Socially Responsible Investing Fund Performance: Evaluation and Comparison to Conventional Funds in Luxembourg

(1)

1 Socially Responsible Investing Fund

Performance:

Evaluation and Comparison to Conventional Funds

in Luxembourg

Abstract

Prompted by the importance of socially responsible investing (SRI), SRI has gained a lot of attention from scholars. However, previous literature concluded rather conflicting results as some researchers determined either under-, over- or similar performance of SRI funds compared to conventional funds (non-SRI funds). In this paper the literature is extended by analyzing the financial performance of SRI equity mutual funds compared to the market and to non-SRI equity mutual funds in Luxembourg. For this exploration, firstly the Sharpe and Treynor ratios were calculated, and secondly, a variety of models were estimated in order to compute the abnormal returns. This was tested on a total of 60 randomly selected SRI and non-SRI funds with a high versus low score according to the MSCI ESG fund rating. The results indicated that SRI funds performed similarly compared to the market. Furthermore, the results demonstrated no significant difference in financial performance between SRI mutual funds and non-SRI mutual funds. In conclusion, this paper provides evidence to suggest SRI and non-SRI mutual funds are converging.

Bruno Rein – 11716797

University of Amsterdam – Amsterdam Business School

BSc Business Administration – Specialization Finance

Supervisor – Drs. P.V. Trietsch, M.Phil

(2)

2

Statement of Originality

This document is written by Bruno Rein who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

3

4 1. Introduction

Socially responsible investing (SRI) has increased exponentially in the world during the last fifteen years. Funds that specialized in socially responsible opportunities increased both in assets under management (AuM) and in numbers all over the world and particularly in Europe. According to the Association of the Luxembourg Fund Industry, the European SRI fund market doubled from 2010 to 2016 and reached 476 billion Euros in AuM; with Luxembourg being the number one domicile for European SRI funds, accounting for 31% of funds and 35% of AuM.

In this paper, the literature is extended by further investigating the SRI fund market in Luxembourg due to its significance for the European SRI fund market. In addition, this research aims to evaluate whether the performance of SRI funds and conventional funds (non-SRI funds) diverge. The research question of this paper is: Do SRI equity mutual funds differ in performance from non-SRI equity

mutual funds in Luxembourg?

Previous research has demonstrated rather conflicting results when evaluating the performance of SRI funds and comparing it to non-SRI funds. Most research concludes that there is no significant difference in performance between SRI funds and non-SRI funds (e.g., Geczy et al., 2006; Bello, 2005; Statman, 2000). However, there is also sufficient research that documents under- or overperformance between SRI funds and non-SRI funds (e.g., Renneboog et al., 2008; Gil-Bazo et al., 2010). Hence, previous literature provides a rather inconclusive result, especially for Luxembourg.

As previously mentioned, the fund market in Luxembourg is an important one for Europe. Luxembourg is one of the most financially developed fund countries in the world and the second largest fund centre in the world in terms of market capitalization after the United States. According to the Foundation De Luxembourg, Luxembourg is eager to expand and support philanthropy throughout the fund market. For example, the Luxembourg Stock Exchange became the first-ever exchange to set up a separate platform for SRI funds. Accordingly, the financial market in Luxembourg is a significant one in terms of funds and SRI and therefore relevant to thoroughly research this market.

Besides its importance, the Luxembourg market may produce different results when comparing the performance of SRI to non-SRI funds. Firstly, due to Luxembourg’s separate platform for SRI funds it may make SRI investments increase in terms of liquidity, transparency, and visibility which would make these investments more accurately priced (something that SRI has always lacked). Secondly, due to Luxembourg’s favorable tax rate (the lowest in the EU) and advanced fund law makes it cheaper and easier to establish as fund and continue its operations. Lastly, Luxembourg’s fund law allows funds to be easily and cheaply distributed throughout the world. In conclusion, all the above would make SRI and/or non-SRI funds differ in performance compared to other countries.

The central research question is divided into four sub-questions that together will answer the central question. These are:

(5)

5

1. How do SRI funds differ from non-SRI funds?

2. How can SRI and non-SRI funds performance be measured? 3. What is the performance of SRI funds compared to the market? 4. What is the performance of SRI funds compared to non-SRI funds?

In order to answer the central research question 30 randomly selected SRI equity mutual funds are comparedto 30 randomly selected non-SRI mutual funds domiciled in Luxembourg. The funds differ in terms of their MSCI ESG fund rating such that SRI funds have a high score and non-SRI funds have a low score. Performance is evaluated using monthly returns and for the period January 2012 to December 2019. The risk-adjusted performance of these funds is derived by analyzing their Sharpe and Treynor ratios. Furthermore, their Jensen alpha is assessed in a single-factor model (CAPM) and a multifactor model (Carhart four-factor model). This is investigated by running a Newey-West time-series regression on the SRI and non-SRI funds and on their difference.

For evaluating the central research question and the sub-questions, the paper is divided into five chapters. Succeeding the introduction, the second chapter provides an overview of related literature. The third chapter describes the methodology and data. Subsequently, the results of the empirical study are presented and analyzed in the fourth chapter. Finally, this research is ended with a discussion and conclusion in the last chapter.

2. Theoretical Framework

This chapter is divided into six sections. The first section (2.1) provides the theoretical elements of how SRI differs from non-SRI. Thereafter, in the second section (2.2), measures of fund performance are delineated. In the third section (2.3), the performance theories of SRI are described. Subsequently, the literature that depicted the performance of SRI funds compared to the market and to non-SRI funds are outlined in section 2.4 and 2.5 respectively. Lastly, in section 2.6 the hypothesis of this paper is illustrated.

2.1 Socially Responsible Investments

This sub-section has three sections and aims to provide an explanation of how SRI differs from non-SRI. Firstly, SRI is defined. Secondly, screening is defined. Finally, SRI implementation techniques are characterized.

2.1.1 What is SRI?

As SRI does not have a universal definition, the extent to which an investment is SRI can be scaled depending on the perspective of how SRI is assessed. Derwall et al. (2011) defined SRI as only

(6)

6

an investment that combines both financial gains and social responsibility. Boatright (1999) proceeded one step further and defined SRI as investments that need to include ‘people and planet’. Similarly, Berry and Junkus (2013) described SRI as combining ‘personal values and societal concerns’ into the investment choice. For these definitions the requirements needed to be characterized as SRI are marginal. However, Renneboog et al. (2008) defined SRI as an investment mechanism that incorporates ‘social, environmental and ethical considerations’ into the investment choice. Therefore, into this definition more strenuous principles are incorporated. In conclusion, SRI funds limit their investments to securities of companies whose actions are observed as being socially acceptable (Bollen, 2007).

2.1.2 Screening

Unlike conventional funds, SRI funds apply a set of investment screens to choose companies that fit certain social criteria or prevent those that are participating in undesirable activities (Liete et al., 2017). Barnett et al. (2006) defined screening as a process of excluding or including securities in a portfolio that depends on certain criteria. These screens are usually set on ethical, social, ecological, and corporate governance criteria. Furthermore, some even include criteria such as engaging in local communities or shareholder activism (Renneboog et al., 2008).

2.1.3 SRI Techniques

SRI funds, unlike conventional funds, have four techniques for implementing SRI. These are: positive screening, negative screening, best-in-class, and activism. In the first three techniques SRI is implemented in the screening process while in the last technique it is implemented at a later stage.

Positive screening includes securities of companies that set positive examples such as companies that engage in ‘progressive hiring practices or produces renewable energy’ (Humphrey & Lee, 2011). On the other hand, negative screening excludes companies that are participating in undesirable business practices or activities, such as firms involved in the weapon, tobacco or gambling industry (Scholtens, 2013). Thirdly, best-in-class uses the screening process to select and invest in the top 30% - 50% of firms with respect to a particular social criterion such as environmental sustainability (Scholtens, 2013). Lastly, Sparkes (2008) defined activism as using shareholder rights to contend socially responsible objectives.

2.2 Measuring Fund Performance

Fund performance can be measured in two ways: alpha and ratios (Wallis & Klein, 2014). Firstly, the more common approach in literature to outline SRI fund performance is the Jensen alpha (Wallis & Klein, 2014). It is defined as the ‘risk-adjusted measure of portfolio performance’, predicted by the capital asset pricing model (CAPM) given the market return and beta (Jensen, 1967). The CAPM is defined by the following equation:

(7)

7

𝑟_!− 𝑟_",! = 𝛼_$+ ß_%&''𝑟_!(_{− 𝑟}

",!( + 𝜀! (1)

where 𝑟_! is the monthly return in time t of the equally weighted portfolio or individual fund; 𝑟_",! is the return on the risk free rate in time t; 𝑟_!(_{is the return on the market proxy in time t; ß}

%&' is the market portfolio factor loading; 𝛼_$ is Jensen’s alpha; and 𝜀_! is the idiosyncratic risk return.

The model can be extended to include more risk factors such as the Carhart four-factor model. However, in each potential model, the alpha interpretation remains the same – ‘as a risk-adjusted measure of portfolio performance’ (Jensen, 1967). The Carhart four-factor model is defined by the following equation:

𝑟_!− 𝑟_",! = 𝛼₎+ ß_%&''𝑟_!(_{− 𝑟}

",!( + ß*%+𝑟!*%++ ß,%-𝑟!,%-+ ß%.%𝑟!%.%+ 𝜀! (2)

where 𝑟_!*%+_{is the difference in returns between a portfolio of small caps and a portfolio of large caps} in period t; 𝑟_!,%-_{is the difference in returns between a portfolio of value stocks and a portfolio of} growth stocks in period t; 𝑟_!%.%_{is the difference in returns between a portfolio of past winners and} losers in period t; ß_*%+, ß_,%-, ß_%.% are the factor loadings for each factor; 𝛼₎ is the fund’s portfolio four factors risk adjusted return.

The second approach to measure fund performance is to use ratios such as the Sharpe ratio and the Treynor ratio. These ratios are less distinguished in literature about SRI fund performance even though the Sharpe ratio is the most prominent performance measure used (Mistry & Shah, 2013). This might be due to their inherent downsides such as: the assumption of normally distributed returns, they do notdistinguish between upside and downside volatility, and they are based on historical data (Sharpe, 1994). Limitations to the Treynor ratio include that it is based on historical data, it only takes systematic risk into account, and assumes normally distributed returns (Steinki & Mohammad, 2015).

Besides research on the applicable techniques used to measure fund performance, there has also been research concerning the optimal sample size. Using advanced statistical techniques, Ang and Kristensen (2012) indicated the optimal sample size to be between 1.5 and 8.5 years with monthly data. Furthermore, Gilbert et al. (2014) determined that higher frequency data is unfavorable to the betas in single and multifactor models due to their higher opacity exposure.

2.3 SRI Performance Theory

There are two main performance theories of SRI. The first subsection (2.3.1) outlines the underperformance theory while the second subsection (2.3.2) describes the overperformance theory.

(8)

8

2.3.1 SRI Underperformance Theory

SRI underperformance theory can be traced back to Markowitz modern portfolio theory (MPT) (Markowitz, 1952). According to this theory, portfolios that restrict their universe of investments would not be able to diversify properly and therefore their performance would lack those that do not restrict investments. Furthermore, MPT suggests there would be a premium to investors in capital efficient markets from the compensation of systematic risk but not from unsystematic risk. Therefore, due to SRI restricting their investments they would not be able to fully diversify their unsystematic risk as much as non-SRI funds. Rudd (1981), Hickman et al. (1999), and Tippet (2001) applied this theory to SRI. They suggested that by restricting the investment universe to just socially responsible investments it would increase total portfolio risk, transaction costs, and management fees and hence, reduce returns for the individual investor.

Furthermore, Belkaoui and Karpik (1989) suggested that SRI managers have agency and opportunity costs that do not have a financial aim and would, therefore, affect financial returns negatively. A study by Adler and Kritzman (2008) quantified the cost that incurs to investors by choosing SRI. They concluded that investors give up between 0.17% – 2.4% of return each year due to the restrictions on the investment universe.

The shunned-stock theory developed by Darwell et al. (2011) made an approach to explain why SRI funds underperform non-SRI funds. It made three assumptions: SR investors are value-driven, are large enough to affect stock prices, and care more about nonfinancial returns. The theory states that when SR investors shun non-ethical stocks, their prices will go down and therefore their returns will go up. Therefore, non-ethical funds which invest in these stocks will outperform those that do not.

2.3.1 SRI Overperformance Theory

While underperformance theory can be traced back to a fundamental finance theory, overperformance theory cannot. Overperformance theory is multiple small theories that point to a general overperformance for SRI. Bello (2005) proposed that SRI funds have a screening process, through which these social filters are tools for selecting companies with higher management quality and as a consequence have higher financial performance. Hamilton et al. (1993), Guerard (1997), Goldreyer et al. (1999), and Tippet (2001) extended this and suggested that on the company level, companies that meet certain ethical criteria will be more efficient and superiorly managed. Furthermore, Gregory et al. (1997) and Luther et al. (1992) provided that SRI funds invest into smaller companies which can easier adapt to changing market conditions. In conclusion, that suggests that funds which invest into these companies will overperform those that do not.

Kurtz (1997) and Goldreyer et al. (1999) implied that SRI fund managers need more and better knowledge of the companies they invest in than non-SRI fund managers. Consequently, this would then

(9)

9

lead to managers basing their investment allocation decision on better, higher quality, and more comprehensive knowledge which would reduce the decision risk.

In the same paper, as describing the shunned-shock theory, Darwell et al. (2011) proposed the errors-in-expectations hypothesis. This theory predicts that SRI funds overperform due to giving investors superior returns because the market is unable to value SRI to its correct value. Darwell et al. (2011) determined three reasons for this occurrence. The first reason is that valuing SR investments is a subjective matter. The second reason is that non-ethical activities get more attention from the public and therefore positive ethical activities are not valued enough. The last reason states that the accounting standards do not value SRI practices because these are usually intangible and left out in financial statements. Consequently, the firms are incorrectly priced.

2.4 SRI Fund Performance Compared to the Market

When comparing the performance of SRI funds to the market most studies have derived that SRI funds underperform the market. This is depicted in Table 1 below, which shows the results of previous papers. Seven out of the eleven papers found negative performance while two discovered a range of negative and positive performance and two identified positive performance.

Table 1: Performance of SRI Funds Compared to Market - Results of Previous Papers

SRI and Non-SRI indicates the number of SRI and non-SRI funds used in the papers respectively. Table 1 is ordered in terms of performance of SRI from low to high. Performance of SRI funds are the annualized average alpha given from the multifactor models of each paper unless otherwise specified. Some papers give monthly alphas and are annualized by multiplying them by 12.

Article Country SRI Non-SRI Years SRI Performance

Statman (2000) US 31 62 1990-1998 -5.04 %

Renneboog et al. (2008) Europe, North America, Asia Pacific

440 16,036 1991-2003 -4.01% (only Europe)

Bauer et al. (2007) Canada 8 267 1994–2002 -3.18%

Scholtens (2005) Netherlands 12 10 2001-2003 -2.86 %

Bauer et al. (2006) Australia 25 281 1992-2003 -2.17%

Goldreyer & Diltz (1999) US 49 180 1981-1997 -0.49 % (CAPM)

Liete et al. (2017) Sweden 33 102 2002-2012 -0.09% (only for inv.

Eur.)

Mallin et al. (1995) UK 28 140 1986–1993 -3.36% -14.52%

Gregory et al. (1997) US 18 18 1986–1994 -0.71% - 0.24%

Kreander et al. (2005) UK, Sweden, Germany, Netherlands

30 30 1995-2001 2.40 %

(10)

10

2.5 SRI Fund Performance Compared to Non-SRI Funds

On the empirical level, there are very conflicting results when comparing SRI funds performance to non-SRI funds. Therefore, in the first section (2.5.1) the studies that derive similar performance are depicted. In the second (2.5.2) and third section (2.5.3), literature that exemplifies under- and overperformance respectively, is debated. In the last section (2.5.4), explanations for the differences in results and similar performances are given. For an overview of the literature, Table 2 summarizes the number of studies that found similar-, under- or overperformance of SRI funds compared to non-SRI funds.

Table 2: SRI Fund Performance Compared to Non-SRI Funds

Table 2 presents the number of studies that found similar-, under- or overperformance of SRI funds compared to non-SRI funds. Multiple market are the studies that included more than one market such as European and North American. Europe includes studies that studied Europe as a whole market or an individual European country. Index includes studies on indexes.

Multiple Market

Europe US Other Index Total

Similar Performance 3 5 6 2 - 16

Underperformance 1 1 2 1 2 7

Overperformance - 2 1 - 2 5

2.5.1 Similar Performance

Most research concludes that there is similar performance between SRI and non-SRI funds, as demonstrated in Table 2. By studying the European, North American, and Asian Pacific markets, Renneboog et al. (2008) showed that in most countries conventional funds perform similarly to SRI funds. It is the only study investigating SRI funds domiciled in Luxembourg. They explored the performance of 12 SRI funds against 360 non-SRI funds in Luxembourg and discovered a statistically similar performance. However, it is important to note that this study is not an accurate representation of the Luxembourg fund market, since 41 out of 56 of the funds domiciled in Luxembourg were assigned to other European countries.

Other multiple market studies, such as Schröder (2004) and Cortez et al. (2012), concluded that SRI funds perform equivalently to conventional funds. Schröder (2004) derived this by finding the alpha from multiple factor models by studying 30 US funds and 16 funds from Germany and Switzerland. Cortez et al. (2012) studied the US and European market. They found similar performance between SRI funds and non-SRI funds for most European markets.

Most studies focusing only on the fund market of the United States also determined similar performance. By studying the period from 1990 to 2001, Bauer, Koedijk, and Otten (2005) detected no significant difference between US SRI and non-SRI funds which were matched on size and age.

(11)

11

However, the SRI performance improves (SRI perform better than non-SRI) and becomes significant in the period 1998 to 2001. Furthermore, Geczy et al. (2003), Bello (2005), Goldreyer and Diltz (1999), Gregory et al. (1997), and Statman (2000) also documented similar performance between SRI funds and non-SRI funds in the US.

Besides the US market, there are other single country studies. Bauer, Derwall, and Otten (2007), who studied the Canadian market, and Bauer, Otten, and Tourani (2006), who studied the Australian market, evaluated similar performance between SRI and non-SRI funds.

Besides studies investigating globally, there have been specific studies on the European market. Kreander et al. (2005) and Leite and Cortez (2014), who studied the European market discovered no difference in performance between the conventional and SRI funds. Lastly, there has been a couple of individual market studies for Europe. Leite et al. (2017), Fernandez-Izquierdo and Mattalin-Saez (2008) and Scholtens (2005) investigated the performance of Swedish, Spanish, and Dutch funds, respectively, and found statistically no performance difference.

2.5.2 Underperformance

Diverging from the previous literature, a few studies discovered underperformance between SRI funds and conventional funds. For example, the previously mentioned study by Renneboog et al. (2008) observed that SRI funds domiciled in Sweden, Japan, Ireland, and France underperform compared to conventional funds. Rennenboog et al. (2008) states that underperformance may be due to the higher management fees of SRI funds and, to a lesser extent, the higher transaction costs and non-stock holdings. This links back to the underperformance theory and can therefore partially explain why underperformance is observed.

Cortez et al. (2012) investigated that US funds and Austrian funds underperform both conventional funds and socially responsible funds. Cortez et al. (2012) state the use of negative social screens as the reason why underperformance is observed. Again, this provides evidence supporting the underperformance theory. Lastly, Otten et al. (2002), by studying funds in Germany, UK, and the US, discovered that German and US SRI funds underperform their non-SRI peers and relevant indexes. They evaluated this by using the Carhart four-factor model and controlling for investment style. Lastly, besides empirical evidence, a meta-analysis conducted by Renneboog et al. (2007) illustrated that existing studies hint but do not ‘univocally demonstrate’ that SRI perform worse.

Walley and Whitehead (1994) suggested that the average expense ratio of SRI funds would be higher than those of non-SRI funds. Therefore, the average return after fees would be less for SRI funds than for non-SRI funds.

In addition to the studies that compared conventional funds to SRI funds, there are also other studies that conclude underperformance of SRI. Ghoul and Karam (2007) stated that the Dow Jones

(12)

12

Islamic Index outperformed the Domini Social 400 Index during a one-year period. By excluding the companies that had operations in South Africa during the apartheid from the S&P 500 index, Rudd (1979) found that it lacked in returns from the original S&P 500 index.

2.5.3 Overperformance

A few studies concluded overperformance of SRI funds to conventional funds. Firstly, Mallin et al. (1995) studied the performance of ethical funds versus non-ethical funds in the UK. They discovered that ethical funds overperformed the non-ethical trusts considering their Jensen alpha. They suggest the reason for this is due to the high demand of SRI during the period which would produce a premium in terms of return. However, this is not explained by the overperformance theory described in section 2.3.1 but it does enhance it.

The previously mentioned study by Otten et al. (2002) demonstrated that UK SRI funds overperformed their conventional peers. Furthermore, by studying the US market, Gil-Bazo et al. (2010) found that US SRI funds overperform their conventional peers. They state the overperformance is due to the better management of SRI companies that SRI funds invest in. Therefore, this provides evidence supporting the overperformance theory. Luck and Pilotte (1993) discovered that the Domini Social Index outperformed the S&P 500 index during 1990 to 1992. Furthermore, Statman (2000) found the same results by studying the period of 1990 to 1998.

There has been evidence that SRI funds overperform non-SRI funds in crisis periods. The claim is that SRI funds are better at limiting the downside risk in times of crisis than conventional funds (Nofsinger & Varma, 2014). Nofsinger & Varma (2014) discovered that SRI funds outperform conventional funds in periods of market turmoil. Liete et al. (2017) also demonstrated that conventional funds overperform SRI funds in normal market states but match them in periods of market crisis. Therefore, in this study analysis of crisis periods is avoided, in order to determine the significant underlying performance difference between SRI funds and non-SRI funds.

2.5.4 Explanations for Differences and Similar Performance

There are many reasons on the divergence of results in studies investigating this topic. The indisputable reason why this happens is due to the methodologies differing in each paper. For example, the studies examine different periods, some not controlling for crisis periods, different countries, and different sample size. All this concludes in a different evaluation of performance. Secondly, there are a lot of different performance measures used in each study which will obviously also provide different results.

An interesting study by Chatterji et al. (2015) describes little convergence among different SRI rating agencies. This suggests SRI funds in one research could be a conventional fund in another research, which would explain the differences in results. Furthermore, according to Kempf and Osthoff

(13)

13

(2008), SRI funds are in reality not socially responsible but conventional funds in disguise. In order to bypass this, this study expands the current literature by sampling the funds from one rating agency that rates both SRI and non-SRI funds.

A study by Galema et al. (2008) supposed the misinterpretation of the risk-adjusted performance measures are the reason why the results differ so much in SRI performance literature. They presumed it emerges when computing performance, which usually controls for systematic risk. This control would then capture part of the trade-off between SRI performance and financial performance. Moreover, Renneboog et al. (2007) stated that SRI underperformance can be driven by the omission of an ethical risk factor. Therefore, in this study, the robustness checks include an ethical risk factor.

One possible explanation for similar performance between SRI and conventional funds is that they are increasingly becoming similar. Laurel (2011) argued that SR investments are becoming the norm rather than being a choice. Furthermore, Crifo and Mottis (2010) suggest that screening will become the norm. Moreover, Bello (2005) demonstrated that SRI funds can diversify as well as conventional funds. Therefore, there are no diversification losses to the individual investor. Evidence also suggests that SRI and non-SRI funds carry the same amount of SR securities (Utz & Wimmer, 2014). Lastly, Humphrey et al. (2016) denoted that SRI funds do not have very different holdings compared to their non-SRI peers. This suggests that SRI and non-SRI funds are converging and/or that more and more funds are getting higher SRI scores. Consequently, this could provide an explanation as to why so many empirical papers evaluated similar financial performance between them.

2.6 Hypothesis

This section derives the hypotheses of this study based on the previous literature review. As a first step, the financial performance of SRI funds is compared to the market. Most of the literature provides that SRI funds underperform the market. Therefore, the first hypothesis is:

H1: The average alpha for the SRI funds is negative, such that the SRI funds underperform the market.

Additionally, the current literature does not provide a decisive answer whether these funds overperform, underperform or perform similarly. However, most point to the conclusion that they perform similarly. Therefore, the second hypothesis that is tested is:

H2: The difference in performance between SRI funds and non-SRI funds is zero, such that SRI funds perform similarly to non-SRI funds.

The first and second hypothesis can be regarded as conflicting. However, if both SRI and non-SRI underperform the market, then their performance will be similar.

(14)

14 3. Method and Data

This chapter provides an explanation for the method and data used in this paper. The first section provides an outline of the methodology. Subsequently, the robustness checks are explained in section two. In the third section the used data are described. The last section presents the descriptive statistics.

3.1 Methodology

The performance analysis of SRI funds and non-SRI funds were derived by ratios and alphas. Firstly, the Sharpe ratio and the Treynor ratio were calculated (see the definition of these ratios in Appendix 8.1). Secondly, the alphas were found using the CAPM and Carhart four-factor model which was derived in section 2.2 and given as equations 1 and 2. The variable of interest in these models is the intercept, which is the alpha. These models were run on the equally weighted portfolios of the SRI and non-SRI funds. Furthermore, their difference in performance was tested with the following equations 3 and 4:

𝑟_!*/0_{− 𝑟}

!1213*/0 = 𝛼$+ ß%&''𝑟!(− 𝑟",!( + 𝜀! (3) 𝑟_!*/0_{− 𝑟}

!1213*/0= 𝛼)+ ß%&''𝑟!(− 𝑟",!( + ß*%+𝑟!*%++ ß,%-𝑟!,%-+ ß%.%𝑟!%.%+ 𝜀! (4)

where rtSRI represents the monthly excess return in time t of the equally weighted SRI fund portfolio;

and rtnon-SRI represents the monthly excess return in time t of the equally weighted non-SRI fund

portfolio. The alpha is again the main variable of interest. This step is executed to find the difference between SRI and non-SRI funds.

All equations are estimated using ordinary least square regressions. The standard errors are corrected for heteroskedasticity and autocorrelation by using the Newey-West approach.

3.2 Robustness Checks

There are two robustness checks. Firstly, the methodology already includes a robustness check since performance is measured from the CAPM and Carhart four-factor model. Therefore, robustness is simultaneously checked when investigating financial performance. If the models produce similar results, then the results are robust.

The second measure of robustness is to include a socially responsible index in the regressions. As previously mentioned in section 2.5.4, this is to take into account the underperformance effect of the omission of an ethical risk factor (Renneboog et al., 2007; Galema et al., 2008). Therefore, the

(15)

15

financial performance measures can be improved by including an index to proxy for this ethical risk. The following regressions as demonstrated in equations 5 and 6 are run:

𝑟_!− 𝑟_",! = 𝛼_$+ ß_%&''𝑟_!(_{− 𝑟}

",!( + ß4𝑟!51678+ 𝜀! (5) 𝑟_!− 𝑟_",! = 𝛼₎+ ß_%&''𝑟_!(_{− 𝑟}

",!( + ß*%+𝑟!*%++ ß,%-𝑟!,%-+ ß%.%𝑟!%.%+ ß4𝑟!51678+ 𝜀! (6)

where rtindex is the monthly returns in time t of the ethical index. The ethical index is proxied by the

FTSE4Good Global Index. This regression will be run on the SRI and non-SRI funds. Again, if this provides similar results than the models in the methodology, then those models are robust.

3.3 Data

This section describes the data and how they were derived. The first and second subsection provides information on how the SRI and non-SRI funds were collected. The third subsection depicts the data frequency and time interval used. Lastly, the risk-free rate and risk factors used in this paper are described. An overview of the data used is shown in Table 3.

Table 3: Data

Table 3 presents the main variables used in the research with their respective label, unit name, source, and explanation.

Variable Label Unit Name Source Explanation

SRI Funds 𝑟_!*/0 _{Euro (€)} _- _Factset _{Returns on the equally}

weighted portfolio of SRI funds

Non-SRI Funds

𝑟_!1213*/0 _{Euro (€)} _- _Factset _{Returns on the equally}

weighted portfolio of non-SRI fund Risk-free

Rate

𝑟",! Euro (€) 1 Month Euribor Rate European Central Bank

(ECB)

Interest rate at which European Banks lend to

one another Risk Factors 𝑟_!(_{, 𝑟} !*%+, 𝑟_!,%-_{, 𝑟} !%.% US Dollar ($) Fama/French European/Developed/Emerging 3 Factors/Momentum Factor Kenneth French Website

See section 2.2 for definitions of risk factors

Ethical Risk Factor

𝑟_!51678 _{Euro (€)} _{FTSE4Good Global Index} _{Investing.com} _{Return on a socially} responsible index in order to proxy ethical risk

(16)

16

3.3.1 SRI Funds

At the end of 2019, according to FactSet, there were 2404 SRI fund units (roughly 368 funds) domiciled in Luxembourg with a rating of 8 and higher according to the MSCI ESG fund rating. This rating measures the environmental, social, and governance characteristics of the fund’s holdings and rates them from a scale from 0 to 10, with 10 being the highest.

The sample was created by randomly selecting funds from the population by randomly generating a number from 1-2025 and choosing the fund it selected. Each fund was then checked on the criteria: only equity, mutual fund, and sufficient data. Furthermore, the population consisted of fund units which meant that for some funds there were multiple units with different ratings. For those, the highest rating one was chosen rather than the fund unit it selected. If the selected fund did not meet the criteria another fund was selected until there was a sample of 30 SRI mutual equity funds that were domiciled in Luxembourg. A number of 30 mutual equity SRI funds was chosen as this was considered to be a representative and significant sample.

3.3.2 Non-SRI Funds

The non-SRI funds were sampled from the funds being domiciled in Luxembourg with a 0 to 4.3 MSCI ESG rating. According to the rating, funds with a rating until 4.3 are exposed to companies that do not meet ESG characteristics or are worsening these issues and therefore can be characterized as non-SRI. The population consisted of 9975 fund units or roughly 1529 funds. The same procedure was done as with the SRI funds until there was a sample of 30 non-SRI funds.

3.3.3 Data Frequency and Time Interval

For each fund, the monthly returns were downloaded for the sample period from January 2012 to December 2019. January 2012 was chosen in order to avoid the European Sovereign Debt Crisis which emerged around 2010. The ending of the crisis is well debated among finance literature. However, most point to the end of 2011 and therefore January 2012 was chosen (Liete et al., 2017; Ruscakova & Semancikova, 2016). December 2019 was chosen in order to avoid the global COVID-19 pandemic which disrupted financial markets substantially.

In order to access the aggregate performance of the SRI and non-SRI funds an equally weighted portfolio for each was created.

3.3.4 Risk-free Rate and Risk Factors

In order to compute excess returns, the risk-free rate was proxied by the Euribor with 1-month maturity, which was downloaded from the European Central Bank statistical data warehouse.

(17)

17

The risk factors excess market return, small-to-big, book-to-market, and momentum were used. All the risk factors were obtained from the Kenneth French website. Since both SRI and non-SRI funds invested in Europe, globally, and emerging markets, their respective risk factors were downloaded. More specifically, the ‘Fama/French European 3 Factors’ and the ‘European Momentum Factor’ was downloaded for the funds that invested in Europe. ‘Fama/French Developed 3 Factors’ and ‘Developed Momentum Factor’ were used for funds investing globally. ‘Fama/French Emerging 5 Factors’ and ‘Emerging Momentum Factor’ were used for funds that invested in emerging markets.

Each risk factor was downloaded in monthly frequency and was expressed in US Dollars. Therefore, the risk factors were converted into Euros by using the appropriate monthly exchange rates which were obtained from FactSet. Several studies, such as Cortez et al. (2012), Chua et al. (2008), and Liete and Cortez (2014), used a similar approach by using factors expressed in US Dollars and converting them into the currency they need. Furthermore, a study by Glück et al. (2019) suggested not converting the risk factors in the respective currency would severely skew the alphas and the factor loadings.

An ethical risk factor had to be downloaded, in order to perform a robustness check (see equation 5 and 6). This risk factor was proxied by the FTSE4Good Global Index. Monthly returns were downloaded from Investing.com for the period of January 2012 till December 2019.

3.4 Descriptive Statistics

Table 4 presents the summary statistics of the funds used in this paper. It provides an understanding of the data used in this paper. Several important insights are given when comparing the statistics between SRI funds and non-SRI funds.

Table 4: Summary Statistics of the Funds

Table 4 presents the key statistics based on the equally weighted portfolio of the SRI and non-SRI funds for the period 2012-2019. Difference is given by SRI minus non-SRI. Monthly returns and standard deviation of the returns are in percentage terms. Values in the brackets are the p-values.

SRI Non-SRI Difference

Monthly Return 0.78% 0.62% 0.16%

(0.596)

SD of Return 3.30% 3.66% -0.36%

Skewness -1.05 -1.64 0.59

Kurtosis 5.53 8.98 -3.45

(18)

18

Firstly, the average monthly returns for the SRI funds are higher than those of the non-SRI funds. This could lead to an inference that SRI funds overperform non-SRI funds, but the difference is relatively small at only 0.16%. A t-test for the significance of the difference provides a p-value of 0.596 (results can be seen in Appendix 8.2). Therefore, the null hypothesis which claims that the difference is equal to zero, cannot be rejected. This could provide evidence against the second hypothesis with stated SRI and non-SRI funds perform similarly.

A second observation is that SRI funds have a slightly lower standard deviation. Thirdly, the returns are not normally distributed when considering the numbers for skewness and kurtosis. Furthermore, a test of normality shows that the null hypothesis of normality can be rejected at a 0.05% significance level which is important for the assumption of the Sharpe and Treynor ratios (results can be seen in Appendix 8.3).

Besides insights on the returns, there are several interesting observations on the fund characteristics which are depicted in Appendix 8.4. The age of the SRI funds exceeds that of the conventional funds. This provides evidence against the assumption that the SRI market is young and evolving. Moreover, the SRI funds total net assets exceed those of the non-SRI funds. Lastly, against expectations, the net expense ratios of the SRI funds are lower than those of the non-SRI funds.

4. Results

This section illustrates the results. The first subsection (4.1) presents the results from the ratios. The second (4.2) and third (4.3) subsections provide the results from the single factor and multifactor model, respectively. This is then followed by the robustness checks in the final section.

4.1 Ratios

Table 5: Sharpe and Treynor Ratios

Table 5 presents the results from the Sharpe and Treynor ratios for the equally weighted portfolios of SRI and non-SRI and their difference. Difference is given by SRI minus non-SRI. Values in the brackets are the p-values from a t-test.

SRI Non-SRI Difference

Sharpe Ratio 0.317 0.242 0.075

(0.390)

Treynor Ratio 1.132 0.888 0.245

(0.437)

(19)

19

As Table 5 indicates, the Sharpe and Treynor ratios are higher for the SRI funds than for the non-SRI funds. This could indicate that the SRI funds are overperforming in terms of their risk-adjusted returns. However, a t-test on the significance of the difference in Sharpe ratios produces a p-value of 0.390 and for Treynor ratios a p-value of 0.437 (results can be seen in Appendix 8.2). This points evidence towards the second hypothesis since the difference in Sharpe and Treynor ratios are not statistically significant from zero. This would indicate that SRI and non-SRI funds perform similarly. However, these results are inconclusive due to the assumption of normality for the SRI and non-SRI monthly returns not being valid.

4.2 Single-Factor Model

Table 6 presents the results from the CAPM regression. The explanatory power of the model for the returns is relatively high for SRI and non-SRI funds, with adjusted R2_{above 70%. The alpha is}

positive for both SRI and non-SRI funds. Furthermore, the alpha for the SRI funds is higher at 0.297 compared to 0.078 for the non-SRI funds. This suggests that SRI funds overperform non-SRI funds, however, their alphas are not significant. Therefore, none of the outperformances over the market are statistically different from zero. This provides evidence against the first hypothesis, which states that SRI funds will underperform the market.

In order to test the second hypothesis, which states that SRI and non-SRI funds have similar performance, the difference in performance between the funds was tested. This test provided a non-statistically significant value of 0.221. This suggests that the performance difference of SRI funds compared to non-SRI funds are not statistically different from zero. This, therefore, provides evidence towards the second hypothesis.

Besides the hypotheses tests, the CAPM model provides another implication which could affect performance. The market betas propose that both SRI and non-SRI funds move in the same direction as the market, as both are positive and significant at a 0.01 significance level. Furthermore, the difference is negative but not significant, suggesting they are both equally affected by the market risk. These results are inconsistent with Kreander et al.’s (2005) and Gregory et al.’s (1997) observations as both demonstrated that SRI funds are less impacted by market risk than non-SRI funds.

(20)

20

Table 6: Regression Results

Table 5 presents the main results from the single and multifactor model for the equally weighted portfolios of SRI and non-SRI and their difference. Results are based on Newey-West standard errors. Values in the brackets are the p-values.

CAPM Carhart four-factor model

SRI Non-SRI Difference SRI Non-SRI Difference

Alpha 0.297 (0.228) 0.078 (0.762) 0.221 (0.321) 0.228 (0.358) 0.044 (0.874) 0.184 (0.566) Market 0.924*** (0.000) 0.998*** (0.000) -0.074 (0.476) 0.969*** (0.000) 0.982*** (0.000) -0.013 (0.806) SMB -0.117 (0.669) -0.461* (0.083) 0.344 (0.290) HML -0.542*** (0.010) 0.108 (0.629) -0.650*** (0.003) MOM 0.081 (0.591) 0.097 (0.523) -0.017 (0.930) Adj R2 _0.702 _0.750 _0.010 _0.727 _0.752 _0.077

*Significant at the 10% level **Significant at the 5% level ***Significant at the 1% level

4.3 Multifactor Model

The multifactor model produces results similar to the single index model which is depicted above in Table 6. The explanatory power of the multifactor model is higher for this model than the single index model, due to the adjusted R2_{being higher. The alphas are again positive for both SRI and}

non-SRI, suggesting they overperform the market. The alpha is again larger for SRI funds (0.228) than for non-SRI funds (0.044), suggesting SRI funds overperform to the market compared to non-SRI funds. However, the SRI alpha is again not significant; thus the SRI funds do not significantly overperform the market. This again provides evidence against the first hypothesis.

The difference between SRI and non-SRI funds is positive (0.184) but not significant. Again, it supports the second hypothesis as SRI and non-SRI funds performances do not differ statistically from 0. Concluding, this could suggest that SRI and non-SRI funds perform similarly.

The market betas again are both positive and significant for both SRI and non-SRI funds. Furthermore, the difference suggests they do not differ significantly in terms of market risk. In terms of investment style, non-SRI funds are more exposed to large-cap stocks than SRI funds considering the small minus big (SMB) coefficients. The high minus low (HML) coefficient is negative and statistically significant for the difference portfolio suggesting SRI funds are more exposed to low

(21)

book-21

to-market stocks. Lastly, the momentum (MOM) coefficient is slightly positive for both SRI and non-SRI funds but not statistically significant.

4.4 Robustness Checks

The first robustness check is to compare the results from the ratios, single and multifactor models. All models discovered similar results suggesting there is robustness in the results. The ratios showed that SRI funds overperformed non-SRI funds, but these results are inconclusive. The alphas in both models suggest SRI funds overperformed non-SRI funds but the difference, while positive, is insignificant.

The second robustness check is to include a socially responsible index in the single factor and multifactor model. This is done to take into account the underperformance effect of an omission of an ethical risk factor. The results are shown in Table 7.

Table 7: Regression with Ethical Risk Factor

SRIethical_{and non-SRI}ethical_{show the results from the regressions with the ethical risk factor. SRI and non-SRI}

show the results without the ethical risk factor. Results are based on Newey-West standard errors. Values in the brackets are the p-values.

CAPM Carhart four-factor model

SRIethical _Non-SRIethical _SRIethical _Non-SRIethical

Alpha 0.301 (0.248) 0.167 (0.522) 0.202 (0.433) 0.278 (0.365) Market 0.928*** (0.000) 1.132*** (0.000) 0.953*** (0.000) 1.126*** (0.000) SMB -0.102 (0.721) -0.606** (0.023) HML -0.538** (0.013) 0.069 (0.740) MOM 0.096 (0.485) -0.044 (0.762) Ethical Index -0.009 (0.959) -0.252*** (0.004) 0.034 (0.824) -0.314*** (0.001) Adj R2 _0.699 _0.766 _0.725 _0.775

*Significant at the 10% level **Significant at the 5% level ***Significant at the 1% level

Three important insights can be derived from this test. Firstly, the adjusted R2_{remains relatively}

similar but decreases by a small amount when including an ethical risk factor for the SRI funds. On the other hand, the adjusted R2_{increases for the non-SRI funds when including an ethical risk factor.}

(22)

22

The second insight is that the SRI alphas remain relatively similar when including an ethical risk factor. On the other hand, the non-SRI alphas increase when including an ethical risk factor. However, all alphas are still insignificant.

The third insight is that the ethical risk factor for SRI funds is close to zero (-0.009) for the CAPM and positive (0.034) for the multifactor model. This is in accordance with the expected results, since SRI funds are ‘compensated’ for being ethical. Furthermore, the non-SRI funds have two significant and negative values for the coefficient of the ethical index (-0.252 and -0.314). This is again consistent with the expectations since the non-SRI funds are ‘fined’ for being non-ethical.

These findings suggest the results do not differ significantly when including the ethical risk factor for the SRI funds. Hence, one can argue that the financial performance of the SRI funds is not adversely affected by the ethical risk factor and is, as a consequence, robust. However, there is a difference in results when looking at the non-SRI funds. Therefore, the ethical risk factor does affect the performance of non-SRI funds and the results could be argued to be less robust.

5. Discussion and Conclusion

The last section of this paper provides a summary of the results and how they are related to current literature. Following this, the limitation of this study is provided. Lastly, the suggestions for future research are outlined.

5.1 Discussion of Results

The main goal of this study was to examine the performance of SRI equity mutual funds. Particularly, the performance difference between SRI and non-SRI equity mutual funds was investigated. In addition, by analyzing the Luxembourg fund market, the purpose of this research was to expand the gap in scientific knowledge.

There was no support for the first hypothesis which presumed that SRI funds underperform the market by considering their risk-adjusted returns. The results from all models suggested that SRI funds overperformed compared to the market. However, no outperformance was significant. Consequently, none of the outperformances were statistically different from zero. This suggests that SRI funds performed similarly to the market. The current literature on the performance of SRI funds to the market suggests they underperform; therefore, the results disagree with current literature. However, since the literature is not consistent it nevertheless agrees with the results of some studies (Kreander et al., 2005, Geczy et al., 2006).

There was support for the second hypothesis which stated that SRI funds perform similarly to their non-SRI peers. The results from all models suggested that SRI funds overperformed non-SRI funds

(23)

23

by analyzing the risk-adjusted returns of their differences. However, again, all outperformances were insignificant. This suggests that SRI funds perform similarly to non-SRI funds, which is in accordance with the second hypothesis and corresponds with current literature which depicted also similar performance (e.g. Schröder, 2004; Geczy et al., 2006; Scholtens, 2005). Furthermore, it is consistent with the study conducted by Renneboog et al. (2008) which included funds from Luxembourg and evaluated similar performance between SRI and conventional funds. However, it is not consistent with the studies that demonstrated under- or overperformance (e.g. Bauer, Koedijk, and Otten, 2002; Cortez et al., 2012).

The results of the study provide additional evidence to the theory that SRI and non-SRI funds are converging since all results indicate that SRI funds perform similarly to a conventional benchmark and conventional funds. This is in accordance with the theories of Laurel (2011) and Crifo and Mottis (2010) which suggested that SRI and non-SRI funds are increasingly becoming similar.

5.2 Limitations

This study is subject to limitations. Firstly, the biggest limitation is survivorship bias. This bias resulted due to only including surviving funds in the analysis and not including dead funds. This happened due to sampling the funds from before the MSCI ESG fund rating started rating funds. Therefore, there was a trade-off between survivorship bias and getting the funds from the same population. However, studies by Brown and Goetzmann (1992) and Grinblatt and Titman (1989) showed that survivorship bias can only take into account about 0.4% – 0.6% and 0.1% – 0.4%, respectively, of the returns per year. In addition, Renneboog et al. (2007) stated that, over time, only few SRI funds cease to exist. Therefore, one can argue that survivorship bias in this study is relatively small however still there.

A second limitation arises from converting the risk factors from US Dollars to Euros. This potentially could introduce the research to exchange rate risk and bias the factor exposures. However, the size and direction of this bias are unknown.

A third limitation originates from the construction of the equally weighted portfolios. Equally weighted portfolios are more exposed to systematic risk and could lead them to overperform other types of portfolios such as value-weighted ones. A study by Plyakha et al. (2014) found that equally weighted portfolios overperform value-weighted portfolios. This potentially explains why the alphas of the SRI and non-SRI funds were positive in this study.

A fourth limitation is the multicollinearity problem usually associated with multifactor models. The risk factors are usually correlated with each other and this could lead to the coefficients being less

(24)

24

valid. However, multicollinearity would not affect the explanatory power of the models and in this study, the robustness of the multifactor model was checked multiple ways.

A fifth limitation could be small sample size. In this study, only 30 SRI funds were compared to 30 non-SRI funds, which is a relatively small sample size. This small sample size could explain the lack of statistical significance in the models. However, for statistical inferences, a sample size of 30 would be enough. The effect of increasing sample size could only be explained by further research.

5.3 Suggestions for Future Research

Given that this research faces many limitations, there are ample opportunities for future research. Firstly, only a small part of the SRI fund market in Luxembourg was studied. Therefore, one improvement would be to increase the sample size to include more funds. Secondly, to eliminate survivorship bias one can do the same analysis in a couple of years but sample the funds at the beginning of the period to include dead funds. Thirdly, to eliminate the potential exchange rate risk one can create their own risk factors in the respective currency they need. Lastly, one could do the same analysis but with value-weighted portfolios for the SRI and non-SRI funds and see if the results provide a difference. Besides improvements, there are further suggestions to gain a deeper understanding of the topic. Firstly, one could research the performance effect of the different SRI techniques such as positive screening, negative screening, best-in-class, and activism. Furthermore, one could investigate if the performance differs between each technique and which one provides the highest risk-adjusted returns. Secondly, one could research the influence of fund characteristics such as age, size, and expense ratio on SRI funds financial performance. Lastly, there is extensive literature about the performance of SRI equity mutual funds, however, barely any research on SRI funds that specialize in bonds or ETFs. Due to their importance and the growing number of these types of investment vehicles, especially in SRI terms, future research could investigate the performance difference between these SRI funds and their non-SRI peers.

(25)

25 7. References

Adler, T., & Kritzman, M. (2008). The Cost of Socially Responsible Investing. The Journal of Portfolio

Management, 35(1), 52-56. doi:10.3905/jpm.2008.35.1.52

Ang, A., & Kristensen, D. (2012). Testing Conditional Factor Models. Journal of Financial

Economics, 106(1), 132-156. doi:10.1016/j.jfineco.2012.04.008

Barnett, M.L., & Salomon, R.M. (2006). Beyond Dichotomoy: The Curvilinear Relationship Between Social Responsibility and Financial Performance. Strategic Management Journal, 27(11), 1101-1122. doi:10.1002/smj.557

Bauer, R., Derwall, J., & Otten, R. (2007). The Ethical Mutual Fund Performance Debate: New Evidence from Canada. Journal of Business Ethics, 70(2), 111-124. doi:10.1007/s10551-006-9099-0

Bauer, R., Koedijk K. & Otten R. (2005). International Evidence on Ethical Mutual Fund Performance and Investment Style. Journal of Banking and Finance, 29(7), 1751-1767. doi:10.1016/j.jbankfin.2004.06.035

Bauer, R., Otten, R., & Tourani A. (2006). Ethical Investing in Australia: Is there a Financial Penalty?

Pacific-Basin Finance Journal, 14(1), 33–48. doi:10.1016/j.pacfin.2004.12.004

Belkaoui, A., & Karpik, P. G. (1989). Determinants of the Corporate Decision to Disclose Social Information. Accounting, Auditing & Accountability Journal, 2(1).

doi:10.1108/09513578910132240

Bello, Z.Y., 2005. Social Responsible Investing and Portfolio Diversification. Journal of Financial

Research, 28(1), 41-57.

Berry, T. C., & Junkus, J. C. (2013). Socially Responsible Investing: An Investor Perspective. Journal

of Business Ethics, 112(4), 707-720. doi:10.1007/s10551-012-1567-0

Boatright, J., (1999). Ethics in Finance. Blackwell, Oxford.

Bollen, N. P. (2007). Mutual Fund Attributes and Investor Behavior. Journal of Financial and

Quantitative Analysis, 42(3), 683-708. doi:10.1017/s0022109000004142

Brown, S. J., Goetzmann, W., Ibbotson, R. G., & Ross, S. A. (1992). Survivorship Bias in Performance Studies. Review of Financial Studies, 5(4), 553-580. doi:10.1093/rfs/5.4.553

Chang, C. E., Nelson, W. A., & Witte, H. D. (2012). Do Green Mutual Funds Perform Well? Management Research Review, 35(8), 693–708. doi: 10.1108/01409171211247695

Chatterji, A. K., Durand, R., Levine, D. I., & Touboul, S. (2015). Do Ratings of Firms Converge? Implications for Managers, Investors and Strategy Researchers. Strategic Management

Journal, 37(8), 1597-1614. doi:10.1002/smj.2407

Chua, C.T., Lai, S.S., Wu, Y. (2008). Effective Fair Pricing of International Mutual Funds. Journal

Banking Finance. 32, 2307–2324. doi:10.1016/j.jbankfin.2007.06.014

Cortez, M. C., Silva, F., & Areal, N.(2012). Socially Responsible Investing in the Global Market: The Performance of US and European Funds. International Journal of Finance & Economics, 17(3), 254–271.

Crifo, P., Mottis, N. (2010). SRI Analysis and Asset Management: Independent or Convergent? A Field Study on the French Market. ESSEC Research Center, ESSEC Business School, No. 10006.

(26)

26

Derwall, J., Koedijk, K. C., & Horst, J. R. (2011). A Tale of Values-Driven and Profit-Seeking Social

Investors. Journal of Banking & Finance, 35(8), 2137-2147. doi:10.2139/ssrn.1641131

Fernandez-Izquierdo, A., & Mattalin-Saez, J. C. (2008). Performance of Ethical Mutual Funds in Spain: Sacrifice or Premium? Journal of Business Ethics, 81(2), 247–260. doi:10.1007/s10551-007-9492-3v

Galema, R., Plantinga, A., & Scholtens, B. (2008). The Stocks at Stake: Return and Risk in Socially Responsible Investment. Journal of Banking & Finance, 32(12), 2646-2654. doi:10.1016/j.jbankfin.2008.06.002

Geczy, C. C., Stambaugh, R. F., & Levin, D. (2003). Investing in Socially Responsible Mutual Funds. SSRN Electronic Journal. doi: 10.2139/ssrn.416380

Ghoul, W., & Karam, P. (2007). MRI and SRI Mutual Funds. The Journal of Investing, 16(2), 96-102. doi:10.3905/joi.2007.686416

Gil-Bazo, J., Ruiz-Verdú, P., & Santos, A. A. (2009). The Performance of Socially Responsible Mutual Funds: The Role of Fees and Management Companies. Journal of Business Ethics, 94(2), 243-263. doi:10.1007/s10551-009-0260-4

Gilbert, T., Hrdlicka, C., Kalodimos, J., & Siegel, S. (2014). Daily Data is Bad for Beta: Opacity and Frequency-Dependent Betas. Review of Asset Pricing Studies, 4(1), 78–117. doi: 10.1093/rapstu/rau001

Glück, M., Hübel, B., & Scholz, H. (2020). Currency Conversion of Fama/French Factors: How and Why. SSRN Electronic Journal. doi:10.2139/ssrn.3437648

Goldreyer, E., Ahmed, P., & Diltz, J. (1999). The Performance of Socially Responsible Mutual Funds: Incorporating Sociopolitical Information. Journal of Financial Research, 28, 41–57. doi:10.1108/03074359910765830

Gregory, A., Matatko, J., & Luther, R. (1997). Ethical Unit Trust Financial Performance: Small Company Effects and Fund Size Effects. Journal of Business Finance and Accounting, 24, 705–725. doi:10.1111/1468-5957.00130

Grinblatt, M., & Titman, S. (1989). Mutual Fund Performance: An Analysis of Quarterly Portfolio Holdings. The Journal of Business, 62(3), 393. doi:10.1086/296468

Guerard, J. B. (1997). Is there a Cost to Being Socially Responsible in Investing? Journal of

Investing, 6(2), 11-18. doi:10.3905/joi.1997.408416

Hamilton, S., Joe, H., & Statman, M. (1993). Doing Well While Doing Good? The Investment Performance of Socially Responsible Mutual Funds. Financial Analysts Journal, 49(6), 62–66. doi:0.2469/faj.v49.n6.62

Hickman, K.A., Teets, W.R., & Kohls, J.J. (1999). Social Investing and Modern Portfolio Theory.

American Business Review, 17(1), 72-78.

Humphrey, J. E., & Lee, D. D. (2011). Australian Socially Responsible Funds: Performance, Risk and Screening Intensity. Journal of Business Ethics, 102(4), 519-535. doi:10.1007/s10551-011-0836-7 Humphrey, J. E., Warren, G. J., & Boon, J. (2016). What is Different About Socially Responsible

Funds? A Holdings-based Analysis. Journal of Business Ethics, 138, 263–277.

Jensen, M. C. (1967). The Performance of Mutual Funds in the Period 1945-1964. The Journal of

(27)

27

Kreander, N., Gray, G., Power, D.M., & Sinclair, C.D. (2005). Evaluating the Performance of Ethical and non-SRI funds: A Matched Pair Analysis. Journal of Business, Finance and Accounting, 32(7), 1465–1493.

Kurtz, L. (1997). No Effect, or No Net Effect? Studies on Socially Responsible Investing. The Journal

of Investing, 6(4), 37-49. doi:10.3905/joi.1997.37

Laurel, D. (2011). Socially Responsible Investments in Europe: The Effects of Screening on Risk and the Clusters in the Fund Space. SSRN Electronic Journal. doi:10.2139/ssrn.1883427

Lee, D. D., Humphrey, J. E., Benson, K. L., & Ahn, J. Y. K. (2010). Socially Responsible Investment Fund Performance: The Impact of Screening Intensity. Accounting & Finance, 50(2), 351–370. doi:10.1111/j.1467-629x.2009.00336.

Leite, C., Cortez, M. C., Silva, F., & Adcock, C. (2017). The Performance of Socially Responsible Equity Mutual Funds: Evidence from Sweden. Business Ethics: A European Review, 27(2), 108– 126. doi:10.1111/beer.12174

Leite, P., & Cortez, M. C. (2014). Style and Performance of International Socially Responsible Funds in Europe. Research in International Business and Finance, 30, 248–267. doi:10.1016/j.ribaf.2013.09.007

Luck, C. G., & Pilotte, N. (1993). Domini Social Index Performance. The Journal of Investing, 2(3), 60-62. doi:10.3905/joi.2.3.60

Luther, R. G., Matatko, J., & Corner, D. C. (1992). The Investment Performance of UK “Ethical” Unit Trusts. Accounting, Auditing & Accountability Journal, 5(4). doi:10.1108/09513579210019521 Mallin, C.A., Saadouni, B., & Briston, R.J. (1995). The Financial Performance of Ethical Investment

Funds. Journal of Business Finance and Accounting, 22(4), 483–496.

Markowitz, H. (1952). Portfolio Selection. The Journal of Finance, 7(1), 77-91. doi:10.2307/2975974 Mistry, J., & Shah, J. (2013). Dealing with the Limitations of the Sharpe Ratio for Portfolio

Evaluation. Journal of Commerce & Accounting Research, 2(3), 10-18.

Nofsinger, J., & Varma, A. (2014). Socially Responsible Funds and Market Crises. Journal of Banking

& Finance, 48, 180–193.

Otten, R., Bauer, R., & Koedijk, K. C. (2002). International Evidence on Ethical Mutual Fund Performance and Investment Style. SSRN Electronic Journal. doi:10.2139/ssrn.297882

Plyakha, Y., Uppal, R., & Vilkov, G. (2014). Equal or Value Weighting? Implications for Asset-Pricing Tests. SSRN Electronic Journal. doi:10.2139/ssrn.1787045

Renneboog, L., Horst, J. T., & Zhang, C. (2008). The Price of Ethics and Stakeholder Governance: The Performance of Socially Responsible Mutual Funds. Journal of Corporate Finance, 14(3), 302–322. doi:10.1016/j.jcorpfin.2008.03.009

Renneboog, L., Horst, J. T., & Zhang, C. (2007). Socially Responsible Investments: Institutional Aspects, Performance, and Investor Behavior. Journal of Banking & Finance, 32(9), 1723–1742. doi:10.1016/j.jbankfin.2007.12.039

Rudd, A. (1979). Divestment of South African Equities. The Journal of Portfolio Management, 5(3), 5-10. doi:10.3905/jpm.1979.5

Rudd, A. (1981). Social Responsibility and Portfolio Performance. California Management

(28)

28

Ruščáková, A., & Semančíková, J. (2016). The European Debt Crisis: A Brief Discussion of its Causes and Possible Solutions. Procedia - Social and Behavioral Sciences, 220, 399-406. doi:10.1016/j.sbspro.2016.05.514

Scholtens, B. (2005). Style and Performance of Dutch Socially Responsible Investment Funds. The

Journal of Investing, 14(1), 63–72. doi:10.3905/joi.2005.479390

Scholtens, B. (2014). Indicators of Responsible Investing. Ecological Indicators, 36, 382-385. doi:10.1016/j.ecolind.2013.08.012

Schröder, M. (2004). The Performance of Socially Responsible Investments: Investment Funds and Indices. Financial Markets and Portfolio Management, 18(2), 122-142. doi:10.1007/s11408-004-0202-1

Sharpe, W. F. (1994). The Sharpe Ratio. The Journal of Portfolio Management, 21(1), 49-58. doi:10.3905/jpm.1994.409501

Statman, M. (2000). Socially Responsible Mutual Funds. Financial Analysts Journal, 56(3), 30–39. Steinki, O., & Mohammad, Z. (2015). Common Metrics for Performance Evaluation: Overview of

Popular Performance Measurement Ratios. SSRN Electronic Journal. doi:10.2139/ssrn.2662054 Sparkes, R. (2008). Socially Responsible Investment. Handbook of Finance. 2(1).

doi:10.1002/9780470404324.hof002014

Tippet, J. (2001), Performance of Australia's ethical funds. Australian Economic Review, 34(2), 170-178. doi:10.1111/1467-8462.00186

Utz, S., & Wimmer, M. (2014). Are they any Good at all? A Financial and Ethical Analysis of Socially Responsible Mutual Funds. Journal of Asset Management, 15, 72–82. doi:10.1057/jam.2014.8 Walley, N., Whitehead, B. (1994). It’s Not Easy Being Green. Harvard Business Review, 72(3), 46. Wallis, M. V., & Klein, C. (2014). Ethical Requirement and Financial Interest: A Literature Review on

Socially Responsible Investing Fund Performance: Evaluation and Comparison to Conventional Funds in Luxembourg

1