Efficient Market Hypothesis (EMH)

The efficient market hypothesis is the idea that the prices of securities already reflect the information available about them, so that a security is neither a bargain nor overpriced in any way an investor can reliably exploit. It was formalized by the economist Eugene Fama, who later shared a Nobel Prize, and its logic is competition: with millions of profit- seeking investors analyzing the same public information, any obvious mispricing is bid away almost as soon as it appears. What remains is a price that has already absorbed what is known, and moves mainly on genuinely new information, which by definition cannot be predicted.

The practical conclusion follows directly. If prices already reflect available information, then the effort of picking individual winners or timing entries and exits is unlikely to pay for itself, because the advantage it seeks has usually already been priced in. This is the intellectual backbone of passive, index-based investing: rather than pay to search for mispriced securities, hold the whole market cheaply and accept its return. The hypothesis does not promise prices are correct, only that errors are hard to find and harder to profit from after costs.

The hypothesis is usually stated in three forms that differ in how much information they claim prices already reflect. The weak form says prices reflect all past price and volume data, so technical analysis (the study of past price patterns) cannot reliably generate excess returns. The semi-strong form says prices reflect all publicly available information, including financial statements and news, so fundamental analysis of public data cannot reliably win either, because everyone else has the same data. The strong form says prices reflect all information, public and private, so even insider information could not beat the market; this form is the least defensible and is largely contradicted by the fact that insider trading is both illegal and profitable when it occurs. Most evidence is consistent with markets being somewhere around weak-to-semi-strong efficient: hard to beat with public information, not perfectly efficient.

A crucial and often-missed point is that the hypothesis is joint with whatever model of "fair" return is used to test it. To say a return was abnormal, one must first say what a normal return should have been, and that requires a risk model. So a finding that some strategy "beat the market" can mean the market was inefficient, or it can mean the risk model was incomplete and the strategy was simply being paid for a risk the model did not measure. Much of what once looked like evidence against efficiency turned out, on closer inspection, to be compensation for identifiable risk factors rather than free money, which is where the study of factor investing grew from.

The hypothesis has serious challengers, and honesty requires naming them. Behavioral finance documents that investors are systematically prone to errors (overreaction, herding, loss aversion) and argues that these can push prices away from fundamental value for extended periods, with asset bubbles as the dramatic case. The efficient-market response is that exploiting even a real mispricing is dangerous and costly (prices can stay wrong longer than a contrarian can stay solvent), so the difficulty of beating the market can survive even if prices are not always right. The debate is genuinely unresolved among credible economists. What is not seriously disputed is the practical takeaway that most active managers, after fees, fail to beat the market over long periods, which is the part of the hypothesis with the strongest evidence behind it.

The hypothesis is a claim about how fast prices absorb information, so its mechanism is best seen through an example. Suppose a company announces earnings far above what analysts expected. Under the efficient market hypothesis, the share price jumps almost immediately to a new level that reflects the surprise, within seconds or minutes, not over the following days. An investor who reads the news that morning and buys is too late: the price has already moved, and what they pay already embeds the good news.

The consequence for strategy is what matters. Imagine two investors over many years. One pays 1.0% a year to an active manager who tries to find mispriced stocks; the other pays 0.05% to hold a broad index fund. For the active manager to leave the second investor behind, the manager must beat the market by enough to cover the roughly 0.95% annual cost gap, every year, reliably. The efficient market hypothesis argues that identifying enough mispricing to clear that hurdle, consistently, is very hard, and the record of active funds bears this out: on a $100,000 balance, that fee gap alone is $950 in the first year, compounding against the active investor thereafter.

This does not prove the market is never wrong. It argues that the errors are hard to find in advance and expensive to chase, so for most investors the low-cost, whole-market approach is the rational default. All figures are illustrative.

The case for it

• It explains why obvious bargains are rare and why most active managers fail to beat the market after fees.
• It provides the rigorous argument for low-cost, broadly diversified index investing.
• Its weak and semi-strong forms are well supported by decades of evidence on the difficulty of consistent outperformance.

The case against it

• Asset bubbles and crashes are hard to square with prices that always reflect fair value.
• Behavioral finance documents systematic, predictable investor errors that can move prices away from value for long stretches.
• Every test of efficiency is joint with a risk model, so "the market was beaten" and "the risk model was wrong" cannot always be told apart.
• The strong form is contradicted by the reality that private information can and does produce profits.