قیمت اخلاق و حکومت ذینفعان : عملکرد وجوه متقابل مسئول اجتماعی
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|1595||2008||21 صفحه PDF||سفارش دهید||1 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Journal of Corporate Finance, Volume 14, Issue 3, June 2008, Pages 302–322
Do investors pay a price for investing in socially responsible investments (SRI) funds, or do they obtain superior returns? This paper investigates these under- and overperformance hypotheses for all SRI funds across the world. Consistent with investors paying a price for ethics, SRI funds in the US, the UK, and in many continental European and Asia-Pacific countries underperform their domestic benchmarks by − 2.2% to − 6.5%. However, with the exception of some countries such as France, Japan and Sweden, the risk-adjusted returns of SRI funds are not statistically different from the performance of conventional funds. We also find that the underperformance of SRI funds is not driven by loadings on an ethics style factor. There is mixed evidence of a smart money effect: SRI investors are unable to identify the funds that will outperform in the future, whereas they show some fund-selection ability in identifying funds that will perform poorly. Finally, corporate governance and social screens yield lower risk-adjusted returns.
Although economics textbooks usually state that human behavior is driven by the maximization of self-interest, many people deviate from exclusively selfish behavior (Fehr and Gachter, 2000 and Fehr and Gachter, 2002). For example, recent experimental evidence indicates that altruism or selflessness is a powerful feature of human demeanor (Fehr and Fischbacher, 2003). An individual's utility partially depends on the utility of other members of the community, and ethical and social considerations are important determinants of economic behavior.2 Economic theories of social norms (see Akerlof, 1980 and Romer, 1984) point out that, even when individuals maximize self-interest, social norms that are financially costly to the individual may nevertheless persist if individuals are sanctioned by loss of reputation when disobeying the norm.3 Using a repeated game framework, Bovenberg (2002) formalizes various roles of social norms and values that facilitate economic cooperation. He argues that social considerations of corporate stakeholders (including consumers, employees and shareholders) may incite corporations to care for public goods such as the natural environment, even when such social considerations do not yield a direct benefit to the stakeholders. This paper studies the impact of ethics and stakeholder governance on the risk-adjusted performance of the money-management industry. We define stakeholder governance as the amalgam of good corporate governance (protecting shareholders' interests), sound stakeholder relations (protecting the interests of other stakeholders, including those of employees and local communities), and environmental care (protecting the environment). Stakeholder governance (as defined by Tirole (2001:4)) takes thus a broader view: it is “the design of institutions that induce or force management to internalize the welfare of stakeholders”. Over the past decade, ethical mutual funds, often also more broadly called socially responsible investment (SRI) funds4 which screen their investment portfolio based on ethical, social, corporate governance or environmental criteria, have experienced an explosive growth around the world. The assets in the SRI portfolios reached $2.3 trillion in 2005 or approximately 9.4% of the total universe of professionally managed assets in the US (Social Investment Forum, 2005). This type of funds provides an ideal setting for studying the economic effects of ethical/governance/social strategies for the following reason. Investors in SRI funds explicitly pursue two types of goals: the economic rational goal of wealth-maximization and social responsibility. SRI investors are socially conscious and derive non-financial utility by holding assets consistent with their ethical and social values. Still, by investing in mutual funds rather than giving money to charity, SRI investors desire to enhance their financial utility as they expect positive risk-adjusted returns on their investments. Academic interest has followed the rapid growth of the SRI industry. Most studies focus on SRI fund performance in individual countries (mainly, the US and the UK). Hamilton et al., 1993 and Goldreyer and Diltz, 1999, Statman (2000), and Bello (2005) show that the performance of SRI funds in the US is not significantly different from that of non-SRI funds. In contrast, Geczy et al. (2003) show that the financial costs of SRI screens on mean-variance optimizing investors can be substantial. Specifically, the SRI constraints impose a cost of more than 1.5% per month on investors believing in asset selection skills, i.e. investors who rely heavily on individual funds' historical risk-adjusted returns to predict future performance.5 The UK evidence (Luther et al., 1992, Mallin et al., 1995 and Gregory et al., 1997) concludes that the difference in performance between SRI and non-SRI funds is not statistically different from zero. A similar conclusion is drawn by Bauer, Otten and Tourani Rad (2006) and Bauer, Derwall and Otten (2006) for Australian and Canadian SRI funds, respectively. Multi-country studies are undertaken by Schroder (2004) for US, German and Swiss SRI funds; by Bauer et al. (2005) for US, UK and German funds; and by Kreander et al. (2005) for SRI funds in a few European countries. A detailed review of this SRI literature can be found in Renneboog, Ter Horst and Zhang (in press). Still, it is rather difficult to draw definitive conclusions on SRI performance as some of the performance evaluation methods used in the above papers are deficient. For example, some studies use the CAPM model to evaluate fund performance without controlling for other risk factors such as size, book-to-market value, and momentum. In addition, given that the above studies are not only based on different methodologies but also on fairly small samples, different sample periods and benchmarks, international comparisons of SRI performance are difficult to make. This paper makes the following four contributions to the literature. First, we relate the SRI fund performance to the performance of conventional funds in each country. This setup allows us to examine whether SRI fund investors pay a price for the ethical, corporate governance, social and environmental policies that firms have adopted across different institutional settings and in different phases of the economic cycles. In order to pursue social objectives, SRI funds employ a set of investment screens that restrict their investment opportunities. On the one hand, the exclusion of companies based on SRI screens may constrain the risk-return optimization and negatively influence fund performance. For instance, SRI funds typically do not invest in ‘sin' stocks, i.e. publicly traded companies involved in producing alcohol or tobacco and in gambling, although these stocks have historically outperformed the market (see Hong and Kacperczyk, 2005). SRI investors who derive non-financial utility from investing in companies meeting high ethical/social standards may be content with a lower rate of return. Therefore, we expect SRI funds to do worse than their benchmarks and than conventional mutual funds (the underperformance hypothesis). On the other hand, the labor-intensive screening process applied by SRI funds may generate value-relevant information and yield superior fund performance (the overperformance hypothesis). Under this hypothesis, the SRI screens are usually also used as filters to identify managerial competence and superior corporate governance, or to avoid the potential costs of corporate social crises and environmental disasters. Second, the paper investigates whether or not ethical investors are able to select the SRI funds that will generate superior performance in subsequent periods (a smart money effect). Geczy et al. (2003) show that the fund selection process of SRI investors determines the performance of the SRI fund portfolios relative to that of conventional portfolios. While the Geczy et al. study assumes that investors make fund selection decisions in a Bayesian way based on a funds' past performance, expenses and turnover, a number of other financial and non-financial fund attributes may significantly influence SRI investors' decision processes (see also Renneboog, Ter Horst and Zhang, 2006). We contribute to this line of research and examine the performance of SRI investors' portfolios by tracking the actual asset allocation decisions of investors (i.e. the decisions to invest or withdraw money) instead of making assumptions on investors' fund selection processes. Third, the paper studies the impact of SRI screens on fund returns and risk loadings, an issue that plays a central role in the SRI fund industry but has not yet been explored in the academic literature. More specifically, we analyze the question whether or not the screening intensity (the number of screens employed) and screening criteria (i.e. sin, ethical, social, corporate governance, and environmental screens) influence the risk-adjusted returns and risk exposure of SRI funds. We examine the joint impact of other fund characteristics, such as fund size, age, the fee structure and the reputation of fund families on fund returns and risk. Finally, we study the risk and return characteristics of SRI funds using a unique dataset consisting of nearly all SRI mutual funds around the world (Europe, North America, and Asia-Pacific). Hence, our sample covers a larger set of countries over a longer period of time than previous studies. The paper yields several interesting results. First, the average SRI funds in the US, the UK, and most continental European and Asia-Pacific countries strongly underperform their Fama-French-Carhart (FFC) benchmarks. In particular, the risk-adjusted returns of the average SRI funds in Belgium, Canada, France, Ireland, Japan, Netherlands, Norway, Singapore, and Sweden are between − 4% and − 6% per annum. The FFC-adjusted alphas of UK and US SRI funds are − 2.2% and − 3.4%, respectively. These results imply that the firms included in SRI funds and hence meeting high ethical/social standards and strict stakeholder governance criteria may be overpriced by the market or that SRI funds are too expensive. It seems that investors pay a price for ethics. It is important to analyze the risk-adjusted returns difference between SRI and conventional mutual funds as it may well be the case that conventional funds are also underperforming the benchmarks especially when active management is costly. We find that the SRI funds' alphas are in almost all countries lower than those of conventional funds. Still, as the conventional funds do indeed also underperform the FFC benchmarks, their risk-adjusted returns are mostly not statistically different from those of their SRI counterparts. Exceptions are France, Ireland, Sweden, and Japan where the underperformance hypothesis for SRI funds unequivocally holds. When we augment the Fama-French-Carhart four-factor model by an ethics factor, we confirm that the SRI funds have a higher exposure to this ethics factor. However, the difference between five- and four-factor alphas of SRI funds is economically small. The fact that a higher fraction of the return variation of ethical funds can be replicated by the well-known risk factors may indicate that SRI funds gradually converge to conventional funds by holding similar assets in their portfolios (or that conventional funds become more ethical or socially responsible). Second, we find mixed results on a ‘smart money' effect in the SRI fund industry: although ethical investors are unable to identify the funds that will outperform, there is some fund-selection ability to identify the ethical funds that will perform poorly. Third, the screening activities and processes of SRI funds have a significant impact on the risk-adjusted returns. Funds adopting a community involvement policy or employing an in-house SRI research team to screen portfolios have better returns than SRI funds without such processes policies. Fund returns decrease with screening intensity on social and corporate governance criteria (proxied by the number of social and governance screens applied). This is also consistent with the underperformance hypothesis of SRI funds stating that high SRI screening intensity constrains the risk-return optimization and does not help fund managers to pick underpriced stocks. Finally, while fund size erodes the returns of conventional funds, there is no such effect for SRI funds. This implies that larger SRI funds are not subject to decreasing returns to scale. However, fund management fees significantly reduce the risk-adjusted returns of both SRI and conventional funds. The remainder of the paper is organized as follows. Section 2 develops the hypotheses and Section 3 describes the data on SRI and conventional funds, investment screens, and performance benchmarks. Section 4 presents the returns and risk characteristics of SRI funds and Section 5 focuses on the investors' portfolios of SRI funds, more specifically on the smart money effect. While Section 6 examines the determinants of returns and risk of SRI funds, Section 7 concludes.
نتیجه گیری انگلیسی
This paper contributes to the literature on socially responsible investments and stakeholder governance as it studies the risk and return characteristics of nearly all ethical/SRI mutual funds around the world. Our main hypothesis is that ethical/social/environmental/governance considerations influence the stock prices and that investors pay a price for the use of SRI screening by funds. The main reason why SRI investors may be willing to pay such a price for ethics or social responsibility is based on aversion to corporate behavior which is deemed unethical/asocial. Investors of SRI funds may thus explicitly deviate from the economically rational goal of wealth-maximization by pursuing social objectives. Alternatively, one would expect SRI funds to outperform its benchmark portfolios or conventional funds if SRI screening processes unveil new value-relevant information not completely embedded in the share prices. We find that SRI funds in many European, North-American and Asia-Pacific countries strongly underperform domestic benchmark portfolios (such as the Fama-French-Carhart factors). In particular, the average risk-adjusted returns of the SRI funds in Belgium, Canada, France, Ireland, Japan, Malaysia, the Netherlands, Singapore, Sweden, the UK and the US range from − 2.2% to − 6.5% per annum. However, when we compare the alphas of the SRI funds with those of matched conventional funds, we conclude that there is no statistically significant evidence that SRI funds underperform their conventional counterparts in most countries (as the differences in alphas are not statistically different from zero). Exceptions are the SRI funds in France, Ireland, Sweden, and Japan for which we find evidence consistent with investors paying a price for ethics: these SRI alphas are 7% to 4% per annum below the conventional alphas. Only these latter results are consistent with the underperformance hypothesis of SRI funds. As the underperformance of SRI and actively managed conventional funds may be due to transaction costs (such as fees), we compute the gross alphas of all fund portfolios before deducting management fees. The gross returns of about half of the country portfolios underperform the benchmarks by more than 3% per annum. We also measure the potential welfare costs of inadequate diversification – which may be induced by the SRI screens – by considering investors' opportunity costs of bearing idiosyncratic risk. We do not find any such costs. In case the conventional four-factor pricing model does not capture an ‘ethics or SRI' style, the alpha may reflect the expected returns associated with the missing factor. In addition, the screening activities of SRI funds vary across countries and they may influence the SRI fund risk loadings on this ethics or SRI style. An alternative explanation is that, to the extent that SRI funds invest in companies that are considered ethical or socially responsible, the companies meeting high ethical standards may be overpriced by the stock markets. We find that SRI funds do indeed have higher loadings on the ethics factor than conventional funds. However, after controlling for the ethics factor, the SRI funds still underperform the benchmarks (now comprising five factors) in many countries and the SRI alphas are significantly lower than those of conventional funds in Canada, Ireland and Japan. Overall, we find that adding the ethics factor to the four-factor model has only limited influence on the risk-adjusted returns of ethical funds. Consequently, the underperformance of ethical funds is not driven by a missing ethics style factor. Furthermore, the fact that – over time – a higher fraction of the return variation of ethical funds can be replicated by the well-known risk factors, may indicate that SRI funds gradually converge to conventional funds by holding similar assets in their portfolios (or that conventional funds become more ethical or socially responsible). We also examine whether a ‘smart money' effect exists for SRI funds by investigating the relation between SRI money flows and future performance. We find mixed results: while there is some fund-selection ability in identifying poorly performing ethical funds, ethical investors are unable to identify the funds that will outperform their benchmarks in subsequent periods. Our results on the determinants of SRI funds' returns and risk loadings suggest that the screening activities of SRI funds matter. Funds with a higher number of corporate governance and social screens yield lower risk-adjusted returns. All else equal, funds with one additional screen are associated with a 1% lower 4-factor-adjusted return per annum. This is consistent with the underperformance hypothesis of SRI funds stating that high SRI screening intensity constrains the risk-return optimization and does not help fund managers to pick underpriced stocks. In addition, employing an in-house SRI research team increases the 4-factor adjusted return by 10 basis points per month. Finally, we find that fund size erodes the returns of conventional funds. Interestingly, the effect is not significant for SRI funds, which implies that larger SRI funds are not subject to decreasing returns to scale. Fund management fees significantly reduce the risk-adjusted returns of both SRI and conventional funds.