Efficient Markets Hypothesis: Does the Theory Survive This Crisis?

The efficient markets hypothesis, the bedrock of modern finance, is under attack due to — get this — irrational behavior.  Believe it or not, it appears that some market participants behaved irrationally.  So where does that evidence leave us?

What is the efficient markets hypothesis?  The Wikipedia definition is pretty good:

"… the efficient-market hypothesis (EMH) asserts that financial markets are "informationally efficient", or that prices on traded assets (e.g., stocks, bonds, or property) already reflect all known information, and instantly change to reflect new information."

The implication of the efficient markets hypothesis is that trying to pick the best stocks or bonds will not do better than buying an index fund.

There have been a number of anomalies reported that cast doubt on the efficient markets hypothesis.  Most notably, the behavioral finance researchers have found plenty of biases in human decision-making.  These include recency bias (a tendency to put greater weight on recent performance than it deserves), herd behavior (doing what everyone else is doing), and framing bias (in which risky bets to make a gain are shunned, but equally risky bets framed as avoiding a loss are embraced).  The housing boom-bust, and the related booms-busts in mortgage-backed securities, reinforce the irrationality of much human decision-making.

But where does that leave us?  The efficient markets hypothesis may be overstated, but what actionable consequences does that leave us with?  The behavioral finance researchers have not presented a better way of estimating value than current prices.  They have simply cautioned us that current prices are not terribly accurate estimators of value.  Here are the lessons as I see them:

  • To investors, buy and hold a diversified portfolio still looks to be the best strategya, but do not take long-run average returns to be accurate indicators of what you will actually earn over your holding period. 
  • For companies, be very suspicious of strategies based on the assumption that the company can predict prices better than the market can.  (I once knew a mining executive who assumed that most of his gold production would be sold within 10 percent of market tops.  Talk about hubris!)

These conclusions generalize.  Even if a theory is found to be inaccurate in some regards, that does not imply that alternative behaviors are automatically proved to be optimal.  For example, finding that the classical economic conditions of perfect competition are not met does not necessarily imply that our government's intervention in the markets is optimal.