Socially Responsible Firms

Allen Ferrell is Greenfield Professor of Securities Law, Harvard Law School. The following post is based on the paper co-authored by Professor Ferrell, Hao Liang and Professor Luc Renneboog.

The desirability of corporations engaging in “socially responsible” behavior has long been hotly debated among economists, lawyers, and business experts. Two general views on corporate social responsibility (CSR) prevail in the literature. The CSR “value-enhancing view” argues that socially responsible firms, such as firms that promote efforts to help protect the environment, promote social equality, improve community relationships, can and often do adhere to value-maximizing corporate governance practices. Indeed, well-governed firms are more likely to be socially responsible. In short, CSR can be consistent with shareholder wealth maximization as well as achieving broader societal goals. The opposite view on CSR begins with Milton Friedman’s (1970) well-known claim that “the only social responsibility of corporations is to make money”. Extending this view, several researchers argue that CSR is often simply a manifestation of managerial agency problems inside the firm (Benabou and Tirole, 2010; Cheng, Hong, and Shue, 2013; Masulis and Reza, 2014) and hence problematic (“agency view”). That is to say, socially responsible firms tend to suffer from agency problems which enable managers to engage in CSR that benefits themselves at the expense of shareholders (Krueger, 2013). Furthermore, managers engaged in time-consuming CSR activities may lose focus on their core managerial responsibilities (Jensen, 2001). Overall, according to the agency view, CSR is generally not in the interests of shareholders.

The empirical literature testing these two views is mixed and thus has left the issues raised in the debate largely unresolved. Much of the literature is largely focused only on the ex post effects of CSR. That is, the principal research focus is measuring shareholder reactions’ to CSR as captured by abnormal stock returns (e.g., Dimson, Karakas, and Li, 2013), the cost of capital (e.g., El Ghoul et al., 2011), and ownership changes (e.g., Cheng et al., 2013), or on the financial consequences of CSR spending (e.g., Lee and Faff, 2009). However, both the value-enhancing and agency views are concerned to a significant extent with managerial incentives, which are ex ante in nature.

By means of a rich and partly proprietary CSR dataset with global coverage across a large number of countries and covering thousands of the largest global companies, we test these two views by examining whether traditional corporate finance proxies for firm agency problems, such as capital spending cash flows, dividend payouts and leverage, are associated with increased CSR. We also examine managerial incentive-performance perspective in the spirit of Jensen and Murphy (1990), and hence investigate hypotheses concerning the relationship between CSR and managerial pay-for-performance. In the corporate finance literature, executive compensation helps align the interests of managers and of shareholders, and higher pay-performance sensitivity leads to less severe agency problems (and thus shareholder value-enhancement). Therefore, weak managerial pay-for-performance can be viewed as a proxy for agency problems at the firm. Accordingly, the CSR value-enhancing view would hypothesize that CSR is associated with stronger pay-for-performance sensitivity whereas the agency view would predict the opposite.

Of course, CSR and agency problems can emerge simultaneously as they are both choices of the firm in some sense. This simultaneity (or endogeneity) creates an obvious empirical challenge for investigating the relationship between CSR and firm agency problems. Several studies resort to policy and market-wide shocks as quasi-experiments to help identify a causal relationship between CSR and agency proxies (e.g., Hong et al., 2012; Cheng et al., 2013; Flammer, 2013), but this approach is hard to apply in a multi-country context. Therefore, we employ exogenous variation in country-level laws as instrumental variables for firm-level agency problem.

Our primary data on CSR are from MSCI’s Intangible Value Assessment (IVA) database and the Vigeo corporate ESG database. Both databases are built by means of different proprietary data sources and employ different rating metrics, which enables us to cross-validate our results. The IVA indices measure a corporation’s environmental and social risks and opportunities, and are compiled using company profiles, ratings, scores, and industry reports, and are available from 1999 to 2011. Its coverage comprises the top 1,500 companies of the MSCI World Index (expanding to the full MSCI World Index over the course of the sample period); the top 25 companies of the MSCI Emerging Markets Index; the top 275 companies by market cap of the FTSE 100 and the FTSE 250 (excluding investment trusts); and the ASX 200. The whole IVA sample (including the RiskMetrics ratings) covers 91,373 firm-time observations from 59 countries. The Vigeo corporate ESG data set focuses more on CSR compliance, as it applies a check-the-box approach to rate how a firm and the country where it operates comply with the conventions, guidelines, and declarations by international organizations such as UN, ILO, and OECD.

We do not find empirical evidence that CSR is associated with ex ante agency concerns, such as abundance of cash and a weak connection between managerial pay and corporate performance. Rather, higher CSR performance is closely related to tighter cash—usually a proxy for better-disciplined managerial practice in the traditional corporate finance literature (Jensen, 1986)—and higher pay-for-performance sensitivity. In addition, firms in countries with better legal protection on shareholder rights receive higher CSR ratings. We also find that the relation between CSR and large shareholders’ ownership exhibits a non-monotonic relationship. Finally, we find that CSR can counterbalance the negative effects of managerial entrenchment, and lead to higher shareholder value as proxied by Tobin’s Q. Our empirical results (based on an instrumental variables-estimation) suggest that good governance is associated with higher CSR, and that a firm’s CSR practice is consistent with shareholder wealth maximization. Therefore, our findings support the positive stance on CSR, which is also found in Dimson et al. (2013), Deng et al. (2013), and Ioannou & Serafeim (2010, 2012).

The full paper is available for download here.

Both comments and trackbacks are currently closed.