Journal of Business Ethics (2010) 93:237–253
Springer 2009

DOI 10.1007/s10551-009-0218-6

It’s a Matter of Principle: The Role William R. Pasewark

of Personal Values in Investment Decisions Mark E. Riley

ABSTRACT. We investigate the role of personal values Proponents of alternative investment choice
in an investment decision in a controlled experimental models cite the shift of funds into ‘‘socially respon-
setting. Participants were asked to choose an investment sible’’ and ‘‘ethical’’ investments as evidence that
in a bond issued by a tobacco company or a bond issued individuals seek investments consistent with their
by a non-tobacco company that offered an equal or personal values (Beal and Goyen, 1998; Social
sometimes lower yield. We then surveyed the participants
Investment Forum, 2008). The Social Investment
regarding their feelings toward tobacco use to determine
whether these values influenced their investment deci-
Forum (2008) claims 11% of all assets ($25.1 trillion)
sion. Using factor analysis, we identified investment- and under professional management have an orientation
tobacco-related dimensions on which participants’ re- toward social responsibility.1 The degree of interest
sponses tended to load. Two of these factors, relating to in socially responsible investing (SRI) is likely to
the societal impact of investment decisions and the health increase over the next few years. Gevlin (2007)
effects of tobacco, were highly significant in determining found that 41% of sampled investors planned to add
whether participants selected a tobacco or non-tobacco socially responsible investments to their portfolio
related investment. More importantly, we found that over the next 3 years.
when the rate of return on a tobacco-related investment While previous studies identify characteristics and
exceeds the rate of return on an investment not involving attitudes of socially responsible investors, specific
tobacco by 1%, the intensity of participant concerns about factors that influence the investment decision are not
the societal effects of their investment decisions was
well understood (McLachlan and Gardner, 2004).
especially important in determining investment choices.
This finding indicates that traditional wealth-maximiza-
The purpose of this study is to determine the role of
tion approaches, which do not consider the personal personal values in an investment decision. We utilize
values of the investor, omit an important factor that affects a controlled experimental setting in which partici-
investment decisions. pants are asked to choose between investing in a
bond issued by a tobacco company and a bond issued
KEY WORDS: investment, personal values, socially by a non-tobacco company. We manipulate the
responsible investing yield of the non-tobacco company’s bond to levels
equal to or lower than the yield of the tobacco
company’s bond. We then investigate whether
personal values regarding tobacco use influence the
Traditional approaches to asset valuation often as- choice between two investment alternatives. We
sume investors act with a goal of maximizing wealth find that within a simulated investment decision,
(Shefrin, 2005; Tetlock and Mellers, 2002). These investors are sensitive to both financial and
approaches typically view personal values of the non-financial factors when making an investment
investor as irrelevant (Beal et al., 2005; Markowitz, decision. Most importantly, we find that in this
1959). Alternative views suggest that investors may be experimental setting, the investment decision was
motivated by a sense of well being (Auger et al., 2003; significantly affected by the interaction between
Cullis et al., 1992; Gao and Schmidt, 2005) or the differences in rates of investment return and the
desire to facilitate social change (Beal et al., 2005). personal values of the investor.
238 William R. Pasewark and Mark E. Riley

Background To date, studies have classified investors as socially

responsible based on a general inclination toward
The development of socially responsible investing (SRI) non-specific social issues. Such studies ignore three
aspects of social responsibility. First, these studies do
The beginnings of SRI are attributed to the Religious not consider the multitude of social issues that face
Society of Friends who, in 1758, asked their members investors. For example, socially responsible investors
to refrain from doing business with organizations that may consider some or all of the following issues
bought or sold humans (Schueth, 2003). Likewise, in important: the environment, alcohol, tobacco,
1770s, Methodists discouraged their members from genetic engineering, gaming, weapons, labor rela-
participating in businesses that harmed the health of tions, and animal testing. Webley et al. (2001) and
workers (Wesley, 1872). During the 1950s, trade Lewis and Mackenzie (2000b) used participation the
unions influenced pension fund managers to make Friends Provident fund to classify investors as ‘‘eth-
investments consistent with union positions, and in ical’’ or conventional. That fund utilizes an invest-
the 1960s, it became evident that more mutual fund ment strategy which considers a wide range of
managers were incorporating moral issues in their ethical issues such as sustainable lifestyles, quality of
investment decisions (Malkiel and Quandt, 1971). In life, the environment, energy efficiency, human
the 1900s, equity funds were founded to accommo- rights, employment practices and more. However,
date the religious requirements of the Islamic com- socially responsible investors might differ as to the
munity (Hussein and Omran, 2005) and a growing specific issues about which they feel strongly. For
environmental movement (Fowler and Hope, 2007). example, an individual investor might be opposed to
The term ‘‘socially responsible investing’’ and investing in companies that harm the environment,
‘‘ethical investing’’ became popular in the 1990s and but not feel strongly about investments related to
refers to the consideration of corporate respon- gaming.
sibility and societal concerns in investment decisions. Second, socially responsible investors might differ
During this time, investment managers developed in the intensity of their feelings regarding a single
the technique of ‘‘screening,’’ the practice of iden- social issue (Hummels and Timmer, 2004). For exam-
tifying companies that meet certain ethical criteria ple, an investor who moderately opposes the use of
(Michelson et al., 2004).2 alcohol might choose to not invest in companies that
Recent literature suggests the degree to which an produce alcoholic beverages. A second investor with a
investor is influenced by personal values can be stronger opposition to alcohol consumption might
measured on a continuum that ranges from a strictly deny capital to retailers that sell alcoholic beverages as
ethical orientation to a strictly financial orientation well as manufacturers who produce alcoholic bever-
(Hummels and Timmer, 2004). However, McLach- ages. Given the wide range of issues that may or may
lan and Gardner (2004) note that classifying investors not be important to individual investors’ personal
is difficult because existing literature is not consistent value systems, it becomes extremely difficult to con-
regarding how to identify a ‘‘socially responsible struct a portfolio that is consistent with all the values
investor.’’ Studies attempting to identify consistent embraced by an individual investor or to design a
demographic patterns associated with SRI have, in mutual fund that appeals to a broad spectrum of
fact, yielded inconsistent results. Rosen et al. (1991) investors with varied value systems.
found that socially oriented investors tend to be Third, investors considered to be socially respon-
white-collar workers who are younger and better sible vary in their degree of financial commitment.
educated, but lower-salaried, than other investors. On Mackenzie and Lewis (1999) find that few socially
the other hand, more recent studies (McLachlan and responsible investors are willing to commit their
Gardner, 2004; Williams, 2007) did not find investors entire portfolio to socially oriented investments. In
in socially responsible mutual funds to be demo- fact, investors classified as ‘‘ethical investors’’ in one
graphically different than investors in conventional study averaged only 28.35% of ‘‘ethical holdings’’ in
funds. their total portfolio (Webley et al., 2001).
 The Role of Personal Values in Investment Decisions 239

Motivations of socially responsible investing Public Employees’ Retirement System (CalPERS),

the largest pension fund in the US, claims to have
Past studies cite three motivations for socially respon- foregone $400 million by screening potential invest-
sible investments (Beal et al., 2005). These are a ments in countries that violate human rights. The
desire to achieve return, a desire to affect social California State Teachers Retirement System (CalS-
change, and a desire for personal satisfaction. TRS) claims that avoiding tobacco investments re-
duced potential investment gains by $1 billion.
CalSTRS recently announced that they will no longer
Return on investment avoid tobacco stocks (Palmeri, 2008).

A theoretical argument is sometimes made that

socially responsible investors must be willing to accept
Social change
a lower rate of return. The lower rate of return is
sometimes referred to as an ‘‘ethical penalty’’
A second motivation for investing in socially
(Michelson et al., 2004; Tippet, 2001) and is attributed
responsible corporations is the potential for influ-
to increased fund management costs as well as higher
encing social change. Investors, for example, might
costs incurred by socially responsible companies as
expect that withholding capital from corporations that
they attempt to monitor and maintain their socially
harm the environment could cause these corporations
responsible status. In fact, research has shown that
to institute more environmentally friendly business
most socially responsible investors are not willing to
practices.
reduce their holdings in socially responsible invest-
Although socially responsible investors may seek to
ments upon the discovery that they yield lower returns
change business practices, research indicates that
(Lewis and Mackenzie, 2000a; Webley et al., 2001).
shareholder-based social change is minimal (Beal et al.,
An alternative theory suggests socially responsible
2005). Klonoski (1986) notes that investors, while
investments yield higher returns because greater
technically owners, are often unable to influence
awareness of socially desirable behavior makes these
decision making within a corporation. Haigh and
firms more attractive (Cullis et al., 1992). Others
Hazelton (2004) find that because they represent an
believe that since these firms are subject to greater
insufficient share of the market, even managers of
scrutiny, they typically operate in a more efficient
socially responsible mutual funds are not effective at
manner (Schwartz, 2003). Surprisingly, Gevlin
social change. In fact, findings indicate that instances of
(2007) finds that more than half of investors expect
shareholder success in changing corporate policy have
socially responsible investments to be less risky and
been limited to highly visible companies that have
to have better returns than other investments.
currently newsworthy business operations (Graves
Most research comparing investment returns finds
et al., 2001; Rehbein et al., 2004). Interestingly, socially
very few differences between conventional and so-
responsible investors seem resolved to their limited
cially responsible investments.3 Statman (2000)
ability to make change. Lewis and Mackenzie (2000b)
compares the Domini Social Index4 and the S&P
find that most socially responsible investors prefer a
500 Index between 1990 and 1998 and finds similar
passive approach to investing, in which investors utilize
volatility and returns in both. Likewise, returns for
mutual funds that filter socially inappropriate invest-
socially responsible and conventional mutual funds
ments, rather than active investment, in which own-
tend to be statistically indistinguishable (Bauer et al.,
ership rights are utilized to lobby for change.
2005; Benson et al., 2006; Shank et al., 2005).
Interestingly, one study found that the primary
reason for indistinguishable returns is that the com-
position of investments in socially responsible funds Personal satisfaction
did not differ significantly from that of the general
market (Bello, 2005). Investigating personal satisfaction with regard to
On the other hand, there is significant anecdotal investing is difficult because individuals derive
evidence supporting an ethical penalty. The California satisfaction in many different ways. For example, an
240 William R. Pasewark and Mark E. Riley

investor might be interested in investing in the collective behavior regarding socially responsible
gaming industry because he or she enjoys gambling investment, they are subject to certain limitations.
and is familiar with the industry participants. Con- First, such studies do not provide an understanding
versely, another investor might find gambling ethi- of individuals’ investment decision-making pro-
cally improper and would not be comfortable cesses. Second, they are subjected to a variety of
investing in a mutual fund that includes gaming confounding influences. For example, these studies
stocks in its portfolio. are unable to control for economic conditions or
Evidence suggests that a broad range of non- changes in the portfolios of the mutual funds studied.
economic factors affects the investment decision Another method for investigating the role of
(Jeffrey, 2006; Nagy and Obenberger, 1994). Beal ethics in investment decisions is to survey investors
et al. (2005) find that fewer than half of investors to determine their attitudes and demographic char-
sampled consider wealth maximization to be the acteristics. Studies utilizing this method typically
most important factor in an investment decision. sample those who either request a prospectus of a
Similarly, Sparkes (1998) finds that 35% of investors socially responsible mutual fund (Beal et al., 2005) or
would invest ethically, even if returns were slightly are existing investors of a fund promoting social
lower than comparable conventional funds. Lewis responsibility (Beal et al., 2005; Lewis and Mac-
and Mackenzie (1999) asked socially responsible kenzie, 2000a, b; McLachlan and Gardner, 2004;
investors generalized questions regarding their will- Williams, 2007). These studies identify attitudes and
ingness to hold socially responsible investments characteristics that distinguish socially responsible
given ex post evidence of investment return. They investors from conventional investors. However,
found most (94.8%) would not shift funds away from they are limited in the ability to determine whether
socially responsible funds if the return were two socially responsible investors engage in different
percentage points lower and that only 35.8% would decision processes than investors in conventional
reduce socially responsible investments if the return funds. Again, these studies are conducted in envi-
were five percentage points lower. ronments with potentially confounding variables. In
While research supports the idea that socially some cases, investors may select socially responsible
responsible investors are willing to accept a lower mutual funds for reasons other than personal values.
return, there is no specific evidence regarding what For example, these funds may be selected because
kind of values influence a specific type of invest- they provide higher returns, incur lower manage-
ment. For example, certain values might entice a ment fees, or offer better portfolio diversification.
socially responsible environmentalist to invest in a Studies that survey actual investors in socially
‘‘green’’ fund; however, that same fund might have responsible and conventional mutual funds are also
little appeal to a socially responsible investor focused limited because investors are classified in a strictly
on values related to pacifism. dichotomous manner as either socially responsible
or conventional.5 Mutual funds are typically classi-
fied as socially responsible based on positions
Methodologies for investigating personal values regarding a wide variety of social issues, including
in investment decisions environmental considerations and vice avoidance.6
Investors seeking environmentally friendly invest-
Previous studies investigating the role of personal ments might have significantly different personal
values in investment decisions have used two values than those seeking investments that are free
approaches. First, studies have compared financial of ‘‘vice.’’ Further, investors’ definitions of ‘‘vice’’
aspects of socially responsible and conventional may vary widely.
mutual funds. Bollen (2007) measures the demand Sandberg (2007) stresses that understanding SRI
for socially responsible investments by observing the requires observation of consistency between personal
flow of cash into these funds. Others studies observe values and a potential investment. Previous studies
differences in diversification (Bello, 2005) and return provide initial insight regarding this relationship
(Bauer et al., 2005; Benson et al., 2006; Shank et al., when investor classes are defined broadly and inves-
2005). While these studies provide insight into tors are observed collectively. However, a more
 The Role of Personal Values in Investment Decisions 241

thorough understanding of this relationship might be effect on the investment decision, and eliminate the
gained by observing the degree of consistency noise associated with multiple issues.
between individuals’ personal values and their Second, we recognized that individuals might have
investment choices in a controlled environment. varying strengths of beliefs regarding multiple aspects
We attempt to gain a better understanding of SRI of a social concern (Hummels and Timmer, 2004).
by using methodologies not previously employed in For example, regarding tobacco use, we suspected that
studying SRI. Specifically, we utilize an experi- the intensity of an individual’s beliefs about the use of
mental design in which participants make an invest- tobacco in public places could differ substantially from
ment decision under controlled conditions. To the intensity of beliefs related to the health effects of
remove the effects of confounding variables, we limit tobacco. Narrowing the investment choice to a single
the investment opportunities, hold investment rat- social concern permitted the investigation of the
ings constant, and specify expected returns. We also multiple aspects of one social issue.
simplify participant perceptions regarding social Our second research question relates to the
responsibility by examining a single ethical issue, interaction between personal values and financial
tobacco use. opportunities:
We seek insight regarding two research questions.
The first question is similar to those investigated by RQ 2: How do personal values interact with finan-
previous studies, but is examined using different cial opportunities when individuals make investment
methodology: decisions?

RQ 1: Do personal values influence investment To investigate this question, we vary the returns in an
decisions? experimental investment choice. The yield on the
bond of the non-tobacco company is manipulated to
In previous studies, this question has been investigated be equal to or lower than the yield on the tobacco
by (1) observing differences in investment activities of company’s bond. Specifically, we are interested in
conventional and socially responsible funds or (2) how feelings regarding tobacco might combine with
determining whether the personal characteristics of higher rates of return on the tobacco-related bond to
those who invest in conventional funds differ from influence the investment decision process.
those who invest in socially responsible funds. In
this study, we attempt to directly observe causality
between personal values and investments. We ask Research design
participants to choose between an investment in a
tobacco company’s bond or a non-tobacco company’s In this experiment, participants were asked to make a
bond. We next ask the participants about their views hypothetical investment based on two actual invest-
regarding a single social issue (the use of tobacco). ment opportunities. One of the investments was a
Finally, we estimate the extent to which these views bond issued by a tobacco company.7 The other
affect their investment decisions. investment was a bond issued by a company that
We purposely constrained the social content produces specialty steels and alloys (the ‘‘non-tobacco
surrounding the investment decision to one issue, company’’). After choosing between the two invest-
tobacco use. Our intent was to overcome limitations ments, each participant responded to two follow-up
in previous studies in at least two ways. First, par- surveys regarding his or her investment preferences
ticipants included in previous studies invested in and opinions about tobacco-related issues, respec-
funds that attempt to address multiple social and tively.
ethical issues. It is unclear whether investors are
attracted to these funds because of the conglomera-
tion of social issues rather than a specific issue Task
addressed by the fund. Narrowing the investment
decision to one social issue permitted us to assess Participants were given the opportunity to make a
values related to that particular issue, measure the hypothetical $10,000 investment in the bonds of one
242 William R. Pasewark and Mark E. Riley

of two actual companies. The $10,000 investment bond would be used to expand steel and alloy pro-
would be part of a well-diversified $250,000 port- ducing facilities. Second, the investment choices
folio. One of the two bonds had been issued by a differed in yield (except in the first treatment which
tobacco company, while the other bond was issued served as a control).
by a non-tobacco company. Both bonds matured On the same day the participants made the
approximately several years from the date of the investment decision, they were asked to complete the
experiment and paid semi-annual interest over that first of two follow-up surveys, the ‘‘investment sur-
period of time. Both bonds also had similar levels vey.’’ The intent of this survey was to determine the
of risk indicated by identical ratings of BBB by role played by financial considerations in the partic-
Standard and Poor’s, a leading credit rating agency. ipants’ decision-making processes. The survey con-
Participants were provided with information regard- sisted of 14 items that individuals might consider
ing each bond issue, such as the interest rate, when making an actual investment decision, such as
maturity date, credit rating, and the use of the bond risk, cash flow, earnings, and use of proceeds (iden-
proceeds. They were also provided with a one-page tified in the Appendix). Prior to use, the survey was
summary of the issuing company by Value Line that piloted with individuals who had extensive experi-
included information such as the nature of the ence making actual investment decisions of this
company’s business, prospects for the near future nature.
(including risks), and 10–15 years worth of financial When completing the investment survey, partic-
data.8 ipants were asked to consider investments in general
We utilized a 1 9 3 between-subjects design, in rather than the specific investment they had just
which the difference in bond yield served as the made. The survey asked the participant to rate each
independent variable, and the investment decision item on a five-point scale from ‘‘Very Important’’ to
was the dependent variable. Each participant was ‘‘Not Very Important’’ with regard to an investment
randomly assigned to one of three experimental decision.
treatments. We manipulated the yields on the bond After making their investment choices, partici-
of the non-tobacco company so that these yields pants responded to a second follow-up survey, the
differed between treatments. In the first treatment, ‘‘tobacco survey.’’ The purpose of the tobacco sur-
the yields provided by the two bonds were identical. vey was to assess participants’ personal values
Since the bonds matured within 2 months of one regarding the use of tobacco. The survey consisted
another and they were both rated identically (BBB) of eight items relating to medical, legal, and societal
by Standard and Poor’s, the choice in this condition aspects of tobacco use (see Appendix). In addition,
was between essentially equivalent investments.9 In the survey asked about the participants’ personal
the second and third treatments, we reduced the yield tobacco use. We delayed the delivery of this survey
on the non-tobacco company bond to levels that to decrease the association with the investment
were 50 basis points (one half of a percent) lower and decision.
100 basis points (a full percentage point) lower,
respectively, than the yield on the bond of the
tobacco company.10 Our intent was to allow par- Participants
ticipants to consider, under various rate differential
scenarios, their personal values regarding investing in Our participants consisted of 235 undergraduate and
a tobacco manufacturing company. The instruction graduate business students at two large public uni-
page for the decision task is presented in the versities. The experiment and follow-up surveys
Appendix, along with the bond descriptions.11 were delivered in class, and participation was vol-
In general, the investment choices in this experi- untary, non-compensated, and anonymous.12 All
ment differed in only two ways. First, the investments participants had taken classes that would familiarize
differed regarding the use of the bond proceeds. The them with bond instruments as an investment
proceeds of the tobacco company bond would be choice. Responses provided by 19 of the participants
used for facilities related to the manufacturing of were unusable due to failure to answer one or more
tobacco products. The proceeds of the non-tobacco questions used in our analysis. Therefore, our sample
 The Role of Personal Values in Investment Decisions 243

comprised 216 participants who provided answers to In summary, the results presented in Table II
all of the questions. indicate that a sizable number of participants in each
Table I contains demographic data for our par- treatment found each of the two investment alterna-
ticipants. The participants in our sample are pre- tives attractive. Even when the non-tobacco bond
dominantly male (57.9%). A majority of participants was at a yield disadvantage, relative to the tobacco
were between 21 and 25 years in age (81.5%). Only bond, almost half the participants chose the non-to-
10.6% of our participants describe themselves as bacco bond. This suggests that some participants
current (consistent or casual) tobacco users,13 and might have been willing to sacrifice yield to invest in a
79.2% indicate they have never used tobacco.14 manner consistent with their values. The factor
analysis and logistic regression presented below
allowed us to better isolate the extent to which
Analysis and results financial concerns and personal values, as well as
interactions between such factors, influenced partic-
Descriptive statistics and initial data analysis ipant decisions in our study.

Each participant was presented with one of three

scenarios (treatments) in which he or she decided Factor analysis
between investing in a bond issued by a tobacco
company or a bond issued by a non-tobacco com- We perform an exploratory factor analysis to identify
pany. The yield to maturity of the non-tobacco common factors captured in responses to debriefing
bond differed in each treatment. In all three treat- questions the participants answered. We use princi-
ments, the tobacco company bond offered a yield to pal components factor analysis with varimax rotation
maturity of 6.731% and both bonds shared the same to identify constructs that potentially influence an
credit rating. In Treatment 1, the non-tobacco bond investment choice. Each construct is identified by
offered the same yield to maturity. An investor combining responses to two or more questions
evaluating the two potential investments based on answered by our participants, facilitating creation of
rates of return and credit ratings might view the two a reasonably parsimonious model of the investment
investments in Treatment 1 with indifference. decision process.
The non-tobacco company bond offered a lower We include in our factor analysis questions from
yield to maturity than the tobacco company’s bond in both the investment survey and tobacco survey pre-
Treatment 2 and in Treatment 3. The spread between sented in the Appendix. Our analysis identified five
the two bonds’ yields was 50 basis points (0.5%) in logical constructs based on combinations of responses
Treatment 2 and 100 basis points (1%) in Treatment to questions answered by our participants. The factors
3. Sandberg (2007) notes that understanding SRI are identified in Table III. We omit loadings with
entails observation of consistency between personal absolute values of 0.40 or below from Table III.15
values and investment. The latter two treatments Each resulting factor had a Cronbach alpha statistic of
were intended to test whether investors whose per- >0.60, indicating reasonable reliability among items
sonal values were inconsistent with tobacco use within each factor.
would be willing to sacrifice return to obtain con- Three of the five identified factors reflect financial
sistency between their investment choice and their aspects of investment analysis. The variables that load
personal values. Table II presents data on the invest- on the first factor, Corp Data, represent participants’
ment choices of participants in each treatment. responses to questions about the importance they at-
When the returns offered by the two bonds were tach to corporate performance and financial trends
equal (Treatment 1), 47 of 73 investors (64.4%) chose when making investment decisions. Participants with
the non-tobacco bond. The subtraction of 0.5% or high values for Corp Data consider financial statement
1% of return from the non-tobacco company’s bond figures and trends relatively important when making
results in movement away from that company’s investment decisions. Participants with high values
bond, as 71 of 143 (49.7%) participants in Treatments for Risk and Repay also place high importance on
2 and 3 chose the tobacco company’s bond. what might be thought of as traditional investment
244 William R. Pasewark and Mark E. Riley

TABLE I considerations. Participants with high values for the

Participant demographics factor labeled Risk prefer investments that they feel
exhibit low levels of risk, offer high degrees of safety,
n % and are suitable for conservative investors. Participants
who exhibit high values for the Repay factor are also
Panel A: Participants included in the study concerned about investment safety in that they seek
Surveys distributed 235 investments in bonds of companies they believe have
Surveys with missing data 19 the ability to meet principal and interest payments.
Included in analysis 216 The two other identified factors relate to personal
Panel B: Gender values regarding the use of the investment proceeds
Male 125 57.9 or the use of tobacco. These factors represent less
Female 91 42.1 traditional considerations with respect to potential
Total 216 100.0
investments. Participants with high scores for the
Panel C: Age Health factor are likely to believe tobacco is linked to
21–25 176 81.5 significant health problems and that tobacco prod-
26–30 24 11.1 ucts are both addictive and unsafe. Before he or she
31–35 6 2.7 decides to make an investment, a participant with a
36–40 4 1.9 high score on the Society factor considers whether
41–45 2 0.9 the proceeds from that investment will be used in a
46–50 2 0.9 manner that, in his or her opinion, benefits society.16
51–56 1 0.5
56–60 1 0.5
Logistic regression
Total 216 100.0
Panel D: Tobacco use The experimental decision entailed a dichotomous
Consistent user 7 3.2
choice, whether to invest a hypothetical $10,000 in
Casual user 16 7.4
the bonds of a tobacco company or a non-tobacco
Former user 22 10.2
Never used 171 79.2 company. Since this dichotomous choice variable is
the dependent variable in our study, we employ
Total 216 100.0 logistic regression to identify variables that explain
participants’ investment choices.

TABLE II
Investment selection by rate differential

Treatment Rates Rate differentiala Investment selection Total

Tobacco Non-tobacco

1 Tobacco – 6.731% 0 26 (35.6%) 47 (64.4%) 73 (100.0%)

Non-tobacco – 6.731%
2 Tobacco – 6.731% 0.5% 39 (54.9%) 32 (45.1%) 71 (100.0%)
Non-tobacco – 6.231%
3 Tobacco – 6.731% 1% 32 (44.4%) 40 (55.6%) 72 (100.0%)
Non-tobacco – 5.731%
Total 97 (44.9%) 119 (55.1%) 216 (100.0%)
a
The rate differential refers to the excess of the tobacco company bond’s yield over the yield offered by the non-tobacco
company’s bond.
 The Role of Personal Values in Investment Decisions 245

TABLE III
Factor analysis of investment decision predictors’ (varimax rotation)

Item Description Factor 1 Factor 2 Factor 3 Factor 4 Factor 5

(Corp Data) (Health) (Risk) (Society) (Repay)

I8 The investment has higher than average 0.770

revenue projections for the next several years
I 12 The investment has higher than average 0.724
earnings projections for the next several years
I 10 The investment has higher than average cash 0.724
flow projections for the next several years
I9 The investment has demonstrated high rates 0.671
of cash flow growth in the past 5–10 years
I 11 The investment has demonstrated high rates 0.605
of earnings growth in the past 5–10 years
I4 The investment has recently reported results 0.536
that were significantly better than expected.
I7 The investment has demonstrated increased 0.526
revenue growth in the past 5–10 years
T5 Tobacco products are not safe to use 0.755
T2 Tobacco use is responsible for a significant 0.743
portion of health problems in our society
T3 Significant medical problems are strongly 0.708
linked to tobacco use
T4 Tobacco use potentially leads to addictive 0.656
behavior
T1 Tobacco use should be eliminated in public 0.498
places
I5 The investment has lower risk compared to 0.820
the market in general
I1 The investment has a high degree of safety 0.746
I6 The investment is suitable for conservative 0.624
investors
I 14 The investment proceeds will be used in a 0.866
way that I find productive
I 13 The investment proceeds will be used in a 0.850
way that benefits society.
I3 The investment is likely to repay the principal 0.830
at maturity
I2 The investment has the ability to meet 0.801
interest payments
Eigen value 3.154 2.523 2.011 1.552 1.388
Cronbach alpha 0.779 0.693 0.644 0.787 0.609

This table presents factor patterns generated by a principal components analysis with varimax rotation for 19 of the
questions from the investment and tobacco questionnaires. Factor loadings with absolute values under 0.40 are omitted to
enhance readability (N = 216).

We utilized two models in our analysis. The decision. The initial model also included demo-
purpose of the initial logistic regression model was to graphic variables relating to gender and education.
identify which of the six factors the participants After identifying the most important factors from the
considered important in making their investment initial model, we employ those factors in a second
246 William R. Pasewark and Mark E. Riley

model that includes treatment conditions and inter- with an increase of slightly more than 1.3 times in
action effects. Our initial logistic regression model is the odds of a participant choosing the non-tobacco
represented by Eq. 1. bond over the tobacco bond.18 Strong beliefs about
the negative health effects associated with tobacco
PðInvest ChoiceÞ ¼ k0 þ k1 Corp Data use (Health) were also significantly (p = 0.039) re-
þ k2 Health þ k3 Risk þ k4 Society lated to the investment selection with higher scores
on this factor increasing the likelihood of choosing
þ k5 Repay þ k6 Gender þ k7 Age þ e; ð1Þ
the non-tobacco bond. Neither of the demographic
in which Invest Choice = 1 if the participant chose factors (Gender and Age) considered in Eq. 1 was
the non-tobacco company’s bond, 0 if the partici- significant in predicting the participants’ investment
pant chose the tobacco company’s bond, Corp choices.19
Data = factor score representing importance to In order to answer our second research question,
participant of corporate performance data in an which asks whether values interact with financial
investment decision, Risk = factor score represent- opportunities, we investigated interaction effects
ing the importance to participant of risk factors in an between the treatment conditions in our survey and
investment decision, Repay = factor score represent- the factors identified in our factor analysis. Initially,
ing the importance to participant of the ability of a we ran a model (results not presented) including
borrower to repay the interest and principal of debt interactions of each of the five factors (CorpData,
instrument, Health = factor score representing par- Health, Risk, Society, and Repay) with two treatment
ticipant’s agreement with statements indicating that variables (Treatment 2 and Treatment 3). In that model
tobacco has adverse effect on heath, Society = factor (not presented), the following interactions were
score representing importance to participant that significant: Society and Treatment 2 (p = 0.098);
the proceeds from his or her investment are used Society and Treatment 3 (p = 0.033); Repay and
in a manner that is productive or benefits society, Treatment 3 (p = 0.082). None of the interactions
Gender = 1 if the participant was male, 0 if the par- between the treatment variables and Corp Data,
ticipant was female, Age = 1 if the participant’s age Health, or Risk were significant. Next, a model
is over 25, 0 if the participant’s age is 25 or under. (results not presented) including interactions of both
The results of the initial logistic regression are treatment variables with Repay and Society was run.
presented in Table IV. Wealth-maximizing models of The treatment interactions with Society remained
investment choice suggest that investors make statistically significant, but neither of the treatment
investment decisions based on strategies for maxi- interactions with Repay was significant. Therefore,
mizing wealth (Shefrin, 2005; Tetlock and Mellers, our final model included Treatment 2, Treatment 3,
2002). None of the three factors related to what might Health (based on its significance in Table IV), Society,
be termed a ‘‘traditional’’ investment strategy (Corp and interactions of the treatment conditions with
Data, Risk, and Repay) is significant in predicting Society as predictors. Treatment 1, the treatment in
investment choice using the model in Eq. 1.17 which the interest rates associated with the bonds
The two variables relating to personal values were equal, was used as a control condition. The
regarding tobacco use, Society and Health, were sta- final model is represented below as Eq. 2 and its
tistically significant in explaining the investment results are presented in Table V.20
choice. Society is highly significant (p < 0.0001) in
PðInvest ChoiceÞ ¼ k0 þ k1 Treatment 2
determining whether a non-tobacco investment was
selected instead of a tobacco investment. Participants þ k2 Treatment 3 þ k3 Health þ k4 Society
with high values on this factor are especially con- þ k5 Treatment 2
Society þ k6 Treatment 3
cerned about the societal implications and produc-

Society þ e; ð2Þ
tive use of the proceeds from their investments. The
significance of the variable Society in the initial model in which Invest Choice = 1 if the participant chose
indicates that this aspect of personal values played an the non-tobacco company’s bond, 0 otherwise;
important role in many participants’ decisions. A Treatment 2 = 1 if the non-tobacco bond’s yield
one-unit increase in the value of Society is associated exceeded the tobacco bond’s yield by 0.5%,
 The Role of Personal Values in Investment Decisions 247

TABLE IV
Logistic regression of choice between tobacco and non-tobacco investments

P(Invest Choice) = k0 + k1 Corp Data + k2 Health + k3 Risk + k4 Society + k5 Repay + k6 Gender + k7 Age + e

Variable k Standard error Wald statistic Significance % Change in oddsa

Corp Data -0.237 0.151 2.473 0.116 -21.1

Health# 0.315 0.152 4.280 0.039 37.0
Risk -0.015 0.151 0.010 0.922 -1.5
Society
0.858 0.172 24.746 0.000 135.8
Repay 0.066 0.152 0.191 0.662 6.9
Gender 0.447 0.316 2.000 0.157 56.4
Age -0.214 0.382 0.315 0.575 -19.3
Constant 0.013 0.245 0.003 0.959 -98.7

n 216
v2 36.232
Cox and Snell (Pseudo) R2 0.154

Invest Choice = 1 if the participant chose the non-tobacco company’s bond, 0 if the participant chose the tobacco
company’s bond; Corp Data, factor score representing importance to participant of corporate performance data; Risk,
factor score representing the importance of risk factors in participant’s investment decision; Repay, factor score repre-
senting the importance to the participant of the ability of a borrower to repay the interest and principal of debt instrument;
Health, factor score representing participant’s agreement with statements indicating that tobacco has adverse effect on
heath; Society, factor score representing importance to participant that the proceeds from his or her investment are used in
a manner that is productive or benefits society; Gender = 1 if the participant was male, 0 if the participant was female;
Age = 1 if the participant’s age is over 25, 0 if the participant’s age is 25 or under.
a
The percentage change in odds is equal to (exp(B) - 1)
100. This value is the percentage change in odds that the
participant chooses the non-tobacco investment when the value of the predictor variable changes by one unit.
#,

Significant at the 0.05 and 0.01 levels, respectively.

0 otherwise; Treatment 3 = 1 if the non-tobacco significant interaction effect between Treatment 3,

bond’s yield exceeded the tobacco bond’s yield by the condition in which the tobacco bond’s yield
1.0%, 0 otherwise; Health = factor score represent- exceed the non-tobacco bond’s yield by a full
ing participant’s agreement with statements indicat- point, and Society, the factor indicating the extent
ing that tobacco has adverse effect on heath; to which a participant considers the effects of his or
Society = factor score representing importance to her investment on society. Thus, when the yield
participant that the proceeds from his or her difference between the bonds grows to a full point,
investment are used in a manner that is productive the importance of a participant’s concerns about the
or benefits society. societal implications of his or her investment grows.
The results presented in Table V indicate that In other words, as the opportunity cost of investing
Health, the factor reflecting the extent to which a in accordance with one’s values grows, the strength
participant believes smoking presents health haz- of those values takes on added importance. The
ards, is a significant (p = 0.058) predictor of most important implication of this finding is that,
investment choice in this experimental setting. when returns on investments with socially unde-
Higher loadings on the Health factor are associated sirable characteristics exceed returns on socially
with a higher probability of choosing the non-to- responsible investments, the strength of investors’
bacco bond. In a finding more directly bearing personal values becomes particularly important in
on our second research question, we detected a determining their investment choices.21
248 William R. Pasewark and Mark E. Riley

TABLE V
Logistic regression of choice between tobacco and non-tobacco investments

P(Invest Choice) = k0 + k1 Treatment 2 + k2 Treatment 3 + k3 Health + k4 Society + k5 Treatment 2
Society +

k6 Treatment 3
Society + e

Variable k Standard error Wald statistic Significance % Change in oddsa

Treatment 2^ -0.732 0.363 4.061 0.044 -51.9

Treatment 3 -0.482 0.373 1.669 0.196 -38.3
Health# 0.299 0.158 3.605 0.058 34.9
Society# 0.406 0.225 3.245 0.072 50.1
Treatment 2
Society 0.527 0.381 1.907 0.167 69.3
Treatment 3
Society^ 0.914 0.425 4.619 0.032 149.5
Constant^ 0.614 0.254 5.864 0.015 84.8

n 216
v2 40.923
Cox and Snell (Pseudo) R2 0.173

Invest Choice = 1 if the participant chose the non-tobacco company’s bond, 0 otherwise; Treatment 2 = 1 if the non-
tobacco bond’s yield exceeded the tobacco bond’s yield by 0.5%, 0 otherwise; Treatment 3 = 1 if the non-tobacco bond’s
yield exceeded the tobacco bond’s yield by 1.0%, 0 otherwise; Health, factor score representing participant’s agreement
with statements indicating that tobacco has adverse effect on heath; Society, factor score representing importance to
participant that the proceeds from his or her investment are used in a manner that is productive or benefits society.
a
The percentage change in odds is equal to (exp(B) - 1)
100. This value is the percentage change in odds that the
participant chooses the non-tobacco investment when the value of the predictor variable changes by one unit.
#,^
Significant at the 0.10 and 0.05 levels, respectively.

Implications, extensions, and limitations consistent with previous studies that utilized different
methodologies (see Beal et al., 2005 and Sparkes,
This study expands existing research by utilizing a 1998 who survey existing and potential investors of
unique experimental approach to determine the ef- socially responsible funds).
fects of values on an investment decision. Previous Second, we found that personal values interact
research classified an investor dichotomously as ethi- with expected rates of return to determine an
cally minded (or not) according to whether they investment choice. If a tobacco-related investment
participated in a socially responsible mutual fund (e.g., offered a rate of return that was 1% greater than a
Lewis and Mackenzie, 2000b; Webley et al., 2001). non-tobacco investment, the propensity to choose a
This research measures the personal values held by the non-tobacco investment was highly dependent on
investor in a multi-faceted way. In addition, previous the participant’s concern about the societal impli-
studies did not observe investor behavior within a cations of his or her investment.
particular investment decision. In this experiment, The results of this study have implications for both
participants were presented with investment choices investment decision-making and social responsibility
related to specific instruments and monetary amounts. literatures. First, much of the literature pertaining to
As such, we were able to directly relate specifically investment incorporates financial factors only. This
held values to an investment decision. study lends additional support to utilizing non-
We identify two important conclusions in this financial data in investment decision-making models.
study. First, consistent with studies that used a non- Second, literature concerning SRI has typically
experimental approach we found that investors classified investors dichotomously as socially
consider personal values in addition to financial responsible or not. We found that SRI can relate to a
factors in choosing investments. These results are single issue, such as tobacco use, and that classifying
 The Role of Personal Values in Investment Decisions 249

an investor based on a variety of values may not be decisions are likely more complex than the choice
realistic. In addition, we found that investors differed required in this experiment. Actual investment
in their opinions regarding tobacco use over a wide decision processes might involve other factors that
range. Accordingly, the dichotomous classification we held constant, such as greater variation in risk,
of socially responsible investors is potentially an capital gains, dividends (in the case of stock), and tax
oversimplification. Future studies would likely bene- consequences. As a result, the generalizability of our
fit by measuring social responsibility on a continu- results is limited in this regard.
ous, rather than a dichotomous, scale. Likewise, the experiment in the study addresses an
Awareness that individuals are more likely to seek investment decision that addresses a single social is-
investment opportunities consistent with their per- sue, tobacco use. The findings that the investment
sonal values has practical implications in two areas. decision regarding a tobacco-related bond is affected
First, conventional mutual funds tend to emphasize by beliefs regarding the social impact of smoking are,
financial aspects of the funds. Evidence that some of course, not generalizable to other investment
investors incorporate social values in the investment decisions that are influenced by the context of other
decision should potentially motivate mutual fund social influences.
managers who seek these investors to create greater Finally, future research may benefit by employing
awareness of the criteria for selecting fund invest- different investment vehicles. In particular, invest-
ments. Second, this study confirmed previous findings ment in common stock might affect the results since
that socially oriented investors are willing to accept that type of investment represents ownership and
lower returns. Corporate executives should consider could heighten the stakes with regard to social
the possibility that creating an awareness of socially responsibility.
desirable behavior can reduce the cost of capital.
Our experiment is subject to several limitations.
Consideration of these limitations could provide Notes
opportunities for future research. The participants in
1
our investments were arguably younger than the The industry reports an 18% growth rate from
‘‘typical’’ investor. While some studies have found 2005 to 2007 of 18%, increasing from $2.29 trillion in
youth to be one factor that distinguishes socially 2005 (Social Investment Forum, 2008).
2
Negative screening refers to eliminating invest-
responsible investors (Rosen et al., 1991), others
ments that do not have certain criteria, while positive
have not found demographic differences (McLachlan screening includes those that meet established criteria.
and Gardner, 2004; Williams, 2007). However, the 3
An exception would be Hussein and Omran
fact that we find a clear link between the importance (2005) who found that Islamic funds yield economically
participants place on the effects of their investments and statistically significant positive abnormal returns
on society and their investment choices does not rule from 1996 to 2003.
4
out the possibility that there might be important The Domini family of mutual funds invests in
generational differences in the manner in which securities based on certain social responsibility criteria.
5
personal values affect investment choices. Future Definitions of social responsibility lack both speci-
research might attempt to answer the question of ficity and consistency (McLachlan and Gardner, 2004).
whether younger investors attach more weight than Shank et al. (2005) note that ‘‘there are as many per-
older investors to personal values when making spectives of social responsibility investing as there are
options for investors.’’
investment decisions.22 In addition, the participants 6
See http://www.socialinvest.org/resources/mfpc/
in our experiment utilized hypothetical funds in the screening.cfm for summaries of screens for socially moti-
investment choice. Potentially, the use of personal vated funds.
wealth could affect the results. 7
We considered several issues in selecting a tobacco
We purposely simplified the investment decision corporation as an investment choice. First, to maintain
in this study to isolate key variables. The investment realism, we wanted actual companies with actual bond
decision in this experiment consisted of a choice issues. Second, we wanted the corporation to participate
between two relatively uniform bonds, except for in an industry that inspired a wide range of opinions.
the intended use of the proceeds. Actual investment While some consider tobacco to be a staple product,
250 William R. Pasewark and Mark E. Riley

others consider the product harmful. As opposed to than differences in the values between particular sub-
other social issues, we believed that tobacco use had a samples.
17
longer history of scientific study that enabled individuals This finding should not be interpreted to mean that
to form an educated opinion regarding the issue. Third, concerns about corporate performance or risk are not
we desired a corporation that purposefully engaged in the important considerations in a number of investment
controversial social activity as their primary business. In decisions. The insignificance of these factors in this
this scenario, the tobacco company sold bonds specifi- experiment was likely due to the similarity of the risk
cally to expand tobacco production. and performance levels of the companies issuing bonds
8
Value Line is an independent provider of informa- in this experiment.
18
tion and analyses regarding corporate investments to sub- This variable’s value ranges from -3.18 to 2.19 in
scribers. Value Line was not particularly sanguine about our sample. Due to the factor analysis techniques used
either company’s near-term prospects. Value Line’s com- to construct this variable, its mean is zero and its stan-
ments about one of the companies’ stocks noted that it dard deviation is 1.0.
19
expected that stock to ‘‘lag the market in the year ahead.’’ We were also interested whether tobacco users
Value Line’s commentary on the other company’s shares (consistently or casually) might be more inclined in-
stated that they ‘‘are not particularly appealing.’’ vest in a tobacco producing company. Inclusion of
9
We asked participants to offer open-ended com- this variable in the model in Table IV had some
ments regarding what influenced their choice of invest- influence on the levels of significance, primarily re-
ments. In no case was length of maturity mentioned. lated to the independent variable related to health.
10
In the first condition, both bonds yielded 6.731%. We conducted T tests of the factor means to deter-
In the second and third conditions, the yield on the mine whether tobacco users differed in any of the
non-tobacco company’s bond was lowered to 6.231 and values determined in Table III. The T test revealed
5.731%, respectively. that tobacco users differed with regard to one factor,
11
The descriptions of the bonds in the other two Factor 2 which relates to health. Smokers were less
treatments were identical to the description presented in likely to believe that tobacco use is linked to medical
Appendix, except for the non-tobacco bond’s effective problems, leads to addictive behavior, and that to-
yield, which was varied for each treatment. bacco is not safe to use. We determined that a signif-
12
Approval for use of human subjects was received by icant degree of collinearity existed between Factor 2
both co-authors from their respective universities. (Health) and a dummy variable related to whether or
13
This compares to a US average of 20.8% smokers of not the participant used tobacco and did not use the
adults aged 18 and over (Center for Disease Control, tobacco use variable in further analysis.
20
2008, p. 57) and to 19% among college students (John- The results of the interaction models that are not
ston et al., 2008). Lower rates are potentially explained by presented are available from the authors upon request.
21
the data being collected in parts of the US that does not It might be argued that more significant results
produce tobacco. Additionally, subjects may have under- might be found had we included a treatment with a
stated their smoking status due to a social desirability bias even greater yield differences (e.g., 1.5%). While that
(Nunnaly and Bernstein, 1994, pp. 382–384). might be the case, we felt that similar bond ratings
14
Chi-square tests indicate that demographic factors would not have been maintained under wider rate dif-
are not associated with investment choices made by ferences. In other words, in an actual financial market a
subjects in this study. company such as the non-tobacco company used in this
15
Kachigan (p. 252) lists 0.3, 0.4, and 0.5 as values study with a rate of 5.231% would likely be perceived
most often used as lower bounds in practice. Our lower as having less risk and carry a higher bond rating.
22
bound falls in the middle of this range. Similar results We attempted to determine whether those of
were found by using higher loading criteria. investing age (presumed to be 26 years and over) had
16
We considered whether investment preferences values (as measured by the resulting factors) that differed
expressed by subjects in Treatment 3 resulted from values from younger participants. T tests of factor means of the
that were distinctly different from subjects included in divided sample revealed no significant differences be-
Treatments 1 and 2. We performed a T test of the means tween factor means of older and younger participants.
for each factor for subjects in each treatment. These tests While a population of ‘‘seasoned investors’’ would have
revealed no difference between the factor means for the been preferable for this experiment, there is some assur-
subjects in each treatment. The results increase the likeli- ance that those of investing age had similar views
hood that investment choice was motivated by the treat- regarding investment and tobacco use as their younger
ment or the values held by the individual subject, rather counterparts.
 The Role of Personal Values in Investment Decisions 251

Appendix: Experimental instrument First follow-up survey

and follow-up questionnaire summary
The first follow-up survey was administered imme-
diately after the experiment. Subjects indicated
Experimental instrument summary whether they considered the item to be very
important (5), important (4), unsure or Neutral (3),
Instructions: You have been given $10,000 to invest. less important (2), or not important (1).
The investment will be part of a $250,000 portfolio
that you personally own. Currently, your portfolio is The investment
well diversified and is earning a return that meets
I1 Has a high degree of safety
your intended goals.
I2 Has the ability to meet interest payments
The conditions of the investment are: I3 Is likely to repay the principal at maturity
I4 Has recently reported results that were significantly better
than expected
• You must invest the entire $10,000. I5 Has lower risk compared to the market in general
• The investment must be in a fixed-rate bond I6 Is suitable for conservative investors
of one company described below. I7 Has demonstrated increased revenue growth in the past
• Please carefully consider the reasons for mak- 5–10 years
ing your investment. We will ask you about I8 Has higher than average revenue projections for the next
several years
the reasons for your choice after you make I9 Has demonstrated high rates of cash flow growth in the past
your selection. 5–10 years
I10 Has higher than average cash flow projections for the next
several years
Carefully review the attached analyst’s descrip- I11 Has demonstrated high rates of earnings growth in the past
tions for the corporations. When you have finished 5–10 years
indicate the corporate bond that you would pur- I12 Has higher than average earnings projections for the next
chase. several years
I13 Proceeds will be used in a way that benefits society
Investment choices:
I14 Proceeds will be used in a way that I find productive

Bond Issuer Reynolds American, Corp. Carpenter Technology, Corp.

Maturity date 6/1/2018 4/20/2018

Yield to maturity rate 6.731% 6.731%a
Bond Rating BBB BBB
Use of proceeds The net proceeds from the sale of these bonds The net proceeds received by the company
may include the repayment and refinancing from the sale of the debentures offered hereby
of outstanding debt, additions to working will be used for general corporate purposes.
capital, capital expenditures or the financing Note: Proceeds of past bond issues with this
of possible acquisitions or business expansion. description have been primarily used to ex-
Note: Proceeds of past bond issues with this pand steel and alloy producing facilities
description have been primarily used to ex-
pand tobacco producing facilities
Other Sold at 107.632; $204.90 accrued interest; Sold at 101.93; $345.60 accrued interest;
$387.50 semiannual interest payments; $349.50 semiannual interest payments;
10,000 at maturity 10,000 at maturity
a
In one of the three treatments, the rate on the Carpenter Technology Corp. bond was 6.731%. In the second and third
treatments, this rate was changed to 6.231 and 5.731%, respectively.
252 William R. Pasewark and Mark E. Riley

Second follow-up survey Cummings, L. A.: 2000, ‘The Financial Performance of

Ethical Investment Trusts: An Australian Perspective’,
The second follow-up survey was administered a Journal of Business Ethics 25(1), 79–92.
few weeks after the experiment. Subjects indicated Fowler, S. J. and C. Hope: 2007, ‘A Critical Review of
whether they strongly agree (5), agree (4), unsure or Sustainable Business Indices and Their Impact’, Journal
of Business Ethics 76(3), 243–253.
neutral (3), disagree (2), or strongly disagree (1).
Gevlin, K.: 2007, ‘The Coming of Age of Socially
Responsible Investing’, Financial Planning 37(8), 56.
Gao, L. and U. Schmidt: 2005, ‘Self Is Never Neutral:
T1 Tobacco use should be eliminated in public Why Economic Agents Behave Irrationally’, Journal of
places. Behavioral Finance 6(1), 27–37.
T2 Tobacco use is responsible for a significant Graves, S. B., K. Rehbein and S. Waddock: 2001, ‘Fad
portion of health problems in our society. and Fashion in Shareholder Activism: The Landscape
T3 Significant medical problems are strongly linked of Shareholder Resolutions, 1988–1998’, Business and
to tobacco use Society Review 106(4), 293–315.
T4 Tobacco use potentially leads to addictive Haigh, M. and J. Hazelton: 2004, ‘Financial Markets: A
behavior Tool for Social Responsibility?’, Journal of Business
T5 Tobacco products are not safe to use Ethics 52(1), 59–71.
Hummels, H. and D. Timmer: 2004, ‘Investors in Need
of Social, Ethical, and Environmental Information’,
Journal of Business Ethics 52(1), 73–84.
References Hussein, K. and M. Omran: 2005, ‘Ethical Investment
Revisited: Evidence from Dow Jones Islamic Indexes’,
Auger, P., P. Burke, T. Devmney and J. Louviere: 2003, Journal of Investing 14(3), 105–119.
‘What Will Consumers Pay for Social Product Fea- Irvine, W. B.: 1987, ‘The Ethics of Investing’, Journal of
tures?’, Journal of Business Ethics 42(3), 281–304. Business Ethics 6(3), 233–242.
Bauer, R., R. Otten and D. C. G. Koedijk: 2005, Jeffrey, H. J.: 2006, ‘Expanding the Range of Behavioral
‘International Evidence on Ethical Mutual Fund Per- Factors in Economic Simulations’, Journal of Behavioral
formance and Investment Style’, Journal of Banking & Finance 7(2), 97–106.
Finance 29(7), 1751–1767. Johnston, L. D., P. M. O’Malley, J. G. Bachman and J. E.
Beal, D. and M. Goyen: 1998, ‘Putting Your Money Schulenberg: 2008, ‘Monitoring the Future National
Where Your Mouth Is: A Profile of Ethical Investors’, Survey Results on Drug Use, 1975–2007: Volume II,
Financial Services Review 7(2), 129–144. College Students and Adults Ages 19–45’ (NIH Pub-
Beal, D. J., M. Goyen and P. Phillips: 2005, ‘Why Do lication No. 08-6418B). (National Institute on Drug
We Invest Ethically?’, Journal of Investing 14(3), 66–78. Abuse, Bethesda, MD).
Bello, Z. Y.: 2005, ‘Socially Responsible Investing and Klonoski, R. J.: 1986, ‘The Moral Responsibilities of
Portfolio Diversification’, The Journal of Financial Stockholders’, Journal of Business Ethics 5(5), 385–390.
Research 28(1), 41–57. Lewis, A. and C. Mackenzie: 2000, ‘Morals, Motives and
Benson, K. L., T. J. Brailsford and J. E. Humphrey: 2006, Money: The Case of U.K. Ethical Investing’, Human
‘Do Socially Responsible Fund Managers Really Relations 53(2), 179–191.
Invest Differently?’, Journal of Business Ethics 65(4), Lewis, A. and C. Mackenzie: 2000, ‘Support for Investor
337–357. Activism Among UK Ethical Investors’, Journal of
Bollen, N. P. B.: 2007, ‘Mutual Fund Attributes and Business Ethics 24(3), 215–222.
Investor Behavior’, Journal of Financial and Quantitative Mackenzie, C. and A. Lewis: 1999, ‘Morals and Markets:
Analysis 42(3), 683–708. The Case of Ethical Investing’, Business Ethics Quarterly
Center for Disease Control: 2008, ‘Early Release Estimates 9(3), 439–452.
Based on Data from the January–September 2008’, Malkiel, B. G. and R. E. Quandt: 1971, ‘Moral Issues in
National Health Interview Survey, (http://www.cdc. Investment Policy’, Harvard Business Review 49(2), 37–
gov/nchs/about/major/nhis/released200903.htm#8). 47.
Cullis, J. G., A. Lewis and A. Winnett: 1992, ‘Paying To Be Markowitz, H.: 1959, Portfolio Selection (Yale University
Good: UK Ethical Investments’, Kyklos 45(1), 3–23. Press, New Haven).
 The Role of Personal Values in Investment Decisions 253

McLachlan, J. and J. Gardner: 2004, ‘A Comparison of Shefrin, H.: 2005, A Behavioral Approach to Asset Pricing
Socially Responsible and Conventional Investors’, (Elsevier Academic Press, Burlington, MA).
Journal of Business Ethics 52(1), 11–25. Sparkes, R.: 1998, ‘Through a Glass Darkly: Some
Michelson, G., N. Wailes, S. van der Laan and G. Frost: Thoughts on the Ethics of Investment’, Epworth Re-
2004, ‘Ethical Investment Process and Outcomes’, view, July, 13–27.
Journal of Business Ethics 52(1), 1–10. Social Investment Forum: 2008, http://www.social
Nagy, R. A. and R. W. Obenberger: 1994, ‘Factors invest.org/news/releases/pressrelease.cfm?id=108
Influencing Individual Investor Behavior’, Financial Statman, M.: 2000, ‘Socially Responsible Mutual Funds’,
Analysts Journal 50(4), 63–68. Financial Analysts Journal 56(3), 30–39.
Nunnaly, J. C. and I. H. Bernstein: 1994, Psychometric Tetlock, P. E. and B. A. Mellers: 2002, ‘The Great
Theory (McGraw-Hill, New York). Rationality Debate’, Psychological Science 13, 94–99.
Olsen, K. R.: 2006, ‘A Literature Review of Social Tippet, J.: 2001, ‘Performance of Australia’s Ethical
Mood’, The Journal of Behavioral Finance 7(4), 193–203. Funds’, Australian Economic Review 34(2), 170–178.
Palmeri, C.: 2008, ‘CalPERS: The Price of Good Webley, P., A. Lewis and C. Mackenzie: 2001, ‘Com-
Intentions’, Business Week August 11. mitment Among Ethical Investors: An Experimental
Rehbein, K., S. Waddock and S. B. Graves: 2004, Approach’, Journal of Economic Psychology 22(1), 27–42.
‘Understanding Shareholder Activism: Which Cor- Wesley, J.: 1872, ‘‘The Use of Money: Sermon 50’’ The
porations Are Targeted?’, Business and Society 43(3), Sermons of John Wesley (in public domain).
239–268. Williams, G.: 2007, ‘Some Determinants of the Socially
Rosen, B. N., D. M. Sandler and D. Shani: 1991, ‘Social Responsible Investment Decision: A Cross-Country
Issues and Socially Responsible Investment Behavior: Study’, The Journal of Behavioral Finance 8(1), 43–57.
A Preliminary Empirical Investigation’, Journal of
Consumer Affairs 25(2), 221–234. William R. Pasewark
Sandberg, J.: 2007, ‘Should I Invest with My Conscience?’, Texas Tech University,
Business Ethics: A European Review 16(1), 71–86. Lubbock, TX, U.S.A.
Schueth, S.: 2003, ‘Socially Responsible Investing in the E-mail: [email protected]
United States’, Journal of Business Ethics 43(3), 189–194.
Schwartz, M. S.: 2003, ‘The ‘Ethics’ of Ethical Investing’,
Mark E. Riley
Journal of Business Ethics 43(3), 195–203.
Northern Illinois University,
Shank, T., D. Manullang and R. Hill: 2005, ‘‘‘Doing
Well While Doing Good’’ Revisited: A Study of DeKalb, IL, U.S.A.
Socially Responsible Firms’ Short-Term Versus Long-
Term Performance’, Managerial Finance 31(8), 33–47.