In the fitful debate about the role of the market in the financial crisis, the efficient-market hypothesis looms as a sort of loyalty test. If you’re a conservative, free-market type—a Randian, Austrian, Milton Friedman-meets-Ronald-Reagan Chicagoan with a supply side streak—you view the notion of market efficiency as broadly correct; the market really can process available information “efficiently,” certainly better than cartoon bureaucrats in Washington can, and thus produce roughly “correct” prices (this is also one explanation for why the market is so hard to beat).
If you’re more distrustful of the market, brought up short by Stiglitzean market breakdowns and convinced of imperfect competition and wandering equilibria, you view the efficient market, and its accomplice, the rational-expectations hypothesis, as dangerous fantasies belied by empirical reality. Not surprising, that latter point of view has been accumulating adherents after a period in which the efficient-market hypothesis seemed to reign supreme. That reign ended with the kind of market breakdown that proponents of the “strong” version of the efficient hypothesis had all but ruled out as impossible. This has proved to be a problem for the efficient crowd.
Still, what has been missing in all this to-and-froing is a rigorous examination of the conceptual grounds for the efficient-market hypothesis. Now we have it in a paper written by University of Florida law and philosophy professor Charles Collier ’72 titled “An Inefficient Truth,” published last year in the journal Critical Review. Part of the difficulty of wrapping your mind around the efficient-market hypothesis is that it’s rhetorically slippery, not only because it comes in a variety of flavors—weak, strong, semi-strong—but because each of them involves different assumptions and claims about what markets can and cannot do. Moreover, while the efficient market emerged from an intensely abstract and mathematical school of financial economics, defining and rationalizing it often depends on elusive, and often counterintuitive, definitions of words and dubious tests. In fact, as Collier burrows into the matter, many rationalizations for market efficiency resemble tautologies: prices in an efficient market are “correct” because they are produced by an efficient market.
The issue comes down to testability. How do we know that these prices are “correct,” based on available information? Can the hypothesis be falsified, as Karl Popper once defined scientific statements? This proves to be difficult, not least because financial markets are forward-looking, and so we are always analyzing current prices as reflections of future projections. How do we know prices reflect a rational interpretation of available data or a collective set of views, some of which may be “right” while others are clearly “wrong”—that makes no progress toward equilibrium? After all, in the short run, market prices clearly move in a random walk. How do we interpret that? As Collier writes, “Prices may be wandering randomly not because the market is reacting to unpredictable new information, but because the consensus is unwittingly traversing the periphery of a colossal bubble that formed for no fundamental reason and is about to burst for no good reason either.”
Even the University of Chicago’s Eugene Fama, the father of the efficient-market hypothesis, recognized the difficulty of proving the concept. To test the hypothesis requires both actual prices and a sense of the corresponding equilibrium price in the future. “Market efficiency per se is not testable,” wrote Fama in a 1991 paper. “It must be tested jointly with some model of equilibrium, an asset-pricing model. . . . We can only test whether information is properly reflected in prices in the context of a pricing model that defines the meaning of ‘properly.’ As a result, when we find anomalous evidence on the behavior of returns, the way it should be split between market inefficiency or a bad model of market equilibrium is ambiguous.”
As Collier comments, this is like seeing a car driven poorly and not being able to figure out if it’s the fault of the car or the driver.
Attempts to establish that equilibrium model, through financial economic innovations such as the capital-asset pricing model, have tended to unravel because of simplified assumptions and a belief in normal probability distributions; as a number of commentators like Tufts’ Amar Bhidé have suggested, market complexities outstrip the models. Collier particularly questions the well-known arbitrage argument of the efficient hypothesis. This says that liquid, efficient markets with many investors free to buy or sell will quickly arbitrage away anomalies—that is, investor opportunities for windfall gains. In real markets, however, many small investors may be supplanted by a handful of large, sophisticated investors, who may find themselves holding what they view as undervalued assets, which, as funding pressures mount, they eventually have to bail out of before the market “corrects” in their direction. Collier’s example: the plight of Long-Term Capital Management, which imploded before its “rational” bets came home. “This analysis implies that arbitrage is ineffective in bringing about pricing efficiency. If even the low-hanging fruit—or rather, if especially the low-hanging fruit—cannot be picked off because of funding considerations, there is little chance that prices will reflect all available information as well as the ability to act on that information.”
Collier covers a lot of ground. He raises the question whether there is a kind of uniform rationality shaping markets—or whether investors adopt a range of impulses, from self-fulfilling prophecies to “election” behavior, in which speculators judge not whether the price is right or wrong but what the crowd thinks about the price (this observation was famously articulated by John Maynard Keynes in his metaphor of the newspaper beauty pageant). “If an ‘electorate’ that is ignorant or irrational is determining market prices,” he writes, “then trading based on a correct and rational bet against current prices would perform poorly relative to a weighted market index.” Prices may never catch up and return to a theoretical equilibrium. Markets can easily grow “unbalanced and unstable.”
Recent history has tossed some cold water on the efficient-market hypothesis, particularly the strong form. But given that as denizens of the markets we are always living for the future, it’s possible that we are too quick to describe the panic of 2008 as a bout of irrationality; in the longer view, perhaps that represents a rational correction based on new information, a milestone on the road to a new equilibrium, a new “normal.” Given that, Collier’s discussion of the untestability of the idea is devastating. If the hypothesis cannot be falsified, then it comes down to a matter of belief, faith or ideology. Such a market can thus be inflated into a broad metaphor for all aspects of life that involve valuing and judging. How would the world be different, he asks, if markets were inefficient? As he concludes, since the efficient hypothesis cannot be confirmed or disproved, “one is entitled to assume either that it cannot be spelled out or that the resulting picture would strikingly resemble the financial markets that we have actually been stuck with for all these years.”