|
THE IRRATIONAL BEHAVIOR OF
INDIVIDUAL INVESTORS
As Part One made abundantly clear, there are always times
when investors are irrational. Behavioral finance, however,
says that this behavior is continual rather than episodic.
Overconfidence
Researchers in cognitive psychology have documented that
people deviate in systematic ways from rationality in making
judgments amid uncertainty. One of the most pervasive of
these biases is the tendency to be overconfident about beliefs
and abilities and overoptimistic about assessments of the
future.
One class of experiments illustrating this syndrome
consists of asking a large group of participants about their
competence as automobile drivers in relation to the average
driver in the group or to everyone who drives a car. Driving
an automobile is clearly a risky activity where skill plays an
important role. Answers to this question easily reveal
whether people have a realistic conception of their own skill
in relationship to others. In the case of college students, 80 to
90 percent of respondents invariably say that they are more
skillful, safer drivers than others in the class. As in Lake
Wobegon, (almost) all the students consider themselves above
average.
In another experiment involving students, respondents
were asked about likely future outcomes for themselves and
their roommates. They typically had very rosy views about
their own futures, which they imagined to include successful
careers, happy marriages, and good health. When asked to
speculate about their roommates’ futures, however, their
responses were far more realistic. The roommates were
believed to be far more likely to become alcoholics, suffer
illnesses, get divorced, and experience a variety of other
unfavorable outcomes.
These kinds of experiments have been repeated many
times and in several different contexts. For example, in the
business management best-seller
In Search of Excellence
,
Peters and Waterman report that a random sample of male
adults were asked to rank themselves in terms of their ability
to get along with others. One hundred percent of the
respondents ranked themselves in the top half of the
population. Twenty-five percent believed that they were in
the top 1 percent of the population. Even in judging athletic
ability, an area where self-deception would seem more
difficult, at least 60 percent of the male respondents ranked
themselves in the top quartile. Even the klutziest deluded
themselves about their athletic ability. Only 6 percent of male
respondents believed that their athleticism was below
average.
Daniel Kahneman has argued that this tendency to
overconfidence is particularly strong among investors. More
than most other groups, investors tend to exaggerate their
own skill and deny the role of chance. They overestimate
their own knowledge, underestimate the risks involved, and
exaggerate their ability to control events.
Kahneman’s tests show how well investors’ probability
judgments are calibrated by asking experimental subjects for
confidence intervals. He asks a question such as the
following:
What is your best estimate of the value of the Dow
Jones one month from today? Next pick a high value,
such that you are 99% sure (but not absolutely sure)
that the Dow Jones a month from today will be lower
than that value. Now pick a low value, such that you are
99% sure (but no more) that the Dow Jones a month
from today will be higher than that value.
If the instructions are carried out properly, the probability
that the Dow will be higher (lower) than your high (low)
estimate should be only 1 percent. In other words, the
investor should be 98 percent confident that the Dow will be
within the given range. Similar experiments have been carried
out on estimates for interest rates, the rate of inflation,
individual stock prices, and the like.
In fact, few investors are able to set accurate confidence
intervals. Correct intervals would lead to actual outcomes
being outside the predicted range only 2 percent of the time.
Actual surprises do occur close to 20 percent of the time.
This is what psychologists mean by overconfidence. If an
investor tells you he is 99 percent sure, he would be better
off assuming that he was only 80 percent sure. Such precision
implies that people tend to put larger stakes on their
predictions than are justified. And men typically display far
more overconfidence than women, especially about their
prowess in money matters.
What should we conclude from these studies? It is clear
that people set far too precise confidence intervals for their
predictions. They exaggerate their skills and tend to have a far
too optimistic view of the future. These biases manifest
themselves in various ways in the stock market.
First and foremost, many individual investors are
mistakenly convinced that they can beat the market. As a
result, they speculate more than they should and trade too
much. Two behavioral economists, Terrance Odean and Brad
Barber, examined the individual accounts at a large discount
broker over a substantial period of time. They found that the
more individual investors traded, the worse they did. And
male investors traded much more than women, with
correspondingly poorer results.
This illusion of financial skill may well stem from another
psychological finding, called hindsight bias. Such errors are
sustained by having a selective memory of success. You
remember your successful investments. And in hindsight, it is
easy to convince yourself that you “knew Google was going
to quintuple right after its initial public offering.” People are
prone to attribute any good outcome to their own abilities.
They tend to rationalize bad outcomes as resulting from
unusual external events. History does not move us as much as
a couple of anecdotes of success. Hindsight promotes
overconfidence and fosters the illusion that the world is far
more predictable than it really is. The people who sell
worthless financial advice may even believe that it is good
advice. Steve Forbes, the publisher of
Forbes
magazine,
knows better and quotes the advice he received at his
grandfather’s knee: “It’s far more profitable to sell advice
than to take it.”
Many behavioralists believe that overconfidence in the
ability to predict the future growth of companies leads to a
general tendency for so-called growth stocks to be
overvalued. If the exciting new computer technology, medical
device, or retail outlet catches the public fancy, investors will
usually extrapolate success and project high growth rates for
the companies involved and hold such beliefs with far more
confidence than is justified. The high-growth forecasts lead to
higher valuations for growth stocks. But the rosy forecasts
are often not realized. The earnings may fall, and so may the
price-earnings multiples of the shares, which will lead to poor
investment results. Overoptimism in forecasting the growth
for exciting companies could then be one explanation for the
tendency of “growth” stocks to underperform “value” stocks.
Do'stlaringiz bilan baham: | 
