Fear and Greed Drive the Stock Market

Terrance Odean
University of California, Berkeley

SKAGEN Funds
New Years’s Conference 2011
• What do we know about investor behavior?

• Why does it matter?

– Market prices
– Investor welfare
 Investor Biases
• Confusion about probability → Underdiversify.
• Overconfidence → Trade too much.
• Desire to reduce regret → Cling to losers.
• Limited attention → Buy attention grabbing
stocks.
• Confusion about probability → Chase
performance.
 Underdiversification
• Bill Gates
vs.
• Enron employees
– 62% of Enron employee 401(k) investments
were in company stock
 Investor Biases
• Confusion about probability → Underdiversify.
• Overconfidence → Trade too much.
• Desire to reduce regret → Cling to losers.
• Limited attention → Buy attention grabbing stocks.
• Confusion about probability → Chase performance.
 When All Traders Are Above Average
(i.e., overconfident) Odean, 1998, Journal of Finance

• Trade more.
• Earn less.
• Underdiversify.
• Increase market volatility.
Do Investors Trade Too Much?
Odean, 1999, American Economic Review
Excluding Non-Speculative Trades
Odean, 1999, American Economic Review
Trading is Hazardous to Your Wealth
Barber and Odean, 2000, Journal of Finance
Percent Annual Return Monthly Turnover Monthly Turnover and Annual Perform ance of
Individual Investors

25

20

15

10

0
1 (Low 2 3 4 5 (High
Turnover) Turnover)

Net Return Turnover

"Overall, men claim more ability than
do women, but this difference emerges
most strongly on … masculine task[s].”

Deaux and Ferris (1977)

 Boys will be Boys
Barber and Odean, 2001, Quarterly Journal of Economics,

100

80

60

40

20

0
All Women All Men Single Single Men
Women
“Own Benchmark” Annual Net Returns
Barber and Odean, 2001, Quarterly Journal of Economics

All All Men Single Single

Women Women Men
-1

-2

-3

online
Data = Expertise
 Learning to be Overconfident
(Gervais & Odean, 2001, Review of Financial Studies)

• Self-attribution bias. People take too much

credit for their own successes.
 Illusion of Control

• Who rolls the dice.

 Investor Biases
• Confusion about probability → Underdiversify.
• Overconfidence → Trade too much.
• Desire to reduce regret → Cling to losers.
• Limited attention → Buy attention grabbing stocks.
• Confusion about probability → Chase performance.
 Rate at which Gains are
Realized Relative to Losses
Panel A: Discount Households
2.0
PGR/PLR

1.0

Taxable TDAs

0.0
jan feb mar april may june july aug sep oct nov dec

Source: Barber and Odean, 2003, Journal of Public Economics

 Disposition Effect in Taiwan
Barber, Lee, Liu, Odean, 2007, European Financial Management
 Relative Rate at which Investors Repurchase
Stocks Previously Sold
8
Likelihood of repurchas relative to stocks sold

7
Originally Sold for Gain
for loss with zero subsequent return

Orignially Sold for Loss

5

0
2

-4

12

20

28

36

44

52

60
-4

-3

-2

-2

-1
<

Percent return since sale

19
Source: Barber, Odean, & Strahilevitz, 2010.
 Investor Biases
• Confusion about probability → Underdiversify.
• Overconfidence → Trade too much.
• Desire to reduce regret → Cling to losers.
• Limited attention → Buy attention grabbing
stocks.
• Confusion about probability → Chase performance.
All that Glitters: The Effect of Attention and
News on the Buying Behavior of Individual
and Institutional Investors
Barber and Odean, Review of Financial Studies, 2008

• Thousands of stocks.
• Bounded rationality and processing ability.
• Limit search to stocks that catch attention.
• Investors buy stocks that catch their attention.
 Simulation Volume Sort

30

20
Percent Order Imbalance

10

0
1 2 3 4 5 6 7 8 9 10a 10b
-10

-20

-30

-40
Partitions Sorted on Same Period Trading Volume
 Abnormal Volume Sort Individuals
Imbalance by # trades

40

Large Discount Brokerage

Percent Order Imbalance .

Large Retail Brokerage

Small Discount Brokerage

0
1 2 3 4 5 6 7 8 9 10a 10b

-40

Partitions of Stocks Sorted on Current Day's Abnormal Trading Volume

Source: Barber & Odean 2008.

 Order Imbalance: Volume Sorts
TAQ & ISSM Data: 1983-2000
20.00

Individual Investors (Number of Trades)

10.00
Individual Investors (Value of Trades)
Imbalance
Buy / Sell

0.00
Lowest Highest
Abnormal Abnormal
Volume Volume
-10.00

-20.00

Abnormal Trading Volume

 Simulation Return Sort

20

15
Percent Order Imbalance

10

0
1a 1b 2 3 4 5 6 7 8 9 10a 10b
-5

-10

-15
Partitions Sorted on Previous Period's Return

25
 Return Sort Individuals
Imbalance by # trades

30

25
Percent Order Imbalance

Large Discount
Brokerage
20
Small Discount
15 Brokerage
Large Retail Brokerage
10

0
1a 1b 2 3 4 5 6 7 8 9 10a 10b
-5

Partitions of Stocks Sorted on Previous Day's Return

Source: Barber & Odean 2008.

 Order Imbalance: Return Sorts
TAQ & ISSM Data: 1983-2000
12.00

Individual (Number of Trades)

8.00
Individual (Value of Trades)

4.00

0.00

Lowest Previous Highest Previous

Day Return Day Return

-4.00

Previous Day's Return

27
 Investor Biases
• Confusion about probability → Underdiversify.
• Overconfidence → Trade too much.
• Desire to reduce regret → Cling to losers.
• Limited attention → Buy attention grabbing stocks.
• Confusion about probability → Chase
performance.
Mutual Funds: Money Pours into
Last Year’s Winners
45
39
Percentage of Buys

40
35
30
25
20 15
15 10
7 9
10 6
3 3 4
5 2
0
9

t)
st)

es
or

(B
(W

1
10

Performance Decile

Source: Barber, Odean, and Zheng

• What do we know about investor behavior?

• Why does it matter?

– Market prices
– Investor welfare
 Do Retail Trades Move Markets?
Barber, Odean, & Zhu, Review of Financial Studies 2009

• Purchases and sales of individual investors are highly correlated

• Higher percentage of individual investor buy orders this week

→ more buy orders next week
→ higher returns next week
→ lower returns next month

• Higher percentage of individual investor buy orders this year

→ lower returns next year
→ in the subsequent year high idiosyncratic risk stocks bought by
individuals underperform those sold by 13 percentage points.
Monthly Percentage Four-Factor Abnormal Returns for Value-Weighted Portfolios
formed on the basis of Weekly Proportion Buyer-Initiated Trades using
Small and Large Trades: February 1983 to December 2000
Monthly
Four-Factor Alpha
(%) t-statistic

Panel A: Contemporaneous Returns

Proportion
Buyer-Initiated Small Large Small Large
Quintile Trades Trades Trades Trades
1 (Sold) -2.398 -7.398 -9.79 -38.96
2 -1.205 -5.718 -6.57 -29.36
3 -0.422 -1.091 -3.37 -11.73
4 0.413 4.111 4.20 31.91
5 (Bought) 1.786 8.062 10.92 35.87

B-S (5-1) 4.184 15.460 11.99 39.37

Panel B: Subsequent Returns

Proportion
Buyer-Initiated Small Large Small Large
Quintile Trades Trades Trades Trades
1 (Sold) -0.637 0.421 -5.16 3.57
2 -0.160 0.797 -1.87 8.06
3 0.161 0.276 1.70 3.53
4 0.427 -0.219 4.81 -2.79
5 (Bought) 0.733 -0.362 5.22 -3.96

B-S (5-1) 1.370 -0.782 6.55 -5.54

Barber, Odean, Zhu, 2009, Review of Financial Studies

Monthly Percentage Abnormal Returns for High Idiosyncratic Risk Stocks for Value-
Weighted Portfolios formed on the basis of Annual Proportion Buyer-Initiated Trades
using Small Trades: 1984 to 2001

Barber, Odean, Zhu, 2009, Review of Financial Studies

 Just How Much Do Investors
Lose from Trade?
Mean Daily Profits
Institutions Individuals
Millions $NT Million $NT
Gross Trading 178.0 (178.0)
Profits
Gross Market- 46.4 (46.4)
timing Profits
Commissions (25.6) (216.9)

Transaction Tax (27.0) (228.4)

Net Profits 171.8 (669.7)

Source: Barber, Lee, Liu, and Odean, Review of Financial Studies, 2009
 Economic Significance
• Net Annual Individual Trading Losses are 2.2% of GDP.

• Individual Trading Losses Represent a 3.8% reduction

in annual portfolio performance.

• Institutional Gains Represent a 1.5% increase in annual

portfolio performance.

• 46.2 % of individuals’ gross daily losses are to foreign

investors.
– $3.5 million US per day

Source: Barber, Lee, Liu, and Odean, Review of Financial Studies, 2009
 U.S. vs. Taiwan
• Taiwanese investors trade more actively.

• U.S. investors face more asymmetric

information risk.
 Do institutional investors have
biases?
• Disposition effect
• Extrapolating recent past?
• Self-attribution bias
• Overconfidence
– models
– risk and leverage