Dissecting the Equity Premium
Puzzle
Gabriel Chen
Although stocks have
outperformed bonds over the last century by a surprisingly large margin,
investors appear unwilling to hold them than “reasonable” levels of risk
aversion would suggest. For the 129-year period of 1872 through 2000, average
real return on the S&P 500 was 8.8 percent per year and average real return
(yield) on commercial paper (rolled over at six months) was 3.2 percent. Holding
stocks versus bonds over that period would have resulted in an average equity
premium of 5.6 percent (Fama and French, 2001). Why is this equity premium so
large or why should anyone be willing to hold bonds? In order to explain the
enormous discrepancy between the returns on stocks and fixed income securities,
I borrow Benartzi and Thaler’s (1995) concept of myopic loss aversion, which is
a combination of two behavioral concepts – loss aversion and the evaluation
period of an investor’s portfolio. Their argument, however cogent, should not be
vaunted as the only explanation for
the equity premium puzzle since alternatives like the ambiguity aversion model
could also provide a possible resolution to this phenomenon.
People tend to feel losses much more acutely
than they feel gains of comparable magnitude. Since investors are likely to be
loss averse, they perceive a loss as having to sell something for a lower
amount than they bought it for and hence have great difficulty coming to terms
with their losses (Shefrin, 2002, p.147). If the investor’s stock holdings fall
in value, he or she may regret the specific decision made to invest in stocks. Regret
is “the pain we feel when we realize that we would be better off if we had not
taken a certain action in the past,” and investors would want to minimize their
future regret by seeking the optimum trade-off between risk and return
(Barberis and Thaler, 2003, p.1081). The result of a wrong decision is
perceived as more negative than the damage that could be incurred through doing
nothing. To minimize an investor’s regret, it would be prudent for him or her
to hold the asset for as long as possible, making the risky asset seem more
attractive, provided that the investment is not evaluated frequently (Benartzi
and Thaler, 1995).
Thaler, Tversky,
Kahneman and Schwartz showed how the way information is presented affects the
frame people adopt in their decision-making. In their experiment, one group of
subjects – Group I - were asked to imagine themselves as portfolio managers and
to allocate their portfolio between Funds A and B. Returns on Fund A, and
likewise for Fund B, would be drawn from a normal distribution calibrated to
mimic bond and stock returns respectively. After each monthly observation, the subjects
were asked to allocate their portfolio between the two funds over the next
month. They were then shown the realized returns over that month, and asked to
allocate once again. A second group of investors – Group II – were shown the
same series of returns, except that it was aggregated at the annual level i.e.,
subjects only see cumulative annual returns. The experiment found that the
average final allocation chosen by subjects in Group I was much lower than that
chosen by people in Group II. The result is consistent with the idea that
“people code gains and losses based on how information is presented to them,”
as the subjects in Group I saw monthly observations and hence more frequent
losses. If they adopted the “monthly distribution as a frame,” they would be
“more wary of stocks” and would “allocate less to them” (Barberis and Thaler,
2003, p.1082).
Because investors
focus too much on the potential for short-term losses, they hold overly
conservative portfolios. This is effectively myopic loss aversion at work, as
it leads investors to hold too little in equities and too much in bonds. Stocks,
when looked at in months, “have made money 62 percent of the time since 1926,
with the average loss having been almost as large as the average gain,” but when
looked at in five-year segments, “have made money 90 percent of the time, and the
average loss was only 63 percent the size of the average gain” (Shefrin, 2002,
p.147). If investors’ historical reluctance to hold stocks stem from their
evaluation horizons being too short, how often would such an investor have to
evaluate his portfolio in order to be indifferent between the historical
distribution of returns on stocks and bonds? Benartzi and Thaler (1995) argued that
“there is no single evaluation period that applies to every investor.”
Nevertheless, “if one had to pick a single most plausible length for the evaluation
period, it would be a year” since individual investors file taxes annually and
receive their most comprehensive reports from their brokers, mutual funds, and
retirement accounts once a year.
However, we must distinguish
decision-making by the individual investor from decision-making by those managers
who oversee assets held by organizations like pension funds and endowments. Pension
funds essentially have an infinite time horizon. Yet, they commonly have an
allocation in about 60 percent stocks and 40 percent bonds and treasury bills.
Why do pension funds not invest a higher proportion in stocks? Benartzi and
Thaler (1995) argued that in this context, myopic loss aversion is produced by
agency costs that arise from conflicts between the fund manager and
stockholders. Pension fund managers do not have an infinite period of time to
work out near-term losses. While the fund would exist as long as the company
remained in business, the fund manager could not expect to be in the job
forever and thus would have to achieve his or her personal investment goals
within a short horizon.
The psychology of
decision-making is certainly more complex than it would seem, making it
difficult for behaviorists to reconcile and pinpoint with rationality the
trading strategies of investors. For example, individuals may make different
choices depending on how a given problem is presented to them, so that framing
influences their decisions. In choosing investments, investors could perhaps be
persuaded to allocate more of their wealth to stocks rather than bonds by simply
showing them an impressive history of long-term stock returns relative to those
on bonds, rather than just the volatile short-term stock returns. The action taken
by this individual investor would naturally differ from that of the average
investor painted in the equity premium puzzle.
Barberis,
Nicholas; Thaler, Richard. “A Survey of Behavioral Finance.”
Handbook of the Economics of Finance 2003.
Benartzi,
Shlomo; Thaler, Richard. “Myopic Loss Aversion and the Equity
Premium Puzzle.” The Quarterly
Journal of Economics 1995, 110.1:73-
92.
Fama, Eugene; French, Kenneth. “The Equity Risk Premium.”
Working Paper
522, Center for Research in Security Pries,
University of Chicago,
Graduate
School of Business (April 2001).
Shefrin,
Hersh. Beyond Greed and Fear. Oxford
University Press, 2002.