Shareholder Value, Stakeholder Management, and Social Issues: What's the Bottom Line? Author(S): Amy J

Shareholder Value, Stakeholder Management, and Social Issues: What's the Bottom Line? Author(s): Amy J. Hillman and Gerald D. Keim Source: Strategic Management Journal, Vol. 22, No. 2 (Feb., 2001), pp. 125-139 Published by: John Wiley & Sons Stable URL: http://www.jstor.org/stable/3094310 Accessed: 20/04/2010 23:11

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=jwiley.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

John Wiley & Sons is collaborating with JSTOR to digitize, preserve and extend access to Strategic Management Journal.

http://www.jstor.org Strategic Management Journal Strat. Mgmt. J., 22: 125-139 (2001)

SHAREHOLDERVALUE, STAKEHOLDER MANAGEMENT,AND SOCIALISSUES: WHAT'S THE BOTTOMLINE? AMY J. HILLMAN*and GERALDD. KEIM Ivey School of Business, University of Western Ontario, London, Ontario, Canada

We test the relationship between shareholder value, stakeholder management, and social issue participation. Building better relations with primary stakeholders like employees, customers, suppliers, and communitiescould lead to increased shareholder wealth by helping firms develop intangible, valuable assets which can be sources of competitive advantage. On the other hand, using corporate resources for social issues not related to primary stakeholders may not create value for shareholders. We test these propositions with data from S&P 500 firms and find evidence that stakeholder management leads to improved shareholder value, while social issue participation is negatively associated with shareholder value. Copyright ? 2001 John Wiley & Sons, Ltd.

INTRODUCTION ments or other nongovernmentalorganizations. But, is this the appropriaterole for business in Globalizationhas increasedcalls for corporations society? Should the mandateof business extend to use firms' resources to help alleviate a wide beyond its traditionalstakeholders (shareholders, variety of social problems. The pharmaceutical customers,suppliers, employees, local communi- industry, for example, is asked to donate free ties, and government)?These are essentially nor- drugs and vaccines to ThirdWorld nations where mative questions. the afflictedcannot pay. Firms engaged in manu- Empirical researchers interested in the way facturing are encouraged to apply developed firms interact with stakeholders,however, can nation's laws and norms to issues such as child examine related but somewhat more objective labor and environmental pollution in less questions. For example, when firms do expand developed countries,regardless of local laws or their activities beyond those associated with the customs. direct stakeholderrelationships, what is the effect These calls for expanded responsibilitiesfor on the economic viability that createdthe wealth business are intuitively appealing to those who of the firm? That is, if the economic success of see existing governmentsas unable or unwilling firmsraises societal expectationsto considermore to deal with such problems. Firms may indeed than the interests of primarystakeholders when have resources that could be used to help with making resource decisions, can firms respond to issues that are typically dealt with by govern- these social issues and continue to be eco- nomicallyviable? In a world of increasinglycompetitive capital markets,how are a firm's share- Key words: shareholder value; stakeholder man- holders affected by firm decisions to respond to agement; social issues; market value added these increasedresponsibilities? *Correspondenceto: Amy J. Hillman, Ivey School of Busi- ness, Universityof WesternOntario, 1151 RichmondStreet Previous literaturehas studied the relationship North,London, Ontario N6A 3K7, Canada. between firm financial performance and firm

Copyright ? 2001 John Wiley & Sons, Ltd. Received 24 August 1999 Final revision received 14 August 2000 126 A. J. Hillman and G. D. Keim social responsibilityor social performance(e.g., formanceand stakeholdermanagement as well as Aupperle,Carroll, and Hatfield,1985; Griffinand the resource-basedview of the firm. Next, we Mahon, 1997; McGuire, Sundgren,and Schnee- use a sampleof S&P 500 firmsto empiricallytest weis, 1988; Pava and Krausz, 1996; Waddock the proposedrelationships. Finally, we discuss the and Graves, 1997a) but to date there is no clear implicationsof our results for future researchin empiricalrelationship. For example,Waddock and social performance,stakeholder management, and Graves (1997a) find a recursive relationship financial performanceas well as for practicing between social performanceand financial per- managers. formance. They find empirical support for both the propositionthat social performanceleads to improved financial performanceand that better VALUE CREATIONAND DECOUPLING financialperformance leads to social performance. SOCIAL PERFORMANCE Do socially responsible strategies create value for shareholders?Or, is social performancea Corporate social performance is a multi- discretionaryactivity funded by slack cash flow? dimensionalconstruct defined by Carroll (1979) The relationshipbetween social performance as having four components:economic responsi- and financial performancemay be better under- bility to investorsand consumers,legal responsi- stood by separatingsocial performanceinto two bility to the government or the law, ethical components:stakeholder management and social responsibilities to society, and discretionary issue participation.Corporate social performance responsibilityto the community.CSP incorporates (CSP) is a multidimensionalconstruct (Carroll, the interactionbetween the principles of social 1979) that is related to stakeholdermanagement responsibility,the processes of social responsive- although not synonymous (Clarkson, 1995). We ness, and the policies and programsdesigned by believe a key distinctionbetween the two compo- corporationsto addresssocial issues (Wartickand nents of CSP, stakeholdermanagement and social Cochran, 1985). Despite the lack of a shared issue participation,pertains to their respective precise definitionin the literature,CSP is gener- roles in the firm's value creationprocess. Build- ally conceived as a broadconstruct comprised of ing betterrelations with primarystakeholders like stakeholdermanagement and social issue man- employees,customers, suppliers, and communities agement (Clarkson, 1995; Swanson, 1995; (Freeman,1984) could lead to increasedfinancial Wood, 1991). returns by helping firms develop intangible but valuableassets which can be sources of competi- Stakeholder management tive advantage.For example, investing in stakeholder relationsmay lead to customeror supplier In this paper, we adopt what Mitchell, Agle, loyalty, reduced turnover among employees, or and Wood (1997) would classify as a 'narrow' improved firm reputation.These valuable assets definition of stakeholdersin that we consider in turn lead to a positive relationshipbetween primarystakeholders as those stakeholderswho stakeholdermanagement and shareholdervalue 'bear some form of risk as a result of having wherein effective stakeholdermanagement leads investedsome form of capital,human or financial, to improved financial performance.Participating something of value, in a firm' (Clarkson,1994: in social issues not related to the firm's direct 5). These stakeholdersare those without whose relationshipwith primarystakeholders, however, participation the corporation cannot survive may not create similar value for shareholders. (Clarkson, 1995). Primary stakeholdersinclude Instead, we expect that social issue participation capital suppliers(shareholders), employees, other is negatively related to shareholdervalue. Thus, resource suppliers, customers, community resi- we posit that shareholdervalue may be affected dents, and the natural environment (Clarkson, differently depending upon the nature or scope 1995; Starik,1995). Clarksonargues that 'primary of the socially responsiblestrategy/activity. stakeholder groups typically are comprised of In the following section of this paper,we build shareholdersand investors,employees, customers, a theoreticalrationale to supportthese claims and and suppliers, together with what is defined as advanceour hypotheses.Our theoreticaldevelop- the public stakeholdergroup: the governments ment draws upon existing literaturein social per- and communitiesthat provide infrastructuresand

Copyright ? 2001 John Wiley & Sons, Ltd. Strat. Mgmt. J., 22: 125-139 (2001) Shareholders, Stakeholders and Social Issue 127 markets, whose laws and regulations must be outperformits rivals and create value for share- obeyed, and to whom taxes and other obligations holders (Atkinson,Waterhouse, and Wells, 1997; may be due' (1995: 106). While not all com- Barney, 1991; Teece, 1998). Resources that are munity residents are employees, suppliers, cus- most likely to lead to competitiveadvantage are tomers or investors, they do provide various those that meet four criteria: they should be forms of importantinfrastructure for the firm and valuable, rare, inimitable, and the organization in turn are impacteddirectly by tax revenuesand must be organized to deploy these resources physical environmental protection (or effectively (Barney, 1991). Using these criteria, degradation). resourcesthat may lead to competitiveadvantage Clarksonasserts that 'the survivaland continu- include socially complex and causally ambiguous ing profitabilityof the corporationdepends upon resources such as reputation,corporate culture, its ability to fulfill its economic and social pur- long-term relationshipswith suppliers and cus- pose, which is to create and distribute wealth tomers, and knowledge assets (Barney, 1986; or value sufficient to ensure that each primary Leonard,1995; Teece, 1998). stakeholdergroup continuesas part of the corpo- Some strategy researchershave explored the ration's stakeholdersystem' (1995: 107). Thus, firm as an institutionalsetting that can facilitate an organizationcan be viewed as a set of inter- learning and the creation and disseminationof dependent relationships among primary stake- value-producingknowledge (Grant, 1996; Moran holders (Chakravarthy,1986; Donaldson and and Ghoshal, 1996; Nahapietand Ghoshal, 1998; Preston,1995; Evan and Freeman,1988; Greenley Spender, 1996). This institutional context can and Foxall, 1996; Harrisonand St. John, 1994; include, for example, a history of repeatdealings Hill and Jones, 1992; Jones, 1995; Kotter and with actors such as employees, customers, sup- Heskett, 1992). For example,purchasing a quality pliers, and local communitiesthat generaterepu- productat a reasonableprice is a consumerobjec- tational capital and trust (Barney and Hansen, tive. If desiredvalue is not delivered,fewer prod- 1994; Ring and Van de Ven, 1992, 1994). ucts will be purchased. This, in turn, affects By developing longer-termrelationships with presentand futureexpectations resulting in lower primary stakeholderslike customers, suppliers, stock prices, possibly leading to lay-offs, and communities,as well as present and future reductionsin purchasesof inputs from suppliers, employees, firms expandthe set of value-creating and lower taxes being paid by the firm, etc.- exchanges with these groups beyond that which negative consequences for all primary stake- would be possible with interactionslimited to holders. markettransactions. Our emphasishere is on the Managing relationships with primary stake- value that can be createdby interactions,between holders, however, can result in much more than firms and primary stakeholders, which are just their continued participationin the firm. relational rather than transactional since trans- Effective stakeholdermanagement-relations with actionalinteractions can be easily duplicatedand primary stakeholders to include customers, thus offer little potential for competitive advan- employees, suppliers, community residents and tage. Relationshipsinvolve investmentsby both the environment-can constituteintangible, soci- (or multiple) parties and thereby include a time ally complex resources that may enhance firms' dimension;reputation is importantand fair deal- ability to outperformcompetitors in terms of ing and moral treatmentby both (or multiple) long-termvalue creation. parties enhance the value of relationships. The resource-basedview of the firm (Barney, Harrisonand St. John (1996) describeexamples 1991; Penrose, 1959; Wernerfelt,1984) contends of 'webs of interdependencies[that can be] cre- that a firm's ability to perform better than the ated among stakeholders'as organizationalmeans competitiondepends on the unique interplayof to deal with increasinglyuncertain and competi- human, organizational,and physical resources tive environments.Cooperation among competi- over time (Amit and Schoemaker,1993; Barney, tors and other firms operating in geographic 1991; Dierickx and Cool, 1989; Lippman and locales to support infrastructureinvestments in Rumelt, 1982). Many scholars now argue that communitiesare relationaltransactions that lead intangible, difficult-to-replicateresources must to value creation(Hart, 1995; Sharmaand Vred- undergirdthe business processes if a firm is to enburg, 1998). Other examples of activities con-

Copyright ? 2001 John Wiley & Sons, Ltd. Strat. Mgmt. J., 22: 125-139 (2001) 128 A. J. Hillman and G. D. Keim sistent with long-term value creation through nancial performanceare the relationshipsa com- relationshipswith key stakeholdersare coopera- pany develops with customers and the relation- tive planning and design efforts that unite firms ships internal to the firm that shape customer with suppliers and customers and rewarding relations and impact customer service. Legnick- managers/employeeson the basis of customer Hall (1996) emphasizesthe importanceof loyalty- satisfactionmeasures or other measuresof exter- producing relationships with customers that nal reputation(Lado and Wilson, 1994; Martin, extend beyond traditionalfirm boundariesas a Mitchell, and Swaminathan,1995; Mudambiand source of competitiveadvantage. Atkinson et al. Helper, 1998; Nayyar, 1995; Oliver, 1988; Rao, (1997) argue that employees and communities 1994). Because of the relational aspects that shouldalso be includedin this list of relationships underliethese activities, the time dimensionwill that drive financialperformance, such that effec- constitute an important,intangible, path depen- tive stakeholdermanagement with primarystake- dent quality of the relationshipwith that stakeholders is seen as driving financialperformance. holder group. In turn, these relationshipswill be Bennett Stewart, creator of the financial man- difficult for other firms to duplicate at least in agementsystem based on EconomicValue Added the short run. (EVA), arguesthat 'to increaseshareholder value, We are not alone in emphasizingthe impor- a company must addressthe needs of its stake- tance of improvingrelations with primarystakeholders more efficiently and effectively than the holders as competition increases. Chakravarthy companies against which it competes' (Birchard, (1986), Pfeffer (1998), and Prahalad (1997) 1995: 49). Therefore,we propose the following: express similar views and Jones (1995) in his instrumental stakeholder theory contends that Hypothesis 2: Stakeholder management leads firmsthat contractwith their primarystakeholders to improved shareholder value creation. on the basis of mutualtrust and cooperationwill have a competitive advantage over firms that Social issue participation do not, all else equal. Therefore, we propose the following: If stakeholdermanagement is positively related to shareholdervalue creation and the nature of Hypothesis 1: Stakeholder management is posi- causality is such that effective stakeholderman- tively associated with shareholder value cre- agementleads to improvedshareholder value creation. ation, does this relationshipalso extendto another component of corporate social performance- Next, we address the question of causality. If social issue participation? effective stakeholder managementis positively We have suggested above that investing in associated with financial performance,in what relationshipswith primarystakeholders can lead to direction is the causality? The primary stake- valuable, intangible competencies that are holder interdependenceperspective holds that importantin gaining and maintainingcompetitive effective stakeholdermanagement leads to finan- advantage.Using corporateresources to pursue cial performance.Firms can be more successful social issues that are not directly related to the by developing (up to some margin)relationships relationshipwith primary stakeholdersmay not with customers, employees, communities and create such advantages.Social issue participation governments(Harrison and St. John, 1994; Kotter refersto elementsof corporatesocial performance and Heskett, 1992). that fall outside of the direct relationshipsto pri- This sentiment is reflected by Robert Wood mary stakeholders.For example, common forms Johnson(quoted in Prestonand Sapienza, 1990), of social issue participationmay include:avoiding who led Sears in its postwar growth, when he nuclear energy, not engaging in 'sin' industries listed 'four parties to any business in order of (alcohol, tobacco, and gambling),refraining from importance'as 'customers,employees, communi- doing business with countriesaccused of human ties, and stockholders.'He contends that if the rights violations,refusing to sell to the military, interestsof the first three groups are looked after, etc. While each of these may be an important then the stockholdersbenefit. Similarly, Kaplan issue for some membersof society, the fundamen- and Norton (1996) argue that the drivers of fi- tal difference between social issue participation

Copyright ? 2001 John Wiley & Sons, Ltd. Strat. Mgmt. J., 22: 125-139 (2001) Shareholders, Stakeholders and Social Issue 129 and stakeholdermanagement is the absence of scarce resources.MVA is calculatedas: direct ties to the relationshipsbetween the firm and its primarystakeholders. That is, social issue MVA = market value - capital participationmay be characterizedas pertainingto a more 'broad' definitionof social responsibility where market value refers to the equity market beyondthe primarystakeholder exchange (Mitchell valuation of the company and capital refers to et al., 1997) that recognizes companies can be the debt and equity invested in the company. affectedby or affect almost anyone. MVA is simply the difference between the cash Normatively some groups (even within the that both debt and equity investorshave contrib- company)may desire taking stances on such is- uted to a companyand the value of the cash that sues, but participationin such does not necessarily they expect to get out of it. Essentially,MVA is provide the basis for value creation that stake- the stock market'sestimation of net presentvalue. holder managementdoes. For example, while the Thus, MVA is unique in its ability to capture gamblingindustry may be viewed as undesirable shareholdervalue creation because it captures by a segmentof society, firms that choose not to both the valuation(the degree of wealth enrich- be in this industryare not necessarilymaking a ment for the shareholders)and performance(the decision that could providefor sustainedcompeti- overall qualityof capitalmanagement) (Ster Ste- tive advantage.Other firms could easily make the wart, 1996). We use MVA in our analysis not same choice not to participate.Choice of industry cross-sectionally,but by examiningthe change in or overseas investment locations in themselves MVA between one year and the next in order to cannotprovide for the intangiblesources of com- more accuratelyreflect changes in the measure petitiveadvantage so importantin today'scompeti- that are attributableto events in the prior year tive landscape.Similarly, an internationalcorporate ratherthan total capitalizationacross time. That giving programmay providesome value to share- is, the measureof MVA for 1996 representsthe holders in the form of tax deductions.However, change in marketvalue added between 1995 and tax advantagesare readily duplicatedby other 1996. This operationalizationis more appropriate firmsand, therefore,this type of advantagecannot in causal models, such as those we use to test providethe basis for competitiveadvantage. Thus, Hypotheses 2 and 4, because it representsnot we contendthat the very natureof the relationship total capitalizationthat may have to do with betweenshareholder value and social issue partici- events outside the timeframeof interest,but only pation could be different from that with stake- the portion of MVA that is created/destroyed holder managementbecause of the lack of a link during our sample. to importantunderlying sources of competitive While many different operationalizationsof advantagefor the firm. shareholdervalue creation, or firm performance, Thus, we propose the following: could have been used, we chose MVA for a variety of reasons. First, accountingmeasures of Hypothesis 3: Social issue participation is firm performanceare inherentlymore short term negatively related to shareholder value cre- in nature (Briloff, 1972, 1976; Fisher and ation. McGowan, 1983; Hayes and Aberathy, 1980; Ouchi, 1980), tap only historicalaspects of per- Hypothesis 4: Social issue participation leads formance(McGuire, Schneeweis, and Hill, 1986) to decreased shareholder value creation. and are subjectto a great degree of manipulation by managers(Bentson, 1982; Briloff, 1972, 1976; Fisher, 1979; Livingstone and Salamon, 1971; METHODS McGuireet al., 1988; Solomon, 1970; Watts and Zimmerman,1978, 1990). Therefore, Variable operationalization accounting measures of performance,such as Return on Shareholder value creation is operationalized as Assets and Returnon Equity, are less useful for MarketValue-Added, or MVA. MVA was chosen the projectat hand because they are not success- because it is a measurethat capturesthe relative ful in capturingthe long-termvalue of the com- success of firms in maximizingshareholder value pany or value created for shareholders. In through efficient allocation and managementof addition, accounting measures of performance

Copyright ? 2001 John Wiley & Sons, Ltd. Strat. Mgmt. J., 22: 125-139 (2001) 130 A. J. Hillmanand G. D. Keim have difficulty capturingintangible relationships ment Services, Inc. to track the Fortune 1000. In (Barney, 1991; Dierickx and Cool, 1989; Itami, this data base, MVA is calculatedbased on data 1987), such as those with stakeholders. For availablefrom Compustat. example, it is extremely difficult to capturethe Stakeholder Management (SM) is a variable value of customer service or reputationon a that has been rarely quantified.Two exceptions balancesheet (Bentson, 1982; Wattsand Zimmer- are Greenley and Foxall (1997), who use survey man, 1990). Accountingmeasures of performance methodology to measure a firm's orientation are better suited for measuring tangible asset towardsmultiple stakeholders,and Waddockand utilizationand, thus, are inadequatefor capturing Graves (1997b), who use the Kinder,Lydenburg, the type of performanceof interestin this paper- Domini (KLD) index as a measureof stakeholder shareholdervalue creation. performance.Because our question of interest When comparedto other market-basedmeas- involves stakeholder management performance ures of shareholdervalue creation,MVA also has outcomes,we were more interestedin quantifying advantages. Lubatkin and Shrieves (1986) and this relationshipbased on firm behavior rather Rappaport(1992) assert that market-basedmeas- than beliefs and thus turn to the KLD data base ures of performanceare preferableto accounting for our data. measures because of the ability to capture the KLD is a commonlyused measureof corporate future value of income streams more appropri- social performance(e.g., Graves and Waddock, ately. MVA was chosen for this reason and also 1994; Ruf, Muralidhar,and Paul, 1993; Sharfman, because MVA is more thanjust representativeof 1996; Waddock and Graves, 1997a). The KLD the future stream of income as it takes account index of social performanceis compiled by an of debt and equity invested in the company. It independent rating service that focuses exclu- has been shown in finance literaturethat firms sively on ranking approximately800 firms (to that apply net presentvalue, or NPV, performance include the Standard& Poor's 500) on a range measures and invest in positive NPV strategies of nine areas of social performance.These areas maximize the wealth of stockholders(Copeland include:community relations, employee relations, and Weston, 1983). Simple firm calculatedNPV environmentalperformance, product character- measures,however, are also subjectto accounting istics, treatmentof women and minorities,military problemsregarding the anticipationof futurecash contracting, production of alcohol or tobacco, flows and discount rates. Therefore, by using involvement in the gambling industry, involve- MVA, which approximatesthe stock market's ment in nuclear energy, and investmentin areas estimation of net present value, subjective involved with human rights controversies.KLD accountingissues are avoided. uses a variety of sources to capture these data Another market-basedmeasure that approxi- including annual surveys, annual reports, proxy mates the stock market'sestimation of net present statements,and quarterlyreports, as well as exter- value is Tobin's Q (Tobin and Brainard,1968). nal data sources such as articles in the general Tobin's Q is calculated by dividing the firm's business press and agencies. This rating scheme, marketvalue by a firm's asset replacementcosts. in additionto being adoptedin recent empirical While Tobin's Q is commonly used in strategy testing of corporatesocial performance,has been research,we have chosen MVA over Tobin's Q tested for constructvalidity against other meas- because the valuationof asset replacementcosts ures of CSP by Sharfman(1996) and has been in Tobin's Q suffers from the same issues iden- found to be one of the best measures of CSP tified with many accountingmeasures of perform- availableto date. ance- difficulty in valuing intangible assets. In order to adaptthe KLD measureto capture Therefore,because shareholdervalue creation is primary stakeholder managementand create a the performancevariable of interest, MVA is variable SM (stakeholdermanagement), we cus- the most appropriatechoice because it captures tomized this scale to exclude issues outside of shareholdervalue creation without being subject the primarystakeholder domain of CSP. These to accountingmeasure shortfalls. excluded issues were then used in creating the MVA data for this study was taken from the variable Social Issue Participation (SIP). In order Ster StewartPerformance 1000 data base. This to divide these measures into the categories of is a database compiledby Ster StewartManage- stakeholdermanagement and social issue partici-

Copyright ? 2001 John Wiley & Sons, Ltd. Strat. Mgmt. J., 22: 125-139 (2001) Shareholders, Stakeholders and Social Issue 131 pation, we screened items based on their direct of SM and SIP as gestalt measures and used relationship to primary stakeholders. As in the simple summing of the dimensions of the KLD case of the Waddock and Graves (1997b) study, measure adapted for the study at hand. the items for stakeholder management chosen Control variables are also included in our came from five existing categories of the KLD analysis to ensure that any relationship found measures: employee relations, diversity issues, between shareholder value creation, as measured product issues, community relations, and environ- by MVA, stakeholder relations, as measured by mental issues. These five categories parallel the SM, and social issue participation, as measured primary stakeholder groups (other than capital by SIM, are not a result of other confounding suppliers) for corporations: employees (items variables. Because size has been suggested in from employee relations and diversity issues), previous articles (Ullman, 1985; Waddock and customers (items from product issues and com- Graves, 1997a) to be a factor that affects both munity relations), the community (items from firm performance and the larger construct of CSP, community relations, environmental relations and we have included control variables in our analysis diversity issues), and suppliers to the extent that for net sales and net income. Size is a relevant among the diversity issues are reports of dealings variable because size may be related to the with minority-owned suppliers. Ideally, we would urgency and salience of stakeholder relations. like to have broader measures of supplier In addition, previous literature has indicated a relations. While none of these measures captures need to control for industry (Waddock and the full range of relations with these primary Graves, 1997a) and risk (Aupperle et al., 1985; stakeholders, each provides some important evi- Pava and Krausz, 1996; Waddock and Graves, dence pertaining to the nature of stakeholder 1997a). Industry and risk are also included as relations with these groups. control variables to ensure that differences in The SIP variable includes the KLD categories MVA across our sample are not merely an effect of Other, Alcohol/ tobacco/gambling exclusionary of industry differences or differences in risk pro- screens, military exclusionary screens, nuclear files. Industry has been operationalized in this power exclusionary screens, and non-U.S. con- study using the standard 2-digit SIC code. Firm cerns over investment in Burma and Mexico. For risk has been operationalized using beta as individual item components of SM and SIP, reported in Standard & Poor's. please refer to the Appendix. The KLD categories are rated on a scale rang- Analysis ing from -2 (major concerns), -1 (concern), 0 (neutral), +1 (strength), to +2 (major strength). While ideally an event study methodology would Each category in the SM and SIP measures is allow us to evaluate changes in shareholder given equal weighting in that each may range wealth associated with stakeholder management from -2 to +2. Prior use of KLD as a measure and social issue participation, the multi- of CSP has used differential category weightings dimensionality of each of the constructs and the (Graves and Waddock, 1994; Ruf et al., 1993; lack of discrete events associated with such Waddock and Graves, 1997a) based on either activities make this methodology difficult. There- academic opinion about importance of the catego- fore, we use regression analysis as the primary ries (Graves and Waddock studies, 1994, 1997) methodology to test our hypotheses. or the analytic hierarchy process (Ruf et al., Testing of the hypotheses was performed for 1993). However, since theoretical work in stake- the years 1996, 1995, and 1994. The change in holder management and social issues participation MVA between 1995 and 1996 is used as our has yet to identify a ranking of importance for dependent variable in testing Hypotheses 2 and the various stakeholder groups and issues (and 4. The stakeholder management and social issue indeed, Mitchell et al. (1997) assert that no such behavior measured took place during the year universal ranking can be made), we have chosen 1994 (reported by KLD in 1995) and the share- in this paper to give equal importance to the holder value measure is that created/destroyed in categories adopted from KLD identified above in 1995. We chose to model a lagged effect between order to construct our variables SM and SIP. our independent variables and our dependent vari- Given this, we chose to construct our measures ables because the effect of stakeholder man-

Copyright ? 2001 John Wiley & Sons, Ltd. Strat. Mgmt. J., 22: 125-139 (2001) 132 A. J. Hillman and G. D. Keim agement or social issue participation is not Thus, Hypothesis 2 stating that effective stake- expected to have an immediateeffect on shareholder managementleads to improved financial holder value and due to reportingpracticalities performanceis supported.In orderto ensure that (KLD measures are gestalt measures over the our results were not the result of a 1-year anom- year and not logged as specific timed behavior aly, we also checked our analysis with a 3-year duringthe reportingyear). We considerit likely, lag and found no significant differences from however, that the stakeholdermanagement and this model. social issue participationthat is observed in the Hypothesis 3 posited that social issue partici- year 1994 will take fairly quick effect in the pation would be negativelyrelated to shareholder market'sestimation of the firm. value creation.Table 1 presentsthe resultsof this Merging the Stem Stewart Performance1000 test. As hypothesized,social issue participationis data base used for the MVA variable with the significantly(-0.286, p < 0.01) and negatively KLD data base, along with data available from correlatedwith shareholdervalue creation. Compustatfor our control variables, yielded a Hypothesis4 contendsthat social issue partici- final sample size of 308 firms. In order to make pation will lead to decreased shareholdervalue sure that this remaining sample did not differ creation. This hypothesis was tested using SIP from those firms droppeddue to data availability, measures for 1994 as our independentvariable, we tested for the difference in means for our along with the control variables, and MVA control variables (industry,risk, and size). We change 1995-96 as our dependentvariable. Table found no significant differences. Testing of 3 presents the results of this analysis. As Hypotheses1 and 3 was performedthrough corre- expected, SIP has a negative relationshipto the lation analysis and Hypotheses 2 and 4 were creation of shareholdervalue and is significant tested using regressionanalysis. (p < 0.05). Thus, Hypothesis 4 also receives supportin our analysis.

RESULTS Additional analyses In order to test Hypothesis 1, we examined the As noted in the introduction,broader investi- correlationbetween MVA, as measuredby the gations of the relationship between corporate change between 1996 and 1995, and SM in 1994. social performance and financial performance As representedin Table 1, SM and MVA are have found a recursive relationship (Waddock significantlyand positively correlated(0.244, p < and Graves, 1997a). In order to evaluate the 0.01). Thus, Hypothesis 1 is supported. reverseorder causality, that financialperformance While a positive and significantcorrelation is leads to stakeholdermanagement and social issue evident between MVA and SM, Hypotheses 2 participation,we did additionalanalyses using the focuses on issues of causality. In order to test change in MVA between 1993 and 1994 as our Hypothesis 2, we ran regression analyses with independentvariable and SM and SIP from 1994 MVA (change between 1995 and 1996) as our as our individualdependent variables. The model dependentvariable, SM for 1994 as our explana- predicting SM was not significant (p = 0.134), tory independentvariable, and control variables indicating that the reverse causality is not sup- of beta, net income, sales, and industry from ported. Using SIP as our dependent variable 1994. Table 2 presentsthe resultsof this analysis. yielded a significantmodel, but the only predictor The overall model is significant(p < 0.01) with variable of significancewas the control variable an adjustedR2 of 0.414 and SM is positively and of net income with a negative effect. MVA was significantlyassociated with improved MVA (p not significant,again indicatingthat the reverse < 0.01). Table 2 also shows that the control causalityis not supported. variablesrepresenting size (net income and sales) Additionally,although we believe that MVA are significant, but risk and industry are not.' is the most appropriateoperationalization of shareholder value creation, many studies of

1 While it is surprising that risk is not significant in our model, the finding is greatly influenced by the appearance of the relationship between risk and return is as expected in a net income in the model. Without net income in the model, typical two-parameter model.

I, a.

Table 1. Correlations and descriptive statistics

S.D. MEAN MBASS MVA ROA ROE SALES NET SIC BETA SIP SM

MBASS 462.245 39.402 1.000 MVA 4913.883 2212.939 -0.033 1.000 ROA 0.055 0.064 -0.095 0.165** 1.000 ~" ROE 0.170 0.152 -0.066 0.161* 0.708** 1.000 SALES 18518.398 10124.786 -0.030 0.404** -0.089 0.060 1.000 NET 1049.636 523.056 -0.037 0.623** 0.268** 0.352** 0.779** 1.000 SIC 14.914 37.397 0.052 -0.101 -0.140* -0.232** -0.028 -0.133* 1.000 BETA 0.371 1.068 -0.190** 0.044 0.048 -0.036 -0.090 -0.015 -0.003 1.000 SIP 1.948 -1.016 0.090 -0.286** 0.086 -0.006 -0.319** -0.299** 0.019 0.055 1.000 SM 2.005 0.842 -0.080 0.244** 0.101 0.105 0.160* 0.189** -0.013 -0.028 -0.154* 1.000 ?

*p < 0.05; **p < 0.01 MBASS = Market-to-BookAssets; MVA = MarketValue Added; ROA = Returnon Assets; ROE = Returnon Equity; SALES = Net Sales (proxy for size); NET = Net E Income(proxy for size); SIC = Industry;BETA = Risk; SIP = Social Issue Participation;SM= StakeholderManagement

t 5/ 134 A. J. Hillman and G. D. Keim

Table 2. Regression results for market value-added three cases yielded no significant results for our (MVA 95-96): stakeholder management independent two variables of interest: SM and SIP. Thus, variable the findings using these more accounting-based Variables measures of firm financial performance are not consistent with those using MVA. SM 94 0.128** (124.397) Finally, there has been some precedent set in Sales 94 -0.202* (0.021) the literature for examining the individual dimen- Net Income 94 0.758** (0.376) sions of KLD as they pertain to CSP. Given this, Industry 94 -0.007 (16.567) we also conducted our with MVA as Risk 94 0.041 analysis (660.363) our variable the five dimensions Intercept 146.757 (990.972) dependent using R2 0.426 of our stakeholder management (SM) variable Adjusted R2 0.414 entered individually as presented in Table 4. F 35.132** Results of this analysis indicate that the dimension of community relations is the primary driver Standardizedregression coefficients are shown;standard errors of the between MVA and shareholder are in parentheses relationship N = 308 value creation. *p < 0.05; **p < 0.01

DISCUSSION Table 3. Regression results for market value-added (MVA 95-96): social issue participation independent We have that a more variable argued fine-grained approach to studying the relationship between Variables corporate social performance and financial performance is important because of differences SIP 94 -0.127* (132.538) underlying two dimensions of CSP: stakeholder Sales 94 -0.215** (0.21) management and social issue participation. Our Net income 94 0.762** (0.376) results using MVA as a measure of shareholder Industry 94 -0.005 (16.589) wealth creation indicate a Risk 94 0.043 positive relationship (661.510) with stakeholder and a Intercept 104.044 (994.786) management negative R2 0.424 relationship with social issue participation. Our Adjusted R2 0.412 F 34.904** Table 4. Regression results for market value-added Standardizedregression coefficients are shown;standard errors (MVA 95-96): five individualcategories of stakeholder are in parentheses managementindependent variables N = 308 *p < 0.05; **p < 0.01 Variables

PRD (Product) -0.074 (351.574) corporate social performance and financial per- ENV (Environment) -0.046 (293.524) formance in the have used more traditional ER (Employeerelations) 0.019 (276.568) past DIV 0.046 (305.401) measures. In order to frame this (Diversity) accounting-based COM (Community) 0.225** (378.267) study in the context of the existing literature Sales 94 -0.196* (0.021) and to test for the sensitivity of our results to Net Income 94 0.689** (0.373) 0.012 performance measure, we also ran our analyses Industry 94 (16.265) Risk 94 0.056 (643.943) using three additional variables often used to Intercept -800.273 (1007.584) measure financial performance: Return on Assets R2 0.474 (ROA), Return on Equity (ROE), and the ratio Adjusted R2 0.454 of Market to Book Assets (often called the Q F 23.359** ratio as it approximates Tobin's Q). Table 1 also Standardized coefficientsare shown;standard errors indicates the descriptive statistics and correlation regression are in parentheses of these additional dependent variables to our N = 308 predictor variables. Regression analyses in all *p < 0.05; **p < 0.01

Copyright ? 2001 John Wiley & Sons, Ltd. Strat. Mgmt. J., 22: 125-139 (2001) Shareholders, Stakeholders and Social Issue 135 results also indicatethat the directionof causality sample scored negatively for the dimension of is from stakeholdermanagement/social issue par- CommunityRelations, where firms with negative ticipation to shareholder wealth creation/ scores capture 16.9 percent of Diversity Issues, destruction. Additional analyses support this 20.9 percent of Product Issues, 21.7 percent of directionalcausality in that the reverse causality EmployeeRelations and 29.1 percentof Environ- is not statisticallysupported. Thus, our findings mental Issues. This skewness may also have an are consistentwith our theoreticallybased predic- effect on our results. Therefore,we have reason tions that stakeholdermanagement can lead to to believe that the findings in our additional shareholderwealth creationand that participation disaggregationmay be a result of the data rather in social issues does not lead to shareholder than an indication that only one dimension of wealth creation. stakeholdermanagement is positively related to Our results, however, should be interpreted shareholderwealth creation.These analyses,how- with caution.Additional analyses using alternative ever, also indicatepromise for furtherresearch in measures of financial performance,ROA, ROE this area. and Market-to-BookAssets, are not significant. As discussedin the Methodssection, we strongly believe that this is a result of the problems IMPLICATIONS AND CONCLUSION associated with these operationalizations,rather than an indicationof lack of robustnessof our Business firms face an increasinglycompetitive findings.Conceptually, MVA is the closest oper- environment.The developmentof a world market ationalization available to us to capture our for investmentcapital, in particular,increases the dependentvariable of interest:shareholder wealth importanceof competing for investment capital. creation. However, this is an area for future Such increasedcompetition, we believe, encour- research. ages firms to searchfor sources of organizational Finally,while our fundamentalargument is that advantagethat cannot be easily or quickly dupli- CSP is multidimensionaland that disaggregation cated in order to continue to attractinvestment is necessaryto betterunderstand the relationships capital. Sustainableorganizational advantage may studied herein, our additionalanalysis also indi- be built with tacit assets that derive from cates promise in disaggregatingstakeholder man- developing relationshipswith key stakeholders: agementeven furtherinto individualcomponents. customers,employees, suppliersand communities Unfortunately,while the KLD data are the best where businesses operate. availableto researchersstudying corporate social Implicationsof our researchare that investing performance,these data have unique issues in in stakeholdermanagement may be complemen- their constructionand aggregationthat cannot be tary to shareholdervalue creationand may indeed disentangledhere. For example,while community provide a basis for competitive advantage as relationsis the only positive and significanteffect importantresources and capabilitiesmay be cre- found in our regression analyses, employee ated that differentiatea firm from competitors. relationsand diversityissues are also significantly On the other hand, participatingin social issues correlatedwith MVA. Interestingly,product is- may be seen at best as a transactionalinvestment sues and environmentalissues have an insignifi- easily copied by competitors. cant but negative relationship.This may be a We think these findingshelp shed light on the resultof the actualcomposition of the dimensions dilemmafaced by managerswhen called upon to tracked by KLD. The Appendix shows that the serve an expandedrole in society. Our findings dimensionsof communityrelations and diversity suggest that if the activity is directly tied to issues track more 'areas of strength'than 'areas primarystakeholders, then investmentsmay bene- of concern.' The other three dimensions have a fit not only stakeholders but also result in more equal balance between strengthsand weak- increased shareholder wealth. Participating in nesses. How the individual items within each social issues beyond the direct stakeholders,how- category are summed to form a score for each ever, may adversely affect a firm's ability to category is undisclosed by KLD. In addition, a create shareholderwealth. frequency analysis of the individual dimensions We are not making the normative assertion indicatesthat only 1.2 percentof the firms in our that firms should not engage in such activities.

Indeed, many firms have multidimensional per- inquiry. Understanding how stakeholder demands formance goals that may include social issue may differ and how managers prioritize each activism. However, the conflict between these would be a valuable area of future research. Are goals and shareholder wealth creation should be resources devoted to stakeholder relations subject recognized. The use of a firm's resources always to diminishing returns? If so, questions about has an opportunity cost. Implementing a social marginal returns and optimal levels of investment issue participation strategy appears to come at should be addressed. In addition, the motivation the cost of forgone opportunities to increase behind social issue participation and the effects shareholder value. of such on the organization beyond shareholder Moran and Ghoshal argue for a reorientation wealth represents a gap in our understanding of of business strategy 'to reflect the fact that what social issues. We hope these results will spur is good for society does not necessarily have to further research on these and other related issues. be bad for the firm, and what is good for the firm does not have to come at a necessarily REFERENCES cost to society' (Moran and Ghoshal, 1996: 45). Consistent with this view, the emphasis on share- Amit R, SchoemakerP. 1993. Strategic assets and holder value creation today should not be con- organizationalrent. Strategic ManagementJournal strued as coming at the expense of the interests 14(1): 33-45. of other stakeholders. Atkinson A, WaterhouseJ, Wells R. 1997. A stake- primary Participation by holder to measure- firms in all the social issues that on the approach strategicperformance beckon, ment. Sloan ManagementReview Spring: 25-37. other hand, may not lead to the same competitive AupperleK, CarrollA, HatfieldJ. 1985. An empirical value creation prospects as stakeholder man- examinationof the relationshipbetween corporate agement. social responsibilityand profitability.Academy of In our ManagementJournal 28: 446-463. addition, findings may provide insight J. 1986. culture: can it be a into the of between social Barney Organizational pattern relationship sourceof sustainedcompetitive advantage? Academy performance and financial performance in past of ManagementReview 11: 656-665. literature. Evidence here suggests the two dimen- BarneyJ. 1991. Firm resourcesand sustainedcompeti- sions of corporate social performance-stake- tive advantage.Journal of Management17: 99-120. holder and social issue BarneyJ, HansenM. 1994. Trustworthinessas a source management partici- of to financial competitive advantage. Strategic Management pation-have opposing relationships Journal,Winter Special Issue 15: 175-190. performance. This may partially explain why Bentson G. 1982. Accountingnumbers and economic aggregating the two together into a measure of values. AntitrustBulletin Spring: 161-215. corporate social performance may lead to ambigu- BirchardB. 1995. How many masterscan you serve? CFO 11: 48-53. ous results. Furthermore, as noted in our Methods Briloff R. 1972. UnaccountableAccounting. Harper & section, our operationalization of financial per- Row: New York. formance using market value added may be an BriloffR. 1976. The Truthabout CorporateAccounting. improvement over accounting measures of return Harper& Row: New York. in understanding the effect of intangible assets CarrollA. 1979. A three-dimensionalconceptual model such as stakeholder This of corporatesocial performance.Academy of Man- relationships. suggests Review 4: 497-505. that future research extend the agement may decoupling ChakravarthyB. 1986. Measuringstrategic perform- of social performance and further explore the ance. StrategicManagement Journal 7(5): 437-458. differences between the dimensions as well as ClarksonM. 1994. A risk based model of stakeholder reconsidering measures of financial performance. theory:Toronto. Proceedings of the Second Toronto on Stakeholder Centrefor Cor- These findings create other opportunities for Conference Theory, porateSocial Performanceand Ethics, Universityof further research. First and foremost are the Toronto:Toronto. methodological issues discussed in our Discussion ClarksonM. 1995. A stakeholderframework for analyz- section. Further research focusing on alternative ing and evaluating corporate social performance. measures of performance and further disaggre- Academyof ManagementReview 20: 92-117. T, Weston J. 1983. Financial Theory and gation of our constructs is promising. In addition, Copeland CorporatePolicy. Addison-Wesley:Reading, MA. the processes by which stakeholder relations are DierickxI, Cool K. 1989. Asset stock accumulationand managed and the balancing of diverse demands sustainabilityof competitiveadvantage. Management of stakeholder groups is a ripe area for further Science 35: 1504-1511.

Copyright ? 2001 John Wiley & Sons, Ltd. Strat. Mgmt. J., 22: 125-139 (2001) Shareholders, Stakeholders and Social Issue 137

Donaldson T, Preston L. 1995. The stakeholder theory Lippman S, Rumelt R. 1982. Uncertain imitability: an of the corporation: concepts, evidence and impli- analysis of interfirm differences in efficiency under cations. Academy of Management Review 20: 65-91. competition. Bell Journal of Economics 13: 418- Evan W, Freeman E. 1988. A stakeholder theory of the 438. modem corporation: Kantian capitalism. In Ethical Livingstone J, Salamon G. 1971. Relationship between Theory and Business, Beauchamp T, Bowie N (eds.). the accounting and the internal rate of return meas- Prentice-Hall: Englewood Cliffs, NJ; 75-93. ures: a synthesis and analysis. In Income Theory Fisher F. 1979. Diagnosing monopoly. Quarterly and Rate of Return, Livingstone J, Burs T (eds.). Review of Economics and Business 19: 7-33. Ohio State University Press: Columbus, OH; 37-59. Fisher F, McGowan J. 1983. On the misuse of account- Lubatkin M, Shrieves R. 1986. Towards reconciliation ing rates of return to infer monopoly profits. Amer- of market performance measures to strategic man- ican Economic Review 73: 82-97. agement research. Academy of Management Review Freeman E. 1984. Strategic Management: A Stakeholder 11: 497-512. Approach. Pitman: Boston, MA. Martin X, Mitchell W, Swaminathan A. 1995. Recre- Grant R. 1996. Toward a knowledge-based theory of ating and extending Japanese automobile buyer- the firm. Strategic Management Journal, Winter supplier links in N. America. Strategic Management Special Issue 17: 109-122. Journal 16(8): 589-620. Graves S, Waddock S. 1994. Institutional investors and McGuire J, Schneeweis T, Hill J. 1986. An analysis corporate social performance. Academy of Manage- of alternative measures of strategic performance. In ment Journal 37: 1035-1046. Advances in Strategic Management, Vol. 4. Lamb Greenley G, Foxall G. 1996. Consumer and noncon- R, Shrivastava P (eds.). JAI Press: Greenwich, CT; sumer stakeholder orientation in UK companies. 107-153. Journal of Business Research 35: 105-116. McGuire J, Sundgren A, Schneeweis T. 1988. Corporate Greenley G, Foxall G. 1997. Multiple stakeholder orien- social responsibility and firm financial performance. tation in UK companies and the implications for Academy of Management Journal 31: 854-872. company performance. Journal of Management Stud- Mitchell R, Agle B, Wood D. 1997. Toward a theory ies 34: 259-284. of stakeholder identification and salience: defining Griffin J, Mahon J. 1997. The corporate social perform- the principle of who and what really counts. Acad- ance and corporate financial performance debate: 25 emy of Management Review 22: 853-886. years of incomparable research. Business and Society Moran P, Ghoshal S. 1996. Value creation by firms. 36: 5-31. Best Paper Proceedings, Academy of Management Harrison J, St. John C. 1994. Strategic Management of Annual Meeting, Cincinnati, OH. Organizations and Stakeholders. West Publishing: Mudambi R, Helper S. 1998. The 'close but adversarial' St Paul, MN. model of supplier relations in the U.S. auto industry. Harrison J, St. John C. 1996. Managing and partnering Strategic Management Journal 19(8): 775-792. with external stakeholders. Academy of Management Nahapiet N, Ghoshal S. 1998. Social capital, intellectual Executive 10: 46-60. capital, and the organizational advantage. Academy Hart S. 1995. A natural-resource-basedview of the firm. of Management Journal 19: 775-792. Academy of Management Review 20: 986-1014. Nayyar P. 1995. Stock market reactions to customer Hayes R, Aberathy W. 1980. Managing our way to service changes. Strategic Management Journal economic decline. Harvard Business Review 58(4): 16(1): 39-53. 67-77. Oliver C. 1988. The collective strategy framework: an Hill C, Jones T. 1992. Stakeholder-agency theory. Jour- application to competing predictions of isomorphism. nal of Management Studies 29: 131-154. Administrative Science Quarterly 33: 543-561. Itami H. 1987. Mobilizing Invisible Assets. Harvard Ouchi W. 1980. Markets, bureaucracies, and clans. University Press: Cambridge, MA. Administrative Science Quarterly 25: 121-141. Jones T. 1995. Instrumental stakeholder theory: a syn- Pava M, Krausz J. 1996. The association between thesis of ethics and economics. Academy of Manage- corporate social responsibility and financial perform- ment Review 20: 404-437. ance: the paradox of social cost. Journal of Business Kaplan R, Norton D. 1996. The Balanced Scorecard. Ethics 15: 321-357. Harvard Business School Press: Boston, MA. Penrose E. 1959. The Theory of the Growth of the Kotter J, Heskett J. 1992. Corporate Culture and Per- Firm. Wiley: New York. formance. Free Press: New York. Pfeffer J. 1998. The Human Equation: Building Profits Lado A, Wilson M. 1994. Human resource systems by Putting People First. Harvard Business School and sustained competitive advantage: a competency Press: Boston, MA. based perspective. Academy of Management Review Prahalad CK. 1997. Corporate governance or corporate 19: 699-727. value added? Rethinking the primacy of shareholder Legnick-Hall CA. 1996. Customer contributions to value. In Studies in International Corporate Finance quality: a different view of the customer-oriented and Governance Systems, Chew D (ed.). Oxford firm. Academy of Management Review 21: 791-824. University Press: New York; 46-56. Leonard D. 1995. Wellsprings of Knowledge: Building Preston L, Sapienza H. 1990. Stakeholder management and Sustaining the Sources of Innovation. Harvard and corporate performance. Journal of Behavioral Business School Press: Boston, MA. Economics 19: 361-375.

Rao H. 1994. The social construction of reputation: Accounting Review 53: 112-133. certification contests, legitimation and the survival of Watts R, Zimmerman J. 1990. Positive accounting organizations in the American automobile industry: theory: a ten year perspective. Accounting Review 1895-1912. Strategic Management Journal, Winter 65: 131-156. Special Issue 15: 29-44. Wernerfelt B. 1984. A resource-based view of the firm. Rappaport A. 1992. CFOs and strategists: forging a Strategic Management Journal 5(2): 171-180. common framework. Harvard Business Review Wood D. 1991. Corporate social performance revisited. 70(3): 84-91. Academy of Management Review 16: 691-718. Ring PS, Van de Ven A. 1992. Structuring cooperative relationships between organizations. Strategic Man- 483-498. agement Journal 13(7): APPENDIX: KLD MEASUREMENT Ring PS, Van de Ven A. 1994. Developmental processes of cooperative interorganizationalrelationships. SCREENS Academy of Management Review 19: 90-118. Ruf B, Muralidhar K, Paul K. 1993. Eight dimensions Items used for stakeholder management (SM): of corporate social performance: determination of relations. Areas concern. Fines relative the Community of importance using analytic hierarchy or civil or involvement in process. Best Papers Proceedings, Academy of Man- penalties paid, major agement Annual Meeting, Atlanta, GA. litigation or controversies, relating to a com- Sharfman M. 1996. The construct validity of the munity in which it operates; the company's Kinder, Lydenberg & Domini social performance relations with a community in which it operates data. Journal Business Ethics 15: 287-296. ratings of have become strained due to a recent plant clos- Sharma S, H. 1998. Proactive Vredenburg corporate or breach of with the environmental strategy and the development of coming general agreements petitively valuable organizational capabilities. Stra- community. tegic Management Journal 19(8): 729-754. Areas of strength. The company in recent years Solomon J. 1970. Alternative rate of return concepts has consistently given over 1.5 percent of pretax and their implications for utility regulation. Bell earnings to charity or otherwise demonstrated Journal of Economics 1: 65-81. the is known for inno- J-C. 1996. the basis of a generous giving; company Spender Making knowledge vative such as of dynamic theory of the firm. Strategic Management giving, support nonprofit agen- Journal, Winter Special Issue 17: 45-62. cies promoting self-sufficiency among the eco- Starik M. 1995. Should trees have managerial standing? nomically disadvantaged; the company supports Toward stakeholder status for non-human nature. education through a long-term commitment to Journal of Business Ethics 14: 204-217. at the or Stem Stewart. 1996. The Stern Stewart improve programs primary secondary Performance or the is a recent 1000. Ster Stewart Management Services: New level, company prominent sup- York. porter of job training programs; prominent partici- Swanson D. 1995. Addressing a theoretical problem by pation in public/private initiatives that support reorienting the CSP model. Academy of Management housing initiatives for the economically disadvan- Review 20: 43-64. Teece D. 1998. value from assets: taged. Capturing knowledge relations. Areas concern. The the new markets for know-how, and intan- Employee of economy, has relations with its unions rela- gible assets. California Management Review 40(3): company poor 55-79. tive to others in its industry; the company has Tobin J, Brainard W. 1968. Pitfalls in financial model had recent lay-offs of more than 15 percent of building. American Economic Review 58: 99-122. employees in 1 year or 25 percent of employees Ullman A. 1985. Data in search of a theory: a critical in 2 the has fines examination of the social years; company paid significant relationships among per- or over or been formance, social disclosure, and economic perform- penalties employee safety ance of U.S. firms. Academy of Management Review involved in major safety controversies; the com- 10: 540-557. pany has a substantially underfunded pension plan Waddock S, Graves S. 1997a. The corporate social or an inadequate benefits plan. performance-financial performance link. Strategic Areas The has Journal 303-319. of strength. company strong Management 18(4): union relations relative to others in its Waddock S, Graves S. 1997b. Quality of management industry; and quality of stakeholder relations. Business & the company has maintained a long-term policy Society 36: 250-279. of company-wide cash profit sharing; the com- Wartick S, Cochran P. 1985. The evolution of the pany has a substantial sense of worker corporate social performance model. Academy of of financial infor- Review 10: 758-769. involvement/ownership, sharing Management mation with or Watts R, Zimmerman J. 1978. Towards a positive employees, employee participation theory of determination of accounting standards. in management decision making; the company

Copyright ? 2001 John Wiley & Sons, Ltd. Strat. Mgmt. J., 22: 125-139 (2001) Shareholders, Stakeholders and Social Issue 139 offers its employees strongretirement benefits, or emissions of toxic chemicalsin the United States; other innovative benefits, relative to others in the company is among the top producers of its industry. ozone-depletingchemicals; the company's legal Diversity issues. Areas of concern. The com- emissions of toxic chemicals into the air and pany has paid substantialfines or penalties or water are among the highest of the companies been involved in major controversiesrelated to followed by KLD; the company is one of the its affirmativeaction record. largest producersof agriculturalchemicals. Areas of strength. The company's CEO is a Areas of strength. The company has policies woman or memberof a minoritygroup; the com- to reduce emissions throughelimination of toxic pany has made notableprogress in the promotion chemicals; the company is a substantialuser of of women and minorities, particularlyto line recycled materials; the company's environmen- positions; women and/or minorities hold board tally sensitive property,plant, and equipmentis seats in the company;the companyhas outstand- among the most superior environmentally;the ing benefitprograms addressing work/family con- company derives substantial revenues from cerns; the company has a strong and consistent developing, using, or marketing fuels with recordof supportfor women- and minority-owned environmental advantages, or has undertaken businesses (purchasingfrom or investing in); the notable energy conservationprojects; the com- company has taken innovative hiring initiatives pany derives substantialrevenues from alternative or other human resource programs directed at fuels including natural gas, wind power, and employmentof the disabled. solar energy. Product issues. Areas of concern. The com- Items used for social issue participation pany faces major recent product safety contro- (SIP) versies; the company faces a major marketing Non-U.S. Issues. Areas of concern. Operations controversyor has paid fines or penaltiesrelated in Burma;operations in Mexico are controversial to advertising practices, consumer fraud, or especially related to employees or the environ- governmentcontracting practices. ment. Areas of strength. The company has a long- Areas of strength. The company has established standing company-widequality programjudged a substantial,innovative charitable giving pro- to be amongthe best in the industry;the company gram outside of the United States. is an industryleader in researchand development, Other. Areas of concern. Notably high levels as evidenced by expenditureas a percentageof of compensationto top managementor board, sales, effective new product development, or companyis involved in tax disputes,the company unusualinventiveness; part of the company'sba- owns a substantialportion of a company with sic mission is provision of products or services social concerns. for the economicallydisadvantaged. Areas of strength. Company has notably low Environment issues. Areas of concern. The compensationfor top managementor board,com- company's liabilities for hazardous waste sites pany owns a substantialpart of a company with exceed $30 million or the companyhas significant social strengths. involvement in more than 30 federal Superfund Exclusionary screens. sites; the company has recently paid significant Alcohol/tobacco/gambling. fines or penalties, has a pattern of regulatory Military weapons contracting or supplies to problemsor has been involved in major contro- Departmentof Defense. versies involving environmentaldegradation; the Nuclearpower electricalutility, designs or con- company'semissions are among the highest legal structsnuclear energy plants or uranium.