My Market Theory? Forget Theories
George Soros, Letters to the Editor, January 8, 2001

In discussing whether markets are efficient on the Dec. 28 editorial page, both sides ignore a crucial feature of financial markets, namely their ability to influence events ("Are Markets Efficient?": "Yes, Even if They Make Errors" by Burton G. Malkiel and "No, Arbitrage Is Inherently Risky" by Andrei Shleifer).

Both sides treat stock prices as if they were passive reflections of underlying fundamentals. Patently, that is not the case. The so-called fundamentals that are supposed to determine stock prices are not independently given; they are contingent on the behavior of the financial markets. There are myriad ways in which stock prices can affect the fortunes of companies: They determine the cost of equity capital; they decide whether a company will be taken over or can acquire others; stock prices influence the companies' capacity to borrow and their ability to attract and reward management through stock options; stock prices serve as an advertising and marketing tool.

Instead of a one-way connection in which financial markets discount the future more or less accurately, there is a two-way connection in which financial markets also shape the future they are supposed to discount. Instead of a single outcome, there is a range of possibilities. Which of those possibilities will materialize depends not only on the future evolution of the so-called fundamentals, but also on the present behavior of financial markets. In these circumstances, it would be irrational for market participants to base their decisions solely on their expectations about the fundamentals because the fundamentals do not determine market prices; on the contrary, they are themselves shaped by market conditions. What matters is the future course of market prices and not the fundamentals they are supposed to reflect. If market prices deviate from a theoretical equilibrium there can be no assurance that they will ever return to it.

Mr. Shleifer builds his case against efficient markets on the arbitrage opportunity between Royal Dutch and Shell. He points out that the arbitrage is risky because if prices can deviate from equilibrium there is no reason why they should not deviate even further. The point was proven by what happened in the Long Term Capital Management crisis. The argument is valid as far as it goes, but it does not go far enough. In the Royal Dutch/Shell case there is an equivalence that is not altered by any deviation from it. That is a special case. As a general rule, not only do market prices diverge from a theoretical equilibrium, but the divergence can also shift the theoretical equilibrium.

Take the recent boom/bust sequence in Internet and telecommunication stocks. The Internet boom accelerated the development of Internet and allowed a business model to proliferate that was based not on profit expectations, but on the expectations of selling stock to the public at ever- increasing prices. The telecommunication boom has also accelerated the spread of new technology and the bust was caused not by an unsound business model, but by the overextension of credit. In both cases, the innovations will continue, but at a slower pace, and the slowdown will affect the fundamentals of individual companies almost as much as their stock prices; there are also some macro-economic consequences.

These developments cannot be explained in terms of efficient markets and rational expectations. I have proposed an alternative explanation based on the two-way connection between fundamentals and valuations, which I call reflexivity.

I find it both amazing and amusing that economists have been able to ignore my arguments. They claim that the phenomenon of reflexivity is already taken into account in the concept of multiple equilibria that has recently gained acceptance. Maybe so, but I have yet to see how the theories of efficient markets and rational expectations can be reconciled with multiple equilibria. If market prices can consistently diverge from a theoretical equilibrium when the equilibrium is independently given, as in the case of Royal Dutch/Shell, how much more room is there for divergence in a world of multiple equilibria? The divergence amounts to more than noise; it is a causal factor in determining which equilibrium will prevail. By ignoring it, efficient market theory presents a totally misleading picture of financial markets.

Why is my reasoning dismissed out of hand, without giving it serious consideration? Because it leads to the conclusion that financial markets are inherently unpredictable. And what is the value of a scientific theory that does not yield usable predictions? I contend that it would be better to recognize the uncertainties inherent in the behavior of financial markets than to cling to a purportedly scientific theory that distorts reality. If financial markets are potentially unstable, there is a need for financial authorities who make it their business to prevent excesses. The debate about efficient markets may be arcane, but the consequences for our lives are very real indeed.

Are Markets Efficient?

By Burton G. Malkiel, author of "A Random Walk Down Wall Street." He is a director of the Vanguard Group and of the American Stock Exchange, Wall Street Journal, Dec. 28, 2000.

There is an old story about a finance professor and a student who come upon a $100 bill lying on the ground. The student stoops to pick it up. "Don't bother," the professor admonishes. "If it were really a $100 bill, it wouldn't be there."

This story illustrates what financial economists mean by efficient markets. Markets can be efficient even if investors are subject to overconfidence and errors in judgement. Markets can be efficient even if they make errors in the valuation of individual stocks and exhibit greater volatility than can apparently be explained by fundamentals such as earnings and dividends.

Many of us economists who believe in this efficient market theory do so because we view markets as amazingly successful devices for reflecting new information rapidly and, for the most part, accurately. Above all, we believe that financial markets are efficient because they don't allow investors to earn above-average returns without taking above-average risks. In short, we believe that $100 bills are not lying around for the taking.

While the efficient market theory has been the mantra of my generation, it has come under increasing attack from a new breed of economists. Their work has emphasized psychological and behavioral elements of stock-price determination; they believe future stock prices are somewhat predictable on the basis of past stock-price patterns and certain "fundamental" valuation metrics.

In their view, value stocks -- those with low ratios of stock prices to earnings and book values -- are alleged to outperform growth stocks, while small-company stocks supposedly do better than large-capitalization stocks. They believe that stock prices sometimes underreact to news, creating some short-run momentum, as well as sometimes overreact to events, creating price reversals that can be exploited by investors. Behavioralists also emphasize that the arbitrage activities of rational professional investors, who might be expected to bring stock prices back to fundamental values, are often impossible to execute and, in any event, risky and therefore limited.

These attacks on the efficient market theory are far from convincing. Some of the market patterns discovered may have rational causes; others may be spurious. But none of them are dependable in all time periods. And there is no evidence that rational investors can exploit any of the alleged mispricing in securities markets to earn above-average returns.

Many of the statistical patterns behaviorists emphasize could have rational as well as psychological explanations. Some long-run evidence suggests that growth stocks produce lower returns than value stocks. Behavioralists argue that this reflects investor overconfidence about optimistic growth forecasts and overpricing of growth stocks. But it is also possible that stocks selling at low valuations relative to their book values reflect some degree of financial distress and riskiness, so they should offer higher rates of return.

Behavioralists also believe stock prices exhibit reversals because individuals overreact to recent events. But reversals for the market as a whole could be caused by the tendency of interest rates and risk perceptions to fluctuate, meaning that stock prices are simply rationally adjusting to underlying economic conditions.

Many of the predictable patterns behaviorists claim to find in stock prices may be the result of endlessly mining the vast financial data banks now available until they cough up some seemingly significant, but wholly spurious, relationship. Moreover, findings of underreaction appear in the data about as frequently as overreaction and so could be random occurrences consistent with market efficiency. Many supposedly exploitable price patterns tend to become marginal or even disappear when alternative measurement approaches are used.

Even the strongest empirical regularities in the stock market aren't dependable. For example, small-cap stocks have historically outperformed large-cap stocks, while value stocks have outperformed growth stocks. But investors acting on that finding would have suffered very disappointing investment results indeed over this past decade of high-tech investment.

And what of the behavioral tenet that when psychological contagion pushes price-earnings ratios well above, and dividend yields well below, their historical averages, poor investment results must follow? An investor who followed that advice would have sold out in the mid-1990s when the stock market was less than half its present value. While the "Dogs of the Dow" strategy of buying the depressed highest dividend yields in the Dow worked brilliantly in back tests, when mutual funds were recently introduced based on that technique the dogs didn't hunt.

Finally, even if systematic pricing patterns persist over time, it may be impossible to exploit them. One pattern my colleagues at Princeton and I attempted to exploit was the overreaction phenomenon leading to return reversals. We simulated a strategy of buying those stocks with the poorest three- to five-year performance (the losers) and selling those stocks with the best three- to five-year performance (the winners). We found that statistical patterns of return reversals held up and were strongly significant: The losers later enjoyed better performance and the winners performed more poorly. What the losers did not produce, however, were excess returns: The returns for both groups were the same. There was statistical evidence of mean reversion but no inefficiency that could be exploited for gain.

As Richard Roll, a brilliant financial economist and active money manager, has said, "I have personally tried to invest money, my client's and my own, in every single anomaly and predictive result that academics have dreamed up. And I have yet to make a nickel on any of these supposed market inefficiencies. An inefficiency ought to be an exploitable opportunity. If there's nothing investors can exploit in a systematic way, time in and time out, then it's very hard to say that information is not being properly incorporated into stock prices. Real money investment strategies don't produce the results that academic papers say they should."

As further evidence of how difficult it is to outguess the collective wisdom of the market, consider the relative performance of index funds (that simply buy and hold the entire market portfolio) and active mutual-fund managers. While more than half the active managers are beating the indexes so far in 2000, the long-run results are devastating. Over the past three-year, five-year and 10-year periods, more than 75% of active managers underperformed index funds when both are measured after expenses. Those that do outperform in one period are not typically the ones who outperform in the next.

There are some exceptions like Peter Lynch and Warren Buffet, but you can count the truly outstanding long-term overachievers on the fingers of one hand. If markets were nearly as inefficient as some would believe, it would be easier for well paid professionals to profit at the expense of those who make systematic mistakes in processing information. As Rex Sinquefield, co-author of a thorough compendium of past stock returns, has said: "There are three classes of people who don't think markets work; the Cubans, the North Koreans and active money managers."

To be sure, we sometimes know in advance of isolated instances of mispricing. My favorite this year occurred when 5% of Palm Pilot shares were spun off by its parent 3Com, which retained the other 95%. The market immediately priced Palm Pilot shares at a valuation that made 3Com's ownership interest "worth" more than $50 billion, much more than the entire $28 billion market capitalization of the parent company. It was as if the operational component of 3Com had a negative value. The mispricing persisted for a while because not enough Palm Pilot shares were available for borrowing to effect a profitable arbitrage. But over time, and with more Palm shares available, the mispricing was corrected. The story illustrates that even occasional irrationality in market prices doesn't create a profitable trading opportunity.

In summary, I remain skeptical that markets are systematically irrational and that knowledge of such irrationalities can lead to profitable trading strategies. Indeed, the more potentially profitable a discoverable pattern is, the less likely it is to survive. This is the logical reason one should be cautious not to overemphasize apparent departures from efficiency.

It is always possible that new patterns will emerge and be discovered, but such patterns must be exploited immediately because they are unlikely to last. The advice I give my students is slightly different from that of the finance professor of the story: "If you see a $100 bill on the ground, pick it up right away because it surely won't be there for long."

Are Markets Efficient?

By Andrei Shleifer, an economics professor at Harvard and author of "Inefficient Markets" (Oxford University Press, 2000), Wall Street Journal, Dec. 28, 2000.

The extraordinary performance of the stock market until recent months has led many skeptics -- from Federal Reserve Chairman Alan Greenspan to bestselling economist Robert Shiller -- to complain about irrational exuberance. Market enthusiasts have responded by pointing to the efficient markets theory, which holds that the market is far better equipped to assess the prospects of American companies than any pundit.

The idea that "the market knows best" was developed at the University of Chicago in the 1960s. It has gained enormous intellectual dominance since then and is now drilled into the heads of tens of thousands of business students around the world. Yet a growing number of economists, myself included, are of the belief that because arbitrage is risky business, markets are necessarily inefficient.

Risky Bets

The efficient markets theory holds that the trading by investors in a free and competitive market drives security prices to their true "fundamental" values. The market can better assess what a stock or a bond is worth than any individual trader. If the stock market is efficient, we don't need to worry about irrational exuberance or a crash, and we don't need to make up stories to explain the New Economy. But if the market isn't efficient, we are in for a meltdown, or at least a long period of mediocre returns.

Perhaps the most compelling evidence in favor of market efficiency is the inability of even the shrewdest investors to consistently beat the market. It isn't just that individual investors trail the passive benchmarks, such as the Standard & Poor's 500, by 2% to 3% a year. Nearly all mutual and pension funds also fail to beat the market on a consistent basis. Even the savviest investors -- George Soros, Warren Buffett, Julian Robertson -- occasionally stumble. If markets were irrational, the argument goes, then the very best investors would find strategies to make money consistently and without risk. The fact that they do stumble proves that the market truly knows best.

This last argument -- that even the best managers aren't able to outsmart the market -- is the most plausible, and has become the bedrock of the efficient market theory. Unfortunately, it is false, and for a very simple reason. In financial markets, bets against security mispricing, sometimes referred to as arbitrage, are bets that prices will converge to true values. With rare exceptions, such bets are inherently risky.

An overpriced stock today can become even more overpriced tomorrow, bringing losses to even the cleverest short-seller. A bargain today can become an even better bargain next month, bringing grief to a value investor. Even the shrewdest investors must bear these risks and so lose money on occasion. Some of the risks they face can be hedged, but many can't. Because rational arbitrage is always risky, it is inherently limited in its ability to bring prices to their true values. A free, competitive market is almost necessarily inefficient.


To illustrate this point, consider how efficient markets theory goes wrong. One very clear example is the pricing of the shares of Royal Dutch and Shell. Royal Dutch and Shell are independently incorporated in the Netherlands and England, respectively. In 1907, they formed an alliance agreeing to merge their interests on a 60-40 basis while remaining separate and distinct entities. All their profits, adjusting for corporate taxes and control rights, are effectively split into these proportions.

Information clarifying the linkages between the two companies is widely available. This makes for an easy prediction for the efficient markets theory: If prices are right, the market value of Royal Dutch should always equal 1.5 times the market value of Shell. In this case, the efficient markets theory reflects the law of one price: Identical securities must sell at the same price in different markets. If not, there would be clear and easy arbitrage opportunities from dumping the relatively expensive stock and buying the cheaper one.

The nearby chart shows the deviations of market values of Royal Dutch and Shell from the 60-40 parity from 1990 to 1999. In the early 1990s, Royal Dutch traded at a 5% to 7% discount from parity, while in the late 1990s it traded at up to a 20% premium. A closer look at the chart clarifies why the market doesn't bring the relative prices to efficiency.

A shrewd investor who noticed, for example, that in the summer of 1997, Royal Dutch traded at an 8% to 10% premium relative to Shell, would have sold short the expensive Royal Dutch shares and hedged his position with the cheaper Shell shares. Sadly for this investor, the deviation from the 60-40 parity only widened in 1998, reaching nearly 20% in the autumn crisis. This bet against market inefficiency lost money, and a lot of money if leveraged.

In this case, it is said that when Long Term Capital Management collapsed during the Russian crisis, it unwound a large position in the Royal Dutch and Shell trade. Smart investors can lose a lot of money at the times when an inefficient market becomes even less efficient. In fact, as the LTCM experience illustrates, their businesses might not survive long enough to see markets return to efficiency.

The inefficiency in the pricing of Royal Dutch and Shell is a fantastic embarrassment for the efficient markets hypothesis because the setting is the best case for that theory. The same cash flows should sell for the same price in different markets. It shows that deviations from efficiency can be large and persistent, especially with no catalysts to bring markets back to efficiency. It also shows that market forces need not be strong enough to get prices in line even when many risks can be hedged, and that rational and sophisticated investors can lose money along the way, as mispricing deepens.

But if markets fail to achieve efficiency in this near-textbook case, what should we expect in more complicated situations, when the risks of arbitrage are greater? Who would dare to sell short Internet stocks to bring their prices down to earth when a company trading at five times its fundamental value can easily rise to 10 times its value? Or who would bet against the overpriced S&P 500 as a whole? What would have happened to the sellers of the market who heeded Alan Greenspan's concerns in 1996?

If smart investors lose money whenever markets move away from efficiency, it is no puzzle that even the mightiest stumble. On average, they make money (the track records of Messrs. Soros, Buffett and Robertson are nothing to complain about) but the money they make is not without risk. The fact that arbitrage is risky in no ways implies that the market knows best. To the contrary, it shows that a misvalued market can become even more misvalued. As the great financial economist Fischer Black once wrote humorously, the market is "efficient" when security prices are within a factor of two from value.

In inefficient markets, active investment management pays off in the long run. Contrarian strategies -- betting against the mispricing -- do better over the longterm than indexation. Value stocks have in fact outperformed growth stocks over long periods in the U.S. and European markets. But these strategies are inherently risky precisely because markets can move further away from efficiency. The Internet bubble of 1998-99 killed the relative performance of value investors; Mr. Robertson was only one of the victims. The question for active investors is whether they can take the pain of volatility long enough before the bubble bursts.

Market Knows Better

The fact that markets aren't efficient doesn't imply that the government should regulate them. Far from it. There are many benefits of inefficient markets. The Internet boom would not have been possible -- at least not on the same scale -- without financing from irrationally exuberant investors. The millions of Americans now benefitting from stocks might have stuck with savings accounts without the boom. The proposals to reform Social Security -- both Democratic and Republican -- would not have even started if markets were moribund. Yet to keep the government away from markets, we do not need to proclaim that "markets know best." The weaker but more accurate proposition, that the market knows better than the government, is more than sufficient.