48 ANNAMALAI INTERNATIONAL JOURNAL OF BUSINESS STUDIES & RESEARCH
LOSS AVERSION AND OVERCONFIDENCE :
DOES GENDER MATTERS?
Dr. Hanuman Prasad, Bharti Mohta
Vol : 4 Issue : 1
Abstract Nov’12 PP : 48-54
Behavioral finance is one of the emerging areas of research in the field of finance. Behavioral finance is
cross-disciplinary in nature incorporating finance with psychology. For investors, managing their
investments is like managing any enterprise. They apply all their skills, knowledge and expertise in
managing their money to get the best from investments. Despite this, some psychological factors affect
their decision. This study of systematic errors in cognitive reasoning and perception, and ultimately
what these errors reveal about the individual’s underlying thought processes, is often referred to as
investor heuristics and biases. This article defines behavioral finance and integrates two most important
biases of investors – Overconfidence and Loss Aversion with Gender. The objective of the study was
achieved by administering a well structured questionnaire and collecting empirical evidence on
investor’s own perceptions of bias. Analysis was done on a final set of 128 investors consisting of both
male and female investors. An Average Bias Score (ABS) was calculated and analyzed in SPSS. Results
indicate that male and female investors have different perception for different biases. Males were found
to be over-confident than females and females were more loss-averse than males. These behavioral biases
are present in each and every one of us in varying degrees and awareness about them is one of the best
ways of coping with them.
Keywords: Behavioral Finance, Biases, Gender, Over-Confidence, Loss Aversion, Investors.
Introduction Behavioral finance is based on the alternative
notion that investors, or at least a significant minority
Behavioral Finance is an emerging discipline that of them, are subject to behavioral biases that mean
represents a collection of alternative approaches to their financial decisions can be less than fully rational.
refine the classical finance definition of economic Evidence of these biases has typically come from
rationality. In particular, Behavioral Finance draws cognitive psychology literature and has then been
on the psychology and cognitive science literatures applied in a financial context.
to examine why individual decision-making often
deviates from rational choices in systematic ways. Some of these biases include
Previous research on these issues have focused on
investor heuristics, biases, and framing effects. • Overconfidence and over optimism –
Specifically, this paper will analyze: overconfidence investors overestimate their ability and the accuracy
and loss aversion, with special reference to gender. of the information they have.
Associate Professor, Faculty of Management Studies, • Representativeness – investors assess
Mohanlal Sukhadia University, Udaipur-313001, situations based on superficial characteristics rather
Rajasthan, e-mail :
[email protected] than underlying probabilities.
Research Scholar, Faculty of Management Studies • Conservatism – forecasters cling to prior
Mohanlal Sukhadia University, Udaipur-313001, beliefs in the face of new information.
Rajasthan, e-mail :
[email protected] LOSS AVERSION AND OVERCONFIDENCE : DOES GENDER MATTER? 49
• LossAversion–losses are emotionally felt twice First positively demonstrated by Tversky and
as strongly by people compared to comparable gains. Kahneman, the theory of risk aversion stipulates
that losses are emotionally felt twice as strongly
As defined by Hersh Shefrin, bias is nothing else by people compared to comparable gains. There
but the “predisposition towards error” (Shefrin, 2007). have been some studies which suggest that losses can
In other words, a bias is a prejudice or a propensity be as much as twice as psychologically powerful as
to make decisions while already being influenced by gains. The aversion to loss is so great with most
an underlying belief. There are many common biases investors that the thought of selling a stock when one
humans exhibit. is down (showing a paper loss) is truly abhorrent and
Overconfidence too painful to face. Many investors hang on to losing
stocks in the hope that one day, the stock will come
Overconfidence is often regarded as the most good. They do so even when there is no information
prevalent judgment bias. Overconfidence is another to suggest this recovery will occur. In other words,
bias that affects decision-making, both in the their distaste for losses renders them unable to make
corporate world and individual investments. a sound investment decision when losing money and
According to Shefrin, overconfidence “pertains to they start to gamble.But refusing to sell a stock when
how well people understand their own abilities and it is down (unless there is a good reason to hold) is
the limits of their knowledge” (Shefrin, 2007). In akin to a failure to acknowledge reality. Many
general, people tend to overestimate their ability to investors continue to hold loss-making stocks in the
perform well. This, in turn, leads to impulsive hope they will one day recover.
decisions as managers who think they know more
than they really do are overly confident in their own This theory of loss aversion is present both in
abilities. Therefore, they search for less help and business and in everyday life. Previous studies have
direction in making major decisions. Barber and shown that biases, heuristics, and framing effects have
Odean (2001) did work on gender based studies. a negative impact on decision making and result in
They have summarized psychological studies that find loss of productivity and value maximization; but do
a higher degree of overconfidence among men than students display the same amount of bias in everyday
among women. They find that men trade more than life when it comes to investment decisions. The paper
women which are consistent with overconfidence will also explore this question.
models. In itself, overconfidence can generally be Background of the study
viewed as a positive trait as it leads to survival both in
short and long run. The negativity of the bias presents A large body of psychological literature suggests
itself in those situations when individuals don’t that females tend to be more risk averse than males
recognize their limitations and therefore, make faulty (Byrnes, Miller, & Schafer, 1999,). This finding has
decisions based on erroneous premises. been relatively robust across a wide variety of
domains and using a wide variety of definitions of risk.
Loss Aversion (Regret Aversion) Wilson and Daly (1985) suggest that risk taking in
Loss aversion (Kahneman & Tversky 1979), males has evolved due to the adaptive mating
which is well established in choice scenarios, is advantage conferred on males willing to take risks to
commonly assumed to be due to the anticipation that accumulate mates and resources that may attract
losses will have a greater effect on feelings than mates. Females, in contrast, have evolved a lower
equivalent gains. Loss aversion or “prospect theory” preference for risk in response to different
is related to individual’s stronger desire to avoid losses evolutionary pressures faced because of their greater
than experience comparable gains (Tversky and biological investment in any offspring (Buss, 1999).
Kahneman, 1979).
50 ANNAMALAI INTERNATIONAL JOURNAL OF BUSINESS STUDIES & RESEARCH
Empirical investigation of gender differences in return investments by their investment brokers than
risk taking do point in the direction of less risk taking those offered to men (Wang, 1994).
by women than by men ( Eckel and Grossman, 2002
and Croson and Gneezy, 2004). Research in the Barber and Odean (2001) reported that men
domain of investing has also found consistent support 45 percentage more active in trading common stock
for this basic gender difference. For example, non- than women, attributed to men being more confident
professional women investors have been found to about investing than women. Women also appear less
allocate less of their portfolios to volatile assets (Barber inclined to invest in the currency market to avoid
& Odean, 2000; Bernasek & Shwiff, 2001; Chow potential losses from the fluctuation of exchange rate
& Riley, 1992; Cohn, Lewellen, Lease, & (Powell & Ansic, 1997). Finally, in an escalation of
Schlarbaum; 1975; Jianakoplos & Bernasek, 1998; commitment experiment by Bateman (1986), males
Sunden & Surette, 1998). This risk aversion on the were found more apt to stand by an initial investment
part of women has also been demonstrated in decision than were females.
professional financial analysts. Rationale of the Study
Studies have found that men tend to more risk From the above discussion, it is clear that
tolerant on average than women (Byrnes, Miller, and investors are prone to behavioral biases and their role
Schafer, 1999). This gender difference carries over in decision making cannot be undermined. Rather, we
into personal investment decisions. Men tend to attempt to study the most prevalent of these biases,
allocate more wealth to risky assets than do women i.e., Overconfidence and Loss Aversion (Regret
(Jianakoplos & Bernasek, 1998) and are 45% more Aversion) in the budding Indian investors. We are
active in trading common stock than women (Barber studying only Overconfidence and loss aversion
& Odean, 2001). Thus, men are more often perceived because these are the most common among the biases
to be competent, confident, and independent while to which people are prone to not only in business
women are more often perceived to be warm, decisions but also in day to day activities. For this,
nurturing, and concerned about the feeling of others we have taken business students who are investing in
(Zenmore, Fiske, & Kim, 2000). Findings such as the stock market. It is our assumption that the business
the tendency for women to shy away from competition students do not have much knowledge and expertise
(Niederle & Vesterlund, 2007) and for women with about the behavioral biases and hence are more prone
a greater distaste of competition being less likely to to them. Further, we have not only identified these
choose an occupation in law, business or management biases among the student investors, we have also
and more likely to choose one in education or health classified the biases on the basis of gender, i.e. intensity
(Kleinjans, 2009) are consistent with such gender- of Overconfidence and Loss Aversion is different in
stereotyped beliefs and gender roles. males and females student investors.
Gender-stereotyped beliefs exist within the So, the objectives of this study are:
business realm. For example, a study conducted by
Williams, Paluck, and Spencer-Rodgers (2010) l To give an overview of the two biases of our study
revealed that both university students and working
ü Overconfidence
adults provided higher salary estimates for men than
women to neutral job titles in both white and blue ü Loss aversion (Regret Aversion)
collar professions. Moreover, they also believed that
men should earn more than women. In financial l To find out the differences in male and female
investment decisions, women are typically perceived investors in terms of biases, .i.e. Overconfidence and
to be more risk averse than men. Not surprisingly, Loss aversion.
women tend to be offered lower risk/lower expected
LOSS AVERSION AND OVERCONFIDENCE : DOES GENDER MATTER? 51
Research Methodology The Questionnaire pertaining to Over
Confidence the responses yielded a Cronbach Alpha
For testing Over Confidence and Loss Aversion score of 0.752 and for that of Loss Aversion the
with respect to gender we designed a questionnaire Cronbach Alpha was 0.751.
where in the questions relating to both was mixed.
This was done in order to make sure that while filling Over Confidence and Loss Aversion are the two
the participants may not be able to judge that the main common behavioral biases among investors and
questions pertain to one particular topic. are present in varying degrees in both men and women.
This study is an attempt to identify whether there is a
The questionnaire was prepared using Google significant difference of these behavioral biases in Indian
documents and was sent to participants through e- men and women. To study the impact on gender in the
mail. Participants comprised of various categories of behavioral biases the following hypothesis is formulated:
people namely – students, working professionals,
professional investors and amateur investors. E-mails H01: There is no difference between the Over-
were sent out to more than 300 people, almost half Confidence exhibited by males and females.
of them were women since this is a gender based
H02: There is no difference between the Loss
study.
aversion exhibited by males and females.
We received responses from 135 people
approximately, out of which 61 were female and rest
Analysis and Interpretation:
of it were male. Three responses for women were The data received from the questionnaire was
deleted due to incomplete information, making a total further refined by separating the data from the two mixed
of 58 responses for consideration. Out of 74 questionnaires. Question numbers 1, 3,5,6,9 were on
responses received from males only 70 were Over Confidence and the remaining questions i.e. 2, 4,
considered and four of them were rejected due to 7,8,10 were used to measure Loss Aversion. Once
incomplete information. Thus the final set of responses we had data for the two different questionnaires, both
was 128, of which 58 were of women and 70 were sets were further refined to extract responses for males
of men. and females. This was done using Microsoft Excel.
Reliability of the Instrument: The Aggregate score was calculated in excel and
a Graphical analysis was done using Minitab software
From the questionnaire, question numbers
so as to generate the basic statistics of the Aggregate
1,3,5,6,9 were asked to measure Over Confidence
Score. Below are the 4 graphs revealing the
and the remaining questions i.e. 2,4,7,8,10 were used
descriptive statistics of the total aggregate score of
to measure Loss Aversion. The questionnaire was
the four categories – Aggregate data of Over-
designed on a 5-point Likert scale and an internal
consistency check was done by calculating Cronbach Aggregate Over-Confidence (Males)
Normal
Alpha for both set of questionnaires. Mean 15.5
14
StDev 2.996
N 70
12
Questionnaire Cronbach alpha 10
Frequency
Over Confidence 0.752 6
4
Loss Aversion 0.751
2
0
8 10 12 14 16 18 20 22
Aggregate Score
52 ANNAMALAI INTERNATIONAL JOURNAL OF BUSINESS STUDIES & RESEARCH
Aggregate Over-Confidence (Females)
Bias Score (ABS) which was the arithmetic average of
Normal
the aggregate score. This was done separately for males
Mean 9.776
20
StDev
N
2.136
58
and females. Finally, t-tests were performed for identifying
whether there is any significant difference between the
15
Average Bias Score (ABS) for male and female
Frequency
10
responses for both the questionnaires respectively.
5
Hypothesis t value Sig. (2-tailed) Result
0
6 9 12 15 18 H01 12.202 .000 Rejected
Aggregate Score
Aggregate Loss-Aversion(Males)
Over Confidence and Gender
Normal
30 Mean
StDev
11.03
2.924
There is a general notion that males tend to be
25
N 70
more over confident than their female counterparts.
20
To test this we subjected the Average Bias Score
Frequency
(ABS) for Over-Confidence to an independent t-test,
15
comparing the ABS for males and females
10
respectively.
5
0
Hypothesis t value Sig. (2-tailed) Result
4 8 12 16 20
Aggregate Score
H02 -8.361 .000 Rejected
Aggregate Loss-Aversion (Females)
Normal
16 Mean 15.62
The Null hypothesis is Rejected, which means
14
StDev
N
3.286
58 that there is difference between the over confidence
12 displayed by males and females. The ABS for Over
10 Confidence is higher for males suggesting that males
Frequency
8 are more over confident than females.
6
4 Loss Aversion and Gender
2
0
Loss Aversion is another commonly occurring
8 12 16
Aggregate Score
20
bias said to be found more in females than males. To
test this, we again applied an independent t-test on
the ABS for Loss Aversion on male and female data
Confidence in males and females as well as Aggregate respectively.
data of Loss-Aversion in males and females.
The Null hypothesis is Rejected, meaning that
The final analysis on the refined data was done the means of the two populations are significantly
in SPSS. different and females are more Loss averse than their
The aggregate score for the respective male counterparts.
questionnaires was then used to calculate the Average
LOSS AVERSION AND OVERCONFIDENCE : DOES GENDER MATTER? 53
Conclusion Jianakoplos, N., & Bernasek, A. (1998). Are women
more risk averse? Economic Inquiry, 36, pp
Behavioral biases such as overconfidence and 620-630.
loss aversion lead people to make mistakes and hence
lose out more than people who judge correctly. This Niederle, M., & Vesterlund, L. 2007, “Do Women
paper was an attempt to link the biases with gender Shy Away from Competition? Do Men Compete
and see whether there is any difference between the Too Much?” Quarterly Journal of Economics,
biases exhibited by males and females. Our study vol. 122, no. 3, pp. 1067–1101.
found that there is a clear difference between the levels
Kahneman, D. & Tversky, A. (1979). Prospect
of biases displayed between males and females. While
Theory: An Analysis of Decision under Risk.
males seem to be more over confident than females
Econometrica 47,pp 263-291.
they are less likely to show loss averse behavior than
females do. Powell, M., & Ansic, D. 1997, “Gender Differences
in Risk Behavior in Financial Decision-Making:
Being aware of these facts is an advantage to
An Experimental Analysis”, Journal of
both kinds of investors as these facts can act as
Economic Psychology, vol. 18, no. 6, pp. 605–
yardsticks while investing. Also, an important
628.
implication is that mixed team comprising of males
and females can nullify these effects to a great extent. Shefrin, Hersh. Behavioral Corporate Finance.
We have only studied the two most common occurring Decisions that Create Value. McGraw- Hill/Irwin.
biases. However, further studies can be done to New York, 2007.
explore the relation of other behavioral biases with
gender. Sunden, A., & Surette, B. (1998). Gender differences
in the allocation of assets in retirement plans.
References American Economic Review, 88, 207-221
Barber, B., & Odean, T. (2000). Boys will be boys: Williams, M.J., Paluck, E.L., & Spencer-Rodgers,
Gender, overconfidence and common stock J. 2010, “The masculinity of money: Automatic
investments. Quarterly Journal of Economics, stereotypes predict gender differences in
116, pp 261-292. estimated salaries”, Psychology of Women
Quarterly, vol. 34, no. 1, pp. 7–20.
Bernasek, A., & Shwiff, S. (2001). Gender, risk, and
retirement. Journal of Economic Issues, 35, Appendicies
pp345-356.
Questionnaire
Chow, K. V., & Riley, W. (1992). Asset allocation
and individual risk aversion. Finanical Analysts Notice of Confidentiality: Your participation in
Journal, 48, 32-37. this survey is greatly appreciated. We respect your
right to confidentiality. Therefore, your responses will
Cohn, R. A., Lewellen, W. G., Lease, R. C., & not be released without consent and used only for the
Schlarbaum, G. G. (1975). Individual investor risk purpose of my study.
aversion and investment portfolio composition.
Journal of Finance, 30,pp 605-620 Name: Gender: Male/Female
Age: Please tick as appropriate
Croson, R. and U. Gneezy (2004), “Gender
Difference in Preferences,” mimeo, The University 18-30 31-40 41-60 > 60
of Chicago.
54 ANNAMALAI INTERNATIONAL JOURNAL OF BUSINESS STUDIES & RESEARCH
Section I
Please use the 7 point rating scale to evaluate the following statements.
Strongly Agree Some Neither Some Strongly
Agree what Agree Disagree Disagree
what
Agree nor Disagree
Disagree
1. I am good at mathematical calculations.
2. There are 2 stocks in your portfolio – A & B. A is
appreciating in value and there are chances that it
will further appreciate. B is losing value and it
continues to do so. You are considering to sell A.
3. I think I will be able to get a new job in less than 2
months, paying close to my current salary.
4. You have to check a box at the end of the final
exam and potentially get 10 extra points. If, however,
the number of students who check the box is greater
than ½ of the class, the professor will deduct 10
points from the final score. You will check the box.
5. I think you have better athletic ability as compared
to my peer age group.
6. I prefer stock market investments over fixed income
yielding instruments (even if the rate of return is low).
7. Imagine that your performance was not up to the
mark in the last test. You are given the opportunity
to replace the failing grade from the last exam with
whatever you would get on the next exam. You have
very little time to study and a small chance that your
grade on the next exam would be higher; you are
still taking the risk hoping for a higher grade.
8. You have already won Rs.1,000, and now have a
choice of either receiving another Rs.500 without doing
anything or they could flip a fair coin and receive another
Rs.1,000 if heads come up or lose the original Rs.1,000
in case tails come up. You will take the chance.
9. I prefer own intelligence and analysis over professional
guidance while taking investment decisions?
10. You bought a new pair of shoes. You will throw
away the old pair of shoes you have (It is still in
working condition)?
Copyright of Annamalai International Journal of Business Studies & Research is the property
of Annamalai International Journal of Business Studies & Research and its content may not
be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.
