مسئولیت اجتماعی شرکت ها در صنعت بانکداری: انگیزه و عملکرد مالی

The current study investigates the association between corporate social responsibility (CSR) and financial performance (FP), and discusses the driving motives of banks to engage in CSR. Three motives, namely, strategic choices, altruism, and greenwashing, suggest that the relationship between CSR and FP is positive, non-negative, and non-existent, respectively. We obtained our sample, which covered 2003–2009, from the Ethical Investment Research Service (EIRIS) databank and Bankscope database. The data consists of 162 banks in 22 countries. We then classified the banks into four types based on their degree of engagement in CSR. This study proposes the use of an extended version of the Heckman two-step regression, in which the first step adopts a multinomial logit model, and the second step estimates the performance equation with the inverse Mills ratio generated by the first step. The empirical results show that CSR positively associates with FP in terms of return on assets, return on equity, net interest income, and non-interest income. In contrast, CSR negatively associates with non-performing loans. Hence, strategic choice is the primary motive of banks to engage in CSR.

The question of whether adopting corporate social responsibility (CSR) can improve a corporation’s financial performance (FP) is an old yet continually and heatedly debated issue.1 Thus, a better understanding of the link between CSR and FP would be valuable to corporate managers, stockholders, and stakeholders. For example, what resources should managers direct to socially responsible activities? How should stockholders react to resource allocations for a social purpose? How can public policy best promote socially responsible behavior (Simpson and Kohers, 2002)? A corporation is generally encouraged to adopt CSR because of its perceived benefits to both macro and micro-performances. Macro-performance includes environmental improvement and reduction in social inequality. Micro-performance includes reputation enhancement, potential to charge a premium price for products as well as the enhanced ability to recruit and to retain high-quality workers.2 The most attractive lure is that firms adopting CSR can gain financial benefits that are greater than the ensuing costs, thereby improving FP in the long run. Accordingly, adopting CSR can be beneficial to both corporate shareholders and stakeholders, which creates a potential win–win situation. Numerous empirical studies attempt to justify this positive relationship between stock return3 and accounting performance.4 However, previous results often offer contradictory conclusions. For example, the review of Margolis and Walsh (2003) of 127 empirical studies, dating from 1972 to 2002, reveals the relationship between CSR and FP. Among those reviewed, 54 studies point toward a positive relationship, 20 show mixed results, and 28 reveal insignificant relationships. Only seven studies show a negative relationship. The meta-analysis by Margolis et al. (2007) on 167 studies over the past 35 years shows that the overall effect is positive yet small. Moreover, they suggest that future studies should be redirected to gain a better understanding of the reasons for companies to pursue CSR, the mechanism connecting prior CSR to subsequent FP, and the methods that companies employ to manage the pursuit of both CSR and FP (see also Shen and Chang, 2009, Shen and Chang, 2012-a and Garcia-Castro et al., 2010). The conflicting results may be attributed to the fact that corporations with having different motives in conducting CSR may exhibit different CSR-FP relationships. According to Baron, 2001 and Dam et al., 2009, and Bénabou and Tirole (2010), an engagement in CSR reflects the different motives of a corporation, which include altruism, strategic choices, and greenwashing. The altruism motive indicates that companies conduct CSR activities for their own sake (Baron, 2001), thereby negatively affecting FP. However, the strategic motive improves FP through CSR engagement. Finally, greenwashing attempts to enhance the corporate image without significantly changing the business (Frankental, 2001).5 According to Dam et al. (2009), if no clear cost differences are observed between responsible and irresponsible corporations, then these firms are merely greenwashing, hence no effects will become evident in their earnings. Thus, the conflicting conclusions on the relation between CSR and FP may be attributed to the different motives of corporations. Attributing the differences in the results to the different samples, methods, and periods used is plausible as well. The banking system plays an important role in economic development (Levine, 2005, Shen and Lee, 2005, Beck et al., 1999 and Beck et al., 2010) because its safety and soundness create several external benefits to society. Generally, for economic growth, banks serve as financial intermediaries by facilitating cash flow between lenders and borrowers. According to Schumpeter (1912), well-functioning financial institutions, such as banks, enhance technological innovation by supporting entrepreneurs with the best chances of successfully introducing innovative products and production processes. Therefore, a healthy banking system is the key to sustained prosperity (King and Levine, 1993). Banks also play an important role during a credit crisis. For example, when Basel I was implemented in 1989, many banks were compelled to increase capital, to decrease lending, or both. Decrease in lending, or the credit crunch mentioned in the literature review, became one of the factors that caused the US recession in 1991. Furthermore, by using considerable resources from society, banks are required to provide feedback to the community more often than other industries. For example, bank assets may mainly come from depositors, but not from shareholders. Additionally, when banks are in distress, governments bail them out or take over at the expense of taxpayers. In March 2011, Bank of England Governor Mervyn King accused banks of exploiting gullible or unsuspecting customers for short-term profit. He criticized the culture of short-term profit and bonuses in the banking system, and suggested that traditional manufacturing industries by contrast adhered more to higher “moral” standards when conducting business. According to King, “They care deeply about their workforce, about their customers and, above all, are proud of their products.” Therefore, as banks employ public resources paid for by society, they are highly scrutinized by the media, government, and academe regarding their CSR activities. Scholtens (2009) mentions that in member countries of the Organization for Economic Cooperation and Development (OECD), specialized banks offer savings accounts to the public while promising that the money will be used to finance so-called community investments in the environment. Thus, in most countries, banks are involved in economic activities aimed at sustainable development. Banks are aware of their use of public resources, and this awareness explains why many banks append a CSR section in their annual reports to explain how they give back to society.6 However, direct studies on the CSR-FP link in the banking industry and the motives for engaging in CSR are considerably scant. Simpson and Kohers (2002) present one of the few studies using bank data but they do not pursue the CSR issue. Other studies on the banking sector engaged in CSR activities focus on a similar issue, but not on the association between CSR and FP. For example, Scholtens and Dam (2007) compare CSR engagements between banks that adopt the Equator Principles and those that do not. Their finding indicates that real cost may be associated with the implementation of the said principles. They observe that banks adopting the Equator Principles have significantly higher CSR policies and lower returns on assets (ROA). Chih et al. (2010) examine the determinants of financial firms that adopt CSR. They observe that financial firms with larger assets and worse ROA adopt CSR. Scholtens (2009) provides a framework for assessing CSR with international banks, but does not examine the concerned relationship. Cuesta-González et al. (2006) focus on social performance but not on the financial performance of top Spanish banks.7 These studies do not directly address the relationship between CSR and FP, thus drawing a conclusion for the banking sector is difficult. Hence, in contrast to the numerous studies using non-bank data, empirical evidence for the CSR-FP link in the banking industry is rare. Bank theories are typically related to delegated monitoring for alleviating the moral hazard of borrowers and reducing risk sharing. In addition, the theories are related to optimal contracts and credit rationing for lender–borrower relationship8 as well as to bank urn and liquidity shock (Freixas and Rochet, 1998). Few bank theories are related to CSR. One way to incorporate CSR with modern banking theories is to relate it to the role of bank reputation. In the theoretical model of Chemmanur and Fulghieri (1994), high-reputation banks are shown to have incentives to perform more rigorous pre-loan evaluations of the borrowers’ unobservable prospects compared with low-reputation banks. Bushman and Wittenberg-Moerman (2012) document that high-reputation banks are associated with the stronger profitability and credit quality of borrowers in the 3 years following the loan initiation as well as with the quality of the reported accounting numbers of borrowers. Some studies state that bank reputation certifies borrower quality because borrowers have positive abnormal stock returns at the time of loan announcements from reputational banks (Billet et al., 1995 and Ross, 2010). Hence, according to the above reputation assertion, banks conducting CSR would select and attract more creditworthy borrowers, which contribute higher profit and better asset quality to the financial institutions. The current study aims to investigate the relationship between CSR and FP in the banking sector, as well as the driving motive of banks to conduct CSR. Our study differs from other studies and contributes to the literature in four aspects. First, we set up a theoretical model of banking profit function considering three motives pertaining to the ways that CSR affects FP. In this model, CSR is one of the determinants, thereby helping us to understand the explicit channels and motives with regard to CSR affecting FP. Similar to Dam et al. (2009), we establish the profit functions to examine the three motives of altruism, strategic choices, and greenwashing. While their study focuses on the non-banking sector in the US, we examine the relationship between CSR and FP in the banking sector by using global banking data. Employing these relationships, we investigate the motives that drive CSR in the banking sector. The three motives of strategic choices, altruism, and greenwashing suggest that the relationship between CSR and FP is positive, non-negative, and non-existent, respectively. Our model is a compromise between those companies that develop CSR profit models in the non-banking sector and those that develop non-CSR profit function in the banking sector. Second, the current study adopts an extended version of the two-step method of Heckman (1978) to minimize the selection bias. This approach is different from those of previous studies, which use the traditional Heckman two-stage method with a dichotomous CSR variable. Although a number of studies consider multi-degree CSR, they do not eliminate the self-selection bias as we do in the current study. Studying the relationship between CSR and FP is often subject to the criticism of causal inference.9 This criticism arises because the bank’s choice to conduct CSR activities may not be random, but rather a deliberate decision made by managers to self-select their preferred options, thereby creating a selection bias problem. 10 Thus, we consider the endogeneity problem by combining the two-step method of Heckman (1978) and multi-degree CSR. Specifically, we classify banks into four types based on their degree of engaging in CSR. The degree ranking is from 1 to 4. The ranking was derived by using a survey conducted by the Ethical Investment Research Service (EIRIS). In the current study, the four-scale measure used is interchangeably referred to as the CSR index or CSR regime. Some studies dichotomize banks into those that adopt CSR and those that do not, but this dichotomy ignores the reality that the majority of banks fall between the two extremes. 11 These studies use the Financial Times Stock Exchange (FTSE) indices (e.g., FTSE4GOOD) or the Dow Jones Sustainability Indexes (DJSI) to classify corporations as CSR or non-CSR corporations. Scholtens (2009) indicates that these indices use a “best-in-class” approach, which assures that top corporations are selected in the indices. Accordingly, firms included in FTSE4GOOD and DJSI are those that strongly engage in CSR activities, otherwise, they are excluded. Banks that conduct only a certain degree of CSR are often classified as non-CSR banks. Our four-scale CSR index enables us to investigate the performance of banks engaged even in a medium degree of CSR activities. The findings may shed light on the reasons for the previous mixed results. Third, we assess the total effect of CSR, which is referred to as the treatment effect. In the Heckman model, our study contributes to the literature by demonstrating that the coefficient of a CSR dummy variable detects only the direct effect of CSR. However, because the residuals of decision and performance equations are correlated, there is an indirect effect through the correlation. Past studies typically focus on only the direct effect. The direct effect is the influence of adopting CSR, referred to as the average treatment effect (ATE). Meanwhile, the indirect effect comprises the correlation between unexplained parts in two equations that appear in the first and second steps (the two equations are referred to as the decision and the performance, respectively). Total effect is the treatment effect on the treated group (ATT) (Clatworthy et al., 2009). This study suggests further investigation of the indirect effect to consider the full-fledged effect of CSR, using the calculation of the treatment effect by Greene (2003, p. 788) and by Hamilton and Nickerson (2003) after conducting Heckman’s two-step method. Finally, we used a large sample of banking data (from 2003 to 2009) from 22 countries. Our study extends the work of Simpson and Kohers (2002), which focuses on US bank data as well as on the binary CSR strategy to investigate global countries and multiple CSR strategy. The large sample avoids the small sample bias, which may vary on a particular area, sample period, and frequency. The results of the current study therefore provide more reliable data compared with those of others. One caveat should be noted. Chatterji and Levine (2008), and Chatterji et al. (2010) argue that the organizations rating the social performance of enterprises cannot truly discern which firms are socially responsible, resulting in metrics that are often invalid and can be misleading to stakeholders. Thus, our results depend on the correctness of the CSR rating. If studies that find a positive correlation between CSR metrics and financial performance overstate the relationship between the actual CSR and financial performance, the customers or other stakeholders are misled by the erroneous CSR metrics (e.g., by successful “greenwashing” campaigns). The remainder of this paper is organized as follows. Section 2 discusses the motivations of CSR and the model of CSR in banks profit functions. Section 3 illustrates the construction of our CSR index. Section 4 discusses the econometric model used in this study and data sources and basic statistics. Section 5 presents the empirical results for the extended Heckman two-stage model and robustness testing. Finally, Section 6 summarizes the results.

This study investigates the association between CSR and FP in global banking sector and discusses the motives driving banks to engage in CSR. Our theoretical model for banking industry suggests that CSR could facilitate brand differentiation for strategic and altruistic banks, but not for greenwashing banks. Strategic banks utilize this advantage to increase NII, NonII, and net profits (ROA and ROE), as well as decrease NPL. Although altruistic banks are not interested in utilizing these advantages, their products are different from the perspectives of clients, thus resulting in an increase in NII and NonII. However, engaging in the altruistic CSR also increases cost; hence, the net effect on profit (ROA and ROE) is uncertain. Greenwashing banks provide only lip service, which does not affect their income and cost. Depending on the driving motives of strategic, altruistic, and greenwashing, CSR could generate a positive, uncertain, or no influence on net income. To explore this issue, we extend the Heckman two-step method to minimize the selection bias given the multi-level CSR data. Past studies using the similar type of data often ignore this bias. Our empirical results find that CSR positively affects NII, NonII, ROA, and ROE, and negatively affects NPL. Our evidence is consistent with existing theories of financial intermediary’s reputation, because we show that bank with higher CSR would have higher financial earnings and asset quality. Hence, in terms of profit measures, banks conducting more CSR activities outperform those banks that do not engage in CSR. The strategic motive seems to be predominant in the banking sector that engages in CSR. By using the aggregate CSR index, we can conclude that strategic motives mainly drive the CSR activities of banks. Our results are robust to the subprime crisis consideration, different specifications for the CSR dummies and more CSR classes. Thus, CSR and FP are positively linked. We suggest that governments should provide less support to banks engaging in CSR activities if the activities are based on strategic motives. Governments could support banks engaging in CSR activities with altruistic motives by providing loans to firms engaged in environmental protection. Studies merely investigating the relationship between CSR and FP tend to ignore the motive. Future research work on the following issues may be interesting. First, although our study is based on the overall aggregate index, examining the relation between CSR and FP using the sub-components of these indices, for example, environmental, social, and governance, is helpful in understanding the issue further. Second, as discussed in the Robust Section, no nonlinear relationship exists between CSR and FP in our case. Future studies can re-visit this nonlinear issue when more CSR choices are available. Finally, the follow-up effects of the private and social costs of the subprime financial crisis can also be considered in future studies.