The hypothesis that financial markets are efficient — that asset prices at any given moment reflect all available information and therefore cannot be systematically outperformed by any individual investor — has been, for the better part of half a century, the central dogma of academic finance. Formulated in its modern guise by Eugene Fama in the 1960s, the efficient market hypothesis provided the theoretical underpinning for an entire generation of financial regulation, investment practice, and economic policy. It justified the deregulation of markets on the grounds that prices generated by the unimpeded interaction of rational agents were, by definition, the best available estimate of underlying value. It discouraged active portfolio management in favour of passive index-tracking strategies. And it furnished, to those who found it congenial, a powerful argument against government intervention: if the market was already incorporating all relevant information, then the regulator who presumed to know better was guilty not merely of hubris but of a logical impossibility.
The hypothesis was always more nuanced than its popularisers suggested. Fama himself distinguished between weak, semi-strong, and strong forms of efficiency, and the strong form — the claim that prices reflect not only public but also private information — was acknowledged, even by its proponents, to be empirically implausible. Yet the weaker versions, which merely asserted that prices incorporate all publicly available information, proved extraordinarily resilient, surviving decades of apparent counter-evidence through a combination of sophisticated econometric defence and the absence of a comparably elegant alternative theory.
The behavioural finance revolution, which gathered momentum from the 1980s onward, mounted what appeared to be a devastating empirical challenge to market efficiency. Researchers documented a catalogue of systematic anomalies — momentum effects, value premiums, calendar regularities, overreaction to news — that were difficult to reconcile with the hypothesis that prices accurately reflected fundamental values. More fundamentally, the psychologists Daniel Kahneman and Amos Tversky demonstrated that human decision-making under conditions of uncertainty was characterised not by the rational calculation of expected utility that economic theory presupposed but by a repertoire of cognitive heuristics and biases that produced systematic departures from rationality. If the agents who populated financial markets were not rational, the argument ran, then the prices they generated could not be efficient.
The defenders of efficiency responded with an argument of considerable ingenuity. Even if individual investors were irrational, they contended, their irrationalities would tend to cancel one another out in the aggregate, and the residual pricing errors would be exploited and eliminated by the minority of sophisticated arbitrageurs whose rationality ensured that market prices remained approximately correct. This defence — sometimes called the 'noise trader' argument — was formally elegant but rested upon an empirical assumption that the financial crisis of 2008 spectacularly falsified: the assumption that arbitrage was both riskless and unlimited, so that rational traders would always have the resources and the incentive to correct the mispricing generated by irrational ones.
What the crisis revealed was that the relationship between individual rationality and market efficiency was far more fragile than the hypothesis supposed. Arbitrageurs, far from being the stabilising force that theory predicted, had in many cases amplified the distortions they were supposed to correct, either because they were themselves subject to the same cognitive biases as other market participants or because the institutional constraints under which they operated — leverage limits, redemption pressures, career risk — prevented them from maintaining their positions long enough for rationality to prevail. The market, it turned out, could remain irrational considerably longer than any individual could remain solvent.
The efficient market hypothesis has not been abandoned — in its weaker forms, it continues to provide a useful benchmark against which deviations can be measured — but it has been comprehensively demoted from the status of established fact to that of an idealised approximation whose relationship to reality is, at best, intermittent. The lesson is not that markets are irrational but that their rationality is conditional, fragile, and dependent upon institutional and psychological conditions that cannot be assumed to hold. A financial system built upon the presumption of efficiency is a system that has mistaken a sometimes useful simplification for a permanent truth, and the consequences of that mistake, as the world discovered in 2008, can be catastrophic.