Social Preferences and Portfolio Choice
Arno Riedl
Paul Smeets
October, 2014
Abstract
We empirically investigate the effect of social preferences on portfolio choice. We use
administrative investor data and link them to behavior in controlled experiments and survey
responses. We find that social preferences systematically affect portfolio choice. Specifically,
social preferences rather than (biased) risk-return expectations are predictive for investment in
socially responsible mutual funds without tax benefits; but not with tax benefits. Our results also
suggest that selfish investors hold socially responsible mutual funds without tax benefits for
signaling reasons. Together with the growth in socially responsible investments this indicates
long-run effects of social preferences and social signaling on stock prices.
Keywords: portfolio choice, social preferences, socially responsible investments, mutual funds,
administrative investor data, experiment, survey
JEL Classification: G11, D64, C90
Arno Riedl (corresponding author): CESifo, IZA, Netspar, Department of Economics (AE1), School of Business and Economics,
Maastricht University, PO Box 616, 6200 MD Maastricht, The Netherlands (e-mail: [email protected]); Paul Smeets:
Department of Finance (LIFE) and European Centre for Corporate Engagement, School of Business and Economics , Maastricht
University, PO Box 616, 6200 MD Maastricht, The Netherlands (e-mail: [email protected]). A former version of
this paper was previously circulated with the title “Socially responsible investments: Return Expectations or social preferences?”.
We are grateful to Robeco for providing us with the administrative data used in this paper and we particularly thank Peter Jurriaans,
Catrien Kleinheerenbrink, Manon Middelink and Jorg Sunderman. This paper benefited especially from the comments and
suggestions of two anonymous referees, the editor (Kenneth J. Singleton), Clifton Green, Chris Parsons and Nicolas Salamanca. We
are also grateful to the valuable comments of Rob Bauer, John Beshears, Thomas Dohmen, Piet Eichholtz, Uri Gneezy, Arvid
Hoffmann, Christine Kaufmann, Stephan Meier, Thomas Post (Netspar discussant), Sebastien Pouget, Walid Saffar (FMA
discussant), Tao Shu (EFA discussant) and Leonard Wolk. We thank seminar and conference participants at FMA 2014 in
Maastricht, EFA 2013 in Cambridge, Society for Experimental Finance in Zurich, the Toulouse School of Economics, UC San
Diego Rady School of Management, UC San Diego Applied Microeconomics, the FMA meetings (Chicago and Maastricht), 12 th
Colloquium on Financial Markets (Cologne), Sorbonne University (MISTRA workshop), EEA 2012 (Malaga), ESA 2012 (New
York), Netspar annual conference 2012 (Amsterdam), MBEES 2012 (Maastricht). We also thank Philip Abele, Oana Floroiu, John
Kramer, Mohammedreza Maghroor, Tobias Ruof, Simone Vermeend and Thorsten Voss for their help as research assistants. Paul
Smeets received financial support from MISTRA and the European Centre for Corporate Engagement (ECCE). Part of this paper
was written while Paul Smeets was visiting the Rady School of Management (UC San Diego).
1. Introduction
Socially responsible investors are a conundrum in finance because they deviate from the
market by excluding potentially high-return ‘sin’ companies from their portfolio or by
focusing on companies with potentially lower returns but good environmental policies,
respect of human rights, employee relations, and so forth (Social Investment Forum
(SIF) (2012)). Paradoxically, it also appears that this deviation from the market is not
vanishing but rather increasing. In Europe socially responsible investments (henceforth
SRI) are growing in volume (EUROSIF (2012)) and in the United States approximately
one in nine dollars of professionally managed assets are involved in socially responsible
investments (SIF (2012)).
Although a few papers show that SRI may sometimes perform financially better
or not worse than conventional investments (Bauer, Otten and Koedijk (2005), Derwall,
et al. (2005), Kempf and Osthoff (2007), Edmans (2011)) it is the prevailing opinion in
finance that investing in a socially responsible manner is financially costly. For instance,
Fabozzi, Ma and Oliphant (2008) and Hong and Kacperczyk (2009) find that divesting
from ‘sin’ industries that involve weapons, tobacco, alcohol or gambling is costly
because these companies tend to perform better than ‘non-sin’ companies. Moreover,
Krüger (forthcoming) finds that stock prices sometimes react negatively to positive
news of a company's corporate social responsibility (CSR). Also Renneboog, Ter Horst
and Zhang (2008) find that SRI equity funds perform worse than conventional equity
funds.
This indicates that the observed deviation from the market portfolio cannot be
explained on the basis of financial performance alone. Hence, other explanations like
biased return expectations or wrong risk perceptions may lay at the basis of SRI.
Alternatively, and more fundamentally, SRI decisions may be driven by investor
preferences other than financial returns. On financial markets, however, biases in riskreturn expectations as well as not return based preferences may be driven out of the
market by competitive forces. Therefore, the question whether these factors can explain
the phenomenon of SRI is ultimately an empirical one.
1
In this paper we investigate whether independently elicited social preferences of
individual investors can explain SRI or if such investment decisions are merely driven
by (perhaps biased) risk-return perceptions. In addition, we probe further details of SRI
patterns, like how they relate to investors' risk preferences, preferences for (or aversion
to) equity, value and size of the portfolio, investment knowledge, tax incentives and
more.
In contrast to conventional mutual funds, socially responsible mutual funds have
a focus on broader societal issues, which suggests that other-regarding motives could
play a role in SRI. There is some first albeit indirect evidence that this might indeed be
the case. Bollen (2007) finds that ex post investors are more likely to hold on to bad
performing SRI funds than to hold on to bad performing conventional funds. Hong and
Kostovetsky (2012) report that Democratic fund managers select stocks that score
higher on social responsibility than stocks selected by Republican fund managers.
Importantly, these studies do neither directly measure social preferences nor elicit riskreturn perceptions. It is, therefore, unclear whether investors' behavior is indeed
influenced by their social preferences, by (biased) expectations and perceptions or other
unobserved investor characteristics.
There is mounting evidence from the laboratory and the field that people do care
about the well-being of others and that this affects their behavior (see, e.g., Ledyard
(1995), Fehr and Gächter (2000), Karlan (2005), Egas and Riedl (2008), Falk and
Heckman (2009)).1 Such social (aka other-regarding) preferences constitute a profound
deviation from the standard neoclassical homo economicus assumption. Despite the
cited evidence the standard neoclassical assumption is still prevalent in the finance
literature as well as most of the economics literature. A much used argument against the
importance of social preferences in finance and economics is that in the market place
they will be driven out by competitive forces (Levitt and List (2007), List (2009)).
Further, Dufwenberg et al. (2011) analyze the theoretical general equilibrium properties
of an economy under the assumption that social preferences exist. They argue that under
1 For theoretical approaches modeling such behavior see, among others, Andreoni (1990), Rabin (1993),
Fehr and Schmidt (1999), Bolton and Ockenfels (2000) (see Sobel (2005) for a critical discussion).
2
some conditions it may be impossible to identify social preferences from market
behavior. On the other hand, however, there are theoretical arguments supporting the
view that tastes for assets may also affect asset prices (e.g., Fama and French (2007)).
Hence, ex ante it is not clear whether on the financial market investors with social
preferences will indeed be able or willing to act according to their preferences,
especially when return expectations or risk perceptions disfavor SRI. With our study we
aim to provide an answer to this empirical question.
We directly measure social preferences and provide first evidence on whether
such preferences affect real portfolio choice by investing in a socially responsible
manner. Understanding the factors influencing such portfolio choice is important
because observed deviations from the market portfolio could affect stock prices in the
long run if they were caused by preferences (Hong and Kacperczyk (2009)). In contrast,
biased performance expectations would probably only generate short run effects on
asset prices, because potential mispricing of socially responsible companies should
disappear as investors learn over time (Bénabou and Tirole (2010), Derwall, Koedijk
and Ter Horst (2011)). Understanding the role of social preferences in investment
decisions is also important for mutual fund managers and pension funds. If investors
solely care about risk and return, fund managers should only invest in a socially
responsible manner if they expect at least equal risk-adjusted returns on socially
responsible companies. This is not the case, however, when investors are willing to
invest in a socially responsible manner even at the cost of lower expected returns.
In our paper, to assess the role of social preferences and other factors in portfolio
choice, we use a unique combination of three data sources. First, we gather
administrative data from a large mutual fund provider in the Netherlands; second, in
order to directly and independently measure social and risk preferences we conduct
incentivized artefactual field experiments (Harrison and List (2004)) with individual
investors; third, we collect additional investor information through a survey.
The mutual fund provider offers a large variety of socially responsible and
conventional mutual funds. Individual investors buy and sell their funds directly online
without the interference of an intermediary. The administrative data contain the monthly
3
portfolio holdings of retail investors, their returns, and basic demographics. The data
also comprise the total amount and proportion of the portfolio invested in SRI funds
with and without tax incentives as well as all investments in conventional mutual funds.
For our analysis we link the administrative data to sampled survey responses and
behavior in the experiments.
In order to investigate direct effects of social preferences on portfolio choice in a
clean way, it is necessary to have an independent measure of such preferences. To
explore the pure role of social preferences, this measure should also be affected as little
as possible by social reputation considerations and strategic fairness (Kreps, Milgrom,
Roberts, and Wilson (1982)) or social image concerns (Ellingsen and Johannesson
(2008)). In order to get such a measure of intrinsic social preferences we let investors
participate in a controlled and anonymous one-shot trust game experiment (Berg,
Dickhaut, and McCabe (1995), Fehr and List (2004), Karlan (2005), Falk, Meier and
Zehnder (2013)). The trust game is a two player sequential move game where the firstmover can transfer money to the second-mover. The transferred amount is tripled by the
experimenter. The second-mover can send back nothing, parts, or all of the received
money to the first-mover. It is the behavior of investors in the role of second-movers
which endows us with the measure of intrinsic social preferences. In a case where the
second-mover behaves like the prototypical homo economicus s/he should not send back
any money. The more an investor returns, the stronger are his or her intrinsic social
preferences.
In the experimental literature, social preferences are also often measured by a
dictator game. We use second-mover behavior in the trust game instead for two main
reasons. First, second-mover behavior in the trust game captures a broader definition of
social preferences as it can reflect outcome based fairness considerations, preferences
for reciprocity, or a combination of both (Cox (2004)).2 Second, it has been shown that
behavior in dictator games can be prone to framing and presentation effects, which has
put into question its usefulness as a mean of eliciting social preferences (List (2007),
2
This broad definition could be a disadvantage when one would be interested in disentangling different
forms of social preferences. Our research question is different and we are ignorant regarding the
precise nature of the elicited social preferences. We discuss this issue in more detail in Section 4.3.
4
Bardsley (2008)). Moreover, second-mover behavior in trust games has already been
shown to have predictive power for pro-social field behavior, in domains other than
financial markets (Karlan (2005), Baran, Sapienza and Zingales (2010)). We avoid
repeated game based reputation concerns by the one-shot nature of our experiment and
minimize potential other reputation and social image effects through the anonymity of
investors. Taken together, we are confident that the amount sent back by the secondmover in the one-shot trust game is indeed a good measure of intrinsic social
preferences.
Next to the trust game, investors also participate in a financially incentivized risk
preferences elicitation task (Holt and Laury (2002) and Dohmen et al. (2011)). This
gives us an independent measure of risk preferences and allows us to control for them
when examining the factors determining investors' portfolio choice.
The third data source of our empirical research strategy is survey data. Such data
are useful for eliciting factors that are otherwise difficult to gather (see, e.g., Guiso,
Sapienza, and Zingales (2013)). In identifying whether social preferences are an
important determinant for investing in a socially responsible manner we need to control
for individual differences in return expectations and risk perceptions regarding
conventional and socially responsible investments. As these variables are not available
in the form of administrative data on the individual level, we use survey questions to
gather them. In addition, we also use survey questions to get data on whether investors
talk about their investment activities and other control variables that are potentially
important for investment behavior, like investment knowledge, income level, education,
age, gender, etc.
Our results can be summarized as follows. We find clear and robust evidence
that intrinsic social preferences matter in portfolio choice and, thus, are not driven out in
competitive financial markets. Specifically, investors with stronger intrinsic social
preferences are significantly more likely to hold SRI (equity) funds without tax
incentives. Interestingly, this is not the case for SRI (bond) funds with tax incentives. In
addition, about half of the socially responsible investors expect lower ex ante returns on
SRI funds and, in contrast to social preferences, expectations about the returns of and
5
risk perceptions on SRI funds are unrelated to investments in SRI funds. Together this
supports a preference-based explanation for portfolio distortions rather than a beliefbased explanation.
We also see that investors who talk about their investment are those with weak
social preferences. At the same time we observe a relation between talking about
investment and investing in SRI funds without tax incentives. This suggests that there is
a group of rather selfish investors who hold SRI funds without tax incentives for social
reputation reasons.
Importantly, these results hold while controlling for other possible explanatory
variables such as a preference for equity, risk preferences and other investor
characteristics. Regarding these variables we find that the equity ratio of an investor's
portfolio is also predictive for holding SRI funds without tax benefits. Social
preferences itself are, however, unrelated with the share of equity in an investor's
portfolio. There is also a positive correlation between holding of SRI funds without tax
benefits and the total portfolio value of an investor whereas holding of SRI funds with
tax benefits is predicted by trading activity, portfolio returns, and investment
knowledge.
Our finding that intrinsic social preferences are related to investments in SRI
funds without tax incentives but not with such incentives resonates with the literature on
intrinsic and extrinsic motivation (Gneezy and Rustichini (2000 a,b), Bénabou and
Tirole (2006), Ariely, Bracha and Meier (2009)). It shows that SRI funds with and
without extrinsic incentives may attract different investor types. At the same time our
evidence also suggests that different investor types can hold SRI funds without tax
incentives for different reasons. Investors with strong social preferences hold them
because they value them intrinsically while some investors with weak social preferences
hold them for social reputation reasons. Together with the steady increase of SRI both
results suggest long-term price effects.
6
2. Related literature and hypotheses
In this section, we derive our hypotheses regarding potential main predictors of
portfolio choice in the form of SRI and review related literature. We first discuss
potential financial reasons to hold SRI funds, followed by a brief discussion of the
empirical literature regarding non-financial motives in SRI. Finally, we provide a brief
review of theories that discuss how social preferences may be reflected in market values
and why they may or may not be driven out by competitive forces.
2.1 Financial motives and SRI funds
According to modern portfolio theory, investors create portfolios based on an optimal
risk-return trade-off. This implies that investors should never hold SRI funds if it would
worsen the risk-return profile of their portfolio. Hence, given the increasing interest in
SRI (EUROSIF (2012), SIF (2012)) one should expect that SRI funds perform at least
as well as the market. A few papers indeed suggest that SRI funds do sometimes not
perform worse than conventional funds, but the received wisdom seems to be that SRI
funds underperform compared to other funds (Bauer, Otten and Koedijk (2005), Kempf
and Osthoff (2007), Fabozzi, Ma and Oliphant (2008), Renneboog, Ter Horst and Zhang
(2008), Hong and Kacperczyk (2009), Edmans (2011)). Therefore, it seems unlikely that
the growth in SRI can be explained by financial reasons alone. Yet, if investors differ in
their subjective (out-of-equilibrium) expectations it may still be that some investors
(incorrectly) expect SRI funds to financially outperform other funds. We will, therefore,
control for these expectations in our analysis.
An alternative financial motivation could be differential tax treatment. In the
Netherlands, where our data set is gathered, certain types of SRI funds indeed offer tax
benefits. In order to explore the role of tax breaks we will analyze the determinants of
investing in SRI funds with and without such preferential tax treatment.
2.2 Non-financial motives and SRI funds
Evidence on the potential effect of investors' preferences over non-material values on
investment decisions is only recent and scarce (e.g., Bollen (2007), Kaustia and Torstila
(2011), Hong and Kostovetsky (2012), Di Giuli and Kostovetsky (2014), Kumar and
7
Page (forthcoming)). The evidence provided by these studies also is only indirect
because preferences over immaterial values are not independently measured. Therefore,
it is still unclear whether investors deviate from holding the market portfolio because of
their preferences or because of some other (unobserved) variable.
In their important study, Hong and Kostovetsky (2012) show that political preferences
(Republicans and Democrats, respectively) of professional mutual fund managers
influence the stocks they select. However, this does not yet answer the question whether
social preferences of individual investors will affect their investment behavior. First,
individual investors trade for their own account, while professional managers trade with
the money of others. It is likely that (social or political) preferences differently affect
investment behavior when own or others' money is at stake (Anderson et al. (2013)).
Second, Hong and Kostovetsky (2012) focus on political preferences and their results
regarding social preferences are proximate. In contrast, we will measure social
preferences directly and independently with an incentivized experiment. Third, we
control for expectations about the returns and risk on SRI funds compared to
conventional funds and can disentangle a social preference-based explanation for SRI
from a belief-based explanation. Fourth, there is interesting work in the experimental
political science literature reporting that social preferences, measured as voluntary
giving, of Republicans and Democrats do on average not differ (Fowler and Kam
(2007)).
2.3 Social preferences and market forces
As laid out in the Introduction, the standard argument against the importance of social
preferences on (financial) markets is that they will be driven out by competitive forces
and, hence, will not affect (asset) prices. On the other hand, there are a few theory
papers that discuss in more detail how social preferences may or may not influence
market values and whether social preferences may indeed be driven out in the market
place.
In their seminal work Heinkel, Kraus and Zechner (2001) develop a model in
which some investors refrain from investing in non-responsible companies.
Consequently, non-responsible companies will have higher expected returns, because
8
their risk is borne by fewer investors. Graff Zivin and Small (2005) argue that some
individuals are willing to outsource philanthropy to corporations, because they might be
more efficient with their donations than individuals by reducing corporate taxes. Baron
(2007) also models socially responsible behavior by firms as donations. He shows that
the cost of social responsibility is borne by the social entrepreneur when going public
rather than by the shareholder as long as corporate social responsibility is anticipated by
shareholders. Fama and French (2007) theorize that asset prices may be influenced by a
combination of disagreement among investors on asset returns and tastes for assets.
Finally, Gollier and Pouget (2014) develop a model in which investors can improve
social responsibility of firms by excluding non-responsible companies from their
portfolio or by activism against non-responsible firms.
Our paper contributes to this discussion by providing empirical evidence for a
key assumption of these models: that social preferences indeed guide investment
decisions of (a subset of) investors. Specifically, we hypothesize that social preferences
are not driven out of the market and that investors with stronger social preferences are
more likely to invest in socially responsible mutual funds. In addition, it may be that
investors with purely financial motives could hold SRI funds for reasons of preferential
tax treatment and/or social reputation.
Moreover, it has been shown that extrinsic incentives sometimes can crowd out
intrinsic pro-social motivations (Gneezy and Rustichini (2000 a,b), Bénabou and Tirole
(2006)). Therefore, it may be that investors with strong social preferences prefer not to
invest in funds with tax benefits. Taking these effects into account we hypothesize that
social preferences are positively related to investments in SRI funds without tax benefits
but are unrelated (or at least weaker related) to investments in SRI funds with tax
benefits. Further, we may see that investors with weak social preferences invest in SRI
funds without tax benefits only if there is a possibility to use it as a signal of social
reputation.
As we will describe in more detail below (Section 3.1) SRI funds without tax
benefits are equity funds whereas SRI funds with tax benefits are bond funds.
9
Therefore, we will control for a preference of investors for equity funds and hypothesize
that social preferences are unrelated with a preference for equity.
3. Data and variables
In this section, we first describe the administrative investor data, followed by a
description of the survey and details on the experiments. Thereafter, we describe our
main variables.
3.1 Administrative investor data
We utilize administrative individual investor data from one of the largest mutual fund
providers in the Netherlands, covering the period January 2002 – August 2012. The
mutual fund provider offers a wide range of investment funds, including equity funds,
bond funds and mixed funds. Within these categories the funds can be global, sectorspecific, SRI funds, and so on.3 Especially important for our study is that the
administrative data show for each investor whether or not s/he holds a socially
responsible mutual fund, the shares invested in SRI funds, and all other funds, on a
monthly basis.
Moreover, for investors holding SRI funds we can distinguish between money
invested in SRI funds that offer tax benefits and SRI funds without tax benefits. The
former are SRI bond funds for which the Dutch government gave tax incentives that
could reach a maximum of 2.2% of the amount invested in the month relevant for the
study.4 The SRI funds without tax benefits are equity funds comparable to SRI equity
funds offered in the United States (SIF (2012)) and the rest of Europe (EUROSIF
3
Figure A1 in the appendix shows a screenshot of the product selector of the mutual fund provider.
The product selector shows for each fund to which category it belongs and whether the provider
classifies the fund as sustainable, emerging markets, global, etc. At the same screen, investors can read
about the details of the fund including the details regarding stock selections based on social
responsibility criteria. In addition, the product selector gives information such as past performance,
Morningstar ratings and fees.
4
If not stated otherwise all used administrative investor data refer to the month when the survey and
experiment were conducted (June 2011).
10
(2012)). SRI funds with tax benefits are defined by Dutch tax law, which is also the
definition we use. For SRI funds without tax benefits we use the classification of the
mutual fund provider of socially responsible and sustainable funds.5
3.2 Survey
The administrative data provide information on 3,382 socially responsible investors,
which were all invited to participate in the survey. Next to the socially responsible
investors, we randomly selected about 35,000 investors of the approximately 145,000
remaining accounts in the database.6 All selected investors received an email containing
a link to the online survey. The response rate was 8% for conventional investors and
12% for socially responsible investors (see Table A1 in the appendix for a comparison
of the two samples regarding several important variables contained in the administrative
data). We deliberately invited disproportionately more socially responsible investors, in
order to increase the statistical power when comparing them to conventional investors.
Relative to the invited sample, there are slightly more men, older investors and investors
with a larger portfolio, among the respondents. We control for these and other
demographic variables in our analyses.
In the online survey investors answered questions and took part in experiments
with monetary incentives (for details see below). At the beginning of the survey
respondents received some general information. In addition, they were also informed
that they would take part in several experiments, but were not informed about the
content of the experiments until they actually took place. In the introduction to the
survey also the general procedure regarding possible money earnings in the experiments
was explained. In the first part of the survey, we asked about general investment issues
like the assets held, the number of investment accounts and investment goals. In this
5
Our survey (see below) indicates that 83% of all investors (also those who do not hold SRI funds)
respond positive or neutral to the statement that socially responsible investments have a positive
influence on society. Only 26% of the investors indicate in another statement that they believe that
SRI funds are a marketing trick to sell more funds. We are therefore confident that funds defined as
SRI funds are also perceived as such by most investors.
6
We excluded investors that were no longer holding the account at the time we conducted the survey.
We also did not invite investors who never placed a single trade or were younger than 18 years.
11
part, investors also participated in a risk preferences elicitation experiment. Thereafter,
more questions on investment behavior followed. Somewhere in the middle of the
survey investors participated in an experiment eliciting their intrinsic social preferences.
We asked all survey questions regarding SRI and other behavior that could be
interpreted as related to social goals after the experiments.
Survey questions have many advantages but also some known limitations. For
instance, participants might differ from non-participants and the answers of respondents
may depend upon the framing of the questions. We discuss a potential response bias in
our results below and conclude that if a response bias is present, it likely weakens the
effect sizes we identify and that we err on the conservative side. Regarding framing
effects, it is important to note that all investors received similarly framed questions. We
are primarily interested in potential differences in beliefs and attitudes of socially
responsible and conventional investors and framing effects should be equal for these
groups. Surveys also have major benefits. In our case, it allows us to gather information
about return expectations and risk perceptions, which would otherwise remain
unobserved. Moreover, we can collect information on additional important control
variables, like self-rated investment knowledge, income level, gender, age etc. (see also
Guiso, Sapienza and Zingales (2013) for a discussion of the pros and cons of surveys
for studying financial decisions).
3.3 Experiments7
Investors participated in a risk preferences elicitation experiment and in an interactive
experiment with other investors where we elicited their social preferences. Investors
were informed that at the end of the survey it would be determined randomly (with a
chance of one out of ten) whether they will receive the earnings from the experiment or
not.8 Those who were selected for payment got one of the experiments paid out at
random. Investors received their earnings via bank transfer at the first working day after
they completed the survey and payments were guaranteed by the authors’ university. We
7
The experiment instructions are available as supplementary materials in the appendix.
8
For a recent validation of this procedure, see Dohmen et al. (2011).
12
used a unique identification number to link the choices in the experiments and responses
to the survey to our administrative data. In order to ensure anonymity of investors we
hired an external company specialized in conducting online research to handle the
payments. This company does not have access to the trading records or other
information of the investors. Survey participants were informed about this at the
beginning of the survey.
Risk preferences elicitation experiment
We elicit risk preferences with incentivized multiple price list lotteries, similar to Holt
and Laury (2002) and Dohmen et al. (2011). Investors faced 20 different decision
situations and for each situation they decided between receiving a specific sure amount
and a lottery with a 50% chance of winning 300 euro and a 50% chance of winning
nothing. The sure amount was minimally 0 euro and maximally 190 euro and increased
in steps of 10 euro from one to the next decision situation. The presented choice options
can be found in Table A2 in the appendix. As common in such risk elicitation
experiments, it was determined randomly which of the 20 decision situations will be
relevant for earnings.
The choices made by participants in each of the 20 decision situations inform us
about their risk preferences. We use the point at which individuals switch between the
lottery and the certain outcome as a measure of risk attitude. As the sure amount is
ordered from low to high, a higher switching point indicates a more risk tolerant
participant.
Social preferences elicitation
To measure intrinsic social preferences, we use a variant of the trust game experiment
introduced by Berg, Dickhaut and McCabe (1995). The trust game is a two-player
sequential game. Both the first-mover and the second-mover are endowed with 50
euro.9 The first-mover decides on the amount he or she wants to send to the second9
Since its introduction (Berg, Dickhaut and McCabe (1995)) it is standard practice in the literature
using trust game experiments to endow both participants with the same initial amount (e.g., Fehr and
List (2004); Falk, Meier and Zehnder (2013); Falk and Zehnder (2013)). The main reason is to avoid
experimenter induced unequal positions ex ante.
13
mover, which can be any multiple of 5 euro, including zero and 50. The amount sent is
tripled by the experimenter and the second-mover decides how much of the received
money to return to the first-mover. Hence, the earnings of the first-mover are 50 euro
minus the amount sent plus the amount returned by the second-mover. The earnings of
the second-mover are 50 euro plus triple the amount sent by the first-mover minus the
money sent back.10
We use second-mover behavior to measure intrinsic social preferences.11 In
order to obtain a comprehensive measure of intrinsic social preferences as well as for
practical implementation reasons, we used for second-movers the so-called strategy
method (Selten (1967)). That is, a second-mover decided how much to send back, for
each of the 10 possible non-zero amounts sent by the first-mover – ranging from 5 euro
to 50 euro – before knowing the actually sent amount. Specifically, the experiment
instructions informed second-movers that “[f]or technical reasons you should make
your decision without knowing how much money the person to whom you have been
linked has actually sent you. Therefore, for each possible amount that the other person
could send you, we would like to ask you to indicate, how much you would like to
return. However, only the decision that is relevant for the amount that has actually been
sent is decisive for your income and the income of the person to whom you have been
linked.” If the first-mover did not sent anything then both, first- and second-mover,
earned the 50 euro they were endowed with.
Next to generating a comprehensive measure of intrinsic social preferences
another important advantage of the strategy method is that it simulates sequential moves
10 The money sent by the first-mover and tripling of this amount by the experimenters is 'free lunch' for
the second-mover and one may argue that second-movers could act differently would they need to
earn these rights. Unfortunately, there is no evidence available on if and how second-mover behavior
in trust games would change when first-mover transfers and tripling of the transfer were not free. In
our study it likely implies that it is harder to detect a relation between social preferences measured by
the trust game and holding of SRI funds. This would lead to an underestimation of the link between
intrinsic social preferences and socially responsible investments and we, therefore, likely err on the
conservative side.
11 We also have data on the behavior of first-movers in the trust game, but do not report on them here
for brevity and because it intermingles trust and intrinsic prosocial behavior (Cox (2004)).
14
for each possible choice of the first-mover without deceiving subjects and without the
necessity that players' choices are indeed sequential in time. Similar versions of the
strategy method have recently been successfully used in trust game experiments (see
e.g. Baran, Sapienza and Zingales (2010), Falk, Meier and Zehnder (2013), Falk and
Zehnder (2013)).
Each investor was either a first- or a second-mover. Every working day, we
randomly matched first-movers to second-movers. After choices were made, we
implemented the one choice out of the 10 possible choices of the second-mover that
corresponded to the actual choice of the first-mover in case s/he made a non-zero
transfer. For example, if the first-mover transferred 30 euro to the second-mover, we
took the amount that the second-mover wanted to return for that transfer. In the example
the second-mover would receive 3 x 30 = 90 euro. If the second-mover, for instance,
decided to return 45 euro, the earnings of the second-mover would be 45 + 50
(endowment) is 95 euro and the earnings of the first mover would be the endowment
50 – 30 + 45 = 65 euro. In case the actual choice of the first-mover was to send zero
both earned their initial endowment of 50 euro.
Moreover, second-movers in the trust game are randomly assigned to one of two
conditions. Under one condition, they are matched to a first-mover who is a randomly
chosen investor participating in the survey and the experiment. In the other condition, a
second-mover is randomly matched to a first-mover who is a socially responsible
investor participating in the survey and the experiment. We inform subjects in the
introduction to the experiment in which condition they are, without telling them that
there are two different conditions.
Investors received instructions of the experiment online and had to answer a
couple of comprehension questions about the rules of the game and how the payment is
calculated before the experiment started. These questions were correctly answered by
89.5% of the investors.12 The trust game was played only once. The investors were
12 We conduct our main analysis with all investors and confirm in unreported analyses that the results
remain qualitatively unchanged when excluding investors who answered incorrectly to at least one
question after three trials.
15
informed about this and also about that they and the other participants in the experiment
would remain anonymous during and after the experiment.
The fact that the trust game is played only once rules out repeated game effects.
Moreover, second-movers know that their behavior will never be revealed to anybody
and is only anonymously known to the experimenters, which minimizes prosocial
behavior in the trust game that is due to (social) reputation and image effects. We are,
therefore, confident that we can interpret second-mover behavior as an independent
indicator of intrinsic social preferences. In Section 3.4 we introduce a quantitative
measure of intrinsic social preferences in detail.13
3.4 Variables
All variables discussed here are also described in Table 1. Table 2 shows summary
statistics for all variables and – when appropriate – test results comparing socially
responsible and conventional investors. We discuss in sequence the variables from (1)
the administrative transaction data, (2) the survey questions, and (3) the experiments.
< TABLE 1 ABOUT HERE>
13 One may argue that there is an issue of saliency because the experiment payoffs are small relative to
investors' assets and incomes. We are confident, however, that this does not jeopardize our results for
the following reasons. First, as the most important effect of no or too low stakes is an increase of noise
in the data (Camerer and Hoghart (1999)), it would most likely reduce the chance to detect a relation
between social preferences measured in our experiment and socially responsible investments in the
field. Therefore, effects of social preferences we observe can be interpreted as lower bounds. Second,
payoffs in the experiment were not unreasonable for the time investors had to spend on the
experiment. On average it took participants 45 minutes to complete the whole survey and all
experiments. Therefore, the potential payment may not be too far off participants’ opportunity costs,
given that they most likely participated in their leisure time. Third, although there have been some
moderate quantitative stake size effects reported in experiments similar to ours, qualitatively the
effects are similar for low and high stakes (Oosterbeek, Sloof and Van De Kuilen (2004)). For a
relatively recent discussion of stake size effects see Falk and Heckman (2009).
16
Administrative data
SRI funds holdings
We distinguish between socially responsible and conventional investors using the
administrative data.14 Table 2 (top) shows that 14% of the investors in our sample hold
at least one SRI mutual fund. This percentage is close to the 18% for Dutch investors in
general (Millward Brown (2011)). The table further shows that socially responsible
investors on average hold 14.9% of their portfolio in SRI funds. 19.5% of the socially
responsible investors only have SRI funds with tax incentives, 68.4% only have SRI
funds without tax incentives, and 12.1% hold both types of SRI funds.
Equity ratio and trading activity
The equity ratio of an investor is defined as the fraction of the overall portfolio that is
invested in equity funds (Equity ratio). Table 2 shows that the average (median) Equity
ratio of socially responsible investors is with 55.0% (52.1%) slightly higher than 51.3%
(47.6%) for conventional investors. The difference is small but statistically significant
(p=0.003)15. From the table it can also be seen that socially responsible and
conventional investors slightly differ in their trading activities. The variable Number of
transactions reports the number of trades an investor made in the 12 months prior to the
experiment. It shows that socially responsible investors are a bit more active (average
number of trades: 25.4, median: 10) than conventional investors (average number of
trades: 22.8, median: 6). Again this difference is small but statistically significant
(p<0.001).
14 If not stated otherwise, the administrative data refer to the month in which investors participated in
the experiments and survey. In principle, it is possible that investors only hold SRI funds for a very
short period. Therefore, in unreported analyses we conduct all tests and regressions also for
investments into SRI funds exactly one month after the survey and experiment in 2011. The results
remain qualitatively the same.
15 To save on space and for convenience in the main text we do not report the kind of tests used. They
can be found in Table 2. If not mentioned otherwise all p-values are for two-sided tests.
17
Total portfolio value, mean portfolio return, volatility portfolio return
As a proxy for wealth, we use the (logarithm) of the total portfolio value in the month in
which the investors participated in the survey and experiment. Table 2 shows that the
average (median) Total portfolio value of socially responsible investors is 106,678 euro
(57,666 euro), compared to 73,250 euro (36,496 euro) for conventional investors. This
difference is statistically significant (p<0.001). The table also reports the Mean portfolio
returns defined as the average of portfolio returns per month since the investor opened
his or her account. In comparison to conventional investors these returns are
significantly smaller for socially responsible investors (SRI: 0.2094, conventional:
0.2363; p=0.0378). As a proxy for portfolio volatility Table 2 reports the variable
Volatility portfolio returns, defined as the average of the standard deviations of monthly
portfolio returns. It shows that on average the returns of socially responsible investors
are significantly more volatile (4.3258) than those of conventional investors (3.9643)
(p<0.001).
<TABLE 2 ABOUT HERE>
Survey data
Return expectations and risk perceptions
To measure the returns that investors expect on SRI equity funds compared to
conventional equity funds, we used the statement: “I expect that the returns of socially
responsible equity funds compared to conventional equity funds are: Much lower, A bit
lower, The same, A bit higher, Much higher, I do not know.” Only 3% of the socially
responsible and 10.3% of the conventional investors choose “I do not know.” To
measure risk perceptions of SRI equity funds compared to conventional equity funds,
we asked investors to rate their agreement to the following statement: “Socially
responsible equity funds are more risky than conventional equity funds.” The agreement
with each of these statements had to be rated on a 1-7 Likert scale from 1 ‘Disagree
completely” to 7 “Agree completely.”
In order to explore the determinants for holding or not holding SRI funds, we
first examine whether conventional and socially responsible investors differ in their
expectations regarding the returns of SRI. Figure 1 depicts the distribution of Expected
18
returns on SRI equity funds in comparison to conventional equity funds (as defined in
Table 1). The figure suggests that socially responsible investors are slightly less
pessimistic about returns of SRI than conventional investors. For instance, 51.9% of the
socially responsible investors and 59.7% of the conventional investors expect to earn
much or a bit lower returns on SRI funds than on conventional funds. The difference in
distributions is small and statistically not significant (p=0.692, see Table 2).16 Figure 2
shows that the distribution of Perceived risk on SRI equity funds relative to
conventional equity funds (as defined in Table 1) is similar for socially responsible and
conventional investors. For both investor types the mean (median) score is with 3.5 (4)
and 3.6 (4) almost identical and statistically indistinguishable (p=0.385, see Table 2).
Hence, both socially responsible and conventional investors think that both fund types
produce similar returns and carry similar risk.
<FIGURE 1 SOMEWHERE HERE>
<FIGURE 2 SOMEWHERE HERE>
The discussed variables reflect that both investor types are rather pessimistic
about return expectations of SRI funds and perceive the riskiness of such funds
similarly. This suggests that other motives than return expectations or risk perceptions
must (also) play a role in the decision to invest in SRI funds. One possibility could be
that investors hold SRI funds for risk diversification reasons. Even if investors perceive
the risk of SRI equity funds in isolation as about the same as the risk of conventional
equity funds, they may want to reduce the overall portfolio risk by including SRI funds
into their portfolio. Our survey data show that this motive is virtually absent. Only 5.1%
of all SR investors indicate to hold SRI funds because of diversification benefits.
Investment knowledge and communication about investment
With the survey we gathered more information on investors not provided by the
administrative data (see Tables 1 and 2). In these questions investors had to respond to
16 For the tests we encode the answers to the question on a Likert-scale from 1 = much lower to 5 =
much higher.
19
several statements on a 1-7 Likert scale from 1 “Disagree completely” to 7 “Agree
completely.”
Similar to other studies (Dorn and Huberman (2005), Graham, Harvey and
Huang (2009), Van Rooij, Lusardi and Alessie (2011)) we measured self-assessed
investment knowledge with the statement: “My investment knowledge is good.” Socially
responsible investors rate with an average of 4.19 their Investment knowledge higher
than conventional investors who rate their knowledge on average with 3.83. The
difference is significant (p< 0.001).
With the statement “I often talk about investments with others” we elicit
communication patterns regarding investment. Socially responsible investors score on
average with 3.11 higher on Talk about investments than conventional investors
(average: 2.91). The difference is statistically significant (p = 0.001). Below we will
discuss a possible interpretation of this variable and will use it as a proxy for the extent
to which investors can potentially signal their social mindedness by talking about it.
Education, gender, age, income
We also asked for the highest achieved education level. Slightly more socially
responsible investors hold a University degree (49.9%) in comparison to 46.2% of the
conventional investors (p=0.136).
Gender composition and age of socially responsible and conventional investors
differ only little. Of socially responsible investors 18.1% are Female in comparison to
20.9% of conventional investors (p=0.075). On average the Age of socially responsible
investors is slightly lower (57.8 years) than of conventional investors (59.1 years)
(p=0.002). Income is self-reported gross family income per year. 18.0% of the socially
responsible investors and 17.7% of the conventional investors did not report their
income (p=0.894). For our subsequent analysis we created the dummy variables Low
Income (below 60,000), Median Income (between 60,000 and 100,000) and High
Income (above 100.000) such that each category comprises about one third of the
sample. Socially responsible and conventional investors are nearly identically
distributed over the different income categories (p>0.537).
20
Experiment data
Risk attitudes
Table 2 (third row from bottom) also reports the experimentally elicited Risk
preferences. Recall that the risk neutral switching point is 150 euro (50% chance to win
300 euro and 50% chance to win nothing). The average switching point for socially
responsible investors is at 113.29 euro compared to 112.23 euro for conventional
investors. The difference is statistically not significant (p=0.723). This implies that both
groups of investors are on average similarly risk averse.
Intrinsic social preferences
As explained in the design section we use second-mover behavior in a one-shot
anonymous trust game experiment to elicit intrinsic social preferences. Through the use
of the strategy method we have 10 non-zero monetary return decisions for each investor
in the role of second-mover. Figure 3 reports the average return ratios of second-movers
for each of the 10 possible non-zero transfers of first movers. The average return ratios
are defined as the amount returned by the second-mover over the amount sent by the
first-mover. It shows that for each possible first-mover transfer, second-movers return
more if they are a socially responsible investor than if they are a conventional investor.17
<FIGURE 3 ABOUT HERE>
In order to arrive at a measure of intrinsic social preferences we need to
aggregate the return decisions. To this end we construct the natural measure of Mean
intrinsic social preferences by calculating the average return ratio across all 10 return
decisions. In other words, for each possible non-zero first-mover transfer (i.e., 5 euro,
10 euro, …, 50 euro) we calculate the ratio of the back-transfer and take the average.
Table 2 (next to last row) shows that this measure is larger for socially responsible
investors (1.53) than for conventional investors (1.42). The difference is marginally
significant (p=0.087). Alternatively, we could also use a measure proposed by Baran,
17 The amount that investors return in the trust game is not significantly different for the two matching
conditions described in Section 3.3. The average return ratios amount to 1.44 and 1.40, respectively
(F-test, p = 0.216). For the remaining analysis we pool the data of both matching conditions.
21
Sapienza and Zingales (2010) who argue that the amount returned by second-movers for
the maximum transfer (50 euro in our case) is the best measure for social preferences
because the stakes are highest for this decision. For completeness and to check
robustness of our measure we also construct a second measure capturing this idea: Max
intrinsic social preferences. It is defined as the absolute amount a second-mover
investor returns for the largest possible first-mover transfer of 50 euro. Using this
measure we find that socially responsible investors on average send back 77.46 euro
and conventional investors 71.61 euro. The difference is statistically marginally
significant (p=0.061; Table 2 last row). In the following we only report results using our
Mean intrinsic social preferences and will call it just Social preferences for brevity. All
main results are similar when using the alternative measure and are reported in the
appendix.
4. Determinants of portfolio choice
Our foremost interest is whether social preferences are predictive for holding SRI funds.
To investigate this we use our measure Social preferences, introduced above. Further we
explore how other important variables and characteristics of investors, like the equity
ratio of their portfolio, risk-return expectations, risk preferences etc. contribute to
explaining holding SRI funds. We first examine what determines the likelihood that an
investor holds at least one SRI fund in his or her portfolio. For that we run probit
regressions in which the dependent variable is binary taking on the value 1 if an investor
holds an SRI fund and 0 otherwise.
Next to social preferences we control for a number of other potential
determinants of holding SRI funds. An important alternative explanation for why
investors may hold SRI funds are (potentially biased) risk-return expectations. We
therefore include variables regarding return expectations and perceived risks of SRI
funds relative to conventional funds as explanatory variables, using the answers to the
respective survey questions (cf. Figures 1 and 2). For the regression analyses we create
a dummy variable for expected returns on SRI (Lower expected returns on SRI) that
takes on value 1 if an investor believes that the expected return of SRI funds is lower
22
than the expected return of conventional funds and zero otherwise. Similarly, for risk
perception on SRI we create a dummy variable (Lower perceived risk on SRI) that takes
on value 1 if an investor believes that the return risk of SRI funds is lower than the one
of conventional funds and zero otherwise.18
Another set of potential explanatory variables is related to investors' portfolios.
We have seen that there are some differences between SRI investors and conventional
investors regarding the equity ratio in their portfolio, their trading activities, the value of
their portfolio,19 and portfolio returns as well as volatility (cf. Table 2). As these
differences may contribute to explaining holding SRI funds, we control for them using
the variables Equity ratio, Log number of transactions, Log total portfolio value, Mean
portfolio returns, and Volatility portfolio returns as introduced above.
Besides the Log total portfolio value, we control for investment knowledge with
two other measures used and validated in related literature: first, investors answers to a
financial knowledge question (Investment knowledge) where they had to rate themselves
on a 7-point Likert-scale from very poor to very good (Van Rooij, Lusardi and Alessie
(2011), Dorn and Huberman (2005), Graham, Harvey and Huang (2009)), and second, a
dummy variable indicating whether an investor has a University degree.
We also control for investors' risk preferences independently measured by the
switch amount in our experimental risk preference elicitation task (Risk preferences). As
further control variables we include gender (Female, which takes on value 1 if the
investor is a woman, zero otherwise) and age (Age) of investors. We also use survey
responses to control for Low income, High income and Untold income, with medium
income being the omitted reference category (for the precise definitions of these
variables, see Table 1).
18 We use dummies for return expectations and risk perceptions of SRI funds instead of the scores
themselves because it reduces the noise in the data, as there are relatively fewer observations in the
extreme categories (cf. Figures 1 and 2). Moreover, generally the scores cannot be interpreted as linear
variables.
19 The value of an investor's portfolio is also a validated proxy for investment knowledge (Calvet,
Campbell and Sodini (2009), Keloharju, Knüpfer and Linnainmaa (2012)) for which we also control
with two separate survey measures (see below).
23
Finally, investors may use SRI funds for signaling prosocial attitudes. For
instance, in a theoretical contribution Glazer and Konrad (1996) argue that people give
to charities not necessarily (only) because of prosocial preferences but because they
want to signal social status. Nelson and Greene (2003) argue in favor of charity giving
as a signal of goodness and recently Fehrler and Przepiorka (2013) provide
experimental empirical evidence for signaling benefits of altruistic acts. It is difficult to
gather direct evidence on such signaling, consequently we do not have observations in
our data with which we could directly measure it. We, therefore, resort to a proxy.
Investors in our data set buy funds directly online without interference of an
intermediary. Thus, if investors want to signal prosocial attitudes with their SRI funds
they need to communicate it to others in one or the other way. In the survey, investors
reported on how often they communicate about their investments by indicating their
(dis)agreement with the statement “I often talk about investments to others.” on a 1-7
Likert scale.20 We use answers to this question as our proxy for signaling and call the
variable Talk about investments. We hasten to note that there are many other potential
reasons for people to talk about investments. Therefore, we view the signaling
interpretation only as suggestive. We will also discuss the role of signaling in more
detail below (cf. Section 4.3).
<TABLE 3 ABOUT HERE>
The second column of Table 3 ('Model Probit') reports the regression results.21
For ease of interpretation we present marginal effects. The results show that stronger
social preferences have a significantly positive effect on the likelihood to invest in a
socially responsible manner (p=0.037). To illustrate the economic effect: an investor
with a 1 point higher mean return ratio is 4.95 percentage points more likely to have
20 This question was asked at the beginning of the survey before any question on prosocial behavior. At
that stage of the survey, no reference to socially responsible investments had been made yet.
21 The total number of observations in the regressions is lower than the overall response rate to the
survey. The reason is that investors were randomly assigned to different experiments and to a different
role in the trust game. For instance, first-movers in the trust game do not appear in our regression
analyses.
24
SRI funds in the portfolio, which is a relatively large effect compared to the 14% of our
sample that holds SRI funds.
Importantly, the reported effect of social preferences holds while controlling for
a number of variables, specifically return expectations and risk perceptions regarding
SRI funds. Indeed, expectations about the returns of SRI funds (in comparison to
conventional funds) have the correct sign but are statistically insignificantly related to
the likelihood to invest in a socially responsible manner (Lower expected returns on
SRI, p=0.336). Similarly, differences in risk perceptions about SRI funds do not
significantly contribute to the likelihood of holding SRI funds (Lower perceived risk on
SRI, p=0.421).
The probit estimate further documents that of the portfolio characteristics the
number of transactions (Log number of transactions) and the value of the portfolio (Log
total portfolio value) have a significantly positive effect on the likelihood to hold SRI
funds.22 Regarding the portfolio size, an investor with a 100% larger portfolio is about
3.9 percentage points more likely to invest in a socially responsible manner (p=0.002).
This relatively strong effect is intuitive as investors with larger portfolios likely spread
their larger wealth over various funds, including SRI funds. Similarly, an investor who
is 100% more active in terms of transactions is about 2.8 percentage points more likely
to hold SRI funds (p=0.015). Neither the equity ratio nor portfolio returns or their
volatility affect the likelihood to hold SRI funds (p>0.166).
Interestingly, of the other variables only Talk about investments exhibits a
significant influence on the likelihood that an investor holds SRI funds. Specifically, an
investor who rates him-/herself one point higher on a 1-7 scale regarding talking about
ones investments is 2.69% more likely to invest in a socially responsible manner
(p=0.020). As discussed above, this variable can be interpreted as a proxy for signaling
prosocial attitudes. Its significance may, therefore, indicate that not only social
22 As it could be that investors with small portfolios disproportionately allocate their funds to SRI we
also run regressions where we include interaction terms between social preferences and portfolio
value. As the interaction terms were never significant we relegate these regressions to the appendix
(Table A5). This also applies to the regressions reported in Table 4 below (see appendix Table A6).
25
preferences have a positive effect on the likelihood to hold SRI funds but that they are
also held for their signaling value.
Having shown that the likelihood to hold SRI funds is affected by social
preferences, we next explore the more ambitious question whether social preferences
are also predictive for the percentage of such funds an investor holds in the portfolio.
For that purpose, we conduct a Tobit regression that accounts for the censoring in the
SRI share at 0% and 100%. The third column of Table 3 ('Model Tobit') shows the
results and reports the marginal effects of the explanatory variables.
The results show that investors' social preferences are also predictive for the
share of SRI funds in their portfolio, albeit at a higher significance level. A one point
higher return ratio is associated with an increased investment into SRI funds of 4.03%
of the portfolio (p=0.092). As in the probit regressions return expectations and risk
perceptions are not significantly related to the percentage of the portfolio that is
invested in a socially responsible manner (Lower expected returns on SRI, p=0.260;
Lower perceived risk on SRI, p=0.759).
Of the portfolio characteristics the Equity ratio is now significantly positively
related to the share of SRI funds an investor holds (p=0.042) and, as before, the same
holds for portfolio size (Log total portfolio value). Specifically, a 100% larger total
portfolio size leads to about an extra 2.8% in SRI funds (p=0.024). In contrast to the
probit regressions Low income is now predictive for the share of SRI funds an investor
holds (p=0.066). The statistically significant positive relation between investing in a
socially responsible manner and Talk about investments already detected in the probit
regression also holds in the Tobit model (p=0.062).
To summarize, we find clear evidence that social preferences are an important
determinant for the likelihood to invest in a socially responsible manner as well as for
the fraction of SRI in an investor's portfolio. Importantly, this holds when controlling
for a number of other potential determinants of investor behavior. Intriguingly, return
expectations and risk perceptions regarding SRI relative to conventional funds do not
correlate with the likelihood or percentage of SRI holding. Taken together, this clearly
shows that social preferences are not driven out in the market place. Moreover, it
26
indicates that differences in investments in SRI funds are better explained by social
preferences than by differences in risk-return expectations. The equity-ratio, which may
reflect a preference for equity over bonds, has some predictive power for the percentage
of SRI an investor holds but not its likelihood. Of the other portfolio characteristics the
number of transactions is positively correlated only with the likelihood to hold SRI
funds whereas the size of an investor's portfolio is correlated also with the percentage of
SRI holdings. In addition, an investor who talks more often about investment is also
more likely to hold (a larger share of) SRI funds.
4.1 The role of tax benefits in investing in a socially responsible manner
As explained in Section 2, investors could hold SRI mutual funds with or without tax
benefits. In this section we explore whether investors’ social preferences are predictive
for holding both types of SRI funds or whether they predict only the holding of one type
of SRI funds, without tax benefits or with tax benefits. There may be a differential effect
of social preferences for at least two reasons. First, SRI funds with tax benefits may also
attract selfish investors for purely financial reasons. Second, the literature on crowdingout effects of social behavior through financial incentives suggests that SRI funds with
tax benefits may be less attractive for investors with a strong non-financial social
motivation for holding SRI funds.
An important difference between SRI funds with and without tax benefits is that
the former are bond funds whereas the latter are equity funds (cf. Section 2). We
therefore need to rule out that an eventual differential effect of social preferences on the
probability of holding SRI funds with or without tax benefit is due to the difference in
asset class between these funds, rather than the presence or absence of tax benefits. In
this respect we identify four main considerations one needs to account for.
First, due to differences in volatility, risk averse investors might prefer SRI bond
funds with tax benefits over SRI equity funds without tax benefits. Therefore, we
control for risk aversion using our independent experimental measure of investors' Risk
preferences. We do not expect a large effect, as 99% of socially responsible investors
only invest part of their portfolio in SRI funds. Their overall portfolio usually combines
27
conventional equity and bond funds with SRI equity and/or SRI bond funds, supposedly
to achieve the desired level of portfolio risk.
Second, investors with stronger social preferences might have a preference for
equity funds over bond funds (or vice versa) for other reasons. For instance, only equity
holders have a true 'share' in a firm and voting rights, and it might be that it is investors
with social preferences who value such rights high. Hence, without taking the equity
ratio in the investors' portfolios into account a correlation between types of SRI funds
and social preferences may be spurious and actually represent a correlation between
types of SRI funds and a general inclination to hold more equity. Therefore, we will
control for the equity ratio.23
Third, investors who buy funds with tax benefits might differ in their trading
pattern from those who buy funds without tax benefits. We therefore also control for the
total portfolio value and the number of transactions made by the investor in the previous
twelve months.
Fourth, it could be that investors who invested in SRI funds with tax benefits
experienced different returns and volatility than those who invested in funds without tax
benefits, which might drive the preference for one type of SRI fund over the other. We
therefore control for monthly portfolio returns and the average monthly standard
deviation of portfolio returns of investors' portfolios.
Specifically, we run a multinomial logit regression in which the dependent
variable takes on four different values. The base group, which is not reported, consists
of conventional investors. The other groups are investors who (1) only hold SRI funds
with tax benefits, (2) only hold SRI funds without tax benefits, and (3) hold both types
of SRI funds. As in the previous section we report the results for Social preferences
measured with Mean intrinsic social preferences.24
<TABLE 4 ABOUT HERE>
23 Further below we will also test directly the correlation between social preferences and holding of
equity.
24 The results remain robust when using Max intrinsic social preferences (see Table A4 in the appendix).
28
The results are shown in Table 4, which reports relative-risk ratios (RRR). We
find that, in comparison to being a conventional investor, investors with stronger Social
preferences are more likely to hold SRI funds without tax benefits but not SRI funds
with tax benefits or a mix of both. Specifically, column (2) of Table 4 shows that an
investor with a one point higher return ratio in the trust game experiment is 54.6% more
likely to only hold SRI funds without tax benefits than being a conventional investor
(p=0.020). Column (1) of the table shows that investors with stronger social preferences
are not significantly less (or more) likely to only hold SRI funds with tax benefits
(p=0.378) and column (3) that they are also not significantly more (or less) likely to
hold a mix of both types of SRI funds (p=0.154). To test if social preferences
differentially affect holdings in the two different types of SRI funds, we have also run
regressions where the base group consists of investors holding only SRI funds without
tax benefits.25 We find that, in comparison to investors holding only SRI funds without
tax benefits, investors with stronger social preferences are less likely to only hold SRI
funds with tax benefits (RRR=0.469, p=0.057) but equally likely to hold both types of
SRI funds (RRR=1.480, p=0.512). Hence, stronger social preferences predict the
holding of SRI funds without tax benefits.
In line with our results reported in the previous section, return expectations
(Lower expected returns on SRI) and risk perceptions (Lower perceived risk on SRI) are
statistically insignificant. Whereas, relative to conventional investors, the Equity ratio is
not predictive for holding SRI funds with tax benefits or holding both types of SRI
funds (p=0.106 and p=0.331, respectively), it is predictive for holding SRI funds
without tax benefits (p=0.058). From the other variables, characterizing the portfolios of
investors, we see that investors who make more transactions (Log number of
transactions) are only more likely to hold SRI funds with tax benefits (p=0.019) and
investors with larger total portfolio size (Log total portfolio value) are in general more
likely to hold SRI funds without tax benefits (p=0.014) or both types (p=0.054).
For the other controls we see that investors who rate their Investment knowledge
higher are more likely to only hold SRI funds with tax benefits than to only hold
25 For the sake of brevity we only report the results regarding social preferences here. The whole
regression is reported in Table A7 in the appendix.
29
conventional funds (p=0.008). There is no investment knowledge effect for the other
investor groups. We also find that having a University degree or being Female increases
the likelihood to hold both kinds of SRI funds (p=0.048 and p=0.031, respectively) and
being of older Age has a positive effect on holding only SRI funds with tax benefits
(p=0.060).
Finally, the variable Talk about investments is, relative to conventional investors,
predictive for the likelihood to hold SRI funds without tax benefits (p=0.052) but not for
holding SRI funds of the other two types (p>0.137). This result is consistent with
findings from a lab experiment with students of Ariely, Bracha and Meier (2009) who
show that monetary incentives can erode the signaling benefits of prosocial behavior. In
our case, the monetary benefits of SRI funds with tax benefits can reduce the signaling
value. We further discuss the interpretation of Talk about investments in Section 4.3
below.
To summarize, we find that, while investors return expectations and risk
perception of SRI funds do not have any explanatory power, investors with stronger
social preferences are substantially more likely to hold SRI funds without tax benefits.
Importantly, this holds while controlling for a number of variables, including the equity
ratio, which are also potential determinants of SRI and/or might correlate with social
preferences.
4.2 Investors' equity ratio and social preferences
In the above analysis we have shown that the differential effect of social
preferences on holding SRI funds with and without tax benefits holds when controlling
for the equity ratio in investors' portfolios. This establishes that social preferences are a
robust predictor for holding SRI funds without tax benefits. It does, however, not
exclude the possibility that investors with stronger social preferences also have a
preference for equity funds. To explore this we use an investor's equity ratio in his or
her portfolio as the dependent variable and regress it on social preferences. In other
words, we test if stronger social preferences are predictive for holding more equity.
30
We examine different regression models and specifications and report the results
in Table 5. Specifications (1 - 3) report Tobit regressions that account for censoring of
the dependent variable equity ratio at 0 and 1. Specification (1) uses only social
preferences as explanatory variable while in specification (2) control variables are
added that are also used in the other regressions. In specification (3) an interaction
variable between social preferences and holding SRI funds is added, which allows to
test if it is possibly SRI investors with strong social preferences who are more likely to
hold equity. As there are well-known concerns when using interaction variables in Tobit
regressions (Ai and Norton 2003), specification (4) repeats the estimation of
specification (3) using OLS.
As can be seen from the table, in no specification do we find a statistically
significant relation between social preferences and the equity ratio. In specifications
(2 – 4) the relation is even slightly negative. In all cases, the coefficient is close to zero
and the standard error large. In specifications (3) and (4) also the interaction between
SRI holding and social preferences is statistically insignificant.
These results clearly show that there is no relation between social preferences
and a preference for holding equity. The regressions give also additional interesting
insights. We find the intuitive result that Volatility of portfolio returns is correlated with
the equity ratio in all specifications (at least at the 5% significance level). Further, the
Log total portfolio value is negatively correlated with the equity ratio (significant at 5%)
and more knowledgeable investors have a higher share of equity in their portfolios
(significant at 1%). All other variables are insignificant.
In our view, the above results already convincingly show that the positive
relation between social preferences and investment in SRI funds without tax benefits is
not driven by a preference for equity by investors with strong social preferences. Still, in
theory, one might argue that investors with strong social preferences might not have a
preference for equity per se, but might have a preference specifically for SRI equity
funds. Although it has been shown that some institutional investors engage as active
shareholders and try to push companies to be more socially responsible (see e.g. Barber
(2007)), we think that this is an unlikely explanation in our context for three reasons.
31
First, we only have individual investors in our sample and they have very limited voting
power (shareholder activism is typically pursuit by institutional investors rather than by
individuals). Second, the mutual fund provider of our study is not an active shareholder
and therefore investors cannot influence companies to be more socially responsible
through the provider. Third, companies cannot only be influenced by shareholders, but
also by debt holders as lenders can set social responsibility criteria when providing a
loan to a company.
Taken together, our results provide strong evidence that investors with strong
social preferences prefer SRI funds without tax benefits over SRI funds with tax
benefits and that this preference is not correlated with a preference for equity.
<INSERT TABLE 5 ABOUT HERE>
4.3 Discussion of motives to hold SRI funds
We have presented evidence that social preferences are robustly influencing portfolio
choice. In this section we try to dig a bit deeper and discuss further motives for
investors with(out) strong social preferences to (not) invest in a socially responsible
manner.
We have seen that social preferences explain holding of SRI funds without tax
benefits but not with tax benefits. With respect to the latter, conventional investors
presumably hold them for the tax benefits and one may ask the question whether
investors with stronger social preferences 'shun' this SRI type in order to distinguish
themselves from conventional investors. There is some evidence suggesting that this is
indeed that case. In Table 4 we have seen that investors with stronger social preferences
are less likely (27.4%) than conventional investors to hold SRI funds with tax benefits,
albeit statistically not significantly. By the same token, comparing the likelihood of
holding SRI funds with and without tax benefits shows that investors with stronger
social preferences are much less likely (53.1%) to invest into SRI funds with tax
benefits than without tax benefits (p=0.057), indeed suggesting an aversion to the
32
former in favor of the latter (cf. Table A7 in the appendix). This is consistent with the
hypothesis that the monetary incentives of SRI funds with tax benefits may to some
extent crowd out the intrinsic motivation of investors with strong social preferences
(Gneezy and Rustichini (2000 a,b), Bénabou and Tirole (2006), Ariely, Bracha and
Meier (2009)).
Next we want to shed some more light on the signaling interpretation of the
variable Talk about investments. We have seen that this variable is, in comparison to
only holding conventional funds, positively correlated with holding SRI funds without
tax benefits (cf. Table 4). This result is consistent with findings from a lab experiment
with students of Ariely, Bracha and Meier (2009) who show that monetary incentives
can erode the signaling benefits of pro-social behavior. It also suggests two
interpretations, depending on how social preferences are related to Talk about
investments. A positive correlation, that is when pro-social investors also share the fact
of their investment with others, would be consistent with an interpretation that these
investors try to distinguish themselves in that way from other investors who only invest
in SRI funds with tax benefits. In contrast, a negative correlation of social preferences
with Talk about investment, that is when rather selfish investors share the fact of their
SRI with others, would be consistent with an interpretation that these investors try to
signal pro-sociality. As these are otherwise selfish investors their investment in SRI
funds without tax benefits can then be viewed as being due to social reputation
concerns.
To better understand this we explore the covariates of social preferences, using a
regression analysis with Social preferences as dependent variable and Talk about
investment and a set of controls as independent variables. Table 6 reports the results.
The significant coefficient of Talk about investment is negative (p=0.001) which
indicates that it is the less pro-social investors who talk about their investment. Taking
into account the results from Section 4.1 (Table 4), this is indeed consistent with the
interpretation that relatively selfish investors who hold SRI funds without tax benefits
do this for social reputation reasons and signal it via talking about their investments. It
also means that we do not find support for the idea that investors with stronger social
preferences try to distinguish themselves from other investors by talking about their
33
investments. We consider this interpretation of our results as reasonable but hasten to
add that we are aware that there are many potential reasons for people to talk about
investments. Therefore, we view this signaling interpretation as interesting but only
suggestive that warrants further investigation in future research.
Of the other variables we find that the total value of an investor's portfolio (Log
total portfolio value) is negatively correlated with social preferences (p=0.081). Hence,
wealth in terms of portfolio predicts less pro-social preferences. In addition, Risk
preferences are statistically significant predictors (p=0.052), whereas the other variables
are insignificant. The positive coefficient of Risk preferences indicates that more prosocial investors also tend to be more risk tolerant.
In our study we have chosen a set-up that allows us to define social preferences
broadly without the need to distinguish between the many different forms of social
preferences proposed in the literature. This does not mean that we think that the
question what specific forms of social preferences investors, and people more generally,
have is not interesting. Quite to the contrary, we believe it to be an intriguing one.
Specifically in our context, altruism and warm-glow are two interesting forms among
many others proposed in the literature. From the literature on this topic (see, e.g.,
Engelmann and Strobel (2004, 2006), Bolton and Ockenfels (2006), Fehr, Naef and
Schmidt (2006), Blanco, Engelmann and Normann (2011)) it can be seen that the
question is far from easily answered and is still unsettled. Recently, it has been even
proposed that individuals do not adhere to one form of social preferences but actually
have mixed motives (Bolle, Breitmoser, Heimel and Vogel (2012)).
Our experiment is the first to elicit social preferences of individual investors.
Therefore, we have deliberately chosen for a set-up where different notions of social
preferences would (qualitatively) lead to the same measure of intrinsic social
preferences. This has the downside that we cannot distinguish between, e.g., warm-glow
and altruism. The few experiments that have been designed to disentangle warm glow
from other forms of intrinsic social preferences show that one needs a rather special
experimental design for that (Crumpler and Grossman (2008), Remoundou, Drichoutis
and Koundouri (2013), Ottoni-Wilhelm, Vesterlund and Xie (2014)). As our experiment
34
was designed with different research questions in mind, we have to leave the question
which specific form of intrinsic social preferences drives investors to buy SRI mutual
funds for future research.
5. Conclusions
We have measured social preferences of investors directly and in a controlled way using
a trust game experiment and showed that such non-financial preferences are predictive
for portfolio choice and, hence, not driven out in competitive financial markets.
Specifically, investors with stronger social preferences are more likely to hold socially
responsible investment (SRI) funds without tax incentives. Importantly, this result is
robust in that we control for a large set of other plausible predictors for SRI. We find
that (biased) return expectations and risk perceptions are not predictive for differences
in investment patterns, whereas the equity ratio and other portfolio characteristics of
investors are. In addition we find suggestive evidence that (some) rather selfish
investors hold SRI funds without tax benefits in order to signal pro-social attitudes.
Recently it has been observed that socially responsible investment is growing
(EUROSIF (2012), SIF (2012)). In our own dataset we see a statistically significant
increase in the SRI share from 13.57% in 2002 to 15.22% in 2011 (OLS, p<0.01).
Together with our results of the effect of social preferences, this indicates that stock
prices of socially responsible companies are likely affected in the long run. In
consequence, social preferences as well as preferences for social reputation may
influence pricing on financial markets, especially when the proportion of socially
responsible investors in the market increases further.
Knowledge about the role of social preferences for portfolio choices can also
provide insights into the role of investor preferences for other prominent examples of
deviations from the market portfolio, like the home and the local bias (Coval and
Moskowitz (1999), Ivković and Weisbenner (2005), Pool, Stoffman and Yonker (2012)),
and disproportional investments in employer stocks (Cohen (2009)) and the investor’s
car manufacturer (Keloharju, Knüpfer and Linnainmaa (2012)). Future research could
35
investigate whether these investment phenomena are also driven by preferences rather
than (biased) risk-return expectations.
We experimentally measured social preferences using a trust game and related it
to field behavior regarding investment choices. Hence, our paper also contributes to the
discussion about the stability of social preferences across different decision domains
(Karlan (2005), List (2006), Benz and Meier (2008), Falk and Heckman (2009), Baran,
Sapienza and Zingales (2010), Stoop, Noussair and Van Soest (2012)) and whether the
effect of experimentally elicited social preferences is nullified in a market environment
(Levitt and List (2007), List (2009), Bartling and Weber (2013)). In this respect it
should be noted that our estimates regarding the relation between behavior in the
experiment and in the field are on the conservative side. Investors in our study were
unaware that we matched their survey responses and experimental behavior to their
(anonymized) trading records. This mitigates the potential problem that socially
responsible investors want to behave consistently pro-socially in the experiment (for
evidence on consistency, see for instance Gneezy et al. (2012)). Moreover, our evidence
also shows that introducing extrinsic rewards such as tax benefits in the field could
eliminate the relation between prosocial field behavior and prosocial behavior in an
experiment without extrinsic rewards to prosocial behavior. Hence, when little prosocial behavior is observed in the field this does not necessarily imply that people are
not pro-social but rather that other incentives may have overwritten the pro-social
tendency.
We deliberately use a broad definition of social preferences as a first approach to
the question whether social preferences influence portfolio choice. Future research
could test how specific models of other-regarding preferences are related to socially
responsible investments. For instance, are socially responsible investors more altruistic
than conventional investors, are they more inequity averse, do they receive more warm
glow from doing good, or are they more reciprocal and want to give back something
positive to specific companies or society as a whole?
Our findings can have important practical implications for banks and mutual
fund providers. SRI funds without tax benefits primarily attract investors with strong
36
social preferences and selfish investors who want to signal pro-sociality. Thus, it seems
optimal to focus the marketing strategies for these products on the societal aspects
rather than the financial aspects. In fact, the intrinsic and signaling motivation of
socially responsible investors might be undermined by advertisements that are focused
too much on returns (see Gneezy and Rustichini (2000a,b), Ariely, Bracha and Meier
(2009) for crowding out effects in non-finance domains). SRI funds with tax benefits
attract more selfish investors. Therefore, if the government would decrease the tax
incentives these investors might stop investing in a socially responsible manner, in
particular, because investors in SRI funds with tax benefits report good investment
knowledge and might be well aware of outside investment opportunities. Accidentally,
after we have conducted our survey and experiments, the Dutch government decided to
reduce the tax benefits on SRI funds. This tax benefit reduction indeed caused an
outflow of money from these funds with tax benefits and a slight inflow to SRI funds
without tax benefits (Financieele Dagblad (2012)), providing additional casual support
of our findings and interpretation.
37
References
Anderson, O., Holm, H.J., Tyran, J.-R., & Wengström, E. (2013). “Risking Other
People’s Money: Experimental Evidence on Bonus Schemes, Competition and
Altruism”. IFN Working Papers 989, Research Institute of Industrial Economics,
Stockholm.
Ai, C., & Norton, E. C. (2003). “Interaction Terms in Logit and Probit Models”.
Economics Letters, 80(1), 123-129.
Andreoni, J. (1990). “Impure Altruism and Donations to Public Goods: a Theory of
Warm-Glow Giving”. The Economic Journal, 100(401), 464-477.
Ariely, D., Bracha, A., & Meier, S. (2009). “Doing Good or Doing Well? Image
Motivation and Monetary Incentives in Behaving Prosocially”. American
Economic Review, (99)1, 544-555.
Baran, N. M., Sapienza, P., & Zingales, L. (2010). “Can we Infer Social Preferences
from the Lab? Evidence from the Trust Game” (No. w15654). National Bureau
of Economic Research.
Barber, J. (2007). “Monitoring the Monitor: Evaluating CalPERS' Activism”. The
Journal of Investing, 16, 66-80.
Bardsley, N. (2008). “Dictator Game Giving: Altruism or Artefact?” Experimental
Economics, 11(2), 122-133.
Baron, D. P. (2007). “Corporate Social Responsibility and Social Entrepreneurship”.
Journal of Economics & Management Strategy, 16(3), 683-717.
Bartling, B., & Weber, R. A. (2013). “Do Markets Erode Social Responsibility?” (No.
4491). CESifo Working Paper available at SSRN: http://ssrn.com/abstract=2369463
or http://dx.doi.org/10.2139/ssrn.2369463.
Bauer, R., Koedijk, K., & Otten, R. (2005). “International Evidence on Ethical Mutual
Fund Performance and Investment Style”. Journal of Banking & Finance, 29(7),
1751-1767.
Bénabou, R., & Tirole, J. (2006). “Incentives and Prosocial Behavior”. American
Economic Review, 96(5), 1652–1678.
Bénabou, R., & Tirole, J. (2010). “Individual and Corporate Social Responsibility”.
Economica, 77(305), 1-19.
Benz, M., & Meier, S. (2008). “Do People Behave in Experiments as in the Field?—
Evidence from Donations”. Experimental Economics, 11(3), 268-281.
Berg, J., Dickhaut, J., & McCabe, K. (1995). “Trust, Reciprocity, and Social History”.
Games and Economic Behavior, 10(1), 122-142.
38
Blanco, M., Engelmann, D., & Normann, H.T. (2011). “A Within-subject Analysis of
Other-regarding Preferences“. Games and Economic Behavior, 72(2), 321-338.
Bolle, F., Breitmoser, Y., Heimel, J., & Vogel, C. (2012). “Multiple Motives of Prosocial Behavior: Evidence from the Solidarity Game”. Theory and Decision,
72(3), 303-321.
Bollen, N. P. (2007). “Mutual Fund Attributes and Investor Behavior”. Journal of
Financial and Quantitative Analysis, 42(03), 683-708.
Bolton, G. E., & Ockenfels, A. (2000). “ERC: A Theory of Equity, Reciprocity, and
Competition”. American Economic Review, 90(1), 166-193.
Bolton, G. E., & Ockenfels, A. (2006). “Inequality Aversion, Efficiency, and Maximin
Preferences in Simple Distribution Experiments: Comment”. The American
Economic Review, 96(5), 1906-1911.
Calvet, L. E., Campbell, J. Y., & Sodini, P. (2009). “Measuring the Financial
Sophistication of Households”, American Economic Review, 99(2), 393–398.
Camerer, C. F., & Hogarth, R. M. (1999). “The Effects of Financial Incentives in
Experiments: A Review and Capital-labor-production Framework”. Journal of
Risk and Uncertainty, 19(1-3), 7-42.
Cohen, L. (2009). “Loyalty-based Portfolio Choice”. Review of Financial Studies,
22(3), 1213-1245.
Coval, J. D., & Moskowitz, T. J. (1999). “Home Bias at Home: Local Equity Preference
in Domestic Portfolios”. The Journal of Finance, 54(6), 2045-2073.
Cox, J. C. (2004). “How to Identify Trust and Reciprocity”. Games and Economic
Behavior, 46(2), 260-281.
Crumpler, H., & Grossman, P. J. (2008). “An Experimental Test of Warm Glow
Giving”. Journal of Public Economics, 92(5), 1011-1021.
Derwall, J., Günster, N., Bauer, R., & Koedijk, K. (2005). “The Eco-efficiency Premium
Puzzle”. Financial Analysts Journal, 61(2), 51-63.
Derwall, J., Koedijk, K., & Ter Horst, J. (2011). “A Tale of Values-driven and Profitseeking Social Investors”. Journal of Banking & Finance, 35(8), 2137-2147.
Di Giuli, A., & Kostovetsky, L. (2014). “Are Red or Blue Companies More Likely to
Go Green? Politics and Corporate Social Responsibility”. Journal of Financial
Economics, 111(1), 158-180.
Dohmen, T., Falk, A., Huffmann, D., Schupp, J., Sunde, U., & Wagner, G. (2011).
“Individual Risk Attitudes: Measurement, Determinants and Behavioural
Consequences”. Journal of the European Economics Association, 9(3), 522–550.
39
Dorn, D., & Huberman, G. (2005) “Talk and Action: What Individual Investors Say and
What They Do”. Review of Finance, 9(4), 437–481.
Dufwenberg, M., Heidhues, P., Kirchsteiger, G., Riedel, F., & Sobel, J. (2011). “OtherRegarding Preferences in General Equilibrium”. Review of Economic Studies,
78(2), 613-639.
Edmans, A. (2011). “Does the Stock Market Fully Value Intangibles? Employee
Satisfaction and Equity Prices”. Journal of Financial Economics, 101(3), 621–
640.
Egas, M., & Riedl, A. (2008). “The Economics of Altruistic Punishment and the
Maintenance of Cooperation”. Proceedings of the Royal Society B: Biological
Sciences, 275, 871-878.
Ellingsen, T., & Johannesson, M. (2008). “Pride and Prejudice: The Human Side of
Incentive Theory.” American Economic Review, 98(3), 990-1008.
Engelmann, D., & Strobel, M. (2004). “Inequality Aversion, Efficiency, and Maximin
Preferences in Simple Distribution Experiments”. American Economic Review,
94(4), 857-869.
Engelmann, D., & Strobel, M. (2006). “Inequality Aversion, Efficiency, and Maximin
Preferences in Simple Distribution Experiments: Reply”. American Economic
Review, 96(5), 1918-1923.
EUROSIF, 2012, European SRI study 2012, http://www.eurosif.org.
Fabozzi, F.J., Ma, K.C., & Oliphant, B.J. (2008). “Sin Stock Returns”. Journal of
Portfolio Management, 35(1), 82-74.
Falk, A., & Heckmann, J. J. (2009) “Lab Experiments are a Major Source of Knowledge
in the Social Sciences”. Science, 326(5952), 535–538.
Falk, A., & Zehnder, C. (2013). “A City-Wide Experiment on Trust Discrimination”.
Journal of Public Economics, 100, 15-27.
Falk, A., Meier, S., & Zehnder, C. (2013). “Do Lab Experiments Misrepresent Social
Preferences? The Case of Self-Selected Student Samples”. Journal of the
European Economic Association, 11(4), 839-852.
Fama, E. F., & French, K. R., (2007). “Disagreement, Tastes, and Asset Prices”, Journal
of Financial Economics, 83(3), 667-689.
Fehr, E., & Gächter, S. (2000). “Cooperation and Punishment in Public Goods
Experiments”. American Economic Review, 90(4), 980-994.
Fehr, E., & List, J. A. (2004). “The Hidden Costs and Returns of Incentives – Trust and
Trust-worthiness among CEO’s”. Journal of the European Economic
Association, 2(5), 743–771.
40
Fehr, E., & Schmidt, K. M. (1999). “A Theory of Fairness, Competition and
Cooperation”. Quarterly Journal of Economics, 114(3), 817-868.
Fehr, E., Naef, M., & Schmidt, K. M. (2006). “Inequality Aversion, Efficiency, and
Maximin Preferences in Simple Distribution Experiments: Comment”. American
Economic Review, 96(5), 1912-1917.
Fehrler, S., & Przepiorka, W. (2013). “Charitable Giving as a Signal of Trustworthiness:
Disentangling the Signaling Benefits of Altruistic Acts”. Evolution and Human
Behavior, 34(2), 139-145.
Financieele Dagblad, 2012, “Ook de Groene Belegger Zoekt Rendement” (Also Green
Investors Seek for Returns), Dutch newspaper article.
Fowler, J. H., & Kam, C. D. (2007). „Beyond the Self: Social Identity, Altruism, and
Political Participation”. Journal of Politics, 69(3), 813-827.
Glazer, A., & Konrad, K. A. (1996). „A Signaling Explanation for Charity”. The
American Economic Review, 86(4), 1019-1028.
Gneezy, A., Imas, A., Nelson, L. D., Brown, A., & Norton, M. I. (2012). “Paying to Be
Nice: Consistency and Costly Prosocial Behaviour”. Management Science,
58(1), 179–187.
Gneezy, U. & Rustichini, A. (2000a). “A Fine Is a Price”. Journal of Legal Studies,
29(1), 1–17.
Gneezy, U., & Rustichini, A. (2000b). “Pay Enough or Don’t Pay at All”. Quarterly
Journal of Economics, 115(3), 791–810.
Gollier, C., & Pouget, S. (2014). “The “Washing Machine”: Investment Strategies and
Corporate Behavior with Socially Responsible Investors”. Working paper.
Graham, J. R., Harvey, C. R., & Huang, H. (2009). “Investor Competence, Trading
Frequency, and Home Bias”. Management Science, 55(7), 1094–1106.
Guiso, L, Sapienza, P., & Zingales, L. (2013). “The Determinants of Attitudes Toward
Strategic Default on Mortgages”. Journal of Finance, 68, 1473-1515.
Harrison, G. W., & List, J. A. (2004). “Field Experiments”. Journal of Economic
Literature 42(4), 1009-1055.
Heinkel R., Kraus, A. & Zechner, J. (2001). “The Effect of Green Investment
on
Corporate Behavior”. Journal of Financial and Quantitative Analysis, 36, 431449.
Holt, C. A. & Laury, S. K. (2002). “Risk Aversion and Incentive Effects”. American
Economic Review, 92(5), 1644–1655.
Hong, H., & Kacperczyk, M. (2009). “The Price of Sin: The Effects of Social Norms on
Markets”. Journal of Financial Economics, 93(1), 15–36.
41
Hong, H., & Kostovetsky, L. (2012). “Red and Blue Investing: Values and Finance”.
Journal of Financial Economics, 103(1), 1–19.
Ivkovic, Z., & Weisbenner, S. (2005). “Local Does as Local is: Information Content of
the Geography of Individual Investors' Common Stock Investments”. Journal of
Finance, 60(1), 267–306.
Karlan, D. S. (2005). “Using Experimental Economics to Measure Social Capital and
Predict Financial Decisions”. American Economic Review, 95(5), 1688–1699.
Kaustia, M., & Torstila, S. (2011) “Stock Market Aversion? Political Preferences and
Stock Market Participation”. Journal of Financial Economics, 100(1), 98–112.
Keloharju, M., Knüpfer, S., & Linnainmaa, J. T. (2012). “Do Investors Buy What They
Know? Product Market Choices and Investment Decisions”. Review of Financial
Studies, 25(10), 2921–2958.
Kempf, A., & Osthoff, P. (2007). “The Effect of Socially Responsible Investing on
Portfolio Performance”. European Financial Management 13, 908-922.
Kreps, D.M., Milgrom, P., Roberts, J., & Wilson, R. (1982). “Rational Cooperation in
the Finitely Repeated Prisoners' Dilemma”. Journal of Economic Theory, 27,
245-252.
Krüger, P. (forthcoming). “Corporate Goodness and Shareholder Wealth”, Journal of
Financial Economics.
Kumar, A. & Page, J. (forthcoming). “Deviations from Norms and Informed Trading”.
Journal of Financial and Quantitative Analysis.
Ledyard, J. O. (1995). “Public Goods: A Survey of Experimental Research”. In
Handbook of Experimental Economics, edited by J. Kagel and A. Roth,
Princeton University Press, 1995.
Levitt, S.D. & List, J.A. (2007). “What Do Laboratory Experiments Measuring Social
Preferences Reveal About the Real World?” Journal of Economic Perspectives,
21, 153-174.
List, J. A. (2006). “The Behavioralist Meets the Market: Measuring Social Preferences
and Reputation Effects in Actual Transactions”. Journal of Political Economy,
114(1), 1–37.
List, J. A. (2007). “On the Interpretation of Giving in Dictator Games”. Journal of
Political Economy, 115(3), 482-493.
List, J. A. (2009). “Social Preferences: Some Thoughts from the Field”. Annual Review
of Economics, 1, 563-579.
Millward Brown, 2011, Retail Investor 2011, Yearly publication.
42
Nelson, P. J., & Greene, K. V. (2003). “Signaling Goodness: Social Rules and Public
Choice”. University of Michigan Press.
Oosterbeek, H., Sloof, R., & Van De Kuilen, G. (2004). “Cultural Differences in
Ultimatum Game Experiments: Evidence from a Meta-analysis”. Experimental
Economics, 7(2), 171-188.
Ottoni-Wilhelm, M., Vesterlund, L., & Xie, H. (2014). “Why Do People Give? Testing
Pure and Impure Altruism” (No. w20497), National Bureau of Economic
Research.
Pool, V., Stoffman, N. & Yonker, S. (2012). “No Place Like Home: Familiarity in
Mutual Fund Manager Portfolio Choice”. Review of Financial Studies, 25(8),
2563-2599.
Rabin, M. (1993). “Incorporating Fairness into Game Theory and Economics”.
American Economic Review, 83(5), 1281-1302.
Remoundou, K., Drichoutis, A. C., & Koundouri, P. (2013). “Isolating Warm Glow in
Charitable Auction Giving” (No. 131). Grantham Research Institute on Climate
Change and the Environment.
Renneboog, L., Ter Horst, J., & 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.
Selten, R. (1967). “Die Strategiemethode zur Erforschung des Eingeschränkt Rationalen
Verhaltens in Rahmen Eines Oligopolexperiments”. In Sauerman, H. (ed),
“Beiträge Zur Experimentellen Wirtschaftsforschung”. Tübingen: J.C.B., Mohr
(Paul Siebeck).
SIF (Social Investment Forum) (2012). “Report on Socially Responsible Investing
Trends in the United States”. http://www.socialinvest.org.
Sobel, J. (2005). “Interdependent Preferences and Reciprocity”. Journal of Economic
Literature, 43(2), 392–436.
Stoop, J.T.R., Noussair, Ch.N., & van Soest, D.P. (2012). “From the Lab to the Field:
Cooperation Among Fishermen”. Journal of Political Economy, 120, 1027-1056.
Van Rooij, M., Lusardi, A., & Alessie, R. (2011). “Financial Literacy and Stock Market
Participation”. Journal of Financial Economics, 101(2), 449-472.
Zivin, J. G., & Small, A. (2005). “A Modigliani-Miller Theory of Altruistic Corporate
Social Responsibility”. Topics in Economic Analysis & Policy, 5(1), article 10.
43
Table 1 – Variables
Survey
Administrative
Type
Variable
Socially responsible investor
Measure
Dummy variable equal to 1 if an investor holds a socially
responsible (SRI) mutual fund in his portfolio at the
provider in the month that he participated in the
experiments.
Total amount invested in SRI funds at the provider as a
Percentage SRI in total portfolio
percentage of the total portfolio at the provider, in the
month in which the investor participated in the
experiments.
Dummy variable equal to 1 if an investor only holds a
Only holds SRI with tax benefits
SRI fund with tax benefits in the month he participated
in the experiments.
Only holds SRI without tax benefits Dummy variable equal to 1 if an investor only holds a
SRI fund without tax benefits in the month he
participated in the experiments.
Holds both SRI with and without Dummy variable equal to 1 if an investor holds both a
SRI fund with and without tax benefits in the month he
tax benefits
participated in the experiments.
Percentage of the overall portfolio that is invested in
Equity ratio
equity funds (risky share).
The number of transactions the investor made in the 12
Number of transactions
months before s/he participated in our experiment. To
account for extremes, we trim this measure by excluding
the 1th and the 99th percentile.
Total euro amount invested at the provider in the month
Total Portfolio Value
that the investor participated in the experiments.
Average portfolio returns per month since the investor
Mean portfolio returns
opened his or her account (in percent). To account for
extremes, we trim this measure by excluding the 1th and
the 99th percentile.
Standard deviation of the monthly portfolio returns since
Volatility portfolio returns
the investor opened her account (in percent). To account
for extremes, we trim this measure by excluding the 1th
and the 99th percentile.
Response to statement “I expect that the returns of
Expected returns on SRI
socially responsible equity funds compared to
conventional equity funds are:
much lower
a bit lower
the same
a bit higher
much higher
I do not know”
(much lower 1-5 much higher)
Response to statement “Socially responsible equity funds
Perceived risk on SRI
are more risky than conventional equity funds” (fully
disagree 1-7 fully agree)
Dummy equal to 1 if an investor believes that the returns
Lower expected returns on SRI
44
Lower perceived risk on SRI
Investment knowledge
University degree
Talk about investments
Female
Age
Low income
Medium income
High income
Incentivized experiment
Untold income
Risk preferences
Mean intrinsic social preferences
(average return ratio)
Max intrinsic social preferences
(return at maximally possible
transfer of 50 euro)
on SRI equity funds are a bid or much lower than on
conventional equity funds.
Dummy equal to 1 if an investor mildly to fully agrees
that the risk on SRI equity funds is lower than the risk on
conventional equity funds.
Response to statement “My investment knowledge is
good” (fully disagree 1-7 fully agree)
The investor reports to have completed a university
degree
Response to statement “I often talk about investment
with others” (fully disagree 1-7 fully agree)
Investor is a woman
Age of investor
Investor's gross family income is below 60,000 euro per
year
Investor's gross family income is between 60,000 euro
and 100,000 euro per year
Investor's gross family income is above 100,000 euro per
year
Investor does not disclose his income
Amount at which the investor switches from choosing the
risky lottery to choosing the risk-free option in the risk
preference task. A higher amount indicates more risk
tolerance.
Average return ratio across the transfer ranges from 0 to
3. It is calculated as follows. First the return ratio for
each possible first-mover transfer in the trust game is
calculated. That is, if the first mover sends 5 euro, the
amount the second mover returns is divided by 5, if the
first mover sends 10 euro the amount the second mover
returns is divided by 10, and so on. Second, the average
of these ratios across the range of 5 to 50 euro firstmover transfers is calculated.
Amount a second-mover investor sends back for the
maximally possible transfer (50 euro) a first-mover could
send in the trust game. Ranges from 0 to 150.
45
Table 2 – Summary statistics comparison socially responsible and conventional investors
This table presents the summary statistics for socially responsible and conventional investors
separately. All variables are defined in Table 1. If not otherwise indicated in Table 1, the
statistics represent the portfolios of investors in the month in which they participated in the
experiment and the survey. Standard deviations are in parentheses. P-values are from two-sided
Mann-Whitney tests (a) or Chi-square tests (b).
Socially responsible investors
(14%)
Mean
Median
N
Percentage SRI in total portfolio
Only holds SRI with tax benefits
Only holds SRI without tax benefits
Holds both SRI with and without tax benefits
Equity ratio
Number of transactions
Total portfolio value
Mean portfolio returns
Volatility portfolio returns
Expected returns on SRI
Perceived risk on SRI
Lower expected returns on SRI
Lower perceived risk on SRI
Investment knowledge
University degree
Talk about investments
Female
Age
Low income
High income
Untold income
Risk preferences
Mean intrinsic social preferences
Max intrinsic social preferences
14.9%
19.5%
68.4%
12.1%
55.0%
(28.2)
25.4
(42.4)
106,678
(190,033)
0.2094
(0.4509)
4.3258
(2.3130)
2.7055
(1.0041)
3.5377
(1.2653)
0.5193
(0.5001)
0.4345
(0.4962)
4.1916
(1.3073)
0.4990
(0.5005)
3.1122
(1.5123)
0.1810
(0.3852)
57.7542
(12.1359)
0.3223
(0.4678)
0.2087
(0.4068)
0.1798
(0.3844)
113.2909
(41.5708)
1.5269
(0.6695)
77.4643
(34.4927)
46
7.86%
-
747
747
747
747
52.1%
747
10
747
57,666
747
0.2647
747
3.8065
747
2
506
4
504
1
491
0
504
4
642
0
487
3
642
0
746
57
716
0
484
0
484
0
484
110
550
1.73
140
100
140
Conventional investors
(86%)
Mean
Median
N
51.3%
(32.0)
22.8
(45.5)
73,250
(127,344)
0.2363
(0.4914)
3.9643
(2.3491)
2.8206
(1.3229)
3.5742
(1.2458)
0.5974
(0.4905)
0.3895
(0.4877)
3.8276
(1.4734)
0.4622
(0.4987)
2.9093
(1.5217)
0.2094
(0.4069)
59.1040
(11.6319)
0.3228
(0.4676)
0.1965
(0.3974)
0.1772
(0.3819)
112.2275
(42.8290)
1.4173
(0.6749)
71.6121
(34.8929)
difference
P
47.6%
4494
0.0027a
6
4494
0.000a
36,496
4494
0.000a
0.2913
4494
0.0378a
3.5142
4494
0.0000a
2
2776
0.692a
4
2750
0.385a
1
2489
0.001b
0
2750
0.057b
4
3881
0.000a
0
2609
0.136b
3
3881
0.001a
0
4585
0.075b
59
4377
0.002a
0
2590
0.984b
0
2590
0.538b
0
2590
0.894b
110
3129
0.723a
1.51
763
0.087a
80
763
0.061a
Table 3 – Determinants of likelihood to own SRI and of percentage of SRI holding
In this table Model Probit presents marginal effects of a probit regression in which the
dependent variable takes on the value of 1 if an investor holds an SRI mutual fund in the month
investors participated in the experiment and survey; 0 otherwise. Model Tobit presents marginal
effects of Tobit regressions in which the dependent variable is the percentage of the portfolio
that is held in SRI mutual fund in the month investors participated in the experiment and survey.
The Tobit regression accounts for left-censoring at 0% and right-censoring at 100%. In both
models Social preferences are measured as Mean intrinsic social preferences. Results for Max
intrinsic social preferences can be found in the appendix (Table A3). All variables are defined in
Table 1. Standard errors are in parentheses. * is 10% ** is 5% and *** is 1% significance.
Model
Dependent variable
PREFERENCES
Social preferences
BELIEFS
Lower expected returns on SRI
Lower perceived risk on SRI
PORTFOLIO CHARACTERISTICS
Equity ratio
Log number of transactions
Log total portfolio value
Mean portfolio returns
Volatility portfolio returns
OTHER CONTROLS
Investment knowledge
University degree
Risk preferences
Female
Age
Low income
High income
Untold income
Probit
Likelihood SRI
Tobit
Percentage SRI
0.0495**
(0.0238)
4.0306*
(2.3914)
-0.0296
(0.0308)
-0.0247
(0.0307)
-3.4272
(3.0383)
-0.9505
(3.0904)
0.0765
(0.0551)
0.0277**
(0.0113)
0.0390***
(0.0125)
-0.0120
(0.0257)
0.0020
(0.0065)
11.3187**
(5.5489)
1.8416
(1.1528)
2.8033**
(1.2393)
-2.3101
(2.5159)
0.2555
(0.6463)
0.0072
(0.0139)
0.0434
(0.0325)
-0.0003
(0.0004)
0.0031
(0.0456)
-0.0017
(0.0015)
0.0452
(0.0418)
-0.0492
(0.0396)
0.0284
(0.0487)
-0.0982
(1.3893)
5.2613
(3.2348)
-0.0199
(0.0392)
0.6038
(4.4640)
-0.1663
(0.1496)
7.1881*
(3.9064)
-4.4087
(4.5014)
5.0211
(4.5174)
Talk about investments
0.0269**
(0.0116)
#
2.1587*
(1.1554)
Constant
-70.2326***
(17.2477)
N
644
644
LR chi-square(17)
43.18
32.74
Prob > chi-square
0.0005
0.0121
Log-likelihood
-295.33
-758.52
# table shows marginal effects of probit regression, constant not reported
Table 4 – Determinants of likelihood to own SRI fund with and without tax benefits
This table presents relative-risk ratios of a multinomial logit regression in which the dependent
variable can take on four different values. The baseline group (not reported) consists of
conventional investors, the second group of investors who only hold SRI funds with tax
benefits, the third group only holds SRI funds without tax benefits, and the fourth group holds
both types of SRI funds. Social preferences is measured as Mean intrinsic social preferences.
Results for Max intrinsic social preferences can be found in the appendix (Table A4). All
variables are defined in Table 1. Standard errors are in parentheses. * is 10% ** is 5% and ***
is 1% significance.
PREFERENCES
Social preferences
BELIEFS
Lower expected returns on SRI
Lower perceived risk on SRI
PORTFOLIO CHARACTERISTICS
Equity ratio
Log number of transactions
Log total portfolio value
Mean portfolio returns
Volatility portfolio returns
OTHER CONTROLS
Investment knowledge
University degree
Risk preferences
Female
Age
Low income
High income
with tax
benefits
(1)
SRI funds
without tax
benefits
(2)
with and without
tax benefits
(3)
0.7256
(0.2641)
1.5462**
(0.2897)
2.2876
(1.3265)
1.9883
(1.1015)
1.6025
(0.8340)
0.7189
(0.1671)
0.6944
(0.1661)
0.5510
(0.3673)
1.0081
(0.6772)
0.1838
(0.1928)
1.6044**
(0.3221)
1.3039
(0.3145)
0.3223**
(0.1520)
0.8617
(0.1189)
2.2162*
(0.9309)
1.1280
(0.0962)
1.2647**
(0.1206)
1.0205
(0.2044)
1.0334
(0.0506)
3.6367
(4.8304)
1.3477
(0.3218)
1.8753*
(0.6132)
1.9393
(1.4173)
1.0224
(0.1345)
2.0271***
(0.5426)
0.3878
(0.2242)
0.9972
(0.0062)
0.8261
(0.7008)
1.0491*
(0.0267)
0.9369
(0.5914)
0.0000
(0.0002)
0.9783
(0.1045)
1.4862
(0.3723)
0.9985
(0.0030)
0.8022
(0.3000)
0.9829
(0.0113)
1.3622
(0.4139)
0.8963
(0.2957)
0.9068
(0.2397)
5.4660**
(4.6855)
0.9963
(0.0087)
5.0530**
(3.8057)
0.9663
(0.0323)
2.0291
(1.6369)
0.2416
(0.2845)
Untold income
Talk about investments
Constant
N
LR chi-square(51)
Prob > chi-square
Log-likelihood
1.2224
(0.8466)
1.1784
(0.2167)
0.0000***
(0.0000)
1.1650
(0.4099)
1.1881*
(0.1054)
0.0060***
(0.0077)
644
100.78
0.0000
-355.13
0.9042
(0.8812)
1.4560
(0.3683)
0.0000***
(0.0000)
Table 5 – Equity ratio and social preferences
This table presents Tobit (specifications 1–3) and OLS (specification 4) regressions in which the
dependent variable is the Equity ratio in an investor's portfolio. The explanatory variable of
main interest is Social preferences (all specifications). In specifications 2-4 control variables are
added. Tobit regressions account for left (right) censoring of the variable Equity ratio at 0 (1).
Social preferences is measured as Mean intrinsic social preferences. All variables are defined in
Table 1. Standard errors are in parentheses. * is 10% ** is 5% and *** is 1% significance.
PREFERENCES
Social preferences
Tobit
(1)
Tobit
(2)
Tobit
(3)
OLS
(4)
0.0032
(0.0163)
-0.0052
(0.0179)
-0.0114
(0.0196)
-0.0129
(0.0191)
-0.0058
(0.0770)
0.0266
(0.0462)
-0.0158
(0.0751)
0.0330
(0.0451)
0.0081
(0.0084)
-0.0179**
(0.0075)
0.0285
(0.0205)
0.0141***
(0.0050)
0.0070
(0.0084)
-0.0188**
(0.0075)
0.0285
(0.0205)
0.0139***
(0.0050)
0.0069
(0.0082)
-0.0189**
(0.0073)
0.0293
(0.0200)
0.0126**
(0.0049)
0.0318***
(0.0102)
0.0080
(0.0246)
-0.0002
(0.0003)
-0.0527
(0.0325)
-0.0010
(0.0012)
-0.0116
(0.0302)
-0.0494
(0.0334)
-0.0545
(0.0348)
-0.0051
(0.0090)
0.6266***
(0.1101)
0.0314***
(0.0102)
0.0065
(0.0246)
-0.0002
(0.0003)
-0.0517
(0.0325)
-0.0010
(0.0012)
-0.0128
(0.0302)
-0.0477
(0.0334)
-0.0543
(0.0348)
-0.0057
(0.0090)
0.6433***
(0.1109)
0.0298***
(0.0099)
0.0055
(0.0240)
-0.0001
(0.0003)
-0.0474
(0.0317)
-0.0009
(0.0011)
-0.0114
(0.0295)
-0.0439
(0.0326)
-0.0498
(0.0340)
-0.0053
(0.0088)
0.6459***
(0.1080)
SRI & INTERACTION
SRI
SRI * Social preferences
PORTFOLIO CHARACTERISTICS
Log number of transactions
Log total portfolio value
Mean portfolio returns
Volatility portfolio returns
OTHER CONTROLS
Investment knowledge
University degree
Risk preferences
Female
Age
Low income
High income
Untold income
Talk about investments
Constant
0.5532***
(0.0258)
N
LR chi-square
Prob > chi-square
Log-likelihood
Adj. R-squared
900
0.04
0.8457
-318.20
673
42.62
0.0001
-165.92
673
44.28
0.0002
-165.08
673
0.0388
Table 6 – Covariates of social preferences
This table presents Tobit regression results where Social preferences is the dependent variable.
The regression accounts for left-censoring at 0 (the minimum ratio that could be sent back in the
trust game) and right-censoring at 3 (the maximum ratio that could be sent back in the trust
game). Social preferences is measured as Mean intrinsic social preferences. All variables are
defined in Table 1. Standard errors are in parentheses. * is 10% ** is 5% and *** is 1%
significance.
Model
Dependent variable
PORTFOLIO CHARACTERISTICS
Equity ratio
Log number of transactions
Log total portfolio value
Mean portfolio returns
Volatility portfolio returns
OTHER CONTROLS
Investment knowledge
University degree
Risk preferences
Female
Age
Low income
High income
Untold income
Talk about investments
Constant
N
LR chi-square(14)
Prob > chi-square
Log-likelihood
Tobit
Social Preferences
-0.0405
(0.0955)
0.0110
(0.0200)
-0.0353**
(0.0179)
0.0153
(0.0490)
0.0092
(0.0120)
0.0252
(0.0244)
0.0129
(0.0588)
0.0014**
(0.0007)
-0.0394
(0.0778)
-0.0011
(0.0028)
-0.0420
(0.0723)
-0.0174
(0.0798)
-0.0626
(0.0832)
-0.0690***
(0.0214)
1.8034***
(0.2592)
673
19.99
0.1304
-745.41
Figure 1 – Return expectations of SRI funds
This figure presents the separate distributions of return expectations for socially responsible and
conventional investors. The variable ‘Expected return SRI’ depicts the response to the statement
‘I expect that the returns of socially responsible equity funds compared to conventional equity
funds are:’ from ‘much lower’ to ‘much higher’. The category ‘I don’t know’ is excluded from
the figure; it was chosen by 3% of the socially responsible and 10.3% of the conventional
investors.
Figure 2 – Risk perceptions of SRI funds
This figure presents the separate distributions of risk perceptions of SRI funds for socially
responsible and conventional investors. The variable ’Expected risk SRI’ is the response of
investors on a 1-7 Likert scale to the statement: “Socially responsible equity funds are more
risky than conventional equity funds” where 1 is fully disagree and 7 fully agree.
Figure 3 – Intrinsic social preferences
This figure shows the average return ratio for second-movers in the trust game for each possible
positive transfer by the first mover. The average return ratio is calculated for socially
responsible and conventional investors separately. We used the strategy method to elicit these
return ratios as described in Section 3.4. A return ratio of 1 means that the second mover sends
back exactly the amount received from the first mover.
Appendix A: Supplementary tables and figures
Table A1 – Respondents and overall sample characteristics
This table compares the mean characteristics of all invited investors to those for the respondents
to the survey and experiments. The variables are defined in Table 1. Note that for our research
design, we on purpose oversampled socially responsible investors in the survey to increase the
power of our analyses in which we compare SR to conventional investors. The response rate for
SR investors is 12% and that for conventional investors is 8%
Female
Age
Total portfolio value (euro)
% Holds only SRI funds
without tax benefits
% Holds only SRI fund with
tax benefits
% Holds SRI funds with and
without tax benefits
Invited sample
(n = 39,379)
24.7%
55.5
61,509
Respondents
(n = 3,254)
20.6%
57.9
74,259
7.6%
10.2%
1.8%
2.9%
0.6%
0.8%
Table A2 – Choice list in risk preferences elicitation experiment
1)
2)
3)
4)
5)
6)
7)
8)
9)
10)
11)
12)
13)
14)
15)
16)
17)
18)
19)
20)
Safe Payment
€0 for sure
€10 for sure
€20 for sure
€30 for sure
€40 for sure
€50 for sure
€60 for sure
€70 for sure
€80 for sure
€90 for sure
€100 for sure
€110 for sure
€120 for sure
€130 for sure
€140 for sure
€150 for sure
€160 for sure
€170 for sure
€180 for sure
€190 for sure
or
or
or
or
or
or
or
or
or
or
or
or
or
or
or
or
or
or
or
or
Lottery
50 percent chance of winning €300 and 50 percent chance of winning €0
50 percent chance of winning €300 and 50 percent chance of winning €0
50 percent chance of winning €300 and 50 percent chance of winning €0
50 percent chance of winning €300 and 50 percent chance of winning €0
50 percent chance of winning €300 and 50 percent chance of winning €0
50 percent chance of winning €300 and 50 percent chance of winning €0
50 percent chance of winning €300 and 50 percent chance of winning €0
50 percent chance of winning €300 and 50 percent chance of winning €0
50 percent chance of winning €300 and 50 percent chance of winning €0
50 percent chance of winning €300 and 50 percent chance of winning €0
50 percent chance of winning €300 and 50 percent chance of winning €0
50 percent chance of winning €300 and 50 percent chance of winning €0
50 percent chance of winning €300 and 50 percent chance of winning €0
50 percent chance of winning €300 and 50 percent chance of winning €0
50 percent chance of winning €300 and 50 percent chance of winning €0
50 percent chance of winning €300 and 50 percent chance of winning €0
50 percent chance of winning €300 and 50 percent chance of winning €0
50 percent chance of winning €300 and 50 percent chance of winning €0
50 percent chance of winning €300 and 50 percent chance of winning €0
50 percent chance of winning €300 and 50 percent chance of winning €0
Table A3 – Determinants of likelihood to own SRI and of percentage of SRI holding
(Max intrinsic social preferences)
In this table Model Probit presents marginal effects of a probit regression in which the
dependent variable takes on the value of 1 if an investor holds an SRI mutual fund in the month
investors participated in the experiment and survey; 0 otherwise. Model Tobit presents marginal
effects of Tobit regressions in which the dependent variable is the percentage of the portfolio
that is held in SRI mutual fund in the month investors participated in the experiment and survey.
The Tobit regression accounts for left-censoring at 0% and right-censoring at 100%. In both
models Social preferences is defined as Max intrinsic social preferences measured in units of 10
euro. All variables are defined in Table 1. Standard errors are in parentheses. * is 10% ** is 5%
and *** is 1% significance.
Model
Dependent variable
PREFERENCES
Social preferences
BELIEFS
Lower expected returns on SRI
Lower perceived risk on SRI
PORTFOLIO CHARACTERISTICS
Equity ratio
Log number of transactions
Log total portfolio value
Mean portfolio returns
Volatility portfolio returns
OTHER CONTROLS
Investment knowledge
University degree
Risk preferences
Female
Age
Low income
High income
Probit
Likelihood SRI
Tobit
Percentage SRI
0.0093**
(0.0046)
0.7195
(0.4619)
-0.0292
(0.0308)
-0.0242
(0.0307)
-3.3899
(3.0396)
-0.8892
(3.0913)
0.0753
(0.0551)
0.0275**
(0.0114)
0.0384***
(0.0124)
-0.0121
(0.0257)
0.0019
(0.0065)
11.2635**
(5.5490)
1.8340
(1.1534)
2.7504**
(1.2343)
-2.3213
(2.5163)
0.2445
(0.6472)
0.0069
(0.0140)
0.0429
(0.0325)
-0.0003
(0.0004)
0.0028
(0.0456)
-0.0017
(0.0015)
0.0454
(0.0419)
-0.0485
(0.0397)
-0.1159
(1.3918)
5.2264
(3.2381)
-0.0196
(0.0393)
0.5520
(4.4695)
-0.1693
(0.1497)
7.1800*
(3.9098)
-4.3356
(4.5010)
Untold income
0.0280
(0.0486)
0.0266**
(0.0116)
#
4.9786
(4.5191)
Talk about investments
2.1272*
(1.1552)
Constant
-68.7024***
(17.0486)
N
644
644
LR chi-square (17)
42.92
32.31
Prob > chi-square
0.0005
0.0137
Log-likelihood
-295.4594
-758.7329
# table shows marginal effects of probit regression, constant not reported
Table A4 – Determinants of the likelihood to own SRI fund with and without tax benefits
(Max intrinsic social preferences)
This table presents relative-risk ratios of a multinomial logit regression in which the dependent
variable can take on four different values. The baseline group (not reported) consists of
conventional investors, the second group of investors who only hold SRI funds with tax
benefits, the third group only holds SRI funds without tax benefits, and the fourth group holds
both types of SRI funds. Social preferences is defined as Max intrinsic social preferences and
measured in units of 10 euro. All variables are defined in Table 1. Standard errors are in
parentheses. * is 10% ** is 5% and *** is 1% significance.
PREFERENCES
Social preferences
BELIEFS
Lower expected returns on SRI
Lower perceived risk on SRI
PORTFOLIO CHARACTERISTICS
Equity ratio
Log number of transactions
Log total portfolio value
Mean portfolio returns
Volatility portfolio returns
OTHER CONTROLS
Investment knowledge
University degree
Risk preferences
Female
Age
Low income
High income
with tax
benefits
(1)
SRI funds
without tax
benefits
(2)
with and without
tax benefits
(3)
0.9535
(0.0702)
1.0797**
(0.0364)
1.1767
(0.1181)
1.9546
(1.0768)
1.6154
(0.8393)
0.7224
(0.1678)
0.6987
(0.1669)
0.5498
(0.3671)
0.9690
(0.6532)
0.1993
(0.2074)
1.6132**
(0.3241)
1.3167
(0.3181)
0.3241**
(0.1527)
0.8658
(0.1191)
2.1941*
(0.9203)
1.1257
(0.0959)
1.2592**
(0.1195)
1.0232
(0.2050)
1.0314
(0.0505)
3.4867
(4.6006)
1.3417
(0.3219)
1.8560*
(0.6092)
1.9058
(1.3761)
1.0177
(0.1327)
1.9886***
(0.5261)
0.3907
(0.2256)
0.9972
(0.0063)
0.8318
(0.7039)
1.0487*
(0.0266)
0.9425
(0.5940)
0.0000
(0.0002)
0.9763
(0.1046)
1.4776
(0.3703)
0.9984
(0.0030)
0.8028
(0.3001)
0.9824
(0.0113)
1.3620
(0.4134)
0.8962
(0.2957)
0.9049
(0.2412)
5.7248**
(4.9465)
0.9961
(0.0087)
5.1779**
(3.9405)
0.9656
(0.0323)
2.0424
(1.6546)
0.2453
(0.2885)
Untold income
Talk about investments
Constant
N
LR chi-square(51)
Prob > chi-square
Log-likelihood
1.2312
(0.8523)
1.1913
(0.2179)
0.0000***
(0.0000)
1.1597
(0.4076)
1.1857*
(0.1051)
0.0072***
(0.0090)
644
100.10
0.0000
-355.47
0.9080
(0.8829)
1.4647
(0.3737)
0.0000***
(0.0000)
Table A5 – Determinants of likelihood to own SRI and of percentage of SRI holding
(interaction between social preferences and portfolio size added)
In this table Model Probit presents marginal effects of a probit regression in which the
dependent variable takes on the value of 1 if an investor holds an SRI mutual fund in the month
investors participated in the experiment and survey; 0 otherwise. Model Tobit presents marginal
effects of tobit regressions in which the dependent variable is the percentage of the portfolio that
is held in SRI mutual fund in the month investors participated in the experiment and survey. The
Tobit regression accounts for left-censoring at 0% and right-censoring at 100%. In both models
‘intrinsic social preferences’ are measured as ‘Mean intrinsic social preferences’. All variables
are defined in Table 1. Standard errors are in parentheses. * is 10% ** is 5% and *** is 1%
significance.
Model
Dependent variable
PREFERENCES
Intrinsic social preferences
BELIEFS
Lower expected returns on SRI
Lower perceived risk on SRI
CONTROLS
Equity ratio
Mean portfolio returns
Volatility portfolio returns
Q1 Total portfolio value
Q2 Total portfolio value
Q3 Total portfolio value
Log number of transactions
Talk about investments
Investment knowledge
University degree
Risk preferences
Age
Female
Probit
Likelihood SRI
Tobit
Percentage SRI
0.0777*
(0.0402)
7.0848*
(4.0575)
-0.0303
(0.0310)
-0.0220
(0.0310)
-3.5181
(3.0317)
-0.7411
(3.0958)
0.0823
(0.0559)
-0.0110
(0.0260)
0.0022
(0.0066)
-0.0706
(0.0968)
0.0135
(0.1124)
-0.0237
(0.0962)
0.0288**
(0.0113)
0.0266**
(0.0117)
0.0062
(0.0141)
0.0461
(0.0328)
-0.0003
(0.0004)
-0.0017
(0.0015)
0.0023
11.8880**
(5.5831)
-2.1933
(2.5153)
0.2985
(0.6461)
-3.5003
(11.3762)
8.3255
(10.8568)
1.4514
(9.9594)
1.9203*
(1.1480)
2.1899*
(1.1559)
-0.1738
(1.3910)
5.6861*
(3.2455)
-0.0201
(0.0392)
-0.1538
(0.1527)
0.8744
Low income
High income
Untold income
Q1 Portfolio value x social preferences
Q2 Portfolio value x social preferences
Q3 Portfolio value x social preferences
(0.0460)
0.0459
(0.0424)
-0.0520
(0.0398)
0.0257
(0.0490)
-0.0539
(0.0686)
-0.0725
(0.0676)
-0.0256
(0.0611)
Constant
N
LR chi-square
Prob > chi-sqaure
Log-likelihood
645
43.97
0.0036
-295.1524
(4.4785)
7.4265*
(3.9331)
-4.5049
(4.5142)
4.9102
(4.5456)
-4.6581
(6.7021)
-9.0838
(6.7144)
-2.5372
(6.0831)
-42.6306***
(15.7510)
645
33.41
0.0563
-758.3913
Table A6 – Determinants of the likelihood to own SRI funds with and without tax benefits
(interaction between social preferences and portfolio size added)
This table presents relative-risk ratios of a multinomial logit regression in which the dependent
variable can take on four different values. The baseline group (not reported) consists of
conventional investors, the second group of investors who only hold SRI funds with tax
benefits, the third group only holds SRI funds without tax benefits, and the fourth group holds
both types of SRI funds. Social preferences is measured as Mean intrinsic social preferences.
All variables are defined in Table 1. Standard errors are in parentheses. * is 10% ** is 5% and
*** is 1% significance.
PREFERENCES
Intrinsic social preferences
BELIEFS
Lower expected returns on SRI
Lower perceived risk on SRI
CONTROLS
Equity ratio
Mean portfolio returns
Volatility portfolio returns
Risk preferences
Q1 Total portfolio value
Q2 Total portfolio value
Q3 Total portfolio value
Log number of transactions
Talk about investments
Investment knowledge
University degree
Age
Female
with tax
benefits
(1)
SRI funds
without tax
benefits
(2)
with and without
tax benefits
(3)
0.7520
(0.4046)
2.1451**
(0.7016)
1.3185
(1.0692)
2.2592
(1.3214)
1.7709
(0.9664)
0.7195
(0.1682)
0.7057
(0.1701)
0.6441
(0.4313)
0.8840
(0.6083)
0.1276*
(0.1425)
0.3135**
(0.1456)
0.9205
(0.1280)
0.9985
(0.0065)
3.9573
(6.0659)
3.0396
(4.5828)
0.5772
(0.8798)
1.7928**
(0.4103)
1.2609
(0.2500)
2.3194***
(0.6765)
0.3785*
(0.2209)
1.0551**
(0.0284)
0.9364
2.2705*
(0.9652)
1.0224
(0.2069)
1.0340
(0.0510)
0.9983
(0.0030)
0.6721
(0.6409)
1.2396
(1.1254)
1.3667
(1.0957)
1.1330
(0.0962)
1.1790*
(0.1055)
0.9658
(0.1041)
1.4928
(0.3782)
0.9816
(0.0115)
0.7700
3.8733
(5.2593)
2.2043
(1.7742)
1.0509
(0.1419)
0.9981
(0.0093)
0.0000
(0.0007)
0.4337
(1.0836)
0.1854
(0.4769)
1.5188
(0.3880)
1.5114
(0.3977)
0.8503
(0.2286)
6.3897**
(5.5428)
0.9742
(0.0329)
6.1103**
Low income
High income
Untold income
Q1 Portfolio value x social preferences
Q2 Portfolio value x social preferences
Q3 Portfolio value x social preferences
Constant
N
LR chi-square (66)
Prob > chi-square
Log-likelihood
(0.8270)
0.8100
(0.5499)
0.0000
(0.0003)
1.5931
(1.1349)
0.0000
(0.0006)
0.2057
(0.2520)
2.0528
(1.8763)
0.0000***
(0.0000)
(0.2905)
1.3840
(0.4230)
0.8370
(0.2803)
1.1371
(0.4037)
0.6925
(0.3780)
0.5529
(0.3017)
0.5638
(0.2732)
0.0856**
(0.1016)
645
116.95
0.0001
-347.2597
(4.8099)
1.8395
(1.4962)
0.2717
(0.3213)
1.1938
(1.1533)
0.8597
(1,509.5150)
1.7478
(2.4435)
3.0958
(4.6142)
0.0008**
(0.0027)
Table A7 – Determinants of the likelihood to own SRI funds with and without tax benefits
(base group is investors holding only SRI funds without tax benefits)
This table presents relative-risk ratios of a multinomial logit regression in which the dependent
variable can take on four different values. The base group (not reported) consists of investors
who hold only SRI funds without tax benefits. The first group of investors are conventional
investors. The second group consists of investors who only hold SRI funds with tax benefits and
the third group holds both types of SRI funds. Social preferences is measured as Mean intrinsic
social preferences. All variables are defined in Table 1. Standard errors are in parentheses. * is
10% ** is 5% and *** is 1% significance.
PREFERENCES
Social preferences
BELIEFS
Lower expected returns on SRI
Lower perceived risk on SRI
PORTFOLIO CHARACTERISTICS
Equity ratio
Log number of transactions
Log total portfolio value
Mean portfolio returns
Volatility portfolio returns
OTHER CONTROLS
Investment knowledge
University degree
Risk preferences
Female
Age
Low income
High income
conventional
SRI funds
with tax benefits
(1)
(2)
with and without
tax benefits
(3)
0.6467**
(0.1212)
0.4692*
(0.1865)
1.4795
(0.8830)
1.3911
(0.3233)
1.4401
(0.3444)
2.7659*
(1.6223)
2.3077
(1.2878)
0.7665
(0.5274)
1.4518
(1.0085)
0.4512*
(0.1895)
0.8865
(0.0756)
0.7907**
(0.0754)
0.9799
(0.1962)
0.9676
(0.0473)
0.0829**
(0.0919)
1.4224*
(0.3032)
1.0310
(0.2621)
0.3158**
(0.1574)
0.8338
(0.1199)
2.2390
(1.6410)
1.1947
(0.2951)
1.4828
(0.4952)
1.9003
(1.4104)
0.9893
(0.1343)
1.0222
(0.1092)
0.6729
(0.1685)
1.0014
(0.0030)
1.2465
(0.4662)
1.0174
(0.0117)
0.7341
(0.2231)
1.1157
(0.3680)
2.0721***
(0.5858)
0.2610**
(0.1603)
0.9987
(0.0067)
1.0298
(0.9350)
1.0673**
(0.0290)
0.6878
(0.4670)
0.000
(0.0002)
0.9269
(0.2557)
3.6779
(3.2274)
0.9978
(0.0090)
6.2990**
(5.0916)
0.9830
(0.0339)
1.4896
(1.2464)
0.2695
(0.3240)
Untold income
Talk about investments
Constant
N
LR chi-square (51)
Prob > chi-square
Log-likelihood
0.8583
(0.3020)
0.8417*
(0.0747)
165.7917***
(212.3499)
1.0492
(0.7889)
0.9918
(0.1963)
0.0009**
(0.0031)
644
100.78
0.0000
-355.13
0.7761
(0.7816)
1.2254
(0.3197)
0.0002*
(0.0008)
Figure A1 – Website of the mutual fund provider
Investors buy funds via the product selector on the website of the provider. The product selector
presents the investment category and information regarding the performance, fees, investment
policies etc.
Appendix B: Instructions Used in Online Experiments
Instructions used in the online experiments
(Translated from Dutch)
Below horizontal lines indicate screen transitions in the online experiment.
Risk preferences elicitation (this title was not part of the instructions)
You are participating in a choice experiment in which you will make financial decisions. In the
experiment there are no right or wrong decisions and you are free to decide in any way you like.
In the tables below you will find two options on each line. You can choose between:
-Option A: a fixed amount that you will receive ‘with certainty’
-Option B: an ‘all or nothing’ lottery, in which you have a 50% chance of winning €300 and a 50%
chance of winning nothing.
The tables that you will find below just serve as examples; you do not have to make any decisions
on this page yet.
In each row you should choose either A or B.
Each row can be the row relevant for your payment. Therefore, consider your decisions carefully for each
row.
Row 1:
Row 2:
Row 3:
Row 4:
Row 5:
Row 6:
Row 7:
Row 8:
Row 9:
Row 10:
Row 11:
Row 12:
Row 13:
Row 14:
Row 15:
Row 16:
Row 17:
Row 18:
Row 19:
Row 20:
Option A
€0 with certainty
€10 with certainty
€20 with certainty
€30 with certainty
€40 with certainty
€50 with certainty
€60 with certainty
€70 with certainty
€80 with certainty
€90 with certainty
€100 with certainty
€110 with certainty
€120 with certainty
€130 with certainty
€140 with certainty
€150 with certainty
€160 with certainty
€170 with certainty
€180 with certainty
€190 with certainty
Option B
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
Examples
1. Imagine that you choose in row 8 for Option B. In that case you will receive: €300 with a 50% chance
and €0 with a 50% chance.
2. Imagine that you choose in row 12 for Option A. In that case you will receive: €110 with certainty.
Experiment
In the following table you will find in each line two options. You should choose for each row between
Option A and Option B. In case that at the end of the experiment you are selected as for payout, one of the
20 rows will be randomly selected to determine your earnings. In case you selected Option B for this
specific row, it will be determined with a 50% probability whether you will receive €0 or €300.
Row 1:
Row 2:
Row 3:
Row 4:
Row 5:
Row 6:
Row 7:
Row 8:
Row 9:
Row 10:
Row 11:
Row 12:
Row 13:
Row 14:
Row 15:
Row 16:
Row 17:
Row 18:
Row 19:
Row 20:
Option A
€0 with certainty
€10 with certainty
€20 with certainty
€30 with certainty
€40 with certainty
€50 with certainty
€60 with certainty
€70 with certainty
€80 with certainty
€90 with certainty
€100 with certainty
€110 with certainty
€120 with certainty
€130 with certainty
€140 with certainty
€150 with certainty
€160 with certainty
€170 with certainty
€180 with certainty
€190 with certainty
Option B
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
50% chance to win €300 and 50% chance to win €0
1. What are the reasons for the choices that you made?
2. Did you consider the interest rate (for example, the interest on savings) while making your
decisions?
 Yes
 No
 Don’t know
3. Did you consider the amount of your personal financial wealth while making your decisions?
 Yes
 No
 Don’t know
4. I consider this choice experiment easy to understand
Totally disagree 1 2 3 4 5 6 7 Totally agree
5. I consider this experiment to be interesting
Totally disagree 1 2 3 4 5 6 7 Totally agree
Trust Game: Instructions for first-mover (this title was not part of the
instructions)
Explanation
You are anonymously and randomly linked to ANOTHER INDIVIDUAL INVESTOR AT <NAME OF
THE BANK>. You and this other investor will remain anonymous during and after the experiment. The
other person will receive instructions that are similar to the instructions you will receive, but this person
has a different role. Just like you, s/he will receive a starting capital of €50. You can send nothing, part or
all of this €50 to the person you are linked to. This person will subsequently receive three times the
amount that you send.
Subsequently, this person can decide to return nothing, part or the entire amount that s/he received from
you. S/He cannot return more than received, i.e. s/he will always retain the starting capital.
N.B. The amount that this person will return to you will not be increased by any amount.
Your income is the sum of:
1. The Euros that you did not send to the other person (€50 – the amount sent)
2. The money that was returned to you from the other participant.
In a formula: €50 – the amount that you send to the other participant + the amount that the other person
returns to you.
The income of the other participant is the sum of:
1. €50.
2. 3x the amount that you send.
3. Minus the amount that s/he returns.
In a formula: €50 + (3x the amount that you send) – amount that s/he returns.
In the table below, you can see what the outcomes are for each amount that you can send to the other
person.
Examples
For clarification purposes, in the following three random examples will be presented, indicating possible
situations that you may encounter during the experiment.
Example 1
You send €0 to the other player. The other player receives €0 and cannot send you anything back. You and
the other player thus each retain €50. Your income and the income of the other player are therefore equal
to €50.
Example 2
You send €50 to the other participant and will thus retain €0. The other participant to whom you have
been linked will receive 3 times the amount that you send to them. They will thus receive an additional
amount of €150 which implies that they hold €200 in total. Imagine that the other person decides to return
€100 in that case. Your income and that of the other person will then both be equal to €100.
Your income will be:
€50 – the amount that you send + the amount that the other participant returns:
50 - 50 + 100 = €100
The income of the other player will be equal to:
€50 + (3 x the amount that you send) – amount that they return:
50 + (3 x 50) - 100 = €100
Example 3
You send €30 to the other player and thus retain €20. The other participant to whom you have been linked
will receive 3 times the amount that you send. They will thus receive an additional amount of €90 and
hold €140 in total.
Imagine that the other player decides to return €10 in that case. Your income will then be €30 and the
other participant will earn €130.
Your income will be:
€50 – the amount that you send + the amount that the other participant returns:
50 - 30 + 10 = €30
The income of the other player will be equal to:
€50 + (3 x the amount that you send) – amount that they return:
50 + 3 x 30 - 10 = €130.
Practice Questions
You will now see several practice questions. These questions are based on random examples and are
meant to help develop a better understanding and feeling as regards to the consequences of your
decisions. In the experiment itself, all calculations will be made by the computer.
For our scientific research it is important that you answer these questions as accurately as possible.
Your income equals: €50 – the amount that you send + the amount that the other participant returns
The income of the other participant will equal: €50 + (3 x the amount that you send) – amount that they
return.
Situation 1
Imagine that you send 0 Euros to the other person.
1a. What will be your income?
... Euros
1b. What will be the income of the other person?
... Euros
Your income equals: €50 – the amount that you send + the amount that the other participant returns
The income of the other participant will equal: €50 + (3 x the amount that you send) – amount that they
return.
Situation 2
Imagine, you send 40 Euros to the other person and they decide to return €50.
2a. What will be your income?
... Euros
2b. What will be the income of the other player?
... Euros
Now the experiment will start.
Your decision
In the table below, you can see what the outcomes are for each amount that you can send to the other
person. You are linked to ANOTHER INDIVIDUAL INVESTOR AT <NAME OF THE BANK>.
How much money do you send to the other person?
Pulldown menu: €0 - €5 - €10 - €15 - €20 - €25 - €30 - €35 - €40 - €45 - €50
What are the reasons for the choices that you made?
Did you consider the amount of your personal financial wealth while making your decisions?
 Yes
 No
 Don’t know
We would like to ask you to make an estimation of the amount that the other person will return. You
can gain an additional €5 by doing this. We would like to ask you to indicate what, according to you, will
be the minimum and maximum amount that you sent and what the other person will return to you. In
other words, we ask you to indicate a range of values for the amount that the other person will return,
according to you.
We would like to ask you to indicate the amount that the other person will return, according to you. For
example, if you sent €30, then the other receives 3x30- €90. In that case, they can return between €0 and
€90.
Instead of giving a range, you could also choose an exact amount for your prediction. You should then fill
out the minimum and maximum amount as the same value. If you exactly estimate the amount that the
other returns, you will gain the maximum of 5 Euros.
In case your exact estimation is incorrect, or the true amount lies outside the range that you indicated, you
will earn nothing. In case the true amount lies within the range that you predicted, you will receive a
lower amount if you selected a larger range. The amount that can be gained will be reduced proportional
to the size of the range. In case you selected the maximum interval possible (i.e. if the minimum amount
equals €0 and the maximum amount equal €90), you will earn nothing.
It is important for you to know that it is in your best financial interest to make the most accurate
estimation of what the other member will return.
The other participant receives 3x the amount that you have sent (thus, 3x€<SENT>) and you will
therefore hold in total €<3 x SENT> + the starting capital of €50. Therefore, they may return
between 0 and <3 x SENT > Euros to you.
How much do you think that the other participant to whom you have been linked will return?
Minimum: ... Euros
Maximum: … Euros
We would now like to ask you to indicate an interval for the amount that the other people return to
you
You will not be paid for this, but we nonetheless ask you to make an accurate and honest estimation.
Imagine that the other player receives 3x the amount that you have sent (thus 3x€<SENT>) and you
therefore hold in total €<3 x SENT> + the starting capital of €50. Therefore, they may return between 0
and €<3 x SENT> Euros to you.
How much do to think that the average individual investor at <Name of the Bank> would return?
Minimum: ... Euros
Maximum: ... Euros
How much do to think that the average socially responsible investor at <Name of the Bank> would
return?
Minimum: ... Euros
Maximum: ... Euros
How much do you think that the average individual investor at <Name of the Bank>, who invests
more than €100.000, would return?
Minimum: ... Euros
Maximum: ... Euros
How much do to think that the average Dutchman would return?
Minimum: ... Euros
Maximum: ... Euros
How much do to think that an average Dutch university student would return?
Minimum: ... Euros
Maximum: ... Euros
Why do you expect that socially responsible investors would return more or less than average
investors?
Why do you expect that individual investors who invest more than €100.000 would return more or
less than average investors?
According to your estimation, what percentage of all the investors at <Name of the Bank> holds one
or more socially responsible investment funds in the portfolio?
I consider this choice experiment easy to understand
Totally disagree 1 2 3 4 5 6 7 Totally agree
I consider this experiment to be interesting
Totally disagree 1 2 3 4 5 6 7 Totally agree
In the following questions you will be asked for your impression of the average socially responsible
investor.
Compared to the average investor, a socially responsible investor will be:
More cooperative
Richer
More often a woman
Older
More trustworthy
More risk averse
1234567
1234567
1234567
1234567
1234567
1234567
Less cooperative
Poorer
More often a man
Younger
Less trustworthy
Less risk averse
Trust Game: Instructions for second-mover (this title was not part of the
instructions)
Explanation
You are anonymously and randomly linked to ANOTHER INDIVIDUAL INVESTOR AT <NAME OF
THE BANK>. You and the person that you are linked to will remain anonymous during and after the
experiment. The other person will receive instructions that are similar to the instructions you will receive,
but this person has a different role. Just like you, s/he will receive a starting capital of €50.
The person to whom you have been linked can send you nothing, part or all of his/her €50. You will
subsequently receive three times the amount that this person sends to you. You can subsequently decide to
return nothing, part or all of the money that you received. You cannot return more to the other person than
you received, i.e. you will always retain the starting capital.
N.B. The money that you return will not be multiplied.
Your income will be the sum of:
1. €50.
2. 3x the amount that the other person sent.
3. Minus the amount that you return
In a formula: €50 + (3x the amount that the other sends) – the amount that you return.
The income of the other participant is the sum of:
1. The Euros that the other did not sent (€50 – the amount that has been sent).
2. The amount that you return to the other person.
In a formula: €50 – the amount that the other sends + the amount that you return to the other.
Examples
For clarification purposes, here are three random examples of possible situations that you may encounter
during the experiment.
Example 1:
The other participant sends you €0. You receive €0 and you can therefore return nothing. You and the
other participant retain each €50. Your income and that of the other participant will be €50.
Example 2:
The other participant sends you €50 and s/he therefore retains €0. You receive 3 times the amount that
s/he sent. Therefore, you will receive an additional €150 and you will hold €200 in total.
Imagine that you decide to return €100 in that case.
Your income and that of the other will be both €100.
Your income will be:
€50 + (3x the amount that they send) – the amount that you return
50 + (3 x 50) - 100 = €100
The income of the other participant will be:
€50 – amount that they send + amount that you return
50 - 50 + 100 = €100
Example 3:
The other participant sends you €30 and s/he therefore retains €20. You receive 3 times the amount that
the other person sends. You will thus receive an additional €90 and you will hold €140 in total.
Imagine that you decide to return €10 in that case. Your income will then be €130 and the other
participant will earn €30.
Your income will be:
€50 + (3x the amount that they send) – the amount that you return
50 + (3 x 30) - 10 = €130
The income of the other participant will be:
€50 – amount that they send + amount that you return
50 - 30 + 10 = €30.
Practice Questions
You will now see several practice questions. These questions are based on random examples and they are
meant to help develop a better understanding and feeling as regards to the consequences of your
decisions. In the experiment itself, all calculations will be made by the computer.
For our scientific research it is important that you answer these questions as accurately as possible.
Your income will equal: €50 + (3x the amount sent) – the amount that you return.
The income of the other participant will equal: €50 – amount that they send + amount that you return.
Situation 1
Imagine that the other sends 0 Euros.
1a. What will be your income?
... Euros
1b. What will be the income of the other player?
... Euros
Your income will equal: €50 + (3x the amount sent) – the amount that you return.
The income of the other participant will equal: €50 – amount that they send + amount that you return.
Situation 2
Imagine, the other person sends you 40 Euros and you decide to return €50.
2a. What will be your income?
... Euros
2b. What will be the income of the other person?
... Euros
Now the experiment will start.
Your decision
For technical reasons you will have to make your decision without knowing how much money the person
to whom you have been linked has actually sent you. Therefore, for each possible amount that the other
person could send you, we would like to ask you to indicate, how much you would like to return.
However, only the decision that is relevant for the amount that has actually been sent is decisive for your
income and the income of the person to whom you have been linked. You have been linked to
ANOTHER INDIVIDUAL INVESTOR AT <NAME OF THE BANK>.
When you make a decision, it is important that you realize that each amount could be the actual amount.
We will now present you with 11 possible amounts that the other person could send you.
Your income will equal: €50 + (3x the amount sent) – the amount that you return.
The income of the other participant will equal: €50 – amount sent + amount that you return.
Could you please decide for each possibility how much you would like to return? Please indicate
this in the last column.
The other sends
0 euro
5 euro
10 euro
15 euro
20 euro
25 euro
30 euro
35 euro
40 euro
45 euro
50 euro
The other retains:
50 euro
45 euro
40 euro
35 euro
30 euro
25 euro
20 euro
15 euro
10 euro
5 euro
0 euro
I receive:
0 euro
15 euro
30 euro
45 euro
60 euro
75 euro
90 euro
105 euro
120 euro
135 euro
150 euro
I will hold in total:
50 euro
65 euro
80 euro
95 euro
110 euro
125 euro
140 euro
155 euro
170 euro
185 euro
200 euro
I return:
... euro
... euro
... euro
... euro
... euro
... euro
... euro
... euro
... euro
... euro
... euro
If you click on ‘Next’, your decision will be final; you will then not be able to return to this page
anymore.
What are the reasons for the choices that you made?
Did you consider the amount of your personal financial wealth while making your decisions?
 Yes
 No
 Don’t know
We would like to ask you to make an estimation of the amount that the other person has sent to you.
You can gain an additional €5 with this. We would like to ask you to indicate what will be the minimum
and maximum amount that the other sends, according to you. In other words, we ask you to indicate a
range of values for the amount that the other person will send according to you.
Instead of giving a range, you could also choose an exact amount for your prediction. You should then fill
out the minimum and maximum amount as the same value. If you exactly estimate the amount that has
been sent, you will gain the maximum of 5 Euros.
In case your exact estimation is incorrect, or the true amount lies outside the range that you indicated, you
will gain nothing. In case the true amount lies within the interval that you predicted, you will receive a
lower amount if you selected a larger range. The amount that can be gained will be reduced proportional
to the size of the range. In case you select the maximum interval possible (i.e. if the minimum amount
equals €0 and the maximum amount equal €50), you will gain nothing.
It is important for you to know that it is in your best financial interest to make the most accurate
estimation of what the other member will return.
How much do you think that the other participant to whom you have been linked will send to you?
Minimum: ... Euros
Maximum: … Euros
We would like to ask you to indicate what range of amounts that you expect that other people in
this role would return to you.
You will not be paid for this, but we nonetheless ask you to make an accurate and honest estimation.
Imagine, the other participant has a starting capital of €50 and s/he can thus send between 0 Euros and 50
Euros.
How much do you think that the average individual investor at <Name of the bank> would send?
Minimum: ... Euros
Maximum: ... Euros
How much do you think that the average socially responsible investor at <Name of the bank> would
send?
Minimum: ... Euros
Maximum: ... Euros
How much do you think that the average individual investor at <Name of the bank>, who invests
more than €100.000, would send?
Minimum: ... Euros
Maximum: ... Euros
How much do you think that the average Dutchman would send?
Minimum: ... Euros
Maximum: ... Euros
How much do you think that the average Dutch university student would send?
Minimum: ... Euros
Maximum: ... Euros
Why do you expect that socially responsible investors would send more or less than average
investors?
Why do you expect that individual investors who invest more than €100.000 would send more or
less than average investors?
According to your estimation, what percentage of all the investors at <Name of the bank> holds one
or more socially responsible investment funds in the portfolio?
I consider this choice experiment easy to understand
Totally disagree 1 2 3 4 5 6 7 Totally agree
I consider this experiment to be interesting
Totally disagree 1 2 3 4 5 6 7 Totally agree
In the following questions you will be asked for your impression of the average socially responsible
investor.
Compared to the average investor, I think a socially responsible investor will be:
More cooperative
Richer
More often a woman
Older
More trustworthy
More risk averse
1234567
1234567
1234567
1234567
1234567
1234567
Less cooperative
Poorer
More often a man
Younger
Less trustworthy
Less risk averse