SEARCH

SEARCH BY CITATION

Keywords:

  • Corporate Governance;
  • Corporate Social Responsibility (CSR);
  • Behavioral Theory of the Firm (BTOF);
  • Attainment Discrepancy;
  • Organizational Slack

ABSTRACT

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. THEORY AND HYPOTHESES DEVELOPMENT
  5. METHODS
  6. RESULTS
  7. LIMITATIONS
  8. DISCUSSION AND CONCLUSION
  9. ACKNOWLEDGEMENTS
  10. REFERENCES
  11. Appendices

Manuscript Type: Empirical

Research Question: Is the relationship between corporate governance mechanisms and corporate social responsibility (CSR) contingent on satisfaction with firm performance?

Research Findings/Insights: Our results suggest that while effective corporate governance discourages both positive (proactive stakeholder relationship management) and negative (violation of regulations and standards) CSR, higher slack and positive attainment discrepancy lead to higher positive and lower negative CSR, respectively. More significantly, we find that the association between effective corporate governance and both positive and negative CSR depends on satisfaction with firm performance as indicated by the levels of slack and attainment discrepancy. Put simply, the impact of corporate governance on positive CSR is more pronounced under low slack/negative attainment discrepancy conditions, and that on negative CSR is more pronounced under high slack/positive attainment discrepancy conditions.

Theoretical/Academic Implications: Our study provides robust support for the behavioral theory of the firm. Previous research has not adequately considered the role of satisfaction with firm performance in studying the impact of corporate governance on managerial decision-making. We show that the association between corporate governance and CSR dimensions depends on differences in decision-making latitude originating from relative firm performance compared to those of peer firms.

Practitioner/Policy Implications: First, to understand how effective corporate governance can constrain positive CSR and more importantly reduce negative CSR. Second, to appreciate that the effectiveness of an organization's governance mechanisms is contingent on slack and performance and the marginal returns from improving governance mechanisms when things are going well may be low.


INTRODUCTION

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. THEORY AND HYPOTHESES DEVELOPMENT
  5. METHODS
  6. RESULTS
  7. LIMITATIONS
  8. DISCUSSION AND CONCLUSION
  9. ACKNOWLEDGEMENTS
  10. REFERENCES
  11. Appendices

Management scholars have been interested in understanding the impact of corporate governance mechanisms such as ownership and boards of directors on corporate social responsibility (CSR) ratings (Coffey & Fryxell, 1991; Johnson & Greening, 1999; Waddock & Graves, 1997). Scholars examining ownership implications argue that institutional owners, the dominant class of owners, are myopic and concerned with quarterly performance targets, and therefore, reduce CSR expenditures, given the long-term horizons and uncertain outcomes associated with them (Coffey & Fryxell, 1991). Other scholars argue that institutional investors cannot exit the firm very easily, therefore undertake more CSR to mitigate the risk of adverse regulatory action, higher compliance costs, consumer retaliation, and so on (Neubaum & Zahra, 2006; Spicer, 1978). To resolve this paradox, scholars posit that different types of institutional owners may have different interests in CSR. For example, Johnson and Greening (1999: 564) argue that “some categories of institutional investors act more as traders concerned predominantly with quarterly earnings and that others act as long-term investors … more concerned with a firms social performance because it may impact financial performance over time.” Nonetheless, the empirical evidence continues to be mixed (Coffey & Fryxell, 1991; Coffey & Wang, 1998; Graves & Waddock, 1994; Johnson & Greening, 1999; Kassinis & Vafeas, 2002; Neubaum & Zahra, 2006).

Similarly, scholars examining board implications find that the proportion of independent directors on the board has a diametrically opposite impact on CSR depending on the studies considered (Coffey & Wang, 1998; Johnson & Greening, 1999; Kassinis & Vafeas, 2002; Kesner & Johnson, 1990; Wang & Coffey, 1992). Some scholars argue that since the selection of a greater number of independent directors signals the firm's intent to pay greater attention to its external environment and legitimacy (Pfeffer & Salancik, 1978), it should be associated with increased CSR expenditures (Johnson & Greening, 1999). Others argue that as directors are hired primarily to protect shareholders' interests: “an effective board may actually serve to screen and eliminate philanthropic intentions” (Coffey & Wang, 1998: 1598). Further, because a vast majority of these directors are hired for their financial expertise (Fligstein, 1991), it may be much easier for them to evaluate historical financial information than to make uncertain strategic decisions such as on R&D, internal innovation, entrepreneurship, and CSR (e.g., Baysinger & Hoskisson, 1990; Deutsch, 2005; Lorsch & MacIver, 1989). In summary, as further encapsulated in Appendix 1, the relationship between various governance mechanisms and CSR is still far from clear.

To resolve the ambiguity surrounding these findings, this study makes three distinctive advances. First, we theorize that one of the reasons for lack of clarity on the relationship between corporate governance and CSR could relate to the substitution effect (e.g., Rediker & Seth, 1995), which refers to the interdependence among various governance mechanisms. Unlike previous research that usually assesses the implications of various corporate governance mechanisms in isolation (Coffey & Wang, 1998; Johnson & Greening, 1999; Kesner & Johnson, 1990), we adapt the recommendations of Agrawal and Knoeber (1996) and use four governance variables in our model: independent director representation, concentrated institutional shareholding, managerial ownership, and strength of shareholder rights. Managerial ownership – the first level of governance – is expected to provide a direct incentive to managers to undertake value-maximizing behavior (e.g., Amihud & Lev, 1981; Davis, 1991; Denis, Denis, & Sarin, 1997; Gedajlovic & Shapiro, 2002; Morck, Shleifer, & Vishny, 1988). Independent directors, tasked with supervision of managerial decision-making on behalf of shareholders, are the second layer of governance arrangements. Concentrated institutional owners – the third layer – are assumed to have both the ability and the means to supervise managerial decision-making, and thus are expected to act as a secondary means of securing principals' (owners') tighter control over their agents (managers). Lastly, the threat of takeover by other firms operates as the final check on the agents, which essentially implies that if the firms are not well managed they would be good candidates for takeover by those who believe they can manage them better. Our choice of these governance mechanisms not only addresses substitution possibilities within the internal governance mechanisms (managerial ownership, institutional ownership concentration, outsiders on boards) but also considers the potential effects of strong shareholder rights (or the lack thereof) for CSR.

Second, previous research has come under increasing criticism for combining positive and negative dimensions of CSR (Chiu & Sharfman, 2009; Godfrey, Merrill, & Hansen, 2009; Kacperczyk, 2009; Mattingly & Berman, 2006; Strike, Gao, & Bansal, 2006). This literature suggests that positive CSR acts such as sustainable practices, commitment-based employment practices, corporate philanthropy and effective relations with local community are not on the same continuum as avoiding negative CSR acts such as violations of regulatory guidelines on environment or equal employment opportunities, health and safety concerns, or controversial actions such as on human or employment rights. While positive CSR involves proactive stakeholder relationship management, negative CSR involves reactive compliance with minimum standards, and hence these should not be combined. In deference to these studies, we make two separate composite ratings – positive and negative CSR – and run separate regressions models for each of them. We believe this helps us in significantly advancing the debate on the nature of the relationship between governance and CSR.

Finally, previous research (e.g., Waddock & Graves, 1997) suggests that when firms perform well, they are more likely to invest in CSR. We formally incorporate this idea by using theoretical concepts based in the behavioral theory of the firm (Cyert & March, 1963) and examine how the concept of attainment discrepancy – the difference between actual and aspired performance – determines levels of CSR. We suggest that when a firm is perceived to be doing well, independent directors or concentrated owners may: 1) not feel the need for close monitoring; and 2) place greater trust in managers' judgment, giving them greater latitude in decision-making. Moreover, in such situations, managers are also likely to deal with their monitors from a position of strength. Conversely, if the firm is perceived to be not doing well, managers may not have much decision-making latitude even under relatively weak governance conditions. We make similar arguments about another behavioral theory of the firm (BTOF) factor, the concept of slack and how it relates to decision-making about CSR. Thus, we theorize that under identical governance conditions, managers could have vastly different decision-making latitude based on the two BTOF factors, namely, attainment discrepancy and slack.

THEORY AND HYPOTHESES DEVELOPMENT

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. THEORY AND HYPOTHESES DEVELOPMENT
  5. METHODS
  6. RESULTS
  7. LIMITATIONS
  8. DISCUSSION AND CONCLUSION
  9. ACKNOWLEDGEMENTS
  10. REFERENCES
  11. Appendices

Corporate Governance and CSR

In order to clarify the nature of relationship between corporate governance and CSR, it is important to make a distinction between positive and negative CSR so that we can separately examine the implications of corporate governance for both enabling effective decision-making (e.g., proactive sustainability practices) and preventing poor decision-making (e.g., violation of environmental regulations). This is important not just from an empirical perspective – previous research (e.g., Mattingly & Berman, 2006) highlights that these two dimensions of CSR do not load together in factor analysis, which may be one of the reasons for inconsistent findings – but also from a theoretical viewpoint, which would suggest that effective governance should always curtail negative CSR, while determining the levels of positive CSR based on a cost-benefit analysis. Therefore, to the extent that good governance is associated with better monitoring, in general, we expect it to be associated with lower negative CSR, as the failure to comply with rules and regulations can lead to penalties and bad publicity that effective monitors would consider avoidable. This insight is valuable as previous governance research has predominantly focused on the upsides of effective governance (i.e., value creation, accounting profits, etc.) and rarely considered the benefits of preventing potential downside losses and associated costs.

However, the relationship between effective governance and positive CSR is a little more complex because while positive CSR has potential benefits for firm performance, these benefits are more long-term and uncertain in nature. Thus, mainly those who have a long-term interest in the firm may prefer them. In contrast, if these governance mechanisms focus on the short-term, then the costs of CSR will likely outweigh the benefits. For example, institutional owners concerned with meeting their short-term goals and achieving their performance targets may not want managers to invest in CSR due to goal conflicts relating to time horizons and uncertainty of outcomes (e.g., Bushee, 1998). This may result in a pressure on the managers to reduce positive CSR, which would be especially true for those institutional investors who are primarily short-term, momentum traders (Neubaum & Zahra, 2006), those who prefer to remain passive (Edwards & Hubbard, 2000; Pound, 1992; Wahal, 1996), or those who have fairly diversified and indexed portfolios (Dharwadkar, Goranova, Brandes, & Khan, 2008).

Similarly, since selection of a greater number of independent directors signals a firm's intent to pay greater attention to its external environment and legitimacy (Pfeffer & Salancik, 1978), it could lead to higher other hand, because these directors are primarily appointed to protect shareholders' interests, independent boards may consider higher positive CSR to be not in the interest of the firm. Baysinger and Hoskisson (1990) and Lorsch and MacIver (1989) support this contention by arguing that a vast majority of independent directors are hired by financial institutions for their financial expertise, which means they likely find it much easier to evaluate historically available financial information rather than uncertain strategic information. They are primarily the agents of the shareholders, a majority of which are financial institutions with short-term interests. Thus, they should find it much easier to justify short-term gains than long-term uncertain investments. In turn, investment in R&D, internal innovation, entrepreneurship, and other functions with uncertain returns tends to display a negative relationship with greater outsider representation on boards (Baysinger, Kosnik, & Turk, 1991; Deutsch, 2005; Hill & Snell, 1988; Hoskisson, Hitt, Johnson, & Grossman, 2002; Zahra, 1996). Because CSR shares these characteristics, greater outsider representation may lead to lower CSR investments.

While some attention has been paid to boards and owners in this domain, limited research has examined the implications of managerial ownership and shareholder rights for CSR. In both cases, the effects on negative CSR are clear. High managerial ownership and greater shareholder rights that allow for market interventions should reduce negative CSR. Similar to our earlier arguments, their effect on positive CSR would depend on the time horizons of the CEO as well as the type of owners of the firm, and short-term horizons on part of managers and owners will reduce CSR.

Our arguments at the firm level can also be augmented using Mackey, Mackey, and Barney (2007), who propose that the relationship between CSR and firm value may depend on demand for and supply of socially responsible investments, and only when its demand exceeds supply the firm may benefit from it. They further argue that “from a broader theoretical perspective, the entire effort to discover how socially responsible activities can increase the present value of a firm's future cash flows is problematic. After all, the essential point of many business and society scholars is that … the firms should sometimes engage in activities that benefit employees, suppliers, customers, and society at large, even if those activities reduce the present value of the cash flows generated by the firm” (2007: 818). While this value-reducing behavior may be acceptable to socially responsible investors, it may not be acceptable to other investors. As the Social Investment Forum (2010) suggests that about one in eight dollars under professional management in the United States today uses some form of socially responsible investing (SRI) (Geczy, Stambaugh, & Levin, 2003), we argue that given the current levels of SRI, the demand for SRI is likely to be lower than the supply of SRI and therefore effective governance structures will ensure that managers act in the interest of their principals. In summary, overall the relationship between effective governance and negative CSR is clear-cut as the downside costs will encroach on firm value. Similarly, the benefits-costs tradeoffs at the firm level along with the broader implications of demand and supply of SRI would suggest that under the current circumstances, effective governance should also reduce positive CSR.

Hypothesis 1a. Effective corporate governance is negatively associated with positive CSR.

Hypothesis 1b. Effective corporate governance is negatively associated with negative CSR.

Behavioral Theory of the Firm (BTOF) and CSR

Ever since Cyert and March (1963) developed the view of organizations as coalitions of stakeholders, management scholars have been cognizant that unresolved conflict is an important feature of organizations and the coalitions within them. Most types of resource allocation decisions within organizations are therefore an outcome of the coalition bargaining processes, which may either depend on the primacy of certain stakeholders or be subject to negotiations across coalitions of stakeholders. Most resource allocation decisions that lead to corporate social programs can be viewed in this light, yet research has not accorded due importance to such processes (Barnett, 2007; Scherer & Palazzo, 2007).

In particular, research has not considered the role of shareholder satisfaction, and how that might affect managers' ability to allocate resources for CSR. This is an important issue because the benefits from CSR, like benefits from any other strategic action such as innovation activity, are uncertain and unclear. If such is the case, shareholders should be more inclined to trust their agents to make the “right” decisions when they are satisfied than when they are not. When a firm has abundant discretionary resources and its actual performance exceeds aspirations, managers will have significant discretionary powers, from a coalitional perspective, even under strong governance conditions. However, when the converse is true, decision-making latitude for managers will be far more constrained and possibly driven by short-term concerns for cost cutting in reaction to weak performance. Therefore, it is important to consider these two key factors from BTOF – attainment discrepancy and organizational slack – that have implications not only for corporate governance (i.e., balance of power between owners and managers) but also provide insight into why firms may engage differently in positive and negative CSR.

Attainment Discrepancy and CSR.  BTOF highlights the importance of firms as systems with aspirations and performance expectations (Cyert & March, 1963). The firm's relative performance with respect to its past performance or industry peers has important and well-documented implications for resource allocation decisions (Lant, 1992). Specifically, firms performing below the industry average aspire to reach the industry average, and firms performing above the mean for their industry aspire to improve on their past performance (Bromiley, 1991; Fiegenbaum & Thomas, 1988). Lant (1992: 624) developed and labeled this concept as “attainment discrepancy,” which simply reflects the difference between actual and aspired performance. If actual performance is better than aspired performance, a positive attainment discrepancy results; and if it is lower, a negative attainment discrepancy occurs.

In the case of positive attainment discrepancy, shareholders should repose greater trust in managers and allow them higher discretion in resource allocations. However, in the case of negative attainment discrepancy, managerial discretion may be limited (Bromiley, Miller, & Rau, 2001) as shareholders are less trusting of managerial decision-making and more prone to pressuring them towards meeting shareholders' goals. For example, if a firm aspired towards a return on asset of 3 per cent and its actual profits turn out to be 5 per cent, shareholders would be more willing to allow managers to invest a portion of the higher earnings on CSR than if its actual profits turn out to be 1 per cent, in which shareholders may want to reduce CSR allocations to a bare minimum. Thus, managers probably have higher discretion in paying attention to social domains in the case of positive attainment discrepancy than in case of negative attainment discrepancy. Alternatively, poor performance on this BTOF dimension may pressure managers into cutting corners in order to improve performance, a condition likely to be associated with increased negative CSR.

Hypothesis 2a. Positive attainment discrepancy is positively associated with positive CSR.

Hypothesis 2b. Positive attainment discrepancy is negatively associated with negative CSR.

Organization Slack and CSR.  Organization slack, an important behavioral theory construct, signifies the existence of a “cushion of actual or potential resources” that enables the firm to adapt to internal or external necessities for strategic change (Bourgeois, 1981: 30). The availability of resources not only provides firms with the opportunities to commit resources to social causes (e.g., Waddock & Graves, 1997), but also makes them less resistant to stakeholders' demands.

While a few recent studies have examined the role of slack on determining levels of CSR, they generally used financial performance as a proxy for slack (e.g., Amato & Amato, 2007; Waddock & Graves, 1997). The use of this proxy is problematic because the relationship between slack and financial performance itself is not very clear; some previous research indicates a positive relationship and others reveal a curvilinear relationship such that there must be a point beyond which slack becomes a wasted resource (e.g., Nohria & Gulati, 1996). The other problem with the use of financial performance as a measure of slack is that unlike recent studies on the role of slack in determining various organizational outcomes (e.g., George, 2005), it does not really distinguish between high discretion (uncommitted liquid resources) and low discretion (absorbed costs) components of slack, which is important because the latter is hard to recover and may provide little discretion to management. Therefore, as a first step, research should use only high-discretion slack measures in studying the CSR (Arora, 2008), which with the exceptions of Navarro (1988) and Seifert, Morris, and Bartkus (2004), studies have not done.

Second, high discretion slack itself can be divided into two categories – available slack and potential slack, where potential slack refers to firm's capacity to quickly raise cash resources, if required. Navarro (1988) studied the impact of potential slack (measured by debt/equity ratio) only. Therefore, an improved research design should include both available and potential slack. Along the lines of our previous arguments regarding attainment discrepancy, we expect higher slack to enable higher positive CSR and lower slack to induce increased negative CSR.

Hypothesis 3a. High discretion slack is positively associated with positive CSR.

Hypothesis 3b. High discretion slack is negatively associated with negative CSR.

Integrating Corporate Governance, Behavioral Theory of the Firm, and CSR

The implications of governance for CSR should be contingent on the resource endowment situation of the firm, and these are best understood with the help of a visual depiction. While Figure 1 shows that the relationship between governance and CSR is moderated by satisfaction with firm performance, Figure 2 provides more specific details on the nature of the hypothesized relationships. In Figure 2, along the horizontal axis, we consider managerial decision-making latitude about CSR based on satisfaction with firm performance as determined by availability of slack resources and achievement of financial aspirations. Along the vertical axis, we consider corporate governance effectiveness as determined by levels of managerial ownership, independent director representation, concentrated institutional ownership, and shareholder rights. While considering the relationships portrayed in Figure 2, it is important to keep in mind that we are talking about general trends and not specific firm strategies. Some firms may – and indeed do – defy these expectations, but in general, we expect these relationships to hold.

image

Figure 1. Satisfaction with Firm Performance as a Moderator in the Relationship Between Corporate Governance and CSR

Download figure to PowerPoint

image

Figure 2. Specific Relationships Between Corporate Governance, Satisfaction with Firm Performance, and Positive and Negative CSR

Download figure to PowerPoint

We theorize that effective corporate governance and low slack and negative attainment discrepancy represent a situation of least managerial discretion regarding positive CSR expenditures. Also, while low slack and negative attainment discrepancy may provide the enabling conditions for involvement in negative CSR, effective governance should curtail that tendency. Overall, we expect effective governance factors to dominate behavioral theory factors in influencing managerial discretion regarding CSR expenditures. In sharp contrast, we theorize that ineffective corporate governance in conjunction with high slack and high attainment discrepancy should enable managers to undertake a high level of positive CSR, while reducing their need to be involved in negative CSR. This quadrant reflects a situation of high managerial discretion with respect to CSR activities. Therefore, we expect behavioral theory factors to dominate corporate governance factors.

In contrast to these two situations (wherein one factor dominates the other), firms in the top right quadrant have both strong corporate governance and high satisfaction with firm performance. Under these circumstances, the managerial tendency to engage in positive CSR is likely to be curtailed in the face of effective governance, given our earlier arguments. Moreover, the implications for negative CSR are very clear in this quadrant – both resource and governance factors in combination should reduce negative CSR. Finally, firms in the bottom left quadrant have weak governance and a weak resource situation. While the resource situation may constrain positive CSR (despite the weak governance context), the combination of ineffective governance and a weak resource situation may provide the enabling condition for very high levels of negative CSR.

Hypothesis 4a. Effective corporate governance is less negatively associated with positive CSR under conditions of high slack and positive attainment discrepancy.

Hypothesis 4b. Effective corporate governance is more negatively associated with negative CSR under conditions of high slack and positive attainment discrepancy.

METHODS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. THEORY AND HYPOTHESES DEVELOPMENT
  5. METHODS
  6. RESULTS
  7. LIMITATIONS
  8. DISCUSSION AND CONCLUSION
  9. ACKNOWLEDGEMENTS
  10. REFERENCES
  11. Appendices

Study Sample

We draw our sample from the S&P 500 and KLD Domini 400 Universe. First, we collect social performance ratings for all the firms, included in these two indices, during the period of 2001–2005 by the firm Kinder, Lyndenberg, and Domini (KLD). Second, we obtain information about institutional ownership, corporate governance, and financial performance for these firms. The information about institutional ownership comes from the CDA/Spectrum Thomson Financial's 13F database; corporate governance is based on the RiskMetrics (IRCC) database; and financial performance comes from the Compustat North America database. We lead the dependent variable by a year to ensure that the independent variables predate the dependent variable. Thus, the information about the independent variables pertains to the years 2000 to 2004. Missing data brought the final sample down to 1522 observations for 518 firms. Appendix 2 details these firms by type of industry.

Dependent Variables

Corporate Social Responsibility.  We use archival ratings of CSR as the dependent variable, obtained from the firm KLD Inc. The use of KLD ratings of corporate social responsibility is fairly standard in the literature; the ratings consider all firms in Standard & Poor 500 or KLD Domini 400 Universe for the period 2001–2005 on 82 indicators in eight major social performance categories: community relations, employee relations, diversity, environment, product quality, governance and transparency, human rights, and other concerns.

While previous research has tended to transform all these dimensions into a composite index or rating (e.g., David, Bloom, & Hillman, 2007; Hillman & Keim, 2001; Waddock & Graves, 1997), there is growing consensus now that positive and negative CSP are different constructs that should not be combined (Chiu & Sharfman, 2009; Godfrey et al., 2009; Kacperczyk, 2009; Mattingly & Berman, 2006; Strike et al., 2006). Recent research suggests that KLD Strengths that primarily relate to corporate philanthropy, gender and racial diversity, good union relations, green products or processes, innovation, and the like is not on a continuum with concerns that relate to violations of regulatory frameworks set by agencies such as Equal Employment Opportunity Commission (EEO), Occupational Safety and Health Administration (OSHA), Environment Protection Agency (EPA), and/ or Fair Trade Commission (FTC), and hence should not be combined. Given these findings, we make two separate composite ratings – positive and negative CSR ratings – and run separate regressions models for each of them.

Independent Variables

Corporate Governance.  To code for corporate governance in different firms, we use four approaches. First, concentrated ownership may be the most effective mechanism to reduce agency problems (e.g., Hoskisson et al., 2002; Kang & Sorensen, 1999; Morck et al., 1988), and financial institutions have both better incentives and better means to monitor mangers. While previous research has generally used total institutional ownership, following Laidroo (2009), we sum up institutional ownership for only those institutions that own 5 per cent or more shares in a firm as they are more likely to have incentives to monitor.

Second, managerial ownership provides the managers with incentives to undertake value-maximizing and not value-destroying behavior (e.g., Amihud & Lev, 1981; Davis, 1991; Denis, Denis, & Sarin, 1997; Gedajlovic & Shapiro, 2002; Morck et al., 1988). Thus, we include managerial ownership in our empirical model, and following Chen (2008) we measure it as the percentage of total equity owned by the CEO. This is also in line with the research that has shown CEO power as a predictor of number of organizational phenomena from opportunistic behavior to executive compensation (Finkelstein & Boyd, 1998).

Third, greater independent director representation provides an important characteristic of a corporation's interest in both improving governance and seeking external legitimacy (Ahmed & Duellman, 2007; Johnson & Greening, 1999). Therefore, we include the proportion of independent directors as a corporate governance variable in our model. Following Chhaochharia and Grinstein (2007) and Chen (2008), we define independent directors as those directors who were neither employed by nor affiliated in any other way to the firm.

Fourth, we borrow Gompers, Ishii, and Metrick's (2003) governance index (GIM), which measures shareholder rights at different corporations and is fairly common in research on corporate governance (e.g., Bebchuk & Cohen, 2005). All information about the corporate governance variables comes from RiskMetrics (formerly, IRCC) database.

Attainment Discrepancy in Financial Performance.  We use two measures, one accounting and one market-based, to calculate the attainment discrepancy in financial performance. First, we calculate the return on assets (ROA) and market to book ratio (MBR). The ROA serves as an accounting measure of performance and is fairly standard in past research (e.g., Waddock & Graves, 1997). The MBR is a modified version of Tobin's Q, as commonly used in existing literature (e.g., Dutta, Narasimhan, & Rajiv, 2005; Richard, Murthi, & Ismail, 2007), which we adopt because it is difficult to assess the replacement cost of assets, which provides the denominator in the Tobin's Q equation. Furthermore, MBR captures the relative success of firms in maximizing shareholder value through the efficient allocation and management of scarce resources.

After calculating the ROA and MBR, we measure attainment discrepancy using the same method as Bromiley (1991) – we code industry average as the aspired performance for firms that perform below the industry average, whereas for firms performing above this average, we multiply their past performance by 1.05, which signifies a 5 per cent increase.1 Finally, attainment discrepancy is the difference between aspired and actual performance. Thus, a positive attainment discrepancy indicates that actual performance exceeded aspired performance and a negative attainment discrepancy its converse.

Organization Slack.  We use two measures – cash and account receivables; and debt-to-equity ratio – to compute slack. Cash and accounts receivables (available slack) and low debt-equity ratio (potential slack) are widely used measures of high-discretion slack (e.g., Navarro, 1988). We log transform cash and accounts receivables as this variable is often skewed and may violate assumption of normality.

Control Variables

We control for industry, firm size, research and development intensity, product differentiation, market growth, demand instability, and industry structure concentration, capital intensity, dividend payouts, and CEO age and CEO tenure.

Industry.  To control for differences in industrial munificence, we use the 13-industry classification of Waddock and Graves (1997). The sample provides a fairly representative cross-section of the US economy: approximately 19.6 per cent of the firms belong to the computers, autos, and aerospace industries; 11.5 per cent come from the telephone and utilities; and another 11.5 per cent from the bank and financial services industries, and the rest are spread across the spectrum of remaining industries quite evenly.

Firm Size.  Larger firms tend to attract more attention and pressure to respond to stakeholders' demands (Burke, Logsdon, Mitchell, Reiner, & Vogel, 1986). Whereas most previous research measures size according to total sales and assets of the corporation, we prefer to measure it as the number of employees. Because we use measures like ROA and R&D intensity, which involve total assets and sales in their denominator, a size measure based on assets or sales could cause multicollinearity. The use of the number of employees avoids this problem without any loss of information. Further, we log transform this variable as firm size is often skewed and may violate assumption of normality.

Research and Development Intensity. McWilliams and Siegel (2000) demonstrate that research and development intensity offers other important variable with a bearing on the relationship between CSR and financial performance. They also argue that a model of the relationship between CSR and financial performance that fails to include R&D intensity will be misspecified and fundamentally flawed, because such a relationship could reflect simply the impact of R&D on firm performance. Therefore, we control for R&D intensity, defined as R&D expenditure divided by sales.2

Product Differentiation.  Previous research (e.g., Finkelstein & Boyd, 1998; McWilliams & Siegel, 2000) has also highlighted the need to control for product differentiation in studying CSR. It is because those firms that invest in CSR are also likely to invest on advertising and branding for product differentiation. Because Compustat data on advertisement expenditure is sparse and unreliable, we use SGA intensity ratio – measured as ratio of selling, general and administrative expenses to sales ratio – to proxy for product differentiation.

Industry-Level Managerial Discretion.  We also control for certain important differences in industry structures that limit or advance managerial discretion. For this purpose, we use market growth, demand instability, and industry structure concentration suggested by Finkelstein and Boyd (1998). We use the same methodology as Finkelstein and Boyd (1998) to capture these three constructs: Market growth or industry munificence is captured by the regression slope coefficient divided by mean sales, where the coefficient is based on regression of time against value of total sales, while estimates for any given year are based on the five preceding years, i.e., estimate for year 2001 is based on data from 1996–2000. Demand instability is measured similarly by dividing the standard error of the regression slope coefficient by mean sales for the industry. Industry concentration is measured using the Herfindahl Index, computed by using all the firms for that industry in the sample.

Capital Intensity.  Capital intensity tends to place limitations on managerial discretion. Finkelstein and Boyd (1998) believe that by creating rigidity in organizations, capital intensity makes it difficult for firms to accommodate changes in strategy. Thus, it is important to control for capital intensity, which is measured as value of total plant, property and equipment divided by number of employees.

Cash Dividend.  Since Jensen (1986) first developed the argument for returning free cash (cash in excess of a firm's requirements or that would be invested in low-return projects) as dividends to the shareholders, it has become axiomatic in agency theory to control for cash dividends paid out before detecting agency losses. Because we argue that CSR might signify agency losses in some circumstances, it is important to control for cash dividends paid out in our empirical model.

CEO Age and Tenure.  Finally, we also control for CEO age and tenure because these factors have been shown to be significant in determining CEO power, which is an important predictor of a CEO's owning responsibility for strategic change, especially in a high discretion environment (Finkelstein & Hambrick, 1990; Haleblian & Finkelstein, 1993; Hambrick, Geletkanycz, & Fredrickson, 1993).

Analyses

We employ time-series, cross-sectional regression analysis (random effects model), because cross-sectional studies can lead to biased or misleading estimates (Finkelstein & Boyd, 1998). The use of a longitudinal methodology enables us to isolate the effects of specific actions and treatments over time and across sections (Hill & Phan, 1991). By including temporal lags and controls for prior states of variables, it also helps us in empirically establishing the causality mechanisms (Hambrick, 2007). Moreover, the use of a random-effects model is preferable from the perspective of generalizing the findings beyond the sample under study (Maddala, 2002).

Additionally, since corporate governance, CSR and firm performance may be endogenously related, we perform sensitivity analysis using the Hausman-Taylor panel data regression for endogenous covariates. The estimators, originally proposed by Hausman and Taylor (1981) are based on instrumental variables models, and assume that some of the explanatory variables may be correlated with the individual-level random effects, but that none of the explanatory variables are correlated with the idiosyncratic error. On reviewing the performance of various panel data models using Monte Carlo simulation experiments, prominent econometricians like Baltagi, Bresson, and Pirotte (2003) highly recommend this model, and this technique has been used in studies like Palia (2001) to perform endogeniety tests. For the purpose of this modeling, we assumed that all governance and social responsibility variables might be endogenously related to each other. The results from this analysis are available on request.

RESULTS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. THEORY AND HYPOTHESES DEVELOPMENT
  5. METHODS
  6. RESULTS
  7. LIMITATIONS
  8. DISCUSSION AND CONCLUSION
  9. ACKNOWLEDGEMENTS
  10. REFERENCES
  11. Appendices

We report the means, standard deviations, and correlations in Table 1. The mean value for positive CSR ratings is 2.53 with a standard deviation of 2.49, and the mean value for negative CSR ratings is 2.62 with a standard deviation of 2.27; the correlation between the two dimensions is .31 (p < .01). Both these dimensions are correlated with firm size, dividends paid out, cash and accounts receivables, debt-equity ratio, SGA expenses, CEO age, CEO tenure, CEO ownership, governance index, institutional ownership, and attainment discrepancy in ROA. The descriptive statistics, including the correlation matrix, with respect to the other measures also appear in the table.3

Table 1. Descriptive Statistics and Correlation Matrix
#Variable NameMeanS.D.12345678910111213141516171819
  1. Notes:

  2. 1. Table reports descriptive information for all variables, if applicable, after Winsorization (at .01 level) but before standardization [normalization to (0,1)] for regression models.

  3. 2. Firm size and cash are represented by logged values of number of employees and cash and account receivables, respectively.

  4. 3. Correlation values higher than .06 are significant at p < .01, between .04 and .06 at p < .05, and between .03 to .04 at <.1.

 1CSR positive ratings2.532.49                   
 2CSR negative ratings2.622.37.31                  
 3Firm size9.541.48.34.39                 
 4Dividend per share.56.6.17.33.14                
 5R&D intensity.07.09.03−.05−.31−.15               
 6Market growth.05.04.01.09.08.13−.17              
 7Demand instability.07.03−.02.02.05−.24.01−.22             
 8Capital intensity4.971.280.31−.24.31.09.15−.16            
 9Industry concentration.11.14−.01.07.08−.12−.02−.07.93−.07           
10Cash & accounts receivables7.251.74.47.35.5.25.12.08−.120−.05          
11Debt equity ratio1.012.08.11.13.05.12−.06.04−.12.06−.07.23         
12SGA expenses.26.15.12−.24−.3−.16.52−.2.12−.230−.04−.16        
13CEO Age (years)59.948.5.08.13.15.13−.05.04−.01.08.02.17.01−.1       
14CEO Tenure (years)11.788.32−.09−.1.0100.050−.01.03.03−.050.39      
15CEO ownership.02.07−.08−.14−.08−.09.010.1−.08.08−.14−.06.04.07.36     
16Percent independent directors.61.19−.03.04.04.04−.13.04−.02.02.02.03−.01−.02.01−.01.02    
17Governance index9.762.51−.09−.05.05.14−.15.02−.03.01.01−.1−.08−.14.05−.04−.13.08   
18Block institutional ownerships.15.12−.16−.1−.11−.17−.06−.05.08−.1.06−.24.06−.05−.02−.01.05−.02−.01  
19MBR Attainment discrepancy−1.023.87−.01.02.02.09−.09.04−.06.08−.020.29−.080.02−.05.05.04−.02 
20ROA Attainment discrepancy−.03.07.05−.02.07.17−.24.09−.09.07−.01.06−.01−.16.07.07−.03.05.08−.11.09

In Tables 2 and 3, we provide the results of the random-effects panel data regression models on standardized (normalized) variables for the period 2001–2005.4 While Table 2 reports results for positive CSR ratings as the dependent variable, Table 3 reports results with negative CSR ratings as the dependent variable.

Table 2. Attainment Discrepancy, Slack, Governance, and Positive CSR Ratings: Random-Effects Panel Data Regression Results
DV: Positive CSR RatingsModel 1Model 2Model 3Model 4Model 5
  1. Significance levels: †p < .10; *p < .05; **p < .01; ***p < .001.

  2. Coefficients for dummy industry variables not reported for the sake of brevity.

Constant.641.171.121.171.01
Industry dummy coefficients
Firm size.36**.18**.19**.18**.20**
R&D intensity−.01−.06−.06−.06−.04
Product differentiation.13**.12*.12*.12*.12*
Cash dividends paid-out.05.09*.07.09**.08
Market growth.07**.08**.08**.08**.08**
Demand instability−.20*−.21*−.22*−.21−.19
Capital intensity.16**.15**.15**.15**.15**
Industry concentration.28*.24.24.24.24
CEO Age.02.01.01.01.01
CEO Tenure−.04*−.03−.02−.03−.02
Percent independent directors (ID) −.13*−.13*−.11*−.10*
Governance index (GIM) −.02−.02−.02−.03
Block institutional ownership (ALL5) −.04*−.06*−.04*−.06**
CEO ownership (CEO_OWN) −.12**−.12**−.12**−.14**
Cash & accounts receivables (CASH) .34***.34***.34***.33***
Debt equity ratio (DE) .00.01.00.01
MBR Attainment discrepancy (MBRD) .05*.07*  
ROA Attainment discrepancy (ROAD)   .02.03
ID × CASH  .07* .07*
GIM × CASH  −.08* −.07
ALL5 × CASH  .05* .06*
CEO_OWN × CASH  −.10* −.10*
ID × DE  .01 .00
GIM × DE  .02 .01
ALL5 × DE  .01 −.01
CEO_OWN × DE  .02 .02
ID × MBRD  .00  
GIM × MBRD  −.02  
ALL5 × MBRD  −.12*  
CEO_OWN × MBRD  .07*  
ID × ROAD    .00
GIM × ROAD    −.03*
ALL5 × ROAD    .00
CEO_OWN × ROAD    −.01
R2.27***.36***.40***.34***.40***
Change in R2 .09***.05***.07***.06***
N17731522152215221522
Table 3. Attainment Discrepancy, Slack, Governance, and Negative CSR Ratings: Random-Effects Panel Data Regression Results
DV: Negative CSR RatingsModel 6Model 7Model 8Model 9Model 10
  1. Significance levels: †p < .10; *p < .05; **p < .01; ***p < .001.

  2. Coefficients for dummy industry variables not reported for the sake of brevity.

Constant.13−.81−1.07−.73−.90
Industry dummy coefficients
Firm size.42**.34**.33**.33**.31**
R&D intensity.04.03.03.01.01
Product differentiation.04.00.00−.02−.03
Cash dividends paid-out.13**.10*.10*.11*.10*
Market growth.14**.14**.15**.14**.14**
Demand instability.10.15.15.11.10
Capital intensity.24**.08.08.08.07
Industry concentration.19.32.34.33.35
CEO Age.03*.02.02.02.02
CEO Tenure−.04*−.02−.01−.02−.01
Percent independent directors (ID) −.02−.02*−.02−.02*
Governance index (GIM) −.03−.04−.03−.03
Block institutional ownership (ALL5) −.05*−.05*−.05*−.06*
CEO ownership (CEO_OWN) −.05−.06*−.04−.05*
Cash & accounts receivables (CASH) −.21***−.21***−.23***−.24***
Debt equity ratio (DE) .06*.08*.04*.05
MBR Attainment discrepancy (MBRD) −.02−.04*  
ROA Attainment discrepancy (ROAD)   −.04**−.05**
ID × CASH  .04 .04
GIM × CASH  −.11** −.11**
ALL5 × CASH  .00 .00
CEO_OWN × CASH  −.06 −.05
ID × DE  .02 .01
GIM × DE  .03 .01
ALL5 × DE  .03 .03
CEO_OWN × DE  −.02 −.01
ID × MBRD  −.03*  
GIM × MBRD  −.05*  
ALL5 × MBRD  .00  
CEO_OWN × MBRD  .01  
ID × ROAD    .02
GIM × ROAD    .00
ALL5 × ROAD    .02
CEO_OWN × ROAD    −.06*
R2.38***.42***.44***.43***.46***
Change in R2 .03***.03***.04***.03***
N17731522152215221522

In Table 2, Model 1 shows the results for the regression with control variables alone, whereas Model 2 includes the results for main effects with attainment discrepancy in MBR, and Model 4 shows that with attainment discrepancy in ROA. Models 3 and 5 represent the complete regression models, including the interaction terms with attainment discrepancy in MBR and ROA, respectively.

The table shows that firms that are larger in size, have high capital intensity, emphasize greater product differentiation, operate in high growth markets, or more concentrated industries are more likely to invest in positive CSR. On the other hand, firms that operate in conditions of greater demand instability or whose CEOs have longer tenure are less likely to invest in positive CSR. These are interesting findings that future research may like to explore in greater detail.

More interestingly, however, Tables 2 and 3 provide strong support for our predictions. Hypothesis 1a predicted that effective monitoring would reduce positive CSR. Table 2 reveals that high independent director representation (β = −.132 in Model 3 and β = −.103 in Model 5, p < .05 in both models), concentrated institutional ownership (β = −.056 in Model 3 and β = −.058 in Model 5, p < .05 in both models), and managerial ownership (β = −.118 in Model 3 and β = −.137 in model 5, p < .01 in both models) all have a significant negative impact on the levels of positive CSR. However, GIM (shareholder rights) did not reveal statistically significant result. Overall, these results confirmed agency theory predictions that expect tighter monitoring to lead to declines in positive CSR, supporting Hypothesis 1a.

Similarly, Hypothesis 1b predicted that effective monitoring would reduce negative CSR.

Table 3 reveals that high independent director representation (β = −.024 in Model 8 and β = −.023 in Model 10, p < .05 in both models), concentrated institutional ownership (β = −.052 in Model 3 and β = −.056 in Model 5, p < .05 in both models), and managerial ownership (β = −.059 in Model 3 and β = −.053 in Model 5, p < .05 in both models) all have a significant negative impact on the levels of negative CSR. However, GIM (shareholder rights) once again did not reveal a statistically significant result. These results confirmed that tighter monitoring also leads to declines in negative CSR, supporting Hypothesis 1b.

In Hypothesis 2a, we predict that positive attainment discrepancy is associated with higher positive CSR. Table 2 shows that, according to both market-based and accounting measures, attainment discrepancy is positively associated with positive CSR (β = .066, p < .05 in Model 3 and β = .026, p < 10 in Model 5).

Similarly, in Hypothesis 2b, we predict that positive attainment discrepancy is associated with lower negative CSR. Table 3 shows that attainment discrepancy is negatively associated with negative CSR (β = −.037, p < .05 in Model 8 and β = −.045, p < .01 in Model 10). These results indicate strong support for Hypothesis 2b.

In Hypothesis 3a, we predict that high-discretion slack is associated with higher positive CSR. Table 2 shows that available slack – cash and accounts receivables – is significantly related to positive CSR (β = .341 in Model 3 (MBRD) and .333 in Model 5 (ROAD), p < .001 in both). The potential slack – debt-to-equity ratio –, which represents the opposite of the cash and accounts receivables interpretation, is not significantly related to positive CSR. Overall, we find support for Hypothesis 3a as well.

In Hypothesis 3b, we predict that high-discretion slack is associated with lower negative CSR. Table 3 shows that available slack – cash and accounts receivables – is significantly related to negative CSR (β = −.214 in Model 8 (MBRD) and −.237 in Model 10 (ROAD), p < .001 in both). Similarly, the potential slack – debt-to-equity ratio –, which represents the opposite of the cash and accounts receivables interpretation, is positively related to negative CSR (β = .076, p < .05 in Model 8 (MBRD) and .046, p < .10 in Model 10 (ROAD). Thus, we find strong support for Hypothesis 3b as well.

Hypotheses 4a and 4b integrate corporate governance and behavioral theories, and test for the interaction effects. We provide a wide array of interaction plots to illustrate these interactions (Figures 3, 4, and 5), and this is where the really interesting findings emerge. These plots clearly depict that the impact of governance on positive CSR is minimal when firms perform well – as indicated by the relatively flatter slopes of lines for high performance under weak and strong governance. Note also the significance of the interaction is the t-test for difference in slopes (Aiken & West, 1991). In plotting these interactions, we plotted only those interactions that had statistically significant differences in slopes. There is a sharp decline in positive CSR for firms who have not performed up to satisfaction and operate under strong governance. That is, the following combinations: 1) low cash and higher proportions of independent directors (Figure 3); 2) low cash and higher proportion of concentrated institutional owners (Figure 4); and 3) low attainment discrepancy and high CEO ownership (Figure 5), all lead to reduced positive CSR. This indicates that the effects of strong governance vary over the range of behavioral theory factors, and are more important under conditions of weaker performance or lower slack. Also, there is a sharp decline in negative CSR for firms with high slack and positive attainment discrepancy and under strong governance. That is the following combinations: (1) low debt-equity ratios and high institutional ownership concentration (Figure 6); (2) high positive attainment discrepancy and high board independence (Figure 7); and (3) high positive attainment discrepancy and high CEO ownership (Figure 8) are associated with reduced negative CSR.

image

Figure 3. Interaction Effect of Independent Director Representation and CASH on Positive CSR Ratings

Download figure to PowerPoint

image

Figure 4. Interaction Effect of Institutional Ownership and CASH on Positive CSR Ratings

Download figure to PowerPoint

image

Figure 5. Interaction Effect of Managerial Ownership and MBR Attainment discrepancy (MBRD) on Positive CSR Ratings

Download figure to PowerPoint

image

Figure 6. Interaction Effect of Institutional Ownership and Debt-Equity Ratio on Negative CSR Ratings

Download figure to PowerPoint

image

Figure 7. Interaction Effect of Independent Director Representation and MBR Attainment discrepancy (MBRD) on Negative CSR Ratings

Download figure to PowerPoint

image

Figure 8. Interaction Effect of Institutional Ownership and ROA Attainment discrepancy (ROAD) on Negative CSR Ratings

Download figure to PowerPoint

The overall trend based on the interactive effects indicates that strong governance curtails positive CSR in firms that are not doing well, and reduces negative CSR in high performing firms with slack. The graphs also reveal that behavioral theory factors have important implications for both positive and negative CSR.

LIMITATIONS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. THEORY AND HYPOTHESES DEVELOPMENT
  5. METHODS
  6. RESULTS
  7. LIMITATIONS
  8. DISCUSSION AND CONCLUSION
  9. ACKNOWLEDGEMENTS
  10. REFERENCES
  11. Appendices

Though our research design makes a number of methodological improvements, it is not free of limitations. First, our study relies on KLD ratings for measuring positive and negative CSR. However, some recent research has highlighted issues with using KLD ratings. For example, Chatterji, Levine, and Toffel (2009) found that in the case of environmental ratings, KLD might not be optimally using publicly available data. This is especially relevant in view of the fact that the economic leverage that a firm can extract from its social performance depends on its past performance and credibility among key stakeholders. Barnett (2007) calls this leverage stakeholder influence capacity (SIC). According to him, “just as a firm's ability to notice, assimilate, and exploit new knowledge depends on its prior knowledge, its ability to notice and profitably exploit opportunities to improve stakeholder relations through CSR depends on its prior stakeholder relationships” (Barnett, 2007: 803). If firms are rational actors, as the instrumental positivist perspective suggests, they likely are aware of both their business strategies and stakeholder influence capacities. Thus, beyond financial performance, they may take into account their stakeholder influence capacities before making CSR investments. We lack access to any such information for this study, but additional research might consider the impact of SIC on CSR. While despite these imperfections, KLD continues to be the best available option for researchers interested in studying this important facet of corporate life; future research should consider looking for alternative data sources.

Second, our sample consists of only firms from the United States. For generalizing our results beyond this single national context, we recommend that researchers explore the concept of social performance from an institutional theory perspective and ask whether mimetic and normative pressures encourage firms to follow others in their industry in allocating higher expenditures to CSR. This line of inquiry could reveal whether firms, when evaluating CSR expenditures, are engaged in a rational pursuit or mere imitative behavior.

Lastly, our research design does not reveal why owners and directors do what they do. Further research should examine the CSR motives of institutions designed for effective corporate governance. For the sake of parsimony, we focus only on the macro effects of institutional ownership, but for different owners, it may be more useful to study the motives of each type, such as pension funds, mutual funds, managerial owners, family owners, and so forth. Similarly, inside directors may be segregated into subgroups on the basis of demographics or their attitudinal identifications (e.g., Hillman, Nicholson, & Shropshire, 2008).

Despite these limitations, our study of the relationship among organizational slack, attainment discrepancy, and corporate governance contributes to an improved understanding of the factors that lead to CSR. It also provides theoretical and practical guidelines to researchers and managers alike to help them determine their hierarchy of objectives and the strategies necessary to meet the demands of various stakeholders.

DISCUSSION AND CONCLUSION

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. THEORY AND HYPOTHESES DEVELOPMENT
  5. METHODS
  6. RESULTS
  7. LIMITATIONS
  8. DISCUSSION AND CONCLUSION
  9. ACKNOWLEDGEMENTS
  10. REFERENCES
  11. Appendices

The main intent of this study was to understand if the relationship between corporate governance mechanisms and corporate social responsibility dimensions was contingent on the levels of slack and performance attainment discrepancy. In doing so, the study sought to resolve the ambiguity in previous findings by considering positive and negative dimensions of CSR, a comprehensive bundle of governance mechanisms, and two important factors, namely, attainment discrepancy and slack, based in the behavioral theory of the firm. The results indicate that effective governance has a symmetric effect on CSR and that it reduces both positive and negative CSR. Second, our results also suggest that greater slack and positive attainment discrepancy lead to higher positive and lower negative CSR. Finally, we find that the associations between effective governance and positive and negative CSR depend on the level of slack and positive attainment discrepancy. That is, the impact of governance on positive CSR is more pronounced under low slack conditions and the impact of governance on negative CSR is more pronounced under high slack conditions.

Our findings highlight the importance of both performance and organizational contexts in studying the impact of governance on managerial decision-making regarding CSR. Overall, our findings provide strong affirmation of the need to integrate behavioral theory insights into corporate governance theory in predicting CSR investments, as we demonstrate that organization slack, attainment discrepancy, and corporate governance jointly determine the levels of CSR.

Our findings support many of the hypotheses and provide more nuanced insights concerning the governance-CSR relationship. First, by considering positive and negative CSR as two distinct activities, we find that effective governance reduces both activities simultaneously in a symmetric fashion. It is possible that previous research that used an all-inclusive CSR measure failed to capture this finding, and that may have led to mixed results. While the benefits and costs of CSR are subject to vigorous debates, we expect that these debates are limited to positive CSR activities. As the costs of negative CSR activities are obvious, and the benefits negligible, the relationship between effective governance and negative CSR should be unequivocally negative and that is what we find. This also provides credence to our findings about positive CSR; irrespective of the benefits/costs tradeoffs, effective governance suppresses positive CSR.

The benefits of dimensionalizing CSR are also obvious when considering the impact of behavioral theory factors. When things look good (i.e., high slack and positive attainment discrepancy), there is more positive CSR and less negative CSR, as one would expect. The significant finding in this instance is that lower levels of slack may actually be associated with increased negative CSR. These findings are consistent with the theoretical predictions of BTOF.

Finally, our findings about the interactive effects highlight the importance of considering the performance and organizational context in understanding the governance CSR relationship. Just as dimensionalizing CSR provides us with unique insights that have not been considered hitherto, contextualizing the relationship also adds to our understanding about the potential mixed results encountered in past research. Our findings suggest that when firms are not doing well, at least by shareholders' expectations, the impact of governance structures is felt more strongly. In such circumstances, shareholders do not seem to trust managers' judgment or give them much discretion for investing in CSR. However, when the firms appear to be doing well, managers seem to possess a significant discretion in determining CSR activities. In fact, independent directors are even associated with higher positive CSR levels. Thus, meeting targets and shareholders' expectations seems to provide managers with more degrees of freedom in terms of resource allocations to corporate social programs. Thus, our findings about attainment discrepancy are not only unique but are also important as we begin to broaden the scope of the agency theory perspective for understanding CSR.

Our results have important implications for understanding when governance matters both for upsides and downsides. While a lot of attention has been paid to the value creation aspect of governance or reduction of agency costs aspects of governance, limited research has actually considered how good governance can prevent managers from making poor decisions. Our findings that good governance reduces negative CSR are unique in that respect and focused on an ignored aspect of governance – preventing bad things from happening.

Our study also has other practical significance for CSR researchers, advocates and managers. From the perspective of CSR advocates, understanding the constraints and concerns of managers is very important. Only by addressing these concerns can firms create enabling conditions for higher CSR. For example, if managers believe they cannot undertake CSR projects because they fear a stock price slide, advocates of CSR could focus on institutional owners instead and convince them that it is in their own best interest to be more proactive. By targeting these owners, advocates would remove an important CSR constraint. Similarly, canvassing boards of directors and recommending they be more aware and responsible toward meeting the needs of all stakeholders, not just shareholders, could provide another effective strategy. Alternatively, if managers provide signals to shareholders about the long-term benefits of corporate social investments, they might be in a position to promote such investments.

This issue is especially relevant in view of comments by researchers such as Scherer and Palazzo (2007: 1101): “In the era of globalization, when the ability of the nation-state to regulate business activities is diminishing… . multinational corporations today are able to choose among various legal systems, applying economic criteria to their choice of which set of labor, social, and environmental regulations they will operate under.” From this perspective, it is not sufficient to rely on the capacity of the state to regulate firm behavior in the interest of society, such as preventing environmental pollution or other ethically questionable activities not covered or enforced by local laws, nor can firms be trusted to behave completely ethically on their own, because their economic rationale (or competitive pressures) makes them concerned primarily with minimizing costs (or maximizing profits). However, by understanding what prompts or prevents managers from undertaking CSR investments, we take an important step toward getting them (or stopping them, when undesirable) to do so.

ACKNOWLEDGEMENTS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. THEORY AND HYPOTHESES DEVELOPMENT
  5. METHODS
  6. RESULTS
  7. LIMITATIONS
  8. DISCUSSION AND CONCLUSION
  9. ACKNOWLEDGEMENTS
  10. REFERENCES
  11. Appendices

We would like to thank William Judge (the editor), Till Talaulicar (associate editor), and three anonymous referees for their insightful comments and suggestions. This paper also benefited from the comments of the participants in Academy of Management Annual Meeting, 2008. The first author is also thankful to the Whitman School of Management at Syracuse University for research grant for this project.

NOTES
  • 1

    There is no special reason for using 5 per cent. Past research has used this number, as a proxy, to signify that most firms would like, at least moderate improvements, on their past performance. Admittedly, it is a crude measure as industry and macroeconomic conditions and other factors also likely influence the performance aspiration, nonetheless on an aggregate past research indicates this to be a good proxy for performance aspirations (Bromiley, 1991).

  • 2

    To avoid unnecessary loss of data, missing values of some variables (e.g., R&D and SGA expenses) were replaced by those in preceding or succeeding years, if available.

  • 3

    We also examine the variance inflation factors (VIF) to detect multicollinearity and Breusch-Pagan test on OLS regression to detect heterogeneity. All VIF scores are below 2, and the mean VIF score is 1.22, well within the commonly specified rule of thumb of a score of 10 indicating multicollinearity problems (Cohen, 2003). Thus, our results do not appear to suffer from multicollinearity. The Breusch-Pagan test also confirmed the absence of heterogeneity in variance.

  • 4

    All variables were standardized [normalized to (0,1)] before running regression models.

REFERENCES

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. THEORY AND HYPOTHESES DEVELOPMENT
  5. METHODS
  6. RESULTS
  7. LIMITATIONS
  8. DISCUSSION AND CONCLUSION
  9. ACKNOWLEDGEMENTS
  10. REFERENCES
  11. Appendices
  • Agrawal, A. & Knoeber, C. R. 1996. Firm performance and mechanisms to control agency problems between managers and shareholders. The Journal of Financial and Quantitative Analysis, 31: 377397.
  • Ahmed, A. S. & Duellman, S. 2007. Accounting conservatism and board of director characteristics: An empirical analysis. Journal of Accounting and Economics, 43: 411437.
  • Aiken, L. S. & West, S. G. 1991. Testing and interpreting interactions in multiple regression. Newbury Park, CA: Sage.
  • Amato, L. H. & Amato, C. H. 2007. The effects of firm size and industry on corporate giving. Journal of Business Ethics, 72: 229241.
  • Amihud, Y. & Lev, B. 1981. Risk reduction as a managerial motive for conglomerate mergers. The Bell Journal of Economics, 605617.
  • Arora, P. 2008. Organizational slack and attainment discrepancy as antecedents of corporate social performance. The Academy of Management Best Paper Proceedings, pp. 16 (AN 33653640).
  • Baltagi, B. H., Bresson, G., & Pirotte, A. 2003. Fixed effects, random effects or Hausman – Taylor? A pretest estimator. Economics letters, 79: 361369.
  • Barnett, M. L. 2007. Stakeholder influence capacity and the variability of financial returns to corporate social responsibility. The Academy of Management Review, 32: 794816.
  • Baysinger, B. & Hoskisson, R. E. 1990. The composition of boards of directors and strategic control: Effects on corporate strategy. The Academy of Management Review, 15: 7287.
  • Baysinger, B. D., Kosnik, R. D. & Turk, T. A. 1991. Effects of board and ownership structure on corporate R&D strategy. Academy of Management Journal, 34: 205214.
  • Bebchuk, L. & Cohen, A. 2005. The costs of entrenched boards. Journal of Financial Economics, 78: 409433.
  • Bourgeois, III, L. J. 1981. On the measurement of organizational slack. The Academy of Management Review, 6: 2939.
  • Bromiley, P. 1991. Testing a causal model of corporate risk taking and performance. The Academy of Management Journal, 34: 3759.
  • Bromiley, P., Miller, K. D., & Rau, D. 2001. Risk in strategic management research. In M. A.Hitt, R. E.Freeman, & J. S.Harrison (Eds.), The Blackwell Handbook of Strategic Management: 259288. Oxford: Blackwell.
  • Burke, L., Logsdon, J. M., Mitchell, W., Reiner, M., & Vogel, D. 1986. Corporate community involvement in the San Francisco Bay area. California Management Review, 28(3): 122141.
  • Bushee, B. 1998. The influence of institutional investors on myopic R&D investment behavior. The Accounting Review, 73: 305333.
  • Chatterji, A., Levine, D., & Toffel, M. 2009. How well do social ratings actually measure corporate social responsibility? Journal of Economics and Management Strategy, 18: 125169.
  • Chen, Y. 2008. Corporate governance and cash holdings: Listed new economy versus old economy firms. Corporate Governance: An International Review, 16: 430442.
  • Chhaochharia, V. & Grinstein, Y. 2007. Corporate governance and firm value: The impact of the 2002 governance rules. The Journal of Finance, 62: 17891825.
  • Chiu, S. & Sharfman, M. P. 2009. Legitimacy, visibility, and the antecedents of corporate social performance: An investigation of instrumental perspective. Journal of Management, Published online on Oct 19, 2009.
  • Coffey, B. & Fryxell, G. E. 1991. Institutional ownership of stock and dimensions of corporate social performance: An empirical examination. Journal of Business Ethics, 10: 437444.
  • Coffey, B. & Wang, J. 1998. Board diversity and managerial control as predictors of corporate social performance. Journal of Business Ethics, 17: 15951603.
  • Cohen, J. 2003. Applied multiple regression/correlation analysis for the behavioral sciences. New York: Lawrence Erlbaum Associates.
  • Cyert, R. M. & March, J. G. 1963. A behavioral theory of the firm. Englewood Cliffs, NY: Prentice Hall.
  • David, P., Bloom, M., & Hillman, A. J. 2007. Investor activism, managerial responsiveness, and corporate social performance. Strategic Management Journal, 28(1).
  • Davis, G. F. 1991. Agents without principles? The spread of the poison pill through the intercorporate network. Administrative Science Quarterly, 36: 583613.
  • Denis, D. J., Denis, D. K., & Sarin, A. 1997. Agency problems, equity ownership, and corporate diversification. Journal of Finance, 135160.
  • Deutsch, Y. 2005. The impact of board composition on firms' critical decisions: A meta-analytic review. Journal of Management, 31: 424.
  • Dharwadkar, R., Goranova, M., Brandes, P., & Khan, R. 2008. Institutional ownership and monitoring effectiveness: It's not just how much you own but what else you own. Organization Science, 19: 419440.
  • Dutta, S., Narasimhan, O. M., & Rajiv, S. 2005. Research notes and commentaries conceptualizing and measuring capabilities: Methodology and empirical application. Strategic Management Journal, 26: 277285.
  • Edwards, F. R. & Hubbard, R. G. 2000. The growth of institutional stock ownership: A promise unfulfilled. Journal of Applied Corporate Finance, 13(3): 92104.
  • Fiegenbaum, A. & Thomas, H. 1988. Attitudes toward risk and the risk-return paradox: Prospect theory explanations. The Academy of Management Journal, 31: 85106.
  • Finkelstein, S. & Boyd, B. K. 1998. How much does the CEO matter? The role of managerial discretion in the setting of CEO compensation. Academy of Management Journal, 41: 179199.
  • Finkelstein, S. & Hambrick, D. C. 1990. Top-management-team tenure and organizational outcomes: The moderating role of managerial discretion. Administrative Science Quarterly, 35: 484503.
  • Fligstein, N. 1991. The structural transformation of American industry: An institutional account of the causes of diversification in the largest firms, 1919–1979. In P. J.DiMaggio & W. W.Powell (Eds.), The New Institutionalism in Organizational Analysis: 311336. Chicago, IL: University of Chicago Press.
  • Geczy, C., Stambaugh, R. F., & Levin, D. 2003. Investing in socially responsible mutual funds. Working paper, Wharton School.
  • Gedajlovic, E. & Shapiro, D. M. 2002. Ownership structure and firm profitability in Japan. Academy of Management Journal, 45: 565575.
  • George, G. 2005. Slack resources and the performance of privately held firms. The Academy of Management Journal, 48: 661676.
  • Godfrey, P. C., Merrill, C. B., & Hansen, J. M. 2009. The relationship between corporate social responsibility and shareholder value: An empirical test of the risk management hypothesis. Strategic Management Journal, 30: 425445.
  • Gompers, P., Ishii, J., & Metrick, A. 2003. Corporate governance and equity prices. The Quarterly Journal of Economics, 118: 107155.
  • Graves, S. B. & Waddock, S. A. 1994. Institutional owners and corporate social performance. The Academy of Management Journal, 37: 10341046.
  • Haleblian, J. & Finkelstein, S. 1993. Top management team size, CEO dominance, and firm performance: The moderating roles of environmental turbulence and discretion. The Academy of Management Journal, 36: 844863.
  • Hambrick, D. C. 2007. Upper echelons theory: An update. The Academy of Management Review, 32: 334343.
  • Hambrick, D. C., Geletkanycz, M. A., & Fredrickson, J. W. 1993. Top executive commitment to the status quo: Some tests of its determinants. Strategic Management Journal, 14: 401418.
  • Hausman, J. A. & Taylor, W. E. 1981. Panel data and unobservable individual effects. Econometrica, 13771398.
  • Hill, C. W. L. & Phan, P. 1991. CEO tenure as a determinant of CEO pay. The Academy of Management Journal, 34: 707717.
  • Hill, C. W. L. & Snell, S. A. 1988. External control, corporate strategy, and firm performance in research-intensive industries. Strategic Management Journal, 9: 577590.
  • Hillman, A. J. & Keim, G. D. 2001. Shareholder value, stakeholder management, and social issues: What's the bottom line? Strategic Management Journal, 22: 125139.
  • Hillman, A. J., Nicholson, G., & Shropshire, C. 2008. Directors' multiple identities, identification, and board monitoring and resource provision. Organization Science, 19: 441458.
  • Hoskisson, R. E., Hitt, M. A., Johnson, R. A., & Grossman, W. 2002. Conflicting voices: The effects of institutional ownership heterogeneity and internal governance on corporate innovation strategies. The Academy of Management Journal, 45: 697716.
  • Jensen, M. C. 1986. Agency cost of free cash flow, corporate finance, and takeovers. The American Economic Review, 76: 323329.
  • Johnson, R. A. & Greening, D. W. 1999. The effects of corporate governance and institutional ownership types on corporate social performance. The Academy of Management Journal, 42: 564576.
  • Kacperczyk, A. 2009. With greater power comes greater responsibility? Takeover protection and corporate attention to stakeholders. Strategic Management Journal, 30: 261285.
  • Kang, D. L. & Sorensen, A. B. 1999. Ownership organization and firm performance. Annual Review of Sociology, 25: 121144.
  • Kassinis, G. & Vafeas, N. 2002. Corporate boards and outside stakeholders as determinants of environmental litigation. Strategic Management Journal, 23: 399415.
  • Kesner, I. F. & Johnson, R. B. 1990. An investigation of the relationship between board composition and stockholder suits. Strategic Management Journal, 11: 327336.
  • Laidroo, L. 2009. Association between ownership structure and public announcements' disclosures. Corporate Governance: An International Review, 17: 1334.
  • Lant, T. K. 1992. Aspiration level adaptation: An empirical exploration. Management Science, 38: 623644.
  • Lorsch, J. & MacIver, E. 1989. Pawns or potentates: The reality of America's corporate boards. Cambridge, MA: Harvard Business School Press.
  • McGuire, J. B., Sundgren, A., & Schneeweis, T. 1988. Corporate social responsibility and firm financial performance. The Academy of Management Journal, 31: 854872.
  • Mackey, A., Mackey, T. B., & Barney, J. B. 2007. Corporate social responsibility and firm performance: Investor relations and corporate strategies. The Academy of Management Review, 32: 817835.
  • McWilliams, A. & Siegel, D. 2000. Corporate social responsibility and financial performance: Correlation or misspecification? Strategic Management Journal, 21: 603609.
  • Maddala, G. S. 2002. Introduction to Econometrics. New York, NY: Prentice-Hall.
  • Mattingly, J. E. & Berman, S. L. 2006. Measurement of corporate social action: Discovering taxonomy in the Kinder Lydenburg Domini ratings data. Business & Society, 45: 20.
  • Morck, R., Shleifer, A., & Vishny, R. W. 1988. Management ownership and market valuation: An empirical analysis. Journal of Financial Economics, 20: 293315.
  • Navarro, P. 1988. Why do corporations give to charity? The Journal of Business, 61: 6593.
  • Neubaum, D. O. & Zahra, S. A. 2006. Institutional ownership and corporate social performance: The moderating effects of investment horizon, activism, and coordination. Journal of Management, 32: 108.
  • Nohria, N. & Gulati, R. 1996. Is slack good or bad for innovation? The Academy of Management Journal, 39: 12451264.
  • Palia, D. 2001. The endogeneity of managerial compensation in firm valuation: A solution. Review of Financial Studies, 14: 735764.
  • Pfeffer, J. & Salancik, G. R. 1978. The external control of organizations: A resource dependence perspective. Palo Alto, CA: Stanford University Press.
  • Pound, J. 1992. Beyond takeovers: Politics comes to corporate control. Harvard Business Review, 70(2): 8393.
  • Rediker, K. J. & Seth, A. 1995. Boards of directors and substitution effects of alternative governance mechanisms. Strategic Management Journal, 16(2): 8599.
  • Richard, O. C., Murthi, B. P. S., & Ismail, K. 2007. The impact of racial diversity on intermediate and long-term performance: The moderating role of environmental context. Strategic Management Journal, 28: 12131233.
  • Scherer, A. G. & Palazzo, G. 2007. Toward a political conception of corporate responsibility: Business and society seen from a Habermasian perspective. The Academy of Management Review, 32: 10961120.
  • Seifert, B., Morris, S. A., & Bartkus, B. R. 2004. Having, giving, and getting: Slack resources, corporate philanthropy, and firm financial performance. Business & Society, 43(2): 135.
  • Social Investment Forum 2010. Report on Socially Responsible Investing Trends in the United States. http://www.socialinvest.org/resources/pubs/index.cfm .
  • Spicer, B. H. 1978. Investors, corporate social performance, and information disclosure: An empirical study. The Accounting Review, 53: 94111.
  • Strike, V. M., Gao, J., & Bansal, P. 2006. Being good while being bad: Social responsibility and the international diversification of US firms. Journal of International Business Studies, 37: 850862.
  • Waddock, S. & Graves, S. 1997. The corporate social performance – Financial performance link. Strategic Management Journal, 18: 303319.
  • Wahal, S. 1996. Pension fund activism and firm performance. The Journal of Financial and Quantitative Analysis, 31: 123.
  • Wang, J. & Coffey, B. S. 1992. Board composition and corporate philanthropy. Journal of Business Ethics, 11: 771778.
  • Zahra, S. A. 1996. Governance, ownership, and corporate entrepreneurship: The moderating impact of industry technological opportunities. Academy of Management Journal, 39: 17131735.
  • Zahra, S., Oviatt, B., & Minyard, K. 1993. Effects of corporate ownership and board structure on corporate social responsibility and financial performance. Academy of Management Best Paper Proceedings, 336340 (AN 10317088).

Punit Arora is a PhD candidate in strategic management/business economics at the Whitman School of Management at Syracuse University. He is interested in corporate governance, business sustainability, and entrepreneurship.

Ravi Dharwadkar is a Professor of Management at Whitman School of Management at Syracuse University. He obtained a Ph.D., in management from the University of Cincinnati. His research on corporate governance and organization studies has appeared in the Academy of Management Review, Academy of Management Journal, Strategic Management Journal, Organization Science, and Academy of Management Perspective.

Appendices

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. THEORY AND HYPOTHESES DEVELOPMENT
  5. METHODS
  6. RESULTS
  7. LIMITATIONS
  8. DISCUSSION AND CONCLUSION
  9. ACKNOWLEDGEMENTS
  10. REFERENCES
  11. Appendices

APPENDIX 1

Overview of Behavior and Agency Theory Literature on Corporate Social Performance (CSP)
Study domain, authorsPrimary findings
Corporate governance: 
 Coffey and Fryxell (1991)Institutional ownership is negatively related with CSP.
 Coffey and Wang (1998); Wang and Coffey (1992)Outside directors negatively related with CSP.
 Graves and Waddock (1994)No relationship between institutional ownership and CSP
 Johnson and Greening (1999)Higher proportion of outside directors is linked with higher CSP because they are more concerned with external legitimacy.
 Kassinis and Vafeas (2002)Outside directors positively related with CSP.
 Neubaum and Zahra (2006)Positive relationship between institutional ownership and CSP.
Attainment discrepancy: 
Behavioral theory concept labeled as attainment discrepancy by Lant (1992) has not been applied in CSP literature.
Slack: 
 McGuire, Sundgren, and Schneeweis (1988) Waddock and Graves (1997) Amato and Amato (2007)Slack as measured by past financial performance is positively associated with the social performance.
 Navarro (1988)Slack as measured by Equity to debt ratio is positively associated with the social performance.
 Seifert et al. (2004)Slack as measured by cash flow is positively associated with the social performance.

APPENDIX 2

Sample by Industry
Firms in Different industriesSIC codeNumber of firms
  1. Industry Classification Adapted from Waddock and Graves (1997).

Mining and construction100–199922
Food, textiles and apparel2000–239027
Forest products, paper & publishing2391–278036
Chemicals & pharmaceuticals2781–289039
Refining, rubber & plastic2891–319915
Containers, steel & heavy manufacturing3200–356930
Computers, autos & aerospace3570–3990102
Transportation3991–473110
Telephone & utilities4732–499160
Wholesale & retail4992–599053
Bank & financial services5991–670060
Hotel & entertainment6701–805143
Hospital and others8052–999921
Total 518