Corporate social responsibility versus corporate shareholder responsibility: A family firm perspective

Highlights

• Studies differences in environmental performance between family and non-family firms

• Family firms are more responsible to shareholders in making environmental investments.

• Family firms do at least as well as non-family firms in alleviating environmental concerns.

• When shareholder and societal interests diverge family firms protect shareholder interests over societal interests

• Lack of diversification by controlling families creates strong incentives to act in financial interest of all shareholders.

Abstract

Recent literature suggests that some socially responsible corporate actions benefit shareholders while others do not. We study differences in policy toward corporate social responsibility (CSR) between family and non-family firms, using environmental performance as the proxy for CSR. We show that family firms are more responsible to shareholders than non-family firms in making environmental investments. When shareholder interests and societal interests coincide, i.e., when it comes to alleviating environmental concerns that have potential to harm society and elevate the firm's risk exposure, family firms do at least as well as non-family firms in protecting shareholder interests. However, when shareholder and societal interests diverge, i.e., when it comes to making environmental investments that might benefit society but do not benefit shareholders, family firms protect shareholder interests by undertaking a significantly lower level of such investments than non-family firms. Our findings suggest that lack of diversification by controlling families creates strong incentives for them to act in the financial interest of all shareholders, which more than overcomes any noneconomic benefits families may derive from engaging in social causes that do not benefit non-controlling shareholders.

Introduction

We study the differences in policy toward corporate social responsibility (CSR) between family and non-family firms to shed new light on both corporate posture toward CSR and the tension between type I and type II agency problems in family firms.¹Bowen (1953) states that businessmen should “pursue those policies, to make those decisions, or to follow those lines of action which are desirable in terms of the objectives and values of our society.” On the other hand, in a well-known New York Times Magazine article, Friedman (1970) states that “The social responsibility of a business is to increase its profits.” These two statements reflect a widely debated topic among both academics and practitioners for the past several decades, i.e. is CSR a legitimate part of a firm's business or none of its business?

Several studies, including Hamilton et al. (1993), Bauer et al. (2005), Schröder (2007), Benabou and Tirole (2010), Borghesi et al. (2014), Masulis and Reza (2015), Cheng et al. (2016), Adhikari (2016) and Ghoul et al. (2016) suggest that CSR is orthogonal to corporate shareholder responsibility. However, numerous other studies, such as Guenster et al. (2011), Derwall et al. (2005), Luo and Bhattacharya (2006), Kempf and Osthoff (2007), Sharfman and Fernando (2008), Deng et al. (2013), Servaes and Tamayo (2013), Chava (2014), Eccles et al. (2014), Crifo et al. (2015), Flammer (2015), Lins et al. (2017) and Albuquerque et al. (2018) show that CSR is positively related to shareholder value. A third set of studies shows that some CSR actions are beneficial to shareholders while other CSR actions are not (Oikonomou et al., 2012; Kruger, 2015; Fernando et al., 2017).

Friedman (1970) goes on to argue that individuals are free to do what they wish with their private money, including spending it on social causes; hence, an individual proprietor is free to spend money from his business as he wishes. However, if a manager is running a firm on behalf of shareholders, his fiduciary duty is to spend the firm's money on investments that maximize shareholder value. In the case of firms run by managers, a question arises then as to whether CSR activities are value enhancing or not. Fernando et al. (2017) document that some corporate investments in social causes (spending to reduce negative environmental outcomes and thereby reduce the firm's risk exposure) creates value for shareholders. In contrast, they show that spending on increasing positive environmental outcomes (investments that go beyond both legal requirements and any conceivable risk mitigation rationale), while good socially, is not viewed by shareholders as value-enhancing.

While the findings of Fernando et al. (2017) are based on a cross-sectional analysis of a sample of firms, there is currently no evidence on whether individual firms can differentiate between CSR actions that benefit shareholders and CSR actions that create no value for shareholders. Family firms provide an ideal experimental setting to examine this question. If family firms operate more like sole-proprietorships than widely held non-family firms, one would expect to see more spending by family firms on social causes that bring noneconomic benefits to the controlling family but no financial benefits to non-controlling shareholders (type II agency problem). On the other hand, if family firms are more concerned about their financial profits than non-family firms due to large undiversified stakes in the firm that mitigate the type I agency problem, one would expect them to spend less on social activities that do not enhance shareholder value.

Following Fernando et al. (2017), we focus on the KLD environmental ratings to capture CSR investments.² The financial consequences of environmental policies of firms are likely to be considerably larger than other socially relevant corporate policies.³ As a result, corporate environmental performance is the area that is most likely to provide evidence of socially responsible investing (see Fernando et al. (2017) for a more detailed discussion). KLD provides a set of binary indicator variables, which reflects either environmental strengths or environmental concerns. For each firm, KLD provides five sub-indicators for environmental strengths and seven sub-indicators for environmental concerns. The sub-indicators for environmental strengths capture aspects of a firm's environmental policy that is intended to enhance its environmental friendliness (“greenness”) whereas the environmental concerns capture aspects that are related to various environmental risk exposures (“toxicity”). KLD assigns a value of one if a firm meets or exceeds a predetermined threshold for each sub-indicator and zero otherwise. Our analysis focuses on separately aggregating a firm's environmental strengths and concerns to measure its “greenness” and “toxicity,” respectively.

The main finding by Fernando et al. (2017) is that institutional investors, who they use as their proxy for the “smart money,” avoid “toxic” stocks (firms with negative net environmental scores). This finding is consistent with the notion that corporate policies that mitigate risk exposure are value enhancing.⁴ Interestingly, they also find that institutional investors also avoid “green” stocks (firms with positive net environmental scores). This finding is consistent with the view that any corporate environmental policy that requires action beyond what is legally mandated or cannot be justified by a risk rationale does not create value for shareholders.⁵ To summarize, Fernando et al. (2017) find that the interests of shareholders and society coincide when it comes to reducing a firm's environmental concerns but they diverge sharply when it comes to increasing environmental strengths. In other words, shareholders reward firms that reduce their “toxicity” but punish firms that increase their “greenness.”⁶

Kruger (2015) finds results that are consistent with Fernando et al. (2017); he uses the KLD Socrates database, which contains the underlying events used to generate the more common KLD indicators. He finds that investors react strongly negatively to negative events (that would lead to higher environmental concerns) and weakly negatively to positive events (that would lead to higher environmental strengths). This implies that reducing environmental concerns, as classified by KLD, is value enhancing but increasing environmental strengths does not create value for shareholders.

We show a clear negative association between family firms and environmental strengths. This result is economically significant indicating an environmental strength score 21% lower compared to the mean environmental score of the sample. This finding suggests that, on average, family firms are more responsible to shareholders than non-family firms when it comes to incurring expenditure that makes their firms green. Our findings for environmental concerns provide somewhat weaker evidence that family firms are also more responsible to shareholders in alleviating environmental concerns. In the univariate analysis, we find that family firms have significantly lower environmental strengths and environmental concerns. Given that higher levels of both environmental concerns and strengths are value eroding, this finding implies that when looking at the unconditional means, family firms appear to be more consistent with shareholder wealth maximization in terms of both environmental strengths and concerns. However, in the multivariate analysis we do not find a statistically significant difference between family firms and non-family firms regarding environmental concerns. Nonetheless, taken together, our findings show that, on average, the corporate environmental policies of family firms are significantly more consistent with shareholder wealth maximization than non-family firms since family firms invest significantly less in environmental actions that benefit society but do not benefit shareholders while investing at least as much as non-family firms in environmental actions that benefit both society and shareholders.

We also investigate whether the nature of the family's involvement has any impact on the reported relationships between environmental strengths/concerns and family firms. We find that regardless of whether a family firm has a founder CEO or a descendent CEO, it has lower environmental strengths compared to non-family firms. If the firm is a family controlled firm (i.e., where neither the founder nor a descendant is the CEO), the reduction in the environmental strengths compared to non-family firms is even more pronounced. As with the previous analysis, regardless of who the CEO is, there is no difference between family firms and non-family firms when it comes to environmental concerns. Furthermore, instead of the family dummy (which is the primary focus of this paper) we look at how environmental strengths relate to the degree of family control. Consistent with our prior analysis we find a negative relationship between the degree of family control and environmental strengths and no relationship between the degree of family control and environmental concerns.

We undertake additional analysis to strengthen identification and establish robustness of our findings. First, we repeat our analysis on a propensity score matched sample. We match family firms with non-family firms using a propensity score calculated on size, book-to-market and the two-digit SIC code, and using the nearest neighbor matching (with replacement) approach. When we repeat our analysis on this propensity score matched sample, we find results that are consistent with our prior findings; family firms have lower environmental strengths compared to non-family firms but there is no significant difference between the firms regarding environmental concerns. Second, we use state level regulatory stringency of corporate environmental performance to control for the possibility that the environmental regulatory stringency of each state affects the environmental CSR activities of firms and their location decisions. We control for regulatory stringency using a proxy that has previously been used in the literature (King and Lenox, 2002; Kassinis and Vafeas, 2002; Berrone et al., 2010). Controlling for the state-level regulatory stringency and interacting with the family firm variable does not substantively change our results. For environmental strengths, we find that even when the state regulation is high, family firms have significantly fewer strengths than non-family firms. For environmental concerns we find that that family firms have significantly lower concerns when state regulation is high. Therefore, our overall conclusions remain unchanged.

Since family reputation is an important source of noneconomic utility for family firms, we also investigate what family reputation might mean for environmental performance in family firms. As a proxy for how important preserving the reputation of the family could be, we look at the interaction between the family firm dummy and a dummy indicating whether the founder's name or a portion of it is included in the firm name. We find that reputation does not seem to have an impact on the results for environmental strengths. However, we find that this interaction term is positive for environmental concerns, which implies that family firms with the founder's name have a higher amount of concerns. This could either imply that family reputation effects do not truly exist for large public companies or that our firm name proxy is capturing another unknown effect. However, this test also provides some evidence of family firms having less environmental concerns than non-family firms, among the firms that do not have the founder's name as part of the firm name.

We also find that our results persist across different time periods. Specifically, we investigate whether the recent financial crisis had an impact on environmental performance. We find that both during the crisis and prior to the crisis family firms have consistently lower environmental strengths. Interestingly we find that during the crisis, family firms have lower environmental concerns compared to non-family firms.

As an additional robustness test, we look at an alternative database, the Thomson Reuters' ASSET4 database. The ASSET4 database reports many indicator variables related to the environmental policy of a firm. Using the KLD definitions we categorize these variables as strengths and concerns. We repeat our analysis using these variables and find results consistent with our previous analysis. We also find that our results are robust to different methods of computing standard errors and when we control for environmental concerns (strengths) in the strengths (concerns) regressions or exclude firms that have both strengths and concerns greater than zero.

This paper contributes to a recent literature that is divided on the CSR performance of family firms. Ghoul et al. (2016) look at the CSR performance in nine East Asian countries using the Thomson Reuters' ASSET4 database and find that family firms are associated with a lower CSR score in terms of both environmental CSR and social CSR. In contrast, Berrone et al. (2010) show for a sample of U.S. firms that family firms have lower levels of toxic emissions, indicating superior environmental performance, while Dyer and Whetten (2006) show that family firms have fewer social (including environmental) concerns than non-family firms. Additionally, Block and Wagner (2014) show that family firms do better than non-family firms in some CSR dimensions and worse in others. Ghoul et al. (2016) view CSR activities as “…firm actions that go above and beyond the interests of the firm to further the social good” (Ghoul et al., 2016, pg.1). Their perspective contrasts sharply with the insight provided by Fernando et al. (2017) that some CSR activities (specifically those that mitigate the risk exposure of companies) are strongly consistent with shareholder interests, while other CSR activities are antithetical to shareholder interests.

Our findings contrast with the aforementioned studies and contribute to the literature by showing that when it comes to CSR, family firms are more responsible to shareholders than non-family firms. When shareholder interests and societal interests coincide, i.e., when it comes to alleviating environmental concerns that have potential to harm society and elevate the firm's risk exposure, family firms do at least as well as non-family firms in protecting shareholder interests. However, when shareholder and societal interests diverge, i.e., when it comes to making environmental investments that benefit society but do not benefit shareholders, family firms are significantly more on the side of shareholders by undertaking fewer such investments than non-family firms. These findings contrast sharply with studies of corporate governance in family firms that show family firms being less responsible to their non-family shareholders.⁷ When it comes to CSR, our findings suggest that the actions of the family are more consistent with the interests of shareholders (including non-family shareholders) than the managers of non-family firms, which supports the argument that the type I agency problem is alleviated in family firms, thereby causing family firms to align their CSR activities with shareholder wealth maximization.

The rest of the paper is organized as follows. Section 2 provides a brief review of the relevant literature. Section 3 presents an overview of the data and the methodology used in the study. Section 4 contains the empirical analysis and discussion of the results while Section 5 concludes.