<<

. 8
( 47 .)



>>

future, and “the old attitude of buying solid stocks as a cornerstone for
one™s life savings and retirement has simply disappeared.”12 The perva-
sive pessimism about stocks in the late 1970s coincided with the best
buying opportunity of the century as shown in Figure 3.2.
As we see in Figure 3.2, from the end of 1981 until today, the Dow has
yielded more than 1,000%, and this calculation does not factor in divi-
dends. As the stock market rose throughout the 1980s and 1990s, the atti-
tudes toward stocks shifted from gloom to glee. Investors gradually
increased stock purchases. By the time the stock market peaked in 2000,
almost half of U.S. households were invested in mutual funds.13 At the
same time, Wall Street firms were advocating 70% investment in stocks,
among the highest figures on record.14 The peak enthusiasm for stocks
reached in 2000 corresponded with the beginning of one of the worst
bear markets since the Great Depression.
Sentiment is a predictor of future returns. Optimistic periods tend to
be followed by bad performance, whereas pessimism tends to dominate
Dow Jones Industrial Average




14,000
12,000
10,000
8,000
6,000
4,000
2,000
0
*04
2000
2001
2002
2003
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999




* through July


FIGURE 3.2 The Mother of All Bull Markets Started When Stocks Were Hated
Source: Dow Jones
Crazy World 47



before good things happen. Wall Street is driven by greed and fear. The
funny thing is that our lizard brains tend to make us greedy when we
ought be fearful, and fearful when we ought be greedy.
Our ability to be excited at the wrong time extends to individual
stocks. Professor Terrance Odean examined the actual trading records of
10,000 ordinary investors.15 He focused part of his study on investors
who sold one stock and then bought another stock within a few days. He
compared the performance of the stock that was sold with that of the
stock that was purchased.
How did the investors in this study perform? Remember that if markets
were rational, then the stocks that these investors sold would have had the
same return (on average) as the stocks that the investors bought. What hap-
pened? Professor Odean writes, “over a one-year horizon, the average return
to a purchased security is 3.3 percent lower than the return to a security
sold.” So these investors became excited enough to buy and pessimistic
enough to sell stocks at the wrong time. Their sentiment was an inverse pre-
dictor of success.
The efficient markets hypothesis is usually employed to suggest that
bargains are impossible. “Don™t waste your time looking for cheap
stocks; if they existed someone else would already have bought them.”
The converse is also true, but rarely mentioned; if markets were rational,
then it would be equally impossible to make systematically bad deci-
sions. “Don™t waste your time worrying that you might get excited at the
wrong time and buy expensive stocks; if they existed someone else
would already have sold them.”
Professor Odean found that markets were mean to these people. The
investors studied were completely out of sync with the market. They sold
stocks that went up and bought stocks that went down. None of this is
supposed to occur if the world ran according to the rules of the efficient
markets hypothesis. All stock prices are supposed to be correct, so it
should be impossible to make systematically bad choices.
In an episode of the futuristic cartoon, The Jetsons, there is a mobile,
robotic slot machine that rolls around enticing would-be gamblers by say-
ing, “I™m due.” The implication is that no one has hit the jackpot in a while
48 The New Science of Irrationality



so now is a good moment to invest a few bucks to try to win. The punch line
of the scene comes when some lucky gambler actually does win the jackpot.
The robot pays out the money and then scoots off proclaiming, “I™m due.”
In reality, slot machines are never “due.” They are designed to com-
pletely forget the past; in the language of probability, slot machines are
“memoryless.” The chance of winning immediately after a jackpot is the
same as on a slot machine that has gone years without a winner.
The efficient markets hypothesis states that stock markets should be
memoryless like an idealized slot machine. Nothing should predict the
next day™s stock price changes. So if it were true that optimism preceded
stock (and stock market) declines, then that would be evidence of market
irrationality and market meanness.
The Denial: The believers in the efficient markets hypothesis deny that
sentiment provides any information about future price changes. They
claim that the evidence of disdain for stocks in the late 1970s (e.g., the
“Death of Equities” cover story) is anecdotal and thus not scientific. Pro-
fessor Odean™s study demonstrating irrationality must be flawed.



Photographic Evidence: Scientific Evidence of
Sentiment Predicting Stock Price Changes

Professor Richard Thaler, doyen of the behavioral school, and Professor
Werner DeBondt performed a systematic analysis of sentiment.16 They
hypothesized that people would feel good about rising stocks and feel
bad about falling stocks. Therefore, they predicted that the future perfor-
mance of stocks that made investors feel bad (the losers) would be better
than that of stocks that made investors feel good (the winners).
Professors Thaler and DeBondt thus predicted that a way to make
money is precisely to buy the hated stocks that had been losers. They per-
formed a systematic study of hundreds of stocks over many years. In each
period they constructed a portfolio of previous winners and previous losers.
They then compared the performance of their winner and loser portfolios.
Crazy World 49



The rational and irrational views of markets make competing predic-
tions in this situation. If the efficient markets hypothesis were true, then
nothing would predict the future changes in price. If the market were
rational, then the winners™ portfolio would have the same performance as
the losers™ portfolio.
But, Professors Thaler and DeBondt found that individual stocks
exhibit the same pattern as the market as a whole: Pessimism precedes
rises, and optimism precedes falls. In their words, “Loser portfolios of 35
stocks outperform the market by, on average, 19.6%, thirty six months
after portfolio formation. Winner portfolios, on the other hand, earn
about 5.0% less than the market.” This study was published in the presti-
gious Journal of Finance.
The winner and loser study contradicts the efficient markets hypothe-
sis, which predicts that memoryless markets don™t care about previous
performance. Many other behavioral finance studies have produced evi-
dence that contradicts the efficient markets hypothesis. (Professor Thaler
collected and published 21 such studies in Advances in Behavioral
Finance.17 ) To which the true believers argue that the studies are con-
trived and people who trade at irrational prices would be weeded out.
It wasn™t me.



Claim #3: Some People Get Rich by Selling High
and Buying Low

Over the last 40 years, Warren Buffett has increased the value of Berk-
shire Hathaway at a compounded rate of 22.2% per year. Over the same
period, the S&P 500 stocks have increased by 10.4% a year; $1,000 in-
vested with Warren Buffett at the start of this period would have been
worth $2,594,850 at the end of 2003 versus $47,430 for an equivalent
investment in the S&P 500.18
So Warren Buffett seems to have a pretty good record of buying and
selling at favorable prices”prices that the efficient markets hypothesis
50 The New Science of Irrationality



suggests should never exist. Furthermore, Warren Buffett seems to have
done a lot better than a dart-throwing monkey.
Warren Buffett actively seeks undervalued investments. In his 2003
letter to shareholders he writes:

When valuations are similar, we strongly prefer owning businesses to
owning stocks. During most of our years of operations, however,
stocks were much the cheaper choice. We therefore sharply tilted our
asset allocation in those years towards equities . . . In recent years,
however, we™ve found it hard to find significantly undervalued stocks.

So Warren Buffett believes that undervalued stocks sometimes exist.
He also says that there are not many good values in the stock market
these days. If the efficient markets hypothesis were true, there would not
be any better or worse time to buy stocks; all prices would be fair at all
times. So Warren Buffett has made his fortune by acting precisely in a
fashion that would be silly if markets were rational.
The Denial: The believers in the efficient markets hypothesis deny that
Warren Buffett™s success is due to skill. Their argument is as follows: Put
1,024 people in a room. Have each of them flip a coin 10 times. On aver-
age, one of them will have produced 10 heads in a row. Now call that per-
son Warren Buffett.
In other words, there are lots of money managers and by sheer dumb
luck someone will have a great track record. That great track record, in
this view, predicts nothing about the future.



Photographic Evidence: Predicting Coin Flips

If Warren Buffett™s success in the past was luck, then the efficient mar-
kets hypothesis suggests that for next year he is not expected to outper-
form the market or a dart-throwing monkey. In this thought experiment
the “winner” who produces 10 heads in a row has exactly a 50% chance
Crazy World 51



of producing an eleventh head on the next coin flip, the same odds as a
loser who produced 10 tails in a row.
Professor Thaler has created a money management firm (run with
Russell Fuller) that seeks to systematically exploit irrational market
opportunities. The firm makes investment decisions guided by the find-
ings of behavioral finance. Although the firm is young, the results are
interesting; as of the end of 2003, the firm™s six funds are outperforming
their benchmarks by an average of 8.1% a year.19 To which the true
believers argue that the performance of these funds, like the performance
of Warren Buffett, is luck and not skill.
No one can really know if a particular performance is due to skill or
luck. However, the interesting point is that the efficient markets hypoth-
esis cannot be proven false by any investor™s performance. Regardless
of how many more good years Warren Buffett or Fuller-Thaler have,
defenders of the efficient markets hypothesis can argue that superior per-
formance is sheer luck.
It wasn™t me.



A Hypothesis Masquerading as a Theory

During the war of 1812, the Native American Chief Tecumseh captured the
fort of Detroit through an interesting ruse. In the conflict, federal and state
troops, under the command of Major-General Hull, vastly outnumbered
about 1,000 Native American warriors. Tecumseh had his fighters run out
of the woods and then secretly circle back to emerge again. General Hull
saw”and counted”the same warriors over and over and was fooled into
thinking he faced a vastly larger force. He surrendered without a shot.
In any endeavor it is important to have an accurate estimate of the oppo-
sition. General Hull gave up before the fight started because he treated a
small force like an army. Similarly, the idea that markets are efficient is at
best a hypothesis, which is a weak statement. Investors who surrender to
dreams of market efficiency give up before the investing battle has begun.
52 The New Science of Irrationality



Proven scientific views of the world are called theories. For example,
all of us know that gravity is true, yet it is categorized as a theory. In con-
trast to proven theories, new ideas that may or may not be true are labeled
“hypotheses.” Importantly, even those who defend market rationality label
their idea as only a hypothesis, acknowledging that it has not been
proven.
In fact, the belief in market efficiency might not even qualify for
hypothesis status. The great philosopher of science, Sir Karl Popper,
writes, “In so far as a scientific statement speaks about reality, it must be
falsifiable; and in so far as it is not falsifiable, it does not speak about
reality.”20
In order to reach the standard of hypothesis, an idea must be provable,
which means that there must potentially be evidence that could disprove
it. As we have seen, there is essentially no way to disprove the idea of

<<

. 8
( 47 .)



>>