This chapter reviews the literature relating to socially responsible investment in a funds management context. A quick review of funds management, specifically mutual funds is presented in Section 2.1. This is followed by a brief introduction to socially responsible investment (SRI) funds in Section 2.2, which includes an explanation of the investment screening approach used by these funds. Following this, Section 2.3 summarises the existing evidence on the performance of SRI funds relative to non-SRI funds. Thereafter, Section 2.4 examines how variation in screening intensity affects performance of SRI funds, while Section 2.5 examines how this variation affects the flow-performance relationship of fund flow into SRI funds. Finally, we draw together the extant literature and outline the contribution of this dissertation to the literature (Section 2.6).
2.1 Funds management – mutual funds
A mutual fund is a company that pools money from many investors and invests money in stocks, bonds, short-term money market instruments, or other securities. essay_footnotecitation">[essay_footnotecitation_link" href="http://freedissertation.com/litreview/literature-review-about-socially-responsible-investments.php#ftn1" name="bodyftn1">1] Investors purchase units of these funds. Funds are managed by professional fund managers who trade the fund’s investments based on its investment objectives. Mutual fund managers and investors generally have different objectives – managers are concerned with money-inflows and the resulting management fees, while investors desire high risk-adjusted return on their investment at low fees.
Early studies show that investors rely heavily on past performance (Capon et al.., 1996) and both new and old shareholders react to past performance (Johnson, 2006). Therefore, good performing funds receive increased cash flows, leading to higher manager compensation (Chevalier and Ellison, 1997; Sirri and Tufano, 1998). However, the reaction of investor flow to fund performance is not identical in both directions – the best performing funds receive the largest share of fund inflows but the poor performing funds do not experience equivalently large outflows (Ippolito, 1992; Sirri and Tufano, 1998). This suggests that once money is invested in a fund, it tends to be relatively sticky (Gruber, 1996). The flow-performance relationship is also more prominent in funds that engage in marketing activities or receive greater media attention as fund investors are generally not trained in portfolio analysis and base their investment decision on published information (Sirri and Tufano, 1998).
2.2 SRI funds
Socially responsible investment (SRI) can be defined broadly as ‘an investment process that considers the social and environmental consequences of investments, both positive and negative, within the context of rigorous financial analysis’ (Social Investment Forum, 2007). Often called ethical investments or sustainable investments, this type of investment has become increasingly popular in recent years. According to the Social Investment Forum, the number of SRI funds in the US have increased from 55 in 1995 to 260 in 2007, now accounting for an estimated $2.71 trillion out of $25.1 trillion invested in the U.S. investment marketplace. essay_footnotecitation">[essay_footnotecitation_link" href="http://freedissertation.com/litreview/literature-review-about-socially-responsible-investments.php#ftn2" name="bodyftn2">2]
There is a variety of choices for the US socially responsible investor. In fact, there are now even socially responsible indexes that have been created. essay_footnotecitation">[essay_footnotecitation_link" href="http://freedissertation.com/litreview/literature-review-about-socially-responsible-investments.php#ftn3" name="bodyftn3">3] Since the launch of KLD Research & Analytics, Inc.’s pioneering Domini 400 Social Index (DSI 400) in 1990, which is modeled on the S&P 500 Index, there are 17 different socially and environmentally screened index funds today.
2.2.1 A brief history of SRI
The early stages of the SRI movement can be traced back to the nineteenth century, especially amongst religious movements such as the Quakers and Methodists. Specifically, these groups excluded investments that would go against their beliefs. essay_footnotecitation">[essay_footnotecitation_link" href="http://freedissertation.com/litreview/literature-review-about-socially-responsible-investments.php#ftn4" name="bodyftn4">4] Knoll (2002) records that such non-financial ‘exclusionary’ behavior in investment choice became a highlight in 1960s during the Vietnam War, where funds like the PAX World Fund was set up with a mission to avoid any investments that derive profits from weapons production. Another milestone moment for SRI occurred in the 1980s – investors who were against the apartheid regime in South excluded any investments in South Africa.
The most common factor that distinguishes a SRI fund from a non-SRI or conventional fund is the use of social criteria as part of the portfolio construction process. According to Knoll (2002), ‘SRI means only that nonfinancial criteria are part of the selection process for choosing investments’.When creating a socially responsible portfolio, a fund manager first analyses the entire universe of investments against non-financial criteria to establish a subset of investments that are acceptable in terms of investors’ ethical, social, religious or other preferences. In the SRI industry, these non-financial criteria are referred to as screens.
Investment screens generally fall into either one of two categories – positive or negative screens. Negative screens can be thought as avoidance or exclusionary screening. The most popular examples of negative screens include tobacco, alcohol, gambling or weapons, most of which are associated with the traditional type of screening against ‘sin’ industries.
On the other hand, positive or inclusionary screens aim to select shares that meet superior corporate social responsibility (CSR) standards. This can be thought of as ‘best-in-class’ in terms of performance in a particular social cause. Most popular examples of positive screens relate to issues about the environmental sustainability, human rights advocacy and community involvement. Today, it is most common for funds to use a combination of both positive and negative screens. essay_footnotecitation">[essay_footnotecitation_link" href="http://freedissertation.com/litreview/literature-review-about-socially-responsible-investments.php#ftn5" name="bodyftn5">5]
2.3 Performance and SRI mutual funds
2.3.1 Performance benchmarks
To date, much of the academic research on socially responsible investments focuses on analyzing their performance. For mutual funds, the Jensen’s alpha is a popular measure for performance. This traditional approach of evaluating performance introduced by Jensen (1968) is conducted by estimating the following OLS regression based on the capital asset pricing model (CAPM):
where is the return on mutual fund i over time t, is the return of the market index, is the return on a risk free asset and is the Jensen’s alpha. A positive alpha can be interpreted as the excess fund return adjusted for the market risk, while measures the mutual fund’s exposure to the market risk. essay_footnotecitation">[essay_footnotecitation_link" href="http://freedissertation.com/litreview/literature-review-about-socially-responsible-investments.php#ftn6" name="bodyftn6">6] Early research on performance of SRI portfolios uses a CAPM based single index model like the one above (Hamilton, Joe and Statman, 1993; Goldreyer and Diltz, 1999; Statman, 2000).
However, questions were raised regarding the adequacy a single index model to explain mutual fund performance. Multi-factor models are thus considered in an attempt to better explain the cross-sectional variation in fund return, thus potentially give a better explanation of mutual fund behavior. Recent studies applied the Fama and French (1993) 3-factor model which considers two additional risk factors, size and book-to-market. More commonly used is the Carhart (1997) 4-factor model, which extends the Fama-French model by adding a fourth factor that captures the momentum factor anomaly identified by Jegadeesh and Titman (1993), which is as follows:
where RMRFt is the excess return on a value-weighted aggregate market proxy, SMBt is the difference in returns between a portfolio of small stocks and a portfolio of big stocks, HMLt is the difference in returns between a portfolio of high book-to-market stocks and a portfolio of low book-to-market stocks and PR1YRt is the current month’s difference in returns between the previous year’s best-performing and worst-performing stocks.
2.2.2 Empirical evidence of SRI performance – SRI and conventional funds
Whether SRI creates shareholder value is ultimately an empirical question. Among the first studies conducted on the performance of SRI mutual fund is Hamilton et al. (1993) who investigate the performance of 32 SRI funds and 320 randomly-selected non-SRI funds in the US for the period 1981-1990. By measuring the CAPM-based Jensen’s alpha against a value-weighted NYSE index, the authors discover that SRI funds with a longer history (established pre-1985) has a higher average alpha than its corresponding non-SRI funds (-0.06% for SRI compared to -0.14% for non-SRI). However, in the case of SRI funds with shorter history (post-1985), SRI funds record an average monthly alpha of -0.28%, underperforming non-SRI funds (-0.04%). For the period between 1990-1998, Statman (2000) compares 31 SRI funds to 64 non-SRI funds matched by fund size. Both fund groups have similar average expense ratios: 1.50% and 1.56% for SRI and non-SRI funds respectively. Although the average monthly alpha is -0.42% for SRI funds and -0.62% for non-SRI funds, the difference is not statistically significant and the results remain unchanged if Jensen’s alpha is measured against the S&P 500 Index or the DSI 400.
Goldreyer and Diltz (1999) and Bello (2005) also show that there is no advantage (or disadvantage), on a risk-adjusted basis, in choosing SRI funds. In fact, both types of funds display strong similarity in terms of characteristics of assets held, level of portfolio diversification and both fail to outperform the DSI 400 and S&P 500 between the years 1994-2001 (Bello, 2005). Studies in the UK (Luther et al., 1992; Mallin et al., 1995 and Gregory et al., 1997) also conclude that there is no significant difference in alpha performance between SRI and non-SRI funds. Bauer, Otten and Tourani Rad (2006) and Bauer, Derwall and Otten (2006) also reach similar conclusions for Australian and Canadian SRI funds respectively.
2.3.3 Performance between SRI funds
While most literature on SRI compared the performance of SRI funds with that of conventional, non-SRI funds, little is done to study how SRI funds are different from one another in terms of the investment screens that they apply. For instance, Goldreyer and Diltz (1999) show an interesting result that SRI funds with positive screens outperform SRI funds without. The former recorded a monthly alpha of -0.11% which is statistically different from the latter (-0.81%). Although this finding is only based on a sample of 29 funds, it supports the hypothesis that fund performance is affected by the type of investment screen used.
The above observation suggests a possibility that individual, non-financial screens may work in opposite directions thus creating confounding results when screens are aggregated. For example, it is possible that news pertaining to good employee relations is negatively related to expected returns while positive news on environmentally friendly production is positively related to expected returns (Scholtens and Zhou, 2008). Galema et al. (2008) calculate the Carhart (1997) four-factor alpha of 12 portfolios constructed based on whether stocks score on particular strength or concern screens and found that no portfolio showed any significant outperformance. As such, there is insufficient evidence to conclude that an aggregate analysis eliminates individual dimension of various SRI screens.
2.4 Screening intensity
The earliest known literature that considered the impact of screening intensity on the performance of SRI funds is by Barnett and Salomon (2006). In their paper, they analyse the relation between screening intensity and performance for a sample of 67 US SRI funds and did not find evidence of a negative linear relationship. Interestingly, they find a significant curvilinear relationship between screening intensity and performance – fund performance initially declines, but later rebounds as number of screens increase to a maximum. Based on this result, the authors propose that stakeholder theory and portfolio theory can actually be complementary. They argue that funds with many social screens effectively only select from an investment pool of companies with high socially responsible standards. These companies are expected to have higher performance given their positive stakeholder relationships. However, funds with few social screens are exposed to a similar investment opportunity set as conventional funds, hence they are better positioned to diversify and maximize performance using the traditional mean-variance framework. Funds with middle-level screening may neither be able to diversify sufficiently nor screen effectively to exclude underperforming firms. Hence, as screening intensity increase, fund performance is expected to decrease initially due to loss in diversification benefits but performance improves as the benefits of positive stakeholder relationships begin to take effect.
In another study, Renneboog et al. (2008) investigate the relation between performance and SRI screens. The authors only find weak evidence (coefficients significant at 10% level) in support of a negative relationship between intensity of social screens and performance, but no such relationship for intensity of environmental, ethical or sin screens. The possibility of a curvilinear relationship was not examined in this study. The authors also examine how each type of screening intensity affects each of the Carhart (1997) four factors.
The above studies have thus far only considered the effects of screening intensity on performance, measured by risk-adjusted return. Lee et al. (2010) extends the literature on screening intensity relation by considering its effects on both return and risk. The authors find no evidence of screening intensity having any effect on a fund’s raw return, but there was evidence of a decrease of 70 basis points per screen when the analysis was performed using the Carhart (1997) model. Similarly, there was also no evidence of screening intensity having an impact on unsystematic risk, but there was evidence for a negative relation between screening intensity and the fund’s total risk (measured by standard deviation of returns). No evidence of a curvilinear relationship between screening intensity and performance was present, but there is such a relationship between screening intensity and systematic risk.
2.5 SRI and fund flow
From a fund manager’s perspective, capturing the most fund flow would be a key objective as their remuneration is often linked to how the value of assets under management. Fund flow is an indicator of investor choice and, typically, an investor would make a decision based on the risk-return profile of the fund, their personal level of risk tolerance and investment horizon. Investors in SRI funds are a slightly different group as they are likely to care more about social, ethical or environmental issues than financial performance.
The first study on the determinants of fund flow in the SRI industry was conducted by Bollen (2006), which focused on a univariate analysis of fund flow and past returns for US SRI mutual funds. The results of the study show that, in the US, the volatility of fund flow is lower in SRI funds than in conventional funds. In addition, the fund flow of SRI funds are less sensitive to lagged negative returns than flows in conventional funds, but more sensitive to lagged positive returns.
Benson and Humphrey (2008) also study the flow-performance relationship in U.S. SRI mutual funds. The authors find that on average, SRI fund flow is less sensitive to returns than conventional fund flow. There is also evidence to confirm the asymmetric relation between flow and performance for both conventional and SRI funds. The authors also note that SRI investors are more likely to reinvest in funds they already own. The heterogeneity that exist in the screens employed by SRI funds makes it difficult for SRI investors to find an alternative fund that will meet exactly their non-financial objectives, thus they are less likely to switch funds compared to ‘conventional’ investors.
2.5.1 Review of Renneboog et al. (2006)
Renneboog et al. (2006) examine a database of 410 SRI mutual funds around the world (excluding US), studying the fund flows into and out of the SRI fund industry. They find that SRI investors have similar characteristics to conventional mutual fund investors – they both chase past returns, past return rankings, and persistence in past performance. In particular, SRI funds that can be denoted as persistent winners receive about 30% more inflows tan persistent losers. Unless a fund persistently underperforms, SRI investors care more about past positive returns than about past negative returns. Further, the authors find a positive relationship between screening intensity and fund flow.
To the best of my knowledge, so far this is the only paper that has made an attempt to study the relationship between SRI fund flow and screening intensity. Although their result seems to be consistent with the intuition that higher screening intensity means better social performance and hence higher fund flow, the methodology used by the authors may dampen the reliability of the results.
Firstly, the benchmark sample of mutual funds used consists of only mutual funds that operate in the UK. This benchmark sample is inappropriate as it is too narrow compared to the worldwide sample of SRI mutual funds that the authors in their paper. This approach of using an aggregated worldwide sample also may not be able to isolate country specific screening-flow relations. For example, we know that investors in Australia and UK are more concerned about environmental issues relative to US investors who are care more about the traditional ‘sin’ issues. essay_footnotecitation">[essay_footnotecitation_link" href="http://freedissertation.com/litreview/literature-review-about-socially-responsible-investments.php#ftn7" name="bodyftn7">7]
In their regression model, the authors accounts for differential effects of screening arising from the different type of screens used by including dummy variables for each of the broad screening categories ( ethical, environmental, Islamic, activism policy, sin and social). Using this set-up, it is unclear to me what the benchmark type of screen would be. In addition, there is also a potential dummy variable trap in the authors’ regression model.
2.6 Conclusion and contribution to literature
A survey of extant literature reveals that there is a dense amount of literature on SRI fund performance, but mostly assume that SRI funds are homogeneous in nature in that they do not account for differences in quantity and type of screening used. Based on the three papers on screening-performance relationship reviewed, there is no strong evidence in support of a negative relationship between screening and SRI fund performance.
Studies on the flow-performance relationship of SRI funds reveal that while both SRI and non-SRI fund flow are asymmetric in their response to past performance, the former is less sensitive to returns than the latter (Benson and Humphrey, 2008). Given that screening intensity has no significant effect on returns, then the question is whether a fund manager can command higher fund inflow by adjusting the number of screens used.
So far, there is only one paper by Renneboog et al. (2006) that made an attempt to explain a possible direct relationship between screening intensity and fund flow. My research aims to build on the shortcomings of the Renneboog paper by improving research methodology and remaining more focused by just doing a study on one country (US).
Cite This Work
To export a reference to this article please select a referencing stye below: