Performance of SRI funds during the financial crisis : small cap vs large cap and US vs Europe

(1)

Performance of SRI funds during the financial crisis:

small cap vs large cap and US vs Europe

Anne van Zelst 10723803 University of Amsterdam Under supervision of Dr. Evgenia Zhivotova

July 2018 Abstract

The aim of this paper is to provide an insight in the performance of socially responsible investment (SRI) funds in a period which contains the financial crisis of 2007-2008. It compares the performance of SRI funds versus general funds in Europe and the US and compares both markets. Furthermore, also the performance of small cap and large cap SRI funds is taken into account. A sample of the monthly returns of 97 SRI and 130 general funds from the US and 155 European SRI and 87 general European funds from the period 1 October 2003 until 1 May 2018 are obtained for this research. To measure performance, the CAPM-model and the Four Factor model are used and show that there is no evidence of difference in performance in Europe of between the European market and US market. However, evidence is found that SRI funds underperform general funds in the US. The main conclusion is that SRI funds do not perform differently from general funds, also in the financial crisis.

Keywords: Socially Responsible Investing (SRI), financial crisis, CAPM-model, Four Factor model, performance evaluation, small cap and large cap

(2)

Statement Of Originality

This thesis document is written by Anne van Zelst who declares to take full responsibility for the contents of this document.

I declare that the text and work presented in this document are original and that no source other than those mentioned in the text and its references have been used creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

4. Results……….……….….10 4.1 Summary Statistics………..……….10 4.2 Test results………..………...15 4.3 Regression results………...………..18 5. Conclusion……….……….…...24 6. Discussion……….………….…....25 Reference List……….………27 Appendix……….…………..31

(4)

1.1 Introduction

Socially responsible investing has become more and more popular, due to a changing social and environmental glance. In recent years, the interest in socially responsible

investing (SRI) has increased significantly. In 2007, the Social Investment Forum (SIF) reported that one out of nine dollars of management under assets in the US was involved with SRI. From 2012 to 2016, assets of SRI funds in the US almost doubled to $8.72 trillion (The Forum for Sustainable and Responsible Investment, 2017) and these SRI’s contain a wide variety of different assets classes, such as private equity, stock investment, fixed income, cash, venture capital and real estate (The Forum for Sustainable and Responsible Investment, 2018). Furthermore, The Report on Socially Responsible Investing Trends in the United States (The Forum for Sustainable and Responsible Investment, 2018) reported a 2.71 trillion-dollar investment in socially responsible funds in 2010. This corresponds to a share of approximately 11% of total assets under management in the US. From 2014 to 2016, the USSIF reported an increase in assets under managements of 33% for SRI assets (The Forum for Sustainable and Responsible Investment, 2018).

Several reasons for this growing interest in SRI are given by Schwarz (2003), such as growing investor concerns about environment, increasing business ethics, increasing advertisement of SRI funds and greater exposure to the media. Several studies have explained the growth of SRI funds and their importance, with the insight that investors are not exclusively considering wealth maximization but are also concerned with moral and social welfare. This conflicts with the neoclassical economic theory (Etzioni, 1998; Cummings, 2000; Statman, 2000; & Fehr and Gatcher, 2000). For individual investors, evaluating and monitoring performance of social or environmental conscious companies is usually time-consuming and complex. These barriers created a way for SRI funds to arise as vehicles to screen investments on behalf of the individual investors.

Besides this, the financial crisis of 2007-2008 also influenced the way of company behavior with respect to business ethics. Lewis et al. (2010) suggest in their study that part of the crisis of 2007-2008 was caused by the lack of business ethics. The question that arises with SRI, is if SRI funds perform better when compared to general funds. Secondly, how SRI funds perform compared to general investment funds in the US during identified break periods, especially during the financial crisis of 2007-2008, whilst taking in account risk and performance. Thirdly, if the performance of SRI funds in the European market is comparable to the US market. Finally, if the performances between small and large cap SRI funds in the US are significantly different.

In this paper, the performance of SRI and general funds in the US and Europe will be researched on the basis of the CAPM-model and Four factor model. Also, a t-test and Wilcoxon signed rank-sum test will check if the returns of SRI funds and general funds and small cap and large cap funds differ. This paper is organized as follows: in section 2, the literature review is described. It contains a capture of the definition of SRI, past studies on SRI performance in different regions and the influence of the financial crisis. Section 3 describes the methodology and the data used in this paper. Section 4 explains the results of

(5)

this research. Section 5 will contain the conclusion and section 6 a discussion on limitations and recommendations of this research.

2.Literature Review

2.1 SRI Definition

Throughout the years, there have been different proposals for defining SRI. Rockness and Williams (1998) were one of the first to attempt to find an objective definition for SRI. Cowton (1999) came to a definition for SRI as “a set of approaches which include social or ethical goals or constraints as well as more conventional financial criteria in decisions over whether to acquire, hold or dispose of a particular investment” (p.99). Schueth (2003) defines SRI as “the process of integrating personal values and societal concerns into investment decision-making” (p. 190). Moreover, Budde (2008) stated SRI as “those investments strategies that consistently and explicitly consider social factors as part of the investment process” (p. 14 ). The adapted definition of SRI by Renneboog (2008) is “an investment process that integrates social, environmental, and ethical considerations into investment decision making” (p. 395). These statements about SRI point out that the term ‘socially responsible investment’ has no exact definition, but refers to a set of screening mechanisms for investments. These screenings criteria include social, ethical and

environmental considerations for investing in addition to screening criteria by general funds, as discussed by Cowton (1999) and Schueth (2003) and Renneboog (2008).

2.2 Past studies on SRI Performance

The prospects for SRI are often presented as relatively positive. Heal (2005) and Geunster et al. (2011) argued that socially responsible companies pay attention to their input and waste of their operations, which leads to a more efficient operating company. Moreover, Michelson (2004) argued that the outperformance of ethical funds occurs due to their screening and monitoring practice, because ethical firms focus on good management and sustainability. Going more into detail, Verwijmeren and Derwall (2010) discussed that SRI firms may suffer less from legal prosecutions and experience a more stable relationship with communities and governments. Also, firms with high results of employee satisfaction experience lower bankruptcy (Verwijmeren and Derwall, 2010). This explains that

irresponsible behavior is positively related to systematic risk and socially responsible behavior is slightly negatively related to systematic risk, according to the findings of Oikonomou et al. (2012).

But when examining performance, several studies have shown in their research that there is no significant difference between the SRI indices performance and general indices

(6)

& Otten, 2005; Schröder, 2006; and Utz and Wimmer, 2014). Gregory and Whittaker (2007) acknowledged this in their study, by looking at the performance of SRI and non-SRI funds in the UK. Bauer et al. (2005) found no significant difference between the returns of SRI and conventional funds in the US during the years 1990-2001, while controlling for common factors. Furthermore, Bauer, Derwall and Otten (2007) confirmed this again when doing research on the difference between SRI and conventional fund performance in Canada during.

In addition, Statman (2000) King and Lenox (2001) concluded that financial performance of firms that are stronger involved with the environment was better than weaker environmental firms. Also, Luther et al. (1992) provided evidence of superior

performance of SRI funds, although this evidence was weak due to a small sample size of 15 SRI unit trusts. To continue, Galema et. al (2008) and Kempf and Osthoff (2007) found that SRI funds might outperform conventional funds in the US, while Fernandez and Matallin (2008) and Gil-Bazo et al. (2010) came to the same conclusion by comparing Spanish and US mutual funds.

Another way to examine the performance of SRI funds is to study SRI funds themselves, rather than a comparison of indices or general funds. Derwall et al. (2005) showed some evidence that environmentally clean firms have abnormal positive returns. Other studies supported this finding, by providing abnormal returns using a wide span of measures on environmental performance, although these findings are not statistically significant (Kempf and Osthoff, 2007; Statman and Glushkov, 2009).

On the other hand, El Ghoul and Karoui (2017) performed a study on corporate socially responsible (CSR) funds and found that higher CSR-funds underperformed compared to lower CSR-funds. Also, Borgers et al. (2015) supported this by comparing funds that hold more or less ‘sin’ stocks. Furthermore, Bauer et al. (2006) found costs that could make SRI more expensive and less profitable, which would give screening of SRI a negative relation with the returns. Moreover, several studies implied that including ethical or social

considerations in the investment policy can increase risk, transaction costs and management fees (Rudd, 1981; Hickman et al., 1999; & Tippet, 2001). Furthermore, Belkaoui and Karpik (1989) argued that SRI screening has no economic aim, which would cause an adverse effect on profit and a negative effect on the returns of SRI funds. Carhart (1997) confirmed this argument, by showing a negative relation between fund performance and fund expenses. To support these theories, Renneboog et al. (2008) noted in a study on different regions that SRI funds underperformed its benchmark, but these risk-adjusted returns are not significant different from general funds.

But would these findings on the performance of SRI funds still hold during a financial crisis? While a recession or crisis generally indicates uncertainty and downward returns, Yelkikalan (2012) argues that CSR firms can stabilize their market position. His arguments are that CSR firms strengthen their business strategy, improve or stabilize the investor’s confidence and enlarge their market share. Varma and Nofsinger (2012) elaborated how socially responsible managers outperform mutual fund managers during times of crisis and

(7)

how mutual fund managers outperform social fund managers in non-crisis period. Tripathi and Bhandari (2016) examined the performance of SRI funds during the financial crisis in the Indian market. They found that SRI in India could be beneficial, especially during the

financial crisis of 2007-2008. 2.3 Relevance and hypotheses

Most of the previous studies analyzed the performance of SRI funds versus conventional funds in a period prior to the financial crisis of 2007-2008 (Hamilton et al., 1993; Reyes and Grieb., 1998; Goldreyer and Diltzl., 1999; Statman, 2000; Bauer et al. (2005); Amenc and Le Sourd, 2008; Climent and Soriano, 2011), not during identified break periods. Furthermore, the difference on performance between small and large cap SRI funds has not been profoundly studied until now according to the examined literature for this study. Bauman et al. (1998) argued that stocks of small-cap companies - not distinguishing SRI stocks - significantly outperformed large-cap stocks.

To summarize, there are different theories based on past literature about the under- or outperformance of SRI fund compared to general funds. Proponents of the theory that SRI funds outperform general funds argue that this is caused by improved efficiency or better social or environmental responsibility standards which can lead to a higher market value and more cost-efficiency than general funds and reduced risk. Opponents of the theory that SRI outperforms general funds argue that underperformance of SRI funds happens due to time-consuming screening methods used for SRI, less diversification, overpriced stocks or the exclusion of companies because they do not meet the SRI criteria. With the theories and discussed literature the following question arises: is there a difference in performance of SRI funds during the financial crisis of 2007-2008 in the US and Europe and between large-cap and small cap funds? These following four assumptions will attempt to answer this question:

1. SRI funds have less market exposure than general funds during the periods pre, during and post crisis.

2. SRI funds do not underperform general funds during the periods pre, during and post crisis (alpha is not statistically different).

3. Small cap SRI funds and large cap SRI funds generate no different returns during the period pre, during and post crisis.

4. SRI funds in the US do not outperform SRI funds in Europe during the period pre, during and post crisis.

This study is an adjustment to prior studies, by using a bigger sample than most previous studies, which can be helpful to eliminate selection bias. To build upon this, most previous studies in the field of SRI have only focused on SRI performance during the period before the start. Furthermore, most studies of SRI have only been carried out in a small number of areas, in particular one market. Finally, the comparison between performance

(8)

3. Methodology

3.1 Sample Selection

This study examines the performances during a fifteen-year period from October 2003 until June 2018 of two types of funds: socially responsible investment funds and general or non-SRI funds. Also, the differences between large-cap and small-cap SRI funds are observed to find a conclusion if large-cap SRI-funds outperform small-cap SRI funds. Finally, the performance of SRI is compared post, during and after the financial crisis of 2007-2008 of SRI funds in the US and Europe.

The three identified break periods chosen are consensus as followed: 1st of October 2003 until the 1st of October 2007, because the bankruptcy of Lehman Brothers is chosen as the starting point of the financial crisis. Secondly, some studies argue that the end tail of the financial crisis was at the end of 2011. In order to measure real effects of the crisis, the ending point of the crisis is chosen as 1st of December 2011. The last identified break period ends at the most recent date from which data could be achieved, which is 1st of May 2018.

For this research, data are collected from different data sources. The primary data source for SRI funds in the US in this study is the United States Forum of Sustainable and Responsible Investment (USSIF). Geczy (2005) et al., Areal (2010) and Silva and Wimmer (2012) used similar data when doing research on SRI fund performance. The USSIF provides a list of 204 SRI funds, which are using a screening method for their investment strategies. To improve comparability, equity funds only are retrieved from the dataset. Furthermore, dead funds are also included to minimize the survivorship bias. The USSIF listed SRI funds, give a sample of 97 equity funds, from which 59 large-cap and 38 small-cap funds. For the conventional portfolio a random sample of 70 large cap and 60 small cap mutual funds is used.

The SRI data for Europe used in this research is retrieved from yoursri.com, which provides a list of all SRI funds in Europe. The sample retrieved from yoursri.com gives a sample of 155 European SRI funds. Again, equity only funds are taken in account.

In order to use the CAPM single factor model and Fama-French model properly, the market risk premium and the risk-free rate in the US are gained from the Wharton

database. For the market return in the US, the S&P500 index is used, because this is the best fitting bench-mark according to Azar and Al Hourani (2010) when comparing mutual funds against four different benchmarks. The proxy for the market return in Europe is the Eurostoxx, as is also used in other studies on European funds. As for the risk-free rate in Europe, a portfolio of several government bonds is used as an indicator. The proxy used for the risk-free rate in the US is the three-month US treasury bill. The retrieved data contain time-series monthly returns from 1st of October 2003 until most recent. Monthly data are used, because they are readily available, gives a consistent view and minimize the bid-ask bounce. The data on the monthly returns of the SRI funds from the USSIF website and yoursri.com are gained via DataStream.

(9)

3.2 Variables Performance 3.2.1 Tests

To check whether there is a significant difference in returns of SRI funds and general funds, the non-parametric Wilcoxon signed rank-sum test and a t-test will be used. The returns for the whole identified period will be compared as well for the identified break periods: pre, at and post-crisis. This comparison will be done for the US market, the

European market and the returns from large and small-cap SRI funds in the US. The t-test is identified as follows:

(1)

where,

x̅1 and x̅2 are the returns of the SRI portfolio and the general portfolio s1 and s2 are the standard deviations of the two portfolios

n1 and n2 are the sample sizes of the two portfolios

Wilcoxon signed rank-sum test is an often-used alternative test for the t-test (Snedecor & Cochran 1989) and tests the following hypothesis:

1. H0: The monthly returns of small and large cap funds in the US are equal vs H1: The

monthly returns of small and large cap funds in the US are unequal

2. H1: The monthly returns of SRI and conventional funds in the US are not equal to the

monthly returns of SRI and conventional funds in Europe.

This test is assumed to suit better than the t-test is more suitable for non-normal

distributions than the two-sample t-test (Wilcoxon, 1945), for it this test is less sensitive to outliers. Let 𝑑i be the difference for a matched pair of observations, then:

𝑑i = 𝑥1i − 𝑥2i (2)

where, 𝑥1i is observation from the return of a SRI fund in month i, and 𝑥2i is the observation

from a general fund month i. As when using the t-test, a comparison will be done for the US market, the European market and the returns from large and small-cap SRI funds in the US. Then, the observed signed ranks 𝑟i is,

𝑟i = 𝑠𝑖𝑔𝑛 𝑑i 𝑟𝑎𝑛𝑘(|𝑑i |) (3)

where 𝑠𝑖𝑔𝑛 𝑑i is the sign of the differences and 𝑟𝑎𝑛𝑘(|𝑑i|) is the rank of the absolute value

of the differences of the two samples. Then the W-statistic is calculated, which is the absolute value of the sum of all ranks:

(10)

A z-statistic can be calculated using the w-statistic. Since the sample is large enough, a normal approximation is used. Then the z-statistic is identified as follows, where it is often presented as the sum of the positive signed ranks; w+,

𝑍 =

45 6 7(45)

:;<= >?@ (45)

(5)

where, m(w+) represents the mean of positive signed ranks and 𝑣𝑎𝑟 <BC (𝑤+)represents

the adjusted variance of positive ranks.

Also, to check whether there are structural breaks for the identified period (i.g. the period October 2007 - November 2011 is identified as the break period), the coefficients of two divided parts of the dataset are significantly different. A test introduced by Chow (1960) is used, where the following hypotheses are tested:

1. H0: 𝛽₀= 𝛾0 𝑎𝑛𝑑 𝛽1 = 𝛾1 … and 𝛽₀ = 𝛾0 (no structural break)

2. H1: 𝛽₀ ≠ 𝛾0 𝑎𝑛𝑑 𝛽1≠ 𝛾1 … and 𝛽₀ ≠ 𝛾0 (structural break)

Where 𝛽₀represents the coefficient of the constant in the regression for the non-crisis period, 𝛽₁represents the slope coefficient of the regression of the non-crisis period, 𝛾₀represents the coefficient of the constant in the regression for the crisis period and 𝛾₁represents the slope coefficient of the regression in the crisis period, 𝛽₀represents the i amount of other coefficients of the factors in the regression model in the non-crisis period and 𝛾₀represents the i amount of other coefficients of the factors in the regression model in the crisis period. To identify the crisis period, a dummy variable crisis is used, which is 0 if non-crisis and 1 if crisis period. The Chow-test is defined as follows:

(6) where RSS represents the sum of squared residuals of the regression model of the overall period from October 2003 - May 2018, RSS1 the sum of squared residuals of the regression model with dummy variable crisis=0, RSS1 the sum of squared residuals of the regression model with dummy variable crisis=2, 𝑘is the number of parameters, 𝑛₁and 𝑛2the number of observations in the crisis period and the non crisis period and the test statistic follows the F distribution with k and 𝑛1+ 𝑛2 - 2k degrees of freedom.

3.2.2 CAPM-model

A single-factor model is widely used based on past literature on mutual funds and therefore this model is considered (Bauer et al., 2005; Fransico & Pilar, 2011) . The

(11)

performance of SRI funds and general funds is studied by using the monthly returns of an equally weighted portfolio of all sampled funds. The most successful performance measures proposed by academic literature are Treynor (1965), Sharpe (1966) and Jensen (1968) measures. The Capital Asset Price Model (CAPM) single-factor is used in this research to determine the beta and Jensen’s alpha (1968) for SRI funds and conventional funds:

𝑅𝑖𝑡 − 𝑅𝑓 = 𝛼𝑖 + 𝛽𝑖 (𝑅𝑚 − 𝑅𝑓) + 𝜀𝑖 (7)

Where 𝑅𝑖𝑡 is the average monthly return off the funds in month t, 𝛽is the slope coefficient of the regression for a certain period (pre, at or post crisis) and also represents the systematic risk of the portfolio being exposed to the market return for each period, 𝑅𝑓 is the monthly average return of the US treasury bill, 𝑅𝑚 is the return on the market index and 𝜀𝑖 is the error term. The intercept of the model is given by 𝛼, also often referred to as Jensen’s alpha (1968) and provides an indication of abnormal return of the portfolio on a risk-adjusted basis. Alpha determines the out or underperformance, so it can be seen as the difference between the actual return and the expected return according to the CAPM model. So, the first part of the study will run the single-factor CAPM analysis. However, in previous literature, the use of this single-factor model has been repeatedly argued as insufficient and have failed to address the explanation of the cross-section of expected returns. Therefore, we also run regressions based on the four-factor model (Carhart, 1997). 3.2.3 Four factor model

The four-factor model has been tested in several studies to analyze fund

performance (Bauer et al., 2005; Fernando, Vargas & Marco, 2014). To control for size, value and momentum influences, the four factor model by Carhart is used (1997), which is based on the three-factor model introduced by Fama and French (1993), by adding a last factor that captures momentum anomaly (Jegadeesh and Titman, 1993):

𝑅𝑖 − 𝑅𝑓 = 𝛼𝑖 + 𝑏𝑖(𝑅𝑚 − 𝑅𝑓) + 𝑠𝑖𝑆𝑀𝐵 + ℎ𝑖𝐻𝑀𝐿 + 𝑚𝑖𝑀𝑂 + 𝜀𝑖 (8)

where 𝑏𝑖, 𝑠𝑖, ℎ𝑖 and 𝑚𝑖 are slope coefficients of the time-series regression, and SMB is the equally weighted average of (historical) excess returns of funds with a small market

capitalization over funds with a large market capitalization. HML represents the equally weighted average return of funds with the highest book-to-market ratios minus the returns of the funds with the lowest book-to-market ratio. Finally, MO represents the coefficient of the difference between high and low previous returns of the portfolios, where the previous

(12)

and for European market from Kenneth R. French’s online database. For the performance analysis of the SRI and conventional funds the Fama French US factors are utilized and for the analysis of the European SRI and non-SRI funds the Fama French factors for Europe. All regressions and tests are done in Stata with a p-value of 5%.

4. Results

4.1 Summary Statistics

TABLE I

Summary statistics on SRI funds versus general funds in the US during the identified break periods.

US Whole period pre-crisis crisis post crisis

SRI General SRI General SRI General SRI General Average return 0.70% 0.93% 0.89% 1.15% 0.04% 0.21% 1.00% 1.25% Standard deviation 4.68% 4.82% 2.66% 3.07% 7.28% 7.26% 3.37% 3.68% Number of funds 97 130 97 130 97 130 97 130

Chow test _{Single-factor CAPM model} _{Four factor model}

SRI General SRI General

F-value _797.7997* _2.4270* _313.0057* _3.5903*

This table reports summary statistics on SRI funds and general investment funds in the US from the selected sample. Mean return and standard deviation are calculated based on an equally weighted portfolio of all sampled funds. Mean return and corresponding standard deviation are presented on monthly basis. Sample period for whole period: 10-2003 - 05-2018. Pre-crisis: 10-2003 - 09-2007. Crisis: 10-2007 - 11-2011. Post crisis: 12-2011 - 05-2018. To estimate the F-values of the Chow-test, equation 6 is used.

*Statistically significant at 5% level.

TABLE II

Summary statistics on SRI funds versus general funds in Europe during the identified break periods.

Europe Whole period pre-crisis crisis post crisis

(13)

Average return 0.76% 0.62% 1.70% 1.13% -0.49% -0.25% 0.94% 0.83% Standard deviation 4.12% 3.27% 3.51% 2.40% 5.69% 4.60% 3.10% 2.62% Number of funds 155 87 155 87 155 87 155 87

Chow test _{Single-factor CAPM model} _{Four factor model}

SRI General SRI General

F-value _2.6647* _0.7189 _0.8317 _1.8157

This table reports summary statistics on SRI funds and general investment funds in Europe from the selected sample. Mean return and standard deviation are calculated based on an equally weighted portfolio of all sampled funds. Mean return and corresponding standard deviation are presented on monthly basis. Sample period for whole period: 10-2003 - 05-2018. Pre-crisis: 10-2003 - 09-2007. Crisis: 10-2007 - 11-2011. Post crisis: 12-2011 - 05-2018. To estimate the F-values of the Chow-test, equation 6 is used.

The dataset is divided into the three main time-series regions: pre, at and post-crisis for the regions US and Europe. Table I shows that the average returns of SRI funds and general funds in the US differ no more than 0.25 percentage points, comparing every specified period. Furthermore, the average returns of SRI funds as well for general funds was positive during the financial crisis, though respectively lower when compared to the other periods. Furthermore, table I shows that the volatility of the SRI portfolio and general portfolio do not differ much in each period, which implies that SRI is not riskier than general investing in the US. In Table II, the average returns of SRI funds and general funds in Europe do not differ much in each identified period as well, in particular a maximum of 0.57

percentage points. According to the average returns, SRI funds obtain an overall higher return than general funds. Comparing the standard deviations of European SRI portfolio and general portfolio in table II, it implies that SRI is riskier than general investing. Also, in Europe, a drop in the average returns shows during the financial crisis, resulting in average negative returns for both SRI and general funds.

When looking at the results of the Chow-tests in Table I and II, a structural break is identified in the US (enough significant evidence), but there is no or weak statistical

evidence (critical value is F=2.2698 for European SRI funds) for a structural break in Europe. This might be the case, for the financial crisis had a stronger effect on the monthly returns of investment funds in the US than in Europe. Another possibility is that the identified break periods for Europe should have been chosen differently, for the crisis might have affected the European market later than the US market.

(14)

TABLE III

Summary statistics on returns small cap SRI and conventional funds Small Cap SRI Funds

Whole period Pre-crisis Crisis Post crisis

Average return 0.64% 0.85% 0.00% 0.92%

Standard

deviation 4.99% 3.07% 7.63% 3.56%

Small Cap General Funds

Average return 1.00% 1.16% 0.08% 1.09%

Standard

deviation 5.46% 4.66% 6.84% 3.28%

This table reports summary statistics on small cap SRI funds and general investment funds in the US from the selected sample. Sample consists of 38 small cap SRI funds and 60 small cap general funds. Mean return and standard deviation are calculated based on an equally weighted portfolio of all sampled funds. Mean return and corresponding standard deviation are presented on monthly basis. Sample period for whole period: 10/2003 - 05/2018. Pre-crisis: 10/2003 - 09/2007. Crisis: 10/2007 - 11/2011. Post crisis: 12/011 - 05/2018.

TABLE IV

Summary statistics on returns large cap SRI and conventional funds Large Cap SRI Funds

Whole period Pre-crisis Crisis Post crisis

Average return _0.75% _0.92% _0.21% _1.21%

Standard deviation _4.45% _2.34% _6.54% _3.34%

Large Cap General Funds

Average return _0.85% _1.19% _0.43% _1.29%

Standard deviation _4.33% _3.68% _8.00% _4.25%

This table reports summary statistics on large cap SRI funds and general investment funds in the US from the selected sample. Sample consists of 59 large cap SRI funds and 70 large cap general funds. Mean return and standard deviation are calculated based on an equally weighted portfolio of all sampled funds. Mean return and corresponding standard deviation are presented on monthly basis. Sample period for whole period: 10/2003 - 05/2018. Pre-crisis: 10/2003 - 09/2007. Crisis: 10/2007 - 11/2011. Post crisis: 12/2011 - 05/2018.

(15)

When comparing the results from table I, III, and IV, it can be argued that the

average returns of small cap funds for SRI funds is lower and the average return of large cap SRI is higher than or equal to the average return of all SRI funds proposed in table I.

Moreover, this also holds for general funds, except when looking at the overall period. Also, it becomes clear from table I - IV than the volatility (standard deviation) of all portfolio’s rises in crisis period compared to the other periods, which generally in line with the findings in Table X - XVI, where the regressions show a relatively lower alpha (systematic risk).

Figure I

Normality approach of SRI and general funds

This figure shows a frequency table from the average monthly returns of SRI and general funds in the US and Europe. The dataset ranges from 1 October 2003 – 1 May 2018.

Figure II

Normality approach of large and small cap funds

This figure shows a frequency table from the average monthly returns of large cap and small cap SRI and general funds in the US. The dataset ranges from 1 October 2003 – 1 May 2018.

(16)

As can be seen in figure I, the frequency table of the monthly returns of SRI and general funds give an indication to look normally distributed, but not perfectly. This may be caused by a relatively small sample size (97 SRI funds in the US, 155 European SRI funds, 130 US general funds and 76 European general funds) and also the assumption that stocks prices are usually not normally distributed. Therefore, the results from the Wilcoxon signed rank-sum test are relevant to observe, for it takes in account a non-normally distributed sample. But also, the results from the t-test are useful, because the t-test assumes normality. This also holds for the comparison between performance of small and large cap funds in the US, as figure II also shows histograms tending to have a look of normally distribution.

4.2 Test results

TABLE V

Results from t-test for comparing differences in sample means for small and large cap funds t-value (1) - (2) (3) - (4) (1) - (3) (2) - (4)

(17)

Pre-crisis -0.1207 -0.0454 -0.3808 -0.4413

Crisis -0.1463 -0.2430 -0.0519 -0.1566

Post crisis -0.5338 -0.3312 -0.3217 -0.1350

This table reports results from two sample t-test with unequal variances. To estimate the t-values equation 1 is used. Where,

(1) - (2) = Average returns of small cap SRI funds minus average returns of large cap SRI funds in the US (3) - (4) = Average returns of small cap general funds minus average returns of large cap general funds in the US (1) - (3) = Average returns of small cap SRI funds minus average returns of small cap general funds in the US (2) - (4) = Average returns of large cap SRI funds minus average returns of large cap general funds in the US

Sample period for whole period: 10/2003 - 05/2018. Pre-crisis: 10/2003 - 09/2007. Crisis: 10/2007 - 11/2011. Post crisis: 12/2011 - 05/2018.

TABLE VI

Results from Wilcoxon signed rank-sum tests for comparing differences in sample means for small and large cap funds

p-value (1) vs (2) (3) vs (4) (1) vs (3) (2) vs (4)

Overall period 0.2457 0.3589 0.9552 0.6269

Pre-crisis 0.7273 0.4357 0.7897 0.8777

Crisis 0.7684 0.5527 0.5657 0.3823

Post crisis 0.2388 0.7746 0.8091 0.9781

This table reports results from the Wilcoxon signed rank-sum test. To estimate the p-values, a Stata command for this test is done. In this table,

(1) - (2) = Average returns of small cap SRI funds versus average returns of large cap SRI funds in the US (3) - (4) = Average returns of small cap general funds versus average returns of large cap general funds in the US (1) - (3) = Average returns of small cap SRI funds versus average returns of small cap general funds in the US (2) - (4) = Average returns of large cap SRI funds versus average returns of large cap general funds in the US

In table V, it shows that no t-value is statistically significant at 5% level, for the critical value for the t-statistic at 5% level is ±1.96. This implies that there is no significant difference in the average return values between small and large cap SRI funds, small and large cap general funds, small cap SRI funds and small cap general funds and large cap SRI funds and large cap SRI funds during the identified break periods. Table VI supports this assumption, for there are no significant differences found by the use of the Wilcoxon signed rank-sum test.

(18)

TABLE VII

Wilcoxon signed rank-sum test for comparing differences in sample means for US market and European market p-value (1) vs (2) (3) vs (4) (1) vs (3) (2) vs (4) Overall period 0.7676 0.0484* 0.7609 0.8454 Pre-crisis 0.8455 0.3099 0.0383* 0.9509 Crisis 0.5673 0.2047 0.5211 0.8462 Post crisis 0.8950 0.0643 0.6811 0.4386

This table reports results from the Wilcoxon signed rank-sum test. To estimate the p-values, a Stata command for this test is done. In this table,

(1) - (2) = Average returns of SRI funds in the US versus average returns of general funds in the US (3) - (4) = Average returns of SRI funds in Europe versus average returns general funds in Europe (1) - (3) = Average returns of SRI funds in the US versus average returns of SRI funds in Europe (2) - (4) = Average returns of general funds in the US versus average returns general funds in Europe

In Table VII, there is an indication that SRI funds overall generate a significant different return than general funds in Europe. From this table, it can only be assumed that there is a difference, but not if this difference contains a higher monthly return or lower monthly return. Therefore, also a one-sided Wilcoxon rank-sum test is executed and gives a statistical significant support at 5% level that the average monthly return of SRI funds is higher than the average monthly returns of general funds in Europe (Found p-value = 0.0242). Furthermore, also a one-sided test is executed for the comparison between SRI funds in the US and Europe during the pre-crisis period and gives statistical significant evidence for lower returns of SRI funds in the US compares to SRI funds in Europe (Found p-value= 0.0191). Both significant findings are in line with the descriptives in table I and II.

TABLE VIII

t-statistics for comparing differences in sample means for US market and European market t-value (1) - (2) (3) - (4) (1) - (3) (2) - (4)

Overall period -0.4530 -0.1545 -0.1325 -0.4604

Pre-crisis -0.4453 1.0244 -1.3484 0.1136

Crisis -0.1898 -0.2567 0.3272 0.3603

(19)

This table reports results from two sample t-test with unequal variances. To estimate the t-values equation 1 is used. Where,

(1) - (2) = Average returns of SRI funds in the US minus average returns of general funds in the US (3) - (4) = Average returns of SRI funds in Europe minus average returns general funds in Europe (1) - (3) = Average returns of SRI funds in the US minus average returns of SRI funds in Europe (2) - (4) = Average returns of general funds in the US minus average returns general funds in Europe

In table VIII, no resulting t-values are significant at 5% level. This indicates that the average returns between general US funds and US SRI funds, US SRI funds and European SRI funds, US general funds and European general funds and European SRI funds and European general funds do not differ significantly during all identified periods. A possible explanation for the insignificant results, is that the European and US financial markets are highly

integrated.

4.3 Regression results

TABLE IX

Single-factor CAPM analysis for monthly return data October 2003 - May 2018 (Overall period) US SRI Portfolio US General

Portfolio Euro SRI Portfolio Euro General Portfolio

Alpha -.0010*

(.0001) 0.0074 (0.0045) .0053 (.0032) .0043 ( .0025)

Beta 1.0028*

(.0020) 0.1132 (0.1696) .1489 (.07965) .1103 (.0638)

R2 _0.9988 _0.0029 _0.0353 _0.0306

This table reports results from the single-factor CAPM regressions. To measure the performance of SRI funds and general funds, the estimated model from equation 7 is used. The market proxy used for the regression, is the value-weighted portfolio retrieved from the Wharton Research database. Sample period: 10/2003-05/2018.

*Coefficient is statistically significant at 5% level.

Table IX provides the regression results of the single-factor CAPM model (equation 7) for SRI and general funds in the US and Europe, where the excess market return is the only independent variable in the model. The standard errors in all regressions executed have been corrected for heteroskedasticity and autocorrelation, by using the Newey-West

(20)

monthly return of a SRI portfolio in the US is 0.09% lower risk-adjusted than the market index. When comparing this with the alpha from Table X, where the four-factor CAPM is used to control for size, book-to-market ratio and momentum, an almost identical alpha is observed. This confirms the suggestion of a monthly underperformance of 0.09% (also significant). When comparing the alpha of the SRI portfolio in the US with Europe based on the single-factor CAPM model, the European portfolio has a slightly overperformance of 0.5% compared to the European market index. A non-significance for the other portfolios implies that US general portfolio, European SRI portfolio and a European general portfolio do not over or underperform compared to the market index and A possible explanation for the non-significance of the other portfolios could be the low r-squared values, which insinuates a small explanation of the portfolio’s return by the coefficients alpha and beta.

TABLE X

Four-factor CAPM analysis for monthly return data October 2003 - May 2018 (Overall period)

Alpha Beta SMB HML MO R2 US SRI portfolio -.0010* (.0001) 1.0027* (.0027) .0022 (.0046) -.0059 (.0042) -.0024 (.0024) 0.9988 US General portfolio .0078 (.0042) .0723 (.1561) .01383 ( .1398) .0063 (.1277) -.2999* (.0856) 0.0819 Euro SRI portfolio .0038 (.0033) .1410 (.0780) .8074 (.1829) .0774 (.1752) .0080 (.0910) 0.1586 Euro General portfolio .0035 (.0027) .0908 (.0611) .6154* (.1462) .0760 (.1464) -.0300 (.0791) 0.1464

This table reports empirical results corresponding to the four-factor model regression formulated by equation 8. The market proxy used for the regression, is the value-weighted portfolio retrieved from the Wharton Research database. In this table, SMB represents the difference in monthly return between a small cap portfolio and a large cap portfolio, HML represents the monthly book-to-market factor and MO is the difference in return between a prior 12-month winner and loser portfolio. R2 _{indicates how much of the}

theoretical excess returns is explained by the factors alpha, beta, SMB, HML and MO. Sample period:10/2003-05/2018

Table X provides the regression results for the four-factor model. Only the alpha and beta of the US SRI portfolio in Table VII are statistically significant, which indicates that a SRI portfolio in the US slightly underperforms according to the market index by 0.09% and a significant non-zero beta indicates that a SRI portfolio in the US is more volatile than the market during the overall period. These findings are in line with the results from El Ghoul’s and Karoui’s (2017) study, for they found an underperformance of high CSR funds compared

(21)

to low CSR funds. The statistical significant difference between general and SRI funds in the US contradicts with the findings of Bauer et al. (2005), Kempf and Osthoff (2007) and Galema et al. (2008). They found that SRI funds performed no worse, or even better than general funds in the US during the period prior to the financial crisis. The momentum-coefficient states a significant difference for the US general portfolio, which suggests a link may exist between the difference of the highest and lowest past 12-month returns on the average excess returns of the portfolio.

TABLE XI

Single-factor CAPM analysis for monthly return data October 2003 - September 2007 (Pre-crisis) SRI Portfolio General

Portfolio Euro SRI Portfolio Euro General portfolio

Alpha -.0026*

(.0002) .0089 (.0049) .01252* (.0058) .0075 (.0043)

Beta 1.0047*

(.0084) -.01654 (.1560) .0965 (.1337) .0624 (.1132)

R2 _0.9966 _-0.0211 _0.1007 _0.0059

TABLE XII

Four-factor CAPM analysis for monthly return data October 2003 - September 2007 (Pre-crisis)

Alpha Beta SMB HML MO R2 US SRI portfolio -.0027* (.0003) 1.0003* (.0094) .0126 (.0091) .0012 (.0163) .0013 (.0091) 0.9968 US General portfolio .0084 (.0059) -.0811 (.1941) .2150 (.1502) -.2570 (.2901) .4705* (.1767) 0.1996 Euro SRI portfolio .0151* (.0062) .2152 (.1616) 1.1800* (.3159) -.3835 (.6320) -.4672 (.3272) 0.2386 Euro General portfolio .0095* (.0043) .1696 (.1327) .6886* (.2285) -.13068 (.4126) -.4256 (.2417) 0.1174

(22)

Wharton Research database. In this table, SMB represents the difference in monthly return between a small cap portfolio and a large cap portfolio, HML represents the monthly book-to-market factor and MO is the difference in return between a prior 12-month winner and loser portfolio. R2 _{indicates how much of the}

theoretical excess returns is explained by the factors alpha, beta, SMB, HML and MO. Sample period: 10/2003-09/2007

When analyzing and comparing the alphas from Table XI and XII, a significant

overperformance of SRI and general funds in Europe and a significant underperformance of SRI in the US might be indicated considering the single-factor and the four-factor model in the pre-crisis period. These results are in line with the findings by Climent and Soriano (2011), who found that US SRI funds underperformed conventional funds in the US during the period 1987-2009. Furthermore, in Table XII, the Small-Minus-Big coefficient for SRI and general funds in Europe is statistically significant at 5% level. This might indicate that there is an apparent size-effect on the average returns of funds in Europe in the pre-crisis period, which is in line with Gregory’s, Matatko’s and Luther’s study (1997).

TABLE XIII

Single-factor CAPM analysis for monthly return data October 2007 - November 2011 (At crisis) SRI Portfolio General

Alpha -.0006*

(.0001) .0014 (.0102) -.0043 (.0080) -.0022 (.0065)

Beta 1.0028*

(.0023) .2653 (.2266) .1735 (.1105) .1251 (.0902)

R2 _0.9997 _0.0285 _0.0582 _0.0454

TABLE XIV

Four-factor CAPM analysis for monthly return data October 2007 - November 2011 (At crisis)

Alpha Beta SMB HML MO R2

US SRI

(23)

US General portfolio .0020 (.0100) .2065 (.2158) -.3234 (.3833) .1992 (.2408) -.3669* (.1207) 0.1821 Euro SRI portfolio -.0027 (.0074) .1423 (.1287) .8851 (.3616) .2074 (.4052) .02476 (.1405) 0.1799 Euro General portfolio -.0006 (.0059) .0906 (.1012) .7619* (.3010) .1973 (.3263) .01201 (.1227) 0.1824

theoretical excess returns is explained by the factors alpha, beta, SMB, HML and MO. Sample period: 10/2007-11/2011.

In table XIII and XIV, the resulting values of the Alpha’s and Beta’s do not differ extremely when compared both single and four-factor model. Furthermore, the alphas of the different portfolio’s (except for the US general funds) show a underperformance compared to the market during crisis period. Consistent with the findings of Varma and Nofsinger (2012), SRI funds in Europe outperform general funds during the crisis, because they underperform less compared to the market index . Therefore, the findings from Table XIV for the US are not consistent with the findings of Varma and Nofsinger (2012). For the SRI funds and the general European funds, a positive size-effect could be assumed (only European general portfolio is significant). The momentum-effect seems to have a negative influence on the excess returns of funds in the US and a positive influence on the excess returns of funds in Europe.

TABLE XV

Single-factor CAPM analysis for monthly return data December 2011 - May 2018 (Post crisis) SRI Portfolio General

Alpha -.0002*

(.0000) .0159* (.0049) .0096* (.0034) .0078* (.0032)

Beta 1.0005*

(.0013) -.3045* (.1372) -.0572 (.0891) .0346 (.0889)

(24)

is the value-weighted portfolio retrieved from the Wharton Research database. Sample period: 12/2011-05/2018.

TABLE XVI

Four-factor CAPM analysis for monthly return data December 2011 - May 2018 (Post crisis) Alpha Beta SMB HML MO R2 US SRI portfolio -.0002* (.0000) 1.0001* (.0014) -.0002 (.0018) .0011 (.0021) -.0003 (.0011) 0.9999 US General portfolio .0160* (.0048) -.2750 (.1365) -.0373 (.1561) -.0965 (.1492) -.3510* (.1365) 0.1450 Euro SRI portfolio .0063 (.0041) .0305 (.1033) .2404 (.2615) -.0252 (.1717) .2047 (.1398) 0.0479 Euro General portfolio .0062 (.0035) .0495 (.0827) .4306* (.1880) .0522 (.1266) -.0058 (.1106) 0.0726

theoretical excess returns is explained by the factors alpha, beta, SMB, HML and MO. Sample period:12/2011-05/2018.

Consistent with the findings in previous periods, the resulting values of the Alpha’s and Beta’s do not differ extremely when compared both single and four-factor model in Table XV and XVI. Furthermore, the alphas of the different portfolio’s (except for the US SRI funds) show a overperformance compared to the market during post crisis period. Also consistent with the findings in previous periods, the SRI funds and the general European funds, a positive size-effect could be assumed (only European general portfolio is

significant). Consistent with Bauer et al. (2005) findings, the alpha’s show no statistically significant difference in performance between general funds and SRI funds.

The overall findings of the results in Table IX - XVI indicate that the four-factor model is a better model to approach the returns of investment funds, due to a higher r-squared value for the explanatory independent variables in the four-factor model. This is consistent with the quoted literature (refer). The alphas of the US SRI portfolio imply an

(25)

portfolios in general overperform the market index (but there is only a significant difference during the post crisis period). The alpha values of SRI funds in Europe indicate an

overperformance to the market index, except for the crisis period. Moskowitz (2000) and Kosowski (2006) state in their studies that that active strategies cause an outperformance during times of recession. This argument is not consistent with the results from Table XIII and XIV, except for the performance of general funds in the US.

As it happened in all periods, the value style (HML) seems to be irrelevant in this model, for there is no significant result for the HML coefficient in every period. The

momentum factor seems to have an overall statistically significant negative influence on the returns of general investment funds in the US, which indicates that the past year

performance negatively influences the portfolio’s excess return.

It is also noticeable that the alphas of funds in Europe is comparable, but not in the US. This might indicate that SRI funds in Europe can make up for the costs incurred by screening and monitoring. In line with Bello (2005), this signal might also indicate that the practical costs are small when associated with limited diversification, which should be the case for SRI funds. A possible explanation for the underperformance of the SRI funds in the US could be an underlying risk factor or SRI funds in the US are less efficient in screening. The indicated underperformance of the SRI funds in the US is in line with standard portfolio choice theory, which suggest that the constraint (in this case the SRI screens) would indicate lower performance.

When comparing the beta’s or systematic risk of SRI funds and general funds, SRI funds tend to have a bigger market exposure compared to general funds. This is in line with Rudd (1981), Hickman et al. (1999) and Tippet (2001), who argued that SRI policies might increase risk. During the crisis and afterwards the general funds US portfolio shows a negative beta, suggesting a negative relation to the market index (but no statistical

significant evidence). The statistically significant beta of the SRI portfolio is almost equal to 1 during all periods, which implies that it follows the market index. Furthermore, the betas of the SRI portfolio and general portfolio in Europe suggest a lower volatility than the market, although not statistically significant.

5. Conclusion

The primary objective of this study is to provide insight on the performance of SRI funds during the financial crisis of 2007-2008. The performances of SRI funds and general investment funds in the US and Europe were examined. To determine if there was a structural break during the financial crisis, a Chow test was used. Furthermore, a Wilcoxon rank-sum test and t-test were executed to check whether there is enough evidence to distinguish a difference in performance between SRI and general funds, the US market and the European market and small and large cap funds. To test whether a SRI portfolio out- or

(26)

The findings from this study imply that the financial crisis indicates a structural break in the US, but not in Europe, for the crisis might have affected funds in the US more, or the crisis hit the European market at different time point and had a different duration of effect. Furthermore, when looking at the results of the t-test and Wilcoxon signed rank-sum test, there is no statistical evidence that returns between large cap and small cap SRI and general funds differ. The Wilcoxon signed rank-sum test provides evidence that European SRI funds might have a significant higher return that general funds. To continue, this study support evidence that SRI funds in the US underperform general investment funds. This contradicts with the prior literature that SRI performance equals general investment, although not all alpha values found in this research were statistically significant. When looking at the differences in average returns of large and small cap firms in the US, no statistical evidence can be deducted for a difference in returns. Although in Europe, there a significant size-effect is found for general investment funds. For the European region, there is no statistical evidence that SRI funds under- or outperform general investment funds.

Based on the systematic risk of the portfolio’s, it might be concluded that neither SRI funds, nor general funds both in the US and in Europe are more volatile than the market index. This is in line with former studies on SRI performance compared with indices.

Systematic risk of European SRI funds is observed as lower than systematic risk of SRI funds in the US. Overall, investing in European SRI funds would be more beneficial when

comparing risks and returns compared to the US SRI funds. European SRI funds tend to have no more or less market exposure according to this research, but no statistical evidence has been found.

To summarize, the results of this research show that assumption one does not hold, for the observed beta values of SRI funds were relatively higher than the betas of the general funds, as well in Europe and the US. Furthermore, assumption two is violated in the US market, for there is found a significant underperformance (lower alpha) of SRI funds. On the other hand, this assumption holds in the European market, for there is not found any significant difference between SRI and general fund performance. It does even imply, without statistical evidence, that SRI funds outperform general funds in Europe, which is in line with previous literature. To continue, assumption three holds, for the t-tests and

Wilcoxon signed rank-sum test have shown no significant difference between the returns of small and large cap SRI and general firms. Furthermore, the last assumption might be violated, for the Wilcoxon signed rank-sum test gave significant evidence for

outperformance of European SRI funds compared to SRI funds in the US. However, the t-test did not support this finding. Overall, the conclusion can be made that SRI funds in the US performed worse than SRI funds in Europe during the financial crisis and the returns of small and large cap funds are assumed to be equal.

(27)

6. Discussion

For further research, a more elaborated study should be done on the moment the financial crisis started influencing other markets, in the case of this study; the European market. Also, the duration of the effects of the crisis in separate markets should be investigated more profoundly. Then, the measured effects of the crisis would be more accurate. Furthermore, a more elaborated model could be used to test the performance of investment funds during the financial crisis, for the four-factor model and CAPM-model used in this study generated relatively low r-squared measures. Other factors than size effect, value-effect and momentum effect should be added to the model to test if there are more factors that contribute to the excess returns of investment funds. In addition, to measure performance of SRI funds more properly, a distinction should be made between performance measurement of environmental, socially and ethically focused investment funds. In that case, an out- or underperformance could not only be compared to general funds or the market index, but also among SRI funds themselves. Finally, the performance of small and large cap firms can be measured more thoroughly, so more comparisons for risk and performance can be done.

(28)

Reference List

Amenc, N. & Le Sourd, V. (2008). Socially responsible investment performance in France. Retrieved from the EDHEC Risk and Asset Management Research Centre

Areal, N., Cortez, M. C., & Silva, F. (2010). Investing in Mutual Funds: Does it Pay to be a Sinner or a Saint in Times of Crisis? University of Minh, p. 1-22

Azar, S.A., & Al Hourani, M.A. (2010). The Performance of U.S. Equity Mutual Funds. Journal of Money, Investment and Banking, Vol. 18 No. , p.

Bauer, R., Derwall, J., Guenster, N., & Koedijk, K. (2011). The Economic Value of Corporate Eco-Efficiency. European Financial Management, Vol. 17 No. 4, p. 679-704

Bauer, R., Derwall, J., & Otten, R. (2007). The ethical mutual fund performance debate: New evidence from Canada. Journal of Business Ethics, Vol. 70 No. 2, p. 111-124.

Bauer, R., Koedijk, K., & Otten, R. (2005). International evidence on ethical mutual fund performance and investment style. Journal of Banking & Finance, Vol. 29 No. 7, p. 1751-1767.

Bauer, R., Otten, R., & Rad, A. T. (2006). Ethical investing in Australia: Is there a financial penalty?. Pacific-Basin Finance Journal, Vol. 14 No. 1, p. 33-48

Bauman, S. W., Conover, M. C., & Miller, R. E. (1998). Growth Versus Value and Large-Cap versus Small-cap stocks in International markets. Financial Analysts Journal, Vol. 54 No. 2, p.75-89

Bello, Z. Y. (2005). Socially responsible investing and portfolio diversification. Journal of Financial Research, Vol. 28 No. 1, p. 41-57.

Budde, S. (2008). Compelling returns: a practical guide to socially responsible investing. Retrieved from wiley.com

Carhart, M. (1997). On persistence in mutual fund performance. The Journal of Finance, Vol. 52 No. 5, p. 1276-1302.

Chow, G. C. (1960). Tests of Equality Between Sets of Coefficients in Two Linear Regressions. Econometrica, Vol. 28, No. 3, p. 591-605.

Climent, F., Soriano, P. (2011). Green and Good? The Investment Performance of US Environmental Mutual Funds. Journal of Business Ethics, Vol. 103 No. 2, p. 275-287

(29)

Cowton, C. (1999). Accounting and Financial Ethics: From Margin to Mainstream. Business Ethics: A European View, Vol. 8 No. 2, p. 99-107.

Cummings, L. (2000). The Financial Performance at Ethical Investment Trusts: An Australian Perspective. Journal of Business Ethics, Vol. 25, p. 79-92.

Etzioni, A. (1998). The Moral Dimension: Towards a New Economics. The Free Press, New York.

Fama, E. and French K. (1993). Common risk factors in the returns on stocks and bonds. The journal of financial economics, Vol. 33, p. 3-56.

Fehr, E., Gachter, S.(2000). Fairness and retaliation: the economics of reciprocity. Journal of Economic Perspectives. Vol. 14, p.159–181.

Fernandez, A. & Matallin, J. C. (2008). Performance of Ethical Mutual in Spain: Sacrifice or Premium?. Journal of Business Ethics, Vol. 81, p. 247–260.

Fernando, M., Marco, I., & Vargas, M. (2014). Environmental Mutual Funds: Financial

Performance and Managerial Abilities. Journal of Business Ethics, Vol. 124 p. 551-569 Galema, R. Plantinga, A. Scholtens, B. (2008) The stocks at stake: Return and risk in socially

responsible investment. Journal of Banking & Finance, Vol. 32 No. 12, p. 2646-2654. Geczy, C., Stambaugh, R., & Levin, D. (2005). Investing in socially responsible mutual funds.

University of Pennsylvania – The Warton School, SSRN, Financial department. p.1-53. Gerogry, A., Matatko, J., & Luther, R. (1997). Ethical unit trust financial performance: small

company effects and fund size effects. Journal of Business Finance & Accounting, Vol. 24 No. 5, p. 705-725

Gregory, A., & Whittaker, J. (2007). Performance and performance persistence of ‘ethical’ unit trusts in the UK. Journal of Business Finance & Accounting, Vol. 34 No. 7-8, p. 1327-1344.

Gil-Bazo, J., Ruiz-Verdu, P., & Santos, A. P. (2010). The Performance of Socially Responsible Mutual Funds: The Role of Fees and Management Companies. Journal of Business Ethics, Vol. 94 No. 2, p. 243–263.

Guenster, N., Bauer, R., Derwall, J., & Koedijk, K. (2011). The Economic Value of Corporate Eco- Efficiency. European Financial Management, Vol. 17 No. 4, p. 679-704. Grigoris, G., & Theotokas, I. (2011). The Effect of Financial Crisis in Corporate Social

Responsibility Performance. International Journal of Marketing Studies, Vol. 3 No. 1, p. 2-10

(30)

Heal, G. (2005). Corporate social responsibility: An economic and financial framework. The Geneva papers on risk and insurance-Issues and practice, Vol. 30 No. 3, p. 387-409. Hickman, K. A., Teets, W. R., & Kohls, J. J. (1999). Social Investing and Modern Portfolio

Theory. American Business Review, Vol. 17 No. 1, p. 72-78

Kempf, A., & Osthoff, P. (2007). The Effect of Socially Responsible Investing on Portfolio Performance. European Financial Management, Vol. 13 No. 5, p. 908–922.

King, A. A., & Lenox, M. J. (2001) Does It Really Pay to Be Green? An Empirical Study of Firm Environmental and Financial Performance: An Empirical Study of Firm Environmental and Financial Performance. Journal of Industrial Ecology, Vol. 5 No. 1, p. 105-116. Kosowski, R. (2011). Do mutual funds perform when it matters most to investors? US mutual

fund performance and risk in recessions and expansions. The Quarterly Journal of Finance, Vol. 03 No. 1, p. 607-664

Kreander, N., Gray, R. H., Power, D. M., & Sinclair, C. D. (2005). Evaluating the performance of ethical and non-ethical funds: a matched pair analysis. Journal of Business Finance & Accounting, Vol. 32 No. 7-8, p. 1465-1493.

Lewis, V., Kay, K., D., Kelso, C., & Larson, J. (2010). Was the 2008 financial crisis caused by a lack of corporate ethics?. Global Journal of Business Research, Vol. 4 No. 2, p. 77-84. Luther, R. J., Matatko, J., & Corner, C. (1992). The Investment Performance of U.K. Ethical

Unit Trusts. Accounting, Auditing and Accountability Journal, Vol. 5 No. 1, p. 57–71. McWilliams, A., Siegel, D. (2000). Corporate Social Responsibility and Financial Performance:

Correlation or Misspecification?. Strategic Management Journal, Vol. 21 No. 5, p. 603-609

Michelson, G., Wailes, N., Van Der Laan, S., & Frost, G. (2004). Ethical investment processes and outcomes. Journal of Business Ethics, Vol. 52 No.1, p. 1-10.

Moskowitz, T. J. (2000). Discussion: Mutual fund performance: An empirical decomposition into stock-picking talent, style, transactions costs, and expenses. The Journal of Finance, Vol. 55 No. 4, p. 1695-1703.

Renneboog, L., J. ter Horst, & C. Zhang. (2008). The price of ethics and stakeholder governance: the performance of socially responsible mutual funds. Journal of Corporate Finance, Vol. 14 No. 3, p. 302-328.

Rockness, J. Williams, P. F. (1998). A Descriptive Study of Social Responsibility Mutual Funds. Accounting, Organizations and Society, Vol. 13 No. 4, p. 397-411.

(31)

Review, Vol. 23 No. 4, p. 55-61

Schröder, M. (2006). Is there a difference? The performance characteristics of SRI equity indices. Journal of Business & Financial Accounting, Vol. 1/2 No. 34, p. 331-348. Schueth, S. (2003). Socially Responsible Investing in the United States. Journal of Business

Ethics, Vol. 43 No. 3, p. 189-194.

SIF: 2007, Social Investment Forum Trends Report (Social Investment Forum, Washington). Snedecor, G. W. & Cochran, W. G. (1989). Statistical Methods. State University Press, Iowa. Statman, M. (2000). Socially responsible mutual funds. Financial Analysts Journal, Vol. 56 No. 3, p. 30-39.

The Forum for Sustainable and Responsible Investment. 2016. Report on Sustainable and R Responsible Investing Trends. Available online:

http://www.ussif.org/content.asp?contentid=82 (accessed on 1 May 2018).

Tippet, J. (2001). Performance of Australia’s Ethical Funds. Australian Economic Review, Vol. 34 No. 2, p. 170-178

Tripathi, V., Bhandari, V. (2015a). Socially Responsible Stocks: A Boon For Investors in India. Journal of advances in management research (Forthcoming), Vol. 12. No. 2. Tripathi, V., Bhandari, V. (2016). Performance of Socially Responsible Stock Portfolios – The

Impact of Global Financial Crisis. Journal of Economics and Business Research, Vol. 12 No. 1, p. 42-68

Wimmer, M. (2012). ESG-persistence in Socially Responsible Mutual Funds. Available at SSRN 2138882. p. 9-15

Yelkikalan, N., & Can, K. Ö. S. E. (2012). The effects of the financial crisis on corporate social responsibility. International Journal of Business and Social Science, Vol.3 No.3, p. 292-300.

(32)

Appendix

Table I

Summary Statistics - SRI Funds in the US

This table shows the summary statistics for the equity SRI funds in the US, as identified by The Forum of Sustainable and Responsible Investment (USSIF) website and is retrieved through DataStream. The USSIF

provides an overview of equity funds and makes a distinction between small and large cap funds. The dataset ranges from 1 October 2003 until 1 May 2018. This table provides the average monthly return, the

standard deviation, minimum, maximum and number of observed months.

Fund Name Obs Mean Std. Dev. Min Max

LARGE CAP 176 0.00688 6 0.04811 8 -0.16187 0.14930 9 American Trust Allegiance 29 0.01117 0.03428

7

-0.08863

0.05714 3

Aspiration Redwood Fund 72 0.00934

8 0.03243 4 -0.09138 0.08793 4 Boston Common Large Cap Core Equity

Fund 8 -0.00282 0.00582 6 -0.01147 0.00270 6 Brown Advisory Sustainable Growth Fund 176 0.00771

3

0.04384

-0.16585

0.16338 2

Calvert Equity Fund A 176 0.00708

3 0.04379 4 -0.16632 0.16221 3

Calvert Equity Fund C 176 0.00812

9 0.04386 9 -0.16588 0.16406 7

Calvert Equity Fund I 176 0.00736

2

0.04674

-0.18538

0.17354 7 Calvert US Large Cap Core Responsible

Index Fund A 176 0.00662 2 0.04671 -0.18555 0.17317 7

(33)

Calvert US Large Cap Core Responsible Index Fund C 176 0.00774 4 0.04675 4 -0.18469 0.17295 5 Calvert US Large Cap Core Responsible

Index Fund I 176 0.00200 5 0.03992 2 -0.17596 0.14720 1 Calvert US Large Cap Growth Responsible

Index Fund C 176 0.00269 6 0.03995 9 -0.17495 0.14902

Calvert US Large Cap Growth Responsible Index Fund I 34 0.00614 6 0.03452 3 -0.07684 0.08693 2 Calvert US Large Cap Value Responsible

Index Fund A 34 0.00642 4 0.03450 4 -0.07633 0.08683 5 Calvert US Large Cap Value Responsible

Index Fund C 176 0.00500 8 0.04391 2 -0.18147 0.16104 4 Calvert US Large Cap Value Responsible

Index Fund I 176 0.00515 9 0.04432 9 -0.18641 0.17111 3

Calvert Social Index A 176 0.00436

9 0.04387 5 -0.18209 0.15959 7

Calvert Social Index C 155 0.00965

6 0.06894 5 -0.26433 0.26666 7 DFA Emerging Markets Social Core Equity 150 7.37E-06 0.04806

7

-0.23148

0.16901 7 Domini Social Equity Fund A 173 -2.6E-05 0.00249

9

-0.02369

0.01315 8

Domini Social Investor 176 0.00291

2 0.03875 9 -0.15601 0.16109 3 Dreyfus Premier Third Century B 97 0.00918

1 0.04064 -0.1061 0.10974 1 Epiphany FFV Fund A 117 0.01279 4 0.06055 -0.14988 0.18944 4

Performance of SRI funds during the financial crisis : small cap vs large cap and US vs Europe