cumbers and MacKinnon 2004; Asheim and Coenen 2005). The proximity between actors within the region makes it possible in some circumstances for firms to generate, acquire, and use knowledge faster than firms outside the region (Maskell 1998).
As a result of the recognition of the importance of proximity to knowledge generation and transfer, the RIS approach sees “that specific local and regional resources may still be important in firms’ efforts to obtain global competitiveness” (Asheim and Isaksen 1996 p.77), and that “the enduring competitive advantages in a global economy lie increasingly in local things – knowledge, relationships, motivations, that distant rivals cannot match” (Porter 1998 p.78).
There is no clear consensus of how to define regional innovation systems. This is in part the result of the rapid development of the RIS approach, however, with two iterations of the concept it is possible for more diversity and variation to be captured (Cooke 2004).
Two of the existing definitions are:
Chapter 2 – Innovation and knowledge activities in regions 38 “Basically, a regional innovation system consists of two main types of actors and the interaction between them. The first types of actors are the firms in the main industrial cluster in a region including their support industries. Secondly, an institutional infrastructure must be present i.e. research and higher education institutes, technology transfer agencies, vocational training organisations, business association etc which hold important competence to support regional innovation” (Asheim and Isaksen 1996 p.83). ; and “A regional innovation system is defined as a set of interacting private firms and public authorities, research organizations, and other bodies that function according to organizational and institutional arrangements and relationships conducive to the generation, use and dissemination of knowledge. In this conception, the environment of an innovating firm consists of a multitude of actors involved in the innovation process” (Doloreux 2004 p.11).
Whilst the definitions may agree on the components of regional innovation systems, these being the firms and organisations associated with the knowledge infrastructure of a region, they do differ in their selection of firms and institutions. Ashiem and Isaksen note that RIS firms are from “the main industrial cluster in the region”, whereas other authors note that a number of industrial clusters can comprise a RIS. For example, De Laurentis argues:
“Within the same region different industrial agglomerations can co- exist and different sectoral specializations may take place, in other words different systems of innovation may be found in the same region. To some extent, this gives impetus to the possible differentiation within regions and takes into consideration the differences that exist within the same region” (2006, p1060).
; and Doloreux and Parto (2004) observe that
“…all regions have some kind of regional innovation system, including not only regions with strong preconditions to innovation, but also old industrial regions…., peripheral regions…, rural regions… and regions in transition” (p.17)
Most RIS investigations begin with the firm or a geographical cluster of like firms in a related industrial grouping. To this it adds a supporting and specialised
Chapter 2 – Innovation and knowledge activities in regions 39 institutional environment, and a cumulative base of knowledge contained within the local labour force. As a result, when we talk about RIS we are actually most often talking about a one-industry scenario. This is a limitation in the applicability of the RIS framework; it works from the specific (one industry), up to the regional (more general, encompassing all industries), making the assumption that the specific will apply for the general, or the single area for the whole region.
2.2.1 Iterations of the concept
As discussed in the first chapter, much of the variety in the RIS conceptual development has arisen from the rapid progress of the field. This has meant that the two streams of understanding of the RIS concept have overlapped and caused the conceptual and policy gaps of the second iteration to be filled by analysis from the first.
The first iteration of RIS emerged from the explanation of a number of globally prominent ‘star’ regions. Examples include: Silicon Valley; Route 128; and Third Italy (Piore and Sabel 1984; Saxenian 1994; Storper 1997). These success-story regions were associated with high profile scientists in key public research institutions. Subsequent development in terms of industrial districts, innovative milieu and clusters, led to the term being associated with industrial concentrations, and the various definitions of clusters and the rediscovery of Marshallian externalities being seen as the explanation and driving force behind these star regions. These first iteration RIS investigations then are seen to centre on a geographical cluster of successful firms (usually high technology firms), which are located within a specific industry.
More recently, and partly in response to the RIS approach becoming the “rising start on the regional economic development front” (Cooke and Memedovic 2003 p.18), more and more regions have sought to use the approach to analyse their situation. These regions have included those with less favourable economic performance, and lacking the same level of sophistication in their institutional
Chapter 2 – Innovation and knowledge activities in regions 40 activity which exists in the star regions. These RIS investigations do not necessarily begin with a successful cluster of firms, but instead arise from the geographical location, and the analysis of the endogenous firms located within the region.
RIS has thus become a broader approach. Within the RIS approach, analysis of regions has included: countries (Denmark, (Maskell 2004) and Wales (Cooke 2004) provinces and states (Ontario, Canada (Gertler and Wolfe 2004) and Tuscany, Italy (Dei Ottati 2004); cities (London: (Simmie and Sennett 2001), and sub-urban regions (Rantisi 2002; Asheim and Coenen 2005).
Theory and empirical work has moved away from thinking of regions as regional innovations systems or not, to comprehending RIS as existing in all regions, with distinctions being drawn between weak and strong, according to their possession of certain regional assets. This represents the second iteration.
It is true, however, that this is not a settled position. Debate still continues within the literature on the qualifications required for regions to be considered an RIS (Vilanova and Leydesdorff 2001; Vilanova and Leydesdorff forthcoming), and whether in fact many regions should not be called an RIS at all. The question that arises from this is: if it is acknowledged that innovative and knowledge activities are the drivers of economic growth, and they are both socially embedded processes, then it must follow that all innovative activity would be contained within some system. These activities must be affected on some level by the same socially embedded processes that lead agglomerations of people to identify themselves as belonging to a region. Therefore, if we re-examine the definition of a system as a “set of arrangements of things so related and connected as to form a unity or organic whole” (Carlsson, Jacobsson et al. 2002 p.233), and the three elements of system analysis (examining innovation through interactivity between a set of actors, a focus on learning and knowledge as an economic process, and a close and dynamic relationship with policy development), then it can be seen that most regions, no matter their size or
Chapter 2 – Innovation and knowledge activities in regions 41 economic success, should be available for some level of analysis within the RIS approach.
2.2.2 Identifying actors, interactions and their attributes
Although variety exists in the definition of RIS, there also exists some consistency, notably in the definitions and interactions of the key elements of the RIS firms, knowledge institutions and organisations. The main question for regional innovation analysis is how these conceptual elements manifest into an empirical framework.
There are a number of examples of such transformations within the RIS literature. Each highlights particular elements or characteristics of RIS, and provides differing theoretical linkage points between concepts and empirical valuables. The main proponents of each of these typologies of RIS elements are provided in Table 2.1.
Table 2.1 Defining elements of RIS Typology authors Defining characteristic Elements
(Cooke and Morgan Governance dimension – 1. Grassroots 1998; Cooke 1999; organisation and 2. Network Cooke 2001; Cooke orientation of knowledge 3. Dirigiste 2004)( 2001), Cooke et flows al (2004) Cooke (Cooke and Business innovation 1. Localist Morgan 1998; Cooke dimension 2. Interactive 1999; Cooke 2001; 3. Regionalised national system of innovation Cooke 2004)( 2001), Cooke et al (2004) (Asheim and Isaksen Location of knowledge 1. Territorially embedded 1996; Asheim 2000; organisations & 2. Regionally networked Isaksen 2001; Asheim, character of knowledge 3. Regionalised national system of innovation Coenen et al. 2003) flows (Asheim and Coenen Knowledge base 1. Analytical knowledge base 2005; Asheim and 2. Synthetic knowledge base Coenen 2006)
Cooke and various collaborators have been key figures in establishing the concept framework for regional innovation systems (Cooke and Morgan 1998; Cooke 1999; Cooke 2001; Cooke 2002; Leydesdorff 2002; Cooke, Heidenreich et al. 2004; Cooke and Leydesdorff 2006; De Laurentis 2006), by creating a framework for determining system attributes. Cooke et al note two dimensions to
Chapter 2 – Innovation and knowledge activities in regions 42 an RIS: a governance dimension; and a business innovation dimension (Cooke 2004).
The governance dimension refers not so much to the government levels that are present in the regions, although these levels and associated agencies are important, but to the organisation and orientation of knowledge flows in a region. Cooke identifies three types of systems: Grassroots; Network; and Dirigiste. In the Grassroots RIS, the mode of the system is locally organised. By this it is meant that key aspects of innovative activity such as funding, and research and development, are co-ordinated at the local level. As a result, research expertise is ‘applied’ and related directly to problem solving. In comparison, in the Network RIS the mode of the system is multi-level, with key resources for funding and research and development potentially coming from local, regional, national, or even international sources. Research competence is mixed between highly applied and pure research. The focus of this type of system is on co-ordination between multiple levels of organisation. In the Dirigiste RIS, innovative activities are largely organised from outside of the region, either as the result of national government policy, or in the case of a major multinational firm. The research competence is highly specialised and the focus of the system is on developing and maintaining that specialisation.
The second dimension Cooke et al discuss, is the business innovation dimension. This dimension considers the nature of innovation within the region and nature of the activities of the firms. Again there are three types of RIS identified: Localist; Interactive; and Globalised. The Localist RIS is dominated by neither small nor large firms, but has a focus on endogenous activity, and a relatively local research reach. The Interactive RIS also has a balance of small and large firms, but includes a mix of external and internal activity, both from firms and public research institutes. The Globalised RIS is dominated by global corporations. The research reach of the system is global, and dominated by private as opposed to public knowledge organisations.
Chapter 2 – Innovation and knowledge activities in regions 43 Both of these dimensions classify RIS based on 5-6 measurement elements. In the case of the governance dimension, it is the technology transfer activity, research funding and competence, level of technical specialisation and co- ordination. In the business innovation dimension, elements are the research reach, and innovation, and R&D resources of regions and the ‘associationalism’ of regional actors. In analysing a broader range of regions, particularly less advanced and developed regions, these elements may not be present. However, the overall principle of these dimensions is to capture information about the types of innovative activities undertaken by firms and the orientation of regions defined through firms’ interactivity both within and beyond the region.
The typology developed by Ashiem and Isaksen (1996) is similar in many ways to Cooke et al’s governance dimension in that the territorially embedded RIS is locally focused, the regional network RIS co-ordinates a mix of local and regional, national and international sources of knowledge, and the regionalised national innovation system is completely dominated by external forces in terms of the supply of knowledge. This typology, however, has a finer focus on institutions and the location of these institutions.
Another typology of RIS is focused on knowledge base. The earlier definition of RIS highlighted that the actors and interactions contained by a region sit within an established and cumulative base of knowledge. Ashiem and Coenen (2006) define two types of knowledge base: ‘analytical’ (science based); and ‘synthetic’ (engineering based). They note that the processes of knowledge generation and transfer, particularly in relation to tacit knowledge, differ between the two bases of knowledge. Therefore, a region’s industrial focus will affect the knowledge base and hence the knowledge gathering and innovation activities that are carried out in that region. For this reason, regions are classified on the basis of their knowledge base.
Each of these typologies emphasises different aspects of knowledge gathering and innovation processes within regions. Each is important in the analysis of regional innovation. The key elements are presented in Table 2.2. The four
Chapter 2 – Innovation and knowledge activities in regions 44 elements are i) Industrial knowledge base, ii) Business innovation, iii) Regional knowledge flows, iv) Regional knowledge orientation. The framework allows for a detailed analysis of knowledge in individual segments; sources, processes of transmission, type: codified or tacit, but also a unified analysis of these activities in firms within the three regions under investigation. The framework allows an integrated view of knowledge and innovation to emerge. Therefore heeding Douloureux and Parto’s call for an integrated approach to RIS analysis (2005).
This integrated approach aims to achieve two things. Firstly, by taking the overriding principles of previous RIS research (much of which has focused on 1st iteration regions), the approach become more applicable to a broader range of regions. Secondly, by combining, the elements offer a more complete and in- depth analysis of regional innovative behaviour.
Table 2.2 Key elements of integrated RIS investigation RIS elements Defining characteristic Derived from
Industrial Knowledge Industry base and key Ashiem and Coenen (2006) Knowledge bases base knowledge bases Business innovation Firm innovation activity Cooke et al (2004) Business Innovation dimension Regional knowledge Types and sources of Cooke et al (2004) Governance dimension flows knowledge accessed Ashiem & Isaksen (1996) Regional orientation Identification and Cooke et al (2004) Business Innovation location of key dimension Cooke et al (2004) Governance knowledge organisations Dimension Ashiem and Isaksen (1996) and the orientation and reach of regional firms
2.5 Summary
This chapter has outlined the literature on innovation and knowledge activities in regions. The literature proves that the analysis of innovation processes has become central to the understanding of economic development.
This review has shown how the activities of innovation and knowledge are linked, and how spatial processes present in regions affect the types of knowledge created and transferred, and the mechanisms for these. The link
Chapter 2 – Innovation and knowledge activities in regions 45 between innovation and knowledge is at the heart of the systems approach. However as the review of the RIS literature has shown, the analysis has not taken place in an integrated way. The conceptual framework presented in the concluding part of this review brings together the different strands of analysis in RIS and combines them into the one framework. The vehicle for this integration is the KISA methodology.
The review has also shown the two prominent iterations of the RIS concept within the literature. These two iterations have shaped the empirical research and policy recommendations emerging from the RIS approach. The first iteration identifies RIS in ideal ‘exemplar’ regions, with Silicon Valley being the most obvious example. Within this iteration regions are either are or are not RIS. The second iteration takes the position that RIS are in existence in all regions and regional innovation activities range on a scale from weak to strong. Although this is the most policy relevant research framework to work from – as most regional policy is targeted at strengthening regional innovation capacity, the framework is the more under-developed of the two. In this respect research will also contribute to further empirical development of this iteration through the identification of “key indicators…in regions with regard to innovative capability” (Doloreux and Parto 2005; p17). This has been identified as a necessary development within the RIS literature in order for the RIS approach to maintain relevance (Cooke and Memedovic 2003).
Chapter 2 – Innovation and knowledge activities in regions 46
Chapter 3 - Research design and methods
The preceding chapters have introduced and positioned this research within the literature using a Regional Innovation Systems (RIS) approach, and with the dual aims of understanding the theoretical foundations and implications of RIS for regional economic development. This chapter brings together the conceptual framework established in Chapter Two into operation.
The literature review presented in the previous chapter also established the crucial element of knowledge in the innovative capacity of firms and highlighted how the treatment of knowledge within the RIS approach is not integrated. This research steps towards providing an integrated framework by bringing together the four established elements of RIS and linking them to the KISA methodology. The resulting synthesis will provide an understanding of the innovation capacity of three regions in Outer Western Sydney.
This chapter outlines the research design and research methods employed in the investigation. As De Vaus (2001) notes, when approaching any research problem social scientists ask two fundamental questions:; firstly, what is going on (descriptive research); and secondly why is it going on (explanatory research). Increasingly, especially in applied research, a third can be added: what can be done about this (policy research). It is from these three basic questions that the empirical research design of this investigation has been formed.
Chapter 3 – Research design and methods 47
Although the prominence of the region as a unit of analysis has certainly grown in recent times, any data beyond basic demographics at the regional level in Australia has not become more readily available. Therefore, a component of this research will be descriptive by necessity, identifying what is currently occurring, and establishing a platform of activities for further analysis. This is where the ‘why’ questions will be asked: why this is so; and why do firms carry out these specific activities? This is also where the RIS framework will be tested for its explanatory power. Finally, consideration of policy, principally at the local governance level, will also be made.
3.1. Selection of Research Design
Whilst South West Sydney is the primary research subject, particularly for any resultant local government policy, a comparative methodology is necessary for contextual and empirical validity. Two adjacent Western Sydney regions were selected to provide the comparison: Central West Sydney; and North West Sydney (please see maps on pages in Chapter One). In selecting the cases for comparison, which are outlined in more detail within the next section, the need to preserve the territoriality of these regions, which are located adjacently to each other in metropolitan Sydney is important, so the comparative regions are analysed in and of themselves, but not at the expense of linkages between the regions, and their role together as part of metropolitan Sydney.
Overall, the research design is cross-sectional, using survey questionnaire and interview methods to collect data for the analysis. The cross-sectional research design in an explicit sense is a somewhat under-utilised research design, but much of the research in regional/economic geography utilises aspects of the design in an implicit sense for comparative work, without the clear emphasis on a structured data set. Cross-sectional design allows for the analysis of variables based on established group differences. Data is collected at one point in time and used to examine the extent to which deviations in various categories of the
Chapter 3 – Research design and methods 48 independent variable differ in relation to the dependent variable (De Vaus 2001). In this study, independent variables are: region; firm demographics, such as size and industry; and the types and novelties of innovations introduced by firms. Dependent variables will be: the level, type, and mix of knowledge gathering and innovation activities conducted by firms; the location of these activities; and the regional experience of the firm. The main emphasis on data collection in this research design is to provide a structured data set over the groups to be analysed. The design provides no tendency towards the data collection methods to be used. The advantage of cross-sectional analysis is that it allows for the three research components discussed earlier – descriptive, explanatory, and policy - to be addressed. A common criticism of research into regional innovative activity is that most previous empirical work is based on the case study method, which although offering depth and richness when explaining the cases at hand, provides minimal opportunity for generalisation (Markusen 2003). This is compounded by the fact that most of these case studies are based on success stories with particularly unique circumstances, in line with the first iteration of RIS (Wiig and Wood 1995). The selection of the case study method is understandable given the complexity of innovation processes, and the diversity of actors and elements involved. This also makes purely survey-based methods inadequate for analysis as well. Cross-sectional analysis design offers a form of middle ground, as noted by Lillis and Mundy:
“Studies employing a cross-sectional field study approach draw on a larger number of observations than in-depth case studies, but can also deal with more complex ‘how’ and ‘why’ questions than survey approaches…A cross-sectional field study approach provides researchers with an effective means of capturing complex phenomenon within a confined domain. It also enables them to uncover reasons that might explain conflicting results, ambiguities or tensions in prior research. Elaborated responses gained from data collection techniques such as semi-structured interviews may also highlight previously unhypothesized relations between variables” (Lillis and Mundy 2005)
When considering these factors and the variety of ‘what’, ‘why’, and ‘what can be done about it’ questions that are the focus of this type of research, the cross- sectional design has much to offer due to the structure it provides to data
Chapter 3 – Research design and methods 49 collection and analysis. The lack of available regional data, in conjunction with time and financial constraints, limits data collection to one point in time. However, the data collected will provide enough observations to answer the descriptive component, as well as offering some conclusions on policy direction, and all within the comparative environment. The following diagram outlines the data collection methods used.
Diagram 3.1 Research methodology
As previously stated, the cross-sectional research design does not naturally tend more towards any particular data collection method. Rather, the consistent structure of the data collected is the core feature of the design. This research employs three phases of data collection: statistical audit (gathered from official government statistical sources); survey; and in-depth interviews. The three stages of data collection offer a range of data on socio-economic and industrial conditions within the regions, innovation and knowledge gathering activities, and in-depth responses regarding firm’s strategies for innovation. The various data collection methods also offer the opportunity for triangulation of results and
Chapter 3 – Research design and methods 50 conclusions. The specifics of each of the data collection methods are discussed further in Section 3.3.
3.1.1 Limitations of research design
There are two dimensions that can affect the validity of a research design: internal validity - the ability the research has to support the causal claims it makes; and external validity - the extent of the research’s ability to be generalised.
As outlined, cross-sectional designs have three distinct features: • No time dimension; • Comparisons between groups based on existing differences and • Groups specifically selected rather than randomly allocated (De Vaus 2001)
Cross-sectional analysis presents a unique situation. The lack of a time dimension means that analysis is based on existing differences rather than change over time. Therefore, the primary threat to internal validity is the ability to establish causality. The absence of a time dimension prevents establishment of a sequence of events. A secondary issue is that of meaning. Cross-sectional analysis relies on existing differences in the selected groups to drive the analysis. In this case, the regions have been particularly selected rather than randomly allocated, meaning there is less room for accepting a causal relationship, wherein an effect on the dependent variable is assumed to be caused by the independent variable, but in this case may actually be the result of some other uncontrolled- for factor.
Two clear problems to the internal validity of cross-sectional analysis are therefore highlighted. Firstly, establishing causality without a time dimension, and secondly, the problem of ‘meaning’, which suggests that specific research results have ‘meaning’ within a specific context. Neither of these problems are
Chapter 3 – Research design and methods 51 unique to this design, but both have specific weight in cross-sectional analysis (Marsh 1982).
The first issue of establishing causality, the lack of a time dimension and sequencing of events, is dealt with through statistical controls and the type of statistical analysis used. The purpose of the statistical controls is to make the groups as much like each other as possible and then work through and eliminate, through controls, any confounding variables (De Vaus 2001). As a result, we can be more confident of a relationship between the independent and dependent variable, because we have controlled, as far as possible, any other identified confounding variables. The statistical analysis used in this research is based on correlations and methods of establishing association. This in itself does not imply causality. If X is correlated with Y, it does not mean there is a causal relationships between X and Y. However, it does mean that a causal relationship cannot be ruled out, and in addition, if there is no correlation, then we can rule out a causal relationship (Marsh 1982). “Eliminating variables as causes can be of just as much importance as locating causes” (De Vaus 2001 p.179). Therefore, the value of this research will be in establishing where causality may lay, as attributed by significant correlations, and highlighting where it does not, when no significant correlations are present. The establishment of causality in this sense must be left to further research.
Correlations can be further strengthened by creating a priori reasoning to propose a correlation. A priori reasoning offers outcomes based on theoretical and previous empirical research (De Vaus 2001). A priori models of themselves do not provide proof of cause, but they do provide a theoretical argument with supporting empirical evidence for a particular causal relationship, which can in turn be tested at a later time. To enable this, in addition to research questions, a number of hypotheses are also nominated, to focus the research project.
The second problem of internal validity has to do with expressing meaning, such that correlating variables is worthless if some meaning is not attached to these correlations. On the other hand, variables of themselves are seen as inadequate to
Chapter 3 – Research design and methods 52 capture the complexity and depth of human action and intention, and the context of that action. That is, in dealing with elements of ‘bits’ (variables), the ‘whole’ is missed (Blumer 1956). Methods used to mediate this dilemma and introduce intentional and contextual dimensions into the analysis include asking the subject about their meaning and intention. Subjects will not always be able to provide an answer, and when they do it will be limited to knowledge of their own experiences. As a result, both data collection methods involving participants (surveys and interviews) are pilot-tested prior to deployment, and issues of ‘concept meaning’ are discussed to ensure congruency. The researcher can also provide meaning through establishing similarities across groups, and arguing an explanation. This requires preparation prior to data collection, so data on the elements that the researcher believes will explain intention and context (based on theoretical framework), is also collected (Marsh 1982). Information from Chapter Two is essential in establishing this theoretical framework and conceptual ‘meaning’, which will later be called upon to explain results.
The questionnaire design also has a critical role to play in identifying potential influencing variables. As noted, the questionnaire used in the business survey is pilot-tested before full deployment, to ensure clarity of meaning of the concepts used and therefore clarity in the data collected. Numerous sections of the survey are based on other innovation research surveys, including the section on firm demographics and innovation, based on the Australian Bureau of Statistics’ (ABS) Innovation in Australian Business (2003), which itself is based on the OECD Oslo Manual (OECD 2005)and the sections of KISA and firm networking were based on the OECD KISA multi-country research project (OECD 2006). This again does not cement universal ‘meaning’ of concepts, but adds to the weight of the argument.
This limitation will also be further mediated through the in-depth interviews of representatives from selected firms, which will not only ask participants of their intention and consciousness of aspects of their activities, but also collect information that relates to intent, context, and relationships that the participant
Chapter 3 – Research design and methods 53 may not necessarily be aware of as influential. These can be analysed by the researcher through the conceptual framework established in Chapter Two.
In terms of external validity, there are no specific threats more prominent in this research design than other designs. In fact, cross-sectional research design by its nature is usually better placed to deal with threats to external validity than other research methods (De Vaus 2001). External validity relates to the extent to which results from the research are representative and able to be generalised. Discussions on the sample and its representativeness are presented in Section 3.3.2 on Questionnaire Field Procedures. However, it is important to bear in mind what needs to be generalised. Research results from regional research are not particularly open to being generalised, due to their context-specific nature. From a theoretical point of view, what needs to be generalised are the concepts, so the focus is on pattern matching in these key concepts, such as KISA usage, type and mix of usage, and organisational networking which adds to theoretical generalisation rather than generalising across empirical results.
3.2 Selection of Research cases
The three regions under investigation are all located within the broader region of Greater Western Sydney. South West Sydney was initially selected for the investigation as part of an Australian Research Council linkage project with a major local government authority in the region – Liverpool City Council6. The two other regions were selected subsequently for comparative analysis.
The Greater Western Sydney7 region is one of the most populous in Australia, with a population in excess of 2 million people, and the third largest economy in Australia after Sydney (total metropolitan area) and Melbourne. Greater Western Sydney is very much a part of the metropolitan ‘global city’ of Sydney, although
6 Liverpool City Council is the industry partner for the Australian Research Council Project (LP 0347917) from which this research is drawn. Liverpool City Council became involved in the research because of a general lack of industrial research of any type focused on the region. Given their current growth challenges, the Council identified the need for this research. 7 Outer Western Sydney is the outer suburban fringe of Greater Western Sydney. Please see maps in Chapter One for detail.
Chapter 3 – Research design and methods 54 Sydney is typically believed to be focused geographically on the inner CBD, harbour, and inner suburbs, and industrially on information industries including finance and business services (Fagan, Dowling et al. 2004). It is because of a unique combination of dynamics, operating at the regional, metropolitan and national level, that the areas of analysis for this comparative study were all selected from the same metropolitan area.
There are a number of crucial features of comparison available between the three regions. Firstly, geographically all three regions have similar distances to the CBD of Sydney, and all have major road and public transport links with the centre of Sydney. Consequently, each of the regions has a similar sense of proximity to the inner city areas of Sydney. Secondly, all areas have experienced significant population growth in the last decade. Both the North West and South West regions have been identified as a greenfields growth corridor in the Sydney Metropolitan Strategy8. Central West Sydney will also see significant change with the greenfields residential development of surplus Australian Defence Force land at St Marys, a suburb in Central West Sydney. Therefore, all three regions have experienced rapid population growth, and the associated challenges of balancing population growth with the provision of sustainable employment and industrial development. Differences however do arise between these three regions in terms of social structure, industry, and occupational structure, which will be discussed further in the following chapter.
In terms of institutional endowment, each of the three regions has a similar base. Each of the regions contains campuses of the University of Western Sydney, each has a major hospital and medical facilities, and in addition South West Sydney and Central West Sydney both have teaching hospitals and research facilities. In terms of governance, the sub-structure of local government councils covers all areas, as well as the state government level through the Office of the Minister of Western Sydney. Recently, Sydney has embarked upon a metropolitan planning process which culminated in the release of the Sydney
8 The major metropolitan planning document guiding Sydney’s growth for the next 30 years (NSW Government 2005)
Chapter 3 – Research design and methods 55 Metropolitan Strategy in 2005 (NSW Government 2005). This strategy provides the blueprint for Sydney growth over the coming thirty years: its population growth; residential and industrial development sites; and public infrastructure spending. The strategy divides Sydney into regions based largely on functional and commuting patterns.
The three regions under investigation broadly follow the Metropolitan Plan in their composition, with slight variations in North West and Central West Sydney. The regions’ boundaries are primarily established through metropolitan planning patterns over the past fifty years, rather than the most current incarnation. The current metropolitan strategy identifies a number of ‘Centres’ that are to be the source of further development in the creation of a poly-centric metropolitan area. A centre in South West Sydney – Liverpool, and Central West Sydney – Penrith, are identified. The Metropolitan Strategy provides the only mechanism for regional level governance. Other aspects of policy concerned with regional economic development are administered at the local government level, or the state government level through the Department of State and Regional Development.
Local policy and economic development strategies, and the success of these policies and strategies, are divergent among the three areas and offer a comparison. The Central West region in particular has followed the path of creating an ‘enabling environment’ in order to attract a certain type of economic actor or activity to their area. The North West region has been highly successful in attracting firms that are loosely referred to as high technology firms, through the development of business parks, especially the Norwest Business Park. This attraction activity, supported by a suitable labour force profile, has led to economic development through new companies locating in the region, bringing high-level employment and innovative activity. South West Sydney has followed a more sectoral approach, with an emphasis on manufacturing, and a targeted program of firm relocations from the inner-city industrial area to the South West.
Chapter 3 – Research design and methods 56 3.3 Field Work
Within the cross-sectional research design there is no preference for the particular research methods to be used; rather the focus is on the structure of the data collected. A mixed method approach, including statistical audit, survey questionnaire, and in-depth interviews, was selected as the best way to collect the data required to answer the research questions. The survey questionnaire undertaken with the available sample provides representative results for the three regions under investigation, and is of significant enough size to allow the application of statistical controls within the analysis. This mediates two of the limitations of the research design, these being the representativeness and consequent ability to generalise results, and it furthermore allows the use of statistical controls to add extra confidence to correlations and potential conclusions established through the analysis.
The in-depth interviews with a smaller number of select firms shed light on the ‘how’ and ‘why’ processes, particularly the degree of regional influence in the innovative activities of firms. The table below outlines the research questions, and how the research methods will be employed to answer them.
Table 3.1 Research questions and corresponding data collection methods Research question Data collection method 1. How does innovative activity manifest in the three regions? Audit/ Questionnaire 2. What kind of knowledge firms access and from where? Questionnaire 3. How much knowledge is external and how much is internal? Questionnaire 4. What are the key organisations for external knowledge sourcing? Questionnaire 5. What is the regional component of knowledge flows? Questionnaire 6. What determines the mix of knowledge that firms obtain for their Questionnaire/ Interviews innovation activities and why? Who drives this process? 7. How and why does ‘regional experience’ affect this process? Interviews
Chapter 3 – Research design and methods 57 3.3.1 Statistical audit
The statistical audit is the first stage of the research methodology. It produces the background context for later empirical work. The audit primarily draws data from the Australian Bureau of Statistics’ (ABS) five-yearly population census (1991, 1996, and 2001). This data is the only readily available source of official statistics for small-area geographic analysis. The results of the audit provide the social profile of the regions (shown in Appendix 6), and the economic and industrial profile outlined in Chapter Four. The establishment of the knowledge base (also discussed in Chapter Four) is on the basis of occupational data also collected as part of this audit.
3.3.2 Questionnaire design
In designing the questionnaire for the business survey, two important aspects had to be considered: the content of the questions; and the method of implementation. There are limited examples within the literature of surveys that are similar in both content and geographic status. This is mainly due to the wide availability of innovation data at the firm and small areas level in America and Europe (with the Community Innovation Survey and various OECD publications). Australia has an innovation survey, the Innovation in Australian Business (ABS 2003) survey, which is collected every three years. However, the results of this survey are not available at levels below that of the state, and in many sub-groups such as industry and business size, results even at this level are variable, sometimes with error rates of up to + or – 50%. Unit record data for this survey is also not available.
The Innovation in Australian Business survey and the European Community Innovation survey are both based on the principles established in the OECD’s Oslo Manual (OECD 2005). These surveys, and the principles of the Oslo Manual, provided a framework for the drafting of the questionnaire, and through comparison of results on key questions, provide opportunity for triangulation of
Chapter 3 – Research design and methods 58 results between the innovation surveys, further strengthening the questionnaire design.
A key point of difference should be noted however, which is that survey questions on firm innovations did not investigate whether innovations were patented, or seek data on sales figures and profits. Consequently, there is no way of quantifying the economic value of these innovations beyond the fact that they were considered important enough for the firms to mention them when surveyed. The limitations of this decision are that the overall economic value of innovations cannot be calculated, and it cannot therefore be ascertained whether the activities associated with highly valuable innovations are the same or different to those that are less economically valuable. The merits of this decision, however, lie in lessening the burden on survey participants. Requiring firms to quantify the value of their innovations (or their determination of the value) would be a considerable task, and therefore the decision was made not to pursue this line of questioning.
It should further be noted that collecting data on sales and profits would not necessarily assist in identifying the ‘value’ of an innovation to a particular firm, because again we would be relying on the firm to quantify value, and there is no avenue to independently appraise this value. Finally, it must be remembered that one of the purposes of the survey is to collect data on innovation activities beyond the actual introduction of innovations, and in keeping with a broader definition of innovation and the ubiquitous nature of innovation (Simmie, 2003), it is the other activities associated with the actual innovations that are of as much importance to understanding the processes of innovation. Hence, a firm being able to answer affirmatively to counts of innovation assists in focussing their attention on other activities such as searching and learning activities. The firm interviews completed subsequently to the survey provided firms with more opportunity to elaborate on their innovative activities, and allowed the link between these activities and firm competitive advantage to be tested. This counter balances the highlighted limitation within the survey.
Chapter 3 – Research design and methods 59 A key component of the survey, and indeed the research, is its focus on collecting data about the usage of Knowledge Intensive Service Activities (KISA) by firms. This is an emerging area of innovation analysis. A recent OECD 5-nation study (OECD 2006) investigating the role of KISA on the innovative activities of firms, found higher levels of innovative activity in firms that used and absorbed higher levels of KISA (OECD 2006). Data collection methods of KISA were available from the literature (Aslesen 2004; Martinez- Fernandez 2006; OECD 2006; Albors, Hervas et al. Forthcoming), and provided a methodological and empirically validated base for analysis in this research. In particular, KISA related questions have been previously applied in the Australian context in innovation surveys of the software, tourism, and mining technology services industries (Aslesen 2004; Martinez-Fernandez 2006; Martinez- Fernandez and Krishna 2006; Martinez-Fernandez and Miles 2006). This ensures the KISA questionnaire methodology has a strong empirical foundation in both nationally and internationally validated surveys.
As highlighted in Chapter Two, another key component of RIS is the interactivity among system actors. Within the RIS literature, there is a strong focus on the role of supply chain networks and universities (Miller 2000; Martinez-Fernandez and Bjorkli 2003; Cooke, Kaufmann et al. 2006), however there is a whole range of other institutions that contribute to firms’ knowledge gathering activities, including business and industry associations, and government agencies. This survey aims to capture some of these business networking activities, and consequently a number of questions within the survey are targeted at capturing this data. A roster roll method is used, similar to that used by Vale and Caldeira (2007), where firms are provided with a list of public, industrial and other institutions for their assessment. Particular attention is paid to regional institutions. Space is also provided for firms to list any institutions that were not included in the roster roll. A copy of the information sheet and questionnaire used can be found in Appendices 1 and 2. The information sheet outlines the project for participants, and also details the ethics clearance and conditions for the project.
Chapter 3 – Research design and methods 60
The first version of the questionnaire was pilot-tested at a state and local government sponsored Small Business Exhibition. This exhibition was selected because it provided access to a range of different businesses, and also enabled the questionnaires to be completed face-to-face directly with the participants. This provided the opportunity for discussion between the author and participants in establishing convergence of meaning for key terms, whether the general flow of questions was successful, and the effect of the questionnaire burden on the participants. Sixteen pilot questionnaires were completed. As a result of the pilot- testing, in the final questionnaire more explanation is given on some key concepts and the order of questions was restructured. The original survey responses are not included in the analysis presented in this thesis due to these changes, and the consequent inability to ensure strict comparability with the updated questionnaire.
The second aspect to be resolved relating to the survey was the method of dissemination. Traditionally, the literature notes poor response rates to mail-out surveys, with better results achieved in either telephone assisted completion or face-to-face completion (Groves 2004). Financial and time constraints, together with the sample size required, meant these methods were not appropriate. The chosen method of dissemination was a web-based questionnaire, with an electronic link to the questionnaire delivered to participants by email9.
As explained in further detail in the following section, the sample for the survey was comprised of firms drawn from business associations’ memberships, and local government business contact lists. For the majority of the sample, email addresses of key personnel were available. Email accessibility was of critical importance to the web-delivery model. This delivery method (emailed link to web-based questionnaire) has shown higher response rates in other empirical research (Groves 2002; Groves 2004), and in the case of this research, provided
9 The survey web platform used was SURVEYMONKEY ™
Chapter 3 – Research design and methods 61 the opportunity to, firstly, deliver the questionnaire to its target, and secondly, minimise the impact of gatekeepers, both of which are noted problems with other dissemination methods. However, there were some companies who did not nominate an email address or did not permit email contact. These companies were sent the questionnaire by mail and were invited to return the completed questionnaire by mail or facsimile. They also received the details of the web link, and therefore the opportunity to complete the questionnaire online. The next section details the field procedures undertaken to deploy the questionnaire.
3.3.3 Questionnaire field procedures
The firm survey was undertaken between October 2005 and January 2006. 474 firms were approached to complete the survey by either email or mail, based upon a sample of business association memberships and Liverpool City Council’s contact database. All firms on all of these lists were provided with the opportunity to participate in the survey. 154 surveys were returned, 35 of these surveys were removed from the analysis because the respondent firms were located outside of the study area, or the returned survey was too incomplete to allow proper analysis. Therefore, 119 survey responses are available for analysis. This represents a response rate of 25.1%, which as a response rate is consistent and comparative with other similar business surveys (Groves 2004).
As noted earlier, the survey was administered in two ways, either via email or mail. With those for whom email addresses were available, which was the majority of the sample, the survey was administered via an online website. Firms were invited to participate through an email that contained an explanation of the study, the reasons for the survey, and a link to the survey website. For firms where email addresses where not available, or the firm declined to be contacted by email, paper based surveys were sent via the post, with respondents able to return the completed surveys by mail or facsimile. 93 survey were returned electronically, and 26 by facsimile or mail.
Chapter 3 – Research design and methods 62 About the sample The sample of 476 firms was drawn from a number of accessible firm lists, but principally from business membership lists of Australia’s two peak business associations – Australian Business Limited (ABL), and Australian Industry Group (AIG). Memberships were sorted by firm address, and those firms located within the eight local government areas covered by the study were selected to be part of the sample. These two lists were supplemented by business contact lists supplied by the individual local government authorities. Steps were taken to identify multiple listings of firms on two or more of the membership lists. Firms that appeared on one list were removed from subsequent ones, so that they were only contacted once.
This combination of samples was selected for a number of reasons, including the reliability of the details. Companies in the sample were either paying membership dues to business associations, or had recently participated in or registered for business development activities with a local council, so there was reasonable confidence that the businesses were still in operation. Secondly, this combination of samples was seen as the best way to obtain a representative cross-section of the regional business community, with the larger, more prominent firms captured through memberships of the peak business associations, and the smaller, more locally based businesses accessed via the local government contact lists. The following table outlines the breakdown of the sample by the three types of list which were utilised.
Table 3.2 Survey sample by list List Count ABL 168 AIG 192 Local government lists 116 Total 476
Table 3.3 presents the breakdown of the sample by broad industry category. The following industry categories are not included in the sample: electricity, gas and water supply; government administration and defence; education, health and
Chapter 3 – Research design and methods 63 community services; and personal services. This follows the example of other innovation surveys including the ABS’s Innovation in Australian Business Survey (2003), where these sectors were excluded because they are dominated largely by public sector organisations, and therefore have different dynamics in terms of what motivates innovative activity than do private sector firms (i.e. profit). The sample is dominated by manufacturing and business services firms, and half of the sample is from South West Sydney.
All of the three regions under analysis have significant manufacturing employment bases,[m1] therefore the concentration of manufacturing firms is not surprising, and shows the consistency of the survey sample with the actual firm population in the regions. Furthermore, the purpose of this research is to establish the level and type of innovative activity in these regions, and other studies have consistently shown that the manufacturing industry drives economic activity. In the Innovation in Australian Business Survey (2003), manufacturing is the leading industry in terms of both the proportion of businesses innovating, and contribution to expenditure on innovation. The dominance of manufacturing firms in the sample is therefore not a cause for concern.
The finance and business services category includes the following industries: property and business services; financial and insurance services; and communication services. These industries contain KIBS or Knowledge Intensive Business Services, which are significant in terms of KISA provision, as highlighted earlier, and which are associated with the production of KISA, another key source of innovative activity. The three regions also have high levels of these firms in their total firm population, again ensuring the consistency of the sample with the actual population. Further details of industrial composition are provided in Chapter Four.
Chapter 3 – Research design and methods 64 Table 3.3 Survey sample by region and broad industry grouping
Count of List Region Central West North West South West Industry Sydney Sydney Sydney Grand Total Agriculture 0 0 5 5 Construction 5 4 16 25 Manufacturing 77 54 169 300 Mining 0 0 2 2 Retail Trade 1 5 7 13 Business Services 17 21 45 83 Transport & Storage 4 0 3 7 Wholesale Trade 13 13 15 41 Sample total 117 97 262 476 Total respondents 28 26 65 119 Total no. firms * 35520 26319 33196 95 035 * This is total number of Australian Business Number registrations per region. The limitations and overestimations of this data are further explained in Chapter Four. However this is the only available data at the firm level.
Respondent demographics A response rate of 25.1% was achieved for the survey. Chart 3.1 shows the spread of the survey respondents over the three geographical regions. 28 firms or 23.5% responded from Central West Sydney; 26 firms or 21.8% responded from North West Sydney; and 65 firms or 54.6% responded from South West Sydney. The response rates for each region were also around 25% of each region’s sample: 25% for Central West Sydney; 26.8% for North West Sydney; and 24.5% for South West Sydney. The bias in the sample towards South West Sydney firms was expected because of the prominence of the research project in South West Sydney10, and the policy direction of the research in South West Sydney. The cross-sectional research design focuses on the structure of the data, so although sample composition in terms of overall numbers is not identical across the three regions, structurally they are the same. South West Sydney has twice the number of firms in the sample compared with North West and Central West Sydney, but response rates for the three regions are very similar, thereby preserving the cross-sectional structure.
10 This research is part of an Australian Research Council Linkage project (LP 0347917) with Liverpool City Council, a major local government authority in South West Sydney. The research has been regularly featured in the local media.
Chapter 3 – Research design and methods 65 Chart 3.1 Respondent firms by region
70
60
50
40
30
20
10
0 Central West Sydney North West Sydney South West Sydney
Source: Outer Western Sydney Business Innovation Survey n=119
Chart 3.2 shows the industrial breakdown of the respondent firms. There is a strong representation of manufacturing and finance and business services firms within the respondents, consistent with the earlier discussions of the sample, as displayed by Chart 3.3.
Chart 3.2 Industry of respondent firms by region
35
30
25
20
15
10
5
0 Construction Manufacturing Other Services Transport & Distribution
Central West Sydney North West Sydney South West Sydney Outer Western Sydney Business Innovation Survey n=119
Chapter 3 – Research design and methods 66 Chart 3.3 Industry of survey sample and survey response firms
70.00%
60.00%
50.00%
40.00%
30.00%
20.00%
10.00%
0.00% Agriculture Construction Manufacturing Mining Retail Trade Business Transport & Wholesale Other, Not Services Storage Trade properly defined Sample % Response % Source: Outer Western Sydney Business Innovation Survey, n=119
The response rate achieved in this survey is comparable to other response rates from online surveys (De Vaus 2001). However, despite this response rate and the even spread of respondents from the three regions comparative to their overall contribution to the overall sample of the survey, care must be taken in interpreting the results from a relatively small-sample survey. The limitations of the research design in terms of external validity have already been discusses, but in this case they pose a further issue to external validity. This sample is not statistically significant for populations of firms beyond these regions. The questionnaire responses and associated analysis are providing indicative findings. These findings will have to be tested on larger samples in the same rigorous manner before firm conclusions can be drawn. This being said the work must begin somewhere. The questionnaire is seen as the beginning of empirical data collection rather than the total or end of it.
3.3.4 Interview field procedures
The in-depth interviews are the third stage of the data collection. The focus of the in-depth interviews is on: how firms decide upon the mix of knowledge they access for their innovative activities; the interplay between internal and external sources; the individuals involved in accessing and managing these resources; and
Chapter 3 – Research design and methods 67 the effect of regional experience on these selection mechanisms. The research questions used and the corresponding data collection methods are as originally outlined in Table 3.1.
From preliminary analysis of the questionnaire results, four themes were developed, around which the interview questions were devised. The four themes are outlined as follows.
Firm strategy for KISA and knowledge sourcing The use of KISA was identified as a key area of difference between firms in the three regions, and therefore an important theme for the interviews was identifying firms’ strategies to access and utilise KISA, and the types and location of sources they used. Interview questions addressed aspects of decision- making relating to these activities, including firms’ strategies for searching out and learning behaviours, and distinction between routine and exceptional aspects of KISA in innovation and learning processes.
The role of knowledge professionals This theme explores the role of individual knowledge professionals in the learning and searching activities of firms. Questions on the role of key knowledge professionals within firms, and how firms’ innovative activities relate to recruitment and growth within the firm, were included in the interviews.
Firm knowledge networks This theme relates to the external knowledge of firms, which was identified as another key point of variance across the three regions. This theme is closely related to the first two themes, as it deals with aspects of firms’ external knowledge networks. In the interviews, the context of specific firm networks are explored, as well as how much firms are embedded in the region, and how they see their regional experience shaping these networks.
Local government and innovation policy
Chapter 3 – Research design and methods 68 The aim for the final outcome of this research is to deliver an understanding of the role that local government can play in the encouragement of innovative activity in firms within their locality. The survey and other case study themes will identify areas of intersection between local/regional resources and the innovative activity of firms in these geographies. This fourth theme will allow firms to discuss how they see industry and government relating, particularly local government in the local/regional innovation system.
The full list of interview questions and the related interview protocol are available in Appendices 3 and 4. The interview questions guided discussions with participants, and provided the structure for the cross-sectional analysis.
3.3.5 Rationale for selection of firms for in-depth interviews
A criteria table (see Appendix 5) was developed to assist in selection of appropriate firms to participate in the in-depth interview section of the research. At the conclusion of the business innovation survey, firms were provided with an option to nominate their willingness to participate in the next phase of the research. The criteria table includes a list of those firms that elected to be part of the next round of research. To increase validity and reduce the effect of any bias from the survey, a number of additional firms were also randomly selected from previous Western Sydney Industry Business Award nominees11.
The criteria table identified the following characteristics:
• Regional location of firm;
• Broad industry category of firm;
• Business size of firm - based on employment; and
• Networking activities and fields – business, government and educational.
11 The Western Sydney Industry Business Awards is an annual award program run by the Minister for Western Sydney. The award focuses on business innovation achievement in many categories of business. Nomination requires the completion of an extensive application process.
Chapter 3 – Research design and methods 69 Firms that were selected through the criteria matrix provided adequate comparability across the three regions and various industry categories. Cases were also selected to provide theoretical replication across the key concepts and hypotheses. A pilot-interview was conducted with a business services firm from South West Sydney, to test the questions again for congruence of meaning and general flow. Some minor adjustments were then made before the interviews were conducted in October and November 2006.
Firms selected through the use of the criteria matrix for the interview phase were initially telephoned or, if they were also involved in the Business Innovation survey, initial contact was by email. The interview target was either the business owner (in the case of small businesses) or the general manager, or their nominated representative for larger firms. After securing the participant’s agreement to proceed with the interview, a copy of the interview questions and a general overview of the research were forwarded to the participants. The interviews were recorded digitally with the participant’s permission. Some participants requested certain aspects of their interview remain confidential, and in addition, names and product details (where they could identify the firm interviewed) have also been suppressed in both the interview transcripts and any quotes which were used in the interview analysis. The next section provides detail on how the collected data was analysed.
3.4. Data Analysis
The data analysis of the survey and interview components of the research is reported in three chapters: Chapters Five, Six, and Seven. The audit component is reported in Chapter Four. The results presented in each chapter are summarised in Table 3.4. Each chapter addresses specific research questions and the resulting associated hypotheses.
Chapter 3 – Research design and methods 70 Table 3.4 Research questions and corresponding data collection methods Research question Data analysis methods Chapter How does innovation manifest in these regions? Statistical audit and statistical Chapters 4 and 5 analysis What knowledge do firms access, how and from Statistical analysis Chapter 5 where? What knowledge is internal and what is external? Statistical analysis Chapter 5 What are the key organisations for external Statistical analysis Chapter 6 knowledge sourcing for firms? What is the regional component of knowledge Statistical analysis Chapter 6 flows? What determines the mix of knowledge firms obtain Statistical analysis and Chapter 7 for their innovation activities and why? Who drives qualitative analysis of this process? interviews How and why does regional experience affect this Qualitative analysis Chapter 7 process?
In combination, these chapters provide evidence of: • The spread of innovative and knowledge gathering activities in these regions, including the usage and mix of KISA; • The type and location of external sources of firm knowledge; and • How these are both related to the availability of regional resources.
Finally, based upon these results, conclusions can be drawn about the role of local government and local economic development policy in supporting and encouraging innovation activities in firms within the regions studied. The methods of analysis used to interpret the research results in order to draw these conclusions are outlined briefly in the next sections.
3.4.1 Statistical Analysis
The statistical analysis has two purposes, the first of which is to describe the level and type of innovation and knowledge gathering activities undertaken by firms. The second purpose lies within the statistical analysis’ exploration of patterns of variance in these activities, based on the region and other factors such as firm size, and industry. The statistical methods used are: factor analysis using principal components analysis as the extraction method, and hierarchical cluster analysis, both of which used the statistical package SPSS.
Chapter 3 – Research design and methods 71
The main application of factor analysis is to reduce data through the identification of underlying, latent variables. However, factor analysis also allows for an investigation of the structure of these underlying relationships, through clusters analysis. Kerlinger (1979) describes factor analysis as “one of the most powerful methods for reducing variable complexity to greater simplicity” (p180). Simplicity is essential for a clear interpretation. Meaningful insights into data are obtained by ensuring that the maximum amount of data from original variables measured is reduced into as few “latent or synthetic variables or factors as possible, so as to keep the solution parsimonious” (Marsh 1988 p.4).
3.4.2 Qualitative Analysis
The interviews will be analysed using a qualitative framework of analysis based on pattern matching and case replication. The pattern matching technique seeks to match a particular pattern in the interview data with a predicted one. The case replication (where the pattern in replicated in a number of cases) offers further evidence in support of the matched pattern (Yin 2003). The four themes discussed previously and in the interview protocol document in the appendices are: firm strategy for KISA and knowledge sourcing; the role of knowledge professionals; firm knowledge networks; and local government and innovation policy. The qualitative data will seek to reinforce and explain results presented form the survey.
3.5. Summary
This research employs a cross-sectional design, with statistical audit, survey questionnaire, and in-depth interviews providing the data collection methods. The cross-sectional design compares structured data sets on selected cases; in this instance, three regions of Western Sydney – South West, Central West and North West Sydney. These regions were selected because of similar geographical
Chapter 3 – Research design and methods 72 and institutional bases and experience with large levels of population growth, yet still having divergent economic and employment development. The three regions thus offer contextual comparability within the territoriality of metropolitan Sydney over the previous decade, between 1991 and 2001.
The research methodology was three-part, thereby offering opportunity for triangulation of results between methods. The first stage involved analysis of government statistical sources, to build the socio-economic and industrial backgrounds of the three regions. The second stage was the development and deployment of a survey questionnaire. This questionnaire was developed from previous national innovation surveys and prior research using the KISA analysis obtained from local and OECD KISA surveys. Questions specifically designed to enable institutional analysis of firm networking activities were also included, and the questionnaire was piloted before distribution. The survey was deployed using a web-based dissemination method wherever possible, and a mail delivery model as the alternative method. 154 responses were received, 35 of which were excluded from further analysis because the respondent firms were located outside of the study area. A total of 119 responses were available for analysis, which equated to a response rate of 25.1%.
The in-depth interviews followed in the third and final stage of the survey research. The questionnaire was designed to answer ‘what’ and ‘how much’ questions, and establish either variance or correlation among key variables, whereas the interviews were aimed at ‘why’ and ‘how’ questions, and were designed to provide further evidence and context for both theoretical and empirical conclusions. A pilot-interview was conducted, and after minor amendments, fourteen in-depth interviews were conducted with selected firms throughout the three regions. Firms were selected using a matrix that allowed for adequate comparability across regions and industry categories, and which provided theoretical replication across the key concepts and hypotheses.
Chapter 3 – Research design and methods 73
Chapter 4 – Economic and Industrial Audit of Outer Western Sydney
This chapter provides an audit of available official statistics on the three regions being studied. The chapter builds a socio-economic and industrial profile of the regions under investigation. As summarised in Chapter Two, the literature describes regional innovation systems as being the innovation and institutional base within the regional production structure. Therefore, the following research includes a detailed analysis of the three regions’ production structure, as well as detailing population, labour force, and workforce changes in the decade between 1991 and 2001.
This chapter shows the coverage of official statistics in regards to characteristics of socio-economic status of the population and industry and occupations of the labour force, both resident and place of work. It also shows the limitations of these data sources in fully expressing the innovation and knowledge activities occurring in each of the three regions. This further highlights the necessity of the business survey within the research methodology.
The three regions have been selected in an aspatial sense, although the regional boundaries are based on labour force commuting and market patterns. The geographical activity and socio-political boundaries of regions are never going to be a clean fit. Policy recommendations for regional economic development are the aims of
Chapter 4 – Economic and Industrial Audit 74 this work, therefore some flexibility needs to be allowed in making ‘policy relevant’, or in other words ‘polity boundary’ relevant recommendations rather than apply functional geographical boundaries, which would be hard to establish form the limited regional data available.
As was established earlier, the analysis of innovative activity, and the processes and mechanisms involved in such activities, are increasingly being interpreted through a regional lens. This is due to the increasing recognition that innovation and its associated processes are spread in a geographically uneven manner, concentrating in high levels in some areas yet hardly present at all in other areas. Regional analysis is difficult mainly because of the lack of reliable data available for analysis at this geographic level. Many of the current measures associated with innovation, such as research and development (R&D) expenditure, and patent counts, are not available at these smaller geographical aggregations, particularly in Australia, and even if they were, they would not adequately capture the full range of innovative activities undertaken within an innovation system.
The current situation is such that while on the one hand the realisation exists that the processes of knowledge and innovation are critical to economic development, and that these processes are now largely determined at a smaller geographical scale than the national level, on the other hand, measures and proxies that can adequately account for these activities on smaller geographical scales are few and far between. Thus, while it is acknowledged that regions are a more appropriate unit for analysis and implementation, the techniques to assess these smaller levels are not readily available.
There is a great need to generate a better method to account for how knowledge and innovation manifest at the local level. This chapter will provide the first step towards this goal while addressing the first research question of how innovative activity manifests itself in the three regions. An industrial profile of the regions is established, and some progress is made towards identifying measures suitable for capturing innovation and knowledge activities at this smaller regional scale.
Chapter 4 – Economic and Industrial Audit 75 4.1 Suburban Sydney
The three regions: South West Sydney, including the local government areas (LGAs) of Camden, Campbelltown, Liverpool and Wollondilly; Central West Sydney, including Blacktown and Penrith LGAs; and North West Sydney, including Baulkham Hills and Hawkesbury LGAs; form a geographic ring at the outer western edge of the metropolitan area of Sydney, as shown in the maps at the beginning of this thesis (see Map 1-5 in Chapter One). All three regions are considered suburban, which needs to be briefly defined at this point, to better comprehend what this term connotes in terms of this particular research, and how this will affect the analysis of innovation.
Suburbs are rarely defined in the literature, as noted by Palen (1995):
“Everybody, it seems know what suburbs are. Over the years there has been a marvellous vague agreement in the many volumes and professional articles written on suburbs that the term in so self-explanatory that no definition need be offered” (p.8)
However, most of this ‘knowing’ about suburbs is based on traditional images of low density, single dwelling residential areas, situated on the outskirts of cities, and dependent on the central city for employment, retail and entertainment (Rothblatt and Garr 1986; Fishman 1987; Palen 1995). A more current definition of suburbs, in the Australian context, is provided by Gleeson (2006) as:
“…incorporated or unincorporated spatial communities of moderate density that lie outside the central city but within the metropolitan area. The area’s primary economic activities are non-agricultural and government is usually through independent and sometimes uncoordinated local units. Suburban areas do not have to be primarily residential, but their densities are moderate and both their population and their economic activities are spread throughout a wide area. Contemporary suburbs are not necessarily dependent economically on the central city they surround, but they are tied to the city (and to other suburbs) by a dependency on the automobile” (p.13)
Gleeson’s definition highlights two inter-related elements of the modern suburban geography: the inner city/suburban relationship; and the suburbanisation of activities away from the inner city areas out to the suburbs. The initial relationship of suburban areas with inner city areas was once one of dependence, with the suburbs dependent
Chapter 4 – Economic and Industrial Audit 76 on the inner city for employment, retail and entertainment options. However, as suburbs have developed in Australia, particularly since the 1970’s, much activity and employment has been decentralised (Fagan, Dowling et al. 2004; Sharpe 2005; Gleeson 2006). Retail and entertainment employment provide the most prominent evidence of this, with the advent of shopping centres located in suburbs having reduced the need to travel ‘into town’ for these experiences. As the employment statistics presented later in this chapter demonstrate, retail employment has exploded into the suburbs. The suburbs with major shopping complexes now possess the majority of retail space in the Sydney metropolitan area. Plans are currently proposed to build the largest shopping centre in the southern hemisphere in North West Sydney, at Castle Hill Towers.
The process of suburbanisation of activities away from the inner city areas and towards the suburbs has not been an even mix (Sharpe 2005). The spread of employment has drawn disproportionately from certain industries which, aside from retail, are primarily: manufacturing; transport; distribution; and warehousing. As Gleeson states “manufacturing industries increasingly shifted to suburban locations, spurred by the expansion of production spaces and encouraged by new manufacturing and transport technologies which freed firms from dependence upon inner city and port locations” (2006, p.17).
Manufacturing industries in Sydney have been particularly affected by suburbanisation, and also the global trend of de-industrialisation (Fagan and Dowling 2005). De-industrialisation in Australia, as in many Western countries, has seen the rationalisation of manufacturing operations and the relocation of many manufacturing functions to overseas locations. The process of suburbanisation in manufacturing has seen a gradual push of the industry out of the inner city sites and towards cheaper and more plentiful land supplies in the suburbs. The consequence of these dual processes in Western Sydney is that manufacturers that have suburbanised usually have particular reasons for continuing to manufacture their products in Sydney, as explained by Fagan and Dowling:
Chapter 4 – Economic and Industrial Audit 77 “Hence a significant proportion of the continued growth of manufacturing in Greater Western Sydney has been ‘home-grown’. In some sectors, new small and medium sized enterprises (SMEs) in Greater Western Sydney, initially established to serve the Sydney market, have been able to build on their initial success to develop export (often niche) markets while others have internationalised from the outset, exporting to the growing Asia- Pacific markets” (2005: p76)
There have also been examples of decentralisation of government departments and other industries traditionally considered to be inner city ‘Central Business District’ (CBD) industries such as legal, financial, and business services, into suburban CBDs. In many cases, this is the result of targeted government policy (predominantly in relation to the relocation of government department offices12), but on the whole, growth of these industries in the suburbs has not kept pace with that of the metropolitan area. This is particularly so in the globally competitive knowledge intensive business services sectors (Sharpe, Martinez-Fernandez et al. 2004). The suburbanisation of employment in Sydney is by no means a new trend, but what is unique to Sydney, and furthermore what is exacerbated by the unequal suburbanisation of different categories of industry, is the size of the metropolitan area. The Sydney metropolitan area occupies a large geographical space, with a low population density13. The large geographical area and low density means that Sydney, and especially Western Sydney, is marked by “substantial geographical differences in industry” (Fagan and Dowling 2005 p.74). The disadvantages arising from these unequal metropolitan industrial distributions can be mitigated in one of two ways: increased emphasis on metropolitan accessibility; or increased emphasis on reducing the skewed distribution, by encouraging diversity of industry and industry development. In the current metropolitan planning framework, the former is not a priority given the move to poly-centricity of the metropolitan area, leaving the latter as the most likely avenue of progress.
12 For example, offices of the Australian Tax Department are located in Central West Sydney, and the Sydney Water Board in South West Sydney. 13 Even with current predicted growth rates of the Sydney metropolitan area, at 2030 Sydney will still be defined as low density, “with around 17 persons per hectare, the Sydney region will be around two and a half times less dense than metropolitan Paris, London and Amsterdam, and around four times less dense then metropolitan Singapore and Tokyo” (NSW Government, 2004) Sydney Metropolitan Ministerial Directions Paper May 2004 p7.
Chapter 4 – Economic and Industrial Audit 78 These elements of suburbanisation in Sydney; the uneven suburbanisation of industry, leading to further inequality in industrial diversity of Western Sydney suburbs with the rest metropolitan Sydney, and the unique geographical spread of the metropolitan area, are important considerations within the context of attempting to analyse the industrial innovation capacity of these regions.
4.2 Growth in Outer Western Sydney
4.2.1 Population growth
Each of the three regions under investigation has received significant population increases in the last decade. The largest, in terms of overall numbers of population growth and percentage increase, has been South West Sydney, with a percentage increase of 32%, or the addition of nearly 100,000 new residents. Central West Sydney and North West Sydney have also both experienced percentage increases well above the 12.1% increase of the Sydney metropolitan area, and far above the 11% experienced overall by the state of New South Wales (NSW), as illustrated in Graph 4.1.
The graph also presents population projections for the three regions up to the year 2021 (this is shown on the right side of the red line in the graph). The population projections predict continuing strong growth trends for the next twenty years in all three regions. These three regions will therefore be the site of the majority of ‘greenfields’ (developing currently rural lands) residential housing development in the Sydney metropolitan area14. Again, the most significant development will be in South West Sydney, which it is anticipated will increase by more than 200,000 new residents over the next twenty years. This will see South West Sydney overtake Central West Sydney as the largest region in the Outer West of Sydney, in terms of population. Central West and North West Sydney will each see their populations grow by approximately 100,000 new residents in the same period.
14 Greenfields residential development is expected to account for half of all new residential dwellings in Sydney over the next 30 years (NSW Government 2005).
Chapter 4 – Economic and Industrial Audit 79 Graph 4.1 Regional Population projections 2001-2021
700,000 Actual growth Projections
600,000
500,000
400,000
300,000
200,000
100,000
0 1991 1996 2001 2006 2011 2016 2021
South West Sydney Central West Sydney North West Sydney
Source: ABS Population Projections 2004
4.2.2. Employment growth
Employment in metropolitan Sydney increased from 1,493,187 in 1991 to 1,962,937 in 2001 (ABS 1991, 2001), or an increase of 31.4%. The South West Sydney region grew by 35.8%, Central West Sydney by 38.4%, and North West Sydney by an impressive 46.9%. These employment growth rates when compared with the region’s individual population growth figures appear to indicate that the Outer Western Sydney regions’ development will be sustainable in terms of matching their population growth with equivalent or larger levels of employment growth. However, when a comparison is made of the share of Sydney’s population growth absorbed by these regions against the share of Sydney’s employment growth achieved by these regions, the differences between the two become apparent, as illustrated in Chart 4.1, with the population growth expected to far exceed the employment growth rate. These differences are clear, both in terms of Outer Western Sydney compared with metropolitan Sydney, and between South West Sydney when compared with the other two Outer Western Sydney regions. This comparison goes to the heart of why the regional economic
Chapter 4 – Economic and Industrial Audit 80 development focus underlying this research is so important, particularly in South West Sydney, which is experiencing relatively rapid but unbalanced growth.
Chart 4.1 – Share of Sydney Metropolitan Population and Employment Growth 1991-2001
25.00%
20.00%
15.00%
10.00% % share share metro increase % of Sydney
5.00%
0.00% South West Sydney Central West Sydney North West Sydney
Population growth Employment growth Source: ABS Census 2001 and 1991
4.2.3 Population growth implications
The significance of these differences is two-fold. Firstly, although it is not necessary for employment growth in a region to match population growth, with the situation of dormitory suburbs well documented (Palen 1995), in this situation, with government strategy working towards a multi-centric metropolitan area, such an imbalance may become an issue. Dormitory suburbs have long been used to describe largely residential areas on the outskirts of major cities where residents work elsewhere and just come home to these residential areas to sleep. The potential problems from imbalance are especially of concern when the emphasis on a multi-centred city comes as a result of the decision to invest less in transport and accessibility infrastructure. For this reason, development must focus on taking the jobs to the people (regional economic development), rather than the people to the jobs (regional infrastructure development). The declining share of employment being generated, coupled with reduced transport infrastructure
Chapter 4 – Economic and Industrial Audit 81 investment, has the effect of reducing the accessibility of a range of employment opportunities to a growing percentage of the Outer Western Sydney population.
Secondly, although all of the Outer Western Sydney regions have experienced a smaller share of employment growth relative to population growth, the degree of mismatch is different across the regions, being less severe in North West Sydney and Central West Sydney, and most severe in South West Sydney (as shown in Graph 4.2). In the time period studied (1991-2001), North West Sydney had less population growth, but also had higher levels of employment growth (46.9% compared with 31.4%), which contributed to the smaller differential in the population/employment growth. Central West Sydney has also had a higher percentage increase in employment (38.4%) than South West Sydney (31.4%).
The discussion in the rest of this chapter focuses on presenting the statistical audit of these three regions. Most data discussed will be from the 1991, 1996 and 2001 censuses. Due to the relatively small geographical level of these regions, the census is the only source of available data. As the census is a population census, data is at the level of the individual; so counts of people employed in industries and occupations form the basis of regional industry and knowledge base analysis.
The following section of this chapter is divided into three sections: the first covers the labour force (people who live in the regions)15; followed by the regional industry base (founded on employment counts of people who work in the region), and detailing some specific industrial concentrations; and finally, the regional knowledge base (centred on occupational classifications) is discussed. The chapter concludes with a summary of all the information imparted on the regions to date, as a prelude to the following chapters’ presentation of the empirical research results and analysis.
15 The difference between where people live and where people work is not usually noteworthy, however, when dealing with small geographical scales and a situation where metropolitan cross-commuting is a necessity, these distinctions are evident (as will be shown through the labour force and industrial profile), and therefore need to be acknowledged.
Chapter 4 – Economic and Industrial Audit 82 4.3 Labour force
The following section presents a brief description of the labour force of the three regions, including details of employment, industry of employment, occupation, and income.
4.3.1 Employment
The following two tables (Table 4.1 and Table 4.2) display the employment participation of the labour force residing in the three regions. Table 4.1 shows the levels of participation in the labour force as at 2001, including levels of part time employment, and unemployment, whilst Table 4.2 shows the movement of the labour force over the past ten years (across three census periods), highlighting changes in the labour force composition.
In terms of labour force participation, all three of the regions have higher rates of participation than that of the Sydney Metropolitan Area (SMA)16 at 61.4%, with South West and Central West Sydney having rates of 63%, and North West Sydney being slightly higher at 69%. The trend of similarities between South West and Central West is also noted in the percentages of people employed full time (68.6% and 68.9% respectively), which is slightly higher than the metropolitan average, and finally the unemployment rates (7.5% and 6.9% respectively) again occurring at slightly higher rates than the metropolitan average.
Table 4.1 Labour force composition in the regions 2001
Labour force participation South West Central West North West SMA People in the labour force 177,943 204,132 107,270 1,934,359 Labour force participation rate 63.1% 63.7% 69.7% 61.4% Employed 164,594 189,963 103,372 1,816,225 Employed full time 68.6% 68.9% 66.7% 67.6% Employed part time 31.4% 31.1% 33.3% 32.4% Unemployment rate 7.5% 6.9% 3.6% 6.1% Not in labour force 91,973 102,896 41,915 1,051,123 Source: ABS Census 2001
16 In this analysis, Sydney Metropolitan Area data are comparative figures for the entire statistical district of Sydney, and include the three regions, therefore they offer a Sydney average for comparison.
Chapter 4 – Economic and Industrial Audit 83
North West Sydney has some variable results, with higher rates of labour force participation (69.7%), but comparatively lower rates of full time employment. North West Sydney’s unemployment rate is half that of the other two regions, and almost half that of the SMA. That the unemployment rate in North West Sydney is comparatively low not only to that of South and Central West Sydney, but also to the Metropolitan area, suggests there must be a specific area of difference for the North West Sydney labour force. The time series analysis presented in Table 4.2 shows that this is also an historical trend. Table 4.2 reveals the labour force composition in the regions in 1991, and offers a few aspects of comparison between the regions over time. Not surprisingly, considering the levels of population growth experienced by the three research subject regions, all three have increased their labour forces at double or more the rate of the Sydney metropolitan level. The largest increases have been in South West Sydney, which is in line with the population growth figures presented in Graph 4.1.
Table 4.2 Labour force composition in regions in 1991, and comparisons with 2001
Labour Force Participation 1991 South West Central West North West SMA In labour force 133,582 170,935 85,873 1,736,096 Difference in labour force 1991-2001 44,361 33,197 21,397 198,263 % Increase in labour force 91-01 33.2% 19.4% 24.9% 11.4% Participation rate 64.7% 65.6% 69.1% 62.6% Employed 116,320 150,627 80,927 1,556,448 Full time employed 72.7% 73.2% 69.8% 70.9% Part time employed 27.3% 26.8% 30.2% 29.1% Unemployment rate 12.9% 11.9% 5.8% 10.3% Not in labour force 65,848 81,520 36,025 958,850 Source: ABS Census 1991 and 2001
Labour force participation rates in both South West and Central West Sydney, and across the metropolitan area as a whole, have decreased slightly over the past ten years, whilst participation rates in North West Sydney have risen slightly in the same time period. There are two main areas of difference over the two time periods. The first of these is in the ratios of full time to part time employment, with the levels of part time employment increasing. This is in line with changes across the entire metropolitan area, and indeed the working arena generally, as the
Chapter 4 – Economic and Industrial Audit 84 move to more flexible working conditions has become more entrenched. The second point of difference is in the unemployment rates. In the early 1990’s the Australian economy was in recession, accounting for the higher unemployment rates in 1991. However, North West Sydney had an unemployment rate in 1991 of less than half that of the other two subject regions (5.8% compared with 12.9% in South West Sydney and 11.9% in Central West Sydney), and almost half of the overall SMA rate of 10.3%. North West Sydney appears to be an exceptional case, achieving higher rates of employment growth within the region, and a labour force with much less unemployment. The industrial profile of the North West region goes some way towards explaining the North West economy’s apparent resilience. This is covered in more detail in the following section.
4.3.2 Employment by industry
The breakdown of employed labour force members by broad industry category is presented in Table 4.3. Three industry categories are highlighted in bold; these are the largest categories of persons employed, and as such offer points of comparison among the three regions, and also between the three regions against metropolitan Sydney.
Manufacturing makes up the largest industry category in both South West Sydney and Central West Sydney (17.5% and 16.4% respectively). In North West Sydney, manufacturing is the third largest industry component with 11.7% of the labour force, behind retail trade with the largest component at 14.5%, and property and business services with 12.9%. These other categories (retail trade, and property and business services) form the second and third largest components in the other two regions, and indeed these three categories match (though in different order) the top three concentrations of employment in the wider SMA. This provides some clue about the perceived resilience of the North West Sydney economy over the decade 1991-2001. The higher levels of labour force within property and business services, an industry associated with the aforementioned globally competitive KIBS, appear to have offered some insulation with their high growth employment figures, a factor that is not present in the other two regions.
Chapter 4 – Economic and Industrial Audit 85
Table 4.3 Labour force industrial employment in Outer Western Sydney 2001
Industry South West Central West North West SMA
Agriculture, Forestry, Fishing 1.20% 0.80% 1.80% 0.60% Mining 0.30% 0.10% 0.10% 0.10% Manufacturing 17.50% 16.40% 11.70% 12.20% Electricity, Gas & Water Supply 0.60% 0.80% 0.70% 0.60% Construction 8.50% 8.30% 8.80% 6.90% Wholesale Trade 6.20% 7.40% 7.50% 6.00% Retail Trade 14.50% 15.10% 14.40% 13.40% Accommodation, Cafes & Restaurants 3.60% 3.50% 3.50% 4.80% Transport & Storage 6.10% 5.40% 3.40% 5.00% Communications Services 2.20% 2.60% 1.80% 2.40% Finance and Insurance 4.60% 5.30% 5.10% 6.10% Property and Business Services 9.20% 9.60% 12.90% 14.50% Gov't Administration & Defence 4.00% 3.70% 3.60% 3.40% Education 5.70% 4.80% 7.60% 6.40% Health & Community Services 7.90% 8.30% 8.90% 8.90% Cultural & Recreation Services 1.60% 1.70% 2.20% 2.80% Personal & Other Services 3.50% 3.70% 3.90% 3.60% Not Stated 2.80% 2.50% 1.90% 2.40% Source: ABS 2001 Census
4.3.3 Occupation
This proposition is further reinforced when the occupational breakdown of the labour force in the three regions is examined. This breakdown is presented in Table 4.4. Again, the largest categories are shown in bold, as are the ones that offer the best points of comparison between the three regions.
The largest categories of occupations in South West Sydney and Central West Sydney are both in the intermediate clerical, sales and service occupations, with 19.1% and 20.1% of the labour force respectively. However, the largest component of occupations in North West Sydney is within the professional category, with 21.4% of the labour force. The breakdown of the wider metropolitan area also shows this as the highest occupational category. The level of managers and administrators in North West Sydney is double the level in South West Sydney and Central West Sydney, and even three percentage points higher than the overall metropolitan figures. The higher levels of intermediate production and transport workers, and labourers and related workers in South West Sydney
Chapter 4 – Economic and Industrial Audit 86 and Central West Sydney reflect the higher levels of manufacturing industries present in these two regions, especially when compared to North West Sydney.
Overall, the occupational profile of the labour force within the three regions shows that North West Sydney has higher levels of professional workers - what some may call ‘knowledge workers’ (Sharpe and Martinez-Fernandez Forthcoming). The presence of higher quantities of these workers also goes some way towards explaining the higher levels of finance and business services in North West Sydney. Even then, when we compare the percentage of professionals in the occupational category to the total percentage of those within the finance and business services industry in the region, the rate of professionals is much higher, thereby suggesting that these professionals are either spread across other industry categories, or are ‘exported’ to other areas in the metropolitan area. Either way, the North West Sydney labour force has a higher than average level of professional workers, which may suggest a higher knowledge capacity than the other two regions (Sharpe and Martinez-Fernandez, Forthcoming).
Table 4.4 Labour force occupation in Outer Western Sydney 2001
Occupation South West Central West North West SMA Managers and Administrators 6.40% 5.40% 12.10% 9.00% Professionals 12.50% 12.20% 21.40% 21.20% Associate Professionals 10.20% 10.00% 12.40% 11.80% Tradespersons and Related Workers 15.00% 13.80% 11.40% 11.10% Advanced Clerical and Service Workers 4.10% 3.90% 5.50% 4.50% Intermediate Clerical, Sales and Service 19.10% 20.10% 16.50% 17.20% Intermediate Production and Transport 11.90% 12.70% 5.60% 7.40% Elementary Clerical, Sales and Service 9.40% 10.40% 8.20% 9.10% Labourers and Related Workers 9.20% 9.30% 5.20% 6.60% Source: ABS 2001 Census
4.3.4 Income
The final area of labour force analysis in the three regions is income levels. Table 4.5 outlines a number of measures of income in the three research subject regions, again compared to the Sydney metropolitan area statistics, including median
Chapter 4 – Economic and Industrial Audit 87 individual weekly income, median family weekly income, median household weekly income, mean individual taxable income, and the number of Centrelink (government income support) recipients.
North West Sydney had the highest level of mean taxable income, significantly higher than South West Sydney and Central West Sydney ($5,583 and $6,080 more per annum respectively in each case). North West Sydney’s mean taxable income is also just over $1,000 more than the NSW average17. These higher levels are reflected in higher levels of median weekly income for individuals, families, and households. The levels in North West Sydney are also higher than the SMA medians as well.
Table 4.5 Income levels in Outer Western Sydney Medium Medium Centrelink Weekly Medium Weekly Mean Income Individual Weekly Family Household Taxable Support Income Income Income Income Customers South West Sydney $400-$499 $1,000-$1,499 $800-$999 $37,085 78,522 Central West Sydney $400-$499 $1,000-$1,499 $800-$999 $36,588 87,653 North West Sydney $500-$599 $1,500-$1,999 $1,000-$1,499 $42,668 26,351 SMA $400-$499 $1,000-$1,499 $800-$999 na. 782,328 NSW $300-$399 $800-$999 $800-$999 $41,623 1,474,412 Source: ABS 2001 Census
The final point of comparison is the number of Centrelink income support recipients in each of the regions. Centrelink income support recipients include pensioners (both old age pensioners and invalid pensioners), support to single parents, unemployed people, and full time students18. The levels of customers in North West Sydney are significantly lower than the other two regions, both in actual numbers, but also in comparison as a percentage of the population of the regions. The highest percentage is recorded in South West Sydney, with 29.1% of the population receiving income support payments, followed by Central West Sydney with 28.5%, both of which are higher than the Sydney metropolitan level of 26.2%, and then North West Sydney with 17.7% of the population.
17 Mean taxable income, in this instance a comparison with metropolitan Sydney, is not available, so the state of New South Wales (NSW) is used. 18 A breakdown of recipients by support payment type is not available at the regional level.
Chapter 4 – Economic and Industrial Audit 88
There are a number of possible reasons for this disparity. The age profile of the regions19 shows lower levels of elderly people than the metropolitan average, and higher levels of children and young adults. This also flows through to figures on the levels of participation in educational institutions. These higher levels of income support payments could therefore be directed at single parent and full time student support, instead of ‘old age’ pensioners. Suburban areas typically contain more affordable living accommodations; consequently it would be natural for lower income earners to gravitate to these areas. While this explains the relatively high levels of Centrelink recipients in South and Central West Sydney, it does not account for the lower than average levels in North West Sydney. Previously highlighted socio-economic dynamics observed in the occupational analysis may account for North West Sydney’s lower level of income support recipients.
4.3.5 Labour force implications
The previous four sections, which described the demographics of the labour force of the three regions, have shown that on the whole, the North West Sydney labour force is distinctly different to the other two regions. It appears to be in a more competitive position to the other two regions, with lower unemployment levels, higher labour force participation rates, higher levels of professional and manager occupations, and a higher level of income. All these factors are inter-related and self-reinforcing. The key questions are then firstly, what are the underlying reasons for these results, and secondly, is this related to industrial innovation capacity? The next section examines both the industrial profile, and then the knowledge base of the three regions, to further explore these regional differences.
4.4 Regional industry base
This section details the industrial base of the regions. As was mentioned earlier, the source of the majority of data at the regional geographic level is the 5-yearly
19 Age profile and household structure for the regions is presented in the social profile of the regions in Appendix 7.
Chapter 4 – Economic and Industrial Audit 89 Australian census, and because the census is a population count, data is centred on the individual. Therefore, the following discussions are founded on the industrial base as drawn from counts of employed people in particular industries. The counts are of people who work in the regions not, as in the previous section, on people who live in the regions.
Table 4.6 presents the percentage of people working in the captured broad industry categories for the three regions under investigation and, as before, in comparison with statistics for metropolitan Sydney. The bold highlighting is used to draw attention to both the largest industry concentrations, and areas of comparison between the three regions.
Table 4.6 Regional industrial base (from employment) 2001
Industry South West Central West North West SMA Agriculture, Forestry, Fishing 1.80% 1.00% 3.10% 0.60% Mining 0.90% 0.20% 0.10% 0.20% Manufacturing 18.10% 17.90% 11.40% 12.60% Electricity, Gas & Water Supply 0.60% 1.40% 0.40% 0.60% Construction 6.60% 6.70% 8.00% 5.00% Wholesale Trade 5.30% 7.50% 7.60% 6.40% Retail Trade 17.60% 18.50% 19.30% 13.80% Accommodation, Cafes & Restaurants 3.30% 3.60% 4.30% 4.90% Transport & Storage 4.30% 4.40% 1.90% 5.00% Communications Services 1.00% 1.30% 1.00% 2.40% Finance and Insurance 1.70% 1.80% 2.50% 6.60% Property and Business Services 7.50% 7.20% 13.40% 14.80% Gov't Administration & Defence 4.90% 4.00% 4.30% 3.60% Education 9.30% 8.90% 8.30% 6.70% Health & Community Services 11.00% 9.20% 7.50% 9.30% Cultural & Recreation Services 1.50% 1.80% 2.10% 2.80% Personal and Other Services 3.40% 3.70% 3.90% 3.60% Source: ABS 2001 Census, Journey to Work data
Manufacturing is the largest employing industry in South West Sydney, closely followed by retail trade, and then health and community services. In Central West Sydney, retail trade has a slightly larger employment base than manufacturing, and is closely followed by health and community services, and education employment. In North West Sydney, retail trade also has the highest levels of
Chapter 4 – Economic and Industrial Audit 90 employment, but unlike the other two regions, this is then followed by property and business services.
The strength of employment levels displayed in the industry categories of health and community services, and education in Central and South West Sydney can be attributed to the young age profile of the three regions20, as well as the large percentage of children and young people residing in the regions. The higher employment in these industries may also be due to the location of major regional hospitals in the regions; Liverpool Hospital in the South West is one of the largest employers in the region, as is Nepean Hospital in Central West Sydney. There is a hospital in North West Sydney, but it is not a major hospital. Generally speaking, the areas of health and community services and education21 are a function of population growth and location (Maglen 2001), which is evident in all these regions. They are not as dependent at the regional level on endogenous processes of industrial development as are some of the other industry categories.
North West Sydney, despite having a smaller employment base (almost 50% smaller than Central West Sydney), nevertheless had the highest number of jobs in the property and business service sector, followed in order by manufacturing, education, and construction.
Table 4.7 examines the amount of industrial change that occurred in these broad industry categories in the decade between 1991 and 2001.
20 See full profile in Appendix 5 21 Education includes employment in pre-school, primary and high school education , and also employment in tertiary education institutions such as Universities and TAFE (Technical and Further Education Colleges). These resources are important for evaluating the innovative capacity of the regions, especially in terms of their relationships with local businesses, and this aspect of ‘Education’ will be discussed in later chapters.
Chapter 4 – Economic and Industrial Audit 91 Table 4.7 Industrial change (based on employment) in Outer Western Sydney 1991-2001
Industry South West Central West North West SMA Agriculture, forestry, fishing -0.59% 0.44% 0.59% 0.35% Mining -4.43% -0.15% -0.12% 0.01% Manufacturing 15.71% 14.60% 8.67% 8.64% Electricity, Gas & Water Supply -1.22% 0.60% 0.26% 0.38% Construction 5.37% 3.27% 3.07% 3.80% Wholesale & Retail Trade 43.24% 41.82% 40.85% 25.67% Transport & Storage 6.54% 8.38% 0.99% 5.39% Communications Services -0.50% -0.24% 0.92% 2.84% Finance, Insurance, Property & Business Services 11.67% 11.88% 27.43% 27.24% Government Administration & Defence -2.52% 1.55% 0.43% 3.17% Education, Health & Community Services 20.71% 12.54% 11.67% 14.99% Cultural, Recreation, Personal & Other Services 3.90% 3.94% 4.15% 6.76% Source: ABS Journey to Work data 1991-2001
The manufacturing industry in South West Sydney and Central West Sydney experienced strong growth (15.7% and 14.60% respectively) when compared with both the overall Sydney metropolitan area (8.4%) and North West Sydney (8.67%). This result is particularly promising considering that the manufacturing industry in Australia was rationalised in the 1990’s (Fagan 2006). The growth in manufacturing is, however, dwarfed by the high level of growth in wholesale and retail trade22, which has grown enormously in the same time period: by 43.24% in South West Sydney; 41.82% in Central West Sydney; and 40.85% in North West Sydney. In the overall Sydney metropolitan region, employment in this sector only grew by 25.67%. These figures highlight the increasing suburbanisation of some service activities, as noted earlier, particularly the move in retail and aspects of the wholesaling and distribution function away from the centre of the SMA.
The suburbanisation of KISA, which are usually associated with finance and business service activities, has not occurred to the same extent. As Fagan notes:
“Service industries have continued to suburbanise during the last two decades and, by 2001, the Greater West’s share of total employment in the SMA (Sydney Metropolitan Area) for services like retailing, wholesaling
22 Due to changes in the industrial categories with the 1996 census, some industry groups have had to be aggregated to calculate growth figures across the 10-year period. Groups affected are: Wholesale Trade, and Retail Trade (amalgamated into Wholesale and Retail Trade); Finance and Insurance, and Property and Business Services (amalgamated into Finance, Insurance, Property and Business Services); and Cultural and Recreational Services, and Personal and Other Services (amalgamated into Cultural, Recreational, Personal and Other Services).
Chapter 4 – Economic and Industrial Audit 92 and construction approximated the proportion of the city’s population resident in GWS (Greater Western Sydney) (42%). By contrast, and despite significant suburbanisation of office employment since the mid 1970s, only 17% of Sydney’s total employment in Banking, Financial and Business Services (BFBS) was located in GWS, which lags behind the inner parts of Sydney as a location for employment in this most dynamic component of the metropolitan labour market” (Fagan et al 2004 p: 76)
The finance, insurance, property and business service industry was a significant area of employment growth for the Sydney Metropolitan region, with a 27.24% increase, and North West Sydney’s employment in this sector also increased by 27.43%. South West and Central West Sydney, however, have lower rates of growth (11.67% and 11.88% respectively). This category represented the largest percentage increase in an industry sector within the Sydney metropolitan area, even larger than retail and wholesale employment, and North West Sydney, as evidenced by these figures, has enjoyed more of this “dynamic component” (Fagan et al 2004: 76) than the other two regions.
Further analysis and a finer level breakdown is presented on both the manufacturing, and financial and business services industries in the next section, with the less aggregated industry breakdown shedding a better light on these industries within the three regions. These two industrial areas were selected because they represent the best opportunities for industrial innovation capacity and development. As earlier stated, health, community services, and education are dominated by public sector organisations with differing innovation dynamics. In addition, retail employment, which makes up the majority of the growth in the retail and wholesale trade sector, is characterised by low levels of knowledge intensity, and high part-time and causal employment. Research into industrial drivers for growth in metropolitan Sydney has noted the ‘drag’ of retail and simple-service employment on the economic development of regions (Blakely, Bista et al. 2006).
Chapter 4 – Economic and Industrial Audit 93 4.4.1 Industry concentration and specialisations
The Western Sydney economy has one of the most significant concentrations of manufacturing activity in Australia, and is therefore known as a manufacturing hub (Martinez-Fernandez and Bjorkli 2003). This concentration is not dominated by any particular types of manufacturing industries that have combined to make a “cluster” or represent a competitive advantage such as developing externalities; rather the unique quality of the outer areas of western Sydney is the diversity of industry, instead of specialisation in one or two particular industries (Fagan et al 2004). This produces further difficulties when contemplating appropriate policy options for industrial development at the SMA level, and demonstrates why policy may be able to be more appropriately targeted from the local/ regional level, with a broad industry focus.
Manufacturing
Table 4.8 shows industrial employment in manufacturing at a finer level of aggregation. South West Sydney has higher concentrations (more than 2000 employees) in four manufacturing industries: Machinery and Equipment manufacturing with 3,613 workers; followed by Food, Beverage and Tobacco manufacturing (2,480 employees); Metal Products manufacturing (2,429 workers); and Petroleum, Coal, Chemical and Associated Product manufacturing (2,425 workers). Central West Sydney also has the same four industries with high concentrations of employment, but in a different order, with Metal Products manufacturing shown as the largest with 3,946 workers, followed by Food, Beverage and Tobacco manufacturing with 3,005 workers, Petroleum, Coal, Chemical and Associated Product manufacturing (2,729 workers), and Machinery and Equipment manufacturing (2,455 workers).
Chapter 4 – Economic and Industrial Audit 94 Table 4.8 Detailed breakdown of manufacturing employment in the regions, 2001
Manufacturing South West Central West North West SMA Food, Beverages and Tobacco 2,480 3,005 536 25,647 Textile, Clothing, Footwear and Leather 837 816 420 13,299 Wood and Paper Products 1,227 1,109 591 10,993 Printing, Publishing and Recorded Media 1,324 1,176 1,044 31,947 Petroleum, Coal, Chemical and Associated Products 2,425 2,729 1,624 25,619 Non-Metallic Mineral Products 1,134 1,440 220 7,944 Metal Products 2,429 3,946 654 22,659 Machinery and Equipment 3,613 2,455 1,450 45,442 Other 1,484 1,676 463 14,559 Total 101,044 120,728 65,205 1,673,445
Source: ABS 2001 Census Journey to Work data
North West Sydney does not have any manufacturing industries with employment levels above 2,000 workers. The manufacturing industry with the highest number of employees is the Petroleum, Coal, Chemical and Associated Product manufacturing (1,624 workers), followed by Machinery and Equipment manufacturing (1,450 workers), and Printing, Publishing and Recorded Media (1,044 workers).
Comparing these employment figures at the regional level with total employment in these industries at the metropolitan level, it is noteworthy that although manufacturing is still a significant industry in these three regions, particularly for South West and Central West Sydney, there is also a large amount of manufacturing employment within the rest of the metropolitan area. In industries with the highest levels of employment in the three regions, such as Machinery and Equipment manufacturing, the three regions together only accounted for 16.5% of the total employment in this industry in the SMA. Petroleum, Coal, Chemical and Associated Product manufacturing accounts for 26.5%, Metal Products manufacturing 31.0%, and Food, Beverage and Tobacco Product manufacturing 23.5%. The highest percentages of employment for a manufacturing industry in these Outer Western Sydney regions is Non-Metallic Mineral Product manufacturing, which accounts for 35.2% of total manufacturing in this category
Chapter 4 – Economic and Industrial Audit 95 within the SMA. Non-metallic Mineral Product manufacturing includes the production of glass, plastics and resins, and clay and ceramics.
It should be noted that these industry levels are still significantly aggregated and finer levels of employment counts may provide evidence of further differences between the regions, and highlight industry concentrations and clusters, which have not so far been apparent. With this in mind, the question remains: Is it significant that North West Sydney does not have any manufacturing concentrations of more than 2,000 employees? When asking this question, it must be remembered that North West Sydney had the smallest labour force in comparison, and therefore the actual number of 2,000 may not be so relevant in determining concentrations. This underscores the challenges inherent in determining the advantages of a regional economy by looking only at the size of the industry. The industry size needs to be put into perspective and context through comparison, as emphasised by the above examples of SMA industrial employment, to better determine whether any characteristics that are significant at the local level also carry weight when compared to other areas.
The Location Quotient (LQ) technique is a frequently used method for this purpose (Armstrong and Taylor 2000). The technique calculates a ratio that compares the local economy, in this case the three individual regional economies of Outer Western Sydney, with a reference economy, usually the national economy. When a calculated LQ is more than 1, this means that local employment in a particular industry was more than was expected in comparison with the reference economy, and suggests some degree of specialisation in the local economy in this particular industry.
Table 4.9 presents the LQ calculations for manufacturing in the regions and the SMA. As manufacturing occupies a significant component of the region’s workforce, with LQs of more than 1 in most industries, those industries with LQs of 2 or more will be the focus of further examination.
Chapter 4 – Economic and Industrial Audit 96 South West Sydney has three industries with a LQ greater than 2: Food, Beverage and Tobacco manufacturing; Non-Metallic Mineral Product manufacturing; and Petroleum, Coal, Chemical and Associated Product manufacturing. Of these three, only Food, Beverage and Tobacco manufacturing and Petroleum, Coal, Chemical and Associated Product manufacturing also appear in the category used earlier for comparison, which looked at those with employment counts of 2,000 or more people. The Food, Beverage and Tobacco manufacturing category has the highest LQ ratio, however these quotients are calculated from 2001 data, and anecdotal evidence suggests that these high levels of employment will not be repeated when the 2006 census results are released (sometime in mid 2008). This drop in levels is anticipated because of a number of major manufacturers relocating out of the area, and a reduction in agricultural production, especially market gardening, due to continued ‘greenfields’ residential developments taking up former agricultural land and thereby reducing the advantages that existed between the once close proximity of food production (agriculture) and food manufacturing.
In Central West Sydney, three categories of manufacturing have a LQ of more than 2. The highest ratio was 3.74, and, as for South West Sydney, this was in Food, Beverage and Tobacco manufacturing, and was followed by Non-Metallic Mineral Products and Metal Products manufacturing. Again, as in the case of South West Sydney, the Non- Metallic Mineral products manufacturing did not show up in the earlier employment analysis. The same anecdotal evidence for Food, Beverage and Tobacco manufacturing holds true for Central West Sydney as for South West Sydney. North West Sydney only had one manufacturing category with a LQ above 2: Petroleum, Coal, Chemical and Associated Product manufacturing, which had a ratio of 2.14.
Chapter 4 – Economic and Industrial Audit 97 Table 4.9 Location Quotients for manufacturing industries in the regions, 2001
Manufacturing LQ SWS LQ CWS LQ NWS LQ SMA Food, Beverages and Tobacco 3.69 3.74 1.24 2.31 Textile, Clothing, Footwear and Leather 1.05 0.86 0.82 1.01 Wood and Paper Products 1.56 1.18 1.17 0.85 Printing, Publishing and Recorded Media 1.07 0.80 1.31 1.56 Petroleum, Coal, Chemical and Associated Products 2.06 1.94 2.14 1.32 Non-Metallic Mineral Products 2.23 2.37 0.67 0.94 Metal Products 1.48 2.01 0.62 0.83 Machinery and Equipment 1.33 0.76 0.83 1.01 Other 1.78 1.68 0.86 1.06 Source: Author’s calculations based on ABS 2001 Census, Journey to Work data
On the whole, the location quotients and the employment counts both show that South West and Central West Sydney have a stronger manufacturing sector than North West Sydney. Unfortunately, the limited degree of industrial aggregation available at the regional level limits further and more particular analysis. It cannot be assumed that the same activities are occurring within each region, even though they are both showing high concentrations in the same industry sector. What this does give us, is some idea of the general industrial bases and specialisations that exist within the regions, and where general differences lay.
The next section repeats the same regional industrial analysis, but this time for the Financial and Business Services industries.
Financial and Business Services
Table 4.10 shows the further industrial breakdown of the Financial and Business Services industries.
Table 4.10 Detailed breakdowns of Financial & Business Services Employment in the regions, 2001
Finance and Insurance South West Central West North West SMA Finance 984 1,458 852 56,291 Insurance 421 347 292 27,279 Services to Finance and Insurance 344 342 473 25,622 Property Services 1,962 1,735 1,328 30,411 Business Services 5,585 6,939 7,358 217,605 Source: ABS 2001 Census Journey to Work data
Chapter 4 – Economic and Industrial Audit 98 The first thing to notice about these is the higher counts of employment for North West Sydney, which is only second by some 500 workers to Central West Sydney, even though the North West Sydney workforce has roughly half the total number of that in the Central West Sydney workforce. Secondly, it is important to observe the detailed breakdowns trend across all three regions, with the highest counts being in business services, followed by property services23, and finally finance employment. The third aspect of importance, is the significant amount of employment in these industries at the metropolitan level, with all three regions only accounting for 8.5% of total employment in Financial and Business services in the SMA.
This observation is further reflected by an analysis of the LQs in these industries across the three regions when compared to the SMA. These figures are presented in Table 4.11. None of the categories in any of the three regions had a location quotient above 2, including North West Sydney, and only three categories had a LQ above 1 (reflecting a potential specialisation in comparison with the national economy). Two of these were in the Property services category within both South West and North West Sydney (1.30 and 1.36 respectively), which, as noted earlier, are to be treated with caution as these relative concentrations can be explained by the high levels of residential property development in these regions. The third LQ over 1 is Business services in North West Sydney, although only with a relatively small ratio level of 1.16.
Table 4.11 Location Quotients for Finance & Business services in the regions, 2001
Finance and Insurance South West Central West North West SMA Finance 0.47 0.58 0.62 1.61 Insurance 0.42 0.29 0.45 1.64 Services to Finance and Insurance 0.43 0.36 0.91 1.92 Property Services 1.30 0.96 1.36 1.22 Business Services 0.57 0.59 1.16 1.33 Source: Author’s calculations based on ABS 2001 Census, Journey to Work
This Table shows that although the Finance and Business services sector is considered ‘high growth’ because of both its association with KIBS, and because
23 There is debate about the structure of the property service category. It is argued that employment in this category is dominated by real estate agents offering retail services to home-buyers rather than property services to businesses. In the 2006 census, this category has been changed. Therefore, property services as business services are treated with caution in this analysis.
Chapter 4 – Economic and Industrial Audit 99 this sector is thought to be occupying an increasing share of employment levels in economies, this is not the case in the three research regions; Central West, North West and South West Sydney. The largest category within these three regions, that of Property and Business services, needs to be treated with caution, as this employment may be associated more with ‘retail style’ real estate services, rather than business services.
Analysis of this sector is further hampered by access to only relatively high levels of regional industrial aggregations. However, we can conclude that these regions all have a significant manufacturing base, more so in the cases of South West and Central West Sydney than with North West Sydney. It can also be concluded that each of the regions does not have the same diversity or depth of business services (excluding property services) that is exhibited at the SMA level, although North West Sydney is better placed than the other two regions. The high concentration of global services firms in North Sydney and North Ryde (which is in close proximity to North West Sydney) may also be reflected in the higher levels recorded for North West Sydney. The next section presents the available general data for firms in these regions.
4.4.2 Firm demographics
Up until this point, all of the data presented on the industrial profile of the regions has been based on employment counts. This is because employment counts provide the most detailed and comparable data. However, general information about firms is important in understanding the structure of the industries and firm innovative activity.
Data on the number of firms located in each of the regions is available from the Australian Taxation Department in the form of Australian Business Number (ABN) registrations. This data has its limitations in that it only records single location entities, and as such will only record the ABN registration location for multiple location firms, which is usually the head office. Furthermore, the data is recorded by postcode, and postcode areas do not match local government
Chapter 4 – Economic and Industrial Audit 100 boundaries, and can also change significantly over time, especially in fast growing areas such as those under investigation. Nevertheless, the ABN registration remains the best regional level data available for firms.
Table 4.12 details industry composition based on firm industry definition. It presents a different aspect to each region’s economy. For example, construction industry firms make up 26% of all firms in South West Sydney, yet only account for 6.6% of employment, meaning that the South West Sydney construction industry, as for the industry across metropolitan Sydney as a whole, is dominated by very small firms. The reverse is true of manufacturing, with 8.1% of firms in the manufacturing industry accounting for 18.1% of employment in South West Sydney. As a result, the manufacturing industry in these regions is characterised by a number of larger firms.
The property and business services industry also tends to be dominated by small firms, which again could be influenced by the property services group. Retail trade joins manufacturing in being under-represented in terms of the percentage of firms versus the percentage of employment provided within the industry, suggesting that medium to large firms are also present in larger numbers than low- employment level small firms. This industry structure is also represented in the turnover of firms; Table 4.13 presents the estimated firm financial turnovers in ranges for 2001.
Chapter 4 – Economic and Industrial Audit 101 Table 4.12 Industrial Composition of Firms in Regions, 2001 Central South Industry West North West West SMA NSW Agriculture, Forestry and Fishing 2.1% 4.9% 3.7% 1.4% 9.2% Mining 0.1% 0.1% 0.0% 0.1% 0.2% Manufacturing 8.1% 6.0% 7.8% 6.1% 5.6% Electricity, Gas And Water Supply 0.1% 0.0% 0.0% 0.0% 0.1% Construction 26.0% 22.4% 26.7% 16.2% 16.2% Wholesale Trade 4.3% 5.2% 3.8% 4.3% 3.8% Retail Trade 10.9% 10.3% 10.6% 9.4% 10.0% Accommodation, Cafes And Restaurants 1.1% 1.6% 1.2% 2.1% 2.6% Transport And Storage 9.0% 5.1% 8.6% 6.0% 5.6% Communication Services 3.1% 1.8% 2.6% 1.8% 1.6% Finance And Insurance 4.1% 0.8% 4.2% 9.4% 7.5% Property And Business Services 19.8% 28.3% 19.3% 29.2% 24.2% Government Administration And Defence 0.1% 0.0% 0.1% 0.0% 0.1% Education 1.2% 1.5% 1.2% 1.3% 1.4% Health And Community Services 3.1% 4.5% 3.3% 4.5% 4.3% Cultural And Recreational Services 2.3% 3.2% 2.3% 3.7% 3.4% Personal And Other Services 4.1% 4.3% 3.9% 3.9% 4.0% Source: ABS 2004 Regional Profile (2001 data)
The majority of firms in all three regions under investigation would be classified as small to medium sized firms (less than 30 employees), with 95% turning over less than $1 million per annum in 2001. However, the outlined limitations of the ABN method would have excluded any multi-location firm that did not have its head office located in these regions, which would have excluded many larger firms from these figures.
Some small differences between the three research regions creep into the analysis, particularly in the lower turnover categories. Central West Sydney has higher levels of firms that turn over $50,000 or less (34.4%), and $1 million or more (4.7%). South West Sydney has the highest levels of firms in the $50,000- $100,000 per annum turnover category, while North West has the highest in the $100,000 - $1 million category. In the category of turnover in excess of $1 million, all three regions are well behind the SMA and NSW levels of 6.2% and 5.4% respectively.
Chapter 4 – Economic and Industrial Audit 102 Table 4.13 Estimated Firm turnover (Ranges) % in regions, 2001
Estimated Turnover by firms $50,000 – $100,000 – $1,000,001 (Ranges) $0 - $49,999 $99,999 $1,000,000 and over South West Total 31.7% 34.3% 29.2% 4.4% Central West Total 34.4% 33.5% 27.3% 4.7% North West Total 32.2% 30.5% 32.4% 4.6% SMA 33.2% 29.2% 31.4% 6.2% NSW 34.9% 28.1% 31.6% 5.4% Source: ABS 2004 Regional Profile
The final firm statistical format available is business start-up rates, based on increases in the number of Australian Business Number (ABN) registrations. Business start-up rates are a key indicator of entrepreneurial activity in an economy. Table 4.14 details business start-up rates, by tracking the number of new ABN registrations with the Australian Taxation Office. Some limitations with ABS data have already been mentioned, and these statistics should be used as trend indicators rather than for their actual raw numbers. This is due to the fact that while new business registrations are added to the overall figures on an annual basis, inactive registrations are not discarded in the same time period (several years of determination is required), so only one side of the equation is evident.
Significant growth in registrations is apparent across all regions, along with similar annual growth rates, peaking in 2001-200224 at between 23.7% and 25.2%, and declining to 2003-2004 rates of 17% in all regions.
Table 4.14 New ABN Registrations 2001-2004
Total Firms Active Growth rate Growth Rate Growth Rate Region 2001 Registrations* 2001-2002 2002-2003 2003-2004 Central West Sydney 35,520 71,970 24.9% 21.7% 17.7% North West Sydney 26,319 57,775 23.7% 22.2% 17.3% South West Sydney 33,196 66,575 25.2% 22.5% 17.4% Source: Australian Business Number Register 2005 *Active registrations as at April 2005
These statistics confirm the importance of small businesses in the economies of these individual regions, but also to broader Sydney and NSW. The role of small business is always at the forefront of regional development policies, as evidenced
24 The peak period of registrations in 2001-2002 is explained by the introduction of a new national Goods and Services Tax (GST). For businesses in order to pay tax or receive tax exemptions and tax credits they needed to have an ABN.
Chapter 4 – Economic and Industrial Audit 103 in the discussion in Chapter Two, simply because, as in this case, most economies’ firms and thereby employment are dominated by small and medium sized firms (Birch 1979; OECD 2001). Small firms have some noted differences in their innovative capacity, as has been demonstrated in a number of empirical studies (Storey 1985; Acs and Audretsch 1990; de Jong and Marsili 2006; Watts, Wood et al. 2006). The advantages of small firms in the innovation process include their generally flatter and more flexible organisational structures, and more flexible management structures, all of which are better able to facilitate the conditions for learning and innovative activity (Acs and Audretsch 1990; Acs 2002). Firm size is a key variable in the empirical analysis presented in the following chapters.
4.5 Regional knowledge base
The final area of research for the background industrial analysis, is an occupational analysis of each region’s knowledge base. The importance of understanding a region’s knowledge base is twofold. In earlier discussions on innovation and the importance of geographic location, the relationship between tacit and codified knowledge was established. Tacit knowledge requires a degree of capability within a certain field of codified knowledge. The establishment of an industrial concentration is usually seen as a catalyst [MT2]for this type of relationship i.e. a concentration of industry exists, therefore a concentration of a specific type of knowledge, and hence the ability for tacit knowledge transfers through the proximity of both geographic and knowledge bases (Asheim and Coenen 2005; Asheim and Coenen 2006). However, industry concentration does not always tell the entire story. As Feser and Bergman state (Feser and Bergman 2000; Feser 2003), it is not just what regions make that explains their story but also what they do, and this is where occupational analysis has its focus.
Occupational analysis has, until recently, been a neglected area of research within regional economic and innovation studies. However, as Koo (2005) and others (Markusen and Schrock 2001; Barbour and Markusen 2004; Markusen 2004)
Chapter 4 – Economic and Industrial Audit 104 point out, “as regional competitiveness has become increasingly dependent upon local knowledge bases and worker quality… examining regional economies from a different angle (i.e. occupations) can provide important insights for regional development” (Koo, 2005,p.1). Markusen and Schrock (2001) also note that the increased focus on human capital elements in the production process due to the realisation of the importance of knowledge activities, leads to the fact that “function, skill and connections (have) become more important than organisation, and these are best studied via occupational groupings” (2001, p5). Acs et al also note that “determining the extent to which innovation can be substituted by other, more accessible measures, is essential for a deeper understanding of the dynamics involved in regional innovation” (2002p.1071). And, finally Hommen and Doloreux also remark that “in theoretical terms, considering the intersection of individuals with high levels of knowledge and mechanisms of knowledge generation and circulation, in addition to processes fostering RIS would provide some insights and promote deeper understanding in connecting emergent processes of knowledge creation within the region and across a range of geographical scales” (2003 p.22).
Occupational analysis is aligned in many ways with the concept of ‘knowledge workers’, which has an extensive history, including the works of Peter Drucker (1995), Paul Romer (1995), Robert Reich (1991) and, more recently, Richard Florida (Florida 2002; 2005). In the work of all these authors, there are a group of workers that exist throughout the occupational categories, and who are variously named: symbolic analysts; creative workers; creative class; and knowledge workers. These occupations are seen to be based around key knowledge functions involved in the processing and analysis of information. These occupations have been shown to be closely aligned with the knowledge-based economy, and are particularly linked with innovative activity within firms, and the generation, transmission, and assimilation of knowledge (Sharpe and Martinez-Fernandez, Forthcoming).
The identification of specific knowledge workers (for example, scientists and engineers) with product innovation, especially of a scientific or technological
Chapter 4 – Economic and Industrial Audit 105 nature, is well established in innovation studies as one of the key proxies for innovative activity. Yet, as with many other proxies for innovation, this measure tends to privilege technological innovations, as opposed to the broader definition of innovation which also encompasses organisational processes. Workers involved in the provision of KIBS and KISA come from a broader range of occupational categories such as finance, banking, insurance, business services and marketing.
The traditional occupational groupings in the Australian Census do not readily adapt to this form of analysis. In previous work (Sharpe, Martinez-Fernandez et al 2004, Sharpe and Martinez-Fernandez 2006), a methodology to classify occupations based on the categories of Reich (1991; Cole 2004) is developed, and applied to the regional geography of Sydney. The methodology re-categorises 340 occupational categories from the Australian Standard Classification of Occupations (ASCO) into three broad types of occupations, which Reich refers to as: Symbolic Analysts (knowledge-based occupations in this case); In-person Service Workers; and Routine Workers. Through an analysis of ASCO descriptions, and matching these descriptions against key criteria established for the Reich categories, the occupational knowledge base is established.
Within these three categories, there are a number of further classifications. The knowledge-based occupation category is further subdivided, to enable classification of precise ‘knowledge bases’ that exist in the specific occupational groupings, in a manner similar to Asheim and Coenen (2005), and their classification of industry-based Nordic knowledge bases. Working on a principle expressed by Asheim and Coenen, and detailed in Chapter Two, different knowledge bases have different dynamics. The five identified sub-groups of occupational ‘knowledge-bases’ are: engineering and building; scientific; business and information; crafts and trades; and general management.
Chapter 4 – Economic and Industrial Audit 106 4.5.1 Knowledge based occupational analysis
Chart 4.2 shows the distribution of the identified knowledge-based occupations as a percentage of the total workers within each region and the comparison of SMA.
Chart 4.2 Knowledge based occupations as a percentage of total workforces
Sydney metro 41.30
Central West Sydney 27.85
North West Sydney 34.33
South West Sydney 27.00
0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 % of total workers
Source: Sharpe and Martinez-Fernandez (Forthcoming) based on ABS 2001 Census
From the graph, it is clear that all of the regions under investigation are lacking in overall ‘knowledge worker’ employment when compared to the average of metropolitan Sydney. North West Sydney, however, has approximately 6% more knowledge workers than the two other regions. These figures are supported by fact in as much as the NSW Government’s Metropolitan Strategy notes:
“Currently skill levels are unevenly distributed across Sydney…In particular there is a need to increase the numbers of full-time and highly skilled jobs. This is highlighted by the fact that eight per cent of jobs in Western Sydney are for professionals, whereas in the Sydney CBD, 40 per cent of the jobs are professionals. In Western Sydney, only 17 per cent of positions are in finance, banking and business services – those with the highest income potential – in comparison with 51 per cent located in the global economic corridor”(2005 p.77)
Chapter 4 – Economic and Industrial Audit 107 Chart 4.3 Types of Knowledge occupations by region
North West Sydney
Central West Sydney Scientific
General management
Engineering
Craft
South West Sydney Business and information
Total Knowledge Workers
Sydney metro
0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00
% of total workers
Source: Sharpe and Martinez-Fernandez (Forthcoming) based on ABS 2001 Census
Chart 4.3 delineates the distribution of knowledge workers into the five categories. In the discussion outlined within Chapter 2 of Ashiem and Coenen’s typology of RIS by knowledge base, they note two knowledge bases: analytical; and synthetic (Asheim and Coenen 2005; Asheim and Coenen 2006; Coenen, Moodysson et al. 2006). The authors argue that between these two knowledge bases lie fundamental differences in the processes of gathering and transferring knowledge. The breakdown of knowledge by occupational categories (by what people do, as opposed to what they make), as demonstrated in Chart 4.3, expands the number of knowledge bases to five. These include scientific and engineering knowledge bases that generally match Ashiem and Coenen’s analytical and scientific knowledge bases. Three other categories of knowledge base are also established: business and information; craft; and general management.
The business and information knowledge base refers to occupations that are focused on providing business and financial services, and information gathering and processing. This knowledge base is closely associated with certain forms of KISA and KIBS, as are the scientific and engineering categories. Craft knowledge
Chapter 4 – Economic and Industrial Audit 108 bases relate to occupations that are based on vocational bases and practices learnt through repeated experience. General management knowledge bases refer to occupations that are management focused, and acknowledges the unique set of occupational skills and knowledge that are associated with business management.
Central West Sydney has a strong engineering knowledge base and, to a lesser extent, so does South West Sydney. Scientific employment, the area reflecting more traditional measures of innovation, is in low levels in all of the regions, although it is highest in South West Sydney. The average scientific group at the SMA level is also slight and only fractionally higher (0.94% versus 0.84%) than the South West Sydney level. Craft based knowledges are also in higher levels in Central and South West Sydney. The levels of general management appear to be consistent across South West and Central West Sydney and the SMA, but slightly higher in North West Sydney. Therefore, whilst business and information knowledge is the most prevalent knowledge in the three regions, and indeed the SMA as well, both Central West and South West Sydney also have an engineering knowledge base. This could be related to higher levels of manufacturing employment in these regions. These differing knowledge bases can be expected to impact on the knowledge sourcing activities of the regions due to the conflicting dynamics associated with the various knowledge bases, as shown by Ashiem and Coenen (2006).
4.6 Summary
The aim in this chapter was to present background information for the three regions, and to highlight the context of this research, and the labour force and industrial differences within the regions under investigation. These three regions have absorbed significant population increases in recent history, but have not generated comparative levels of employment. South West Sydney exhibits the highest differentials between the two, and also has the highest projected population growth rate going forward for the two decades. Analysis in this chapter has also highlighted the fact that employment in these regions has suburbanised from the inner city, and arisen in response to the population demand, but in an
Chapter 4 – Economic and Industrial Audit 109 uneven manner. The statistics in this chapter have shown that certain industries have been disproportionately drawn to these regions, particularly the lower order service activities such as retail, and that these industries are noted for their high rates of casualisation and poor levels of income growth.
Manufacturing employment has also been drawn to the suburbs, predominantly more in the South West and Central West Sydney regions. The continued progress of manufacturing in these regions has been subjected to the twin forces of suburbanisation and deindustrialisation. The result of this is that although deindustrialisation of manufacturing has caused high job losses, especially in the last decade, as the firms suburbanised they became focused on local and niche markets that enabled them to prosper, and provide endogenous ‘home grown’ (Fagan and Dowling 2005) manufacturing developments in these regions.
Business and financial services employment has increased significantly throughout metropolitan Sydney over the last decade. All of the three regions have shared in this growth, however not to the same levels as the inner city areas. This sector is closely associated with Knowledge Intensive Business Services (KIBS), a key participating agent in innovation (Miles, Kastrinos et al. 1994; Koschatzky 1999; Muller and Zenker 2001; Thomi and Bohn 2003; Freel 2006; Martinez-Fernandez and Miles 2006). The association with these KIBS is one of the reasons for the exceptional growth in employment that the business and financial services sector has achieved, and demonstrates why the presence of these KIBS provides an indication of innovation levels. North West Sydney has experienced the strongest growth and concentration in this sector compared with the other two regions, where there is a concentration in manufacturing sector employment.
This chapter also uses an occupational analysis to establish the knowledge base of the three regions. Knowledge base is an important indicator of the type of knowledge flows in a RIS (Ashiem and Coenen 2006). These in turn influence knowledge gathering activities. In the knowledge base analysis, all of the regions had an underlying capacity in business and information knowledge, however this
Chapter 4 – Economic and Industrial Audit 110 was much more concentrated in North West Sydney, where it also coincided with a strong industrial business and financial services sector. South West and Central West Sydney on the other hand have a strong engineering knowledge base, which is associated with the stronger manufacturing industrial sector in the two regions.
The knowledge base analysis is the first of five identified elements that comprise the RIS analysis as established in Chapter 2. Table 4.15 begins to collate these characteristics.
Table 4.15 RIS analysis elements Characteristics Central West Sydney North West Sydney South West Sydney Industrial • Manufacturing • Business & • Manufacturing Knowledge Base • Engineering Financial Services • Engineering • Business & Information
The next chapter identifies the next two RIS elements; business innovation activity; and regional knowledge flows.
Chapter 4 – Economic and Industrial Audit 111
Chapter 5 – Industrial Innovation in Outer Western Sydney
As outlined in Chapter 3, the empirical results of this research are presented over three chapters. The statistical audit presented in the previous chapter provided the industrial and knowledge base for the three regions. The chapter also highlighted differences in the regional industrial and occupational structure, which when interpreted through the conceptual framework would lead us to hypothesise about corresponding differences in the innovation outputs and knowledge sourcing characteristics of the firms in these regions.
This chapter provides an understanding of firm innovation and knowledge flows, and their characteristics for the three regions. This chapter is concerned with the exploration of the Business innovation and Regional knowledge flow elements of the conceptual framework. These elements are operationalised through the first three research questions, presented below.
1. How does innovative activity manifest in the three regions?
2. What knowledge do firms access and from where?
3. What knowledge is external and what is internal?
Chapter 5 – Industrial Innovation in Outer Western Sydney 112 This chapter also tested for other factors which have been demonstrated in the literature to also affect innovation outputs and knowledge activities; namely firm size and industry sector. These factors need to be tested for so as conclusions drawn about regionally specific effects are exclusive of these other factors.
Hypothesis 1 – Firm size determines the mix of knowledge that firms access.
Hypothesis 2 – Industry sector affects the mix of knowledge that firms access.
In Chapter Two, the Regional Innovation Systems (RIS) approach was discussed at length. An analysis of previous RIS typologies produced a matrix of RIS features. Four key features were identified:
• Knowledge base
• Firm innovation
• Regional knowledge flow
• Regional orientation
The analysis presented in the previous chapter provided features of the regions’ knowledge base. This chapter focuses on firm innovation and knowledge flows, and their regional characteristics. This is achieved in two ways. Firstly, through an analysis of firms’ introduced innovations over the previous three years, including the type and novelty of the innovation; and secondly, through an analysis of firms’ knowledge gathering activities, in particular their use of knowledge sources and Knowledge Intensive Service Activities (KISA) and their mix, and the location of these activities.
This chapter is structured in three parts. The first part presents results of actual innovations that have been introduced by firms over the previous three years. This is part of the descriptive component of the research as outlined in Chapter
Chapter 5 – Industrial Innovation in Outer Western Sydney 113 Three, and in which it is necessary to establish, in the absence of other data, a platform of activities against which further analysis can be presented. The type of innovation, product, and service or process (either operational process or organisational process), and the degree of novelty of the innovations (radical or incremental) are presented. These counts will provide a checkpoint against which the analysis of knowledge sources, KISA usage, and locations can be presented. Discussions of these analyses are presented in the second and third sections. Finally, a summary section considers the research questions and hypotheses in the light of these results.
5.1 Innovation counts
This section presents the results of questions asked within the survey, regarding the actual introduction of new products, processes and services by firms in the previous three years. Ninety-four firms out of the one hundred and nineteen firms surveyed noted they had innovated in one of the three categories (some in more than one category) in the past three years. This equates to 79% of the firms surveyed.
There is some regional variation in the percentage of firms innovating, as demonstrated in Chart 5.1; with 80% of South West Sydney firms surveyed innovating in the last three years, compared with 67.8% in Central West Sydney, and 88.5% in North West Sydney. These variations in regional innovative activity are even more distinctive when the type of innovation being undertaken by firms in the region is considered.
In South West Sydney, 70.8% of firms surveyed were engaged in product or service innovation, with a lesser number of firms innovating in operational processes (63.1%), and organisational processes (67.7%). Central West Sydney, as expected because of its lower overall rates of innovation, has lower levels in all categories: its highest levels (57.1%) are both in the form of process innovation -operational and organisational. North West Sydney has the highest
Chapter 5 – Industrial Innovation in Outer Western Sydney 114 levels of overall innovations across all three categories when compared with the other regions. All the innovating firms surveyed in North West Sydney were involved in operational process innovations (88.5%), and most (80.8%) were also involved in organisational process innovations as well. High levels of firms participating in product innovation in North West Sydney suggest that many firms were in fact innovating in all three categories. Chart 5.1 shows the breakdown of innovations by type for the firms surveyed, and shows further diversity of innovative activity at the regional level.
Chart 5.1 Innovations by type and by region
100.00%
90.00%
80.00%
70.00%
60.00%
Product or Service Innovation 50.00% Operational process Innovation Organisational Process Innovation
40.00%
30.00%
20.00%
10.00%
0.00% South West Sydney Central West Sydney North West Sydney
Source: Outer Western Sydney Innovation Survey. Multiple responses in innovation type allowed, n=119
Hence, even from the outset there are significant regional variations in innovative activity. This is in line with the industrial profiles presented in the preceding chapter. In South West Sydney, where the industry base in concentrated in manufacturing, there is more product and service innovation whereas in North West Sydney, which has higher concentrations of business and financial services, there are higher levels of process innovation, particularly operational process innovation. One factor that contributes to firm innovativeness is the position of the firm in the firm life cycle. Newly established firms have higher rates of innovation as the new firms will naturally bring new products, services and
Chapter 5 – Industrial Innovation in Outer Western Sydney 115 processed to market. Chart 5.2 shows innovation activities across three categories of firm life cycle: newly established firms (in operational for less than 4 years), established firms (in operation for between 5-9 years) and mature firms (in operation for 9 or more years). These firm categories are compared with innovation activities.
The chart confirms the innovativeness of newly established firms with these firms exhibiting the highest levels of innovation in all categories. The established firm category had the lowest level of innovation in all categories, but particularly in product and service innovation. Innovation levels increase as firms progress to mature firms, although not to the same levels seen in the newly established firms. This suggests that established firms go through a period of consolidation of their activities with less focus on developing new products, services and processes but that a focus on innovation return as the firm matures. New firms are therefore an important source of innovative activity in the regions.
Chart 5.2 Business cycle and firm innovation activity
90.0%
80.0%
70.0%
60.0%
50.0% New firms (1-4 years) Established firms (5-9 years) Mature firms (9+ years) 40.0%
30.0%
20.0%
10.0%
0.0% Product and service innovation Operational process innovation Organisational process innovation
Source: Outer Western Sydney Innovation Survey. Multiple responses in innovation type allowed, n=119
Chapter 5 – Industrial Innovation in Outer Western Sydney 116 Two other factors in addition to the regional factor are also considered in this innovation analysis. They are: firm size; and broad industry base. Firm size is identified as an influencing factor on firm innovation activity. Chart 5.3 shows innovative activity by both innovation type and business size (based on number of employees). The categories of business size used in this chart allow for aggregations much smaller than is usual in business surveys. In these instances, small enterprises could be considered to be firms with less than 100 or 50 employees. Many innovations surveys exclude altogether firms that have less than 5 employees, including the Australia Business Innovation Survey (ABS, 2003). In the case of the current survey sample, businesses with 1-5 employees make up 26.1% of the firms surveyed, with firms of 6-20 employees contributing a further 25.2%. In fact, firms with 50 employees or less make up 65.6% of the sample. This is consistent with the general firm population in the regions. Across all of the various business size categories, this size of firm (ranging between 1 and 50 employees) includes the three most innovative groups, and in reverse size order, with the smallest firms displaying the highest levels of innovation counts, and consistently high counts across all three innovating categories.
The innovativeness of small firms is well noted in the literature (Acs and Audretsch 1990; de Jong and Marsili 2006). This is due to the differences in the activities of these small firms regarding their knowledge gathering (de Jong and Marsili 2006; Fontana, Geuna et al. 2006; Macpherson and Holt 2007). It is argued that small firms, due to their lower levels of internal resources, are more likely to rely on external knowledge sources for their knowledge gathering and innovative activities. This in turn makes these external sources, their location, and the mix utilised an important consideration in the effective analysis of innovation and resultant policy making. With all three regions having a high population of small firms, ascertaining the validity of this association (between external resources and small firms) is critically important and is therefore addressed in the first of two hypotheses to be tested in this thesis.
Chapter 5 – Industrial Innovation in Outer Western Sydney 117 In examining Chart 5.3, there is a clear point of difference in operational process innovations. These types of innovations were introduced by more than 70% of firms with less than fifty employees (higher in smaller firms), but the level of operational process innovation plummets to 50% for firms with between 51-100 employees, before again rising to above 70% for large firms with an excess of 100 employees. This suggests a reduction in the importance of this type of innovation (as opposed to the other types of innovation, product and services, and organisational process, which both remain high) at a certain firm size, but then a return to prominence as the firm grows beyond a critical level (in excess of 100 employees).
Chart 5.3 Innovation by type and by business size
90.0%
80.0%
70.0%
60.0%
50.0% Product or Service Innovation Operational process Innovation Organisational Process Innovation 40.0%
30.0%
20.0%
10.0%
0.0% 1-5 employees 6-20 employees 21-50 employees 51-100 100 plus Not stated employees employees
Source: Outer Western Sydney Innovation Survey. n=119
The position of the firm within the firm life cycle is also relevant here in that many smaller firms may be newly established firms. An analysis of firm life cycle of the small firms surveyed however shows that only 23.7% of smaller firms are newly established, 23.7% fall into the established firm category with the majority (52.6%) of firms in the mature firm category. Therefore the high level of innovativeness in small firms shown in Chart 5.3 can be attributed to the
Chapter 5 – Industrial Innovation in Outer Western Sydney 118 high number of mature small firms. These firms as shown in Chart 5.2 also have high levels of innovativeness.
The second factor, industry base, also has an extensive history within the systems of innovation approach, both through the sectoral innovations systems (Breschi 2000; Malerba 2005) and the underlying industry focus of the first RIS iteration, and through other recent literature on clusters. There can be no doubt that the type of industry is a critical factor in the innovation and knowledge gathering activities, however, as discussed in Chapter Two, RIS includes a number of different industries and industrial concentrations, and as the importance of KISA highlights, a variety of industry sectors are now contained within firms’ innovation and knowledge gathering spheres. Therefore, this research will consider industry as a factor in the variation of innovative activities, but only inasmuch as it provides a focus for identifying the KISA and knowledge sources that operate across industry within the regions rather than the opposite approach, which was the typical approach of the first RIS iteration and ‘cluster’ analysis focus used in many regions. Innovation type by broad industry category is shown in Chart 5.4.
Chart 5.4 Innovation by type and by broad industry category
90.0%
80.0%
70.0%
60.0%
50.0% Product or Service Innovation Operational process Innovation Organisational Process Innovation 40.0%
30.0%
20.0%
10.0%
0.0% Construction Manufacturing Transport & Business & Financial Other Distribution Services
Source: Outer Western Sydney Innovation Survey. n=119
Chapter 5 – Industrial Innovation in Outer Western Sydney 119 The highest percentage of innovation across both innovating industries and types of innovation is the Construction industry in operational process innovation, with 83.3% of firms surveyed innovating in this category. The prominence of operational process innovation may be a result of the ‘project management’ requirements of the industry to logistically manage multiple suppliers. Construction firms, however, contribute a relatively small percentage to the overall firm sample, and therefore these results are only from a small number of firms and may not produce a reliable picture of innovation in this industry.
The next highest percentage comes from Business and Financial Services firms, in organisational process innovations (77.1%), followed by Manufacturing in product and service innovation (76.5%), and then Transport and Distribution (75.0%), also in organisation process innovation. In both Manufacturing, and Business and Financial Services, over 60% of the firms in each category were innovating in all three categories, suggesting some relationship between the three types of innovation within firms in these two industry sectors. With the majority of surveyed firms either from the manufacturing or business services sectors, a regional breakdown of innovative activity is shown for these two sectors in Chart 5.5.
Within manufacturing firms over the three regions, there is an emphasis on product and service manufacturing, with over 70% of all manufacturing firms in each of the three regions having introduced a product or service innovation in the previous three years. The highest levels were in Central West Sydney where 81.8% of manufacturing firms reported product and service innovations. North West Sydney manufacturing firms had higher levels of operational process innovation, and the same level of organisational process innovations as product and service innovations. Therefore, North West Sydney manufacturing firms are already distinguished from the other two regions in terms of their innovative activities, with the other two regions having similar distributions of innovations.
Chapter 5 – Industrial Innovation in Outer Western Sydney 120 Chart 5.5 Innovating firms by region by innovation types
100
90
80
70
60
Product and service 50 Operational process Organisational process
40 % % of firms innovating
30
20
10
0 Manufacturing Finance and Manufacturing Finance and Manufacturing Finance and Business Business Business services services services Central West Sydney North West Sydney South West Sydney
Source: Outer Western Sydney Business Innovation Survey n=119
In the finance and business service categories, differences between all three regions emerge. Central West Sydney is marked by overall low levels of innovating in all categories, with only 50% of surveyed financial and business service firms innovating in products and services, and even less in the other two categories (37.5%). North West Sydney, on the other hand, has the highest overall levels of innovation of either industry in its financial and business service firms, recording levels of 91.7% in both the operational and organisational process category, and also showing the highest levels of product and service innovating across all three regions in the finance and business service industries (75% of surveyed firms). South West Sydney lay somewhere in-between, reporting the highest levels being in organisational process innovation, with 82.8% of business service firms innovating in this field, and this was followed by equal percentages of firms innovating in the other two categories.
Chapter 5 – Industrial Innovation in Outer Western Sydney 121 5.1.1 Novelty of innovation
The diversity between the three regions continues when we examine the novelty of the innovations. The degree of novelty refers to the impact of the innovation on the firm involved. Firms were asked to nominate for each of the categories within which they were innovating, whether they considered the introduced innovations to be incremental (small improvements to existing products, services and processes), or radical (completely new products, services and processes requiring complete changes to firm practices). Chart 5.6 shows the types and novelty of innovations for each of the three regions.
Surprisingly, considering the previous results, Central West Sydney has the highest levels of radical innovation in each of the three categories. In marked difference to the other two regions, the highest levels are in the operational process category. The product and service innovation category also exhibits a large variation between the three regions, with North West Sydney firms exhibiting higher levels of incremental innovation.
Central West Sydney’s high level of radical innovation, despite overall lower results in the previous two charts, shows that the innovative activity these firms are undertaking is more linked to the creation of completely new product and process developments. Radical innovative activity is associated with specialised KISA; therefore corresponding differences in the usage of KISA and the mix of KISA are expected.
Chapter 5 – Industrial Innovation in Outer Western Sydney 122 Chart 5.6 Novelty of innovation by type and region
60.0%
50.0%
40.0%
30.0%
20.0%
10.0%
0.0% Incremental Radical Incremental Radical Incremental Radical Product & Service Innovation Operational Process Innovation Organisational Process Innovation
South West Sydney Central West Sydney North West Sydney
Source: Outer Western Sydney Innovation Survey. n=119
5.1.2. Innovation implications
This section has presented results on firm innovations that have been introduced in the previous three years, by region, firm size, position in firm life cycle and industry, as well as by type, and degree of novelty. These figures show the manifestations of innovation activities across the three regions. Together with the industrial profiles and the knowledge base characteristics presented in Chapter Four, these will provide an initial platform of innovative activities in the three regions, against which further analysis on KISA and knowledge sourcing can be tested.
The results also show the firm innovation characteristics of the RIS. The earlier expectations of differential regional innovation activity based on the differing industrial and occupational structure has been borne out in these results. There is no suggestion of causality in this relationship. In fact the firm characteristics of size and age also had a strong (positive in both cases) association with higher levels of innovative activity.
Chapter 5 – Industrial Innovation in Outer Western Sydney 123 The one unexpected result was the high level of radical innovation in central western Sydney, despite lower overall levels of total innovation activity and a strong manufacturing knowledge base (usually associated with incremental innovation) and lower levels of occupational knowledge workers (not associated with high levels of R&D and radical innovation). The further analyses of knowledge will be necessary in unpacking these results.
The next section analyses knowledge sources used by firms for their innovative activities. The relationship between knowledge sourcing and innovation means these variations should also be evident in the knowledge sources accessed. Knowledge source analysis not only provides an indication of the sources firms use, but because there is some understanding of the type of knowledge that different sources provide (Brokel 2007) and how it is transferred (Nonaka & Takeychi 1995, Polanyi 1966), they provide indications of the knowledge flows in the firms and regions. This is the next of the five RIS features to be analysed.
5.2 Knowledge sources
This second section covers firms’ activities regarding their knowledge searching and learning processes. As discussed earlier, innovation is inextricably linked to a broader field of activities undertaken by firms in acquiring information. Within the survey, firms were asked questions about knowledge sources they used when innovating, and the importance of these sources. Chart 5.7 shows the spread of knowledge sources accessed by firms concerning their innovative activities, and the importance of each of these sources.
Chapter 5 – Industrial Innovation in Outer Western Sydney 124 Chart 5.7 Knowledge sources in firms by level of importance
60.0%
50.0%
40.0%
30.0%
20.0%
10.0%
0.0% Within the Other parts of Clients and Suppliers Competitors Consultants Universities & Government Professional Websites, business the same Customers and paid Higher Agencies conferences, journals etc enterprise advisers Education meetings & group fairs etc
Not Relevant Low Importance Medium Importance High Importance
Source: Outer Western Sydney Business Survey, n=119
The most important knowledge source identified by firms is internal knowledge, ‘within the business’ sources, with 65.5% of all firms listing them as of high or medium importance to their innovative activities. The supply chain in the form of clients and customers, suppliers and competitors also formed an important block of knowledge sources for firms. These two sources; internal and supply chain, form the bulk of knowledge sources for firms. These results are in line with the findings of much other innovation scholarship and innovation surveys (For example see IBM 2006).
In addition, there were a number of other sources identified as being important to firms, but at much lower percentages. Other organisations and institutions within the RIS framework, such as universities and higher education institutions, consultants and paid advisers, and government agencies all have mixed levels of usage and importance. 21.8% of firms regarded consultants and paid advisers as being of high or medium importance as knowledge sources, then 16.8% for universities, and finally 13.5% for government agencies. The low levels of knowledge sourcing from universities, or conversely the high numbers of firms who list that universities are not a relevant source of knowledge, is interesting. In
Chapter 5 – Industrial Innovation in Outer Western Sydney 125 much RIS work, universities and other public research institutions are noted as the key external knowledge sources of firms, especially small firms (Leydesdorff 2006; Leydesdorff, Dolfsma et al. 2006). The types of knowledge services that firms are seeking and why in this case universities do not seem to be providing these services, or again, conversely the concept that firms do not know these knowledge sources exist, is an important policy consideration. Brokel (2007) noted that knowing where to find some knowledge is just as important as finding it, and that tacit knowledge transfer is usually required in knowing where to look.
Industry specific sources such as conferences and meetings, websites and journals were also highly regarded as knowledge sources, with 32.7% of firms describing conferences, exhibitions, fairs and meetings as of high/medium importance, and 30.3% regarding websites and journals as important. These knowledge sources represent two dimensions of interactivity. The first, participation in conferences, fairs and meetings, is highly interactive. Whereas knowledge sourcing on the internet through websites and journals, is the opposite type of interaction, being isolated in terms of face-to-face interactivity.
5.2.1 Knowledge sources and region, firm size and industry
Examination of the breakdown of knowledge sourcing (sources regarded as of high or medium importance by firms) by region is portrayed in Chart 5.8. The results presented in Chart 5.6 outlining overall knowledge sources showed that the most important sources were internal and supply chain, and this is also now shown to be consistent across all three regions.
A degree of regional variation exists in a number of knowledge sourcing areas. Central West Sydney has the lowest levels of access in all but three sources: universities and higher education institutions; government agencies; and journals and websites. These results support earlier findings regarding the higher levels of radical innovation present in Central West Sydney firms. University based
Chapter 5 – Industrial Innovation in Outer Western Sydney 126 knowledge sourcing is often associated with new technological knowledge. Similarly, journals and websites provide an avenue for disseminating and receiving specialised codified knowledge for and from peer group review.
Chart 5.8 Knowledge sources (rated high and medium importance) by region
100.0%
90.0%
80.0%
70.0%
60.0%
50.0%
40.0%
30.0%
20.0%
10.0%
0.0% Within the Other parts of Clients and Suppliers Competitors Consultants Universities & Government Professional Websites, business the same Customers and paid Higher Agencies conferences, journals etc enterprise advisers Education meetings & group fairs etc
South West Sydney Central West Sydney North West Sydney
Source: Outer Western Sydney Business Survey, n=119
Both South West and North West Sydney show just under a third of firms accessing knowledge from within an enterprise group, suggesting a number of firms are operating in a subsidiary relationship with other firms. Also, firms in both of these regions have similar percentages accessing knowledge from competitors. South West Sydney firms showed stronger supplier relationships, with 33.8% of firms using suppliers as an important source of knowledge, compared with 21.4% in Central West Sydney, and 23.1% in North West Sydney. South West Sydney firms also used more consultants and paid advisers. In North West Sydney, 42.3% of firms listed conferences, exhibitions and fairs as an important knowledge source, compared with 29.7% of firms in South West Sydney, and 17.9% in Central West Sydney.
Chapter 5 – Industrial Innovation in Outer Western Sydney 127 Chart 5.9 Knowledge sources (rated high and medium importance) by business size
100.0%
90.0%
80.0%
70.0%
60.0%
50.0%
40.0%
30.0%
20.0%
10.0%
0.0% Within the Other parts Clients and Suppliers Competitors Consultants Universities Government Conferences Websites business of the customers agencies enterprise group Small Firms <50 Large Firms >50
Source: Outer Western Sydney Business Survey, n=119
Knowledge sourcing by firm size is the subject of Chart 5.9. This chart also follows the trend of the previous two charts, with internal and supply chain sources dominating. The immediately evident result displayed by this chart, is the overall higher levels of knowledge sourcing undertaken by small firms compared with larger firms25. One of the hypotheses to be tested in this research is that small firms, because they lack the internal resources of their larger counterparts, use higher levels of external knowledge sourcing. This chart shows that smaller firms do have more external knowledge sourcing than larger firms. This includes sourcing based around the supply chain through customers and clients, and competitors and suppliers, but also through conferences and exhibitions, and journals and websites. This sourcing is not surprising in that they offer smaller firms ready access to new knowledge at an acceptable cost to the firm. Maskell, Bathelt et al (2006) have referred to conferences and exhibitions etc as “temporary clusters” and “hotspots of intense knowledge exchange,
25 The arbitrary break of firms into two groups of 1-50 employees and 50+ employees has some limitations in its analysis of larger firms, as this group entails a wide variety of firms - in the case of this survey, ranging from 50 employees up to over 1000. However, in this instance small firms are the primary emphasis of the analysis, and their range is well covered by the other group (1-50 employees).
Chapter 5 – Industrial Innovation in Outer Western Sydney 128 network building and idea generation” (2006: p997), overriding the need for permanent geographical proximity to enable tacit knowledge transfer, through regular temporary proximity. On the other hand, journals and websites offer ready and fast access to emerging research, products and trends. Together, they offer firms the opportunity to participate in both face-to-face (tacit), and codified knowledge transfer for a reasonable cost to a small firm.
Overall, this means smaller firms are accessing knowledge not only from their supply chain, but also from more specialist and public sector knowledge sources, at a much higher level than larger firms. These sources are also those most likely to be located outside of the regions.
All this knowledge sourcing clearly shows that smaller firms do have more external accessing of knowledge. These firms have higher levels overall of sourcing, both internally and externally, which suggests that although the external sources are important for smaller firms, they are also linked closely to internal resources. These internal resources, the staff of the firm, are vital in appropriating and interpreting the knowledge gained from these various external sources, and then implementing this knowledge into the firms’ innovative activities. The role of these knowledge professionals is addressed in further detail in Chapter 7, and is central to understanding the dimensions of knowledge sourcing in firms.
The third element of knowledge sourcing to be addressed in this section is that of firms by industry. The results for the two main broad industry categories represented in the survey sample, manufacturing, and business and financial services, are displayed in Chart 5.10. Again, the same trend as noted above is reflected in the statistics, with internal sources and the supply chain forming the primary areas of knowledge sourcing. However, there is not the same degree of variation showing between the two industries as was displayed in the figures relating to the business size of the firm.
Chapter 5 – Industrial Innovation in Outer Western Sydney 129 Chart 5.10 Knowledge sources (rated high and medium importance) by broad industry category
100.0%
90.0%
80.0%
70.0%
60.0%
50.0%
40.0%
30.0%
20.0%
10.0%
0.0% Within the Other parts Clients and Suppliers Competitors Consultants Universities Government Conferences Websites business of the customers agencies enterprise group
Manufacturing Business services
Source: Outer Western Sydney Business Survey, n=119
The key areas of variation are the manufacturing firms’ higher levels of knowledge sourcing from clients and customers, enterprise groups, and suppliers. In comparison, business service firms had higher levels of sourcing from government agencies, conferences and exhibitions, and journals and websites. These results are consistent with the different “technological regimes” (Breschi 2000) of these industries which, as the sectoral innovation systems literature highlights, affects the innovative activities and hence knowledge sourcing of firms in these differing sectors (Breschi 2000, Breschi & Malerba et al 2000, Malerba 2005).
These differences, however, are not as distinct as those exhibited in the comparisons between small and large firms that were presented in the previous chart (Chart 5.9). Therefore, the second hypothesis of this thesis, that regional differences in knowledge sources and the mix of internal and external sources results from different industrial environments would, in this first instance, not
Chapter 5 – Industrial Innovation in Outer Western Sydney 130 appear to be supported, especially in comparison to the analysis of firms by their size.
5.2.2Factor analysis of knowledge sources
Principal components analysis is used to identify the factors. In order to do this, the statistical programme (in this case SPSS), standardises each available item score to a mean of zero and a standard deviation of one. From this each factor is given an Eigen value which provides the weight of the factor in explaining the total variance (Hinton 2004).
The analysis to date has focussed on examining individual knowledge sources, but clearly these sources are used in combination with each other. Firms also show differing levels of interactivity in their knowledge sourcing, from highly interactive relationships with customers and suppliers, to isolated activities, like using websites and journals. Identifying the mix of sources that firms are accessing would provide a better understanding of overall firm strategy in regards to knowledge sourcing. The goal of factor analysis is data reduction: explaining the maximum amount of variance with the smallest number of concepts, to bring order to the data and allow for interpretation (Kerlinger 1979). Factor analysis assumes that beneath the differences of multiple variables lie a smaller set of latent factors or components that can explain this result. The analysis summarises “the interrelationships among variables in a concise but accurate manner as an aid in conceptualisation” (Gorsuch 1983 p.2).
Three key decisions need to be made regarding the factors analysis: firstly, the method of extraction; secondly, the number of factors to be extracted; and thirdly, the rotation method to be employed on the extracted factors. The extraction method used in this research in Principal Components Analysis (PCA). PCA is a mathematical technique used to identify the underlying structure characterising a set of highly correlated variables. PCA is used to
Chapter 5 – Industrial Innovation in Outer Western Sydney 131 identify any groups of knowledge sources within the responses of firms and is especially suited to data reduction.
The second decision to be made is the number of factors to extract. This is a matter of judgement; however there are two available methods to inform this judgement. Each extracted factor accounts for a certain amount of the variance. Ideally, the number of extracted factors should explain as much of the variation as possible, but also reduce the number of variables. Eigenvalues state the amount of variance accounted for by extracted factors[MT3]. The eigenvalues are then used in the Kaiser criterion. The criterion states that variables with factors of an eigenvalue of 1 or more (to account for the same or more of the variance of the original variable) provide the number of factors to be extracted. The second method also uses eigenvalues, but this time plotted on a simple line graph. The scree test (Cattell 1966) then determines the place on the graph where the line begins to plateau, and just on the left before this spot, provides the number of factors to be extracted.
In addition to assessing both of these methods, the overall principle of factor analysis also needs to be kept in mind, to reduce the data but in a way that makes sense, and the preference to ‘over-factor’ rather than ‘under-factor’ (Fabringer, Wegener et al. 1999) needs to be considered. It is because of the principle of factor analysis, that a number of versions of extraction may be necessary to discover the best fit. This was so in this case, when both methods were trialled, the Kaiser criterion suggested three factors while the scree test (scree plot shown in appendix 7) suggested four factors. Both three and four factors were extracted from the variables, and the four factor solution provided the best ‘fit’ for results and accounted for 71.5% of the variance between the variables.
The third decision to be made is the method of rotation. The goal of rotation is to maximise the variations between the factors, to gain a clearer pattern of the factor loadings. The rotation method selected was the Varimax method, as this method
Chapter 5 – Industrial Innovation in Outer Western Sydney 132 maximises the differences between the factors while minimising the differences within factors. This is also the most commonly used rotation method.
The PCA is an iterative process, therefore with these decisions made, there were some further basic factor analysis requirements to be met for successful completion. The data met some of these requirements from the outset, including a sample size in excess of 50, and a ratio of variables to sample of 5 to 1. The correlation matrix for the analysis must have two or more correlations greater than 0.30, which also existed (correlation matrix shown in appendix 7), with multiple correlations above 0.30 already present.
Two tests to ascertain the goodness-of-fit, or factorability of the variables, the Kaiser-Meyer-Olkin Measure of Sampling Adequacy (KMO), and the Bartlett’s test of Sphericity, were also performed. The KMO proved to equal 0.805. This falls within the range described the by Degree of Common Variance as ‘meritorious’ (values between 0.80 and 0.89 (Kaiser, Meyer et al)), and the Bartlett’s Test for Sphericity is statistically significant (both are shown in Appendix 7).
With these requirements met, the factor analysis continued. The next step was to derive a pattern of relationships between the variables, ensuring the variables derived explain 50% of the variance, so variables need to have a communality of 0.50 or greater. The Communalities table (Appendix 7) shows the communality of the variables all accounting for more than 0.50 of the variance. The factor analysis could therefore continue, and the four factors were extracted. The highly correlated factor loadings (.05 or higher) from each factor within the rotated component matrix are shown in Table 5.2.
The factor loadings from the component matrix allow some initial interpretation of the four extracted components. Component 1 is the primary factor, accounting for 39% of the variance, and including the variables: internal knowledge sources;
Chapter 5 – Industrial Innovation in Outer Western Sydney 133 clients; customers; and competitors. This result is consistent with the earlier charts that identified these sources as the highest sourced forms of knowledge. This component demonstrates that all external knowledge comes from customers and clients. This component is referred to as customer focused knowledge sources.
Table 5.2 Rotated Component Matrix for Knowledge Sources Component Analysis Knowledge sources Components Customer Enterprise Public Sector Commercial focused focused focused knowledge network
Within the business .706 Other parts of enterprise group . .789
Clients and customers .677 Suppliers .705 Competitors .853 Consultants and paid advisers .590 Universities .820 Government agencies .824 Conferences, Meetings, Fairs etc .821 Journals and websites .633 Extraction Method: Principal Components Analysis Rotation Method: Varimax with Kaiser Normalization
The second component accounts for 13.4% of the variance and is correlated with knowledge source variables: other parts of the enterprise group (head office, enterprise research and development departments); suppliers; and consultants and paid advisers. In contrast to these previous component, this group sources knowledge from an enterprise group that is similar to internal sources, from suppliers, and from consultants and paid advisers. The sourcing from other parts of the enterprise group suggests these firms belong to larger organisations and may be in a subsidiary or branch office role, thus relationships with supplier and consultants may be influenced by this subsidiary role. This component is titled the enterprise focused component.
The third component is correlated with public sector elements of the innovation system: universities; and government agencies, and hence is called the public
Chapter 5 – Industrial Innovation in Outer Western Sydney 134 sector focused component. This component accounts for 11.3% of the variance. In this component, external knowledge sourcing is more in line with traditional knowledge sourcing for firm innovation.
The final component and the smallest extracted, accounting for only 7.8% of the variance, is correlated with conferences, meetings and fairs, and journals and websites. These two sources represent both tacit and codified knowledge sourcing. As was mentioned in Chapter Two, conferences, fairs etc offer the opportunity for the transfer of tacit knowledge through the creation of “temporary clusters” (Maskell, Bathelt et al 2006). They provide participants with similar benefits to the advantages experienced through industrial cluster situations, which are much noted in the literature. The use of journals and websites complements this external knowledge sourcing through meetings, and leads to this component being referred to as the commercial information network focused component.
5.2.3 Knowledge source implications
From this analysis of knowledge sources, the following conclusion can be drawn: that the internal resources of firms, their supply chain relationships, and commercial sources of knowledge including consultants and paid advisers, and participation in conferences and exhibitions and subscriptions to journal and websites, form the core of their knowledge sources.
The analysis also shows different avenues of knowledge sourcing in the regions. Most notably, Central West Sydney’s comparatively high sourcing from public sector sources (universities and government agencies), which is possibly due to their different knowledge needs because of their higher levels of radical innovation. South West Sydney exhibited stronger supplier knowledge sourcing than the other two regions, which again may be due to their focus on incremental product and service innovation. North West Sydney had the highest overall levels
Chapter 5 – Industrial Innovation in Outer Western Sydney 135 of sourcing in most categories, but was much higher in the conferences, meetings, exhibitions etc sources compared to the other two regions.
The factor analysis of knowledge sources identified four different knowledge sourcing strategies of firms. These four factors account for over 70% of the variance in the responses provided to the eleven knowledge source variables. They identify four key arenas of external knowledge sourcing: (i) customers and competitors; (ii) wider enterprise groups and paid consultants; (iii) public sector institutions; and (iv) commercial information networks.
These activities are closely aligned with knowledge intensive service activities, and these activities are the subject of further analysis in the following section.
5.3 Knowledge Intensive Service Activities
This section explores the nature and use of knowledge intensive service activities within the firms and regions under investigation. As discussed in Chapter Two, the broader definition of innovation is strongly linked with service activities, particularly Knowledge Intensive Service Activities - KISA (OECD 2006), and therefore an analysis of how firms use these activities, and from where they source them, provides an insight into the manifestation of knowledge flows in firms. To review the definition of KISA:
“KISA refers to the production or integration of service activities, undertaken by firms and public sector actors – in the context of manufacturing or services, in combination with manufactured outputs or as a stand alone services” (OECD, 2006, p31)
In this section, three aspects of KISA will be analysed. Firstly, the overall usage of different types of KISA across the three regions and by firm sizes and industry will be examined. Secondly, factor analysis will again be used to establish patterns in the multiple usage of KISA. Thirdly, the location of KISA providers
Chapter 5 – Industrial Innovation in Outer Western Sydney 136 in the three regions will be investigated, to assist with drawing conclusions about not only the type of knowledge flows surrounding firms, but also the geography of these flows.
KISA covers many service activity types, including accounting and financial services, legal services, IT services, and research and development (R&D) services. Within the survey, firms were asked about their usage of these service activities and, if they were used, to rate the importance of these services to their firm innovation. The most cited KISA of high or medium importance was Marketing and Promotion (52.1% of all firms). R&D services followed, with 39.3% of all firms using and highly rating these services, and IT services were also utilised by 39.3% of all firms. Chart 5.11 shows the percentage, by type used, of firms using KISA.
Chart 5.11 KISA usages by type
E-commerce
Legal services (IP, patents etc)
Accreditation
Recruitment
Training services (TAFE etc)
I.T. services
Accounting & Financial services
Research (including market research)
Marketing & Promotion
Business Planning advice
Industry development advice
0% 10% 20% 30% 40% 50% 60%
Source: Outer Western Sydney Business Innovation Survey, n=119, multiple answers allowed
5.3.1 Region, firm size and industry size usage of KISA
Chart 5.12 shows the percentages of firms using different types of KISA in the three regions. From the previous chart (Chart 5.11), the most used KISA were
Chapter 5 – Industrial Innovation in Outer Western Sydney 137 marketing and promotion, R&D and IT services. The first two of these, marketing and promotion and R&D, provide the areas of most variance between the three regions. Marketing and promotions related KISA were accessed by 58.5% of South West Sydney firms, by 57.0% of North West Sydney firms, but by only 32.1% of Central West Sydney firms. Both North West and South West had similar trends in the highest used KISA.
Central West Sydney has a different profile to the other two regions. In addition to lower levels of overall usage, R&D and IT services are the most accessed formats, with both being utilised by 39.3% of firms. This is then followed by marketing and promotions. The earlier results of Central West Sydney’s higher levels (comparatively) of radical innovation may be behind these different results in KISA access, with such innovation requiring a higher focus on R&D and IT services (where it has the highest levels of usage across all regions). The usage of specialist legal services KISA for patents and IP is also closely aligned with radical innovative activity. Central West Sydney also has the highest levels of usage of this type of KISA. 28.6% of firms in Central West Sydney accessing the specialist legal services KISA, compared with 23.1% in South West Sydney, and 19.2% in North West Sydney.
The usage of a number of other sources of KISA: industry planning advice; business planning advice; and financial services, also varies across the regions. South West Sydney firms accessed financial KISA at more than twice the rate of North West Sydney firms (35.4% versus 15.4%), and both South West and North West Sydney have more than 30% of firms using business planning KISA, compared with only 21.4% of Central West Sydney firms. North West Sydney has more than 50% of firms accessing marketing and promotion and R&D. This region also has the highest levels of firms accessing: business planning advice (34.6%); accreditation (34.6%); and recruitment services (30.8%). South West Sydney also has high levels of marketing and promotion, and R&D usage, but also high levels of accounting and financial KISA (more than double the levels of North West Sydney), training services (27.7%), and e-commerce (27.7%).
Chapter 5 – Industrial Innovation in Outer Western Sydney 138 Chart 5.12 KISA usages by type by region
E-commerce
Legal services (IP, patents etc)
Accreditation
Recruitment
Training services (TAFE etc)
I.T. services
Accounting & Financial services
Research (including market research)
Marketing & Promotion
Business Planning advice
Industry development advice
0.00% 10.00% 20.00% 30.00% 40.00% 50.00% 60.00% 70.00% South West Sydney Central West Sydney North West Sydney Source: Outer Western Sydney Business Innovation Survey, n=119, multiple answers allowed
Analysis of the KISA usage starts to build up a picture of the type of knowledge accessed by firms in the region, and in the case of Chart 5.12, a pattern also emerges of the KISA regarded as important by firms, and the regional differences between these important KISA.
The next two charts (Chart 5.13 and 5.14) provide the same comparisons between small and large firms, and between manufacturing and business services firms, to see if similar differences are exhibited. Examining Chart 5.13, once again it is evident that small firms are accessing much higher levels of KISA.
Chapter 5 – Industrial Innovation in Outer Western Sydney 139 Chart 5.13 KISA usages by type by firm size
E-commerce
Legal services (IP, patents etc)
Accreditation
Recruitment
Training services (TAFE etc)
I.T. services
Accounting & Financial services
Research (including market research)
Marketing & Promotion
Business Planning advice
Industry development advice
0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% 70.0%
Other enterprises (over 50 employees) SME (Under 50 employees)
Source: Outer Western Sydney Business Innovation Survey, n=119, multiple answers allowed
The accessing of KISA does not have to be an external exercise. As will be shown in the following section on KISA location, internal sources provide the majority of KISA, therefore the variance in the level of KISA accessing between the two firm size categories is surprising. Looking at the overall trend for higher usages, marketing and promotion, and R&D are still the top two in both categories, just at much lower levels in larger firms than smaller ones. Marketing and promotion services were accessed by 62.0% of smaller firms compared with 33.3% of larger firms, and R&D were accessed by 53.2% of small firms compared with 28.2% of larger firms.
Smaller sized firms had higher levels of every type of KISA access except for recruitment services, with 23.2% of larger firms using these services compared with 20.3% of smaller firms. Recruitment services were the third most accessed type of KISA for larger firms. This is understandable given that the delineation of firm size is by number of employees. Therefore, this chart gives an indication that not only is KISA more widely accessed by smaller firms, but the type of
Chapter 5 – Industrial Innovation in Outer Western Sydney 140 KISA, and consequently the type of knowledge most important to these firms, is different.
Chart 5.14 provides the same comparison of KISA type, but this time by the two main industrial categories: manufacturing; and business and financial services. More variance in KISA usage is exhibited between the two industries than was exhibited in the earlier analysis of knowledge sources. This is not surprising, as KISA analysis reveals the type of knowledge accessed, which although it can be from the same source, may also be of a completely different knowledge type. For example, firms could be accessing IT services via internal sources, from paid consultants and advisers, or from suppliers, and other firms may be accessing R&D KISA also internally, or through customers, suppliers, or universities. Both aspects of knowledge sourcing are important: the what; and the where from.
Returning to the industry based analysis, marketing and promotion, and R&D are the most accessed forms of KISA for both manufacturing, and business and financial services. Surprisingly, particularly in the case of R&D, business services firms show higher levels of use of this resource. This could be explained by the inclusion of market research within the R&D category. IT services KISA are accessed at twice the rate in business services firms (43.0%) than in manufacturing firms (20.5%). These levels of usage were also found in e- commerce KISA.
Manufacturing firms had higher levels of usage in: industry development advice26; training services; recruitment; accreditation; and legal services. These last two types (accreditation and legal services KISA) demonstrated only small differences between the two industry types (less than 4 percentage points). The higher levels of difference displayed between both training, and recruitment services (in excess of 10 percentage-point differences), may be more likely to highlight the current skill shortages experienced by manufacturing firms, and their consequent attempts to address these shortages, rather than any fundamental
26 Much higher levels of industry development advice were evidenced in manufacturing firms (21.6%) compared with business services firms (2.1%). This could possibly suggests that this type of KISA is not as well developed for business and financial sector industries as it is for manufacturing industries.
Chapter 5 – Industrial Innovation in Outer Western Sydney 141 differences in the levels of training provided [MT4]or access to recruitment services across the two industries.
Chart 5.14 KISA usages by type by industry
E-commerce
Legal services (IP, patents etc)
Accreditation
Recruitment
Training services (TAFE etc)
I.T. services
Accounting & Financial services
Research (including market research)
Marketing & Promotion
Business Planning advice
Industry development advice
0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% 70.0%
Manufactruing Business services
Source: Outer Western Sydney Business Innovation Survey, n=119, multiple answers allowed
As was the case for knowledge sources, the analysis of KISA would also benefit from factor analysis, to highlight any latent patterns in the multiple usages of KISA types by firms. This is therefore presented in the next section.
5.3.2 KISA factor analysis
In executing factor analysis on the firms’ responses regarding the high and medium ratings of importance for KISA type usage, similar measures to the previous analysis were undertaken. Principal Components Analysis is again the extraction method used, with Varimax rotation providing the rotational method used on the extracted variables. In order to determine the number of factors to be extracted, both the Kaiser criterion and the scree test were employed. The Kaiser criterion using eigenvalues of 1 or more suggested two factors to be extracted,
Chapter 5 – Industrial Innovation in Outer Western Sydney 142 thereby accounting for 52.1% of the variance. The scree test (Cattell 1966) suggested four factors, which accounted for 68.7% of the variance. The latter was selected as the method to be used, and four factors were extracted (scree test shown in Appendix 8).
The correlation matrix (also shown in Appendix 8) shows multiple correlations in excess of 0.30. The Goodness-of-fit tests - the Kaiser-Meyer-Olkin (KMO) Measure of Sampling Adequacy, and the Bartlett’s Test of Sphericity - were performed. The KMO was 0.846, putting it in the same ‘meritorious’ range of values as the previous PCA (0.80 to 0.89) (Kaiser, Meyer et al), and the Bartlett’s test was also statistically significant. In addition, all variables have a communality of 0.50 or more. Both test results, and the communality matrix, are shown in Appendix 8. Highly correlated factor loadings (0.5 or higher) for the four extracted factors in the rotated component matrix are shown in Table 5.4.
The first component, which accounts for 42.0% of the variance, is highly correlated with marketing and promotion, R&D, business planning advice, accounting and financial services, and IT services. Not surprisingly, this component includes the three KISA that were listed as the highest used KISA across all of the analysis categories (marketing and promotions, R&D, and IT services). This combination of key knowledge intensive activities has led some authors in other research to talk of ‘critical’ or ‘core’ KISA (Aslensen 2004, Albors, Herra et al 2007). Albors, Herras et al in their study of the Spanish ceramic tile industry, identified a core set of KISA including: marketing; R&D; IT; and training activities. Their definition of ‘critical’ KISA, through use of these key activities, was highly correlated with firm innovation activities and competitiveness. As this first component closely matches this definition of ‘critical’ or ‘core’ KISA, it will be referred to as core KISA.
Chapter 5 – Industrial Innovation in Outer Western Sydney 143 Table 5.4 KISA Factor Analysis KISA Components Core KISA Standardisation HR KISA Industry dev’t KISA KISA Industry development advice .896 Business planning advice .661 Marketing & promotion .817 Research & development .673 Accounting & Financial services .610 IT services .598 Training services .858 Recruitment services .629 Accreditation .757 Legal services, IP & patents .787 E-commerce .533
The second component accounts for 10.0% of the variance. The first component, core KISA accounts for the majority of explained variance, therefore the three additional extracted components (each accounting for between 8-10% of variance) would, by reference, be identified with non-core KISA activities. This second component, and the first of the three non-core KISA factors, is associated with new patenting activity through specialist legal services KISA, accreditation, and e-commerce. This factor represents processes of standardisation; IP protection through legal patents; accreditation of new products and services; and new ways of administrating transactions through e-commerce. This factor is referred to as standardisation KISA.
The second non-KISA factor, accounting for 8.8% of the variance, is correlated with the KISA variables of training services and recruitment. As these variables are associated with internal human resources capacity of the firms, this factor is referred to as human resources KISA. The final factor, and the third of the non- core KISA factors, is highly correlated with just one variable - industry development advice. This variable has the highest factor loading of any of the variables in any of the components (0.896), and accounts for 7.9% of the variance. It is therefore referred to as industry development KISA.
Chapter 5 – Industrial Innovation in Outer Western Sydney 144 5.4.3 KISA implications
The analysis of KISA thus far has discussed the usage of types of knowledge by firms. This enables us to further characterise the knowledge flows around the firms in these three regions.
In the analysis of KISA usage on an individual basis, marketing and promotion, R&D, and IT services were the most accessed KISA by firms, both in the various breakdowns of firms, and across the three regions (although in different order and lower levels in Central West Sydney). Marketing and promotion, and R&D were also the most important KISA for both small and large firms, and manufacturing and business service firms.
There were some significant findings at the regional level, with Central West Sydney having lower overall levels of KISA usage, but higher (for the region) usage levels for R&D, IT services, and specialist legal services. These results follow earlier differences between the regions in terms of the type and novelty of innovation, with Central West Sydney having the highest levels of radical innovation, and also the different sourcing of knowledge, where it had higher levels of university and government agency sourcing.
North West Sydney had the highest (overall) levels of R&D, accreditation, and recruitment KISA. The R&D levels were surprising considering the North West’s focus on process innovation. However, the definition of R&D in this survey and other KISA surveys (Aslesen 2004; Martinez-Fernandez 2006; Martinez-Fernandez and Krishna 2006; OECD 2006) includes market research, which could explain why there are high levels of R&D KISA exhibited when the industry is stronger in business and financial services.
South West Sydney had the highest levels of marketing and promotions, accounting and financial services, training services, and e-commerce KISA. The focus on the more process orientated KISA of marketing, accounting, and
Chapter 5 – Industrial Innovation in Outer Western Sydney 145 training fits with South West Sydney’s focus on incremental innovation within its manufacturing industry base.
The factor analysis of KISA shows four components of KISA usage. The first, and largest component, core KISA, supports findings established in other KISA research; that there is a group of ‘core’ or ‘critical’ knowledge intensive service activities, which firms access for innovation. This core KISA includes the primary types of knowledge firms’ access: marketing and promotion services; R&D; IT services; and accounting and financial services. In addition, three other components of KISA were identified. They have been referred to as non-core KISA, and are focused on standardisation KISA, human resources KISA and industry development KISA. These factors, together with the four knowledge sourcing factors identified previously, provide a picture of the sources and types of knowledge firms’ use in their innovation activities. The next section in this analysis of knowledge flows examines the location of the sources of these KISA.
5.4 KISA Location
Having looked in the previous section at the types of KISA, and established components of KISA, particularly ‘core’ KISA, the purpose of this section is to obtain an understanding of the location from where these activities are sourced.
In the survey, firms were asked not only about the types of KISA that firms used, but also to nominate a location for each of the types of KISA used. Chart 5.15 shows the locations of these KISA out of five possible locations: in-house; provided by other firms in the local area (within a radius of 20 kilometres and broadly coinciding with the regional boundaries); provided by firms elsewhere in the Sydney metropolitan area; elsewhere in the state of New South Wales (NSW); elsewhere in Australia; or internationally by firms or organisations overseas. Firms could nominate more than one location for these activities if they sourced KISA from multiple locations. The importance of internal knowledge resources is evidenced by ‘in-house’ being the dominant source of KISA
Chapter 5 – Industrial Innovation in Outer Western Sydney 146 provision in marketing and promotion, R&D, IT services, accounting and financial services, and business planning advice. These are the KISA that are associated with core KISA.
External sources were also highly utilised (25-30% of firms) for business planning advice, and marketing and promotion, and to a lesser extent in IT services, and accounting and financial services. This suggests that whilst internal sources may be dominant, many firms are purchasing knowledge in addition to their internal supplies of KISA. The location of these suppliers was principally at the Sydney metropolitan level or the local level.
Chart 5.15 Overall location of KISA
E-Commerce
Legal services (IP, patents etc)
Accreditation
Recruitment services
Training services
IT services
Accounting & Financial Services
Research and Development
Marketing and promotion
Business planning advice
Industry development Advice
0 5 10 15 20 25 30 35 40 45 50
In-house Local (20km radius) Sydney metro area Elsewhere in NSW Elsewhere in Australia Overseas
Source: Outer Western Sydney Business Innovation Survey n=119, multiple responses allowed
External sources were the dominant providers of KISA in the areas of training services, recruitment services, accreditation, legal services and e-commerce. The factor analysis in the previous section defines these types of KISA as ‘non-core KISA’, and indicates these are used as needed. It therefore makes sense that these activities are largely externally sourced. Training sources and recruitment
Chapter 5 – Industrial Innovation in Outer Western Sydney 147 were primarily sourced locally, signalling the importance of the local labour market for these firms, whilst legal, accreditation, and e-commerce services were largely sourced from the Sydney metropolitan area. However, each KISA category also has sources located nationally and internationally.
R&D has the highest levels of overseas sources, closely followed by marketing and promotion, and industry development advice. These three categories have a variety of external sources from the Sydney metropolitan, national, and international level, as well as being dominated by internal sources, thus demonstrating a high level of knowledge ‘mixing’ from different sources in these KISA.
Understanding how firms manage this process and the selection mechanisms for the mixing and matching of internal and external resources is taken up in further detail in Chapter 7, but Chart 5.15 shows that the KISA activities undertaken by firms have both internal and external components, and that the external components come from a variety of geographies. The Sydney metropolitan and local level both feature prominently, but national and international sources exist in all but one category, that of business planning advice.
The following three Charts (Charts 5.16-5.19) show the location of KISA for each of the three regions under investigation. In Central West Sydney, the main locations for KISA provision are in-house and local, except for services such as accreditation, legal services and e-commerce, where most providers were located in the Sydney metropolitan area, rather than in the local 20km radius area. Central West Sydney also has higher levels of overseas providers for KISA in a wide array of categories. This could be tied to the high level of radical innovation carried out by firms in this region, with internationally located KISA in R&D, and industry development providing access to frontier technology and practice.
Chapter 5 – Industrial Innovation in Outer Western Sydney 148 Chart 5.16 Location of KISA Central West Sydney firms
E-Commerce
Legal services (IP, patents etc)
Accreditation
Recruitment services
Training services
IT services
Accounting & Financial Services
Research and Development
Marketing and promotion
Business planning advice
Industry development Advice
0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00
Overseas Elsewhere in Australia Elsewhere in NSW Sydney metro area Local (20km radius) In-house n=28, multiple responses allowed
There is international sourcing in all but three KISA types (business planning advice, accounting and financial services, and accreditation). The Sydney metropolitan area is the key source for the standardisation KISA (legal services, accreditation, and e-commerce). On the whole, there were low levels of overseas suppliers of KISA, and these were mainly in specialist areas including accreditation, research and product development, and marketing and promotion. The local region is an important location for firms as a site for human resources KISA.
North West Sydney also has high levels of KISA provision in-house, but there are also higher levels of supply from the wider Sydney metropolitan area, especially in the legal services category. For North West Sydney, the two major locations for KISA are internal and Sydney metropolitan area. KISA sourced primarily in-house includes core KISA: R&D; marketing and promotion; accounting and financial services; and IT services.
Chapter 5 – Industrial Innovation in Outer Western Sydney 149 Chart 5.17 Location of KISA, North West Sydney
E-Commerce
Legal services (IP, patents etc)
Accreditation
Recruitment services
Training services
IT services
Accounting & Financial Services
Research and Development
Marketing and promotion
Business planning advice
Industry development Advice
0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00
Overseas Elsewhere in Australia Elsewhere in NSW Sydney metro area Local (20km radius) In-house n=26, multiple responses allowed
External sourcing in the North West is focused on non-core KISA activities, including standardisation KISA of legal services, accreditations, and e-commerce and human resources KISA of training and recruitment services, both sourced primarily externally, and mainly from the SMA level, as was the case in Central West Sydney. National and international sourcing of KISA was mainly based around marketing and promotions, R&D, and accreditation. Two of these KISA are associated with core KISA, but in North West Sydney and across all the other regions, these are the two types of KISA sourced from a variety of locations, suggesting multiple locations of KISA provision, and internal sourcing as well as external sources at the metropolitan, national and international locations. This gives further impetus for the analysis of the mechanisms of ‘mixing and matching’ of knowledge in Chapter 7.
At the local level, the KISA types most accessed are recruitment and training services, and marketing and promotion, and IT services, although in all of these categories local sourcing is less than Sydney metropolitan level sourcing.
Chapter 5 – Industrial Innovation in Outer Western Sydney 150
South West Sydney also shows internal locations for core KISA provision. Although South West Sydney has lower levels of KISA sourcing compared with North West Sydney, KISA is sourced from a larger variety of locations, with international sourcing in all but three KISA types; recruitment services, training services, and accounting and financial services. In all of these three types of KISA, local locations have a high level of importance.
Chart 5.18 Location of KISA, South West Sydney
E-Commerce
Legal services (IP, patents etc)
Accreditation
Recruitment services
Training services
IT services
Accounting & Financial Services
Research and Development
Marketing and promotion
Business planning advice
Industry development Advice
0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00
Overseas Elsewhere in Australia Elsewhere in NSW Sydney metro area Local (20km radius) In-house n=65, multiple responses allowed
The most prevalent KISA locations from the region are again recruitment, and training services (suggesting the importance of the local labour market), and also marketing and promotion, and IT services. Unlike in the case of North West Sydney, the local region is the dominant external location for this KISA (as opposed to SMA in North West Sydney).
Chapter 5 – Industrial Innovation in Outer Western Sydney 151 5.4.1 KISA location implications
This section on KISA locations adds to the analysis on knowledge flows by showing the split between internal and external knowledge sourcing, and the geographical dimensions of this external knowledge sourcing. The results presented for the three regions confirm the importance of internal sources for the knowledge gathering activities of firms. Regional differences exist in the extent of use of external knowledge sources, the type of knowledge that is sourced externally, and the geographical location of the source.
Some of these differences can logically be explained by the firm innovation characteristics of the regions, which were established in the first section of this chapter. For example, Central West Sydney’s focus on radical innovation is coupled with higher knowledge sourcing from universities and government agencies, and a wide variety of external KISA geographical sources, including high levels of international sourcing of KISA and the highest of all three regions usage of legal IP KISA services.
In the case of South West Sydney, the strong manufacturing base is coupled with comparatively higher levels of product and service innovation, and incremental innovation. This is also supported by evidence of a high reliance on supply chain knowledge sources, particularly from suppliers. South West Sydney, despite some variety in external KISA sourcing, has a high reliance on external knowledge sourcing from the local region.
North West Sydney has the highest levels of KISA usage, as evidenced in the multiple levels of KISA locations. Despite this, North West Sydney has a high reliance on the Sydney metropolitan area for external KISA sourcing. This is particularly evident in the external sourcing of core KISA.
Chapter 5 – Industrial Innovation in Outer Western Sydney 152 5.5 Summary
The purpose of this chapter was to examine the firm innovation and knowledge flows of the three regions, and examine the characteristics of these two features as established in the theoretical framework of Chapter 2.
The chapter addresses three specific research questions relating to these two features: 1. How does innovative activity manifest in these regions? 2. What knowledge do firms access and from where? 3. What knowledge is external and what is internal?
Research question one specifically relates to establishing the firm innovation levels and characteristics of the region. Research questions two and three establish the knowledge flows and characteristics that firms access across the three regions.
In addition to the three research questions, the chapter also seeks to test two hypotheses relating to knowledge flows, and the effect that industry and business size have on these flows. Specifically, the hypotheses are: Hypothesis 1 – Firm size affects the mix of knowledge sources that firms’ access. Hypothesis 2 – Industry sector affects the mix of knowledge that firms’ access.
The levels of innovative activity across the three regions were established in Tables 5.1, 5.5, and 5.6. Overall, between 67% and 80% of all firms in each of the regions had innovated in the previous three years. Regional differences existed not only in the levels, but also in the type and novelty of innovations introduced. Central West Sydney had lower overall levels of innovation, but the highest levels of radical innovation, particularly in operational process innovation. North West Sydney on the other hand, had the highest levels across the three regions of firms innovating, especially in process innovations, both operational and organisational, and also the highest levels within the business
Chapter 5 – Industrial Innovation in Outer Western Sydney 153 and finance services sectors. The industry profile in Chapter Four shows high levels of employment in these sectors in North West Sydney.
South West Sydney is positioned overall somewhere in between the two other regions in terms of levels of innovative activity. South West Sydney’s innovation is characterised by higher levels of product and service innovation, especially in the manufacturing industry. The innovation is predominantly of an incremental nature. This is also in keeping with South West Sydney’s industrial profile in Chapter Four, which showed manufacturing as being concentrated in chemical and non-metallic mineral products manufacturing. The focus on product and service incremental innovation is typically of these industries, which have high capital costs in associated production machinery.
Three sections in this chapter address research question two and the associated hypotheses. The characteristics of knowledge flows in the regions are analysed through the study of knowledge sources, KISA usage and the location of KISA. Internal knowledge sources and the supply chain sources are the most important sources of their kind in these two areas across all three regions, and across the categories of industries and firms’ size.
Regional variation exists in the additional external knowledge sourced by firms. Central West Sydney relies upon journals and websites, government agencies, and universities as external knowledge sources, while North West Sydney firms used professional conferences, exhibitions, fairs etc, as well as journals and websites, and other parts of enterprise groups as the key external knowledge sources. The sourcing of other parts of the enterprise group suggests a number of firms are operating in a subsidiary relationship to other firms located elsewhere. South West Sydney displays the highest reliance on their supply chain for external knowledge sourcing. In addition, they also rely upon professional conferences, exhibitions, fairs etc, and consultants and paid advisers for external knowledge provision.
Chapter 5 – Industrial Innovation in Outer Western Sydney 154 This analysis provides an understanding of from where firms obtain their knowledge: internally; the supply chain; and a number of key external sources. The factor analysis of KISA that was performed based on firms’ knowledge sourcing supports this. The analysis identified four factors or strategies of firm knowledge sourcing. The first and largest factor, customer focused sourcing identified customers and competitors as the key sources of external knowledge. This is followed by the second factor, Enterprise focused sourcing. This group sources external knowledge from their enterprise group, suppliers (also shaped by membership of larger enterprise group), consultants, and paid advisers. The final two factors are: public sector sourcing, including universities and government agencies; and commercial information network sourcing, which includes conferences, exhibitions, fairs etc, and journals and websites. These two factors cover the additional external knowledge sourcing that firms access in some combination within each of the regions.
Whereas the analysis of knowledge sources provides an indication of where firms access external knowledge, the analysis of KISA provides information on the types of knowledge that firms are accessing. Previous KISA analysis notes a number of ‘core’ KISA activities, which firms tend to access as a group of services on a regular basis, and which are key in their innovative activities (Albors, Hervas et al. 2007). The core KISA includes marketing and promotion services, R&D, IT services, and financial and accounting services. The first three of these KISA are the most accessed forms of KISA in all the three regions. Regional differences exist in the levels of KISA used and the types of KISA used beyond these top three.
Central West Sydney has lower levels of KISA usage overall. However, it has the highest levels of Legal and IP services KISA, industry development KISA, and IT services KISA. The specialist IP and industry development KISA provides further evidence of the radical nature of innovation activity in this region. North West Sydney firms, in addition to the top three KISA, have higher access levels of accreditation related KISA, recruitment KISA, and business planning advice KISA. South West Sydney, on the other hand, has the highest
Chapter 5 – Industrial Innovation in Outer Western Sydney 155 levels of firms accessing market and promotions KISA, financial and accounting services KISA, training services KISA, and e-commerce KISA. This profile of KISA usage supports earlier results which showed incremental, manufacturing based, product and service innovation as being the primary innovative activity in this region. The use of these types of KISA: marketing and promotions (important in the identifying new product features distinct from earlier versions and competitors); e-commerce (for establishing new ability for market transactions); and training services (overcoming noted skill shortages with training); support this notion.
The factor analysis of KISA usage in firms also identified four factors or strategies. The first and largest is Core KISA. The three others activities, which are by definition non-core activities, are: Standardisation KISA, associated with legal IP KISA and accreditation KISA; HR KISA, including recruitment and training services; and Industry development KISA, associated with the single variable of industry development advice.
These two sections of knowledge sourcing, and KISA usage, provide an understanding of the types of knowledge that firms access, and from where. Within the factor analysis the two primary strategies are: customer focused knowledge sourcing; and core KISA usage. This is consistent across all regions.
The section on KISA locations provides the answer to what knowledge is internally accessed and what is externally accessed. Core KISA are predominantly sourced internally. There is some external sourcing of Core KISA, primarily at the Sydney Metropolitan level, but this is always in combination with internal sourcing.
There are regional variations in the locations of external KISA sourcing. Central West Sydney sourced KISA from a wide variety of sources ranging from local through to international sources. The Sydney Metropolitan Area (SMA) is an important source of non-core Standardisation KISA, whereas the local region is
Chapter 5 – Industrial Innovation in Outer Western Sydney 156 more important for HR KISA. Industry development KISA has the highest levels of international sourcing in the region.
North West Sydney does not have the same variety of KISA locations sourcing as Central West Sydney. The SMA is the primary source of external knowledge sourcing, particularly in the non-core Standardisation and HR KISA.
South West Sydney has lower levels of external KISA sourcing than North West Sydney, but similar variety in external locations as Central West Sydney. The SMA is the most important external location for Standardisation KISA whereas, also in the case of Central West Sydney, the local region is the external source for HR KISA. South West Sydney also has the highest levels of international sourcing in Core KISA R&D, and marketing and promotions services.
Finally, this chapter also explored the effects of industry and firm size on knowledge flows. Industry and firm size are both examined in terms of knowledge sources and KISA usage.
Hypothesis 1 proposed that differences exist in the knowledge firms’ access, according to their firm size. The literature notes that small firms, due to less internal capacity, rely more heavily on external knowledge sources than do larger firms. Analysis shows that small firms have much more extensive knowledge sourcing than larger firms. This knowledge sourcing includes both internal and external sources. The type of knowledge accessed also reveals a much higher level of KISA usage, particularly Core KISA marketing and promotions services and R&D, but also non-core KISA elements of Standardisation KISA and Industry development KISA.
Therefore, on the basis of these results, Hypothesis 1 is accepted. Smaller firms do have more extensive external knowledge accessing than larger firms. This external access however, is not at the expense of internal sourcing. At this point, there is no clear evidence that these external sources are more important than the internal sources. A further examination of firms’ mechanisms for combining
Chapter 5 – Industrial Innovation in Outer Western Sydney 157 internal and external knowledge is needed before conclusions can be drawn on this aspect of the results. This topic is further explored in Chapter 7.
Hypothesis 2 proposed that differences exist in firm knowledge accessing due to the firms’ industry. In the survey, industry is only examined at a broad level, and only in the two sectors that constitute the majority of the survey sample: manufacturing; and business and financial services. The idea behind testing the effect of industry at this broad level is to see if there are any fundamental differences evident in knowledge sourcing or access in these two industries. This testing is also to see whether any policy interventions into encouraging knowledge sourcing, and assisting knowledge access, need to be shaped for specific industry categories.
In both areas where industry was analysed (knowledge sources and KISA usage), the same degree of variance is not displayed as was seen in the comparison of firm size. There are, however, small differences in knowledge sourcing from: clients and customers (higher in manufacturing); competitors (higher in business and financial services); and government agencies (again, higher in business and financial services).
In the analysis of the types of knowledge used, the KISA usages highlight additional distinct differences. Core KISA usage levels are similar in both sectors, except for IT services which have much higher usage levels in the business and financial services sector. E-commerce KISA also has much higher usage in this sector. Manufacturing on the other hand, has much higher accessing of HR KISA, Standardisation KISA (although not to the same extent as HR KISA), and Industry development KISA.
Therefore, on the basis of these results, the hypothesis that industry sector affects the knowledge accessed by firms, must also be accepted, although in analysing the sources of knowledge firms used, there is not the same degree of variance as is seen in the firm size analysis. In terms of the types of knowledge accessed by
Chapter 5 – Industrial Innovation in Outer Western Sydney 158 firms, shown through the KISA analysis, there are differences in knowledge access, particularly in elements of non-core KISA.
In conclusion, this chapter has analysed firm innovation, knowledge sourcing, KISA usage, and KISA location. This analysis has provided information that can be used in the assessment of two further RIS elements. Table 5.5 provides the results of each of the RIS elements examined so far; i) Industrial knowledge base, ii) Business innovation and iii) Knowledge flows.
Table 5.5 RIS analysis elements Characteristics Central West Sydney North West Sydney South West Sydney Industrial • Manufacturing • Business & • Manufacturing Knowledge Base • Engineering Financial Services • Engineering • Business & Information Business innovation • Operational • Process • Product & Service Process innovations (both Innovations Innovation types) • Incremental • Radical innovation • Incremental innovation innovation Knowledge flows • Sources – • Sources – • Sources – customer & public customer, customer and sector enterprise & commercial • KISA – Core KISA commercial networks & Standardisation information • KISA – Core KISA KISA networks & HR KISA • KISA – Core KISA & Standardisation KISA
Already the value of the integrated analysis is apparent in that we can detect relationships between industry knowledge base, innovation output and knowledge flows within and external to the firms. The integrated approach provides us with a multi-dimensional view of the knowledge processes operating. So that within the one region, it is not only the industrial knowledge base that dominates, but how this manifests in innovation output and the knowledge sources and networks that firms use. We can see, for example, Central West Sydney has strength in radical innovation, this also co-exists with very specific KISA usages around IP protection, patents and standards and the networks used to source information are commercially orientated. This is in comparison to South West Sydney, which also has a manufacturing knowledge base, but follows a more conventional and expected path, with a focus on incremental
Chapter 5 – Industrial Innovation in Outer Western Sydney 159 product innovation, co-existing with a focus on HR specific KISA and customer driven networks. Again, no causality can be assumed in these relationships, but that there are relationships, and that these are viable at the regional level, is an important finding.
The following two chapters explore this further, with the next chapter looking into the orientation of the knowledge flows and Chapter Seven, the co-ordination of knowledge usages within firms.
Chapter 5 – Industrial Innovation in Outer Western Sydney 160
Chapter 6 – Regional orientation of firms
At the conclusion of the previous chapter, summary findings on the first three elements of the integrated framework were presented. The third element; Regional knowledge flows is explored in more detail in this chapter. A fourth element; Regional orientation of knowledge sourcing is also analysed. This chapter specifically focuses on external knowledge activities of the firms, and the geographical distribution of these external activities. Again, geography is referred to in a socio-political sense. To achieve this, the fourth and fifth research questions are investigated.
4. What are the key organisations for external knowledge sourcing in firms? 5. What is the regional component of these knowledge flows?
Hypothesis 3 – The dimensions of the regional supply chain are reflected in the location of key knowledge organisations for firms in the regions.
The results so far have highlighted a multi-dimensional picture of regional knowledge activities. The confirmed diversity is expected to continue, yet reflect the emerging individual pictures of activity in each of the regions.
Chapter 6 – Regional orientation of firms 161
North West Sydney – knowledge intensive and service orientated Central West Sydney – radical product innovation and established manufacturing knowledge base. South West Sydney – incremental process innovation within more traditional manufacturing knowledge base.
The chapter is structured into three sections. The first section examines the key knowledge organisations accessed by firms, and the importance that firms attribute to them. From this, the regional component will be identified. The second section builds further on this analysis, using Principal Components Analysis to explore relationships between a range of networking activities undertaken by firms. The third section will look at the dimensions of the supply chain, and establish conclusions comparing supply chain activities with the location of key knowledge organisations.
6.1. Locations of key knowledge organisations
This section explores the networking activities of firms. Within the literature of regional innovation systems, and even more broadly within innovation analysis, firms are surrounded by their knowledge network. This network includes firms within their supply chain (the subject of later examination), and other external sources of knowledge such as industry associations and collaboration partners, other firm sources of KISA, and public institutions, universities and government agencies. In this section, these organisations are divided into four categories: industrial; educational; government institutions and agencies; and regional organisations. The breakdown of the organisations into these four categories provides further indication of the types of knowledge provided within the networks.
As part of the background research, firms were asked about their relationships (if any) with a range of institutions and organisations, which were divided into the four categories outlined above. In the survey instrument, a number of prominent
Chapter 6 – Regional orientation of firms 162
and known institutions in each of these categories were listed (in the manner of a roster call method (Boschma and Ter Wal 2007)). In addition, there was also space for firms to add other organisations that were not listed. The results of each of the four categories are presented in Charts 6.1-6.4.
Chart 6.1 Government organisations included in firm knowledge networks % of firms who involve these organisations in their knowledge gathering
60.0%
50.0%
40.0%
30.0%
20.0%
10.0%
0.0% Local Council Area commissions Austrade Ausindustry Invest Aust DSRD OMWS
Central West Sydney North West Sydney South West Sydney
Source: Outer Western Sydney Innovation Survey. n=119
Examining the first of these charts, which looks at network links with government organisations (Chart 6.1), it is immediately apparent that the listed government organisations are much more a part of the North West Sydney firms’ knowledge networks than they are for the other two regions. An explanation of all acronyms used is available in the glossary at the beginning of this thesis. The three levels of government in existence in Australia’s federalist system are represented here. This begins with the local government level, represented by the Local Councils, which is followed by the state government with the Department of State and Regional Development (DSRD) and the Office of the Minister of Western Sydney (OMWS), and finally the Federal or Commonwealth Government through Austrade (export development agency), Ausindutry
Chapter 6 – Regional orientation of firms 163
(Industry Development agency, closely linked with Austrade), and the Area Consultative committees (GROW - these are regionally based committees that make recommendations on the direction of grant expenditure from the Federal Government’s Department of Transport and Regional Services (DOTARS)).
At the local government level, there is varied use of local government resources in the knowledge networks of the firms in the three regions, with nearly 40% of firms in Central West Sydney noting local councils were in their knowledge networks, decreasing to just above 30% in North West Sydney and a low of 24% in South West Sydney. However, in both Central and South West Sydney, local councils were the most cited government organisation used in their knowledge networks. In the case of North West Sydney, they were only fifth, following both the state (with DSRD the highest) and federal government organisations and agencies. Central West Sydney had higher levels of usage of these state and federal agencies in their knowledge networks than did South West Sydney. These results show that government agencies at the state and federal level are key organisations in North West Sydney firms’ knowledge networks. In both Central and South West Sydney, although firms have some contact with other levels of government, local government appears to be the key government organisation utilised in their knowledge networks.
The higher levels of networking recorded with AusTrade and AusIndustry, both of which are associated with export facilitation and industry development, show that either the firms in North West Sydney (and to a lesser extent Central West Sydney) are highly focused on exporting and are accessing the relevant grants associated with these agencies to support their export development, or that they are inexperienced in exporting and are seeking further assistance from these government agencies, or a combination of both. Either way, it shows they are using government partners in supporting their exporting activity.
Chapter 6 – Regional orientation of firms 164
The second chart examining institutional and organisational networking (Chart 6.2) does so from an industrial viewpoint, including industry and business associations (the major Australian business associations were individually listed in the network roster in the survey), industry specific associations, and local Chambers of Commerce. Comparing the three regions, again North West Sydney appears to have a richer array of network contacts. Industry specific associations (particularly in North and South West Sydney), and the two major business associations (Australian Business Limited and Australian Industry Group) are noted as important network contacts. These organisations provide knowledge that is both industry specific advice from specialist associations, and more generalist ‘business management’ advice from the major industry associations. However, the role of the major business associations should be tempered by the fact that membership lists of these organisations were used to develop the survey sample, therefore it is expected that a higher degree of firms would nominate these industry contacts.
Chart 6.2 Industrial based organisations included in firm knowledge networks % of firms who involve these organisations in their knowledge gathering
60.0%
50.0%
40.0%
30.0%
20.0%
10.0%
0.0% Chambers of Commerce ABL AIG Other industry assoc
Central West Sydney North West Sydney South West Sydney
Source: Outer Western Sydney Innovation Survey. n=119
Chapter 6 – Regional orientation of firms 165
The local Chambers of Commerce has a varying role across the three regions, in much the same way as the local council’s roles were varied; although this time the lower levels were recorded in Central West Sydney. The policy environment in the local government areas that make up these regions need to be considered in light of these results. North West Sydney has a strong Chamber of Commerce; and Central West Sydney has a strong local government focus on local/ regional economic development programs. South West Sydney, at the various local government levels, has not had such a determined focus, and the local Chambers of Commerce have not provided the same level of strategic business support, particularly in Liverpool, which has been weakened in recent years by financial difficulties despite having a strong membership base.
Chart 6.3 Educational organisations included in firm knowledge networks % of firms who involve these organisations in their knowledge gathering
60.0%
50.0%
40.0%
30.0%
20.0%
10.0%
0.0% UWS USYD UNSW UTS UOW Other universities TAFE
Central West Sydney North West Sydney South West Sydney
Source: Outer Western Sydney Innovation Survey. n=119
The third chart looks at the educational institutions within the knowledge networks of the firms (Chart 6.3). These institutions are primarily either universities, or the Technical and Further Education (TAFE) colleges. Percentage counts for individual universities located within the Sydney metropolitan area (SMA) are listed individually while universities outside of the SMA are collated
Chapter 6 – Regional orientation of firms 166
into the one group: external universities. The local university is the University of Western Sydney (UWS) and it has campuses in each of the three regions.
North West Sydney has higher levels of networking activity, although this is not as pronounced as in the previous two charts. North West Sydney firms clearly have more interactions with universities, as is evidenced by the high percentage (in excess of 50%) of firms surveyed that have universities within their knowledge network, with the University of Western Sydney being a key knowledge organisation for the region. In fact, the University of Western Sydney is the most frequent university contact for the firms in each of the regions, although higher in North West and Central West Sydney than in South West Sydney. Three other prominent universities in Sydney: the University of Sydney; the University of Technology Sydney; and the University of New South Wales, also feature in firms’ knowledge networks. In particular, the University of Sydney and University of Technology Sydney are important contacts for Central West Sydney, with the University of NSW playing the same role for North West Sydney. For South West Sydney firms, the TAFE College is also a key knowledge organisation. TAFE collages are responsible for vocational education. A quarter of firms in South West Sydney noted the importance of the TAFE College to their innovative activities. This reliance on vocational expertise is related to the regional high usage of HR KISA (which included training services).
As noted earlier, the category of ‘other universities’ refers to any university not listed by name in the survey. All universities in the Sydney Metropolitan area were individually listed on the network roster in the survey, therefore any contacts with ‘other universities’ would be located outside the Sydney metropolitan area. This group of contacts is the highest in North West Sydney, with 24% of firms in that region surveyed including them as key knowledge organisations. This level is much higher than the 5% and less in Central and South West Sydney. There is growing evidence in the RIS literature that access to external sources of knowledge from beyond the region is essential to the
Chapter 6 – Regional orientation of firms 167
vitality of knowledge in the firm and to avoid knowledge lock-in and stagnation (Sternberg 2000; Moulaert and Sekia 2003; Boschma 2005; Dahlander and Wallin 2006). Links with universities outside of the Sydney metropolitan area would be a key source of external regional knowledge. This further highlights the overall organisational and institutional networking capability of North West Sydney firms. The charts presented in this section show a much richer array of contacts and potential sources of knowledge within the networks of these North West Sydney firms. The question that follows then is: why is this so and then, particularly in relation to regionally external university contacts, how are these contacts being used, and in what way do these firms manage these relationships and networks that is so different to the other two regions?
The final set of organisations and institutions that form this part of the network analysis are organisations that are defined as ‘region-based’ organisations. Although there have been organisations within the other categories that could also be considered region-based organisations, such as local councils, Chambers of commerce, and the local university, the organisations presented in Chart 6.4 are considered different because they are a product of the region, whereas local councils are a product of local government, Chambers of commerce are part of the wider array of industry associations, and the local university is part of the higher education sector. These organisations include the two peak regional ‘lobby’ groups, WSROC (Western Sydney Regional Area of Councils) and its sister (although much smaller) organisation MACROC (Macarthur Regional Area of Councils), as well as the Greater Western Sydney Economic Development Board (GWSEDB).
Again, the prevalence of these organisations in the knowledge networks of North West Sydney firms is shown in this Chart. Despite the importance of external regional knowledge sources, as discussed in the previous section, clearly in North West Sydney’s case this has not been at the expense of regional-base network contacts, with the two peak regional organisations being featured, although with much smaller percentage levels than the networking levels noted
Chapter 6 – Regional orientation of firms 168
in the previous three charts. In both Central West Sydney and South West Sydney, the top regional contacts have been with a mix of other organisations, although only a small percentage of firms in each case are involved, in most cases being less than 5%. Overall, of the four sets of knowledge network institutions and organisations studied, the region-based ones have been the least prevalent.
Chart 6.4 Regional organisations included in firm knowledge networks % of firms who involve these organisations in their knowledge gathering
45.0%
40.0%
35.0%
30.0%
25.0%
20.0%
15.0%
10.0%
5.0%
0.0% WSROC MACROC GWSEDB Business Connection BEC Business Capability Network
Central West Sydney North West Sydney South West Sydney Source: Outer Western Sydney Innovation Survey. n=119
6.1.1 Knowledge organisation implications
From this analysis, a number of key conclusions for each region can be drawn. Firstly, North West Sydney has deep network relationships across all four categories of non-firm organisations. The most prominent relationships were with industrial associations and two key government agencies, one at the federal level, AusIndustry, and one at the state level, Department of State and Regional
Chapter 6 – Regional orientation of firms 169
Development (DSRD). North West Sydney has a variety of higher education links, including the highest levels of interaction with external (to the Sydney metropolitan area) universities.
The ‘region based’ organisations have the lowest levels of inclusion in firms’ knowledge networks across all regions. However, they are represented in the highest percentages in North West Sydney firms. The two key organisations for North West Sydney firms are: WSROC (Western Sydney regional Councils lobby group); and GWSEDB (the main economic development agency of the Western Sydney region). The low levels of MACROC involvement are not surprising given that this organisation is focused on South West Sydney. It is surprising, however, when compared to levels in South West Sydney, which are much lower still.
The key knowledge organisations for Central West Sydney firms are the local councils, and the local university – the University of Western Sydney. However, Central West Sydney also shows strong relationships with two Sydney-based universities, but minimal contact with higher educational sources external to Sydney. The main industrial organisations that feature in networking activites for these firms are Australian Business Limited, one of the two peak Australian business associations27, and, to a lesser extent, industry specific associations. Involvement with the local Chamber of Commerce in Central West Sydney is at a lower level than that shown in the other two regions. In terms of ‘region based’ organisation networking, this is minimal for Central West Sydney. The most prominent organisation in this category is WSROC, and this was listed in only 7.5% of surveyed firms’ knowledge networks.
The key knowledge organisations for South West Sydney firms are both industrial based: industry specific associations, and local Chambers of Commerce. Overall, South West Sydney has the lowest levels of firm networking in the areas of both individual organisations, and the depth of networking with a
27 ABL memberships were part of the survey sample; therefore their prominence in some firm networking may be overemphasised because of this.
Chapter 6 – Regional orientation of firms 170
range of organisations, except for the industrial category, where it is second to North West Sydney. The most cited government organisation is the local council, followed by the two federal organisations, although at lower levels comparatively to the other two regions. More than 30% of firms included the local university as a key organisation, and there were some firms who had contact with other Sydney based universities but, as with Central West Sydney, had only minimal contacts external to the Sydney metropolitan area. Vocational education, through local Technical and Further Education Colleges (TAFE), is also included in the knowledge networks of just under 30% of firms. In terms of ‘region based’ organisations, the most cited, perhaps unsurprisingly, is MACROC, the local organisation of South West Sydney Councils. Although, as earlier mentioned, it is surprising that North West Sydney has a higher percentage of firms with MACROC contacts. It is possible that the joint programs run by MACROC and WSROC, would account for this apparent discrepancy, when one considers the higher number of North West Sydney firms with strong connections to WSROC.
The diagram below summarises the locations of firm knowledge organisations across the levels of geography introduced in the previous chapter on KISA supply location, and based on an assessment of the previous charts. It captures the range and depth of network partners listed by North West Sydney firms across the four categories, and the range of networks in the other two regions, which however are primarily limited to the local/ Sydney metropolitan area level, with the exception of industrial networking at the national level through the peak business group in Central West Sydney.
Chapter 6 – Regional orientation of firms 171
Diagram 6.1 Extent of regional knowledge networks
This section has looked at the knowledge organisations that firms in the regions regarded as being significant on an individual organisational level, and across four broad categories. It is clear that firms have a number of connections with different organisations, and although there may be some consistency across the four categories and geographies discussed in this section, it is not enough from which to draw conclusions. It is therefore the subject of the following section to draw out these patterns of networking activity using factor analysis.
6.2 Factor analysis of knowledge organisation networks
Firms have multiple contacts within their knowledge networks, and whilst the analysis of the percentage of firms accessing individual organisations provides an indication of the importance of that organisation within the regional knowledge network, the combination of relationships between the multiple sources [MT5]is also important, as it gives an indication of how firms use their complete network for knowledge gathering.
Chapter 6 – Regional orientation of firms 172
As in the previous two chapters, factor analysis offers the opportunity to explore the multiple responses of firms to the numerous organisations that form their networks. Specifically through principal components analysis, and using a Varimax rotation method, six networking components were identified. The relevant tests, which have been discussed previously, were performed (shown in Appendix 9). A number of variables were excluded because the communality values were not above 0.5028. Five components were extracted. The components and the associated highly corresponding factor loadings are shown in Table 6.1.
The factor analysis accounts for 64.8% of the variance among these variables. The first factor accounts for 30.4% of the variance and is the strongest extracted factor. It is highly correlated with the two nationally based government agencies that support industry development (AusIndustry), and export promotion (AusTrade). The component is also associated with the two nationally based peak business associations, Australian Industry Group (AIG), and Australian Business Limited (ABL). Subsequent data collected in the interviews suggests that this factor reflects the primary networks of most innovative firms. This factor is referred to as nationally based networks.
The following four extracted factors each account for between 7-10% of the variance. Component two is correlated with a major metropolitan university and the two peak regional organisations, the Greater Western Sydney Economic Development Board (GWSEDB), and the Western Sydney Regional Organisation of Councils (WSROC). It is a mix of both metropolitan and regional networks. This component accounts for 9.6% of the cumulative variance and is referred to as metro/ region mix networks 1.
28 The excluded variables include (i) Greater Western Sydney Business Connection, (ii) University of NSW, (iii) University of Technology, (iv) University of Wollongong, (v) Area committees, and (vi) Invest Australia.
Chapter 6 – Regional orientation of firms 173
Table 6.1 Knowledge networks: Factor analysis (Rotated matrix) KISA Components Metro/ Nationally Metro/ region State/ region Region Mix based mix of mix of of Educational organisations organisations organisations organisations organisations Councils .696 AusIndustry .798 AusTrade .744 Dept State & Reg .732 Development
Office Minister Western .753 Sydney Chamber of Commerce .722 ABL .570 AIG .539 Other industry associations .652 Uni Western Sydney .565 Uni of Sydney .773 Other universities .547 TAFE, Vocational education .762 GWS Econ Dev’t Board .772 WSROC .692 Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization.
The third component accounts for 9.1% of the variance, and is correlated with networking through local Councils and state government organisations, the Department of State and Regional Development (DSRD), and the Office of the Minister of Western Sydney (OMWS). OMWS is a state based agency but, as its name suggests, focuses on the Western Sydney region. This component, like the previous one, shows networking over two levels of government, state and local. It is therefore referred to as state/ region mix networks.
The fourth component also shows a mix of core networking contacts at both the metropolitan and local level. The component accounts for 8.9% of the variance and is correlated with the local university, the University of Western Sydney, the local Chambers of Commerce and, at the metropolitan level, industry specific associations29. This component is referred to as metro/region mix networks 2.
29 Many of the listed industry associations are actually national bodies or state chapters of national organisations but have their offices and run the majority of their programs in the Sydney metropolitan area.
Chapter 6 – Regional orientation of firms 174
The final component accounts for 6.8% of the variance and is the smallest of the extracted variables. This component is correlated with two educational variables, Other Universities (outside the metropolitan area) and vocational educational provider TAFE (Technical and Further Education Colleges). This factor also straddles two (at least) levels of geography: the local and national (or international); the local refers to the TAFE colleges, and the national and international to the external (to the metropolitan area) universities. However, as the focus of both correlated variables is education, the factor is referred to as educational networks.
The latter four factors in combination account for approximately 35% of the variance and indicate, not so much through the individual factor correlations, but jointly overall, that knowledge sourcing for firms cuts across a number of geographic levels - local, metropolitan, national, and even international. The first and largest component identifies a number of key major organisations in firms’ knowledge networks. The identification of the two peak business associations is not surprising, as memberships of these organisations were one of the sample selection mechanisms. The identification of the two federal government agencies of AusIndustry and AusTrade is more important, as it shows the strong position that nationally based government agencies have in these regions’ RIS. Two other factors need to be considered in the context of firms’ knowledge networks; the exporting activity of the firm and whether any branch or head office relationships exists. Exporting activity along with the geographical dimensions of supply chain activity is examined in the next section. Branch and head office were noted as knowledge sources for a number of firms in the previous chapter and were associated with the Enterprise focused component also discussed in Chapter Five. Discussion will again return to these components in the cluster analysis presented in Chapter Seven.
Chapter 6 – Regional orientation of firms 175
6.3 Supply chain geography
Throughout the research up to this point, the supply chain of firms - suppliers, customers and clients and competitors - has been highlighted as an important source of knowledge activities for firms. This next section analyses the geography of firms’ supply chains, to better understand the effect of location of suppliers and customers on the knowledge gathering activities of the firms.
In the survey, firms were asked to nominate the importance of five different levels of geography (local, metropolitan, state, national, and international) for the location of their most important customers and suppliers (on a ranking system from no customers or suppliers, not relevant for customers and suppliers, low, medium and high importance). Charts for each of the three regions, and for customers and suppliers, are presented in the following pages (Charts 6.5-6.10), beginning with Central West Sydney.
In terms of customers, those of highest importance to Central West Sydney firms were located in metropolitan Sydney, followed by the rest of the state of NSW, and then local firms. Over 30% of firms note that the international market was not relevant to their business for customers and 10% of firms also said national and local locations were not a relevant customer base. This is surprising given the high levels of networking with export focused government organisations as discussed in the previous section. It may be a situation of thinking and organising for exporting rather than actually undertaking the activity.
Suppliers were a different story, with the highest levels of important sources for suppliers located in the local area, and the Sydney metropolitan area. Supplier locations that are considered ‘not relevant’ or ‘no suppliers’ for Central West Sydney firms are spread across all the categories of location, with the highest at the international level. However, at the local level more than 10% of firms noted that the international level was not a relevant area, or they had no suppliers at this area. The large number of responses indicating ‘low importance’ at the state,
Chapter 6 – Regional orientation of firms 176
national, and international level, does not necessarily signify that firms disregard these areas as sources of suppliers, just that the most important ones are located at the local and metropolitan level.
Chart 6.5 Central West Sydney – Customers by geography (level of importance)
International
Australian
High importance Medium importance State Low importance None/ Not relevant
Metro Syd
Local
0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0% 35.0% 40.0% 45.0%
Source: Outer Western Sydney Business Innovation Survey
Chart 6.6 Central West Sydney – Suppliers by geography (level of importance)
International
Australian
High importance Medium importance State Low importance None/ Not relevant
Metro Syd
Local
0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0%
Source: Outer Western Sydney Business Innovation Survey
Chapter 6 – Regional orientation of firms 177
Overall, Central West Sydney’s customer base is focused from the Sydney metropolitan level through to the national level. The supplier base is more focused at both the Sydney metropolitan and local levels.
North West Sydney presents a similar picture to Central West Sydney in terms of important customers being located in the Sydney metropolitan area, although this is not as pronounced. The local and state level is considered of medium importance for the location of important customers in this region. Out of the three regions, North West Sydney also has the highest amount of firms regarding the international level as an important source of primary customers.
Chart 6.7 North West Sydney – Customers by geography (level of importance)
International
Australian
High importance Medium importance State Low importance None/ Not relevant
Metro Syd
Local
0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0%
Source: Outer Western Sydney Business Innovation Survey
Supplier locations also broadly mirror those of Central West Sydney, with high levels (in excess of 50% of firms) noting that the state and international levels held no important suppliers for them. However, the highly important locations are spread through the five levels, with equal percentages of firms regarding the local, national and international levels of high importance for primary suppliers.
Chapter 6 – Regional orientation of firms 178
The local and metropolitan level is considered of medium importance for the location of primary suppliers for firms in the North West Region.
On the whole, these two charts (6.7-6.8) show that North West Sydney firms have a more extensively located supply chain, with high percentages of firms reporting international and national customers and suppliers. When combined with the extensive knowledge networks of the North West Sydney firms, it is seen that this region has knowledge gathering activities taking place on a broader scale than the other two regions.
Chart 6.8 North West Sydney – Suppliers by geography (level of importance)
International
Australian
High importance Medium importance State Low importance None/ Not relevant
Metro Syd
Local
0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0%
Source: Outer Western Sydney Business Innovation Survey
Chapter 6 – Regional orientation of firms 179
Chart 6.9 South West Sydney – Customers by geography (level of importance)
International
Australian
High importance Medium importance State Low importance None/ Not relevant
Metro Syd
Local
0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0% 35.0% 40.0% 45.0% 50.0%
Source: Outer Western Sydney Business Innovation Survey
South West Sydney firms have most of their customers at the metropolitan and local level, and in fact South West Sydney has the highest percentage of firms (40%) who identify the local area of high importance for major customers. This level is much higher than both the two other regions; in Central West Sydney it is 7%, and in North West Sydney it is 8%. At the international, national, and state levels there are high percentages of firms (in excess of 30% of firms surveyed) that regard these locations as of low importance. In the same locations (international, national, and state), 35-40% of firms regarded these locations as of low importance for customers.
These high levels were even more pronounced for suppliers in these same locations (international, national and state). The local and metropolitan levels were again considered the most important source of suppliers. There were a number of firms who did see the international, national and state level as of high importance for the location of main suppliers, and these were at similar levels to North West Sydney.
Chapter 6 – Regional orientation of firms 180
Chart 6.10 South West Sydney – Suppliers by geography (level of importance)
International
Australian
High importance Medium importance State Low importance None/ Not relevant
Metro Syd
Local
0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0% 35.0% 40.0% 45.0%
Source: Outer Western Sydney Business Innovation Survey
Therefore, South West Sydney’s customer base is very much focused at the local and metropolitan Sydney level, even though there are some international and national customers. The focus on the local customer base is the most significant difference between South West Sydney and the other two regions, and given the importance of the supply chain for the knowledge gathering activities of firms, this concentration at the local level would have an effect on the knowledge gathering activities of firms in these regions.
Diagram 6.2 summarises these results and highlights the key differences in the spread of each regions’ supply chain, based on the locations identified as being of high importance for firms’ main customers and suppliers. The diagram clearly shows the differences in the locations of key customers and supplier relationships between the three regions. These results, together with the earlier established importance of the supply chain as a primary source of knowledge, demonstrate the differences in the orientation of the three regions.
Chapter 6 – Regional orientation of firms 181
Diagram 6.2 Regional supply chain locations
6.5 Summary
The focus of this chapter has been on establishing the orientation and location of key knowledge organisations for firms in the three regions. The focus on orientation and key knowledge organisations builds from the final two features of RIS established in Chapter 2.
The research questions guiding this chapter are: 4. What are the key organisations for external knowledge sourcing? 5. What is the regional component of these knowledge flows?
The following hypothesis is also addressed:
Hypothesis 3 – The dimensions of the regional supply chain are reflected in the location of key knowledge organisations for firms in the regions.
The first section of this chapter discussed the key knowledge organisations for firms by region. These results found that in comparison to the other two regions, North West Sydney firms have a broader range of key organisations they rely
Chapter 6 – Regional orientation of firms 182
upon for knowledge. North West Sydney has the highest levels of external university contacts, and the highest levels of interactivity with ‘region based’ organisations, although at lower overall levels. This shows that external (to Sydney) knowledge contacts were not at the expense of more localised contacts with regionally based organisations.
In Central West Sydney, contact with the local councils and the local university feature the most prominently. Central West Sydney also has prominent relationships with two Sydney based universities: the University of Technology, Sydney; and the University of Sydney. The fact that three of the four key knowledge contacts for firms in this region were universities further reinforces earlier results regarding the close relationships between the type of innovative activity undertaken by firms in Central West Sydney (radical innovation), and the types of knowledge firms seek, and where they seek it from.
In South West Sydney, key knowledge organisations were industry specific associations: the local Chambers of Commerce; TAFE College; and the University of Western Sydney. The location of these organisations is at the local and Sydney metropolitan levels. The inclusion of the TAFE College as a key knowledge organisation highlights the importance of vocational education for the firms in this region, and the importance of being involved with the vocational education provider to shape the training that is provided.
All of these results show that there are quite different key knowledge organisations that are important for external knowledge sourcing in each of the three regions. These organisations range in locations from local to national in both North West Sydney and Central West Sydney, and local to Sydney metropolitan level in South West Sydney. Therefore, when answering the research question, what are the key organisations for external knowledge sourcing, the answers are going to be different for each region.
Factor analysis is again used to detect patterns in associations of these organisations, or in other words, identifying whether firms who regarded one
Chapter 6 – Regional orientation of firms 183
organisation as important necessarily systematically regarded another organisation as important. The factor analysis extracted five factors, accounting for 64.8% of the variance among the organisational variables. The five groups represent strategies used by firms to network with key knowledge organisations. The first group, and accounting for the largest share of the variance (30%), is strongly related to national level organisations, including the two national government agencies supporting industry and export development (AusTrade and AusIndustry); and the two peak national business associations - Australian Business Limited (ABL) and Australian Industry Group (AIG)30. This illustrates the importance of national organisations in the RIS, and why an integrated rather than an exclusive approach to innovation systems analysis and encouragement is essential. The following four components were a mix of local, metropolitan, state, national, and international networking, suggesting that firms use a mix of contacts from various geographic levels.
The regional component of these key knowledge organisations comprises: the local Council; local Chambers of Commerce; the local university; the University of Western Sydney (each region has a campus of the university located within its boundaries); local TAFE Colleges; and a range of regionally based ‘lobby’ organisations. The final group of organisations, the regionally based lobby groups, were not considered as key organisations by very many firms (5% on average). The levels are highest in North West Sydney, which is consistent with the overall trend displayed by the North West region towards more diversity in acknowledged key knowledge organisations. Of the other organisations, the local Council was especially important in Central West Sydney, whereas the local Chamber of Commerce and TAFE College were central resources in South West Sydney. All of the regions regarded the local university as a key organisation. However, if we recall the sources of knowledge that firms noted as important in the last chapter, universities were not key sources. Therefore, this result for the University of Western Sydney may represent a desire on the part of the firms to
30 The two peak national industry associations formed part of the sample so care should be taken in emphasising their role.
Chapter 6 – Regional orientation of firms 184
see the role of the local university as a key knowledge organisation for their region, rather than a current role it is already fulfilling.
To further examine external linkages, the third section analysed supply chain geography of the firms by region. The most significant result from this analysis is the diversity between the three regions in primary customer base. North West Sydney’s customer base extends from the Sydney metropolitan area through to the international level; Central West Sydney’s focus is on the Sydney metropolitan level through to the national level; however, South West Sydney’s customer base is located in the local and Sydney metropolitan level. The supplier locations for all three regions are concentrated at the local and Sydney metropolitan level.
Looking jointly at these two aspects of analysis - the location of key knowledge organisations, and the location of major customers and suppliers - a picture of the external knowledge dimensions of the regions can be built, as shown in diagram 6.3. The diagram also shows the extent of the regional component of external knowledge gathering. Regional sources are key in all of the regions, whether government, industrial, or educational organisations. For Central West and North West Sydney however, these sources are also complemented by other sources external to the region, which come from a variety of locations; from metropolitan Sydney through to the international level. In the case of South West Sydney, the regional component of external knowledge makes up the majority of knowledge flows.
On this basis, hypothesis 3, which posits that the orientation of the regional supply chain is reflected in the location of key knowledge organisations, is accepted. The illustration of South West Sydney in Diagram 6.3 provides clear evidence of this hypothesis. Central and North West Sydney also show that supply chain relationships, particularly with customers, mirror other avenues of external knowledge sourcing.
Chapter 6 – Regional orientation of firms 185
Diagram 6.3 Orientation and location of key knowledge organisations in the regions
This diagram shows that North West Sydney firms operate in a wide sphere of innovation, connected to regional, city, national, and international systems of innovation through the breadth of their interactivity. The regional level is an important source of knowledge for firms, but is by no means the only source. In Central West Sydney, the firms operate in regional and city innovation systems, extending to the national in some areas. External knowledge is available through customers and industry contacts. In South West Sydney, firms are much more based in the regional and city system. There is a high reliance on local knowledge through the supply chain, but also through the local government agency and local university. The primary source of external knowledge is through industry specific associations.
This diagram adds further detail to the relationship between the firms’ supply chain, and organisational networking contacts. South West Sydney is locally focused through its high reliance on local customers and suppliers for the firms based there. The networking activities of South West Sydney, whilst not just restricted to the region, really only extend to the metropolitan area, and are not as extensive as the other comparative regions. This analysis also allows the completion of the RIS analysis across the five elements, as shown in Table 6.2.
Chapter 6 – Regional orientation of firms 186
Table 6.2 RIS analysis elements Characteristics Central West Sydney North West Sydney South West Sydney Industrial • Manufacturing • Business & • Manufacturing Knowledge Base • Engineering Financial Services • Engineering • Business & Information Business innovation • Operational • Process • Product & Service Process innovations (both Innovations Innovation types) • Incremental • Radical innovation • Incremental innovation innovation Knowledge flows • Sources – • Sources – • Sources – customer & public customer, customer and sector enterprise & commercial • KISA – Core KISA commercial networks & Standardisation information • KISA – Core KISA KISA networks & HR KISA • KISA – Core KISA & Standardisation KISA Type and Location • Local Councils • State and federal • Local Chamber of of key knowledge and Sydney government Commerce, local resources based universities agencies, University and • Regional to Universities and local TAFE national level industry college associations • Regional to • Regional to Sydney national level metropolitan level Supply chain • Regional to • Regional to • Regional to orientation national international level Sydney metropolitan level
Both Diagram 6.3 and Table 6.2 once again show the multi-dimensionality of knowledge activities in the three regions. Clear pictures of each of the regions innovation systems are emerging, and following from the second RIS iteration (that RIS are present in all regions, and are to be assessed on a scale from weak to strong) a picture of each region’s strengths is also emerging. The value of the integrated conceptual framework is evident in that if policy recommendations were drawn from either one or two of the RIS elements (as has been the case previously within the literature) they may have had at best neutral and at worse negative impacts, because they would be missing an understanding of other elements of the system.
The implications of the RIS analysis are discussed at length in Chapter 8. However, prior to this, the data gathered and presented to date is not able to provide any direction or causality to this situation. The significance of this RIS analysis for South West Sydney is the recognition that firms in that region are operating in a narrower flow of knowledge than the other two regions. How this
Chapter 6 – Regional orientation of firms 187
affects these firms, and why this picture is so different across the three regions, is the focus of the next chapter, which examines how firms arrive at their particular blend of knowledge gathering activities, internal and external, local and further afield, and why and who drives this mixing and matching.
Chapter 6 – Regional orientation of firms 188
Chapter 7 – Firm strategies for innovation: the process of mixing and matching
This chapter is the final chapter in the discussion of research results, and it brings together the analysis presented in the previous two chapters in order to identify firms’ strategies for innovation, and how and why these are affected by the firms’ regional experience. In the previous two chapters the four elements form the integrated RIS developed in Chapter Two have been investigated through the research methodology. The findings have provided a multi-dimensional view of firm innovation and knowledge activities with the three regions. These findings by necessity have shown the dominant position of the majority of firms in each region in regards to the four elements.
Whilst this has value in characterising the strengths of the RIS in each of the three regions, it does not provide an exacting view of the weaknesses of each of the systems. As policy outcomes and encouraging more certain activities associated with highly innovative and therefore, high growth firms is the aim, we need to specifically identify these characteristics within these highly innovative firms across all the three regions. The literature on RIS provides us with many examples of what these firms look like – small, new, knowledge intensive, open to flows of knowledge and adaptive to new relevant knowledge. We need to test
Chapter 7 - Firm strategies for innovation 189
the applicability of these intuitions on the activities of firms. The research questions that focus this chapter are:
6. What determines the mix of knowledge firms obtain for their innovation activities and why, and who drives this process? 7. How and why does ‘regional experience’ affect this process?
The chapter will test the hypothesis:
Hypothesis 4 – Regional experience affects the mix of knowledge firms’ access by shaping the type and location (internal versus external) of accessible knowledge.
In order to answer these questions and test the related hypothesis, this chapter is divided into two sections. The first section is focused on determining the mix of knowledge accessed by firms. Cluster analysis is used to ascertain patterns in knowledge activities from the factor analysis presented in the previous two chapters. The analysis identifies two clusters of knowledge activities. The second section of the chapter then examines the combination of internal and external sources of knowledge within the firm, to better understand what determines the mix of sources used, and the effect of ‘regional experience’ on the mix within the firm. This second section uses data from in-depth interviews with fourteen firms identified either through the survey or other methods31.
7.1 Cluster analysis of innovative activities in firms
Cluster analysis is ideal for the aims of this research, which are to seek patterns, and classify diverse material. The type of cluster analysis employed is hierarchical cluster analysis. This cluster method is another form of multivariate analysis, and was chosen due to its suitability for smaller sized samples, and the ability to cluster variables as opposed to cases. Ward’s linkage is the selected clustering method. Ward’s linkage uses an analysis of variance approach to
31 The firm selection methods are detailed in Chapter 3
Chapter 7 - Firm strategies for innovation 190
evaluate the distances between clusters by minimising the sum of squares between the two clusters formed at each step. The method is very efficient and tends to produce smaller sized clusters. These cluster steps, however, allow the progress of the agglomerations into clusters to be observed at the various distances, as shown in the dendrogram. This in turn allows judgements to be made on the number of clusters to be identified.
Hierarchical cluster analysis allows for the clustering of variables, and cases. In this situation, variables are initially clustered to show characteristics, and then the cases are clustered, to show firm membership of the clusters. The variables used are the 13 factors created in the previous two chapters. These are: four KISA factors; four knowledge providers; and five organisational networking factors.
Inspecting the resultant dendrogram (Diagram 7.132) at the farthest linkage points, the 13 identified factors from the previous factor analysis divide into two clusters. However, one step before this farthest distance, there are three identified clusters, as shown by the red line. Three clusters offer more differentiation between the groups, and thus the opportunity to get a better fit with the number of cases. Table 7.1 shows the variables associated with each of the three clusters.
32 Full workings of the cluster analysis are available in Appendix 11
Chapter 7 - Firm strategies for innovation 191
Diagram 7.1 Dendrogram using Ward Method
The three clusters identified in the analysis show mixes of knowledge gathering activities. The first cluster is the spectrum-sourced innovation group. The motivation for this group’s innovative activity is no doubt customer based, but for innovation this group draws on a wider spectrum of sources to drive the direction of their innovation. The main sources of knowledge are: universities and government agencies; conferences and exhibitions; and journals and websites. KISA are focused on legal and IP services, accreditation, and industry development advice. Networks are a mix of local and state based government agencies, and Sydney based universities. However, with conferences and exhibitions as a key knowledge source, network reach of these firms would be international.
Chapter 7 - Firm strategies for innovation 192
Table 7.1 Factors and factors characteristics of three clusters (of variables)
Cluster Factors Characteristics 1 • Public sector knowledge sources Knowledge sources- Universities and Spectrum • Commercial information network government agencies, Conferences and sourced knowledge sources exhibitions, Journals and websites innovation • Standardisation KISA KISA – focused on IP, legal, e-commerce ,and accreditation services, and industry group • Industry development KISA • State/Metropolitan/Region location of key development advice knowledge organisations Key knowledge organisations – Mix of state based government agencies, regional organisations, local councils, and Sydney based universities 2 • Customer focused knowledge sources Knowledge sources – Internal , customers, Customer • Enterprise focused knowledge sources competitors, enterprise groups, suppliers, focused • Core KISA and paid advisers innovation • National/ Metropolitan location of key KISA – Core KISA Networks – Mix of nationally based group knowledge organisations government agencies, Chambers of Commerce, industry associations, local and external universities. 3 • Human Resources KISA KISA –Training and recruitment KISA Human • Vocational education Networks –Vocational training institutions Resources • Local location of key knowledge and external universities. focused organisations innovation group
The second cluster is a Customer-focused innovation group. The motivation for innovation in this group of firms comes principally from customers. External knowledge sources are: customers; competitors; suppliers; other parts of the enterprise group; and consultants. The KISA activities are core KISA. Key knowledge organisations include the nationally based organisations. This group of organisations was the largest component identified in the key knowledge networking factor analysis and is correlation with the two main national government agencies supporting industrial and export development.
The third cluster, Human resources focused innovation group is aligned with HR KISA, training, and recruitment services, and key knowledge organisations include local vocational education institutions, and external universities (although this is at a low correlation). Innovation strategies for this group are centred on improving workforce quality.
The cluster analysis was then performed again using the cases. Non-parametric correlations were performed on the factor analysis components that we associated with each cluster, to ensure the clustering was consistent across
Chapter 7 - Firm strategies for innovation 193
variables and cases (correlations shown in Appendix 11). Clusters 1 and 2 were significantly and positively correlated with their factors. Cluster 3 was negatively correlated with factors in Clusters 1 and 2, but not positively associated with their own two factors. This suggests the relationship between human resources development and this cluster may not be a strong as indicated in the cluster analysis. However, Cluster 3 is still significantly different from the other two (due to the considerable negative correlations), and is therefore considered appropriate to be regarded as a separate group.
The memberships of the three cluster groups by cases were established through SPSS analysis. The membership of individual cases’ cluster membership was saved[MT6] as a separate variable to allow further analysis. Of the 113, 7 belonged to Cluster 1, 64 to Cluster 2, and 42 to Cluster 3. These results mean a small number of firms in Cluster 1 are using specialist knowledge sources in their innovation activities. Knowledge sources include universities, government agencies, and conferences and exhibitions. The KISA used is of the non-core or exception variety, standardisation KISA, associated with IP development and accreditation and industry development KISA.
Cluster 2 is the largest group, and it is therefore not surprising that they are associated with the largest factors in each of the three categories. Knowledge sources are customer focused, KISA usage is core KISA, and key knowledge organisations are nationally based export and industry development support agencies, as well as the peak business associations.
Cluster 3 in the cluster analysis, is associated with human resource development through the use of HR KISA, with key knowledge organisations being vocational education institutions and external universities. However, in the correlations with the identified factors associated with the specific clusters, the factors with which this cluster had a negative association were more readily apparent than the factors with which it had a positive association. This means that, rather than firms in this cluster being associated because of certain knowledge gathering strategies, they are associated because they do not have these strategies.
Chapter 7 - Firm strategies for innovation 194
Therefore, the clustering of cases into the three groups produces slightly different outcomes in terms of Cluster 3 characteristics. The title of Cluster 3 is changed to reflect this. The titles of the three clusters are: Spectrum-sourced innovation group; customer–sourced innovation group; and minimal sourced innovation group.
7.1.1 Cluster [MT7]membership cross-tabulations
The second step in this cluster analysis is to compare cluster membership to other variables, in order to further detect characteristics and extract explanations. The first cross-tabulation is cluster membership and innovation performance. Performance on a number of innovation categories established in Chapter 5, such as type of innovation, and degree of novelty, are cross-tabulated with cluster membership. The results are presented in Chart 7.1.
Cluster 1 has the highest levels of innovation in all categories except for organisational innovation. All firms in this cluster are involved in product and service, and operational process innovation, and over 70% of firms were innovating in all other categories, including radical innovation. Clearly, this cluster is highly innovative. This is supported by the characteristics of this cluster as established through the earlier factor analysis. Cluster 2 also has strong innovation performance. It has the highest levels of organisational process innovation, and over 80% of firms innovating in the product and service innovation category. This Cluster is associated with core firm activities, core KISA, and customer focused knowledge sources. The majority of this innovation is incremental, with less than 50% involved in any radical innovations.
Chapter 7 - Firm strategies for innovation 195
Chart 7.1 Cluster membership and innovation performance
Radical Innovation
Organisational Process Innovation
Operational Process Innovation
Product and Service Innovation
0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% 70.0% 80.0% 90.0% 100.0% Cluster 1 Cluster 2 Cluster 3 Source: Outer Western Sydney Business Innovation Survey n=11333
Cluster 3 has the lowest innovation performance of the three clusters. The majority of firms in this cluster were not innovating, with less than 40% of firms innovating in any of the three categories. The majority of innovation being undertaken was also of an incremental type. From the earlier correlations, negative correlations with elements that were central to innovation performance in the other two clusters were noted.
From the cross-tabulation of cluster membership and innovation performance, it is clear that Cluster 1 is the highest performing cluster, followed by Cluster 2. These clusters were associated with two sets of key knowledge gathering strategies, including the types of knowledge accessed, and the methods of access.
With this knowledge of innovation performance of the three identified clusters, further cross-tabulations on other independent variables are completed[MT8] beginning first with region. Chart 7.2 provides the distribution of region across the three clusters.
33 A number of outlying firms were identified in the process of the cluster analysis. These firms were excluded from the cluster analysis
Chapter 7 - Firm strategies for innovation 196
Chart 7.2 Cluster memberships by region
South West Sydney
North West Sydney
Central West Sydney
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Cluster 1 Cluster 2 Cluster 3 Source: Outer Western Sydney Business Innovation Survey n=113
These results confirm earlier findings for innovation performance, and other key characteristics established regarding the regions. Firstly, Cluster 1, which is associated with high innovation performance, is largely made up of Central West Sydney firms and, to a lesser extent, North West Sydney firms. This is not surprising, given their relatively high levels of radical innovation. Additionally, Central West Sydney had the lowest overall levels of innovative activity and knowledge sourcing, and therefore it is not unexpected that Central West Sydney also occupies the largest part of Cluster 3, which represents the lowest levels of innovation performance. Both North West and South West Sydney have equal components of firms that fall into the Cluster 3 category.
The majority of both South West Sydney and North West Sydney firms are located in Cluster 2. This Cluster has strong innovative performance, and is tied to core activities of knowledge gathering such as customer relationships and core KISA usage. In their components of cluster membership, North West and South West Sydney present a similar story; this is in spite of the differing industrial
Chapter 7 - Firm strategies for innovation 197
profiles of the two regions. Chart 7.3 shows the industry sectors of the three clusters.
Chart 7.3 Industry sector and cluster membership
Cluster 3
Cluster 2
Cluster 1
0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0% 35.0% 40.0% 45.0% 50.0% Manufacturing Business and Finance Services Source: Outer Western Sydney Business Innovation Survey n=113
There is minimal difference between the industrial sector components of the three clusters. This in turn leads to speculation that industrial sector (at this broad level) does not affect innovative performance, or firms’ knowledge gathering strategies, in the ways that are typically shown in the literature. This realisation has important policy consequences, particularly for regions without prominent industrial concentrations.
Another key factor throughout the analysis undertaken in this thesis, has been firm size. Chart 7.4 shows cluster membership by firm size, divided into two categories of firm size: less than 50 employees; and 50 or more employees. All of the firms found in Cluster 1, which is the highly innovative group, are small firms. The majority of Cluster 2 is also small firms. Larger sized firms are concentrated in Cluster 3, the lowest performing category. This is consistent with
Chapter 7 - Firm strategies for innovation 198
earlier findings on the innovativeness of small firms, and their higher levels of knowledge access.
This result has a number of important conclusions. Firstly, the range of innovation and knowledge gathering activities of smaller firms is not widely collected in innovation survey, but it is clear from these results that these firms contain a large degree of the innovation activity of regions. The ability to support and encourage these activities is the second conclusion that can be drawn. These firms make up the majority of firms in the three regions under investigation, as they do the majority of regions. Understanding how these firms innovate, and the activities they do in support of this outcome, is essential to creating effective ways to support these enterprises. These mechanisms may not relate to, or fit well with, current regional economic development methods, as a result of the existing focus on the first iteration of RIS.
Chart 7.4 Firm size and cluster membership
Cluster 3
Cluster 2
Cluster 1
0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% 70.0% 80.0% 90.0% 100.0% Small Businesses >50 Larger Businesses <50 Source: Outer Western Sydney Business Innovation Survey n=113
Chapter 7 - Firm strategies for innovation 199
7.1.2 Cluster implications
The hierarchical cluster analysis presented in this section has brought together the analysis presented in the preceding two chapters and combined it into three sets of strategies or modis operandi for external knowledge gathering within firms. Each of the groups is associated with different mixes of knowledge and networking partners.
The three identified clusters are in turn associated with different levels and types of innovation performance. The first, and most innovative cluster, has high levels of product and service and operational process innovation, with a high degree of radical novelty in the innovation. Cluster 2 also has strong innovative performance, which is highest in the organisational process innovation category, and is largely of an incremental nature. Cluster 3 has the lowest levels of innovation performance, with most firms in this group not innovating, or conducting incremental innovation.
Cluster membership is also compared by region. Central West Sydney has the highest levels of membership in Clusters 1 and 3, covering both ends of the innovation performance spectrum. North West and South West Sydney have similar memberships of Clusters 2 and 3, with more firms in Cluster 2 than 3.
Industry sector is shown to have minimal effect on the innovation performance of firms, with cluster membership being comprised of the same levels of firms from manufacturing as from business and financial services firms. Firm size, on the other hand, was shown to have an effect, with all of the firms in Cluster 1, and the majority of firms in Cluster 2, being small firms (less than 50 employees). Larger sized firms were instead concentrated in the lower innovation performing Cluster 3.
In the following section, interview data is analysed to: firstly, examine the mix of knowledge gathering in firms in more detail, particularly the deciding mechanisms that lead firms down these different paths, as illustrated by the three
Chapter 7 - Firm strategies for innovation 200
clusters. Secondly, the next section considers the effect of ‘regional experience’ on these mechanisms. Regional experience is interpreted from data regarding the length of time the firm has been in the region, reasons for the firm locating to that region, and the interviewees’ opinion of the role of their firm within the region.
7.2 Identifying firm strategies
This final section of research results and discussion is drawn from the last part of the research methodology, the in-depth interviews with firms. The qualitative data from the interviews provides a deeper and richer analysis of specific examples of firms’ knowledge gathering activities, and the context in which they occur. By analysing these activities within specific contexts, ‘why’ questions, such as, why do firms select a particular mix of knowledge and why does being in a particular region affect this selection, can be answered.
This section is divided into three parts. The first section focuses on the knowledge sources, and the use and combinations of KISA (both type, and internal and external) by these firms, and the employees within the firm who are responsible for these activities. The second section focuses on the knowledge networks of firms; how these are created and maintained. These sections provide further elaboration on the research results presented for these areas in Chapters 5 and 6. Finally, the third section discusses the ‘regional experience’ of firms and how and why this affects the previous two sets of activities. The in-depth interviews with fourteen firms are used in this analysis.
7.2.1 Knowledge sources and KISA
Through the results outlined in Chapter 5, and through the cluster analysis presented in the first section of this chapter, a number of strategies by which firms approach knowledge gathering are highlighted through the use of different
Chapter 7 - Firm strategies for innovation 201
combinations KISA. The combination of knowledge sources and KISA refers both to the combination of type, as well as the combination of locations. The combination of types was already explored in the first section, and was used to establish the three clusters. Therefore, this section focuses primarily on the combination of location, specifically internal and external, and the location of external knowledge [MT9]in the third section, combined with the discussion on regional experience.
The data collected on KISA and knowledge sources provides analysis of firm capacity and their direction and priorities, and as an element of RIS analysis offers a direction for future direction and development[MT10]. Internally sourced knowledge provides information on the capabilities of the firm, externally sourced shows where the boundaries of the firm lie and, depending on the location of the external source, tell something of the capacity of that location for certain knowledge. The mixing of the two sources further demonstrates firms’ capabilities in terms of assimilating knowledge, as does the extent of their networking with key knowledge organisations. Transitioning of knowledge between internal and external also provides further information on the direction and priorities of the firm.
The terms internal and external knowledge, of themselves are too broad to be analytically useful. Examining the interview transcripts, it can be seen that external and internal knowledge are each broken down into two further categories. External knowledge sourced by firms can be broken down into either: (i) business ‘practice’ knowledge; or (ii) specialist product and process knowledge. Specialist product and process knowledge can be further classified into three elements. Firstly, specialist knowledge used for product and processes development, such as the development of components; secondly, specialist knowledge used for product and process regulations, such as standards and accreditation; and thirdly, specialist knowledge used for market development, such as export information on international markets. These groups of external knowledge relate to KISA, but in different firms the same types of KISA may be used in different ways and for different purposes. These three external categories
Chapter 7 - Firm strategies for innovation 202
(specialist knowledge for product and service development, product and service regulations, and market development) refer to the function of the knowledge; where it is used in the firm, and why. In contrast, the earlier analysis of KISA referred to the type of knowledge used.
Internal knowledge can be segmented into two categories, with the first relating to product knowledge, heavily linked to competitive advantage, and the second being knowledge related to business practice and the functions of the firm. Table 7.2 presents a breakdown of the usages of knowledge and the mix of internal and external knowledge for the fourteen interviewed firms.
The first thing to notice about this table is that all of the firms interviewed used both business practice, and product and process knowledge, internally. This is not surprising, as these activities are closely tied to the perceived competitive advantage of the firm. The points of comparison emerge in studying the types of knowledge relating to both business practice, and product and process development functions, which are sourced internally, compared with those that are sourced externally, and the reasons for this split.
Table 7.2 Mix of internal and external KISA in interview firms Cases External Knowledge Internal Knowledge Business Specialist product & process Business Product & practice practice process Product& Product & Market process process development & development regulation regulation 1 X X X 2 X X X X 3 X X X 4 X X X 5 X X X X 6 X X X 7 X X 8 X X X X 9 X X X X 10 X X X X X 11 X X X X X 12 X X 13 X X X X 14 X X X
Chapter 7 - Firm strategies for innovation 203
The activities highlighted in red are those where knowledge is transitioning from internal to external or vice versa. These examples will be discussed in greater detail in a later section.
The mix process between internal and external knowledge is closely related to firms’ competitive advantage, in that the knowledge activities most closely associated with competitive advantage are more likely to be internally produced, sourced and stored. For example:
“…all the designs for the (products) are done in house, there is very little outsourcing for the (product) design.” (Case 2)
“…we basically design everything, it is our ideas that go into play there, very rarely is it someone else’s, the knowledge of how we can fit the whole thing together because we have the whole range of knowledge across the whole spectrum of business activities.” (Case 5)
“…we do all of the CAD drawing in-house, we taught ourselves, the entire catalogue is now done in the CAD program. When you talk about innovation, product innovation and the way we do things, this is not done this way by anyone else in our industry, it is quite different.” (Case 12)
“It would all be internally really, a lot of stuff if I was to break it down is empirically based, we have a few mad chemists creating new, wonderful formulas, at lot of trial and error…it is an iterative process.” (Case 6)
In all of these examples, the firms directly relate their competitive advantage in either product or process innovation, to their internal knowledge activities. This occurs both with their product, but also with how they organise their business to deliver their product. Returning to the discussion in Chapter 5 on core KISA, the activities associated with the R&D aspects of core KISA are primarily in-house for all interviewed firms, even small ones.
Chapter 7 - Firm strategies for innovation 204
This internal knowledge is embedded in the employees that the firm hires. These knowledge professionals represent a key investment in internal KISA. Firms selected specific skills that they wanted to hold internally, and it is in this skill selection, and the hiring of the people who possess these skills, that firms reveal the type of knowledge they want internally.
“It was a deliberate strategy to target people with these specific skills…to do specific jobs” (Case 4)
“…we recruit them for their specific skills, these skills are not really something you can bring in and train someone in, they have to have a feel for it, they have to have a real flair for it, some people have a flair for designing kitchens some might have a flair for furniture other might like sheet metal components, as silly as it sounds” (Case 7)
For other firms, it is the inability to hire the skills sets that they desire, which leads to the acquisition of different types of internal KISA. The acquisition of internal KISA is always related to firm competitive advantage or an emerging competitive advantage. This is shown in Case 12.
“The inability to get staff led us to making this change – we could not get sheet metal trades workers so we had to come up with another way of doing it, we turned to computers basically.” (Case 12)
“When we were teaching ourselves the CAD system we went to a three-day crash course, and the rest we taught ourselves by using it. We battled like that for about 2-3 years and then M came along. M was university trained in the 3D modelling…he showed us what he could do, and then well he has really taken it to the next level for us, while we were using the program in a limited format, he has taken the program to the level where you do not even have to redraw the product…something that used to take a few days can be done virtually straight away” (Case 12)
Chapter 7 - Firm strategies for innovation 205
The relationship between these knowledge professionals, who provide the internal KISA, and the firm for which they work, can be one of tension. The success of the firms, and the key innovations that are at the core of their competitive advantage, can be threatened by this tension. Two examples from the interviews are provided below.
“We have a very senior industrial chemist here and I must admit that getting information out of him is like pulling teeth, he is always walking around with his black book of secrets…so we have just taken on a new younger guy, he has a PhD and is very clever and he was brought in because he is supposed to be taught or mentored by this senior guy, but the information is not forthcoming ….but we realised that if this senior guy got hit by a bus that we would be very exposed, but he doesn’t want to train anyone, keeps saying that he will always be here and don’t worry, it is a difficult one for us, probably our main problem HR wise around him”. (Case 6)
“I had a brilliant guy, he was good at the innovative phase, the idea phase, he could gush information, but two years ago we were half way through the development of the new product and he just left us…gave us no option. I have felt the burn of people leaving me high and dry and it is not just a theory, it is real, some of these people cannot hack it, they may be brilliant but they cannot work in a commercial environment….he just left, like that, so how do you pick people, because qualifications are not everything” (Case 14)
In both of these examples, key knowledge that was closely tied to the firms’ innovative activities was held by knowledge professionals within the firm, and in both cases, although the knowledge is internal to the firm, there are difficulties with internal knowledge sharing, and with the knowledge escaping the firm when people leave.
Also in both of these cases, the KISA is R&D related, and these activities are also closely tied to IP protection through patenting and legal services. It was shown in Chapter 5 that this type of KISA, associated with standardisation KISA,
Chapter 7 - Firm strategies for innovation 206
is usually externally sought. From the interviews, it is clear that protecting internal knowledge is met with alternate methods among the firms. Patents were the subject of mixed opinions, with some firms not seeing the value of patenting, both in terms of the initial expense needed to gain the patent, and then the protection it affords.
“We don’t waste time with patents; they are not worth the paper they are printed on”. (Case 12)
“…we patented in 1987 which cost me about $340,000 personally to get the patents in 30 odd countries, so it is fairly obvious that we were the first ones to come up with this technology, the problem is because we are such as small company we do not have the money to prosecute people for patent violations”. (Case 14)
“We have had a little bit of a look at it, but it is a lot of work. We are trying to develop the product, and then we have to say, right whilst we are still in developing the product, look at all the paperwork and processes to do with the patent. It is quite a time consuming process, and there are so many other products that you could develop, and you are not even guaranteed that you are going to get this patent, so it makes us very cautious in terms of putting too much time into patents” (Case 2)
Alternatively, for other firms, patents are crucial to the way business is done.
“This is the only way to protect our IP; we spend 6-7% of our revenues on product development and we spend a certain proportion of that on protecting the new products that we are inventing”. (Case 11)
In the firms that were interviewed, there was a tendency for smaller firms to have difficulties in patenting, which was noted by a large firm that had many patents, suggesting that there may be a size dimension to the patenting method.
Chapter 7 - Firm strategies for innovation 207
“…when the company was smaller, yes, we couldn’t patent everything that we wanted to, the company is now successful enough that we now patent everything that we want to”. (Case 11)
As highlighted in the earlier analysis of standardisation KISA, which includes legal and IP services, these services are primarily sought externally, and even internationally, as in Case 11:
“…it is a mix, with about half done internally and half externally, we have 2 full time internal patent lawyers, and externally about another 3-4 attorneys…because the US patents are very important to us, most of them are actually written by US attorneys…we actually originally had our patents written by an Australian and sent over, but we found they were not strong patents in the US” (Case 11)
The question then follows, how do smaller firms protect their intellectual property? In the cases interviewed, a combination of two methods is used. Firstly, continuous innovation, which is only incrementally released to the market, and secondly, use of marketing activities that build brand recognition and reputation for the firm’s products.
In Case 2, the firm found that their products were quickly replicated and mass produced in the Asian markets, and therefore initial advantage through investment in research and development was quickly lost. The strategy became one of significant investment in research and development, particularly in energy efficiency, which is an area of key differentiation, and the subject of emerging government regulations. However, the advantage of this investment in R&D was only retained by releasing these innovations gradually onto the market, which allowed the firm to maintain the benefit for a longer period of time. This enabled the firm to reman competitive with Asian rivals despite price point difference.
“You cannot over-design your product because you will be priced uncompetitive, you will be too efficient…you cannot be too far out of pace
Chapter 7 - Firm strategies for innovation 208
with the market…we put a lot of effort into our innovation, yes our product is definitely better and we have a few more years of new products to come”. (Case 2)
The second aspect of preserving competitive advantage instead of investing in IP protection is in the marketing of both products, and the firm. Marketing and promotion services are identified as another element of core KISA. This KISA activity shows a great deal of external and internal mixing of knowledge resources. Brand development is a particular avenue that allows smaller firms to develop and maintain competitive advantage. In addition to marketing and promotions KISA, these activities also increasingly call upon IT services and e- commerce. As is shown in the following cases:
“…if you have got a good product, and we have got a pretty good product, what will get you over the line is the trust that you can develop with the end user and the buyer, and are you going to be around after the sale to support him, to service him, that is where reputation comes in.” (Case 6)
“…this is another side of our competitive difference, businesses in our industry in this area and other areas in Sydney have had a hard time in recent years, and they have responded by cutting staff, advertising, sold off assets. We weathered that storm and said “in this market we need to spend more not less to retain our business”, so we went against the trend in our area, and it was hard, we stayed static for a period, but it paid off in the long run because our market share has since increased significantly.” (Case 1)
“…we have used external advertising, design and media companies to create our branding and our sub-brands…we use professionals for this…” (Case 8)
“We have a marketing strategy company that we outsource to, we do not have a lot of resources, we do have two marketing assistants here, but not that high level strategic marketing, we are outsourcing that. When we go
Chapter 7 - Firm strategies for innovation 209
overseas with the new product the marketing strategy people will be involved in that too” (Case 2).
In all of these cases KISA was sourced externally, because greater and more specialist expertise was required than the firm held internally. However, external resources were combined with internal resources to allow for internal skill and capacity building, and to enable the maintenance of such branding programmes to be carried out internally. In the discussion at the beginning of this section on the uses of external KISA, three different uses were identified from the firms interviewed. These were: specialist KISA used for product and process development; specialist KISA used for product or process regulation; and specialist KISA used for market development. The above two examples represent specialist KISA used for this third purpose.
Specialist external KISA used for product and process development usually took the form of component development. Such component development was seen as outside the main product development that was happening internally.
“We do get advice with plastic suppliers and adhesive suppliers, where we want a particular product for a particular job they will advise us the best material for our needs” (Case 3)
“…we get the software tools from external people to speed up our capability in developing up the product” (Case 14)
External KISA were also used for the development of product and process standards and accreditation.
“…then I get a consultant in and I know what standard we have to obey and I ask them to specify, because when you read all of this jargon you will go crazy, so I need then to translate really, I need them to explain it to us, so when we design the final product we can take account of all the international standards, so no matter where we want to sell, we will not have a problem”. (Case 10)
Chapter 7 - Firm strategies for innovation 210
In addition to these three areas of specialist KISA provision, firms also access external KISA for business ‘practice’ knowledge, such as human resources, business planning advice, and IT services.
“For the last two years we have been outsourcing our human resources and recruitment functions” (Case 2)
“ I am in a business group that meets one day a month…there are about 16 in the group, all CEOs …we have a resource speaker on a different topic each time, usually for about half the day, and then we have meetings, discuss issues…it is very valuable to me, sort of replaces an MBA program” (Case 5).
There are also a number of instances within the cases investigated, surrounding KISA in transition; those that are moving from being internal to external, and vice versa. In the above example of externally sourced HR KISA, the same company has now moved this function, and the associated KISA, internally. They explain why this knowledge now needs to be internal:
“We have just employed a HR manager…it is going to make it a lot easier, so skill matrices will be finished, training processes in place…in this current environment we need to (referring to current record low levels of unemployment)” (Case 2).
An example of KISA transitioning from external to internal and then back to external is also provided by another firm.
“The printer circuit boards is a good example, when the company started all of the boards were designed and made by an external company, and then after a couple of years we bought that external company…and now we are in the process of outsourcing all of the printer circuit board manufacture” (Case 11)
Chapter 7 - Firm strategies for innovation 211
The transitioning of KISA from internal to external and vice versa suggests changes in the direction of the firms’ priorities. In Case 12, the firm initially sourced HR functions externally, and hence the KISA associated with these functions were also external, but as the firm grew and more control and specialisation was needed in terms of human resources functions, such as training, skills matrixes and staff development, the function and associated KISA was moved to be internal. In Case 11, regarding the transition of the KISA associated with printer technology, the firm made the function and associated knowledge internal when it became vital to the firms’ product development. Then, as this importance waned (largely due to the increased functionality of digital recording), the printer technology and associated knowledge again become less relevant to the core product development, and is therefore now in the process of becoming external again.
These transitioning phases of functions and associated KISA offer clear examples of the changing direction of innovative activities within these firms, and how the internal and external KISA sourcing of the firm is a dynamic and constantly changing process, in line with the continuing development of the firm and its product offerings.
Competitive advantage, and where this lies, determines what knowledge firms access internally, and what is accessed externally. For the firms that were interviewed, knowledge that is closely associated with key competitive advantage, either through product development or business practice, was sought to be held internally. This internal provision is accomplished through the hiring of specific staff, ‘knowledge professionals’, which of itself can be a risky and tense activity.
These results have also shown that this is a dynamic process, ongoing and changing over time, with the evolution of the firm. The analysis presented from the interviews gives a particular picture of mixes of KISA, which is a snapshot at
Chapter 7 - Firm strategies for innovation 212
this particular point in time, but these activities will continue to shift. Therefore, any policy developed focusing on these activities must keep this in mind.
The following section examines firms’ knowledge networks, and the key organisations within these networks.
7.2.2 Firms’ knowledge networks
In the previous chapter, discussion on firms’ knowledge networks was restricted to non-firm organisations, which the discussion in this section will also follow. The same organisational categories used in Chapter 6 will also be used: governmental; industrial; educational; and region based organisations. The purpose of this section is not only to analyse the non-firm networking relationships of the firms under investigation, but to also analyse the knowledge that firms receive from their networks.
In the firms that were interviewed, the key external organisations within their networks were either educational or industrial organisations. The response of firms to these organisations, and the knowledge they acquired from them, varied significantly. Educational institutions were accessed for the latest technology and ‘ideas’, and applications of each of these in the respective firms’ field of expertise. In Case 9, the firm conducts internal research and development, and maintains significant relationships with a number of external overseas universities in order to gain access to global ‘best practice’ knowledge in their field.
“Quite often I will hook them (my staff) into a university where I perceive the best value to be, so if it were patent design then I would go to the University of California, (specific product) management I would go through the University of Birmingham in the UK. For the technology I would go to the University of Denmark, I have sought these contacts out, these are global leaders. I have spent many years
Chapter 7 - Firm strategies for innovation 213
cultivating this group of people, so that I can go over and see them and find out what new things are happening” (Case 9)
For other firms, the local universities provide this access.
“The local campus of the University of Western Sydney, Macarthur Campus, has a technology network and they run programs” (Case 5)
For Case 11, university links are developed not only through contact with universities across the world and one specific university based in Sydney, but also through a large grouping of staff members’ continuing activities with universities, either through continuing joint positions, visiting fellowships, or internship programs for staff-supervised graduate students.
“…a lot of people are here from universities, they come here and stay, some are still Professors…we also have these summer internship programs where we have about 20 graduates come, and some stay on with permanent jobs” (Case 11)
However, for Case 10, the universities are not seen as much of a source of knowledge compared with other educational providers.
“M Technology is much more relevant to us than any university would ever be because they are hands on, practical, in the field, because unfortunately I have to say that my experiences with universities is they are far too much theoretical and for me this is not the right platform. I have tried many times, it is just not suitable, we are completely apart I have to say that”. (Case 10)
Industrial network contacts were evident in the majority of firms interviewed; however, the value of these contacts for the firm also differed widely. The value derived from industrial networking was dependent on the firms’ position within the industry, rather than the particular type of industry involved. Smaller firms relied on these contacts for product and process knowledge, whereas larger
Chapter 7 - Firm strategies for innovation 214
firms regarded participation more as method of shaping the regulatory environment and maintaining their position within the industry. For example:
“…we attend the meetings (of the industry association) every three months, and that is really to keep on top of what is happening in government in terms of energy performance, regulations, greenhouse gas policy…the sort of things that would affect the whole industry” (Case 2) (Large firm)
“I am a member of PIMA (Plastic Injection Moulding Association), these meetings are important. They get in experts to talk about different aspects of plastics…I have learnt a lot of things at those meetings…I solved a very difficult problem I was facing when I spoke with one of the speakers they bought out from England and he gave a talk on metal injection moulding, imagine that - injecting moulding metal instead of plastics, you would not think it was possible but it is, and it was what I needed (for medical devices product)” (Case 3) (Small firm)
“We regularly go the Association of Hydraulic Services Consultants…this is the most important of all our networking activities as these are the people that specify out product…you are really only talking to a few people from a few firms but the knowledge circulates well from there…” (Case 12) (Small firm)
In the factor analysis of key organisations presented in Chapter 7, government agencies were highlighted as key networking contacts. The main contacts were the nationally based industry and export development agencies of AusIndustry and AusTrade. Five of the companies interviewed have ongoing relationships with these two organisations. There is a deliberate inter-relationship between these two agencies, to facilitate them in jointly encouraging industry development and export development. This means that in all the cases analysed here, when a firm has a relationship with AusIndustry, they also have a relationship with AusTrade. In relation to AusTrade, firms note that this is really the only option open to them for assistance with overseas market development.
Chapter 7 - Firm strategies for innovation 215
“We are very much involved with AusTrade…I have a fair bit of time for them, they have helped us and I don’t really know of any market where they have not helped, I used them in Egypt for example, and in South Africa, our Sales and Marketing Manager uses them throughout Asia, our customers in some of these countries tend to be a little bit underhanded so AusTrade people are also nice to have sitting next to you in the meetings so things usually go a bit smoother…” (Case 6)
The focus of the KISA provision by the nationally based government agencies is on specialist market development and regulation KISA, whereas the focus of state agencies is more on business practice KISA.
“DSRD, by far, over the years I have found them to be the best, anything I have found out has been through them, that they have alerted me to, I would say 95% of the things we have learnt about are through them…I am an engineer, that is my background, so I had no idea on marketing or anything like this when I started my first business and DSRD helped.” (Case 10)
“We go to the odd DSRD meeting or training course, go there as much for the training as the networking” (Case 2)
Therefore, in terms of what shapes the relationship between a knowledge organisation and a firm; it is the suitability of the knowledge to that particular firm. Suitability is determined by the size of the firm and its position within the industry, and also the type of knowledge sought. All of the firms interviewed also referred to the length of time they had spent cultivating relationships with key contacts within these knowledge organisations. This process of relationship development is lengthy, in many cases taking several years of ongoing contact for trust to develop.
This is one of the key areas where ‘regional experience’ has an effect; in the ability to create a series of contacts between key knowledge organisations and the firm rather than just specific individuals in the firm, although these relationships are still manifested in people employed by the
Chapter 7 - Firm strategies for innovation 216
firm. Longevity of the firm, and longevity of the firm’s existence in the area, establishes the reputation of the firm, not only in terms of their industry and what they do, but in terms of where the firm is located. Regional experience is explored further in the next section.
7.3 Regional experience and its effect
The final area of discussion regarding the interview results relates to firms’ regional experience. By this is meant how firms relate to their region, what they see as the advantages and disadvantages, what were their reasons for locating here, and why. Regional experience is embodied in the people who are employed by the firm.
7.3.1 It is where the boss lives
For the majority of firms interviewed, their selection of region was based on where the initial entrepreneur lives or lived. The ability to live and work in close proximity was highlighted as a key advantage by most firms.
“The only reason we are here is because the business started in Winston Hills (suburb within the region), in the garage, went from the garage to Seven Hills (another suburb within the region), we had a number of factories there, built another factory at Kings Park, which is also local, we did not want to be too far from home. The business was growing more and more but still we did not want to be travelling too far, that is why we chose this location” (Case 2)
“I started work here in 1980 and I already lived here…I think it is great. I can work until 7pm at night and then just have a quick commute home” (Case 3).
“I live about 5 minutes away, the owner is about 15 minutes away, we were sick of the M4 (major highway). It used to be that if you left home at
Chapter 7 - Firm strategies for innovation 217
6:30am you would miss the traffic, then it was 6am, then it was 5:30am, so we would be getting up earlier and earlier to try and miss the traffic, you would be working 13 hour days and not really even wanting to, just to miss the traffic. One day it took me two and a half hours to get to work, I got to work and let off some steam to the boss about the traffic and he agreed…two weeks later we had bought this site.” (Case 12)
This desire to work close to where key professionals live is further emphasised by the fact that many firms did not believe their location impacted on the functioning of their business. They did not see it as an advantage to be located in a particular place for their customer or supply chain relationships.
“I think I could do this in any region, in fact a lot of guys in this business are moving up to the Sunshine Coast (in Queensland, a neighbouring state) because it is nicer living and they get incentives form the government. I had a extremely interesting offer a few months ago to go to the USA, free real estate, university support, I seriously considered it, because that is where all the action is in this business, except I have got a wife and she thought it was a bad idea.” (Case 14)
“It is not essential for us to be here in our line of business, our biggest customers are all over the country, we could be parked anywhere to be honest and we would still be doing business” (Case 6)
In fact, after a desire to live and work in close proximity, access to labour force was considered the main other key advantage for firms to locate in their region. Locating to these suburban areas of rapid population and labour force growth was seen as a way of ensuring a future supply of ready labour for the growth of the business.
“One of the reasons for moving to this area was that we wanted to move into an area where we though people had some commitment to their job, in other words they generally had a house, a mortgage and a family…and that has proven to be reasonably successful” (Case 5)
Chapter 7 - Firm strategies for innovation 218
“When we relocated 15 years ago from Alexandria, because back then Alexandria was the industrial heart of Sydney, we thought okay where is a central location, where we are never going to have problems getting people, we though Liverpool (in South West Sydney) and we bought this building…but we have had problems getting people in the last 5-6 years” (Case 7)
The availability of cheaper land was also a factor in the initial location of some firms.
“No the first business that we had we came close to here because it was the cheapest area we could rent premises and to have a bigger factory. That was the only reason we came, well not the only reasons it was a good price and we liked the area, but we were not living here at the time and after about 7-8 years having the business here we loved it so much that we moved. So we decided to move instead of moving the business.”(Case 10)
The key disadvantages noted by firms in operating in these regions emerge from the rapid development of these regions. The firms remark the lag in infrastructure development, particularly in regard to roads and city amenities as key drawbacks for operating in these regions.
“…getting here is a nightmare from a road point of view, yes sure it is a big industrial area but the phone and telecommunications systems here are bad and there is no public transport” (Case 6)
“…there are no nice places out here, cafes etc where I can take people out. There are also some safety issues with the CBD at night, I think it is unsafe to be walking around on the street at night.” (Case 12)
And also the prevailing stereotypes of the suburbs.
“Perception wise, you are still a Westie…Doesn’t so much affect us for data collection but on the consulting side yes, you are up against Sinclair, Knight Mertz, and all the larger companies. All in the city and North Sydney, the
Chapter 7 - Firm strategies for innovation 219
perception is if you can’t work in the city you cannot afford it, if you cannot afford it then you must have cash flow problems, or you cannot attract enough business or whatever….I think it influences some (potential customers) but we try and counter it well, we don’t let it worry us, we know that it is there, but if we let it worry us, it starts to seep through into your strategies and how you speak, and present yourself to the market, there is no point making a big deal of it.” (Case 9)
Therefore, for the firms interviewed, the primary reason for locating to a particular region was proximity to where the entrepreneur and key staff resided, or access to cheap land. The initial reasons for the selection of the place of residence is mixed, from a long and ongoing family background in the area, to being a nice place to live, and a lifestyle choice outside of the main part of the city. This selection of location did not in any way affect the breadth of knowledge sourcing these firms undertook outside the region.
“Always located in western Sydney, ever since I moved to Australia, but I worked at Bankstown at Hawker de Havilland, and when I started out on my own I was still travelling a lot, because my network especially from Hawkers is all over the country, so I do a lot of travelling, a lot of work in Perth and South Australia”. (Case 4)
All of the firms interviewed had external knowledge sourcing outside of the region, and many had key overseas knowledge sources
7.4 Summary
This chapter analyses the strategies of firms towards knowledge gathering. It brings together the results of the previous three chapters, which have established the elements of the RIS in the three regions: knowledge base; business innovation activity; knowledge flows; regional orientation; and key knowledge organisations and their location.
Chapter 7 - Firm strategies for innovation 220
Hierarchical cluster analysis provided three clusters of variables from the RIS elements identified, and highlighted three strategies used by firms for knowledge gathering: (i) Spectrum-sourced innovation group; (ii) Customer focused innovation group; and (iii) Human resources innovation group. In Cluster 1, the Spectrum sourced innovation group, knowledge is sourced from a diverse array of external sources, including government agencies, universities and conferences, and journals. KISA used by this group are highly specialised, and primarily from the non-core KISA categories of Standardisation KISA and Industry development KISA. Within Cluster 2, the Customer focused innovation group, as the name suggests, knowledge is sourced from customers and competitors. KISA used by the cluster are Core KISA, and key knowledge organisations are the peak business associations and national government agencies that support industry and export development. The third cluster of variables is associated with HR KISA and key knowledge organisations for vocational education.
Hierarchical cluster analysis was then used to cluster cases. Correlations ensured clustered variables matched the clusters of firms. Clusters 1 and 2 were matched with significant correlations in the key characteristics associated with the clustered variables. Cluster 3 did not exhibit any significant positive correlations with its two associated factors, suggesting that the relationships between HR KISA and vocational education, and Cluster 3, are not as strong as the cluster analysis shows. Cluster 3 [MT11]is not correlated with a number of key knowledge gathering activities, which are shown to reflect firm innovation performance, such as Core KISA, and Customer knowledge sourcing. This refers back to discussions about causality in Chapter 3, that whilst correlations cannot detect causality, causality can detect associations. Associations provide guidance on causality may or may not lie, but irrespective of this, without association there cannot be causality. This highlights that associations can be just as informative in directing future research to areas where causality may potentially lay, and eliminating areas where it does not. Therefore, the associations for this cluster show the characteristics that are not related within the cluster, rather than the ones that are. This knowledge is just as valuable to innovation analysis, as is
Chapter 7 - Firm strategies for innovation 221
evidenced when cluster membership is compared with business innovation performance.
The three clusters are very different in terms of innovative performance. Firms within Cluster 1, which is the smallest cluster, have very high levels of innovation. All of the firms belonging to this cluster were innovating in product and service, and operational process innovation categories. This cluster also has the highest levels of radical innovation. Cluster 2 is the largest group of firms and also has strong innovation performance, with high levels in all innovation categories, but particularly in organisational process innovations. Cluster 3 has the weakest levels of innovation performance; many firms within this cluster are not innovating. The innovation that is occurring is primarily of an incremental nature.
Cluster membership comparisons by region showed that Central West Sydney made up the primary components of Clusters 1 and 3. The extremes of innovation performance are demonstrated within the one region. North West and South West Sydney have similar memberships across the three clusters.
The same comparisons between cluster membership and broad industry are also made. These showed that innovation performance (via cluster membership) was equally spread across the three clusters. This explains the similarities in cluster membership of North West and South West Sydney firms, despite differing industrial bases.
However, differences emerge when firm size and cluster membership are compared. All of Cluster 1 and the majority of Cluster 2 are small firms. The majority of larger firms are located in Cluster 3, which shows that there are a number of larger firms exhibiting minimal innovation activity. This highlights a limitation in other regional innovation studies that concentrate analysis on larger firms, because it indicates that the ability of the analysis to adequately capture the true depth of regional innovation in firms may be lacking if only larger firms are studied.
Chapter 7 - Firm strategies for innovation 222
The firm population characteristics discussed in Chapter 4 indicate that the three regions are dominated by small firms. This is encouraging in terms of the regions’ innovative capacity, but is also important to recognise when shaping policy aimed at encouraging firm innovation in these regions. It also highlights the necessity of considering the broader characteristics of small firm behaviour, and the eventual consequences such a firm population will have on the call for public resources, particularly in relation to training of knowledge professionals (internal knowledge resources), and the variety of external knowledge flows for firms (Martinez-Fernandez and Sharpe forthcoming).
The final section of this chapter draws on in-depth interview data to provide more contexts to the relationship between internal and external knowledge sourcing in firms. Whereas the cluster analysis determined the mix of knowledge types and the external dimensions, the interview analysis highlights the relationship between firms’ identified competitive advantages and their internal knowledge resources.
Internal knowledge within firms falls into two categories: (i) product and process knowledge; and (ii) business practice knowledge. External knowledge gathering is done by these firms for specialist reasons. In terms of product and process knowledge, three additional categories are identified: (i) specialist knowledge for product and process development (usually related to component development); (ii) knowledge for product and process regulation (standards etc); and finally (iii) knowledge regarding market development and regulation.
The drivers of these processes are key knowledge professionals employed by the firms. The interview analysis showed a tense and even precarious relationship exists between firms and knowledge professionals, where these people are not the initial entrepreneurs. This tension is further exacerbated by the limited range of IP protection firms feel they have available to them.
Chapter 7 - Firm strategies for innovation 223
The effect of ‘regional experience’ is minimal. Many firms noted no apparent advantages to their location other than proximity to the entrepreneurs or other key staff’s residences. This is noted by nearly all of the firms interviewed. Many firms noted that they could conduct their business from any location in Australia.
Reasons for the decision to reside in these areas were mixed, from long standing family connection with the area, to lifestyle choices to live on the outskirts of the city.
Locating within a particular region did not limit the knowledge gathering activities of these firms in any way. Many firms have cultivated relationships with key knowledge organisations over a long period of time. These knowledge organisations were located at the Sydney metropolitan level, nationally, and in many cases, internationally. Therefore hypothesis 4, that regional experience affected the mix of knowledge that firms access, by shaping the type and location of accessible knowledge, is rejected.
The next chapter presents a summary of the empirical research results, and draws conclusions regarding RIS in regional innovation analysis and the development of innovation-led regional economic development.
Chapter 7 - Firm strategies for innovation 224
Chapter 8 - Conclusions and recommendations for Urban Regional Economic Development Policy
This thesis analyses industrial innovative activity within three regions of Outer Western Sydney. The purpose of the research is to understand the socio- economic development of these regions, using the lens of regional innovation analysis. Innovation is widely acknowledged as the driver of economic progress, and innovation is also increasingly recognised as a spatial process. This chapter presents the conclusions and policy recommendations for local government based on this research.
8.1 Research Conclusions
In the introductory chapter of this thesis the objectives of this research were established. The objectives were threefold. Firstly, provide a regional innovation analysis of three Outer Western Sydney regions, i) North West Sydney, ii) Central West Sydney and iii) South West Sydney. Second, construct and deploy an integrated RIS framework as the conceptual basis for this research. Thirdly, operationalise this framework through the recently established KISA
Chapter 8 – Conclusions and Policy Recommendations 225
methodology (OECD 2006) and thereby offer further commentary on the usefulness of this methodology in knowledge analysis in regions.
8.1.1 Regional innovative activity in Outer Western Sydney
This research investigated the innovative activities of firms within three regions of Outer Western Sydney. The aim of the research was to identify characteristics of firm innovative activity in the regions, as a means of understanding regional economic performance, and as a guide to developing innovation-led regional economic development policy for one region in particular, South West Sydney.
Population growth in South West Sydney has been enormous in the last decade, and future population projections suggest a continuing high growth rate. Employment growth in the region has not matched population growth. This, of itself, would not necessarily be cause for concern in an urban metropolitan region, with ready access to jobs in other areas, but in a context of poly-centric metropolitan planning, encouraging regional self containment of employment at the expense of metropolitan accessibility, economic and employment development is a key issue for the current and future sustainability of South West Sydney.
The link between innovation and economic development is clear, as is the growing recognition that innovation is affected by spatial processes, which is the basis for the regional innovation analysis presented in this thesis. The research uses a Regional Innovation Systems framework. This enables the analysis of four identified elements of RIS: knowledge base; business innovation; regional knowledge flows; and regional orientation. The results of the analysis are summarised in Table 6.2 and are shown again in Table 8.1.
Chapter 8 – Conclusions and Policy Recommendations 226
Table 8.1 RIS analysis elements Characteristics Central West Sydney North West Sydney South West Sydney Industrial • Manufacturing • Business & • Manufacturing Knowledge Base • Engineering Financial Services • Engineering • Business & Information Business innovation • Operational • Process • Product & Service Process innovations (both Innovations Innovation types) • Incremental • Radical innovation • Incremental innovation innovation Knowledge flows • Sources – • Sources – • Sources – customer & public customer, customer and sector enterprise & commercial • KISA – Core KISA commercial networks & Standardisation information • KISA – Core KISA KISA networks & HR KISA • KISA – Core KISA & Standardisation KISA Type and Location • Local Councils • State and federal • Local Chamber of of key knowledge and Sydney government Commerce, local resources based universities agencies, University and • Regional to Universities and local TAFE national level industry college associations • Regional to • Regional to Sydney national level metropolitan level Supply chain • Regional to • Regional to • Regional to orientation national international level Sydney metropolitan level
These results illustrate the divergent innovation activities of the three neighbouring Outer Western Sydney regions. Variations between the regions were highlighted in each of the four elements. The results suggest five main conclusions and their associated policy implications. Firstly, the importance of regional level innovation analysis is demonstrated. All three of these regions are within the same metropolitan area. Analysis at the metropolitan or state level (the lowest level where any form of innovation statistics is readily available) would not be able to detect this diversity. Knowledge of this level of diversity is important, because the variations represent fundamental differences in the processes and behaviours of firms across the three regions. If the outcome of either industrial or regional economic development policy is the encouragement and support of innovation activities in firms, then the effect this policy has on a region would be different. Policy applied across the entire area may be beneficial in one region, but harmful or simply ineffective in another. In circumstances of limited public funding, this presents clear evidence of the need for targeted
Chapter 8 – Conclusions and Policy Recommendations 227
regional economic development policy with an understanding of the elements of innovative activity in individual regions.
In turn, this highlights the role that local government could play in the development and application of regional economic development policy. Although responsibility for industrial and economic development policy within the Australian federalist system is shared throughout the three levels of government, the local level has the least prominent role, yet is arguably in the best position to tailor policy and public sector involvement in regional innovative activity.
Secondly, the variations in innovative activity did not appear to be related to the industrial profiles of the three regions. The analysis on KISA presented in Chapter Five highlighted some key differences in the types of knowledge accessed by firms from either manufacturing or business and financial services sectors, but this was largely restricted to non-core knowledge intensive activities. In terms of knowledge sources and business innovation performance, results were remarkably similar. This does not mean that there are not industrial differences in innovative activity, but suggests that at the regional level, where and how firms source knowledge is similar across both of these broad industry groups. There is clear variation between the regions in terms of their levels of national and international knowledge sourcing, and hence global embeddedness. Therefore, the focus of regional economic development policy should be on developing capability and access to knowledge within the region rather than targeting particular specialised industries.
Thirdly, firm size, in contrast to industry type, did prove to be an important factor in explaining innovative activity. Small firms displayed higher levels of innovative activity, and higher and more diverse levels of knowledge sourcing. The industrial and economic profiles of the regions that were presented in Chapter Four revealed that the vast majority of firms in these regions are small
Chapter 8 – Conclusions and Policy Recommendations 228
(less than 50 employees). In terms of the innovative capacity of firms, this is a positive feature, but it also provides a challenge for policymakers, who need to take note of the dimensions, limitations, and call on public resources made by small firms in carrying out their innovative activities.
The interviews highlighted the challenges smaller firms face in regard to intellectual property protection for their innovations. Although this is not something that can be addressed specifically at the regional level, it is suggestive of the limitations that smaller firms face and issues to which public sector agencies need to be sensitive.
Fourthly, there also appeared to be a relationship between the age of a firm and its innovativeness, with the highest levels of innovation detected in new firms (up to four years old), followed by mature firms (over nine years old), with a period of consolidation and less innovation taking place in established firms (five to nine years old).
Both of these last two conclusions have important consequences for policy development. Small and medium sized firms have been an area of focus for policymakers, due to their acknowledged higher levels of innovative activity, but policy application has at times been haphazard. This research has shown that smaller firms draw on a wide variety of knowledge sources, including many that are external to the region. These external connections are important in maintaining the flow of new knowledge, both into the firm and, through the firm’s regional interactions, into the region as well. Research has also shown that the types and sources of knowledge accessed by firms change with the age of the firm in line with the evolution of the firm’s innovations and competitive advantage. The interview data also highlighted that this relationship also extends to the types of knowledge that are internal and external to firms. The discovery that innovative activity varies with the business cycle of the firm has also been noted. The analysis of knowledge activities in firms is key in understanding their
Chapter 8 – Conclusions and Policy Recommendations 229
innovative behaviour and capacity. This analysis also shows that knowledge activities in firms are a dynamic process. In order to explore these activities all of these dimensions must be considered. Therefore, policy development at the local level needs to take better account of how, why, and when small firms innovate, and the partners they use in this process.
Finally, whilst the research has highlighted the differences between regions, the cluster analysis presented in Chapter Seven shows that intra-regional variations also exist. Therefore, regional economic development policy must be able to take into account the specific variety of innovation existing within a particular region. In most cases, regions will contain firms represented in all the three cluster categories; from highly innovative firms through to firms with minimal levels of innovation. Policy needs to be focused on the fact that innovation must be undertaken by all firms that want to remain in business. Means by which successful innovative firms can share their knowledge with less performing firms in the region should be explored.
8.1.2 Policy recommendations for South West Sydney
The integrated conceptual framework explored in this research also highlights the integrated nature that policy and policy makers will need to adapt. Factors ranging from computer/ internet access, roads, public transport, residential density, distance to work and quality of life, all affect the component elements of RIS. This way clearly demonstrated through the case study work. Therefore further work on understanding these demands and how this translates to responsibilities for the various levels of government is necessary work for the future. This research has provided a starting point; an empirically proven identification and characterisation of three regional innovation systems. This in itself proves that local government has a role to play in urban regional economic development.
Chapter 8 – Conclusions and Policy Recommendations 230
As explained in Chapter One, this research is part of an Australian Research Council linkage project with Liverpool City Council. The Council is the leading local government authority in South West Sydney. The RIS analysis presented a specific profile for South West Sydney, summarised in Table 8.1.
This profile concurs with the industrial profile of the region as an established manufacturing hub. Innovative activity in the region is focused on incremental product and service innovation, which is consistent with a mature manufacturing sector. Knowledge sources are concentrated around the supply chain. This especially applies to customers as knowledge sources and the types of knowledge used, in particular HR KISA, which are also consistent with the manufacturing profile.
Table 8.2 South West Sydney RIS analysis Elements South West Sydney characteristics Industrial and Knowledge Base Manufacturing industrial concentration Engineering knowledge base Business Innovation Product and service innovation Incremental innovation
Regional Knowledge Flows Key sources – customers and commercial information networks KISA – Core KISA and HR KISA Regional Orientation Key organisations – Local Chamber of Commerce, local university, local TAFE college Location of Key organisations – Regional to Sydney Metropolitan level Supply chain concentration – Regional to Sydney Metropolitan level
The orientation of the region is strongly locally aligned. This is the key point of comparison between South West Sydney and the other two regions. Both Central West and North West Sydney display broader orientations in terms of the location of supply chain and key knowledge organisations. This means that South West Sydney operates in a narrower flow of knowledge than the other two regions. Firms have less opportunity to access regionally external knowledge, which is a source of knowledge that is widely acknowledged to rejuvenate
Chapter 8 – Conclusions and Policy Recommendations 231
knowledge flows. The domestic location of primary customers who are predominantly in the local and metropolitan area also shows less emphasis on the development of international markets and exporting, both of which are significant attributes in firm development and key regional economic and employment growth.
South West Sydney’s regional focus, however, means that the local government authorities in the region are in a position to effect change. Strategic policy recommendations are:
• Extend the orientation of firms in the region, both in terms of supply chain, and location of key knowledge institutions. This can be done through publicly supported networking arrangements with a national and international focus.
• Strengthen government focus on creating public amenity and development within Liverpool (the regional centre of South West Sydney), to encourage the location of KISA providing businesses. This will enable local firms to access a broader range of KISA at the regional level, and will also increase the availability of local knowledge occupations.
• Develop closer relationships with the local university, aimed at encouraging more radical innovation. The relationships with local higher education institutions (University of Western Sydney and the local TAFE College) are already in existence. The emphasis now needs to be on increasing the knowledge intensity of these relationships.
8.1.3 RIS: an integrated approach
Chapter One established the advantages, and challenges, of the RIS approach. The advantage of RIS is the ability to consider regional economic performance through an innovation lens, at the regional level. The RIS approach has
Chapter 8 – Conclusions and Policy Recommendations 232
experienced a rapid rise within the theoretical and policy fields, so much that Lagendijk et al (2005) have spoken of its immediate acceptance as both a theoretical construct and normative policy option. As Chapter Two outlined, the RIS approach has its basis at the intersection of economic geography and innovation scholarship. The approach initially focused on the explanation of ‘star’ or ‘ideal’ RIS such as the often referred to ‘holy trinity’ of Silicon Valley, Baden-Württemberg, and Third Italy. RIS’ rapid accession in policy terms led to an expansion of the number and variety of regions analysed within the approach. The focus is increasingly on lagging regions, and ways and means for these regions to “catch up” (Lagendijk et al 2005), even though the dimensions of analysis and the measures of success for these regions were still largely based on the ‘ideal’ version of RIS.
Past research has also focused on particular elements of RIS to create typologies of regions and their innovative capacity. Each element and associated typology privileges certain aspects of innovative activity. In the expanded focus of RIS now used, these elements may not be present in all regions. Therefore, in line with recent calls from the literature (Doloreux and Prato 2006), this thesis has made an effort to contribute to more readily applicable measures, through the integration of key elements in a single approach. The integrated RIS analysis presented in this thesis focuses on four elements: knowledge base; business innovation activity; knowledge flows; and regional orientation.
Chapter 8 – Conclusions and Policy Recommendations 233
Knowledge Business Base Innovation Activity
Integrated RIS Analysis
Regional Orientation Knowledge Flows
8.1.4 KISA analysis
Instrumental in developing this integrated approach is the analysis of Knowledge Intensive Service Activities (KISA). KISA analysis is an emerging area of innovation research, which offers the ability to analyse specific types and uses of knowledge that are critically linked to the innovation activities undertaken by firms. The advantages of KISA analysis is that it allows the type and mix (such as ‘core’ KISA) of KISA that firms use to be analysed, and from this, estimations and conclusions about innovation activities in firms and regions are able to be made.
The analysis of KISA identified a group of what was termed ‘core’ KISA. These were knowledge intensive service activities that were accessed the most by firms. These activities were accessed by both low KISA and high KISA using firms, just in lower levels by the low KISA usage group. Activities included in ‘core KISA’ were: marketing and promotion; research and development; and IT services. The core activities were in line with other empirical research results (Albors, Hervas et al, 2007).
Chapter 8 – Conclusions and Policy Recommendations 234
A further advantage of KISA analysis within the regional context is its ability to examine a wide variety of regions. KISA will always be present, just at varying levels, with lower and more basic levels (around ‘core’ KISA) in weak RIS, and at higher and more sophisticated levels within stronger RIS.
8.2 Limitations of the research
Regionally focused research is always challenging. The smaller geographical scale, and lack of readily accessible data, makes a creative and innovative approach necessary. However, this research has highlighted the importance of understanding regional variations and the effects that these variations may have on regions. In addition, working within a ‘systems’ framework means these variations will, by implication, affect connected and consequently broader systems at the Sydney metropolitan, state and national level. Therefore, it is hoped that these current challenges in regional research will not be ongoing.
This research has examined three regions on the suburban edge of Outer Western Sydney, and cannot therefore claim to be representative beyond this. The limitations of the cross-sectional analysis research method should also be noted namely, that the research investigated a snapshot of innovative activity in the regions at one point in time. Innovation is a dynamic process, as is knowledge gathering activities and learning behaviours; these will therefore change over time and so too will the manifestations and consequences of firm innovation in these regions.
The lack of time dimension also restricts the establishment of causality. Together with the specific selection of the three regions analysed, rather than random allocation, this means claims cannot be made regarding the causes of regional effects on innovation and knowledge gathering activities.
Chapter 8 – Conclusions and Policy Recommendations 235
8.3 Suggestions for future research
The limits surrounding establishment of causality in this research highlight a vital area for further research. Two subjects are immediately obvious for further regional innovation analysis. Firstly, the area of small businesses and their interaction and activities within regional innovation systems requires additional study. The current research has emphasized the innovative capacity of small firms. This capacity is also related to firms’ position in the firm life cycle, thereby also emphasising the importance of entrepreneurial activity. Whilst there is much research on small and medium sized enterprises in terms of their innovative activity, little research focuses on the knowledge gathering mechanisms and processes behind these innovations, and how these are affected by regional location.
The linkages between knowledge gathering activities and innovation are the second avenue for further research. The KISA analysis drew attention to linkages between certain types of knowledge (i.e. core KISA), which were accessed by most firms, and other types of knowledge (i.e. non-core KISA such as Standardisation KISA, HR KISA, and Industry development KISA), which were accessed only by certain firms. The interview analysis demonstrated the relationship between the types of knowledge that are internal, and those which are sourced externally as they relate to firm capability. Future KISA analysis could provide a deeper understanding of these linkages.
More broadly, further research could also explore the effects of differing systems of governance, particularly at the local level, in order to provide a comparative view of the role that this level of government plays in differing regional innovation systems.
Chapter 8 – Conclusions and Policy Recommendations 236
References
(AEGIS), A. E. G. i. i. S. (2003). 'Stocktake' of NSW as a potential 'Knowledge Hub'. Sydney, University of Western Sydney.
ABS (2003). Innovation in Australian Business. Canberra, ABS.
Acs, Z. (2002). Innovation and the Growth of Cities. Cheltenham U.K., Edward Elgar.
Acs, Z., L. Anselin, et al. (2002). "Patents and innovation counts as measures of regional production of new knowledge." Research Policy 31: 1069-1085.
Acs, Z. and D. Audretsch (1990). Innovation and Small Firms. Cambridge, MIT Press.
Albors, J., J. L. Hervas, et al. (2007). KISA role in traditional manufacturing industries: the case of ceramic tiles in Spain. The Functioning Knowledge Economy. M.-F. M.C. and I. Miles, Edward Alger Publishing.
Albors, J., J. L. Hervas, et al. (Forthcoming). KISA role in traditional manufacturing industries: the case of ceramic tiles in Spain. The Functioning Knowledge Economy, Edward Alger Publishing.
Andersson, M. and C. Karlsson (2004). Regional Innovation Systems in Small and Medium-Sized Regions: A Critical Review and Assessment. The Emerging Digital Economy: Entrepreneurship, Clusters and Policy. B. Johansson, C. Karlsson and R. R. Stough. Berlin, Springer-Verlag.
Andersson, M. and C. Karlsson (2004). The Role of Accessibility for the Performance of Regional Innovation Systems. Knowledge Spillovers and KNowledge Management. C. Karlsson, P. Flensburg and S. A. Horte. Cheltenham, Edward Elgar.
Armstrong, H. and J. Taylor (2000). Regional Economics and Policy. Malden, MA, Blackwell Publishers.
Asheim, B., T. (2000). Industrial Districts: the Contributions of Marshall and Beyond. Oxford Handbook of Economic Geography. G. L. Clark, Feldman, M.P., and Gertler, M.S. Oxford, Oxford University Press: 413-431.
Asheim, B., T. and L. Coenen (2005). "Knowledge bases and regional innovation systems: Comparing Nordic clusters." Research Policy 34: 1173-1190.
Asheim, B., T. and L. Coenen (2005). "Knowledge bases and regional innovation systems: Comparing Nordic clusters." Research Policy 34(October): 1173- 1190.
References 237
Asheim, B., T. and L. Coenen (2006). "Contextualising Regional Innovation Systems in a Globalising Learning Economy: On KNowledge Bases and Institutional Frameworks." Journal of Technology Transfer 31: 163-173.
Asheim, B., T., L. Coenen, et al. (2003). Nordic SMEs abd Regional Innovation Systems - Final Report. Lund, Department of Social and Economic Geography, Lund University, Sweden.
Asheim, B., T. and A. Isaksen (1996). Location, Agglomeration and Innovation: Towards regional innovations systems in Norway? STEP Group report. Oslo, Studies in Technology, Innovation and Economic Policy (STEP) Group.
Aslesen, H. W. (2004). Knowledge intensive service activities and innovation in the Norwegian aquaculture industry. Part project report from the OECD KISA study Oslo, Norway, STEP- Centre for Innovation Research.
Audretsch, D. and E. E. Lehmann (2005). "Does the Knowledge Spillover Theory of Entrepreneurship hold for regions?" Research Policy 34: 1191-1202.
Barbour, E. and A. Markusen (2004). Regional Occupational and Industrial Structure: Does One Imply the Other? Working Paper Series, Project on Regional and Industrial Economics, the Humphrey Institute of Public Affairs, University of Minnesota.
Beer, A. and B. Kearins (2004). Regional Dimensions in National Development: Understanding the Role of Regionally Focussed Economic Development in Achieving National Growth. Inaugural National Regional Research Colloquium.
Beer, A., A. Maude, et al. (2003). Developing Regional Australia. Kensington, UNSW Press.
Best, M., A. Paquin, et al. (2004). "Discovering Regional Competeitive Advantage: Massachusetts High-Tech." BUsiness and Economic Hostory On-line 2: 1- 30.
Birch, D. (1979). The Job Generation Process. Cambridge, Mass, Joint Centre for Neighbourhood and Regional Development.
Blakely, E., S. Bista, K., et al. (2006). Drivers of Economic Growth in the Greater Sydney Metropolitan Region. Canberra, Australian Government Department of Transport and Regional Services, Bureau of Transport and Regional Economics.
Blumer, H. (1956). Symbolic Interactionism: Perspective and Method. Berkerley, University of California Press.
Boschma, R. A. (2005). "Proximity and Innovation: A Critical Assessment." Regional Studies 39(1): 61-74.
References 238
Boschma, R. A. and A. L. J. Ter Wal (2007). "Knowledge Networks and Innovative Performance in an Industrial District: The Case of a Footwear District in the South of Italy." Industry and Innovation 14(2): 177-199.
Braczyk, H.-J., P. Cooke, et al., Eds. (2004). Regional innovation systems : the role of governance in a globalized world. New York, Routledge.
Brenner, T. (2007). "Local Knowledge Resources and Knowledge Flows." Industry and Innovation 14(2): 121-128.
Breschi, S. (2000). "The Geography of Innovation: A Cross-sector Analysis." Regional Studies 34(3): 213-229.
Breschi, S. and F. Lissoni (2001). Knowledge Spillovers and Local Innovation Systems: A Critical Survey. Linc Papers, Serie Economia e Impresa. Milan, CESPRI.
Breschi, S., F. Malerba, et al. (2000). "Technological Regimes and Schumpeterian Patterns if Innovation." The Economic Journal 110: 388-410.
Brokel, T. (2007). "The Regional Dimesnion of Knowledge Transfers - A Behavioural Approach." Industry and Innovation 14(2): 151-175.
Camagni, R. P. (1995). "The concept of innovative milieu and its relevance for public policies in European lagging regions." Papers in Regional Science 74: 317-340.
Camagni, R. P. (2003). Regional Clusters, Regional Competencies and Regional Competition. International Conference on "Cluster management in structural policy - International experiences and consequences for Northrhine-Westfalia", Duisburg, Germany.
Capello, R. (1999). "Spatial Transfer of knowledge in High Technology Milieux: Learning Versus Collective Learning Processes." Regional Studies 33(4): 353-365.
Capello, R. (2001). Spatial and Sectoral Characteristics of Relational Capital in Innovative Activity. ESRC CITIES Competitiveness and Cohesions Programme International Research Workshop on " Innovation and Competitive Cities in the Global Economy", Worcester College, Oxford University.
Capello, R. and A. Faggian (2005). "Collective learning and Relational Capital in Local Innovation Processes." Regional Studies 39(1): 75-87.
Carlsson, B., S. Jacobsson, et al. (2002). "Innovation systems: analytical and methodological issues." Research Policy 31: 233-245.
Carlsson, B. and R. Stankiewicz (1991). " On the nature, function and composition of technological systems." Journal of Evolutionary Economics 1(2): 90-118.
References 239
Carter, A. P. (1994). Measuring the Performance of a Knowledge-based Economy. Working Paper No. 337, Departments of Economics, Brandeis University.
Cattell, R. B. (1966). Handbook of Multivariate Experimental Psychology. Chicago, Rand McNally.
Clark, G. L., M. P. Feldman, et al. (2000). The Oxford Handbook of Economic Geography. Oxford, Oxford University Press.
Coenen, L., J. Moodysson, et al. (2006). "Comparing a Pharmaceutical and an Agro-food Bioregion: One the Importance of KNowledge Bases for Socio- Spatial Patterns of Innovation." Industry and Innovation 13(4): 393-414.
Cole, M. (2004). Neo-Reichian Occupational Breakdown. University of Sydney, Australia.
Cooke, P. (1999). "The role of innovation in regional competitiveness." Australasian Journal of Regional Studies 5(1): 3-22.
Cooke, P. (2001). "Regional Innovation Systems, Clusters and the Knowledge Economy." Industrial and Corporate Change 10(4): 945-974.
Cooke, P. (2002). "Regional Innovation Systems: General Findings and Some New Evidence from Biotechnology Clusters." Journal of Technology Transfer 27: 133-145.
Cooke, P. (2004). Introduction: Regional Innovation Systems - an evolutionary approach. Regional Innovation Systems: The role of governance in a globalized world. P. Cooke, M. Heidenreich and H.-J. Braczyk. London, Routledge: 1-18.
Cooke, P. (2004). The regional innovation system of Wales: evolution or eclipse? Regional Innovation Systems: The role of governance in a globalized world. H.-J. Braczyk, P. Cooke and M. Heidenreich. London, Routledge.
Cooke, P., M. Heidenreich, et al., Eds. (2004). Regional Innovation Systems: The role of governance in a globalized world. London, Routledge.
Cooke, P., D. Kaufmann, et al. (2006). "The Biosciences KNowledge Value Chain and Comparative Incubation Models." Journal of Technology Transfer 31: 115-129.
Cooke, P. and L. Leydesdorff (2006). "Regional Development in the Knowledge- Based Economy: The Construction of Advantage." Journal of Technology Transfer 31: 5-15.
Cooke, P. and O. Memedovic (2003). Strategies for Regional Innovation Systems: Learning Transfer and Applications. Policy papers. Vienna, United Nations Industrial Development Organisation: 1-25.
References 240
Cooke, P. and K. Morgan (1998). The Associational Economy: Firms, Regions and Innovation. Oxford, Oxford University Press.
Cumbers, A. and D. MacKinnon (2004). "Introduction: Clusters in Urban and Regional Development." Urban Studies 41(5/6): 959-969.
Dahlander, L. and M. W. Wallin (2006). "A man on the insude: Unlocking communities as complementary assets." Research Policy 35: 1243-1259. de Jong, J. P. J. and O. Marsili (2006). "The fruit flies of innovations: A taxonomy of innovative small firms." Research Policy 35: 213-229. de La Mothe, J. a. P., G. (1998). Local and Regional Systems of Innovation. US, Kluwer Academic Publishers.
De Laurentis, C. (2006). "Regional Innovation Systems and the Labour Market: A Comparison of Five Regions." European Planning Studies 14(8): 1059- 1084.
De Vaus, D. (2001). Research Design in Social Research. London, Sage.
Dei Ottati, G. (2004). The remarkable resiliance of the industrial districts of Tuscany. Regional Innovation Systems: The role of governance in a globalized world. P. Cooke, M. Heidenreich and H.-J. Braczyk. London, Routledge: 21-43.
Den Hertog, P. (2000). "Knowledge-intensive Business Services as co-producers of innovation in services." International Journal of Innovation Management 4(4): 491-528.
Doloreux, D. (2004). "Regional Innovation Systems in Canada: A Comparative Study." Regional Studies 38(5): 481-494.
Doloreux, D. and S. Parto (2004). Regional Innovation Systems: A Critical Synthesis. INTECH (Institute for New Technologies) Discussion Paper Series. Maastricht, The Netherlands, United Nations University.
Doloreux, D. and S. Parto (2005). "Regional Innovation Systems: Current discourse and unresolved issues." Technology in Society 27: 133-153.
Drucker, P. (1995). The future of the Industrial Man. New York, Transaction Publishers.
Edquist, C., Ed. (1997). Systems of Innovation. London, Pinter.
Edquist, C., L. Hommen, et al. (2001). Innovation and Employment: Process versus Product Innovation. Massachusetts, Edward Elgar.
References 241
Fabringer, L. R., D. T. Wegener, et al. (1999). "Evaluating the use of exploratory factor anlaysis in psychological research." Psychological Research 4: 272- 299.
Fagan, B. and R. Dowling (2005). "Neoliberalism and Suburban Employment: Western Sydney in the 1990s." Geographical Research 43(1): 71-81.
Fagan, B., R. Dowling, et al. (2004). Suburbs in the Global City. Employment Profile of Greater Western Sydney. Sydney, WSROC.
Feldman, M. a. (1999). "The new economies of innovation spillovers and agglomeration: A review of empirical studies." Economics, Innovation and New Technology 8: 5-26.
Feser, E. and E. Bergman (2000). "National Industry Cluster templates: A framework for applied regional cluster analysis." Regional Studies 34(1): 1- 19.
Feser, E. J. (2003). "What regions do rather than make: A proposed set of knowledge-based occupational clusters." Urban Studies 40(10): 1937-1958.
Fishman, R. (1987). Bourgeois Utopias New York, Basic Books.
Florida, R. (2002). "The Economic Geography of Talent." Annals of the Association of American Geographers 92: 743-755.
Florida, R. (2005). Cities and the Creative Class. New York, Routledge.
Fontana, R., A. Geuna, et al. (2006). "Factors affecting university-industry R&D projects: The importance of searching, screening and signalling." Research Policy 35: 309-323.
Freel, M. (2006). "Patterns of Technological Innovation in Knowledge-Intensive Business services." Industry and Innovation 13(3): 335-358.
Freeman, C. (1988). Japan: A new national innovation system? Technology and Economic Theory. G. Dosi, C. Freeman, R. Nelson, G. Silverberg and L. Soete. London, Pinter.
Freeman, C. (1995). "The "National Systems of Innovation" in historic perspective." Cambridge Journal of Economics 19(1): 5-24.
Gertler, M. and D. A. Wolfe (2004). Ontario's regional innovation system: the evolution of knowledge-based institutional assets. Regional Innovation Systems: The role of governance in a globalized world. H.-J. Braczyk, P. Cooke and M. Heidenreich. London, Routledge.
Glaeser, E. (2000). The New Economies of Urban and Regional Growth. The Oxford Handbook of Economic Geography. G. L. Clark, Feldman, M.P., and Gertler, M.S. Oxford, Oxford University Press: 83-98.
References 242
Gleeson, B. (2006). Australian Heartlands: making space for hope in the suburbs. Crows Nest, Allen and Unwin.
Godin, B. (2006). "The Knowledge-Based Economy: Conceptual Framework or Buzzword?" Journal of Technology Transfer 31: 17-30.
Gorsuch, R. L. (1983). Factor Analysis. Hillsdale, L. Erlbaum Associates.
Government, N. (1951). County of Cumberland Plan. Sydney, NSW Government.
Government, N. (2005). Sydney Metropolitan Strategy. Sydney, NSW Government Department of Planning and Natural Resources.
Gregersen, B. and B. Johnson (1997). How do Innovations affect Economic Growth? Some Different Approaches in Economics. Innovation Systems and European Integration (ISE). Aalborg, Denmark, IKE-Group, Department of Business Studies, Aalborg University.
Groves, R. M. (2002). Survey nonresponse. New York, Wiley.
Groves, R. M. (2004). Survey Methodology. New Jersey, J. Wiley.
Hales, M. (2001). Birds were dinosaurs once - the diversity and evolution of research and technology organisations. RISE Final Report, Centrim.
Hommen, L. and D. Doloreux (2003). Is the Regional Innovation System at the End of Its Life Cycle? Innovation in Europe: Dynamics, Institutions and Values, Roskilde University, Denmark.
Howells, J. (2002). "Tacit knowledge, Innovation and Economic Geography." Urban Studies 39(5-6): 871-884.
Howells, J. (2005). "Innovation and regional economic development: A matter of perspective?" Research Policy 34: 1220-1234.
Hussler, C. and P. Ronde (2007). "The imapct of cognitive communities on the diffusion of academic knowledge: Evidence from the neworks of inventors of a French University." Research Policy 36: 288-302.
IBM (2006). Global COE Innovation Survey Report. Melbourne, IBM.
Isaksen, A. (2001). "Building Regional Innovation Systems: Is Endogenous Industrial Development Possible in the Global Economy?" Canadian Journal of Regional Science 24(1): 101-120.
Jensen, P. A. and E. Webster (2004). Examining Biases in Measuring Firm Innovation. Melbourne Institute Woring Paper Series, Melbourne Institute of Applied Economic and Social Research, The University of Melbourne.
References 243
Kerlinger, F. (1979). Behavioural Research: a conceptual approach. New York, Holt, Rinehart and Winston.
Koo, J. (2005). "How to Analyze the Regional Economy with Occupational Data." Economic Development Quarterly 19: 356-372.
Koschatzky, K. (1999). "Innovation networks of Industry and Business-Related Services - Relations Between Innovation Intensity of Firms and Regional Inter-Firm Cooperation." European Planning Studies 7(6): 737-757.
Krugman, P. (2000). Where in the World is the 'New Economic Geography'. The Oxford Handbook of Economic Geography. G. L. Clark, Feldman, M.P., and Gertler, M.S. Oxford, Oxford University Press: 49-60.
Lagendijk, A. (2004). Regions and regional boundaries in the minds and practices of policy makers in a unifying Europe. Changing Economic Behaviour in a Unifying Europe. G. Van Vilsteren and E. Wever. Assen, Van Gorcum.
Lam, A. (2002). "Alternative societal models of learning and innovation in the knowledge economy." International Social Science Journal 54(171): 67-82.
Lawson, C. and E. Lorenz (1999). "Collective Learning, Tacit Knowledge and Regional Innovative Capacity." Regional Studies 33(4): 305-317.
Lever (2002). "Correlating the knowledge-base of cities with economic growth." Urban Studies 39(5/6): 859-70.
Leydesdorff, L. (2006). ""While a Storm is Ranging on the Open Sea": Regional Development in a Knowledge-based Economy." Journal of Technology Transfer 31: 189-203.
Leydesdorff, L., W. Dolfsma, et al. (2006). "Measuring the knowledge base of an economy in terms of triple-helix relations among 'technology, organisation, and territory'." Research Policy 35: 181-199.
Leydesdorff, L., P. Cooke and M. Olazaran (2002). "Technology Transfer in European Regions: Introduction to the Special Edition." Journal of Technology Transfer 27: 5-13.
Lillis, A. M. and J. Mundy (2005). "Cross-sectional Feld Studies in Management Accounting Research - Closing the Gap between Surveys and Case Studies." Journal of Management Accounting Research 17: 119-141.
Lissoni, F. (2001). "Knowledge codification and the geography of innovation: the case of Brescia mechanical cluster." Research Policy 30: 1479-1500.
Lorenzen, M. (2005). "Knowledge and Geography." Industry and Innovation 12(4): 399-407.
Lundvall, B.-A., Ed. (1992). National Systems of Innovation. London, Pinter.
References 244
Lundvall, B.-A. and B. Johnson (1994). "The Learning Economy." Industry and Innovation 1(2): 23-42.
Lundvall, B.-A., B. Johnson, et al. (2002). "National systems of production, innovation and competence building." Research Policy 31: 213-231.
Macpherson, A. and R. Holt (2007). "Knowledge, learning and small firm growth: A systematic review of the evidence." Research Policy 36: 172-192.
Malerba, F. (2005). "Sectoral Systems of Innovation: A framework for linking innovation to the knowledge base, structure and dynamics of sectors." Economics, Innovation and New Technology 14(1-2): 63-82.
Malmberg, A. and P. Maskell (2002). "The elusive concept of localization economies: towards a knowledge-based theory of spatial clustering." Environment and Planning 34: 429-449.
Markusen, A. (2003). "Fuzzy Concepts, Scanty Evidence, Policy Distance: The case for rigour and policy relevance in Critical regional Studies." Regional Studies 37(6&7): 701-717.
Markusen, A. (2004). "Targeting Occupations in Regional and Community Economic Development." Journal of the American Planning Association 70: 253-268.
Markusen, A. and G. Schrock (2001). Occupational Advantage: Detecting and Enhancing Occupational Mix in Regional Development. Project on Regional and Industrial Economics Humphrey Institute of Public Affairs.
Marsh, C. (1982). The Survey Method: The contribution of surveys to sociological explanation. London, Allen and Unwin.
Marsh, C. (1988). Exploring data: an introduction to data anlaysis for social scientists. Cambridge UK, Polity Press in association with Blackwell.
Marsili, O. and A. Salter (2005). "'Inequality' of Innovation: Skewed distributions and the returns to innovation in Dutch manufacturing." Economics, Innovation and New Technology 14(1-2): 83-102.
Martinez-Fernandez, M. C. (2006). "Introduction to Special Issue: Knowledge Intensive Service Activities (KISA) in Software Innovation." International Journal Services Technology and Management. 7(2): 109-174.
Martinez-Fernandez, M. C. and M. Bjorkli (2003). The Role of Knowledge in Large Australian City-Regions: A Traditional Industry in Greater Western Sydney and the Hunter Region. State of Australian Cities, Parramatta, Sydney.
References 245
Martinez-Fernandez, M. C. and V. V. Krishna (2006). "KISA in innovation of Australian software firms." International Journal Services Technology and Management. 7(2): 126-136.
Martinez-Fernandez, M. C. and I. Miles (2006). "Inside the Software Firm: Co- production of Knowledge and KISA in the Innovation Process." International Journal Services Technology and Management. 7(2): 115-125.
Martinez-Fernandez, M. C., T. Potts, et al. (2005). Innovation at the Edges: The Role of Innovation Drivers in South West Sydney. Sydney, University of Western Sydney.
Martinez-Fernandez, M. C., C. Soosay, et al. (2005). Knowledge Intensive Service Activities (KISA) in the Australian Software Industry. Sydney, University of Western Sydney.
Maskell, P. (2004). Learning in the Village economy of Denmark: the role of institutions and policy i sustaining competitiveness. Regional Innovation Systems: The role of governance in a globalized world. H.-J. Braczyk, P. Cooke and M. Heidenreich. London, Routledge.
Maskell, P., H. Bathelt, et al. (2006). "Building Global Knowledge Pipelines: The Role of Temporary Clusters." European Planning Studies 14(8): 997-1013.
Maskell, P., H. Eskelinen, I. Hannibalson, A. Malmberg and E. Vatne (1998). Competitiveness, Localised Learning and Regional Development. London, Routledge.
Meyer, B. (2005). Sydney's Growth Strategies. Suburban Sydney revisted, Parramatta, Univerity of Western Sydney.
Miles, I. (2003). Knowledge Intensive Services: suppliers and clients. Report to the Ministry of Trade and Industry Finland.
Miles, I., N. Kastrinos, et al. (1994). Knowledge-Intensive Business Services: Their Role as Users, Carriers and Sources of Innovation. Manchester, PREST.
Miller, W. F. (2000). "The "Habitat" for Entrepreneurship."
Moulaert, F. and F. Sekia (2003). "Territorial Innovation Models: A Critical Survey." Regional Studies 37(3): 289-302.
Muller, E. and A. Zenker (2001). "Business services as actors of knowledge transformation: the role of KIBS in regional and national innovation systems." Research Policy 30: 1501-1516.
Myrdal, G. (1957). Economic Theory and Under-Developed Regions. London, Gerald Duckworth & Co.
References 246
Nelson, R. and S. Winter (1982). An Evolutionary Theory of Economic Change. Cambridge, Belknap Press/ Harvard University Press.
Nonaka, I. and H. Takeuchi (1995). The Knowledge Creating Company. New York, Oxford University Press.
OECD (1996a). Employment and Growth in the Knowledge Based Economy. Paris, OECD.
OECD (1997). National Systems of Innovation. Paris, OECD.
OECD (1999b). Managing National Systems of Innovation. Paris, OECD.
OECD (2001). Benchmarking Knowledge-based Economies. Pais, OECD.
OECD (2001a). Benchmarking Knowledge-based Economies. Pais, OECD.
OECD (2001b). Innovative Clusters: Drivers of National Innovation Systems. Paris, OECD.
OECD (2002). Redefining Territories: The Functional Regions. Paris, OECD.
OECD (2005). The Measurement of Scientific and Technological Activities: Guidelines for Collecting and Intepreting Innovation Data, The Oslo Manual. Paris, OECD.
OECD (2006). The role of Knowledge Intensive Service Activities (KISA) in Innovation. Paris, OECD.
Palen, J. (1995). Suburbs. New York, McGraw-Hill.
Perroux, F. (1969). L'Économie du XXe siècle. Paris, Presses Universitaires de France.
Piore, M. and C. Sabel (1984). The Second Industrial Divide. New York, Basic Books.
Polanyi, M. (1962). Personal Knowledge: Towards a Post Critical Philosophy. New York, Harper Tourchbooks.
Polanyi, M. (1966). The Tacit Dimension. London, Routeledge.
Porter, M. (1990). The Competitive Advantage of Nations. New York, The Free Press.
Porter, M. (1998). Clusters and Competition: new agendas for companies, governments, and institutions. On Competition. M. Porter. Boston, Harvard Business School Press: 197-287.
References 247
Porter, M. E. (2001). Regions and the New Economics of Competition. Global City Regions. A. Scott. New York, Oxford University Press: 139-157.
Rantisi, N., M. (2002). "The Local Innovation Systems as a Source of 'Variety': Openness and Adaptability in New York City's Garment District." Regional Studies 36(6): 587-602.
Reich, R. (1991). The Work of Nations. New York, Alfred A. Knoff.
Romer, P. (1995). "Beyond the Knowledge Worker." Worldlink(January/February).
Ronde, P. and C. Hussler (2005). "Innovation in regions: What does really matter?" Research Policy 34: 1150-1172.
Rothblatt, D. and D. Garr (1986). Suburbia: An International Assessment. Worchester, Croom Helm.
Sainsbury (2003). Address to European Parliament by UK National Science Minister.
Saxenian, A. (1994). Regional Advantage: Culture and Competition in Silicon Valley and Route 128. Cambridge, MA., Harvard University Press.
Schumpeter, J. (1937). The Theory of Economic Growth. Cambridge, Mass., Harvard University Press.
Scott, A. and M. Storper (2003). "Regions, Globalization, Development." Regional Studies 37(6&7): 579-593.
Searle, G. and B. Pritchard (2005). "Industry clusters and Sydney’s ITT sector: Northern Sydney as ‘Australia’s Silicon Valley’?" Australian Geographer 36(2): 145-169.
Sharpe, S. (2005). From Dormitories to Downtown: suburban change in Sydney. ANZRSAI Annual Conference, Manuaka New Zealand.
Sharpe, S., C. Martinez-Fernandez, et al. (2004). Reichian Analysis of Occupation in Sydney Regions. Australian, New Zealand Regional Science Association (ANZRSAI), Wollongong.
Sharpe, S. and C. M. Martinez-Fernandez (2006). Measuring Regional Knowledge Resources: What do Occupations have to offer? DRUID Summer Conference, Copenhagen , Denmark
Sharpe, S. and M. C. Martinez-Fernandez (Forthcoming). "Measuring regional knowledge resources: what do knowledge occupations have to offer?" Innovation: Management, Policy and Practice.
References 248
Simmie, J. (2002). "Knowledge spillovers and Reasons for the concentration of Innovative SMEs." Urban Studies 39(5-6): 885-902.
Simmie, J. (2003). "Innovation and Urban Regions as National and International Nodes for transfer and sharing of knowledge." Regional Studies 37(6 and 7): 607-620.
Simmie, J. (2004). "Innovation and Clustering in the Globalised International Economy." Urban Studies 41(5/6): 1095-1112.
Simmie, J. (2005). "Innovation and Space: A Critical review of the Literature." Regional Studies 39(6): 789-804.
Simmie, J. and J. Sennett (2001). London: International Trading Metropolis. Innovative Cities. J. Simmie. London, Taylor and Francis.
Smith, K. (1996). Systems Approaches to Innovation: Some Policy Issues. Innovation Systems and European Integration (ISE). Oslo, Norway, STEP Group.
Sternberg, R. (2000). "Innovation Networks and Regional Development - Evidence from the European Regional Innovation Survey (ERIS): Theoretical Concepts, Methodlogical Approach, Empirical Basis and Introduction to the Theme Issue." European Planning Studies 8(4): 389-407.
Storey, D., Ed. (1985). Small Firms and regional Economic Development. Cambridge, Cambridge University Press.
Storper, M. (1995). "The Resurgance of Regional Economies, ten years later: the region as a nexus of untraded interdependencies." European Urban and Regional Studies 2(3): 191-221.
Storper, M. (1997). The Regional World: Territorial development in a global economy. New York, Guilford Press.
Thomi, W. and T. Bohn (2003). Knowledge Intensive Business Services in Regional Systems of Innovation - Initial Results from the Case of Southeast Finland. 43rd Annual Congress of the Regional Science Association, Jyvaskyla, Finalnd.
Vale, M. and J. Caldeira (2007). "Proximity and Knowledge Governance in Localized Production Systems: The Footwear INdustry in the North Region of Portugal." European Planning Studies 15(4): 531-548. van Stel, A. J. (2004). "Knowledge Spillovers and Economic Growth: An Analysis Using Data of Dutch Regions in the Period 1987-1995." Regional Studies 38(4): 393-407.
Vilanova, M. R. and L. Leydesdorff (2001). "Why Catalonia cannot be considered as a Regional Innovation System." Scientometrics.
References 249
Vilanova, M. R. and L. Leydesdorff (forthcoming). "Why Catalonia cannot be considered as a Regional Innovation System." Scientometrics.
Watts, H. D., A. M. Wood, et al. (2006). "Owner-managers, clusters and local embeddedness: small firms in the Sheffield (UK) metal-working cluster." Entrepreneurship and Regional Development 18(May): 185-205.
Wiig, H. and M. Wood (1995). What comprises a Regional Innovation System? An Empirical Study. STEP report. Oslo, STEP Group: 26.
Yin, R. (2003). Case Study Research Designs and Methods. London, Sage.
References 250
Appendices
Appendices 251
Appendix 1 Survey Information sheet for companies
INVITATION FOR COMPANIES TO PARTICIPATE IN NEW RESEARCH
We are seeking your involvement in the ‘Ways to Grow: Auditing and Encouraging Industrial Innovative Capacity and Employment Growth’ project. This project is funded by the Australian Research Council through a Linkage Grant (LP0347917) and will investigate innovative activity among firms in the outer metropolitan regions of Western Sydney – including the local government areas of Liverpool, Campbelltown, Camden, Wollondilly, Blacktown, Penrith, Hawkesbury and Baulkham Hills.
We all know that being innovative is critical to being successful in business. Innovation is defined as any new or significantly improved product or service, or any new or significantly improved organisational, managerial or operational process that creates economic value. Innovation is a complex and involved process that runs to the heart of how businesses operate. Therefore analysing how innovation is also complex – how do businesses innovate? Where do they get their ideas from? What are the main barriers and obstacles to innovating? What services and other businesses do firms rely on when they are innovating?
The specific answers to these questions will be individual to each business, but increasingly we are seeing emerging trends in innovative activity across industries, countries and regions. These trends relate both to stories of economic success and decline. Innovative activity is critically linked with not only the success of the individual businesses involved but also the wider industry, region and locality to which they belong. The aim of the Outer Western Sydney Innovation survey will be to collect previously unavailable information on innovative activity at the local level and then use this information to benchmark and compare outer Western Sydney in terms of innovative activity and economic development with metropolitan Sydney, Australia, internationally and across industry sectors.
WHY IS THIS IMPORTANT TO OUTER WESTERN SYDNEY?
Outer Western Sydney will grow significantly in population over the coming decades. The pressure to ensure this population growth is matched by sustainable employment growth will be both immediate and constant throughout this period. Innovating companies are growing companies. Understanding how innovation works in these companies in outer Western Sydney is the first step in a process of recognising and encouraging innovation. This survey will inform businesses, government and research and educational institutions on the innovation potential of outer Western Sydney and the
Appendices 252
contributions that can be made in terms of industry development, policy, infrastructure and university-industry partnerships, among other things.
DETAILS OF THE SURVEY
The survey contains twenty-one questions and will take approximately fifteen minutes to complete. The survey can be completed online at http://aegis.uws.edu.au/linkageliverpool/ main.html or a PDF version can also be downloaded from this page, printed and the survey returned by fax or mail.
DETAILS OF THE PROJECT
The research will take place over four stages. This is the first stage – a firm innovation survey; this phase of the research will be broad in the number and type of participants, with the aim of discovering the general levels of industrial innovative activity in outer Western Sydney. The second phase will be a series of case studies of targeted firms. The purpose of the case studies will be to develop an in-depth understanding of the processes and mechanisms that firms undertake when innovating and the role that regional factors play. The third section of the research will be semi-structured interviews with key policymakers and educational institutions in the regions, to establish key current policies and governance frameworks. The final stage will be the analysis of the data and the proposal of recommendations for policy development.
The outcomes of the project will be the provision of previously unavailable data and analysis on the innovative activity of firms in outer Western Sydney, including sources of innovation, methods of acquiring new knowledge and drivers of innovative activity, followed by the transformation of this data into regional socio-economic development policies appropriate to the region.
PRIVACY AND CONFIDENTIALITY
The survey is being carried out by researchers from the University of Western Sydney research centre AEGIS (Australian Expert Group in Industry Studies). AEGIS specialises in the analysis of industry innovation and on providing policy advice to public agencies. It is an independent research centre, with no commercial ties to companies in this, or other fields. The research team includes Chief investigator Dr Cristina Martinez-Fernandez and Australian Postgraduate Award (Industry) (APAI) Doctoral Candidate Samantha Sharpe – brief biographies are shown on the next page. Analysis of data collected will provide information for the ARC project, specific reports and the APAI PhD dissertation. Each member of the research team is committed to the principles that guide reputable research. All research will be undertaken with care and respect for the respondent’s input and with a commitment to the quality and validity of the research report/results.
All information on individual firms will be treated in strictest confidence, and no individuals or individual firms will be identified in the research results. We will ask you for your contact details if you would like to receive a copy of the report. You are free to withdraw at any time without repercussions and you may request that your contributions be removed from the project.
NOTE: This study has been approved by the University of Western Sydney Human Research Ethics Committee. The Approval Number is HREC 05/108 If
Appendices 253
you have any complaints or reservations about the ethical conduct of this research, you may contact the Ethics Committee through the Research Ethics Officers (tel: 02 4736 0883). Any issues you raise will be treated in confidence and investigated fully, and you will be informed of the outcome.
RESEARCH TEAM
Cristina Martinez, BA Psy (Hon), Grad. Dip. (Ind. Psy), M Psy, Doc. (Psy), PhD (Planning & Urban Development UNSW) is a Senior Research Fellow at AEGIS. Cristina’s research concentrates on innovation networks, urban and regional innovation systems, innovation policy and the spatial analysis of innovation. Cristina has recently finished a research project titled “Innovation at the Edges” in the Macarthur region.
Samantha Sharpe, BA (Communications) (UWS), M Public Policy (USyd) is an Australian Postgraduate Award (Industry) Doctoral candidate. She has five years experience in local government in policy and economic development roles.
For further information on the project please contact: Dr Cristina Martinez, Ph: (02) 8255 6217 [email protected] or Ms Samantha Sharpe , Ph: (02) 8255 6238 [email protected]
THANK YOU FOR YOUR PARTICIPATION
Appendices 254
Appendix 2 Survey Questionnaire
WAYS TO GROW - AUDITING AND ENCOURAGING INDUSTRIAL INNOVATION AND EMPLOYMENT GROWTH SURVEY QUESTIONNAIRE FOR BUSINESSES
Contact persons for this document: Cristina Martinez, [email protected] Samantha Sharpe, [email protected]
Project website: http://aegis.uws.edu.au/linkageliverpool/main.html
AEGIS is a Research Centre of the University of Western Sydney. Level 8, 263 Clarence Street Sydney NSW 2000 PO Box Q1287 QVB Post Office NSW 1230 Phone: (02) 8255 6200 Fax: (02) 8255 6222 Web: aegis.uws.edu.au
Appendices 255 Questionnaire protocol for Innovation survey In Outer Western Sydney 34
This survey is part of a study examining innovative activity in businesses, throughout three regions in outer Western Sydney – South West, Central West and North West Sydney. The survey consists of 21 questions and will take approximately 20 minutes to complete. The survey is divided into four sections - the first section asks for some details about your business. Section 2 asks about any new products, services or processes your company has introduced over the last three years. Section 3 asks questions about how your company gains new knowledge and the use of knowledge intensive service activities, while section 4 refers to your business collaborative activities.
All information is kept confidential
SECTION 1 – ABOUT YOUR BUSINESS
1. In which local government area is your business located?
Baulkham Hills ̌ Blacktown ̌ Camden ̌ Campbelltown ̌ Hawkesbury ̌ Liverpool ̌ Penrith ̌ Wollondilly ̌ Other ̌
Please indicate, ______
2. What industry does your business mainly operate in?
Agriculture, Forestry, Fishing ̌ Mining ̌ Manufacturing ̌ Construction ̌ Electricity, Gas & Water Supply ̌ Wholesale Trade ̌ Retail Trade ̌ Transport & Storage ̌ Accommodation, Cafes, Restaurants ̌ Communication Services ̌ Finance & Insurance ̌ Education ̌ Property & Business Services ̌ Gov’t Admin & Defence ̌ Health & Community Services ̌ Personal & other Services ̌ Cultural & Recreation Services ̌ Other ̌
3. Which of the following statements best describes your business?
34 This survey is based on the ABS Innovation in Australian Business Survey (2003) and AEGIS Knowledge Intensive Service Activities (KISA) Innovation related surveys.
You may check two boxes if the last statement best describes your business
Starting Business (less than 1 year operation) ̌ New Business (1-4 years of operation) ̌ Established Business (5-9 years of operation) ̌ Mature Business (9 years or more of operation) ̌ Business in expansion (complex and changing structure) ̌
4. How many employees (Full time equivalent) does your business have?
4a) How many of your employees are employed
on a full time basis on a contract basis on a casual basis as apprentices & trainees
4b) How many employees have you recruited in the last financial year (2004- 2005)?
How many of these new employees were employed
on a full time basis on a contract basis on a casual basis as apprentices/trainees
4c) Was this recruitment related to Staff replacement? ̌ Growth/ expansion of the business? ̌ Other, please specify?
5. How many of your staff have the following as their highest educational qualification? If you are self employed/ sole trader, please answer for yourself
Secondary education Certificate or Diploma level Bachelors degree Postgraduate Degree
6. What your annual turnover/sales for last financial year (2004-2005)?
Appendices 257
Less than $50,000 ̌ Between $50,001 and $250,000 ̌ Between $250,001 and $500,000 ̌ Between $500,001 and $1 million ̌ Between $1,000,001 – $2 million ̌ More than $2 million ̌
7. How long has your business been in its current location?
Less than 1 year ̌ 1-4 years ̌ 5-9 years ̌ Longer than 9 years ̌
8. When your company established or re-located to its current location, what were the main deciding factors in selecting this location, and how would you rate these in terms of their importance?
Not Not High Medium Low important relevant
Land availability ̌ ̌ ̌ ̌ ̌ Infrastructure ̌ ̌ ̌ ̌ ̌ Land process, rent ̌ ̌ ̌ ̌ ̌ Close to customers ̌ ̌ ̌ ̌ ̌ Close to suppliers ̌ ̌ ̌ ̌ ̌ Encroaching urban development ̌ ̌ ̌ ̌ ̌ Suitable skilled workforce ̌ ̌ ̌ ̌ ̌ Close to where mgmt/ GM lives ̌ ̌ ̌ ̌ ̌ Close to where staff live ̌ ̌ ̌ ̌ ̌ Favourable government policies ̌ ̌ ̌ ̌ ̌ Government subsidy or inducement ̌ ̌ ̌ ̌ ̌ Other, please give details ̌ ̌ ̌ ̌ ̌
8a) If your company re-located, where was it located before?
Appendices 258
8b) Were any other areas considered in your company’s location? No ̌ Yes ̌ Please list
9. How would you rank the importance of the following areas in terms of the location of your company’s current customers?
Not No
Location of customers High Medium Low relevant customers
Within the local region (20km radius) ̌ ̌ ̌ ̌ ̌ Within the Sydney metro region ̌ ̌ ̌ ̌ ̌ Elsewhere in NSW ̌ ̌ ̌ ̌ ̌ Elsewhere in Australia ̌ ̌ ̌ ̌ ̌ Overseas ̌ ̌ ̌ ̌ ̌
How would you rate the following areas in terms of the location of your company’s current suppliers?
Not No High Medium Low relevant suppliers
Within the local region (20km radius) ̌ ̌ ̌ ̌ ̌ Within the Sydney metro region ̌ ̌ ̌ ̌ ̌ Elsewhere in NSW ̌ ̌ ̌ ̌ ̌ Elsewhere in Australia ̌ ̌ ̌ ̌ ̌ Overseas ̌ ̌ ̌ ̌ ̌
Appendices 259
SECTION 2 - YOUR FIRM’S INNOVATIVE ACTIVITY
An innovation is defined as any new or significantly improved product or service that has economic value to your business. Innovation also includes any new or significantly improved operational process (such as new production methods) and new or significantly improved managerial or organisational processes (such as new accounting or human resources systems) as long as they have economic value.
10. Has your business made any changes in the last three years in terms of introducing a new product, service or process, if yes, how would you describe these changes – incremental (gradual change) or radical (one big change)?
Incremental Radical No New Innovations change change innovation A new or substantially improved product or service? ̌ ̌ ̌ A new way of producing an existing product or service? ̌ ̌ ̌ (new operational process) A new or substantially improved managerial or ̌ ̌ ̌ organisational process
11. Did your company encounter any of the following barriers when introducing new goods or services or developing and implementing new processes? Could you please rate then on a scale of 1 to 3?
1=Critical barrier (prevented innovation) 2=Significant barrier (considerable barrier but not insurmountable) 3=Minor barrier (barrier but able to be overcome with relatively minor costs)
Not a Critical Significant Minor Barriers barrier Excessive costs ̌ ̌ ̌ ̌ Cost or availability of finance ̌ ̌ ̌ ̌ Potential market already dominated ̌ ̌ ̌ ̌ by established businesses Lack of customer demand ̌ ̌ ̌ ̌ Lack of opportunities in local area ̌ ̌ ̌ ̌ Unable to secure strategic partnerships ̌ ̌ ̌ ̌ Lack of necessary organisational and ̌ ̌ ̌ ̌ Management capacity Lack of skilled personnel ̌ ̌ ̌ ̌
Appendices 260
SECTION 3 - KNOWLEDGE SOURCES AND INNOVATION INPUTS
12. What were the sources for the ideas and knowledge that led to new innovations and how important were each of them?
High Medium Low Not Importance Importance Importance relevant Within the business ̌ ̌ ̌ ̌ Other parts of the same enterprise group ̌ ̌ ̌ ̌ Potential market already dominated ̌ ̌ ̌ ̌ by established businesses Lack of customer demand ̌ ̌ ̌ ̌ Lack of opportunities in local area ̌ ̌ ̌ ̌ Unable to secure strategic partnerships ̌ ̌ ̌ ̌ Lack of necessary organisational and ̌ ̌ ̌ ̌ Management capacity Lack of skilled personnel ̌ ̌ ̌ ̌
13. Do you use any of the following services to develop and introduce new products, services or processes, and how important are these services to your company’s innovation activities?
High Medium Low Not Importance Importance Importance relevant 1. Industry development advice ̌ ̌ ̌ ̌ 2. Business planning advice ̌ ̌ ̌ ̌ 3. Marketing and promotion ̌ ̌ ̌ ̌ 4. Research (including market research) ̌ ̌ ̌ ̌ & product development 5. Accounting & finance services ̌ ̌ ̌ ̌ 6. IT services ̌ ̌ ̌ ̌ 7. Training services (e.g. TAFE ̌ ̌ ̌ ̌ & industry courses) 8. Recruitment ̌ ̌ ̌ ̌ 9. Accreditation ̌ ̌ ̌ ̌ 10. Legal services (IP, patents, etc) ̌ ̌ ̌ ̌ 11. E-Commerce ̌ ̌ ̌ ̌ 12. Other ̌ ̌ ̌ ̌
Appendices 261
14. Of the services that your business used, where were these located? You may check more than one location, it more than one service was used
Sydney Other Other Did not In-house Local metro in NSW in Aust Overseas use 1. Industry development advice ̌ ̌ ̌ ̌ ̌ ̌ ̌ 2. Business planning advice ̌ ̌ ̌ ̌ ̌ ̌ ̌ 3. Marketing and promotion ̌ ̌ ̌ ̌ ̌ ̌ ̌ 4. Research ̌ ̌ ̌ ̌ ̌ ̌ ̌ & product development 5. Accounting & finance services ̌ ̌ ̌ ̌ ̌ ̌ ̌ 6. IT services ̌ ̌ ̌ ̌ ̌ ̌ ̌ 7. Training services (e.g. TAFE) ̌ ̌ ̌ ̌ ̌ ̌ ̌ & industry courses) 8. Recruitment ̌ ̌ ̌ ̌ ̌ ̌ ̌ 9. Accreditation ̌ ̌ ̌ ̌ ̌ ̌ ̌ 10. Legal services (IP, patents, etc) ̌ ̌ ̌ ̌ ̌ ̌ ̌ 11. E-Commerce ̌ ̌ ̌ ̌ ̌ ̌ ̌ 12. Other ̌ ̌ ̌ ̌ ̌ ̌ ̌
15. Please indicate the importance of the following organisations in your firm’s innovation process.
High Medium Low Not Importance Importance Importance relevant In-house divisions/ staff ̌ ̌ ̌ ̌ Local Government ̌ ̌ ̌ ̌ State Government ̌ ̌ ̌ ̌ Federal Government ̌ ̌ ̌ ̌ Universities, CSIRO, etc ̌ ̌ ̌ ̌ Private sector businesses ̌ ̌ ̌ ̌ Industry associations ̌ ̌ ̌ ̌ Informal networks ̌ ̌ ̌ ̌ (E.g. customers, suppliers, competitors) Other, please indicate ̌ ̌ ̌ ̌
Appendices 262
16. Thinking about the service providers discussed in the previous questions – what has been your company’s expenditure on these services in the last financial year (2004-2005)?
Less than A$3,000 ̌ Less than A$10,000 ̌ Less than A$20,000 ̌ Less than A$50,000 ̌ Less than A$100,000 ̌ More than A$100,000 ̌ None ̌
SECTION 4: COLLABORATION
17. If your company participated in a collaborative arrangement with another organisation in the last year, what was the type (formal – contractual or informal – not contractual) and nature of the collaboration?
Yes Yes No Collaborative activity Informal Formal collaboration Joint marketing or distribution ̌ ̌ ̌ Joint manufacturing ̌ ̌ ̌ Joint research and development ̌ ̌ ̌ Other joint venture ̌ ̌ ̌ Licensing agreement ̌ ̌ ̌ Other form of collaboration/ alliance ̌ ̌ ̌
Appendices 263
18. Does your company collaborate and/ or network with any of the following GOVERNMENT ORGANISATIONS? If so, what was the nature (formal or informal) and importance of these activities?
• Formal activities are defined as participation in formal programs and projects that include a financial contribution, this can be cash or in-kind.
• Informal activities are defined as non-commercial activities and can include social networking.
Yes No Informal Medium Formal High Low Local Councils ̌ ̌ ̌ ̌ ̌ ̌ ̌ GROW (Federal Government Area ̌ ̌ ̌ ̌ ̌ ̌ ̌ Consultative Committees) AusTrade ̌ ̌ ̌ ̌ ̌ ̌ ̌ AusIndustry ̌ ̌ ̌ ̌ ̌ ̌ ̌ Invest Australia ̌ ̌ ̌ ̌ ̌ ̌ ̌ Department of State & ̌ ̌ ̌ ̌ ̌ ̌ ̌ Regional Development Office oMinister of Western Sydney ̌ ̌ ̌ ̌ ̌ ̌ ̌ Other, please indicate ̌ ̌ ̌ ̌ ̌ ̌ ̌
19. Does your company collaborate and/ or network with any of the following INDUSTRY BASED ORGANISATIONS? If so, what was the nature (formal or informal) and importance of these activities?
Yes No Informal Medium Formal High Low Local Chambers of Commerce ̌ ̌ ̌ ̌ ̌ ̌ ̌ Australian Business Limited (ABL) ̌ ̌ ̌ ̌ ̌ ̌ ̌ Australian Industry Group (AIG) ̌ ̌ ̌ ̌ ̌ ̌ ̌ Other industry specific associations ̌ ̌ ̌ ̌ ̌ ̌ ̌ Other private sector businesses ̌ ̌ ̌ ̌ ̌ ̌ ̌ Other, please indicate ̌ ̌ ̌ ̌ ̌ ̌ ̌
Appendices 264
20. Does your company collaborate and/ or network with any of the following EDUCATIONAL or OTHER ORGANISATIONS? If so, what was the nature (formal or informal) and importance of these activities?
Yes No Informal Medium Formal High Low University of Western Sydney ̌ ̌ ̌ ̌ ̌ ̌ ̌ University of Sydney ̌ ̌ ̌ ̌ ̌ ̌ ̌ University of NSW ̌ ̌ ̌ ̌ ̌ ̌ ̌ University of Technology, Sydney ̌ ̌ ̌ ̌ ̌ ̌ ̌ University of Wollongong ̌ ̌ ̌ ̌ ̌ ̌ ̌ Other universities ̌ ̌ ̌ ̌ ̌ ̌ ̌ TAFE ̌ ̌ ̌ ̌ ̌ ̌ ̌ WSROC (Western Sydney ̌ ̌ ̌ ̌ ̌ ̌ ̌ Regional Area of Councils) MACROC (Macarthur Regional ̌ ̌ ̌ ̌ ̌ ̌ ̌ Area of Councils) Greater Western Sydney Economic ̌ ̌ ̌ ̌ ̌ ̌ ̌ Development Board Greater Western Sydney ̌ ̌ ̌ ̌ ̌ ̌ ̌ Business Connection Business Enterprise Centres (BECs)̌ ̌ ̌ ̌ ̌ ̌ ̌ Industry Capability Network ̌ ̌ ̌ ̌ ̌ ̌ ̌ Other, please indicate ̌ ̌ ̌ ̌ ̌ ̌ ̌
21. How important are/ would the following activities facilitated by your local Council be to your business:
High Medium Low Council activity Importance Importance Importance Provide economic & demographic information on your region ̌ ̌ ̌ Support industry forums ̌ ̌ ̌ Support industry networks ̌ ̌ ̌ Lobby state/ federal government on behalf of your industry ̌ ̌ ̌ Lobby state/ federal government on behalf of your region ̌ ̌ ̌ Provide information about other government services ̌ ̌ ̌ Facilitate work placements & ̌ ̌ ̌ employment skill development programs Other, please indicate ̌ ̌ ̌
Appendices 265
Would you like to receive a copy of the report from this project?
Yes ̌ No ̌
What is your email or mailing address?
Would you agree to be contacted for a discussion about innovation in your business?
Yes ̌ No ̌
If yes, what is the best way to contact you?
Name:
Tel:
Email:
THANK YOU
Appendices 266
Appendix 3 Firm Interview Protocol
Ways to Grow: Auditing and Encouraging industrial innovative activity and employment growth in South West Sydney
Interview PROTOCOL
Overview
Project objectives These case studies are part of a wider research design aimed at auditing industrial innovative activity in South West Sydney (including the local government areas (LGAs) of Liverpool, Camden, Campbelltown and Wollondilly) and examining suitable policy levers for encouraging levels of innovative activity within South West Sydney firms. This includes, initially establishing what local resources (including firms, local labour force, institutional environment and knowledge base) firms use in their innovative activities, and therefore potential avenues for local government policy intervention and encouragement. The linkage between innovative activity and employment growth is also the subject of analysis, with the overall aim of policy implementation to see an increase in innovative activity that leads to a corresponding increase in employment. A comparative analysis method has been used in this study to increase validity and depth of results, with two other regions in outer western Sydney also selected for study – Central West Sydney (including the LGAs of Blacktown and Penrith) and North West Sydney (including Baulkham Hills and Hawkesbury).
The broad research questions of the study are: 1. What drives economic growth in these regions? 2. How does innovative activity manifest within firms in these regions? 3. How important are ‘firm specific’ factors to the levels, type and novelty of innovative activity in firms? And are these factors related to geography, specifically surrounding locality/ region? 4. What are the employment related dynamics of these innovative activities? 5. How do the current themes of economic development policy work relate to innovative activity at the regional level? 6. What is the role, if any, of local government in innovation led regional economic development?
The research design for this study is divided into four stages; 1. Innovation audit using official statistics and secondary data analysis 2. Firm-based survey 3. Case studies and in-depth interviews with firms
Appendices 267
4. Policy analysis The previous two stages of research, the Innovation audit of official statistics and the firm-based survey have addressed the first two of the research questions and sections of the third and fourth questions (particularly relating to quantity i.e. how much new recruitment, rather than questions to do with why or how recruitment took place), the aim of the case studies will be to provide more in- depth answers to questions three and four and test a number of hypotheses relating to these questions through case study theoretical replication.
Case study themes
Based on analysis of the Innovation audit and the firm-based survey and in reference to the research questions a number of themes for case study analysis are outlined briefly below.
1. Firm strategy and KISA
As identified in the survey, a key area of difference between firms in the three regions relates to aspects of firm strategy, especially in the use, type and location of innovation sources and the usage of KISA. This theme will explore aspects of decision-making relating to these activities, including firm strategies around searching and learning behaviour, and the distinguishing between routine and exceptional aspects of KISA in innovation and learning processes. In essence, how individual firm strategies relate to how new knowledge is generated, disseminated and applied and secondly the geographical and scale components of these activities i.e. using local versus national versus international resources. The analysis of this theme will be used to characterise different types of firms according to three inter-related criteria, 1) innovation and learning processes, 2) KISA processes, 3) geographical and scale contexts.
2. Employment and the role of professionals
One of the key targets of this research is to analyse relationships between innovation and employment growth. The types of employment that exist in firms and the utilisation of these in- house resources was identified in the survey to be the primary source of knowledge and ideas in firms’ innovation processes. This theme explores further the role of key ‘knowledge professionals’ within firms, and how firms’ innovative activities relate to recruitment and growth within the firm.
Appendices 268
3. Firm connections and networks
Another key point of variance across the three regions, identified in the survey was the usage of multiple sources of knowledge, KISA and the location of these services, suggesting, in line with ‘regional innovation systems’ reasoning that a ‘learning network’ surrounds firms, and the functioning of these networks contributes greatly to the capability and opportunity of the firm to innovate. This theme is closely related to the first two themes, as it deals with aspects of firm strategy including collaborative behaviour, and with the role of ‘knowledge professionals, as networks are inherently made up of social processes between individuals. This theme will provide further exploration of the types, importance, location and degree of engagement that firms have with the networks that surround them.
4. Local government and innovation policy
The final outcome of this research is to deliver an understanding of the role that local government can play in the encouragement of innovative activity in firms in their locality. The survey and other case study themes will identify areas of intersection between local/regional resources and the innovative activity of firms in these geographies. This theme will allow firms to discuss how they see industry and government relating, particularly local government in the local/ regional innovation system.
Rationale for selection of case studies
A criteria table (see attached) was developed to select firms participate in the case study section of the research. Following the survey, firms had an option to participate in the next phase of the research. The criteria table includes all of these firms that elected to be part of the next round of research. To increase validity and reduce the effect of any bias from the survey a number of firms have been randomly selected from a KOMPASS business database to also be considered in the criteria table.
The criteria identified in the table includes
• Regional location of firms
• Broad industry category of firm
• Business size of firm (based on employment)
• Length of time in current location
Appendices 269
• Innovations, type and novelty
Firms selected through the criteria matrix provide adequate comparability across the three regions and industry categories. Cases were also selected to provide theoretical replication across a number of key hypotheses and propositions presented in the following section.
Hypotheses and propositions
Firm strategy and KISA
1. ‘Region’ in terms of locally available knowledge resources (professionals, knowledge institutions and firm networks) has a significant impact on a firm’s innovative capacity and opportunity
2. Knowledge professionals drive firms strategies relating to KISA
Employment and the role of professionals
1. The levels of innovative activity within a firm are impacted by the presence of key knowledge workers
2. Access to knowledge professionals influences the construction of firms’ competitive advantages
3. The type and novelty of innovative activity undertaken by a firm will influence the employment generation of the firms, with higher levels if radical product and service innovation and managerial process innovation contributing more to employment generation in firms.
Firm connections and networks
1. Firms with strong networks over multiple levels (local, national and international) are more innovative or develop more KISA
2. Key knowledge professionals are the conduits for these networks.
Appendices 270
Local government and innovation policy
No definitive propositions are to be tested in the case studies regarding local innovation policy. However the case studies will provide an opportunity to investigate industry perceptions about the role of government in industrial development policy, especially at the local level. An outcome of the research will be to provide advice to local government on their role in innovation policy. Currently, innovation policy is driven at the national and supranational level, through institutions such as the DITR & OECD. To local policymakers innovation policies of this type offer little to their jurisdiction. Local government’s have shown limited interest in developing innovation policy and probably don’t have the necessary resources for this type of policy development, despite the increasing evidence of the regional territorial dimension to innovative activity.
Possible propositions could be for example: ‘Innovation policies at the local level are perceived by the Firm as not having influence in their innovation strategy and firm performance.
Field procedures
• Firms will be selected to participate in the case study interview stage of the research through the criteria table. As mentioned earlier cases are selected in order to achieve comparability across the three regions, industry categories and business size. Further selection will be based on the ability to test cases against the hypotheses detailed earlier.
• Once firms have been selected through the criteria table, contact will be made with the firms to firstly to reconfirm their consent, then to arrange a suitable time for meeting. The firms will be forwarded a letter with a general overview of the project and the key themes and questions that will be covered in the interview.
• Prior to the interview, survey questionnaires for the individual firms will be reviewed, and if possible the firm’s website and other available documentation will also be reviewed to familiarise the interviewer with the company. For cases that did not complete the survey, these companies will be asked to complete the survey, if possible.
• At the time of the interview the participant will be asked if they agree to have the interview recorded. If so, recording will commence. The interviewer will then go through the
Appendices 271
questions in order (shown following). These questions will have been previously provided to the firm, but the interviewer will carry a spare copy.
• The interview will take approx one hour to complete. Participant companies will be provided with the project website for further information and a copy of the project report when ready.
Appendices 272
Appendix 4 Firm Interview Questions
Interview questions
Firm strategy and value chain position – 1. Can you please tell me what your firm’s main competitive advantage is (for example product innovation, reliability, product quality, problem solving ability etc)? 2. Can you describe where you firm fit into the industry value chain – raw primary resources, design, distribution, final customer? KISA and knowledge professionals 3. Thinking about your firm’s most innovative product or process, what knowledge intensive service activities were used, how and where do you access them from (i.e. in house resources, external service providers)? What point in the product cycle did you require these services, for example early on in the product design phase, or later when marketing the final product? Could you please describe an example? 4. Who are the persons responsible for this, and how do they communicate new information through the organisation? What are their occupations? Where these personnel recruited specifically for these activities? Knowledge employment 5. What is your recruitment strategy? How do you employ new workers? Can you please describe recent practice? When do you recruit and why? Do you undertake any staff training and development programs? Could you please tell us about them? Regional resources 6. Do you consider (specify region) a good place to do business? What are the advantages and disadvantages, what area (if any) does you find it challenging to compete in? What is the best thing of working in this area? Please specify? 7. What attracted the firm to the area in the first instance? Do these attractions still hold, or are there other reasons now that keep you in this area? Collaboration and networking 8. What kind of collaborative and networking activities are your firm involved in? Can you tell me a little about these i.e. how often, with whom? How do you maintain these? Who drives these activities within your firm? What are the logistics – where do you meet? Is it local, metro etc? 9. What are the main benefits of these activities to your firm, especially in terms of innovative activities? Can you please give me an example (best) of collaboration, detail? Local government and innovation policy 10. What local policies and programs are you aware of? 11. Can you identify any gaps in policy, where it should be, or how it is succeeding or failing in terms of your firm’s learning and innovative activities? Discuss. What effect does the regulatory framework in productivity & employment growth in your firm?
Appendices 273
12. What relationship do you have with your council? Do you think it should be different? How much do you think your company should participate in local economic development initiatives with other organisations?
Thank you
Appendices 274
Appendix 5 Criteria Table for interview selection • Forty two firms were entered into the criteria matrix • Fourteen firms were selected from this for an in-depth interview
Criteria Case study ID no Survey id no.
Region South West Sydney Central West Sydney North West Sydney
Industry Manufacturing Business services Size Business cycle Innovations Product and service Operational process Managerial process
Novelty Incremental Radical
KISA usage
Collaborations Yes No
Networking Mulitple sources Government Industry Education Regional Institutions
Appendices 275
Appendix 6 Social profile of regions
Regional Social profiles
Age profile
The three regions under investigation have relatively young age profiles compared with metropolitan Sydney and NSW. The large populations of children and young people are in line with the suburban nature of these locations, and are also reflected in the social/ family demographics.
Table 1.2 Age Breakdown as % of Population 0-14 years 15-24 years 25-64 years 65+ years South West Sydney 23.9% 15.3% 52.7% 8.1% Central West Sydney 24.3% 15.4% 52.8% 7.5% North West Sydney 21.9% 14.9% 54.9% 8.3% Metro Sydney 19.5% 14.1% 54.5% 11.9% NSW 20.1% 13.5% 53.1% 13.2% Source: ABS 2001 Census
The three areas vary in their ethnic profile. South West Sydney has one of the most ethnically diverse populations in Sydney with over a third of residents born overseas and nearly 40% of the population speaking a language other than English. Sydney itself is an ethnically diverse city, a quality all the regions share and is evidenced by the differences in Sydney regions and the whole of NSW.
Table 1.3 Ethnicity Characteristics % Speaking % Aboriginal & % Born overseas in language other than Torres Strait % Born overseas NES country English Islanders South West Sydney 34.1% 28.8% 39.3% 1.5% Central West Sydney 26.4% 19.2% 22.6% 2.4% North West Sydney 21.6% 13.8% 15.9% 0.7% Metro Sydney 31.1% 22.8% 27.6% 1.1% NSW 23.2% 16.0% 19.0% 2.1% Source: ABS 2001 Census
In terms of family structure all three regions are similar in that couple families with children predominate. South West and Central West Sydney have higher levels of one parent families than North West, which in turn has moderately higher levels of Couple family households.
Appendices 276
Table 1.4 Family Structure Couple family/ with Couple family One parent Total number children w/out children family Other families of families % % % % No. South West Sydney 56.7% 24.3% 17.5% 1.5% 147,339 Central West Sydney 54.8% 25.5% 18.2% 1.5% 113,283 North West Sydney 58.9% 29.2% 10.9% 1.0% 54,772 Metro Sydney 51.0% 31.7% 15.2% 2.1% 1,014,441 NSW 47.8% 34.9% 15.5% 1.8% 1,654,583 Source: ABS 2001 Census
Dwellings
Dwellings in these regions are overwhelmingly separate houses, with higher percentages in all regions compared with metropolitan Sydney and even NSW averages. These high levels of separate house dwelling also aligns with other trends of ‘suburban’ development, such as high personal motor vehicle use and a ‘sprawling’ land use pattern.
Table 1.5 Dwelling Structure Separate House Attached House Flat, Unit & Apartment South West 79.9% 10.7% 8.1% Central West 86.1% 7.8% 5.0% North West 87.6% 8.1% 3.1% Metro Sydney 63.1% 11.3% 23.9% NSW 70.3% 9.3% 17.9% Source: ABS 2001 Census
Some divergence appears between the three regions in terms of dwelling tenure, with higher levels of home ownership in the North West region compared to that of the South West and Central West. Not all that surprising considering some of the other differentials between the regions including income and labour force participation.
Table 1.6 Dwelling Tenure Fully owned Being purchased Rented South West 32.6% 34.5% 32.9% Central West 31.9% 34.7% 33.4% North West 43.5% 34.4% 22.1% Sydney SD 39.0% 23.7% 29.0% NSW 41.1% 23.3% 27.5% Source: ABS 2001 Census
Appendices 277
Income
Table 1.7 show income levels for the regions. The Mean Taxable Income shows the clear difference that exists in Baulkham Hills. All of the other Local Government Areas (LGAs) have an average taxable income in the mid to late $30,000. Baulkham Hills average is nearly $48,000, almost $10,000 the high range of the other LGAs. Baulkham Hills difference in income is also reflected in their occupational and educational profile.
Table 1.7 Income levels Centrelink Medium weekly Medium weekly Income individual Medium weekly household Mean Income Support income Family Income income taxable customers South West Sydney Liverpool $400-$499 $1,000-$1,499 $800-$999 $35,592 35,188 Campbelltown $300-$399 $800-$999 $800-$999 $35,581 30,720 Camden $400-$499 $1,000-$1,499 $1,000-$1,499 $39,282 6,009 Wollondilly $400-$499 $1,000-$1,499 $800-$999 $37,884 6,605 Central West Sydney Penrith $400-$499 $1,000-$1,499 $800-$999 $36,537 30,938 Blacktown $400-$499 $1,000-$1,499 $800-$999 $36,639 56,715 North West Sydney Baulkham Hills $500-$599 $1,500-$1999 $1,000-$1,499 $47,971 16,717 Hawkesbury $400-$499 $1,500-$1,999 $800-$999 $37,364 9,634 Metro Sydney $400-$499 $1,000-$1,499 $800-$999 NA 782,328 NSW $300-$399 $800-$999 $800-$999 $41,623 1,474,412 Source: ABS 2001 Census
Education and Qualifications
Table 1.8 shows participation at educational institutions for each of the regions. Despite relatively young populations both South West and Central West Sydney are below the Metropolitan Sydney rate for educational participation and in the case of Central West Sydney, also below the NSW level. North West Sydney sits well above all other comparisons. Educational participation is one of the key measures of ‘human capital’ levels in a community and an indicator of the innovative capacity of this community. Low levels of educational participation have flow on effect throughout the entire local innovation system, as will be discussed later.
The levels of educational participation also naturally trend towards non-school qualifications. Again North West Sydney is well above all the other comparisons with 44% of the population aged 15 and over possessing some level of non-
Appendices 278
school qualification, in Baulkham Hills alone this percentage rises to 47.1% the majority of which are university based qualifications Bachelors Degree or higher.
Table 1.8 Participation at Educational Institutions University Secondary TAFE or other Other Total %population* South West Sydney 99,111 11,745 8,814 2,424 37,979 13.9% Central West Sydney 120,257 13,360 10,457 2,583 42,288 13.2% North West Sydney 56,760 5,498 8,499 1,386 24,237 15.7% Metro Sydney 907,195 117,907 170,460 37,658 463,165 14.7% NSW 1,648,023 188,943 231,169 50,068 692,979 13.9% Source: ABS 2001 Census *Population aged 15 years and older
Table 1.9 Highest Non-School Qualification Graduate Advanced Diploma / Diploma / Postgraduate Graduate Bachelor Diploma Certificate Total % of Degree level Certificate Degree level level population* South West Sydney 3,434 2,335 25,167 20,596 68,210 28.4% Liverpool 1,095 595 7,781 6,120 19,020 30.2% Campbelltown 1,004 684 5,967 5,006 18,533 29.0% Camden 368 369 2,282 2,007 6,842 36.9% Wollondilly 274 299 1,600 1,499 6,087 35.6% Central West Sydney 3,500 2,057 22,201 15,497 54,933 30.6% Penrith 1,118 882 7,393 6,016 24,436 30.7% Blacktown 2,382 1,175 14,808 9,481 30,497 30.6% North West Sydney 4,343 2,263 19,594 12,254 29,314 44.0% Baulkham Hills 3,775 1,774 16,522 9,442 19,476 47.1% Hawkesbury 568 489 3,072 2,812 9,838 36.8% Sydney SD 89,808 41,295 387,736 215,880 482,910 38.7% NSW 109,621 60,952 506,806 312,187 819,902 36.2% Source: ABS 2001 Census *Population aged 15 years and older
Industry and Occupation
Table 1.10 shows the occupational profile of the regions, and again North West Sydney differs clearly, compared with both the South West and the Central West – North West has twice the levels of ‘Manager and Administrator’ occupations than the other two regions, and 10 percentage point more in the ‘Professionals’ category.
Appendices 279
Table 1.20 Broad category occupations of residents (%) Central Metro Occupation South West West North West Sydney NSW Managers and Administrators 6.4% 5.4% 12.1% 9.0% 9.4% Professionals 12.5% 12.2% 21.4% 21.2% 19.1% Associate Professionals 10.2% 10.0% 12.4% 11.8% 11.6% Tradespersons and Related Workers 15.0% 13.8% 11.4% 11.1% 11.9% Advanced Clerical and Service Workers 4.1% 3.9% 5.5% 4.5% 4.2% Intermediate Clerical, Sales and Service 19.1% 20.1% 16.5% 17.2% 16.5% Intermediate Production and Transport 11.9% 12.7% 5.6% 7.4% 7.9% Elementary Clerical, Sales and Service 9.4% 10.4% 8.2% 9.1% 9.3% Labourers and Related Workers 9.2% 9.3% 5.2% 6.6% 8.0% Source: ABS 2001 Census
The high concentrations of ‘Managers and Administrators’ and ‘Professionals’ is balanced out by lower levels of ‘Clerical’ and ‘Production Workers’. The reverse is true for South West and Central West Sydney. This resident occupation profile is consistent with much of Sydney’s northern geography – the marked differences between where management and professional occupations live in the north because of perceived desirability of harbour and exclusive suburbs located there, compared with South West and Central West Sydney’s more ‘Blue collar’ profile and suburban ‘working class’ profile.
The industrial profile of employed residents tells a similar story to the occupational profile, with traditional ‘blue collar’ industries such as manufacturing predominating in South West and Central West Sydney, whilst North West Sydney’s largest industrial concentration is in the Property and Finance industry. The residential property building boom in these growth regions sees Construction employment play a significant role. Retail employment is significant across all regions, and indeed metropolitan Sydney, accounting for around 15% of total employment.
Appendices 280
Table 1.21 Industry categories of resident workforce South West Central West North West Metropolitan Industry Sydney Sydney Sydney Sydney Agriculture, Forestry, Fishing 1.2% 0.8% 1.8% 0.6% Mining 0.3% 0.1% 0.1% 0.1% Manufacturing 17.5% 16.4% 11.7% 12.2% Electricity, Gas & Water Supply 0.6% 0.8% 0.7% 0.6% Construction 8.5% 8.3% 8.8% 6.9% Wholesale Trade 6.2% 7.4% 7.5% 6.0% Retail Trade 14.5% 15.1% 14.4% 13.4% Accommodation, Cafes & Restaurants 3.6% 3.5% 3.5% 4.8% Transport & Storage 6.1% 5.4% 3.4% 5.0% Communications Services 2.2% 2.6% 1.8% 2.4% Finance and Insurance 4.6% 5.3% 5.1% 6.1% Property and Business Services 9.2% 9.6% 12.9% 14.5% Gov't Administration & Defence 4.0% 3.7% 3.6% 3.4% Education 5.7% 4.8% 7.6% 6.4% Health & Community Services 7.9% 8.3% 8.9% 8.9% Cultural & Recreation Services 1.6% 1.7% 2.2% 2.8% Personal & Other Services 3.5% 3.7% 3.9% 3.6% Not Stated 2.8% 2.5% 1.9% 2.4% Source: ABS 2001 Census
18.0%
16.0%
14.0%
12.0%
10.0%
8.0%
6.0%
4.0%
2.0%
0.0%
Manufacturing Construction Retail Trade Property and Business Health & Services Community Services
South West Sydney Central West Sydney North West Sydney Metro Sydney
Appendices 281
Summary of Social Profile
The previous social allows a number of conclusions to be drawn about these regions, their economic prospects and their innovative capacity.
Population growth
Population growth in each of these regions represents both a challenge and an opportunity. Significant employment generation will be necessary to maintain employment self-sufficiency. Employment self-sufficiency refers to the number of jobs in a region compared with a ratio of population, i.e. 80 jobs for every 100 resident labour force members. Employment self-sufficiency is necessary to maintain current journey to work patterns, and in the future further employment generation will be necessary just to stabilize and maintain these current journey to work patterns, let alone improve on them.
Educational participation and qualification attainment
The importance of knowledge and the so-called move to a more knowledge- based economy also increases the role of ‘human capital’ in regional economic development. Human capital is defined as the embodied resources of people, education, qualifications, knowledge, experience and skills (ref). Participation at educational institutions and attained educational qualifications are one of the key components of human capital levels, which in turn affect other basic indicators such as income levels, but also affect innovative capacity, and therefore employment and economic development through their flow on affects in the local system of innovation.
Land Use Patterns
As the development footprint of Sydney expands, the frontier will exist along all three of these regions (see map pg 7), the conflict between using land for residential development (in line with huge population projections and current tendencies towards separate house dwellings) and employment including agricultural employment will intensify. A recognition of the need to generate regional employment to ensure some level of regional employment self-
Appendices 282
sufficiency and transport sustainability will mean that current land use patterns will need to be modified. A South West Sydney Employment Lands Strategy (2004) points to the need for substantial allowances for ‘employment purpose’ lands to be set aside in current residential planning models. Further pressure on this need is also provided by the per dollar yield difference between developed residential and industrial land.
Appendices 283
Appendix 7 Knowledge Sources Factor analysis (PCA)
Scree Plot for Knowledge sources factor analysis
Scree Plot
4
2 Eigenvalue
0
1 2 3 4 5 6 7 8 9 10 Component Number
Appendices 284
Principal Components Analysis of Knowledge sources Correlation Coefficients Matrix
Correlation Matrix
HM HM HM HM - Knowledge HM Knowledge HM Knowledge Knowledge HM sources - HM Knowledge sources - Knowledge HM HM sources - sources - Knowledge Professional Knowledge sources - Other parts sources - Knowledge Knowledge Consultants Universities sources - conferences, sources - Within the of enterprise Clients and sources - sources - and paid & Higher Government meetings, Websites, business group customers Suppliers Competitors advisors Education Agencies fairs etc journals etc Correlation HM Knowledge source 1.000 .392 .711 .382 .454 .255 .220 .222 .347 .402 - Within the business HM Knowledge source - Other parts of .392 1.000 .450 .384 .209 .331 .208 .117 .295 .061 enterprise group HM Knowledge source .711 .450 1.000 .449 .501 .428 .387 .286 .378 .437 - Clients and customer HM Knowledge source .382 .384 .449 1.000 .240 .382 .280 .149 .209 .104 - Suppliers HM Knowledge source .454 .209 .501 .240 1.000 .124 .211 .040 .209 .325 - Competitors HM Knowledge source - Consultants and paid .255 .331 .428 .382 .124 1.000 .322 .348 .335 .267 advisors HM - Knowledge sources - Universities & .220 .208 .387 .280 .211 .322 1.000 .524 .329 .359 Higher Education HM Knowledge source .222 .117 .286 .149 .040 .348 .524 1.000 .316 .298 - Government Agencies HM Knowledge source - Professional .347 .295 .378 .209 .209 .335 .329 .316 1.000 .439 conferences, meetings fairs etc HM Knowledge source .402 .061 .437 .104 .325 .267 .359 .298 .439 1.000 - Websites, journals et
Appendices 285
Principal Components Analysis of knowledge sources
Goodness-of-Fit tests Kaiser-Meyer-Olkin Measure of Sampling adequacy & Bartlett’s test of Sphericity
KMO and Bartlett's Test Kaiser-Meyer-Olkin Measure of Sampling Adequacy. .805
Bartlett's Test of Approx. Chi-Square 360.414 Sphericity df 45 Sig. .000
Communalities matrix
Communalities
Initial Extraction HM Knowledge sources 1.000 .702 - Within the business HM Knowledge sources - Other parts of 1.000 .684 enterprise group HM Knowledge sources 1.000 .775 - Clients and customers HM Knowledge sources 1.000 .670 - Suppliers HM Knowledge sources 1.000 .731 - Competitors HM Knowledge sources - Consultants and paid 1.000 .591 advisors HM - Knowledge sources - Universities & 1.000 .742 Higher Education HM Knowledge sources 1.000 .733 - Government Agencies HM Knowledge sources - Professional 1.000 .779 conferences, meetings, fairs etc HM Knowledge sources 1.000 .741 - Websites, journals etc Extraction Method: Principal Component Analysis.
Appendices 286
Component variance and matrix
Total Variance Explained
Initial Eigenvalues Extraction Sums of Squared Loadings Rotation Sums of Squared Loadings Component Total % of VarianceCumulative % Total % of VarianceCumulative % Total % of VarianceCumulative % 1 3.899 38.992 38.992 3.899 38.992 38.992 2.072 20.715 20.715 2 1.339 13.392 52.384 1.339 13.392 52.384 1.917 19.175 39.890 3 1.135 11.346 63.730 1.135 11.346 63.730 1.748 17.482 57.372 4 .775 7.753 71.483 .775 7.753 71.483 1.411 14.111 71.483 5 .653 6.533 78.016 6 .587 5.872 83.888 7 .547 5.468 89.356 8 .465 4.648 94.004 9 .359 3.586 97.590 10 .241 2.410 100.000 Extraction Method: Principal Component Analysis.
Appendices 287
Component Matrix
Component Matrixa
Component 1 2 3 4 HM Knowledge sources .735 -.347 -.202 -.007 - Within the business HM Knowledge sources - Other parts of .553 -.345 .445 -.248 enterprise group HM Knowledge sources .839 -.242 -.083 .078 - Clients and customers HM Knowledge sources .575 -.282 .451 .240 - Suppliers HM Knowledge sources .536 -.403 -.450 .280 - Competitors HM Knowledge sources - Consultants and paid .606 .181 .410 -.152 advisors HM - Knowledge sources - Universities & .605 .477 .061 .380 Higher Education HM Knowledge sources .509 .647 .107 .209 - Government Agencies HM Knowledge sources - Professional .618 .233 -.105 -.576 conferences, meetings, fairs etc HM Knowledge sources .598 .259 -.537 -.167 - Websites, journals etc Extraction Method: Principal Component Analysis. a. 4 components extracted.
Appendices 288
Rotated Component Pattern Matrix
Rotated Component Matrixa
Component 1 2 3 4 HM Knowledge sources .706 .358 .055 .268 - Within the business HM Knowledge sources - Other parts of .157 .789 -.063 .181 enterprise group HM Knowledge sources .677 .452 .236 .239 - Clients and customers HM Knowledge sources .299 .705 .229 -.180 - Suppliers HM Knowledge sources .853 .039 .030 .017 - Competitors HM Knowledge sources - Consultants and paid -.018 .590 .390 .302 advisors HM - Knowledge sources - Universities & .211 .136 .820 .080 Higher Education HM Knowledge sources -.011 .088 .824 .215 - Government Agencies HM Knowledge sources - Professional .110 .256 .165 .821 conferences, meetings, fairs etc HM Knowledge sources .465 -.157 .319 .631 - Websites, journals etc Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. a. Rotation converged in 6 iterations.
Component Transformation Matrix
Component 1 2 3 4 1 .579 .527 .461 .418 2 -.488 -.353 .735 .311 3 -.560 .757 .124 -.313 4 .335 -.160 .482 -.794 Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization.
Appendices 289
Appendix 8 KISA Factor analysis (PCA)
Scree Plot for KISA factor analysis
Scree Plot
4
2 Eigenvalue
0
1 2 3 4 5 6 7 8 9 10 11 Component Number
Appendices 290
Principal Components Analysis of KISA
Goodness-of-Fit tests Kaiser-Meyer-Olkin Measure of Sampling adequacy & Bartlett’s test of Sphericity
KMO and Bartlett's Test Kaiser-Meyer-Olkin Measure of Sampling Adequacy. .846
Bartlett's Test of Approx. Chi-Square 446.803 Sphericity df 55 Sig. .000
Communalities matrix
Communalities
Initial Extraction HM KISA - Industry 1.000 .815 Development Advice HM KISA - Business 1.000 .600 Planning Advice HM KISA - Marketing and 1.000 .747 Promotion HM KISA - Research and 1.000 .609 Development HM KISA - Accounting and 1.000 .533 Financial Services HM KISA - IT Services 1.000 .592 HM KISA - Training 1.000 .815 services HM KISA - Recruitment 1.000 .710 HM KISA - Accreditation 1.000 .759 HM KISA - Legal Services 1.000 .718 (IP, Patents, etc) HM KISA - E-Commerce 1.000 .665 Extraction Method: Principal Component Analysis.
Appendices 291
Component variance
Total Variance Explained
Initial Eigenvalues Extraction Sums of Squared Loadings Rotation Sums of Squared Loadings Component Total % of Variance Cumulative % Total % of Variance Cumulative % Total % of Variance Cumulative % 1 4.623 42.028 42.028 4.623 42.028 42.028 2.880 26.179 26.179 2 1.103 10.032 52.059 1.103 10.032 52.059 1.965 17.867 44.046 3 .970 8.817 60.876 .970 8.817 60.876 1.396 12.687 56.733 4 .865 7.868 68.744 .865 7.868 68.744 1.321 12.011 68.744 5 .782 7.109 75.854 6 .705 6.412 82.266 7 .525 4.769 87.035 8 .469 4.265 91.299 9 .374 3.400 94.700 10 .328 2.983 97.683 11 .255 2.317 100.000
Extraction Method: Principal Component Analysis.
Appendices 292
Component matrix
Component Matrixa
Component 1 2 3 4 HM KISA - Industry .353 .621 .346 .431 Development Advice HM KISA - Business .714 -.193 -.012 -.229 Planning Advice HM KISA - Marketing and .764 -.039 .317 -.247 Promotion HM KISA - Research and .730 .128 .209 -.127 Development HM KISA - Accounting and .648 -.034 -.125 -.311 Financial Services HM KISA - IT Services .728 .215 .097 -.080 HM KISA - Training .511 -.482 .046 .565 services HM KISA - Recruitment .558 -.489 .396 .049 HM KISA - Accreditation .654 -.095 -.459 .333 HM KISA - Legal Services .644 .069 -.542 -.077 (IP, Patents, etc) HM KISA - E-Commerce .713 .361 -.131 .097 Extraction Method: Principal Component Analysis. a. 4 components extracted.
Appendices 293
Rotated Component Matrix
Rotated Component Matrixa
Component 1 2 3 4 HM KISA - Industry .093 .017 .051 .896 Development Advice HM KISA - Business .661 .333 .221 -.055 Planning Advice HM KISA - Marketing and .817 .098 .194 .181 Promotion HM KISA - Research and .673 .204 .124 .315 Development HM KISA - Accounting and .610 .400 .015 -.038 Financial Services HM KISA - IT Services .598 .312 .071 .363 HM KISA - Training .094 .247 .858 .094 services HM KISA - Recruitment .553 -.088 .629 -.034 HM KISA - Accreditation .126 .757 .392 .124 HM KISA - Legal Services .315 .787 .003 .007 (IP, Patents, etc) HM KISA - E-Commerce .394 .533 .030 .474 Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. a. Rotation converged in 9 iterations.
Component Transformation Matrix
Component 1 2 3 4 1 .730 .522 .334 .289 2 -.064 .109 -.666 .735 3 .359 -.836 .217 .352 4 -.578 .127 .630 .502 Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization.
Appendices 294
Appendix 9 Network Factor Analysis (PCA)
Scree Plot for Network factor analysis
Scree Plot
4
2 Eigenvalue
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Component Number
Appendices 295
Principal Components Analysis of Networks
Goodness-of-Fit tests Kaiser-Meyer-Olkin Measure of Sampling adequacy & Bartlett’s test of Sphericity
KMO and Bartlett's Test Kaiser-Meyer-Olkin Measure of Sampling Adequacy. .720
Bartlett's Test of Approx. Chi-Square 547.115 Sphericity df 105 Sig. .000
Communalities matrix
Communalities
Initial Extraction HM Networking - 1.000 .763 Ausindustry HM Networking - DSRD 1.000 .716 HM Networking - OMWS 1.000 .703 HM Networking - Industry 1.000 .551 associations HM Networking - UWS 1.000 .544 HM Networking - USYD 1.000 .711 HM Networking - Other 1.000 .561 Universities HM Networking - WSROC 1.000 .687 HM Networking - 1.000 .756 GWSEDB HM - Networking - 1.000 .711 Councils HM Networking - Austrade 1.000 .705 HM Networking Chamber 1.000 .588 of Commerce HM Networking ABL 1.000 .532 HM Networking AIG 1.000 .535 HM Networking - TAFE 1.000 .659 Extraction Method: Principal Component Analysis.
Appendices 296
Component Matrix
Component Matrixa
Component 1 2 3 4 5 HM Networking - .706 .307 -.226 -.340 -.061 Ausindustry HM Networking - DSRD .653 -.422 .020 -.311 .117 HM Networking - OMWS .630 -.469 .026 -.271 .112 HM Networking - Industry .369 .398 .474 .175 .037 associations HM Networking - UWS .532 -.060 .410 .272 -.123 HM Networking - USYD .645 -.118 .027 .484 -.214 HM Networking - Other .436 .357 -.332 .357 .075 Universities HM Networking - WSROC .681 .130 -.191 .160 -.379 HM Networking - .653 -.335 -.247 .333 -.214 GWSEDB HM - Networking - .466 -.387 -.145 .156 .547 Councils HM Networking - Austrade .683 .098 -6.6E-005 -.447 -.174 HM Networking Chamber .347 -.048 .663 -.024 .156 of Commerce HM Networking ABL .485 .320 .300 -.315 -.076 HM Networking AIG .398 .342 -.397 -.223 .228 HM Networking - TAFE .358 .358 .015 .259 .579 Extraction Method: Principal Component Analysis. a. 5 components extracted.
Appendices 297
Rotated Component Matrix
Rotated Component Matrixa
Component 1 2 3 4 5 HM Networking - .798 .225 .165 .040 .215 Ausindustry HM Networking - DSRD .342 .166 .742 .130 -.060 HM Networking - OMWS .283 .185 .753 .124 -.078 HM Networking - Industry .165 .139 -.159 .652 .234 associations HM Networking - UWS .036 .443 .165 .565 .003 HM Networking - USYD .027 .773 .177 .259 .120 HM Networking - Other .211 .450 -.112 -.050 .547 Universities HM Networking - WSROC .445 .692 .027 .069 .067 HM Networking - .086 .772 .384 -.046 .045 GWSEDB HM - Networking - -.111 .165 .696 .003 .432 Councils HM Networking - Austrade .744 .181 .283 .183 -.068 HM Networking Chamber .024 -.037 .255 .721 -.012 of Commerce HM Networking ABL .570 .008 .029 .453 .019 HM Networking AIG .539 .014 .091 -.181 .451 HM Networking - TAFE .055 .020 .080 .263 .762 Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. a. Rotation converged in 11 iterations.
Component Transformation Matrix
Component 1 2 3 4 5 1 .545 .558 .466 .332 .253 2 .442 -.131 -.739 .182 .457 3 -.190 -.218 -.020 .917 -.275 4 -.650 .612 -.248 .113 .358 5 -.220 -.499 .419 .062 .724 Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization.
Appendices 298
Appendix 10 Hierarchical Cluster Analysis Ward’s Method
Case Processing Summarya
Cases Valid Missing Total N Percent N Percent N Percent 113 99.1% 1 .9% 114 100.0% a. Squared Euclidean Distance used
Agglomeration Schedule
Stage Cluster First Cluster Combined Appears Stage Cluster 1 Cluster 2 Coefficients Cluster 1 Cluster 2 Next Stage 1 8 10 43.349 0 0 6 2 2 5 95.466 0 0 8 3 7 13 159.271 0 0 11 4 3 6 232.921 0 0 9 5 1 12 312.307 0 0 10 6 4 8 391.747 0 1 7 7 4 11 477.506 6 0 9 8 2 9 576.665 2 0 10 9 3 4 688.720 4 7 12 10 1 2 805.256 5 8 11 11 1 7 937.583 10 3 12 12 1 3 1082.773 11 9 0
Cluster Membership
Case 3 Clusters KS REGR factor score 1 1 KS REGR factor score 2 1 KS REGR factor score 3 2 KS REGR factor score 4 2 KISA REGR factor score 1 1 KISA REGR factor score 2 2 KISA REGR factor score 3 3 KISA REGR factor score 4 2 Net REGR factor score 1 1 Net REGR factor score 2 2 Net REGR factor score 3 2 Net REGR factor score 4 1 Net REGR factor score 5 3
Appendices 299
Appendix 11 Cluster membership correlations
Cluster Cluster Cluster membership membership membership 1 of 3 clusters 2 of 3 clusters 3 of 3 clusters Kendall's tau_b Cluster membership 1 of Correlation Coefficient 1.000 -.298** -.196* 3 clusters Sig. (2-tailed) . .002 .039 N 112 112 112 Cluster membership 2 of Correlation Coefficient -.298** 1.000 -.879** 3 clusters Sig. (2-tailed) .002 . .000 N 112 113 113 Cluster membership 3 of Correlation Coefficient -.196* -.879** 1.000 3 clusters Sig. (2-tailed) .039 .000 . N 112 113 113
KS REGR factor score 1 Correlation Coefficient .110 .589** -.660** Sig. (2-tailed) .173 .000 .000 N 112 113 113 KS REGR factor score 2 Correlation Coefficient .225** .203* -.320** Sig. (2-tailed) .005 .012 .000 N 112 113 113 KS REGR factor score 3 Correlation Coefficient .355** -.321** .153 Sig. (2-tailed) .000 .000 .057 N 112 113 113 KS REGR factor score 4 Correlation Coefficient .001 .096 -.099 Sig. (2-tailed) .985 .235 .219 N 112 113 113 KISA REGR factor score 1 Correlation Coefficient .234** .371** -.497** Sig. (2-tailed) .004 .000 .000 N 112 113 113 KISA REGR factor score 2 Correlation Coefficient .311** .135 -.293** Sig. (2-tailed) .000 .097 .000 N 112 113 113 KISA REGR factor score 3 Correlation Coefficient .325** -.079 -.081 Sig. (2-tailed) .000 .328 .319 N 112 113 113 KISA REGR factor score 4 Correlation Coefficient .255** .126 -.257** Sig. (2-tailed) .002 .120 .002 N 112 113 113 Net REGR factor score 1 Correlation Coefficient .129 .070 -.136 Sig. (2-tailed) .114 .391 .095 N 112 113 113 Net REGR factor score 2 Correlation Coefficient -.105 -.087 .142 Sig. (2-tailed) .198 .284 .080 N 112 113 113 Net REGR factor score 3 Correlation Coefficient .051 -.093 .070 Sig. (2-tailed) .532 .251 .390 N 112 113 113 Net REGR factor score 4 Correlation Coefficient .059 .315** -.352** Sig. (2-tailed) .469 .000 .000 N 112 113 113 Net REGR factor score 5 Correlation Coefficient .076 -.022 -.016 Sig. (2-tailed) .351 .785 .848 N 112 113 113 **. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).
Appendices 300
Appendix 12 Cluster membership cross-tabulation
Innovation performance
Any innovation
Crosstab
Ward Method 1 2 3 Total Innovations no Count 0 3 17 20 % within Innovations .0% 15.0% 85.0% 100.0% % within Ward Method .0% 4.7% 40.5% 17.7% Yes Count 7 61 25 93 % within Innovations 7.5% 65.6% 26.9% 100.0% % within Ward Method 100.0% 95.3% 59.5% 82.3% Total Count 7 64 42 113 % within Innovations 6.2% 56.6% 37.2% 100.0% % within Ward Method 100.0% 100.0% 100.0% 100.0%
Chi-Square Tests
Asymp. Sig. Value df (2-sided) Pearson Chi-Square 23.902a 2 .000 Likelihood Ratio 24.587 2 .000 Linear-by-Linear 20.819 1 .000 Association N of Valid Cases 113 a. 1 cells (16.7%) have expected count less than 5. The minimum expected count is 1.24.
Bar Chart
Ward Method 1 2 60 3
40 Count
20
0 no Yes Innovations
Appendices 301
Innovating across all categories
Crosstab
Ward Method 1 2 3 Total All categories No Count 2 23 33 58 innovation % within All categories 3.4% 39.7% 56.9% 100.0% innovation % within Ward Method 28.6% 35.9% 78.6% 51.3% Yes Count 5 41 9 55 % within All categories 9.1% 74.5% 16.4% 100.0% innovation % within Ward Method 71.4% 64.1% 21.4% 48.7% Total Count 7 64 42 113 % within All categories 6.2% 56.6% 37.2% 100.0% innovation % within Ward Method 100.0% 100.0% 100.0% 100.0%
Chi-Square Tests
Asymp. Sig. Value df (2-sided) Pearson Chi-Square 19.997a 2 .000 Likelihood Ratio 20.960 2 .000 Linear-by-Linear 17.667 1 .000 Association N of Valid Cases 113 a. 2 cells (33.3%) have expected count less than 5. The minimum expected count is 3.41.
Bar Chart
Ward Method 50 1 2 3 40
30 Count
20
10
0 No Yes All categories innovation
Appendices 302
Product and service innovation
Crosstab
Ward Method 1 2 3 Total Innovating - Products No Count 0 8 26 34 and Services % within Innovating - .0% 23.5% 76.5% 100.0% Products and Services % within Ward Method .0% 12.5% 61.9% 30.1% Yes Count 7 56 16 79 % within Innovating - 8.9% 70.9% 20.3% 100.0% Products and Services % within Ward Method 100.0% 87.5% 38.1% 69.9% Total Count 7 64 42 113 % within Innovating - 6.2% 56.6% 37.2% 100.0% Products and Services % within Ward Method 100.0% 100.0% 100.0% 100.0%
Chi-Square Tests
Asymp. Sig. Value df (2-sided) Pearson Chi-Square 32.636a 2 .000 Likelihood Ratio 34.177 2 .000 Linear-by-Linear 29.547 1 .000 Association N of Valid Cases 113 a. 2 cells (33.3%) have expected count less than 5. The minimum expected count is 2.11.
Bar Chart
Ward Method 60 1 2 50 3
40
30 Count
20
10
0 No Yes Innovating - Products and Services
Appendices 303
Operational Process Innovation
Crosstab
Ward Method 1 2 3 Total Innovating - Operational No Count 0 14 25 39 Processes % within Innovating - .0% 35.9% 64.1% 100.0% Operational Processes % within Ward Method .0% 21.9% 59.5% 34.5% Yes Count 7 50 17 74 % within Innovating - 9.5% 67.6% 23.0% 100.0% Operational Processes % within Ward Method 100.0% 78.1% 40.5% 65.5% Total Count 7 64 42 113 % within Innovating - 6.2% 56.6% 37.2% 100.0% Operational Processes % within Ward Method 100.0% 100.0% 100.0% 100.0%
Chi-Square Tests
Asymp. Sig. Value df (2-sided) Pearson Chi-Square 19.836a 2 .000 Likelihood Ratio 21.698 2 .000 Linear-by-Linear 19.184 1 .000 Association N of Valid Cases 113 a. 2 cells (33.3%) have expected count less than 5. The minimum expected count is 2.42.
Bar Chart
Ward Method 50 1 2 3 40
30 Count
20
10
0 No Yes Innovating - Operational Processes
Appendices 304
Organisational Process Innovation
Crosstab
Ward Method 1 2 3 Total Innovating - No Count 2 12 22 36 Organisational Process % within Innovating - 5.6% 33.3% 61.1% 100.0% Organisational Process % within Ward Method 28.6% 18.8% 52.4% 31.9% Yes Count 5 52 20 77 % within Innovating - 6.5% 67.5% 26.0% 100.0% Organisational Process % within Ward Method 71.4% 81.3% 47.6% 68.1% Total Count 7 64 42 113 % within Innovating - 6.2% 56.6% 37.2% 100.0% Organisational Process % within Ward Method 100.0% 100.0% 100.0% 100.0%
Chi-Square Tests
Asymp. Sig. Value df (2-sided) Pearson Chi-Square 13.249a 2 .001 Likelihood Ratio 13.155 2 .001 Linear-by-Linear 9.370 1 .002 Association N of Valid Cases 113 a. 2 cells (33.3%) have expected count less than 5. The minimum expected count is 2.23.
Bar Chart
Ward Method 60 1 2 50 3
40
30 Count
20
10
0 No Yes Innovating - Organisational Process
Appendices 305
Radical Innovation
Crosstab
Ward Method 1 2 3 Total Radical innovation No Count 2 34 32 68 % within Radical 2.9% 50.0% 47.1% 100.0% innovation % within Ward Method 28.6% 53.1% 76.2% 60.2% Yes Count 5 30 10 45 % within Radical 11.1% 66.7% 22.2% 100.0% innovation % within Ward Method 71.4% 46.9% 23.8% 39.8% Total Count 7 64 42 113 % within Radical 6.2% 56.6% 37.2% 100.0% innovation % within Ward Method 100.0% 100.0% 100.0% 100.0%
Chi-Square Tests
Asymp. Sig. Value df (2-sided) Pearson Chi-Square 8.740a 2 .013 Likelihood Ratio 8.983 2 .011 Linear-by-Linear 8.659 1 .003 Association N of Valid Cases 113 a. 2 cells (33.3%) have expected count less than 5. The minimum expected count is 2.79.
Bar Chart
Ward Method 40 1 2 3
30
20 Count
10
0 No Yes Radical innovation
Appendices 306
Region by cluster membership
Crosstab
Ward Method Cluster analysis 3 clusters 1 2 3 Total Region Central West Sydney Count 2 9 15 26 % within Region 7.7% 34.6% 57.7% 100.0% North West Sydney Count 1 15 8 24 % within Region 4.2% 62.5% 33.3% 100.0% South West Sydney Count 4 40 19 63 % within Region 6.3% 63.5% 30.2% 100.0% Total Count 7 64 42 113 % within Region 6.2% 56.6% 37.2% 100.0%
Chi-Square Tests
Asymp. Sig. Value df (2-sided) Pearson Chi-Square 7.025a 4 .135 Likelihood Ratio 7.009 4 .135 Linear-by-Linear 3.416 1 .065 Association N of Valid Cases 113 a. 3 cells (33.3%) have expected count less than 5. The minimum expected count is 1.49.
Bar Chart
Ward Method Cluster 40 analysis 3 clusters 1 2 3 30
20 Count
10
0 Central West Sydney North West Sydney South West Sydney Region
Appendices 307
Industry sector by cluster membership
Crosstab
Ward Method 1 2 3 Total Manufacturing No Count 4 34 26 64 % within Manufacturing 6.3% 53.1% 40.6% 100.0% % within Ward Method 57.1% 53.1% 61.9% 56.6% Yes Count 3 30 16 49 % within Manufacturing 6.1% 61.2% 32.7% 100.0% % within Ward Method 42.9% 46.9% 38.1% 43.4% Total Count 7 64 42 113 % within Manufacturing 6.2% 56.6% 37.2% 100.0% % within Ward Method 100.0% 100.0% 100.0% 100.0%
Chi-Square Tests
Asymp. Sig. Value df (2-sided) Pearson Chi-Square .797a 2 .671 Likelihood Ratio .800 2 .670 Linear-by-Linear .501 1 .479 Association N of Valid Cases 113 a. 2 cells (33.3%) have expected count less than 5. The minimum expected count is 3.04.
Appendices 308
Bar Chart
Ward Method Cluster 40 analysis 3 clusters 1 2 3 30
20 Count
10
0 No Yes Manufacturing
Crosstab
Ward Method 1 2 3 Total Services No Count 4 37 26 67 % within Services 6.0% 55.2% 38.8% 100.0% % within Ward Method 57.1% 57.8% 61.9% 59.3% Yes Count 3 27 16 46 % within Services 6.5% 58.7% 34.8% 100.0% % within Ward Method 42.9% 42.2% 38.1% 40.7% Total Count 7 64 42 113 % within Services 6.2% 56.6% 37.2% 100.0% % within Ward Method 100.0% 100.0% 100.0% 100.0%
Chi-Square Tests
Asymp. Sig. Value df (2-sided) Pearson Chi-Square .190a 2 .909 Likelihood Ratio .191 2 .909 Linear-by-Linear .168 1 .682 Association N of Valid Cases 113 a. 2 cells (33.3%) have expected count less than 5. The minimum expected count is 2.85.
Appendices 309
Bar Chart
Ward Method Cluster 40 analysis 3 clusters 1 2 3 30
20 Count
10
0 No Yes Services
Business size and cluster membership
Crosstab
Ward Method 1 2 3 Total Small and Medium More than 51 employees Count 0 13 24 37 enterprises % within Small and .0% 35.1% 64.9% 100.0% Medium enterprises % within Ward Method .0% 20.3% 57.1% 32.7% 50 or less employees Count 7 51 18 76 % within Small and 9.2% 67.1% 23.7% 100.0% Medium enterprises % within Ward Method 100.0% 79.7% 42.9% 67.3% Total Count 7 64 42 113 % within Small and 6.2% 56.6% 37.2% 100.0% Medium enterprises % within Ward Method 100.0% 100.0% 100.0% 100.0%
Chi-Square Tests
Asymp. Sig. Value df (2-sided) Pearson Chi-Square 41.996a 10 .000 Likelihood Ratio 49.156 10 .000 Linear-by-Linear 1.439 1 .230 Association N of Valid Cases 113 a. 8 cells (44.4%) have expected count less than 5. The minimum expected count is .56.
Appendices 310
Bar Chart
Ward Method 60 1 2 50 3
40
30 Count
20
10
0 More than 51 employees 50 or less employees Small and Medium enterprises
Appendices 311