You are on page 1of 22

BOYS WILL BE BOYS:

GENDER, OVERCONFIDENCE,
AND COMMON STOCK
INVESTMENT

F419 Team 16 9:30AM

Agenda
Introduction
Data and Methods
Results
Competing Explanations for Differences in Turnover
and Performance
Conclusion

INTRODUCTION

Introduction
Overconfidence and Trading on Financial Markets
Reconciling high trading volumes in equity markets with the needs of rational
investors is difficult
Overconfident investor believes the precision of his/her knowledge about the
value of a security is greater than it actually is
Odean 1998

Barber and Odean 2000

Lower expected utility by


trading too much
Unrealistic beliefs about returns
and the precision with which
theyre estimated
Expend too many resources on
investment information

Difficult Task

Individual investors
underperform benchmarks
Those who trade the most
realize the worst performance

Low
Predictability

Noisy
Feedback

Odean 1999

Stocks investors buy


subsequently underperform
those they sell
Suggests poor timing of trades
Disposition Effect

Overconfidence

Introduction
Gender and Overconfidence
Psychology suggests, in areas like finance, men are more
overconfident than women
Men are inclined to feel more competent than women do in financial
matters [Prince 1993]
When feedback is noisy, women tend to underestimate abilities
relative to men [Lenney 1977]
Self-serving attribution bias infer own abilities based on past
successes and failures. Tendency to take too much credit for
successes leads to overconfidence. Multiple studies agree this bias is
greater for men

Our testable hypotheses:


1.Men trade more than women
2.By trading more, men hurt their performance more,
relative to women

DATA AND METHODS

Data and Methods


A. Household Account and Demographic Data

Sample consists of households with common stock investment at a large


discount brokerage firm for which we are able to identify the gender of the
person who opened the households first account.

Data on marital status, children, age, and income are from InfoBase Inc.

Self-reported data are information supplied to the discount brokerage firm at


the time the account is opened by the person on opening the account.

. Household Account and Demographic Data

Data and Methods


B. Return Calculations

Calculated the gross & net return performance of each household to evaluate
investment performance of men and women

The net return performance is calculated after a reasonable accounting for the
market impact, commissions, and bid-ask spread of each trade.

Estimate the gross monthly return on each common stock investment using the
beginning-of-month position statements from household data and the Center for
Research in Security Prices (CRSP) monthly returns file. In so doing, they made two
simplifying assumptions: 1) Assume that all securities are bought or sold on
the last day of the month & 2)We ignore intra-month trading.

Estimate of the bid-ask spread component of transaction costs includes any


market impact that might result from a trade & includes an intraday return
on the day of the trade.

C. Turnover

Monthly portfolio turnover for each household: of the monthly sales turnover +
the monthly purchase turnover.

Data and Methods


D. The Effect of Trading on Return Performance

Calculate an own-benchmark abnormal return for individual investors.

In this abnormal return calculation, the benchmark for household h is the month t
return of the beginning-of-year portfolio held by household h.

Represents the return that the household would have earned if it had held its
beginning-of-year portfolio for the entire year. The own-benchmark abnormal return
is the return earned by household h less the return of household hs beginning-ofyear portfolio.

If household did not trade during the year: own-benchmark abnormal return would
be 0 for all months during the year.

Own-benchmark advantage: it does not adjust returns according to a particular risk


model. No model of risk is universally accepted. It is inappropriate to adjust
investors returns for stock characteristics that they do not associate with risk.

The own benchmark measure allows each household to self-select the


investment style and risk profile of its benchmark, emphasizing the effect
trading has on performance.

Data and Methods


E. Security Selection

Theory: men will underperform women because men trade more and
trading is costly. Alternative cause: inferior security selection.

Investors with similar initial portfolios & similar turnover will differ in
performance if one consistently makes poor security selections.
Measured security selection ability by comparing bought stock returns to
sold stock returns.
Are lower own-benchmark returns earned by men due to more active
trading or to poor security selection?
Findings: stocks men choose to purchase underperform those they sold by
20 basis points/month. The stocks women choose to purchase
underperform those they sold by 17 basis points/month. Difference is not
statistically significant.

Both men and women detract from their returns by trading; men
simply do so more often.

RESULTS

Results
Men vs. Women

Model predicts that men will trade more than women, empirical
evidence supports this claim

# of households
Mean portfolio value
Mean annual turnover
%
Diff in net monthly
return % from
beginning of year
Performance of stocks
purchased versus sold

Women

Men

8,005

29,659

$18,371

$21,975

52.8%

76.8%

-0.143%

-0.221%

-0.17%

-0.20%

Men trade about 45%


more annually!

Is lower return for men


due to excessive
trading or poor stock
selection?

Check whether difference


in returns for stocks sold
vs. bought between men
and women is significant
It isnt

Results
Single Men versus Single Women
A spouse may influence decisions for an account opened by the other.

Partition dataset into married men, married women, single men, and single
women

Married
Women

Married Men

Single
Women

Single Men

4,894

19,741

2,306

6,326

Mean portfolio value

$17,754

$22,293

$19,654

$20,161

Mean annual turnover %

52.92%

73.32%

50.64%

84.6%

Diff in net monthly


return % from beginning
of year

-0.078%

-0.154%

-0.121%

-0.242%

# of households

Single men underperform single women by significantly more than married


men underperform married women
Consistent with predictions of the overconfidence model

Results
Cross-sectional Analysis

Gender differences in turnover and performance may be correlated with


other variables that also predict turnover and performance

Differences
in turnover

Differences
in net
monthly
return

Married Women & Married Men

-1.46%

-0.058%

Single Women & Single Men

-2.19%

-0.085%

Per decade we age

-0.31%

+0.002%

Demographic
Characteristics
Age
Marital Status
Presence of children
in household
Income

Results
Portfolio Risk
Test: 1) Women hold less risky positions than men, 2) Men decreases portfolio
returns by trading more than women
Estimate market risk (beta) and risk associated with small firms in two-factor
monthly time-series regression
Women

Men

-0.044%

-0.083%

Beta, market risk

1.050

1.081

Risk associated with size


(SMB)

0.360

0.519

Intercept, estimate of net


risk-adjusted monthly
return ()

-0.162%

-0.253%

Intercept, estimate of
gross risk-adjusted
monthly return ()

Women earn 0.091%


more in monthly net
returns, or 1.092%
annually
Both tilt towards higher
beta and smaller firms,
however men do so
more

Results
Portfolio Risk
In addition to beta and size, differences in
idiosyncratic risk exposures between men and women
may exist
Test whether men and women differ in volatility of
portfolio in total and in volatility of the stocks they
Our findings are consistent
hold Regression Results

Age, Marital Status, and Income


correlated with volatility
Young, single, and those with more
income are more willing to hold volatile
positions

with the notion that men and


women have different
attitudes towards risk.

COMPETING EXPLAINATIONS FOR DIFFERENCES IN


TURNOVER AND PERFORMANCE

Competing explanations-differences in turnover & performance


Risk Aversion
Rational informed investors will trade more if they are less risk
averse, BUT they will also improve their performance by trading.
Both groups hurt their performance by trading. And men do so more
than women.
This outcome can be explained by differences in the overconfidence
of men and women and by differences in the risk aversion of
overconfident men and women.
It cannot be explained by differences in risk aversion alone.

Competing explanations-differences in turnover & performance


Gambling

Risk-seeking and entertainment.

Risk-seeking: one demonstrates a preference for outcomes with greater


variance but equal or lower expected return. Increasing variance without
increasing expected return is to under-diversify.

Excessive trading decreases expected returns without decreasing variance.


Thus risk-seeking behavior does not explain excessive trading.

It is unlikely that most individuals churn their accounts to appear busy or trade

for fun. Overconfidence offers a simpler explanation for the high trading activity

CONCLUSION

CONCLUSION

Tested the
hypothesis that
overconfident
investors will trade
too much using
gender as a
natural proxy

Our investor
overconfidence
model suggests
that men will have
increased turnover
and thus, worse
performance

Empirical evidence
supports our
claims, showing
the effects are
more pronounced
when comparing
single vs. married
men and women