BSc thesis:
Socially Responsible Investment:
a responsible choice during the financial crisis?
Name: Elvie Kromwijk
Student number: 6057284
Table of content
1. Introduction 2
2. Theoretical framework 4
2.1. Performance of SRI funds 4
2.2. Financial crisis 5
2.3. Four-factor CAPM model 6
3. Methods 7
3.1. Give Socially Responsible Investment funds a higher return than the market index
during the financial crisis? 7
3.1.1. T-test for comparing two means 7
3.1.2. Data 8
3.2. Are Conventional Investment funds better predictable by the four factor CAPM
model than Socially Responsible Investment funds? 9
3.2.1. Four-factor CAPM model with SRI dummy 9
3.2.2. Data 9
3.2.3. Regression 10
4. Results 12
4.1. Give Socially Responsible Investment funds a higher return than the market index
during the financial crisis? 12
4.2. Are Conventional Investment funds better predictable by the four factor CAPM
model than Socially Responsible Investment funds? 13
5. Conclusion 15
6. Discussion 15
7. References 17
Abstract – This thesis provides an answer to the question: ‘To what extent do Dutch Socially
Responsible Investment funds (SRI) differ in performance and predictability from conventional investment funds (CIF) during the financial crisis?’. The average performance of both fund types during the financial crisis is compared through a t-test. While SRI funds have slightly outperformed the market index, this difference is not significant at a significance level of 5%. The predictability of both fund types during the financial crisis is compared by investigating an extended version of the four-factor CAPM model: a SRI dummy is added. Both fund types have performed worse than predicted and their predictability did not significantly differ. So Dutch SRI funds have not significantly differed from CIF in performance and predictability during the financial crisis.
1. Introduction
A growing amount of investors are concerned about the social and environmental effects of their investments and therefore decide to invest in Socially Responsible Investment funds (SRI funds) instead of in Conventional Investment funds (CIF) (Rennebooga, Horst and Zhang, 2011). SRI is an investment process that integrates social, environmental, and ethical considerations into investment decision making (Renneboog, Horst & Zhang, 2008, p. 1723), while conventional investment funds do not integrate those aspects. The first SRI funds started in the early 1990s and the first Dutch SRI fund ‘ASN duurzaam aandelenfonds’ was founded on April 20th 1993 (Morningstar, 2013).
Performance of SRI funds have already been compared with CIF in foreign countries (Galema, Plantinga & Scholtens, 2008; Bauer, Koedijk & Otten, 2005; Jones, van der Laan, Frost and Loftus, 2007; Hamilton, Joe & Statman, 1993; Statman, 2000; Bauer, Otten & Tourani Rad, 2006; Bauer, Derwall & Otten, 2007). Furthermore Scholtens (2005) studies the performance and style of Dutch SRI funds in comparison with CIF for the period 2001-2003. Most of the researchers did not indicate a significant difference in performance between both investment funds types, however their results are contrary.
All of these studies are focused on a period before 2007, so a comparison between SRI funds and CIF during a financial crisis is not yet made. As Bartrama and Bodnarb (2009) argue the financial crisis did affect different sectors differently, this may also be the case for the different investment types. This thesis will amplify the lack in the existing knowledge by comparing both fund types on performance and predictability against the background of the financial crisis by focusing on the Dutch market over the period from 2007 to 2010. The Netherlands is one of the most financially developed countries in Europe and is furthermore among the European countries with the highest percentage of SRI in overall investments (Scholtens, 2005a). This thesis will provide an answer to the question: ‘To what extent do Dutch Socially Responsible Investment funds (SRI) differ in performance and predictability from conventional investment funds (CIF) during the financial crisis?’
This main question is separated into two sub questions followed from the hypothesis in the theoretical framework: ‘Do Socially Responsible Investment funds yield a higher return than Conventional Investment funds during the financial crisis?’ and ‘Are Conventional Investment funds better predictable by the four-factor CAPM model than Socially Responsible Investment funds.’ In order to answer sub question 1, four Dutch SRI fund returns are compared with the Dutch market index returns through a t-test on the mean returns. I find that SRI funds slightly outperform CIF during the financial crisis, however this result is not significant. Sub question 2 is answered by investigating an extended version the four-factor CAPM model: a SRI dummy is added. CIFs are
indeed better predictable by the four-factor CAPM model than SRI funds, however again this result is not significant.
This thesis will first provide an overview of results from earlier comparisons between SRI funds and CIF. Second the potential influence of the financial crisis on performance of both funds types will be explained. Furthermore methods for predicting and comparing funds returns will be discussed, followed by a data description and by an explanation of the empirical models used in this thesis. Then the results will be presented, followed by a conclusion on the performance and predictability of SRI funds during the financial crisis. Finally limitations of this thesis will be discussed and recommendations for future research will be provided in the discussion.
2. Theoretical framework
Friedman (1970) has argued that a company should only care about profits in the interests of their shareholders. However, investors who invest in SRI funds have a utility function which also increases from doing good, beside of from making money (Rennebooga, Horst and Zhang, 2008). So Friedman’s statement may not hold for all types of investors. In order to fulfill demand of all types of investors, SRI funds are founded. As Rennebooga, Horst and Zhang (2011) argue SRI investors are less concerned about negative returns than investors who invest in conventional funds. SRI investors expect companies to focus on social welfare in addition to value maximization (Rennebooga, Horst and Zhang, 2008). Therefore SRI funds may underperform Conventional Investment Funds (CIF) due to additional investment constraints.
2.1. Performance of SRI funds
The performance of SRI funds relative to Conventional Investment Funds (CIF) has been studied by multiple researchers, with contrary results. Bauer, Koedijk and Otten (2005) have found a significant outperformance of SRI funds in the UK for the period 1990 - 2001. However for this same period they have found a significant underperformance of SRI funds in the US. A study of Jones, van der Laan, Frost and Loftus (2007) about the Australian market also indicates a significant underperformance of SRI funds. They conclude that the Australian SRI funds annual underperform the CIF with 0.88% in the period 1986 – 2005. SRI funds’ relative performance was even worse in the latest years of their sample: in the period 2000-2005 SRI funds underperformed CIF with 1.52%.
However, most of the studies in which the performance of SRI funds and CIF are compared, do not indicate significant differences in performance. Hamilton, Joe & Statman (1993) have indicated that SRI funds in the US have insignificantly outperformed in the period 1981-1985 and insignificantly underperformed in the period 1986-1989. Contrary, Statman (2000) has indicated an insignificant outperformance of US SRI funds in the period 1990-1998, while Bauer, Koedijk and Otten (2005) have indicated an insignificant underperformance of the US SRI funds in almost the same period (1990-2001). The similarity is that they both indicated that SRI funds and CIF do not significantly differ from each other. Also Galema, Plantinga and Scholtens (2008) do not indicate a significant difference in performance between both investment funds types in the US during 1996-2006. Furthermore, Canadian SRI funds insignificantly outperformed CIFs in the period 1994-2002 (Bauer, Derwall & Otten, 2007). And Bauer, Otten & Tourani Rad (2006) have found an insignificant underperformance of SRI funds in Australia in the period 1992-2003, while they have found for the same period an insignificant international outperformance of SRI funds.
Scholten (2005) is the first paper which compared the performance of both investment funds types in the Netherlands. He has found an insignificant underperformance for the SRI funds in the period 2001-2003.
2.2. Financial crisis
The period of 2007-2010 is internationally recognized as one of the financial crisis. Even though the financial crisis affected all market types, it did not affect all markets with to the same extent. Bartrama and Bodnarb (2009) argue the financial crisis affect financial markets more than non-financial markets.
Without a financial crisis, on the background SRI funds are expected to perform worse than CIF, since investors who invest in socially responsible investment funds expect companies to focus on social welfare in addition to value maximization. For that reason SRI funds may perform worse than CIFs (Rennebooga, Horst and Zhang, 2008). However this might have been different during the financial crisis. During the financial crisis a less financial focus might be lucrative as Bartrama and Bodnarb (2009) argue. Since social, environmental and ethical aspects are non-financial, SRI funds are less financial orientated than CIFs. Since Bartrama and Bodnarb (2009) argue that the financial crisis affect financial markets more than non-financial markets, it might be the case that SRI funds performed better during the financial crisis. Therefore, the following hypothesis rises.
H1: Socially Responsible Investment funds give a higher return than the market index during the financial crisis.
2.3. Four-factor CAPM model
In order to analyze and predict the performance of funds, stocks and bonds, the Capital Asset Pricing Model (CAPM model) is developed. The fundamentals of the CAPM model are founded by Treynor in 1961, however he has never published the model in a scientific journal (French, 2003). The single factor CAPM model is first published by Sharpe (1964), Linter (1965) and Mossin (1966). The aim of the CAPM model is to predict the performance of funds, stocks and bonds. The single factor CAPM model contains a regression of individual funds, stocks and bond performances on the overall market performance. The factor market index captures market risks (Mossin, 1966).
Fama and French (1993) state that only market index return is insufficient for the prediction of funds, stocks and bonds returns. Therefore they have amplified the CAPM model with two other explanatory factors: size and book-to-market ratio. Both factors are related to economic fundamentals and therefore they are important for return predictions. Size contains the difference
in return between a small firm and a large firm. Small firms tend to have lower earnings on assets than large firms. The book-to-market ratio is the book value of a firm divided by the market value of a firm and is often used to indicate the under- or overvaluation of a firm. This amplified model is well known as the three-factor CAPM model.
While Fama and French (1993) expand the CAPM model to the three-factor CAPM model, Carhart (1997) has argued that the three-factor CAPM model is still incomplete. Carhart (1997) has amplified the three factor model into the four-factor CAPM model by adding the fourth explanatory factor ‘momentum’. The momentum can be calculated by subtracting the equal weighted average of the highest performing firms from the equal weighed average of the lowest performing firms.
Since Carhart (1997) has focused on mainstream CIF with the construction of the four-factor CAPM model, its predictability for SRI fund returns are unknown. Since the market index factor controls for conjuncture fluctuations as crises, the financial crisis might not influence the potential differences in predictability of both investment fund types. As a result, I present my second hypothesis as follows.
H2: Conventional Investment funds are better predictable by the four-factor CAPM model than Socially Responsible Investment funds.
3. Data and empirical model
In order to provide an answer to the question ‘To what extent do Dutch Socially Responsible Investment funds (SRI) differ in performance and predictability from conventional investment funds (CIF) during the financial crisis?’, a data analysis will be done. The main question will be separated into two sub questions followed from hypothesis 1 and 2: ‘Give Socially Responsible Investment funds a higher return than the market index during the financial crisis?’ and ‘Are Conventional Investment funds better predictable by the four-factor CAPM model than Socially Responsible Investment funds?’ . Both sub questions will be discussed separately.
3.1. Do Socially Responsible Investment funds yield a higher return than the market index during the financial crisis?
3.1.1. T-test for comparing two means
In order to show if SRI funds generate a relatively high return during the financial crisis, the average returns of both fund types should be compared. Since both fund returns have unequal variances and are independent from each other, the desirable test statistic is the t-test for comparing two means. Hypothesis 1 (H1) ‘Socially Responsible Investment funds give a higher return than the market index
during the financial crisis’ can be mathematically formulated as 𝜇𝜇1− 𝜇𝜇2> 0. Where 𝜇𝜇1 is the SRI
funds mean return and 𝜇𝜇2 is the mean market return. If H1 is correct, the zero hypothesis (H0:
𝜇𝜇1− 𝜇𝜇2= 0 ) should be rejected.
In order to test H1 the t-statistic should be calculated and be compared with the critical value of the test statistic. The t-statistic can be calculated as the following equation:
𝑡𝑡 = (𝑥𝑥̅1− 𝑥𝑥̅2)
�𝑠𝑠12
𝑛𝑛1 +𝑛𝑛2𝑠𝑠22
~ 𝑡𝑡[𝑑𝑑𝑑𝑑]
𝑥𝑥̅1 is the mean return of the four SRI funds (described under ‘3.1.2. data’) in the period 2008 – 2010,
𝑥𝑥̅2 is the mean return of the market index in the period 2008-2010, 𝑠𝑠12 is the variance of the SRI
funds return, 𝑠𝑠22 is the variance of the market mean return, 𝑛𝑛1is the number of observations for SRI
returns and 𝑛𝑛2 is the number of observations of market return.
The critical value should be calculated with help of the degrees of freedom (df), calculated with the following equation: 𝑑𝑑𝑑𝑑 = ( 𝑛𝑛1𝑠𝑠12+𝑛𝑛2𝑠𝑠22 )2
(𝑛𝑛1𝑠𝑠12)2 / (𝑛𝑛1−1) + (𝑠𝑠22
𝑛𝑛2)2 / (𝑛𝑛2−1)
. The associated value of the degrees of freedom and the significance level of 5% is 1.645.
The calculated t-statistic should be compared with the critical value. If the t-statistic > 1.645 H0 should be rejected and SRI funds did indeed outperform CIF during the financial crisis for a significance level of 5%. If -1.645 < t-statistic < 1.645 both fund types do not significantly differ in performance. And if t-statistic < -1.645, SRI funds significantly underperform the market index.
3.1.2. Data
My focus is on the Dutch financial market in the period 2008-2010. So four SRI funds which have been active in that period are randomly selected from the website www.morningstar.nl. The selected funds are: ASN duurzaam aandelenfonds, Robeco duurzaam aandelen, Allianz duurzaam wereldfonds and ING duurzaam aandelen fonds. For the market return, the AEX-index is selected, which contains all shares of the Dutch financial market. See figure 1 for key characteristics of these funds.
Funds Start Average return
(2008-2010)
Standard deviation (2008-2010)
ASN duurzaam aandelenfonds 24-03-1993 0.16% 10.15
Robeco duurzaam aandelen 01-02-1999 -0.20% 11.55
Allianz duurzaam wereld fonds 15-03-2004 0.03% 10.39
ING duurzaam aandelen fonds 03-03-2000 0.13% 10.95
AEX-index 10-07-1998 -1.05% 15.67
Table 1: key characteristics Dutch SRI funds and AEX-index
For the t-test statistic the mean of the quarterly returns of the selected SRI funds in the period 2008 – 2010 will be used as 𝑥𝑥̅1. The squared standard deviation of these quarterly returns is 𝑠𝑠12. For 𝑥𝑥̅2 the
mean of the quarterly returns of the AEX-index in the period 2008 – 2010 will be used and the square standard deviation of these quarterly returns is 𝑠𝑠22. The values of these variables are
calculated with help of excel and listed in table 2.
variables SRI (1) AEX-index (2)
Mean return ( 𝒙𝒙� ) 0,029583333 -1,054166667
Squared standard deviation ( 𝒔𝒔𝟐𝟐) 108,6880423 245,4732085
Number of observations (n) 48 12
Table 2: variables t-test
3.2. Are Conventional Investment funds better predictable by the four-factor CAPM model than Socially Responsible Investment funds?
3.2.1. Four-factor CAPM model with SRI dummy
The four-factor CAPM model is the most frequently used model for indicating the predictability of funds returns (Carhart, 1997). So also in this study the predictability of SRI funds and CIF will be determined with help of the four-factor CAPM model. This comprehensive model predicts fund performance with help of four factors. The equation of the four-factor CAPM model is: 𝑹𝑹𝒊𝒊𝒊𝒊− 𝑹𝑹𝒇𝒇𝒊𝒊=
𝜶𝜶𝒊𝒊+ 𝜷𝜷𝒊𝒊�𝑹𝑹𝒎𝒎𝒊𝒊− 𝑹𝑹𝒇𝒇𝒊𝒊� + 𝜷𝜷𝟏𝟏𝒊𝒊𝑺𝑺𝑺𝑺𝑺𝑺𝒊𝒊+ 𝜷𝜷𝟐𝟐𝒊𝒊𝑯𝑯𝑺𝑺𝑯𝑯𝒊𝒊+ 𝜷𝜷𝟑𝟑𝒊𝒊𝑺𝑺𝑴𝑴𝑺𝑺𝒊𝒊+ 𝜺𝜺𝒊𝒊𝒊𝒊 . where Rit is the return on SRI
fund i in year t; Rft is the return on risk free bills (Latest ten-year Dutch government bond); α is the
constant (unpredictable part of the model); Rmt is the return on the market index in year t
(AEX-index); SMBt is the difference in return between a small cap portfolio and a large cap portfolio at
time t; HMLt is the difference in return at time t between a portfolio containing value stocks (with a
high book-to-market ratio) and one consisting of growth stocks (with a low book-to-market ratio); Momt is the difference in return between a portfolio of past 12 months winners and a portfolio of
past 12 month losers at time t and ε is the error term (Carhart, 1997). αi indicates how much of the
funds returns are not explained by the model.
Since the four-factor CAPM model is not useful to make a direct comparison between the predictability of two funds types, an extension to the model is required. The four-factor CAPM model can be extended with a dummy variable for SRI. This dummy has value 1 for a SRI fund and 0 for a CIF. Adding the dummy variable for SRI to the four-factor CAPM model generates the following equation: 𝑹𝑹𝒊𝒊𝒊𝒊− 𝑹𝑹𝒇𝒇𝒊𝒊= 𝜶𝜶𝒊𝒊+ 𝜷𝜷𝟎𝟎𝒊𝒊�𝑹𝑹𝒎𝒎𝒊𝒊− 𝑹𝑹𝒇𝒇𝒊𝒊� + 𝜷𝜷𝟏𝟏𝒊𝒊𝑺𝑺𝑺𝑺𝑺𝑺𝒊𝒊+ 𝜷𝜷𝟐𝟐𝒊𝒊𝑯𝑯𝑺𝑺𝑯𝑯𝒊𝒊+ 𝜷𝜷𝟑𝟑𝒊𝒊𝑺𝑺𝑴𝑴𝑺𝑺𝒊𝒊+ 𝜷𝜷𝟒𝟒𝒊𝒊𝑺𝑺𝑹𝑹𝑺𝑺 + 𝜺𝜺𝒊𝒊𝒊𝒊.
The value of β4i can be distinguished as the difference in predictability of returns between SRI funds
and CIF by the four-factor CAPM model. Because for CIF only αi indicate the lack in explanatory
returns by the four-factor CAPM model, while for SRI funds β4i together with αi indicate the lack in
explanatory returns by the four-factor CAPM model. So if β4i and αi have both a positive or both a
negative value, SRI funds are worse predictable by the four-factor CAPM model. However, if β4i and
αi have contrary values, SRI funds are better predictable by the four-factor CAPM model. So if
hypothesis 2 is correct and CIF are better predictable by the four-factor CAPM model than SRI funds, β4i and αi should be both significant positive or both significant negative.
3.2.2. Data
Again is focused on the Dutch financial market in the period 2008-2010. The same four SRI funds as in sub question 1 are used, together with two random selected CIF from the website
www.morningstar.nl. The selected SRI funds are again ASN duurzaam aandelenfonds, Robeco duurzaam aandelen, Allianz duurzaam wereldfonds and ING duurzaam aandelen fonds. And the selected CIFs are: Rolinco and SNS Wereld Aandelenfonds. The quarterly returns of these funds will be used. See figure 1 for key characteristics of these funds.
Funds Start Average return
(2008-2010)
Standard deviation (2008-2010)
ASN duurzaam aandelenfonds 24-03-1993 0.16% 10.15
Robeco duurzaam aandelen 01-02-1999 -0.20% 11.55
Allianz duurzaam wereld fonds 15-03-2004 0.03% 10.39
ING duurzaam aandelen fonds 03-03-2000 0.13% 10.95
Rolinco 02-06-1965 0.13% 12.98
SNS Wereld Aandelenfond 22-11-1999 0.17% 11.31
AEX-index 10-07-1998 -1.05% 15.67
Table 3: key characteristics of Dutch SRI funds, CIF and AEX-index
For the market return is again the AEX-index selected, which is the largest index of the Dutch financial market. For the returns on risk free bills are the quarterly figures of the latest ten-year Dutch government bond used. And data for SMB, HML and MOM of the Dutch financial market are provided by Milius (2012).
3.2.3. Regression
The four-factor CAPM model will be regressed with help of the program StataSE 12.
First the predictability of all individual funds will be measured. Seven new variables should be generated in StataSE 12: all funds minus the risk free bonds (ASN, ROB, ALL, ING, ROL and SNS) and one for the market index return minus the risk free bills (MKT). Then the new created variables (ASN, ROB, ALL, ING, ROL and SNS) will be regressed on MKT, SMB, HML and MOM. There will be tested if αi significantly differs from zero with a significance level of 5%.
This regression will also be done for all funds together. A scale advance will turn in and therefore the reliability of the regression will improve. All funds (SRI funds and CIF) will be taken together in the new variable FUNDS. The variable FUNDS will be regressed on MKT, SMB, HML and MOM. Again there will be tested if αi significantly differs from zero with a significance level of 5%.
Those two regressions indicate if funds returns in general were predictable by the four-factor CAPM model during the financial crisis.
In order to compare the predictability of SRI funds and CIF the extended version of the four-factor CAPM model with the dummy variable for SRI will be used. SRI funds will get the value 1 for SRI and CIF will get the value 0 for SRI. All funds are again together shared under FUNDS and will this time be regressed on MKT, SMB, HML, MOM and SRI. Investigated should be if SRI funds differ in predictability of CIF by look at β4i (the value of the SRI dummy). If αi and β4i have both a significant
positive / both a significant negative value on the significance level of 5%, than CIF are better predictable by the four-factor CAPM model during the financial crisis. Because in that case the unpredictable returns by the four-factor CAPM model of SRI funds (αi + β4i ) differs more from 0 than
the unpredictable returns by the four-factor CAPM model of CIF (αi).
4. Results
4.1. Do Socially Responsible Investment funds yield a higher return than the market index during the financial crisis?
As graphically shown in figure 1, SRI funds have a smaller standard deviation than the market index. In 2008 and the second part of 2009 the returns of SRI funds decreased less than the returns of the AEX. However, in the first part of 2009 and 2010 the SRI funds returns increased less than the AEX returns.
Figure 1: SRI funds performance against AEX-index
Over the whole period 2008-2010 the SRI funds has generated a higher average return than the market index. See table 3.
Funds Average return
(2008-2010)
Standard deviation (2008-2010)
SRI-funds 0.03 % 10.43
AEX-index -1.05 % 15.67
Table 4: SRI funds and AEX-index performance
However, the difference in mean return is not significant. A t-test for comparing two means using independent samples when the populations have unequal variances gives the value t=0.22736, with
-30 -20 -10 0 10 20 30 2008 2009 2010 qu art erl y re tu rn s
SRI funds performance against
AEX-index
AEX-index SRI funds
df = 270,5797. The belonging critical value is 1.645. Since t < 1.645, both mean returns do not significantly differ on a significance level of 5%.
So hypothesis 1 ‘Socially Responsible Investment funds give a higher return than Conventional Investment funds during the financial crisis.’ should be rejected.
4.2. Are Conventional Investment funds better predictable by the four-factor CAPM model than Socially Responsible Investment funds?
First the predictability by the four-factor CAPM model of all individual SRI funds and CIF are indicated. The equation of the four-factor CAPM model is 𝑹𝑹𝒊𝒊𝒊𝒊− 𝑹𝑹𝒇𝒇𝒊𝒊= 𝜶𝜶𝒊𝒊+ 𝜷𝜷𝟎𝟎𝒊𝒊�𝑹𝑹𝒎𝒎𝒊𝒊− 𝑹𝑹𝒇𝒇𝒊𝒊� +
𝜷𝜷𝟏𝟏𝒊𝒊𝑺𝑺𝑺𝑺𝑺𝑺𝒊𝒊+ 𝜷𝜷𝟐𝟐𝒊𝒊𝑯𝑯𝑺𝑺𝑯𝑯𝒊𝒊+ 𝜷𝜷𝟑𝟑𝒊𝒊𝑺𝑺𝑴𝑴𝑺𝑺𝒊𝒊+ 𝜺𝜺𝒊𝒊𝒊𝒊 . As summarized in table 5, all funds perform worse than
predicted by the four-factor CAPM model, since all values of α are negative (-3.76; -3.04; -4.83; -2.60; -2.48; -3.46). However, only the α of Allianz duurzaam wereldfonds is significant for the significance level of 5%.
α Β0 Β1 Β2 Β3 R2adj F-value
ASN duurzaam aandelenfonds -3.76* (0.090) 0.61*** (0.000) 4.29* (0.064) -0.61 (0.410) 0.29 (0.386) 0.85 182.83
Robeco duurzaam aandelen -3.04* (0.089) 0.72*** (0.000) 3.5* (0.061) -0.66 (0.283) 0.20 (0.453) 0.92 33.46
Allianz duurzaam wereldfonds -4.83** (0.044) 0.62*** (0.000) 5.75** (0.024) -0.97 (0.223) 0.44 (0.221) 0.84 15.71
ING duurzaam aandelen fonds -2.60 (0.166) 0.69*** (0.000) 3.18 (0.106) -0.80 (0.237) 0.18 (0.546) 0.90 24.88 Rolinco -2.48 (0.217) 0.88*** (0.000) 4.59** (0.043) -1.40 (0.074) 0.52 (0.128) 0.91 29.83 SNS Wereld Aandelenfonds -3.46* (0.066) 0.69*** (0.000) 4.25** (0.034) -0.91 (0.161) 0.14 (0.603) 0.91 30.05
Table 5: Results four-factor CAPM model all funds separated * value in brackets is the P-value
All funds together regressed on the factors of the four-factor CAPM model give the same result. Still all funds perform worse than predicted by the four-factor CAPM model during the financial crisis. As summarized in table 6, the value of α is negative (-3.36) and significant for the significance level of 5% (P-value = 0.000). However all estimators of the funds returns are significant at the significance level of 5% and furthermore an F-value of 182.83 and R2
adj of 0.91 indicate that the model in total is also
significant.
In order to compare the predictability of both funds types the extended four-factor CAPM model by the SRI dummy should be used. The equation for this model is: 𝑹𝑹𝒊𝒊𝒊𝒊− 𝑹𝑹𝒇𝒇𝒊𝒊= 𝜶𝜶𝒊𝒊+ 𝜷𝜷𝟎𝟎𝒊𝒊�𝑹𝑹𝒎𝒎𝒊𝒊− 𝑹𝑹𝒇𝒇𝒊𝒊� +
𝜷𝜷𝟏𝟏𝒊𝒊𝑺𝑺𝑺𝑺𝑺𝑺𝒊𝒊+ 𝜷𝜷𝟐𝟐𝒊𝒊𝑯𝑯𝑺𝑺𝑯𝑯𝒊𝒊+ 𝜷𝜷𝟑𝟑𝒊𝒊𝑺𝑺𝑴𝑴𝑺𝑺𝒊𝒊+ 𝜷𝜷𝟒𝟒𝒊𝒊𝑺𝑺𝑹𝑹𝑺𝑺 + 𝜺𝜺𝒊𝒊𝒊𝒊 . The difference between both funds types in
predictability by the four-factor CAPM model is indicated by the value of the SRI dummy: β4. As
summarized in table 7, the values of αand Β4 are both negative (-3.28; -0.12). This indicates that SRI
firms are slightly worse predictable by the four-factor CAPM model than CIF. However, the p-value of β4i is 0.885, which indicates that the difference is not significant at a significance level of 5%.
So there is no significance evidence for hypothesis 2: ‘Conventional Investment funds are better predictable by the four-factor CAPM model than Socially Responsible Investment funds.’ Hypothesis 2 should be rejected. α Β0 Β1 Β2 Β3 R2adj F-value Funds (2008-2010) -3.36*** (0.000) 0.70*** (0.000) 4.26*** (0.000) -0.89*** (0.000) 0.29*** (0.006) 0.91 182.83
Table 6: Results four-factor CAPM model all funds together * value in brackets is the P-value
α Β0 Β1 Β2 Β3 Β4 R2adj F-value Funds (2008-2010) -3.28*** (0.000) 0.70*** (0.000) 4.26*** (0.000) -0.89*** (0.000) 0.30*** (0.007) -0.12 (0.885) 0.91 144.13
Table 7: Results four-factor CAPM model + dummy SRI * value in brackets is the P-value
5. Conclusion
To conclude Dutch Socially Responsible Investment funds (SRI) do not differ significantly in performance and predictability from conventional investment funds (CIF) during the financial crisis. Even though this study indicates that SRI funds performance has been slightly better than the market index during the financial crisis and CIFs returns has been better predictable than SRI funds by the four-factor CAPM model, these results do not provide enough evidence for a difference in performance and predictability, because these results are not significant at a significance level of 5%. So Investing in SRI funds has not been a more responsible choice than investing in CIF during the financial crisis. However, it is also not a worse decision. So investors who want to be sure that their money will be used for socially responsible goals, can invest in SRI funds during a financial crisis without deliver up returns and predictability.
6. Discussion
This study has amplified the lack in the existing knowledge about SRI funds performance and predictability, by analyzing the Dutch SRI funds performance and predictability during the financial crisis. Scholten (2005) did already do a comparable study in the Netherlands for the period 2001-2003. While Scholten (2005) indicated an insignificant underperformance of SRI funds, this study indicated an insignificant outperformance of SRI funds. And furthermore SRI funds have been insignificantly worse predictable by the four-factor CAPM model in both periods. This suggests that the financial crisis may have affected both investment types’ returns differently. However, clear evidence for this suggestion is not found in this thesis. A comparison between relative performance in both periods cannot be made, since the data which Scholten (2005) has used was not available for this study. Furthermore, there is no clear evidence for the causality between the financial crisis and the improvement in performance of SRI funds in 2008-2010 relative to 2001-2003, since the ceteris paribus assumption did not hold. In order to investigate if the financial crisis affects both investment funds types differently, recommended is a comprehensive comparison.
Furthermore this study has some limitations. In the first place is the year 2007 not included in this study, while the year 2007 is also part of the financial crisis (2007-2010). The reason for excluding 2007 is that data of the quarterly returns of the investment funds was not available on the internet. And an employer of the financial website www.morningstar.nl has not been willing to provide date for this study. He could only provide information about 2007 to paying customers. Also the individual investment funds were not willing to provide data of 2007. The only possibility left was contacting investors and ask them for the quarterly returns of the year 2007. This was unfortunately
out of the scope of this research. So a comparable study with 2007 included will be recommended for future research.
Secondly not all Dutch SRI funds and CIF have been included in this study. In order to increase the sample size en thereby the reliability of the research, a larger sample should be used. This will be recommended for future research.
Another suggestion for future research is to compare SRI funds with market index and CIF during the financial crisis and during another period in foreign countries. Results of both periods can be compared in order to investigate if the financial crisis did affect both investment fund types differently. Since causality will be still difficult measurable, there should be controlled for other factors as technological improvement etcetera.
7. References
Bartrama, S.M. and Bodnarb, G.M. (2009). No place to hide: The global crisis in equity markets in 2008/2009. Journal of International Money and Finance, 28 (8), 1246–1292.
Bauer, B, Koedijk, K and Otten, R. (2005). International evidence on ethical mutual fund performance and investment style. Journal of Banking & finance, 29 (7), 1751-1767.
Bauer, R., Otten, R. and Tourani Rad, A. (2006). Ethical investing in Australia: Is there a financial penalty? Pacific-Basin Finance Journal, 14 (1), 33–48.
Bauer, R., Derwall, J. and Otten, R. (2007). The ethical mutual funds performance debate: New evidence for Canada. Journal of Business Ethics, 70 (2), 111–124.
Carhart, M.M. (1997). On Persistence in Mutual Fund Performance. The Journal of Finance, 52 (1), 57-82.
Fama, E.F. (1993). Common risk factors in the returns on stocks and bonds. Journal of Financial Economics, 33, 3-56.
French, C.W. (2003). The Treynor capital asset pricing model. Journal of Investment Management, 1 (2), 60-72.
Friedman, M. (1970). The Social Responsibility of Business is to Increase its Profits. Corporate Ethics and Corporate Governance, 173-178.
Galema, R., Plantinga, A. and Scholtens, B. (2008). The stocks at stake: Return and risk in socially responsible investment. Journal of Banking & Finance, 32 (12), 2646–2654.
Hamilton, S., Joe, H., and Statman, M., (1993). Doing well while doing good? The investment performance of socially responsible mutual funds. Financial Analysts Journal, 49 (6), 62–66.
Jones, S., van der Laan, S., Frost, G. and Loftus, J. (2007). The Investment Performance of Socially Responsible Investment Funds in Australia. Journal of Business Ethics, 80, 181–203.
Linter, J. (1965). The valuation of risk assets and the selection of risky investments in stock portfolios and capital budgets. The Review of Economics and Statistics. 47 (1), 13-37.
Milius, L. (2012). Msc Thesis: The Three Factor Model for Dutch Equities. University of Tilburg. Morningstar (2013). ASN Duurzaam Aandelenfonds Inc. Uploaded on Dec 12, 2013. From: http://www.morningstar.nl/nl/funds/snapshot/snapshot.aspx?id=F0GBR04B25.
Mossin, J. (1966). Equilibrium in a capital asset market. EconometricaI, 34 (4), 768-783. Rennebooga, L., Horst, J. ter, and Zhang, C. (2008). Socially responsible investments: Institutional aspects, performance, and investor behavior. Journal of Banking & Finance, 32 (9), 1723–1742.
Rennebooga, L., Horst, J. ter, and Zhang, C. (2011). Is ethical money financially smart? Nonfinancial attributes and money flows of socially responsible investment funds. Journal of Financial Intermediation, 20 (4),562–588.
Scholtens, B. (2005). Style and Performance of Dutch Socially Responsible Investment Funds. Journal of Investing, 63 – 72.
Sharpe, W.F. (1964). Capital Asset Prices: A theory of market equilibrium under conditions of risk. Journal of Finance, 19 (3), 425-442.
Statman, M. (2000). Socially responsible mutual funds. Financial Analysts Journal , 56 (3), 30–39.
