Reading articles and Answering questions based on these articles

ngotdoan

Heniszetal.2014.pdf

Home >Social Science homework help >Reading articles and Answering questions based on these articles

Strategic Management Journal Strat. Mgmt. J., 35: 1727 – 1748 (2014)

Published online EarlyView 20 September 2013 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/smj.2180

Received 1 July 2011 ; Final revision received 16 July 2013

SPINNING GOLD: THE FINANCIAL RETURNS TO STAKEHOLDER ENGAGEMENT

WITOLD J. HENISZ,1* SINZIANA DOROBANTU,2 and LITE J. NARTEY3 1 Management Department, The Wharton School, University of Pennsylvania, Philadelphia, Pennsylvania, U.S.A. 2 Management & Organizations, Leonard N. Stern School of Business, New York University, New York, New York, U.S.A. 3 Sonoco International Business Department, Darla Moore School of Business, University of South Carolina, Columbia, South Carolina, U.S.A.

We provide direct empirical evidence in support of instrumental stakeholder theory’s argument that increasing stakeholder support enhances the financial valuation of a firm, holding constant the objective valuation of the physical assets under its control. We undertake this analysis using panel data on 26 gold mines owned by 19 publicly traded firms over the period 1993 – 2008. We code over 50,000 stakeholder events from media reports to develop an index of the degree of stakeholder conflict/cooperation for these mines. By incorporating this index in a market capitalization analysis, we reduce the discount placed by financial markets on the net present value of the physical assets controlled by these firms from 72 percent to between 37 and 13 percent. Copyright © 2013 John Wiley & Sons, Ltd.

INTRODUCTION

Corporations increasingly allocate resources to buttress the political and social support for their operations. The theoretical and empirical litera- ture examining the returns to such investments has, however, become increasingly equivocal over time. Theoretical models drawing on neoclas- sical economics and institutional theory high- light rationales to be suspicious of such activ- ity. Empirical evidence offers at best a marginal and contingent positive relationship between such efforts and financial performance. We seek to return the focus of the theoretical literature to its roots in instrumental stakeholder theory (Clarkson,

Keywords: political risk; instrumental stakeholder theory; corporate social responsibility; international business; shareholder value *Correspondence to: Witold J. Henisz, Management Depart- ment, The Wharton School, University of Pennsylvania, 3620 Locust Walk, Philadelphia, PA 19104-6370, U.S.A. E-mail: [email protected]

1995; Donaldson and Preston, 1995; Jones, 1995) by demonstrating dramatic positive returns to stakeholder support using a novel and broadly applicable event-based measure of stakeholder conflict/cooperation and an empirical specification that better conforms to the precepts of that theory.

We argue that efforts to enhance support from external stakeholders should not be conceived in terms of the distribution of rents among direct factors of production (Alchian and Demsetz, 1972; Jensen and Meckling, 1976) but rather as investments in political and social capital that alter the behavior of a wider range of stakeholders. Such investments reduce opportunistic hold-up by stakeholders with whom the firm has no explicit buyer or supplier contracts but whose cooperation is nevertheless required in order for the firm to create and capture value (Baker, Gibbons, and Murphy, 2002). Such investments can alter these stakeholders’ behavior in a manner that enhances the probability that a business plan will proceed on schedule and on budget and, ultimately, generate

1728 W. J. Henisz, S. Dorobantu, and L. J. Nartey

sustainable shareholder value (Freeman, 2010; Zingales, 2000).

Our theoretical arguments refocus attention on an important question for instrumental stake- holder theory — Does stakeholder support improve financial and operational performance? — and are examined with empirical metrics that closely match that theory. In contrast, the focus of recent theoretical work has been on narrower conceptions of the underlying mechanisms, while the growing body of empirical studies testing these mechanisms has relied on data whose limitations preclude a comparison of narrower and broader conceptions of stakeholder relations. Without underplaying the contributions provided by these theoretical and empirical works, we believe that our collective understanding of the value of stakeholder engage- ment can be enhanced by a straightforward analy- sis of the relationship between stakeholder support broadly defined and firm value on the basis of data that directly and objectively capture both sides of this relationship.

Our study draws inspiration from the path taken by two closely related literatures that have bene- fitted from a reconsideration of the key questions and concepts as more appropriate data and empir- ical tools became available. First, in the literature examining the financial impact of environmental management, much of the original focus was on distinguishing between competing causal mech- anisms underlying this relationship, including a possible trade-off between competing economic and environmental performance goals; increased economic performance due to higher morale or productivity associated with better environmental performance; or increased economic performance due to positive externalities to other aspects of the firm’s operations (McGuire, Sundgren, and Schneeweis, 1988). Sharfman & Fernando (2008) extended the literature by highlighting that, in addition to improving resource efficiency, environ- mental management can reduce risk, including that of detrimental action by external stakeholders. As this literature developed objective data on environ- mental compliance and embraced broader concep- tions of economic and environmental performance that allowed for both direct and indirect effects to be captured in their relationship, the consen- sus that environmental management may be pos- itively related to financial performance strength- ened (Dowell, Hart, and Yeung, 2000; King and Lenox, 2001).

Moreover, subsequent research also highlighted the sensitivity of these results to the actions of environmental pressure groups (Binder and Neumayer, 2005; Epstein and Schnietz, 2002; Maxwell, Lyon, and Hackett, 2000), a subset of the stakeholders considered in this paper. To the extent environmental performance enters into the decisions of some stakeholders as to whether to support or disrupt a firm’s operations, this effect would be captured in our study, which broadens the theoretical argument to incorporate all stakeholders regardless of the issue or “stake” that ties them to the firm.

Second, the evolution of the literatures exam- ining the financial returns to corporate reputation or status also offers important insights. In that case, the development of objective data on cor- porate reputation (Fombrun and Shanley, 1990) and methods to value intangible assets (Villalonga, 2004), likewise proved critical in fostering the development of the literature beyond its foun- dations (Weigelt and Camerer, 1988). However, researchers struggled to separate the causal impact of reputation or status on performance from the reverse channel. In order to address this challenge, one group of papers relied upon particular industry settings in which independent and objective indica- tors of reputation or status could be operationalized and shown to be independent of prior financial per- formance (Rao, 1994; Stuart, 1999) while a second focused on the contingent impact of reputation or status under certain contingencies or environmen- tal shocks (Roberts and Dowling, 2002). In order to confront the same challenge of causal inference in the stakeholder relations literature (Baron, Har- joto, and Jo, 2011; Margolis, Elfenbein, and Walsh, 2007), we take the former approach in this paper and leave the second for subsequent research.

Our empirical analysis is set in the gold min- ing sector where stakeholder conflict resulting from the social, political, and environmental con- sequences of mining — including the relocation of communities, increased corruption and private rent seeking, and increased pollution and envi- ronmental risks — has led to costly delays and disruptions in project development and execu- tion. The resulting performance shortfalls even- tually triggered corrective investments in stake- holder engagement strategies. The growing empha- sis on political and social support in the industry is affecting even small mining companies, who, moti- vated by their desire to sell their operations to the

Financial Returns to Stakeholder Engagement 1729

majors, are increasingly acknowledging their own need to obtain a “social license” (Boutilier, 2009, 2011; Chiu and Sharfman, 2011; Lassonde, 2003; Thomson and Boutilier, 2011).

“It used to be the case that the value of a gold mine was based on three variables: the amount of gold in the ground, the cost of extraction, and the world price of gold. Today, I can show you two mines identical on these three variables that differ in their valuation by an order of magnitude. Why? Because one has local support and the other doesn’t.” (Yani Roditis, former COO Gabriel Resources, interview by authors)

Our empirical design follows directly from this observation. We link data on financial market valuation to the intrinsic value of the gold mine and demonstrate that the degree of stakeholder conflict/cooperation helps explain the gap or difference between these figures. We undertake this analysis using panel data on 26 gold mines over the period 1993 – 2008.

We manually code over 50,000 stakeholder events from the population of media reports cov- ering these mines. Our sentence-level coding pro- tocol identifies the population of media relevant stakeholders initiating an action or expressing a sentiment as well as the target of that action or statement. It codes the action or expression accord- ing to a scale measuring the degree of conflict or cooperation among political and social actors that has been developed in the international relations and conflict studies literature. We aggregate this time-varying network of collaborative and con- flictual relations into a single time-varying metric of political and social support for the mine and demonstrate that this metric is an important com- ponent, together with characteristics of the mine and the price of gold, in calculating the financial market valuation of the 19 publicly traded par- ent firms. Specifically, by incorporating this met- ric in a market capitalization analysis that also includes macro-level constraints on policy change, we reduce the discount placed by financial markets on the net present value of the gold controlled by these firms from 72 to between 37 and 13 percent.

Our results support the argument that the abil- ity to enhance support from external stakeholders is a key driver of financial performance in that

investors may discount the value of a firm’s assets (and thus the value of the firm) based on the tenor of that firm’s relations with its external stake- holders. This empirical result serves as a bridge between instrumental stakeholder theory, corpo- rate social responsibility, corporate reputation or status, and strategic management. We offer instru- mental stakeholder theorists and social respon- sibility advocates powerful evidence to combat recent skepticism and encourage theoretical exten- sions to the theory of the firm that encompass a broader range of implicit and relational con- tracts. We also provide a novel methodology that allows for the subsequent analysis of the strate- gic determinants of improvements in stakeholder cooperation, corporate reputation and status. We argue that our approach and methodology is gen- eralizable beyond gold mining and natural resource extraction to a wide array of corporate activity, and discuss in the conclusion the next steps in an inter- disciplinary research agenda designed to enhance the theoretical and empirical support for the financial returns to stakeholder engagement.

THE VALUATION OF STAKEHOLDER MANAGEMENT AND THEORIES OF THE FIRM

Multiple theoretical models explore mechanisms that link managerial efforts to build stakeholder support to financial performance. Models focused on principal – agent relationships highlight that managers may seek to improve relationships with stakeholders in order to enhance their own qual- ity of life or their reputation and do so at the expense of shareholders (Jensen, 2002). A second group of models suggest that under certain mar- ket conditions improved relationships with stake- holders benefit shareholders because managerial decisions to alter the nature of the production process or of the product in order to gain stake- holder support increase consumers’ willingness to pay and/or reduce suppliers’ reservation price (McWilliams and Siegel, 2001). A final group of models incorporate strategic activists and suggest that the extent to which stakeholder engagement correlates with financial performance depends on the characteristics of the activists and the firm’s response (Baron, 2001, 2009). We review these models below and highlight that, while theoret- ical extensions to incorporate a broader set of

1730 W. J. Henisz, S. Dorobantu, and L. J. Nartey

stakeholders are warranted by increased man- agerial commitment to stakeholder engagement, we lack both clear empirical evidence regarding the magnitude of the potential returns available through such efforts and the data to test the con- tingencies that are beginning to emerge from these models. We seek to provide both.

Scholars skeptical of efforts to engage stake- holders or seeking to explain negative empirical associations between such efforts and shareholder value have argued that managers use relationships with external stakeholders to pursue self-interest seeking perquisites, career enhancement, or moral peace of mind. Levitt (1958) and Friedman (1962, 1970) famously attack efforts by managers to pursue objectives other than shareholder value maximization as short-term conflict avoidance by managers. They advocate instead a single-minded all-out focus on profit maximization (see Sun- daram and Inkpen (2004) for a literature review of the shareholder value debate). Jensen (2002) models shareholder principals’ loss of control over managerial agents who may seek to pursue per- sonal social interests (Hemingway and Maclagan, 2004) in the presence of multiple hard-to-quantify performance metrics. Authors have also empha- sized the possibility of collusion between man- agers and either institutional shareholders (Cespa and Cestone, 2007) or nonshareholding stake- holders (Surroca and Tribó, 2008) in support of managerial tenure and social responsibility at the expense of (noninstitutional) shareholder returns. Institutional theorists have further highlighted the pressures for managerial conformity that can arise from regulation, peer behavior and civil society independent of the efficiency of adoption (Camp- bell, 2007; Jennings and Zandbergen, 1995; Mar- golis and Walsh, 2003; Marquis, Glynn, and Davis, 2007).

Empirical research supporting these mecha- nisms include studies showing that financial slack is a determinant of corporate social performance (McGuire, Sundgren, and Schneeweis, 1988; Wad- dock and Graves, 1997); headquarter policies rather than local conditions drive social respon- sibility programs in foreign subsidiaries (Husted and Allen, 2006); shareholder activism reallocates discretionary resources away from corporate social performance (David, Bloom, and Hillman, 2007); and insider ownership and leverage (i.e., factors that increase managerial alignment with share- holder interests) are negatively associated with

corporate social responsibility (Barnea and Rubin, 2006). Baron, Harjoto, and Jo (2011) find that (1) social pressure diverts resources away from corporate financial performance toward corporate social performance, (2) social performance has no net effect on financial performance (though further analysis shows that the result is positive in con- sumer goods and negative in industrial industries) and, (3) social pressure is directed at firms with rel- atively high levels of publicly stated commitments to social responsibility (i.e., soft targets).

Scholars supportive of efforts to engage stake- holders or seeking to explain a positive empirical relationship between such efforts and shareholder value shift the focus of the theoretical argument from the costs of ex post managerial discretion to external stakeholders’ influence over ex ante managerial strategy with respect to members of the value chain (see Kitzmueller and Shimshack (2012) for a literature review). This literature seeks to formalize elements of Freeman’s (1984, 2010) stakeholder approach to strategic manage- ment, which emphasizes how the preferences and objectives of the myriad actors with a political, economic, or social stake in the operations should be incorporated within strategy making.

These insights are corroborated empirically by Hillman and Keim (2001) who find that stake- holder management contributes to shareholder value while participation in social issues that are less directly linked to the preferences of primary stakeholders decreases it. More recently, Barnett and Salomon (2006) show a curvilinear relation- ship between social and financial performance, with firms at both ends of the social responsi- bility spectrum exhibiting higher financial per- formance than firms in the middle and argue in subsequent work (Barnett and Salomon, 2012) that firms benefit unevenly from investments in social responsibility because they vary in their capacity to influence stakeholders (Barnett, 2007).

Mackey, Mackey, and Barney (2007) advance the theoretical work by incorporating the sup- ply and demand for corporate social responsi- bility activities into a contemporaneous model of corporate valuation. Several formal accounts analyze the supply consequences for corporate social responsibility activities by publicly traded companies and not-for-profit entities of demand competition from stakeholders for charitable con- tributions and consider the implications for con- sumer welfare, the level of public regulation,

Financial Returns to Stakeholder Engagement 1731

and the relationship between corporate social per- formance and financial performance (Besley and Ghatak, 2007; Kotchen, 2006; Zivin and Small, 2005). If combined with differentiated marketing or regulations that place followers at a disad- vantage, such activity could itself generate eco- nomic rents. Moreover, the financial benefits to such activities need not accrue contemporane- ously. One line of theoretical (Godfrey, 2005) and empirical (Godfrey, Merrill, and Hansen, 2009) work posits and demonstrates that corporate social responsibility can contribute to moral or reputa- tional capital that insulates the firm from nega- tive consequences in the event of future adverse shocks.

A growing body of literature in organizational economics seeks to model and identify empirically the mechanisms by which stakeholder relations might impact shareholder value by highlighting the impact of (stakeholder perceptions of) corporate social responsibility on the supply or price of fac- tors of production (McWilliams and Siegel, 2001) and the extent of shared value creation (Porter and Kramer, 2011). Moral motivations may alter the behavior of numerous stakeholders (Brekke, Kverndokk, and Nyborg, 2003), including con- sumers who may be willing to pay more for a product or service perceived as socially responsi- ble (Arora and Gangopadhyay, 1995; Casadesus Masanell et al., 2009; Elfenbein and McManus, 2010). As a result, where signaling of producer type is more important or effective (McWilliams and Siegel, 2001) or where such consumers are more powerful (Hoepner, Yu, and Ferguson, 2010) or markets vary in their competitiveness (Bagnoli and Watts, 2003; Fisman, Heal, and Nair, 2006) companies should exhibit greater corporate social responsibility.

Similarly, employees may prefer to work for a company they perceive to be socially respon- sible, demand lower wages or benefits, or exert greater effort (Besley and Ghatak, 2005; Bhat- tacharya, Sen, and Korschun, 2008; Greening and Turban, 2000; Kim et al., 2010; Turban and Green- ing, 1997). Suppliers of other factors of produc- tion could make similar choices influencing the cost of capital or production (Mackey, Mackey, and Barney, 2007; Porter and Kramer, 2006; Wad- dock, 2000). Activists themselves can be consid- ered stakeholders whose preferences, strategies, or resources can influence corporate behavior (Baron, 2001, 2009; Baron and Diermeier, 2007; Frooman,

1999; Rehbein, Waddock, and Graves, 2004; Rowley and Berman, 2000).

In this work, efforts at generating competi- tive advantages from key stakeholder relations can generate sustained rents if corporate governance is strong (Shahzad, David, and Sharfman, 2011), customer switching is costly due to the develop- ment of relation-based trust (Du, Bhattacharya, and Sen, 2007; Lacey, 2007; Mohr and Webb, 2005; Salmones, Crespo, and Bosque, 2005), or regula- tion emanating from governments (after lobbying) or industry leaders (Frynas, 2008, 2010) allows first movers to monetize their advantage. The choice as to whether to engage or confront stake- holders is itself a complex function of competi- tor strategies and stakeholders’ choices between engagement and confrontation (Baron, Harjoto, and Jo, 2011).

While these extensions of agency models of managerial discretion draw attention to the cir- cumstances under which the returns to stakeholder engagement may be positive, it must be noted that the circumstances are relatively specific. Such findings highlight that while stakeholder engage- ment may pay for shareholders of a subset of firms, it is costly for others and seem to suggest that, on balance, the rhetoric surrounding stake- holder engagement is oversold. We disagree. The emphasis within these formal models (and the sup- porting empirical analysis) on stakeholders in the economic value chain whose main choice is the price charged for inputs, or willingness to pay for outputs, runs counter to the broader precepts of a stakeholder approach. The addition of homoge- nous activists able to increase production costs or reduce consumer willingness to pay is a welcome extension, but we argue for an even broader con- ceptualization of relevant stakeholders and offer an alternative empirical approach to demonstrat- ing the financial impact of broad stakeholder engagement.

For many production processes, an implicit or explicit social license to operate is a necessary, if difficult to specify, input. For example, $15 bil- lion of gold sitting in a mountainside cannot be transformed into shareholder rents with financial, engineering, and marketing inputs alone. It also requires the political and social support of key stakeholders including not only members of the economic value chain but also government offi- cials, regulators, community leaders, and mem- bers of civil society (Henisz and Zelner, 2005).

1732 W. J. Henisz, S. Dorobantu, and L. J. Nartey

These stakeholders may reside locally, nationally, or internationally. As their degree of conflict with the owner and operator of the proposed gold mine increases, they are able either to divert rents from the efficient operation of the mine to their pre- ferred causes or to coordinate public and private activity (e.g., in the government or among unions or activists) to delay the opening of the mine, sus- pend its operations, or raise the cost of continued development or operations to uneconomical levels.

In this perspective, efforts to build stakeholder support are not meant to increase consumer will- ingness to pay for the gold, motivate employees, or extract rents from suppliers. Instead they are made in order to maintain the right to transform the gold and generate rents from that process — a right granted formally and directly by the government but also informally and indirectly by a broader set of external stakeholders (Aden, Kyu-Hong, and Rock, 1999; Liu, 2009). Activities perceived by stakeholders as socially responsible build up polit- ical and social capital that enhances stakeholder cooperation and reduces stakeholder conflict. As a result, the probability that a business plan can pro- ceed on schedule and on budget is enhanced and the financial market discount applied by investors to the objective valuation (i.e., free of stakeholder influence) of the tangible assets is reduced. The resulting empirical prediction is that holding con- stant the objective value of a firm’s assets, its mar- ket valuation increases with stakeholder support .

EMPIRICAL SETTING AND DATA

We conduct our empirical analysis of this hypoth- esis in the gold mining industry due to the unique availability of data that allows for the identification of the financial impact of stakeholder relations and due to the widespread acknowledgement of the critical role of such relations for profitability. An additional strength of this empirical setting is that none of the channels by which the organizational economics literature posits a relationship between corporate social and financial performance are operative. Consumers are unable to differentiate between gold from one mine or company or another. There is little evidence that miners or managers of mining companies are offering wage or productivity benefits to more responsi- ble mining companies. An insufficient number of investors are allocating capital on socially

responsible grounds to influence managerial behavior. Activists, though prominent, are unable to impose substantial direct costs on mining companies. By contrast, mining is an industry where the valuation of a fixed resource (i.e., a gold mine) could vary wildly based on the degree of stakeholder conflict or cooperation, so the indirect costs that activists and other stakeholders could impose may be substantial. Stakeholder relations can influence the regulatory environment, land permitting, environmental regulation, taxation, labor contracts, terms of capital-intensive imports, and the like. The value of a fixed stock of gold is thus plausibly linked to these stakeholders’ preferences on whether the owner should have the right to transform the underground gold into shareholder capital.

Our specific empirical context is the population of 19 publicly traded mining firms on the Toronto Stock Exchange who own and operate one, two, or three mines outside of the United States, Canada, and Australia as listed in the Raw Materials Database1 — a total of 26 mines in 20 countries that have reached the stage of a feasibility study.2

This sampling criterion allows us (1) to draw upon strict Canadian disclosure requirements3 for financial and operating data of mining firms, and (2) to exploit the clear and direct links between media reports on a single (or two or three) mine(s) and changes in financial performance without contamination from numerous other news stories covering other assets or practices of the same publicly traded company.

In order to obtain sufficient information to esti- mate the intrinsic value of the mines using widely accepted resource valuation models, we gathered all relevant financial and operating data on these companies during the period that they were pub- licly listed following the completion of their feasi- bility studies. We combined financial and operating data from the parent company’s quarterly financial reports filed with System for Electronic Document

1 Raw Materials Database — http://www.rmg.se. 2 A feasibility study is an economic study based on sample drilling results and engineering analysis, which presents enough information to determine whether or not the project should be advanced to the construction and production stage. These estimates are intended to be accurate within a 15 percent error band. 3 The requirements of the Toronto Stock Exchange for continuous disclosure are available at http://www.tsx.com/ en/listings/tsx_issuer_resources/continuous_disclosure.html

Financial Returns to Stakeholder Engagement 1733

Analysis and Retrieval (SEDAR) filing system4

and the stock price information from the Toronto Stock Exchange and Yahoo! Finance. These data allow us to evaluate the current market valuation of the firms as a function of the intrinsic valuation of their gold reserves, the degree of stakeholder conflict/cooperation and the credibility of the pol- icy environment in which they operate as detailed below.

Financial market valuation

We use a simple financial valuation formula that models stock market value as a function of the value of a company’s announced resource stock and the likelihood that the company will suc- cessfully extract these resources without exten- sive planning or operational delays. Formally, we estimate

MVit = α + β mi∑

j =1

( RVjit Pjit

) + �it ,

where

MV it refers to the market value of company i at time t , RV jit reflects the current valuation of mine j of company i at time t , m i is the number of mines owned and operated by company i , where i =1, 2 or 3, and P jit is the probability at time t that company i will advance the exploitation of the resources at mine j according to the announced schedule.

We calculate the market value of company i at time t (MV it ) by multiplying the stock price times the number of common shares outstanding and adding company debt. Shares outstanding and debt information are taken from the COMPUSTAT North America database.

Our analysis examines the extent to which fluctuations in the financial value of the companies in our sample can be explained by changes in the value of the gold in the ground or whether the model needs to be broadened to incorporate nonmarket factors that might affect the timely

4 The SEDAR filing system provides access to most public secu- rities documents and information filed by public companies and investment funds with the Canadian Securities Administrators (CSA).

extraction of the resources and that signal to investors that they should discount the value they ascribe to the gold still in the ground. We discuss below the evaluation of the resource value (RV jit ) and the measurement of two factors affecting operations at the mine (P jit ): the level of stakeholder conflict/cooperation and the credibility of the national policy environment.

Resource valuation

We build upon the industry standard resource valuation model developed by Cairns and Davis (1998), who propose a modified formulation of the Hotelling valuation principle for valuing hard-rock mineral properties (Miller and Upton, 1985a, 1985b) that relies on assumptions used by mining engineers when planning the rate of extraction for a mine.5 We apply the Cairns and Davis model to our empirical context by adding a time dimension and considering the possibility that companies in our sample own multiple mines for which the resource value can be assessed separately. Formally, we evaluate each mine j of company i as follows

RVjit = Wjit Sjit 1 − e−rt (Tjit −yjit )

rt ( Tjit − yjit

) , where

W jit = average forecast operating profit per unit of gold of mine j and time t , S jit = quantity of gold reserves of gold mine j at time t , T jit = quarters of estimated mine life for gold mine j at time t ,6

y jit = quarters of operation of gold mine j at time t , r t = Treasury bond yield with term T jit − y jit at time t .

We compute the average forecast oper- ating profit per unit of gold for mine j of

5 The model retains the central parameters of the Hotelling valuation principle (commodity price, cost of production, and mineral reserve data) but allows for the use of average cost data, which is more readily available for different mines. 6 T denotes total estimated mine life, not the remaining mine life at time t. T is indexed by t because companies often adjust their estimates of the total life of a mine to reflect adjustments in reserve estimates or changes in technology or production schedules.

1734 W. J. Henisz, S. Dorobantu, and L. J. Nartey

company i at time t as W jit = 1/(T jit − y jit ) Tjit −yjit∑

k =t

( GoldPricet − CostExtractionjk

) , or the ave-

rage quarterly profit for the remaining period of operation. Following Cairns and Davis (1998), we use the current price of gold to estimate or forecast profits. For each mine, data on the cost of extraction (CostExtraction jk ), the remaining quan- tity of proven and probable gold reserves S jit , the estimated mine life T jit , and the production start date used to calculate the quarters the mine has already been in operation y jit were collected from company annual and quarterly reports, annual information forms, technical reports, and press releases available on the companies’ websites or filed with SEDAR. For the Treasury-bond yield r t we use Treasury constant-maturity data provided by the Federal Reserve Bank of St. Louis and available for different terms from 1 month to 30 years.

We base our analysis on market valuation controlling for resource valuation as opposed to the more familiar ratio of market to book value or an evaluation of Tobin’s Q for two reasons. First, mining companies typically value reserves on their balance sheet at historical cost as opposed to their net present value based on current estimates of reserves, cost of extraction, and world prices. Because of this industry-specific idiosyncrasy, we argue that market-to-book ratio is not an accurate method of scaling value to a comparable baseline. Our measure of resource value is, by contrast, a better estimate of these mining firms’ replacement value of assets (i.e., the denominator in Tobin’s Q ratio). Second, our empirical model is effectively a structural model that specifies a particular relationship between the variables of interest. Specifically, the model assumes that, in the absence of project-level and country-level risk, β = 1. That is, market value is equal to the net present value of the underlying natural resources owned by the firm. In reality, however, β is never 1. We argue that the formula needs to be modified to account for project-level risk (specifically, the degree of conflict or cooperation with different social, political, and economic stakeholders) and country-level risk (specifically, political risk that reflects the likelihood that the government will unilaterally change policies that alter the value of the project). Replacing market value with market-to-book ratio on the left-hand side of

the equation would misrepresent the underlying structural relationship.

Stakeholder conflict/cooperation

Our research advances the study of the finan- cial impact of stakeholder engagement through the development of media-based stakeholder event data that captures the level of conflict or cooper- ation between the company and its various polit- ical, social, and economic stakeholders. Through- out the nearly four-decade long history of empir- ical literature on the link between stakeholder relations and corporate financial performance (see Margolis, Elfenbein, and Walsh (2007)), mea- surement of the former construct has proven a daunting challenge. Early studies relied on small- sample comparisons of subjectively rated “bet- ter” versus “worse” performers or used corporate responses to surveys. Over time, external sources of data replaced researchers’ own ratings. The data became more objective because it relied upon con- sumer polling, analysis of annual reports, and other public documents regarding corporate practices.

While these were clear improvements both in assuaging concerns on construct validity and in expanding the sample of covered firms, the unit of analysis remained the corporation. Scholars thus explored the effects of corporate-level disclosures, audits and policies on performance but struggled with converting these results into operational guid- ance to frontline managers tasked with resource allocation decisions. Doing so required untenable assumptions that corporate policies translated into operational practices for stakeholder relations and that strategies were not contingent upon country, stakeholder, issue, time, industry, and project con- text. Scholars seeking to loosen these assumptions struggled with a lack of more fine-grained data.

By contrast, scholars looking at the costs to cor- porations associated with irresponsible corporate activity or with external activism targeting the cor- poration have long used more micro-level event data drawn from media reports (Earl et al., 2004). A wide body of literature links media report- ing of adverse events, including product recalls (Davidson and Worrell, 1992), corporate criminal activity (Davidson and Worrell, 1988; Gunthorpe, 1997; Karpoff, Lee, and Vendrzyk, 1999; Karpoff and Lott, 1993), violations of labor law (David- son, Worrell, and Cheng, 1994; Hersch, 1991), and environmental violations (Karpoff, Lott, and

Financial Returns to Stakeholder Engagement 1735

Wehrly, 2005) to negative financial performance. More recently and closely related to our analysis, King and Soule (2007) demonstrate that activist campaigns in the media negatively impact market valuation particularly for campaigns targeting con- sumers or workers in firms who themselves lack a strong prior record of media coverage (i.e., a stock of reputational capital). These studies typi- cally proceed by linking information on the date of an adverse event to abnormal stock market returns or a long-term reduction in market valuation. King and Soule (2007) go further in coding information on the size of the protest, the number of sponsoring organizations, and the type of issue.

Scholars in international relations and conflict studies have gone even further in their coding of event data in their long-standing examination of the impact of “soft power” or the degree of conflict and cooperation among states on subsequent rela- tions between those states, including the incidence of military conflict (for a review of this literature see Schrodt (1995)). In this literature, events are coded as subject-verb-object triples in which one actor undertakes an action or expresses an opin- ion connoting conflict or cooperation with another actor. The intertemporal evolution of dyadic and network conflict and cooperation is analyzed to ascertain the determinants of escalation of international conflict or cooperation.

The closest analogue to this type of data in the management realm lies in the realm of corporate reputation (Fombrun and Shanley, 1990) where the appearance of negative words in the media is associated with subsequent deterioration in earnings and stock prices (Tetlock, Saar- Tsechansky, and Macskassy, 2008) as well as overall stock index levels, trading volume, and volatility (Das and Chen, 2007; Loughran and McDonald, 2011). The tone of earnings press releases is associated with subsequent earnings and short-term stock price movement (Davis, Piger and Sedor, 2007). Text analysis of Amazon seller reviews similarly demonstrates that strong reputations are associated with increasing market power (Ghose and Ipeirotis, 2011).

Bringing such an event-based approach to the study of instrumental stakeholder theory offers numerous advantages over the extant approaches of relying on corporate disclosures, audits, or prin- ciples. First, a much larger sample of firms could potentially be incorporated within an analysis. Sec- ond, the stakeholder opinions and actions that are

included within the media reports are more accu- rate representations of a firm’s relations with its stakeholders than the evaluations of distant topical experts. Third, real-time event reports in the media offer a much more accurate source of information on how a company is perceived by its stakehold- ers at a moment in time than do periodic audits or expert surveys. As stakeholder conflict and cooper- ation fluctuates with events and tactics, a measure that captures these dynamics should outperform those that are more static. Finally, it is easier to separate the actions and opinions of stakeholders from those of the focal firm and therefore examine their interdependence rigorously.

We created a novel database of firm – stakeholder relations from over 50,000 media-reported events linking firms and their stakeholders. Our source was the full set of more than 20,000 media documents in the FACTIVA database that mention the mine or the parent company of the mine. Every article was read carefully to identify stakeholder events, which were then hand-coded according to a detailed coding protocol adapted from the international conflict studies literature (Bond et al., 2003; King and Lowe, 2003). A stakeholder event is an instance in which a media-relevant7

stakeholder acts or expresses sentiment toward the firm or vice-versa. We distinguish between the initiator of the event and its target by coding who (Source actor) did what (Verb or Verb phrase) to whom (Target actor).

We assembled the stakeholder events dataset with the help of a number of undergradu- ate research assistants who underwent extensive training under our guidance and close supervi- sion. To assess the level of inter-rater agreement (IRA) — that is, the level of similarity between different coders’ judgments and the extent to which their work can be considered interchange- able (LeBreton and Senter, 2008) — our research assistants overlapped in the coding of a subset of the articles in our sample. We used this overlap to assess inter-rater agreement on 313 different items, where one “item” is the level of conflict or cooperation between two stakeholders in a given

7 Our reliance on the media to define the population of relevant stakeholders generates an inclusive set of political, social and economic actors without regard to their normative legitimacy. We believe this strategy to be appropriate because we seek to measure the financial impact of stakeholder conflict and cooperation not the choice by managers as to whether to engage with a specific stakeholder.

1736 W. J. Henisz, S. Dorobantu, and L. J. Nartey

quarter. We compute multiple IRA indices used extensively in organizational research, specifically the r WG , AD M , AD Md , and a WG (see LeBreton and Senter, 2008, for complete definitions and formu- las used to compute each index). Regardless of the IRA measure used, we observed a very satis- factory level of inter-rater agreement for the vast majority of the 313 items assessed. For 284 items the level of inter-rater agreement as measured by r WG exceeds 0.70, the value most scholars inter-

pret as indicating agreement (Lance, Butts, and Michels, 2006). Over the 313 items evaluated, the mean of r WG is 0.86 and the standard deviation is 0.27. Similarly, for a WG — an index that adjusts r WG in a way that does not allow the scale and the number of judges to affect the results (Brown and Hauenstein, 2005) — the mean is 0.80 and the stan- dard deviation is 0.09. Using the average deviation from the mean (AD M ) and the average deviation from the median (AD Md ) indices that have been increasingly used in applied psychology and man- agement over the last decade (Smith-Crowe et al., 2013), we obtain an equally satisfactory level of inter-rater agreement. Because this class of indices measures dispersion, lower values indicate higher agreement, and the results can be interpreted on the original scale. For our inter-rater data, the AD M and AD Md have means of 0.77 and 0.72, respec- tively, and standard deviations of 1.02 and 0.99, respectively, on a scale of 1 to 20.

The degree of conflict/cooperation in the verb phrases was coded using a modified version of the (Goldstein, 1992) scale, which we augmented to reflect relations better between firms and stakeholders in the business context. The Gold- stein conflict – cooperation scale is based upon McClelland’s (1971) World Events Interaction Survey (WEIS), which groups international relations events into 22 verb categories ranging from conflictual to cooperative using verbs such as “accuse,” “promise,” “threaten.” Our modified conflict/cooperation scale is a measure of the degree of conflict/cooperation between actors and ranges across 20 categories from extremely con- flictual events valued at −9, denoting the launch of violent attacks with actual or potential serious deaths or injury, to highly cooperative events valued at 10, denoting the provision of armed support or defense; neutral events are valued 0. We extended and slightly adapted the vocabulary of verb phrases to address the difference between firm – stakeholder and stakeholder – stakeholder

interactions in the commercial sphere from the traditional population of interactions in the diplo- matic sphere. Over 11,000 verbs and verb phrases were mapped on the scale through a “fuzzy matching” of synonyms comprising the smaller vocabulary in the original scale (a summary of the categories as well as the full vocabulary is available from the authors upon request). Table 1 provides several examples of sentences coded according to this protocol.

We aggregate the stakeholder events data to reflect the level of a company’s conflict or cooperation with stakeholders at each mine in every quarter of available data and compute a rolling stock of stakeholder conflict/cooperation. The empirical measure is constructed using a moving average that discounts the “relevance” of past reports by weighing less a report dating from the past than a current report. Formally, for each mine j of each company i at time t we calculate

CC ∗jit = w∑

l =0

δl nji ,t −l w∑

m=0 δm nji ,t −m

ccji ,t −l , where

ccjit = level of stakeholder conflict/cooperation at mine j for company i at time t , n jit = number of new media reports for mine j of company i at time t , w = window of the moving average, and δ = discount factor.

The main results presented below are estimated using a window of eight quarters and a discount rate of 0.8; sensitivity checks show that these results are robust to specifications using alterna- tive values. Summary statistics and a correlation matrix for the variables in our dataset are reported in Table 2.

Policy credibility

While our measure of stakeholder conflict/ cooperation proxies for the social license to operate, we also seek to adjust the resource valuation for the credibility of the formal license to operate. Where the current concession, taxation, or royalty agreements lack credibility, investors may apply a discount to the resource valuation irrespective of the degree of stakeholder conflict/ cooperation under the premise that a future government could more readily alter policies in

Financial Returns to Stakeholder Engagement 1737

T ab

le 1.

S am

pl es

of st

ak eh

ol de

r ev

en t

co di

S en

te nc

e te

xt S

ou rc

e (o

r su

bj ec

t) V

er b(

s) T

ar ge

t( s)

(o r

ob je

ct (s

)) C

on fl

ic t/

co op

er at

io n

ca te

go ry

C on

fl ic

t/ co

op er

at io

n sc

al e

A S

G ch

ai rm

an S

te ph

en E

ve re

tt al

so pr

ai se

d R

A M

S I

an d

lo ca

l po

li ce

an d

th an

ke d

th e

S ol

om on

s go

ve rn

m en

t fo

r it

s po

si ti

ve su

pp or

A S

G ch

ai rm

an S

te ph

en E

ve re

tt P

ra is

e; th

an k

L oc

al po

li ce

; S

ol om

on s

go ve

rn m

en t

[E xp

re ss

su pp

or t

ve rb

al ly

] 3

O n

S ep

te m

be r

14 ,

20 07

, P

re si

de nt

N ur

su lt

an N

az ar

ba ye

v of

K az

ak hs

ta n

ce re

m on

ia ll

y ki

ck ed

of f

th e

pr oc

es s

of ex

tr ac

ti ng

go ld

an d

co pp

er or

e at

th e

V ar

va ri

ns ko

ye de

po si

t. H

e w

as qu

ot ed

as sa

yi ng

th at

th is

m in

e is

on e

of m

an y

en te

rp ri

se s

in th

e re

gi on

th at

w il

l “b

ui ld

up th

e po

w er

of K

az ak

hs ta

n’ s

ec on

om y”

P re

si de

nt N

az ar

ba ye

v C

er em

on ia

ll y

ki ck

ed of

f V

ar va

ri ns

ko ye

de po

si t

[o w

ne d

by E

ur op

ea n

M in

er al

s C

or po

ra ti

on ]

[S ho

w su

pp or

t th

ro ug

h ac

ti on

] 4

[G eo

rg e]

S al

am is

[P re

si de

nt of

R us

or o,

R us

si an

fi rm

] sh

ie d

aw ay

fr om

co m

m en

ti ng

di re

ct ly

on th

e im

po rt

an ce

of R

us or

o’ s

R us

si an

co m

po ne

nt bu

t in

st ea

d sa

id :

“W e

w ou

ld n’

t be

an yw

he re

in V

en ez

ue la

if it

w er

en ’t

fo r

th e

gr ea

t co

nn ec

ti on

s w

e’ ve

bu il

t w

it h

th e

V en

ez ue

la n

go ve

rn m

en t

at al

l le

ve ls

.”

S al

am is

- P

re si

de nt

of R

us or

o B

ui ld

co nn

ec ti

on s

V en

ez ue

la n

go ve

rn m

en t

[B ui

ld po

si ti

ve re

la ti

on s

w it

h] 3

M r.

K ab

il a

ha s

or de

re d

fo re

ig n

co m

pa ni

es op

er at

in g

th er

e to

ne go

ti at

e or

se e

th ei

r co

nc es

si on

s so

ld to

ri va

ls .

T he

co m

pa ni

es in

vo lv

ed —

th e

T or

on to

-b as

ed ex

pl or

at io

n co

m pa

ny B

an ro

R es

ou rc

e, a

B el

gi an

-C an

ad ia

n co

ns or

ti um

ca ll

ed M

in de

v an

d B

ar ri

ck G

ol d,

on e

of th

e w

or ld

’s la

rg es

t go

ld co

m pa

ni es

— ar

e in

an un

en vi

ab le

po si

ti on

M r.

K ab

il a

[l ea

de r

of A

D F

L re

be l

gr ou

p] O

rd er

ed . . .

to ne

go ti

at e

or se

e th

ei r

co nc

es si

on s

so ld

F or

ei gn

co m

pa ni

es B

an ro

, M

in de

v &

B ar

ri ck

[T hr

ea te

n] −4

1738 W. J. Henisz, S. Dorobantu, and L. J. Nartey

T ab

le 1.

C o

n ti

n u

S en

te nc

e te

xt S

ou rc

e (o

r su

bj ec

t) V

er b(

s) T

ar ge

t( s)

(o r

ob je

ct (s

)) C

on fl

ic t/

co op

er at

io n

ca te

go ry

C on

fl ic

t/ co

op er

at io

n sc

al e

K ab

il a’

s go

ve rn

m en

t is

fi gh

ti ng

fo r

it s

su rv

iv al

as re

be ls

ba ck

ed by

ne ig

hb or

in g

R w

an da

an d

U ga

nd a

ha ve

pu sh

ed th

ei r

w ay

w es

tw ar

d to

w ar

d th

e ca

pi ta

l ci

ty of

K in

sh as

a. Z

im ba

bw e

an d

A ng

ol a

ar e

su pp

or ti

ng K

ab il

a w

it h

ar m

s an

d tr

oo ps

K ab

il a’

s go

ve rn

m en

t F

ig ht

fo r

su rv

iv al

R eb

el s

[O pp

os ed

in ac

ti ve

m il

it ar

y co

nfl ic

t] −1

K ab

il a’

s go

ve rn

m en

t is

fi gh

ti ng

fo r

it s

su rv

iv al

as re

be ls

ba ck

ed by

ne ig

hb or

in g

R w

an da

an d

U ga

nd a

ha ve

pu sh

ed th

ei r

w ay

w es

tw ar

d to

w ar

d th

e ca

pi ta

l ci

ty of

K in

sh as

a. Z

im ba

bw e

an d

A ng

ol a

ar e

su pp

or ti

ng K

ab il

a w

it h

ar m

s an

d tr

oo ps

R w

an da

go ve

rn m

en t;

U ga

nd a

go ve

rn m

en t

B ac

k R

eb el

s [S

up po

rt in

ac ti

ve m

il it

ar y

co nfl

ic t]

K ab

il a’

s go

ve rn

m en

t is

fi gh

ti ng

fo r

it s

su rv

iv al

as re

be ls

ba ck

ed by

ne ig

hb or

in g

R w

an da

an d

U ga

nd a

ha ve

pu sh

ed th

ei r

w ay

w es

tw ar

d to

w ar

d th

e ca

pi ta

l ci

ty of

K in

sh as

a. Z

im ba

bw e

an d

A ng

ol a

ar e

su pp

or ti

ng K

ab il

a w

it h

ar m

s an

d tr

oo ps

Z im

ba bw

e go

ve rn

m en

t; A

ng ol

a go

ve rn

m en

S up

po rt

w it

h ar

m s

K ab

il a

[S up

po rt

in ac

ti ve

m il

it ar

y co

nfl ic

t] 10

Financial Returns to Stakeholder Engagement 1739

Table 2. Summary statistics and correlation matrix

Variable Observations Mean St. dev. Min Max

Market value 176 304.227 355.215 3.930 2279.556 Resource valuation 147 594.553 522.003 83.819 2294.293 Valuation × stakeholder conflict/cooperation 112 221.013 166.453 0.000 799.493 Valuation × (stakeholder

conflict/cooperation + policy certainty)/2 73 125.524 79.642 9.790 339.052

Stakeholder conflict/cooperation 115 0.507 0.218 0.000 1.000 Policy certainty 176 0.158 0.198 0.000 0.670

Market value

Resource valuation

Valuation × stakeholder

conflict/ cooperation

Valuation × (stakeholder

conflict/cooperation + policy certainty)/2

Stakeholder conflict/

cooperation Policy

certainty

Market value Resource valuation 0.110 Valuation × stakeholder

conflict/cooperation 0.121 0.911

Valuation × (stakeholder conflict/cooperation + policy certainty)/2

0.206 0.913 0.905

Stakeholder conflict/cooperation

0.082 −0.116 0.260 0.092 Policy certainty 0.155 −0.085 −0.241 0.152 -0.203

a manner inimical to investors (Fatás and Mihov, 2003; Henisz, 2004). We measure the credibility of the formal policy environment using the Political Constraint Index (POLCON) which captures the number of veto players across the different inde- pendent branches of government in each country adjusted by the level of political alignment across the different branches (Henisz, 2000a).

ECONOMETRIC ANALYSIS AND RESULTS

Cooperative relationships between the firm and its various stakeholders increase the probability that the mining project will develop without sig- nificant delays. We test whether the expected value of the company’s mining resources matches the market value using random-parameter models that can accommodate individual (i.e., parent-firm level) heterogeneity in the relationship between resource and financial market valuation, as well as random-effects (or random intercept) models and fixed-effects models. A wide array of coeffi- cient estimates obtained in company-by-company regressions (results not shown) indicate that investors assign different dollar values to a unit

increase in the resource valuation (or the expected resource valuation) of different mines. Random- parameter estimators can accommodate such unit to unit variation by relaxing the assumption that the relationship between Y i and X i is defined by a vector of true coefficients βi and assuming instead that the coefficients βi are random variables.

We estimate the following random-parameter model:

MVit = αi + βi m∑

j =1 RVjit Pjit + �it ,

where

αi = α + γ α v i ,α , βi = β + γ β v i ,β , and v i ,α , v i ,β ∼ N (0,1).

Results vary considerably when estimating the regression coefficient on the resource valuation for each company (RV jit ) and the expected resource valuation (RV jit P jit ) that takes into consideration stakeholder support and policy credibility. The results obtained using the random-parameter esti- mation are shown in Table 3. The models assume that the coefficients are random variables drawn from a normal distribution and estimate both the

1740 W. J. Henisz, S. Dorobantu, and L. J. Nartey

Table 3. Random coefficient estimates

(1) (2) (3) (4) (5)

Resource value 0.284*

(2.55) Resource value × stakeholder

conflict/cooperation 0.523*

(2.41) Resource value × policy certainty 0.464*

(2.21) Resource value × (stakeholder

conflict/cooperation + policy certainty)/2 0.631***

(4.02) Resource value × stakeholder

conflict/cooperation × policy certainty 0.867*

(2.52) Constant 183.5*** 177.7*** 208.8*** 176.7*** 195.7***

(4.32) (3.60) (5.30) (4.41) (4.99) lns1_1_1 Constant −0.982*** −0.308 −0.615 −1.026* −0.126

(−4.13) (−1.17) (−1.60) (−2.06) (−0.33) lns1_1_2 Constant 5.070*** 5.205*** 4.731*** 4.720*** 4.706***

(23.95) (23.83) (12.56) (9.81) (12.01) atr1_1_1_2 Constant −0.337 −0.487 −0.121 0.155 −0.231

(−1.05) (−1.55) (−0.27) (0.29) (−0.49) lnsig_e Constant 4.946*** 4.860*** 4.816*** 4.810*** 4.825***

(111.93) (97.21) (73.20) (73.40) (71.17) Observations 289 235 141 141 136

t statistics in parentheses. *p < 0.05; **p < 0.01; ***p < 0.001.

mean and the standard deviation for the intercept and the slope.

Model (1) shows that, when assuming that investors do not factor in the possibility of delays in the planning stages and disruptions in the pro- duction stage, they are willing to pay, on average, about 28 cents for an increase of 1 dollar in the resource valuation of a gold mining company. By contrast, when we consider the possibility that in bringing the mine to production the company will likely encounter various obstacles that delay the extraction of the resources and model it in terms of the level of stakeholder con- flict/cooperation (model 2), in terms of the level of policy credibility (model 3), or in terms of both the level of stakeholder conflict/cooperation and policy credibility (models 4 and 5), the estimated unconditional means for the coefficients are higher and closer to 1. More specifically, if we adjust the resource valuation by the level of stakeholder conflict/cooperation as a proxy for the likelihood of considerable delays or disruptions (model 2), our random estimate suggests that investors are

willing to pay, on average, about 52 cents for every dollar increase in the company’s expected resource valuation. If we adjust the resource valu- ation by the level of country-level political risk as a proxy for the likelihood of delays in the project development schedule (model 3), the random coef- ficient estimates indicate that investors are willing to pay about 46 cents for every dollar increase in the company’s resource cash flows. Finally, if we model the likelihood that the company will face delays and disruptions at its mine(s) in terms of both the level of conflict/cooperation with stakeholder and country-level political risk using the mean of the two (model 4) or their product (model 5) and, if we discount the valuation of the company’s resources by these values, the estimated random coefficient suggests that investors are willing to pay, on average, between 63 and 87 cents for every dollar increase in the expected resource valuation of a mining company.

We also estimate the equivalent of a random- intercept (or random-effects) model in which only the intercept is assumed to be a random

Financial Returns to Stakeholder Engagement 1741

Table 4. Random effects estimates with robust standard errors

(1) (2) (3) (4) (5)

Resource value 0.355*

(2.43) Resource value × stakeholder

conflict/cooperation 0.221***

(5.90) Resource value × policy certainty 0.495***

(3.50) Resource value × (stakeholder

conflict/cooperation + policy certainty)/2 0.594***

(4.50) Resource value × stakeholder

conflict/cooperation × policy certainty 0.988***

(4.21) Constant 168.1** 231.2*** 223.5*** 188.4*** 208.1***

(3.06) (4.81) (3.99) (3.33) (4.02) Observations 289 235 141 141 136

t statistics in parentheses. *p < 0.05; **p < 0.01; ***p < 0.001.

variable, αi = α + γ α v i ,α , while the coefficients on the independent variables are “fixed” (i.e., nonrandom). The results presented in Table 4 suggest that investors are willing to pay about 35 cents for an increase of 1 dollar in the valuation of a company’s gold mine project(s) if the possibility of delays and disruptions is not accounted for; but they are willing to pay between 60 and 99 cents for an increase of 1 dollar in the valuation of a company’s gold mining project(s) if the likelihood of moving ahead according to the announced schedule is defined in terms of the level of stakeholder conflict/cooperation and the country-level policy credibility. The results shown are estimated using robust standard errors and are robust to specifications that control for AR(1) processes. We obtain very similar results if estimating the parameters using fixed effects and panel corrected standard errors (results not shown).

ROBUSTNESS CHECKS AND SENSITIVITY ANALYSIS

We checked the sensitivity of our results to con- siderations related to various definitions of stake- holder support and the measurement of this con- cept, the direction of causality, market trends, as well as concerns related to the mine development schedule considered in the analysis and the fact that one-quarter of the companies in our sample operate more than one mine. Overall, our results proved highly robust.

Stakeholder conflict/cooperation measure

Because of data availability, much of the work on the impact of stakeholder engagement on cor- porate operations and performance is limited to the analysis of stakeholders in the value chain. While our media-based event data allows us also to incorporate various political actors (local, regional, national, and foreign government officials and bureaucrats), and social stakeholders (com- munity representatives and various cultural, reli- gious, ethnic, environmental, and human rights organizations), we verified the robustness of our results to the inclusion/exclusion of stakehold- ers in the company’s value chain, such as pri- vate and state-owned businesses in the mining industry and corporate and individual service providers.

We also consider the sensitivity of our results to the choices we made while constructing our measure of stakeholder conflict/cooperation. First, this measure represents a moving average of event data that weighs less heavily stakeholder relations described in past media reports relative to current information. We estimated the results shown using a discount factor of 0.8 but also confirmed that our results are robust to a wider range of discount factors and rolling windows. Finally, we verified that the level of stakeholder conflict/cooperation does not have different impacts in different countries depending on the overall country-level stakeholder relations (Bond et al., 2003; King and Lowe, 2003).

1742 W. J. Henisz, S. Dorobantu, and L. J. Nartey

Direction of causality

First, to address concerns that our results are driven by managerial agency or time variant firm-level heterogeneity, we regressed stakeholder coopera- tion on lagged financial market valuation using the same set of specifications detailed above. In no cases did we observe a statistically significant rela- tionship. Second, we ran a Granger causality test to confirm that the resource valuation discounted using stakeholder conflict/cooperation and politi- cal risk Granger causes the company market value and also verified that resource valuation using future changes in the values of stakeholder con- flict/cooperation (at times t + 1 and t + 2 ) do not predict market value. Third, we confirmed empir- ically that exogenous increases in gold reserves (i.e., significant jumps in the net present value of a mine) do not trigger adverse reactions from stakeholders, thus eliminating concerns that com- panies’ discovery announcements rather than their interactions with stakeholders influence the level of conflict or cooperation with stakeholders and indirectly market value. Finally, we attempted to identify an instrument that was (1) strongly cor- related with our stakeholder conflict/cooperation variable; (2) unobservable to investors; (3) observ- able to us. Unfortunately, we have access to the same information that investors observe and no more, and thus were unable to isolate fac- tors that are of no interest to investors, but are nonetheless correlated with the degree of stakeholder cooperation.

Market trends

We also reestimated the results on the basis of empirical specifications that account for fluctua- tions in the stock market. To address the possibility that some of the movement in the stock price of the 19 companies in our sample is due to systemic risk, not to company-level changes in the valua- tion of the physical assets or the tenor of the firm’s relationships with its stakeholders, we include in the analysis the S&P/Toronto Stock Exchange Composite Index.8 Our results are robust to this specification.

8 The S&P/Toronto Stock Exchange Composite Index is a capitalization-weighted index measuring market activity of stocks listed on the TSX (www.bloomberg.com).

Mine development schedules

By the end of our panel dataset, 11 mines in our sample had reached production, while the remaining 15 continued to be in either the feasibility or construction stage and therefore more vulnerable to future planning delays or disruptions. Companies try to reassure investors that production will begin in the near future by announcing the planned production start date, but they sometimes have to revise these announce- ments to reflect delays. We account for such real-life uncertainty by estimating the resource value using both companies’ announced produc- tion start date and a range of mine development schedules that accommodate different times for each of the exploration, feasibility, construction, and production stages. The results we present are estimated using companies’ announced production schedules; however, our results are robust to cal- culations that assume very tight (or “optimistic”) development paths as well as more conservative (or “slower”) planned schedules. We also verified that mine development schedules do not vary systematically across the mines in our sample with either the level of stakeholder conflict/cooperation or the level of political risk.

Companies with multiple mines

One of the strengths of the empirical setting for this research is that in the gold mining industry small entrepreneurial firms have a very small number of projects in development, allowing us to examine the effects of the tenor of stakeholder relations “on the ground” on market valuation without interference from related events. In our sample, 14 companies have just one mine, 3 companies have two mines each, and 2 companies have three mines each. It is possible that for companies with more than one mine stakeholder events at one site counterbalance developments at the other site(s), confounding the effect of stake- holder relations on market value. We acknowledge this and reestimate the models with data pertaining to the 14 companies with only one mine with no substantive change in our results albeit with some loss of statistical power.

DISCUSSION

Our theoretical arguments and empirical results point to the existence of a direct positive and

Financial Returns to Stakeholder Engagement 1743

economically substantive relationship between stakeholder support and financial market valua- tion. Future research and analysis should continue to explore not only trade-offs but also comple- mentarity between resource allocations to enhance stakeholder cooperation and productive efficiency. This finding has important implications for future research on corporate social responsibility, instru- mental stakeholder theory, and multinational strat- egy, which we discuss in turn. We also discuss the limitations of our analysis including, in particu- lar, questions regarding its generalizability beyond our sample of 19 publicly traded gold mining com- panies. This discussion reveals exciting topics for future research.

The research on corporate social responsibility has struggled to make the business case for such activity (Vogel, 2005). Initially supportive empir- ical results were rightly attacked on the grounds of spurious or reverse causality and inappropriate metrics. Agency theorists and financial economists countered with theoretical and empirical analy- sis that highlighted how managers’ pursuit of perquisites or individual morality diverted share- holder returns to stakeholder interests. Organiza- tional economists extended these models to allow for redistributions to stakeholders in the direct eco- nomic value chain that also maximize shareholder value, but only under certain circumstances.

We argue, by contrast, for a broader conceptu- alization of the potential financial impact of stake- holders that draws from the core of instrumental stakeholder theory. The value of certain assets can be diminished if external stakeholders directly interfere with or lobby the government to inter- fere with the property rights of the owner of the assets. In our case, financial models that gener- ate a valuation for a gold mine omit the future uncertainty over government regulation, permit- ting, and community relations, or take these factors to be exogenously or environmentally determined. Yet, the actions of managers who devote resources and efforts to mitigate and quantify what they call “above-ground” risks in this industry suggest that they disagree with this characterization and see variation in political and social support as a source of competitive advantage or economic rents.

While managers, scholars of stakeholder rela- tions, and some activists have long asserted the existence of a positive benefit from stakeholder engagement, empirical evidence using corporate- level data has been equivocal at best. The focus

among theorists and empirical scholars has turned to special circumstances where a link may yet exist. Our results point to a need to broaden the scope of such inquiry. Where stakeholder cooper- ation is necessary to transform an asset into share- holder returns, a direct link between productive efficiency and stakeholder cooperation exists. To show this, we use an empirical context in which none of the mechanisms emphasized by organi- zational economists are operative (i.e., consumers are unable to identify the mine or company from which their purchase originates, miners and min- ing company managers are not seen as leaders in social responsibility, investors who emphasize social responsibility are seen as relatively periph- eral and activists have limited power to cause direct harm to investors) and find strong empirical support for the financial value of external stake- holder engagements.

This link offers an opportunity for instrumen- tal stakeholder theory to address the question of which stakeholders are more important and how much managers should invest in each relationship. Current empirical efforts to examine corporate- level reporting and practices are too far removed from the operational practices of greatest concern to stakeholders both external and internal to the corporation. By tracking the actions and statements of media-relevant stakeholders, scholars and prac- titioners can avoid subjective biases, broaden the pool of covered firms, and better identify which practices substantively contribute to market val- uation. In ongoing research, we analyze in more depth the stakeholder networks for each of the 26 mines. Using the same coding protocol we deployed here, we capture every stakeholder event in which the mine or one of its stakeholders is the object of the sentence. The resulting dataset is amenable to analysis using tools developed in the network literature to ascertain the relative impor- tance of various stakeholders in the diffusion of conflict or cooperation as well as the optimal strategies for an organization seeking to enhance its degree of stakeholder support (Nebus and Rufin, 2010; Rowley, 1997; Rowley and Moldoveanu, 2003). While currently time intensive, progress in information extraction software development (King and Lowe, 2003) could allow for lower-cost deployment in the near future.

Such progress is particularly needed in the study of multinational firms who, by virtue of their foreignness, are more likely to engender a

1744 W. J. Henisz, S. Dorobantu, and L. J. Nartey

conflictual relationship with a given host country stakeholder than a cooperative one (Zaheer, 1995). Such reactions may be based on intrinsic nationalism or may be opportunistically fomented by a stakeholder who seeks to stop the project’s development. The political risk literature in international business has made great strides in modeling the formal institutional structures that govern the likelihood of policy change in response to such pressure (Henisz, 2000b), the extent to which investors from one country may be more or less susceptible to this pressure (Holburn and Zelner, 2010), and the existence of experiential learning in the mitigation of these pressures (Delios and Henisz, 2002; Henisz and Delios, 2001). Nonetheless, the tactics that firms deploy in such cases remain largely unexamined despite long-standing exhortations for analysis (Kobrin, 1979). The dynamic and contingent analysis of stakeholder networks can extend the existing liter- ature by examining the impact of political risk on multinational strategy to bring it closer to schol- arship in nonmarket strategy (Baron, 2009; Baron and Diermeier, 2007; Hillman and Hitt, 1999) and strategic corporate social responsibility (Kytle and Ruggie, 2005; Porter and Kramer, 2006, 2011; Post, Preston, and Sauter-Sachs, 2002).

We believe that the scope of such potential inquiry is far wider than the gold mining industry. While we chose this empirical context because it allowed us to isolate the mechanism we posited from numerous potential competing causal expla- nations, the contingent nature of property rights faced by owners of gold mines has widespread analogues in other industrial contexts. The argu- ment clearly applies to other natural resources (e.g., minerals, oil or gas, agriculture and water). Analogous calculations could lead to adjustments in project valuations for the construction of a new facility or infrastructure asset. Industries with substantial upfront investments and long payback periods are similarly influenced by the realization of property rights over those admittedly harder to value up front: investments in property, plant, equipment, intellectual property, production pro- cesses or brand.

Concerns over government and stakeholder sup- port for the right to transform property are height- ened where the good or service manufactured or the production process is politically or socially salient. Such salience is a function of perceptions of spillovers both negative (e.g., environmental

or social costs, disruptions of cultural tradition and heritage, and the reinforcement of preexisting inequity) and positive (e.g., categorization as desir- able technology, creation of high value added jobs, consistency with broader social or political objec- tives, or a critical and undersupplied input into a production process that itself has these character- istics). While the absolute and relative importance of this direct link between perceptions of social responsibility and market valuation will obviously vary enormously across industries and countries, we would argue from the above set of condi- tions that its existence is ubiquitous. In short, the social license to operate is more than rhetoric. It is operationalizable, empirically testable and strate- gically relevant. For these mining firms and their peers in politically and socially salient sectors, pur- suing support from external stakeholders is not just corporate social responsibility but enlightened self-interest.

ACKNOWLEDGEMENTS

We thank Ashish Arora, Thomas Donaldson, Mauro Guillén, Stephen Kobrin and anonymous reviewers from the Academy of Management and Academy of International Business 2011 annual conferences for their feedback. We acknowl- edge funding from the National Science Foun- dation (SES-1039550), Santander Universidades, the Global Initiatives Research Program of the Wharton School, the Wharton Center for Leader- ship and Change Management, and the Christo- pher H. Browne Center for International Politics at the University of Pennsylvania. We thank TSX Datalinx for providing stock price data. We also thank Adam Garson, Arthur Xu, Bertram Ieong, Boyan Gerasimov, Brianna Wilson, Christoph Suter, Daniel Mizsei, Dong Jin Han, Ivan Kout- sarov, Jervis Hui, Kevin Koplan, Lavinia Seow, Louis Balocca, Mi Hyun Lee, Neha Karmeshu, Priyanka Anand, Rishav Kanoria, and Seung- Jae Lee for over 2,500 hours of superb research assistance.

REFERENCES

Aden J, Kyu-Hong A, Rock MT. 1999. What is driv- ing the pollution abatement behavior of manufac- turing plants in Korea? World Development 27(7): 1203 – 1214.

Financial Returns to Stakeholder Engagement 1745

Alchian AA, Demsetz H. 1972. Production, information costs, and economic organization. American Economic Review 62(5): 777 – 795.

Arora S, Gangopadhyay S. 1995. Toward a theoretical model of voluntary overcompliance. Journal of Eco- nomic Behavior & Organization 28(3): 289 – 309.

Bagnoli M, Watts S. 2003. Selling to socially responsible consumers: competition and the private provision of public goods. Journal of Economics & Management Strategy 12(3): 419 – 445.

Baker G, Gibbons R, Murphy KJ. 2002. Relational contracts and the theory of the firm. Quarterly Journal of Economics 117(1): 39 – 84.

Barnea A, Rubin A. 2006. Corporate social responsibility as a conflict between shareholders. Journal of Business Ethics 97(1): 1 – 16.

Barnett ML. 2007. Stakeholder influence capacity and the variability of financial returns to corporate social responsibility. Academy of Management Review 32(3): 794 – 816.

Barnett ML, Salomon R. 2006. Beyond dichotomy. The curvilinear relationship between social responsibility and financial performance. Strategic Management Journal 27(11): 1101 – 1122.

Barnett ML, Salomon RM. 2012. Does it pay to be really good? Addressing the shape of the relationship between social and financial performance. Strategic Management Journal 33: 1304 – 1320.

Baron DP. 2001. Private politics, corporate social respon- sibility, and integrated strategy. Journal of Economics & Management Strategy 10(1): 7 – 45.

Baron DP. 2009. A positive theory of moral management, social pressure, and corporate social performance. Journal of Economics & Management Strategy 18(1): 7 – 43.

Baron DP, Diermeier D. 2007. Strategic activism and nonmarket strategy. Journal of Economics & Manage- ment Strategy 16(3): 599 – 634.

Baron DP, Harjoto MA, Jo H. 2011. The economics and politics of corporate social performance. Business & Politics 13(2): 1 – 46.

Besley T, Ghatak M. 2005. Competition and incentives with motivated agents. American Economic Review 95(3): 616 – 636.

Besley T, Ghatak M. 2007. Retailing public goods: the economics of corporate social responsibility. Journal of Public Economics 91(9): 1645 – 1663.

Bhattacharya C, Sen S, Korschun D. 2008. Using corporate social responsibility to win the war for talent. MIT Sloan Management Review 49(2): 37 – 44.

Binder S, Neumayer E. 2005. Environmental pressure group strength and air pollution: an empirical analysis. Ecological Economics 55(4): 527 – 538.

Bond D, Bond J, Oh C, Jenkins JC, Taylor CL. 2003. Integrated data for events analysis (IDEA): an event typology for automated events data development. Journal of Peace Research 40(6): 733 – 745.

Boutilier R. 2009. Stakeholder Politics: Social Capital, Sustainable Development, and the Corporation . Stan- ford University Press: Stanford, CA.

Boutilier R. 2011. A Stakeholder Approach to Issues Management . Business Expert Press: New York.

Brekke KA, Kverndokk S, Nyborg K. 2003. An economic model of moral motivation. Journal of Public Eco- nomics 87(9 – 10): 1967 – 1983.

Brown RD, Hauenstein NMA. 2005. Interrater agreement reconsidered: an alternative to the rwg indices. Organizational Research Methods 5(2): 159 – 172.

Cairns RD, Davis GA. 1998. On using current informa- tion to value hard-rock mineral properties. Review of Economics and Statistics 80(4): 658 – 663.

Campbell J. 2007. Why would corporations behave in socially responsible ways? An institutional theory of corporate social responsibility. Academy of Manage- ment Review 32(3): 946 – 967.

Casadesus Masanell R, Crooke M, Reinhardt F, Vasishth V. 2009. Households’ willingness to pay for “green” goods: evidence from Patagonia’s introduction of organic cotton sportswear. Journal of Economics & Management Strategy 18(1): 203 – 233.

Cespa G, Cestone G. 2007. Corporate social responsibility and managerial entrenchment. Journal of Economics & Management Strategy 16(3): 741 – 771.

Chiu S-C, Sharfman M. 2011. Legitimacy, visibility, and the antecedents of corporate social performance: an investigation of the instrumental perspective. Journal of Management 37(6): 1558 – 1585.

Clarkson MBE. 1995. A stakeholder framework for ana- lyzing and evaluating corporate social performance. Academy of Management Review 20(1): 92 – 117.

Das S, Chen M. 2007. Yahoo! for Amazon: sentiment extraction from small talk on the web. Management Science 53(9): 1375 – 1388.

David P, Bloom M, Hillman AJ. 2007. Investor activism, managerial responsiveness, and corporate social performance. Strategic Management Journal 28(1): 91 – 100.

Davidson W, Worrell D. 1988. The impact of announce- ments of corporate illegalities on shareholder returns. Academy of Management Journal 31(1): 195 – 200.

Davidson W, Worrell D. 1992. Research notes and com- munications: the effect of product recall announce- ments on shareholder wealth. Strategic Management Journal 13(6): 467 – 473.

Davidson W, Worrell D, Cheng L. 1994. The effective- ness of OSHA penalties: a stock-market-based test. Industrial Relations: A Journal of Economy and Soci- ety 33(3): 283 – 296.

Davis AK, Piger JM, Sedor LM. 2007. Beyond the numbers: managers’ use of optimistic and pessimistic tone in earnings press releases. SSRN eLibrary.

Delios A, Henisz WJ. 2002. Political hazards, experi- ence and sequential entry strategies: the international expansion of Japanese firms, 1980 – 1998. Strategic Management Journal 24(11): 1153 – 1164.

Donaldson T, Preston LE. 1995. The stakeholder the- ory of the corporation: concepts, evidence, and implications. Academy of Management Review 20(1): 65 – 91.

Dowell G, Hart S, Yeung B. 2000. Do corporate global environmental standards create or destroy market value? Management Science 46(8): 1059 – 1074.

Du S, Bhattacharya CB, Sen S. 2007. Reaping relational rewards from corporate social responsibility: the role

1746 W. J. Henisz, S. Dorobantu, and L. J. Nartey

of competitive positioning. International Journal of Research in Marketing 24(3): 224 – 241.

Earl J, Martin A, McCarthy JD, Soule SA. 2004. The use of newspaper data in the study of collective action. Annual Review of Sociology 30: 65 – 80.

Elfenbein D, McManus B. 2010. A greater price for a greater good? Evidence that consumers pay more for charity-linked products. American Economic Journal: Economic Policy 2(2): 28 – 60.

Epstein MJ, Schnietz KE. 2002. Measuring the cost of environmental and labor protests to globalization: an event study of the failed 1999 Seattle WTO talks. International Trade Journal 16(2): 129 – 160.

Fatás A, Mihov I. 2003. The case for restricting fiscal policy discretion. Quarterly Journal of Economics 118(4): 1419 – 1447.

Fisman R, Heal G, Nair V. 2006. A model of corporate philanthropy. Working paper 1331, Wharton School of the University of Pennsylvania, Philadelphia, PA.

Fombrun C, Shanley M. 1990. What’s in a name? Reputation building and corporate strategy. Academy of Management Journal 33(2): 233 – 258.

Freeman RE. 1984. Stakeholder Management: Frame- work and Philosophy . Pitman: Mansfield, MA.

Freeman RE. 2010. Strategic Management: A Stakeholder Approach . Cambridge University Press: New York.

Friedman M. 1962. Capitalism and Freedom . University of Chicago Press: Chicago, IL.

Friedman M. 1970. The social responsibility of business is to increase its profits. New York Times Magazine 13 September: 122 – 126.

Frooman J. 1999. Stakeholder influence strategies. Academy of Management Review 24(2): 191 – 205.

Frynas JG. 2008. Corporate social responsibility and international development: critical assessment. Cor- porate Governance-an International Review 16(4): 274 – 281.

Frynas JG. 2010. Corporate social responsibility and societal governance: lessons from transparency in the oil and gas sector. Journal of Business Ethics 93(2): 1 – 17.

Ghose A, Ipeirotis P. 2011. Estimating the socio- economic impact of product reviews: mining text and reviewer characteristics. IEEE Transactions on Knowl- edge and Data Engineering 23(10): 1498 – 1512.

Godfrey PC. 2005. The relationship between corporate philanthropy and shareholder wealth: a risk manage- ment perspective. Academy of Management Review 30(4): 777 – 798.

Godfrey PC, Merrill CB, Hansen JM. 2009. The rela- tionship between corporate social responsibility and shareholder value: an empirical test of the risk man- agement hypothesis. Strategic Management Journal 30(4): 425 – 445.

Goldstein JS. 1992. A conflict-cooperation scale for WEIS events data. Journal of Conflict Resolution 36(2): 369 – 385.

Greening DW, Turban DB. 2000. Corporate social perfor- mance as a competitive advantage in attracting a qual- ity workforce. Business & Society 39(3): 254 – 280.

Gunthorpe D. 1997. Business ethics: a quantitative analysis of the impact of unethical behavior by

publicly traded corporations. Journal of Business Ethics 16(5): 537 – 543.

Hemingway C, Maclagan P. 2004. Managers’ personal values as drivers of corporate social responsibility. Journal of Business Ethics 50(1): 33 – 44.

Henisz WJ. 2000a. The institutional environment for economic growth. Economics and Politics 12(1): 1 – 31.

Henisz WJ. 2000b. The institutional environment for multinational investment. Journal of Law Economics & Organization 16(2): 334 – 364.

Henisz WJ. 2004. Political institutions and policy volatil- ity. Economics and Politics 16(1): 1 – 27.

Henisz WJ, Delios A. 2001. Uncertainty, imitation, and plant location: Japanese multinational corporations, 1990 – 1996. Administrative Science Quarterly 46(3): 443 – 475.

Henisz WJ, Zelner BA. 2005. Legitimacy, interest group pressures and institutional change in emergent institu- tions: the case of foreign investors and home country governments. Academy of Management Review 30(2): 361 – 382.

Hersch J. 1991. Equal employment opportunity law and firm profitability. Journal of Human Resources 26(1): 139 – 153.

Hillman AJ, Hitt MA. 1999. Corporate political strategy formulation: a model of approach, participation and strategy decisions. Academy of Management Journal 24(4): 825 – 842.

Hillman AJ, Keim GD. 2001. Shareholder value, stake- holder management, and social issues: what’s the bottom line? Strategic Management Journal 22(2): 125 – 139.

Hoepner AG, Yu P-S, Ferguson J. 2010. Corporate social responsibility across industries: when can who do well by doing good? SSRN eLibrary.

Holburn GLF, Zelner BA. 2010. Policy risk, political capabilities and international investment strategy: evidence from the global electric power industry. Strategic Management Journal 31(12): 1290 – 1315.

Husted B, Allen D. 2006. Corporate social responsibility in the multinational enterprise: strategic and institu- tional approaches. Journal of International Business Studies 37(6): 838 – 849.

Jennings P, Zandbergen P. 1995. Ecologically sustainable organizations: an institutional approach. Academy of Management Review 20(4): 1015 – 1052.

Jensen MC. 2002. Value maximization, stakeholder theory, and the corporate objective function. Business Ethics Quarterly 12(2): 235 – 256.

Jensen MC, Meckling WH. 1976. Theory of the firm: managerial behavior, agency costs and owner- ship structure. Journal of Financial Economics 3(4): 305 – 360.

Jones TM. 1995. Instrumental stakeholder theory: a syn- thesis of ethics and economics. Academy of Manage- ment Review 20(2): 404 – 437.

Karpoff J, Lee D, Vendrzyk V. 1999. Defense procure- ment fraud, penalties, and contractor influence. Jour- nal of Political Economy 107(4): 809 – 842.

Financial Returns to Stakeholder Engagement 1747

Karpoff J, Lott J Jr. 1993. The reputational penalty firms bear from committing criminal fraud. Journal of Law and Economics 36(2): 757 – 802.

Karpoff J, Lott J Jr, Wehrly E. 2005. The reputational penalties for environmental violations: empirical evi- dence. Journal of Law and Economics 48: 653 – 675.

Kim H-R, Lee M, Lee H-T, Kim N-M. 2010. Corporate social responsibility and employee – company identifi- cation. Journal of Business Ethics 95(4): 557 – 569.

King AA, Lenox MJ. 2001. Does it really pay to be green? An empirical study of firm environmental and financial performance. Journal of Industrial Ecology 5(1): 105 – 116.

King G, Lowe W. 2003. An automated information extraction tool for international conflict data with performance as good as human coders: a rare events evaluation design. International Organization 57(3): 617 – 642.

King B, Soule S. 2007. Social movements as extra- institutional entrepreneurs: the effect of protests on stock price returns. Administrative Science Quarterly 52(3): 413 – 442.

Kitzmueller M, Shimshack J. 2012. Economic perspec- tives on corporate social responsibility. Journal of Economic Literature 50(1): 51 – 84.

Kobrin S. 1979. Political risk: a review and recon- sideration. Journal of International Business Studies , (Spring) 10: 67 – 80.

Kotchen MJ. 2006. Green markets and private provision of public goods. Journal of Political Economy 114(4): 816 – 834.

Kytle B, Ruggie JG. 2005. Corporate social responsibil- ity as risk management: a model for multinationals. Harvard University’s John F. Kennedy School of Gov- ernment’s Corporate Social Responsibility Working paper 10, Cambridge MA.

Lacey R. 2007. Relationship drivers of customer commit- ment. Journal of Marketing Theory & Practice 15(4): 315 – 333.

Lance CE, Butts MM, Michels LC. 2006. The sources of four commonly reported cutoff criteria: what did they really say? Organizational Research Methods 9(2): 202 – 220.

Lassonde P. 2003. How to earn your social license. Mining Review , (Summer) 7 – 13.

LeBreton JM, Senter JL. 2008. Answers to 20 questions about interrater reliability and interrater agreement. Organizational Research Methods 11(4): 815 – 852.

Levitt T. 1958. The dangers of social responsibility. Harvard Business Review 36(5): 41 – 50.

Liu Y. 2009. Investigating external environmental pres- sure on firms and their behavior in Yangtze River Delta of China. Journal of Cleaner Production 17(16): 1480 – 1486.

Loughran T, McDonald B. 2011. When is a liability not a liability? Textual analysis, dictionaries, and 10-Ks. Journal of Finance 66: 35 – 65.

Mackey A, Mackey TB, Barney JB. 2007. Corporate social responsibility and firm performance: investor preferences and corporate strategies. Academy of Management Review 32(3): 817 – 835.

Margolis J, Elfenbein H, Walsh J. 2007. Does it pay to be good? A meta-analysis and redirection of research on the relationship between corporate social and financial performance. Mimeo, Harvard Business School: Boston, MA.

Margolis J, Walsh J. 2003. Misery loves companies: rethinking social initiatives by business. Administra- tive Science Quarterly 48(2): 268 – 305.

Marquis C, Glynn MA, Davis GF. 2007. Community isomorphism and corporate social action. Academy of Management Review 32(3): 925 – 945.

Maxwell JW, Lyon TP, Hackett SC. 2000. Self-regulation and social welfare: the political economy of corporate environmentalism. Journal of Law and Economics 43: 583 – 617.

McClelland C. 1971. World Event/Interaction Survey . Inter-University Consortium for Political Research: Ann Arbor, MI.

McGuire JB, Sundgren A, Schneeweis T. 1988. Corporate social responsibility and firm financial performance. Academy of Management Journal 31(4): 854 – 872.

McWilliams A, Siegel D. 2001. Corporate social respon- sibility: a theory of the firm perspective. Academy of Management Review 26(1): 117 – 127.

Miller MH, Upton CW. 1985a. The pricing of oil and gas: some further results. Journal of Finance 40(3): 1009 – 1018.

Miller MH, Upton CW. 1985b. A test of the hotelling valuation principle. Journal of Political Economy 93(1): 1 – 25.

Mohr L, Webb D. 2005. The effects of corporate social responsibility and price on consumer responses. Journal of Consumer Affairs 39(1): 121 – 147.

Nebus J, Rufin C. 2010. Extending the bargaining power model: explaining bargaining outcomes among nations, MNEs and NGOs. Journal of International Business Studies 41(6): 996 – 1015.

Porter ME, Kramer MR. 2006. Strategy & society: the link between competitive advantage and corpo- rate social responsibility. Harvard Business Review 84(12): 78 – 92.

Porter M, Kramer M. 2011. Creating shared value. Harvard Business Review 89(1/2): 62 – 77.

Post J, Preston L, Sauter-Sachs S. 2002. Redefining the Corporation: Stakeholder Management and Organi- zational Wealth . Stanford University Press: Stanford, CA.

Rao H. 1994. The social construction of reputation: certification contests, legitimation, and the survival of organizations in the American automobile industry: 1895 – 1912. Strategic Management Journal 15(S1): 29 – 44.

Rehbein K, Waddock S, Graves SB. 2004. Understanding shareholder activism: which corporations are targeted? Business & Society 43(3): 239 – 267.

Roberts PW, Dowling GR. 2002. Corporate reputation and sustained superior financial performance. Strate- gic Management Journal 23(12): 1077 – 1093.

Rowley TJ. 1997. Moving beyond dyadic ties: a network theory of stakeholder influences. Academy of Manage- ment Review 22(4): 887 – 910.

1748 W. J. Henisz, S. Dorobantu, and L. J. Nartey

Rowley T, Berman S. 2000. A brand new brand of corporate social performance. Business & Society 39(4): 397 – 418.

Rowley TJ, Moldoveanu M. 2003. When will stakeholder groups act? An interest-and identity-based model of stakeholder group mobilization. Academy of Manage- ment Review 28(2): 204 – 219.

Salmones M, Crespo A, Bosque I. 2005. Influence of corporate social responsibility on loyalty and valuation of services. Journal of Business Ethics 61(4): 369 – 385.

Schrodt P. 1995. Event data in foreign policy analysis. In Foreign Policy Analysis: Continuity and Change in Its Second Generation , Neack L, Haney PJ, Hey JAK (eds). Prentice Hall: Englewood Cliffs, NJ; 145 – 166.

Shahzad A, David P, Sharfman M. 2011. Corporate governance and CSP: does aligning managers with shareholders help or hurt stakeholders?

Sharfman MP, Fernando CS. 2008. Environmental risk management and the cost of capital. Strategic Man- agement Journal 29(6): 569 – 592.

Smith-Crowe K, Burke MJ, Kouchaki M, Signal S. 2013. Assessing interrater agreement via the average deviation index given a variety of theoretical and methodological problems. Organizational Research Methods 16: 127 – 151.

Stuart TE. 1999. Interorganizational alliances and the per- formance of firms: a study of growth and innovation rates in a high-technology industry. Strategic Manage- ment Journal 21: 791 – 811.

Sundaram AK, Inkpen AC. 2004. The corporate objective revisited. Organization Science 15(3): 350 – 363.

Surroca J, Tribó JA. 2008. Managerial entrenchment and corporate social performance. Journal of Business Finance & Accounting 35(5 – 6): 748 – 789.

Tetlock PC, Saar-Tsechansky M, Macskassy S. 2008. More than words: quantifying language to measure firms’ fundamentals. Journal of Finance 63(3): 1437 – 1467.

Thomson I, Boutilier RG. 2011. Social license to operate. In SME Mining Engineering Handbook , Darling P (ed). Society for Mining, Metallurgy and Exploration: Littleton, CO; 1779 – 1796.

Turban D, Greening D. 1997. Corporate social perfor- mance and organizational attractiveness to prospective employees. Academy of Management Journal 40(3): 658 – 672.

Villalonga B. 2004. Intangible resources, Tobin’s q, and sustainability of performance differences. Jour- nal of Economic Behavior & Organization 54(2): 205 – 230.

Vogel D. 2005. The Market for Virtue: The Potential and Limits of Corporate Social Responsibility . Brookings Institution Press: Washington, DC.

Waddock S. 2000. The multiple bottom lines of corporate citizenship: social investing, reputation, and respon- sibility audits. Business and Society Review 105(3): 323 – 345.

Waddock S, Graves S. 1997. The corporate social performance-financial performance link. Strategic Management Journal 18(4): 303 – 319.

Weigelt K, Camerer C. 1988. Reputation and corporate strategy. Strategic Management Journal 9: 443 – 454.

Zaheer S. 1995. Overcoming the liability of foreignness. Academy of Management Journal 38(2): 341 – 363.

Zingales L. 2000. In search of new foundations. Journal of Finance 55(4): 1623 – 1653.

Zivin JG, Small A. 2005. A modigliani-miller theory of altruistic corporate social responsibility. BE Journal of Economic Analysis & Policy 5(1): 1 – 21.