Against the Gods: The Remarkable Story of Risk (58 page)

There is nothing new about this pattern. In 1933, Alfred Cowles, a
wealthy investor and a brilliant amateur scholar, published a study covering a large number of printed financial services as well as every purchase and sale made over four years by twenty leading fire insurance
companies. Cowles concluded that the best of a series of random forecasts made by drawing cards from an appropriate deck was just as good
as the best of a series of actual forecasts, and that the results achieved by
the insurance companies "could have been achieved through a purely
random selection of stocks."" Today, with large, sophisticated, and
well-informed institutional investors dominating market activity, getting ahead of the market and staying there is far more difficult than it
was in the past.

If investors are unable to outguess one another with any degree of
reliability, perhaps the computer can capitalize on the market's nonrational behavior; machines are immune from such human flaws as the
endowment effect, myopia, and decision regret. So far, computer mod els that instruct the investor to buy when others are frightened and to
sell when others are overconfident have produced mixed or irregular
results. The investors become either more frightened or more overconfident than the computer model predicts, or else their behavior is
outside the patterns the computer can recognize. Yet computerized
trading is a fruitful area for further research, as we shall see shortly.

Human investors do turn in outstanding track records from time to
time. But even if we ascribe those achievements to skill rather than
luck, two problems remain.

First, past performance is a frail guide to the future. In retrospect,
the winners are fully visible, but we have no reliable way of identifying
in advance the investors whose skills will win out in the years ahead.
Timing also matters. Even the most successful investors, people like
Benjamin Graham and Warren Buffett, have had long periods of underperformance that would make any manager wince. Others zoom to
fame on one or two brilliant calls, only to fall flat when their public following grows large. No one knows when their next takeoff will come,
if ever.

The fine performance record of unmanaged index funds is vulnerable to the same kinds of criticism, because the guidance provided by
past performance is no more reliable here than it is for active managements. Indeed, more dramatically than any other portfolio, the indexes
reflect all the fads and nonrational behavior that is going on in the market. Yet a portfolio designed to track one of the major indexes, like the
S&P 500, still has clear advantages over actively managed portfolios.
Since turnover occurs only when a change is made in the index, transaction costs and capital-gains taxes can be held to a minimum.
Furthermore, the fees charged by managers of index funds run about
0.10% of assets; active managers charge many times that, often exceeding 1% of assets. These built-in advantages are due neither to luck nor
are they sensitive to some particular time period; they are working for
the investor all the time.

The second problem in relying on evidence of superior management skills is that winning strategies tend to have a brief half-life.
Capital markets as active and liquid as ours are so intensely competitive
that results from testing ideas on past data are difficult to replicate or
sustain in real time. Many smart people fail to get rich because people not so smart soon follow in their footsteps and smother the advantage
their strategy was designed to create.

Because of the danger that free-riders will hop aboard a successful strategy, it is quite possible that there are investors out there who
beat the market consistently beyond the probability of luck but who
stubbornly guard their obscurity. Nobel Laureate Paul Samuelson, an
eloquent defender of the hypothesis that markets act as though they
were rational, has admitted that possibility: "People differ in their
heights, pulchritude, and acidity, why not in their P.Q., or performance quotient?" But he goes on to point out that the few people
who have high P.Q.s are unlikely to rent their talents "to the Ford
Foundation or the local bank trust department. They have too high
an I.Q. for that."" You will not find them on Wall $treet Week, on
the cover of Time, or contributing papers to scholarly journals on
portfolio theory.

Instead, they are managing private partnerships that limit the number of investors they accept and that mandate seven-figure minimum
participations. Since they participate in the capital appreciation as well
as receiving a fee, adding other people's money to their own gives
them an opportunity to leverage their P.Q.s. It may well be that some
of them would qualify as Snap champs.

In Chapter 19 we shall look at what some of these investors are trying to do. Their strategies draw on theoretical and empirical concepts
that reach back to the origins of probability and to the Chevalier de
Mere himself But those strategies incorporate a more complex view of
market rationality than I have set forth. If there is validity to the notion
that risk equals opportunity, this little tribe is showing the way.

Nevertheless, private partnerships are peripheral to the mainstream
of the marketplace. Most investors either have too little money to participate, or, like the giant pension funds, they are so big that they cannot allocate a significant portion of their assets to the partnerships.
Moreover, the funds may be inhibited by the fear of decision regret in
the event that these unconventional investments go sour. In any case,
when the largest investors begin to experiment with exotic quantitative
concepts, they must be careful not to get in each other's way.

What are the consequences of all this for managing risks? Does the
presence of nonrational behavior make investing a riskier activity than
it would otherwise be? The answer to that question requires putting it
into its historical setting.

Capital markets have always been volatile, because they trade in
nothing more than bets on the future, which is full of surprises. Buying
shares of stock, which carry no maturity date, is a risky business. The
only way investors can liquidate their equity positions is by selling their
shares to one another: everyone is at the mercy of everyone else's
expectations and buying power. Similar considerations apply to bonds,
which return their principal value in cash to their owners but only at
some future date.

Such an environment provides a perfect setting for nonrational
behavior: uncertainty is scary. If the nonrational actors in the drama
overwhelm the rational actors in numbers and in wealth, asset prices are
likely to depart far from equilibrium levels and to remain there for
extended periods of time. Those periods are often long enough to
exhaust the patience of the most rational of investors. Under most circumstances, therefore, the market is more volatile than it would be if
everyone signed up for the rational model and left Kahneman and
Tversky to find other fields to plow.19

Nevertheless, explicit attention to investment risk and to the tradeoff between risk and return is a relatively young notion. Harry
Markowitz laid out the basic idea for the first time only in 1952, which
seems like a long time ago but is really a late-comer in the history of
markets. And with a great bull market getting under way in the early
1950s, Markowitz's focus on the risks of portfolio selection attracted little attention at the time. Academic interest speeded up during the 1960s,
but it was only after 1974 that practitioners sat up and took notice.

The explanation for this delayed reaction has to do with changes in
the volatility of the market. From 1926 to 1945-a period that included
the Great Crash, the Depression, and the Second World War-the
standard deviation of annual total returns (income plus change in capital values) was 37% a year while returns averaged only about 7% a year.
That was really risky business!

Investors brought that memory bank to the capital markets in the
late 1940s and on into the 1950s. Once burned, twice shy. A renewal
of speculative fever and unbridled optimism was slow to develop despite a mighty bull market that drove the Dow Jones Industrial
Average from less than 200 in 1945 to 1,000 by 1966. From 1946 to
1969, despite a handsome return of over 12% a year and a brief outburst
of speculative enthusiasm in 1961, the standard deviation of total
returns was only one-third of what it had been from 1926 to 1945.

This was the memory that bank investors carried into the 1970s.
Who would worry about risk in a market like that? Actually, everyone
should have worried. From the end of 1969 to the end of 1975, the
return on the S&P 500 was only half what it had been from 1946 to
1969, while the annual standard deviation nearly doubled, to 22%.
During 12 of the 24 calendar quarters over this period, an investor in
the stock market would have been better off owning Treasury bills.

Professional managers, who by 1969 had pushed client portfolios as
high as 70% in common stocks, felt like fools. Their clients took an
even harsher view. In the fall of 1974, the maiden issue of The Journal
of Portfolio Management carried a lead article by a senior officer of Wells
Fargo Bank who admitted the bitter truth:

Professional investment management and its practitioners are inconsistent, unpredictable, and in trouble.... Clients are afraid of us, and
what our methods might produce in the way of further loss as much
or more than they are afraid of stocks.... The business badly needs
to replace its cottage industry operating methods.20

For the first time risk management became the biggest game in
town. First came a major emphasis on diversification, not only in stock
holdings, but across the entire portfolio, ranging from stocks to bonds to
cash assets. Diversification also forced investors to look into new areas
and to develop appropriate management techniques. The traditional
strategy of buy-and-hold-until-maturity for long-term bonds, for example, was replaced by active, computer-based management of fixedincome assets. Pressures for diversification also led investors to look
outside the United States. There they found opportunities for high
returns, quite apart from the diversification benefits of international
investing.

But even as the search for risk-management techniques was gaining
popularity, the 1970s and the 1980s gave rise to new uncertainties that
had never been encountered by people whose world view had been shaped by the benign experiences of the postwar era. Calamities struck,
including the explosion in oil prices, the constitutional crisis caused by
Watergate and the Nixon resignation, the hostage-taking in Teheran,
and the disaster at Chernobyl. The cognitive dissonances created by
these shocks were similar to those experienced by the Victorians and the
Edwardians during the First World War.

Along with financial deregulation and a wild inflationary sleighride,
the environment generated volatility in interest rates, foreign exchange
rates, and commodity prices that would have been unthinkable during
the preceding three decades. Conventional forms of risk management
were incapable of dealing with a world so new, so unstable, and so
frightening.

These conditions gave rise to a perfect example of Ellsberg's ambiguity aversion. We can calculate probabilities from real-life situations
only when similar experiences have occurred often enough to resemble the patterns of games of chance. Going out without an umbrella on
a cloudy day is risky, but we have seen enough cloudy days and have
listened to enough weather reports to be able to calculate, with some
accuracy, the probability of rain. But when events are unique, when
the shape and color of the clouds have never been seen before, ambiguity takes over and risk premiums skyrocket. You either stay home or
take the umbrella whenever you go out, no matter how inconvenient.
That is what happened in the 1970s, when the valuations of both stocks
and bonds were extremely depressed compared with the valuations that
prevailed during the 1960s.

The alternative is to discover methods to mute the impact of the
unexpected, to manage the risk of the unknown. Although diversification has never lost its importance, professional investors recognized some
time ago that it was both inadequate as a risk-management technique and
too primitive for the new environment of volatility and uncertainty.

Fortuitously perhaps, impressive technological innovation coincided with the urgent demand for novel methods of risk control.
Computers were introduced into investment management just as concerns about risk were escalating. Their novelty and extraordinary
power added to the sense of alienation, but at the same time computers greatly expanded the capacity to manipulate data and to execute
complex strategies.

If, as Prospect Theory suggested, investors had met the enemy and
it was them, now the search was on for protective measures that made
more sense than decision regret or myopia or the endowment effect. A
new age of risk management was about to open, with concepts, techniques, and methodologies that made use of the financial system but
whose customers were spread well beyond the parochial precincts of
the capital markets.