Seeking Alpha
, Buckingham (100 clicks)
Portfolio strategy, value
Profile| Send Message|
( followers)  

Summary

  • The efficient market hypothesis is a hypothesis that provides an important organizing principle that helps us understand how markets function and prices are set.
  • EMH asserts that financial markets are informationally efficient.
  • As anomalies have been uncovered, there’s mounting criticism of the EMH.

Today begins a four-part series on the efficient market hypothesis. We'll begin with a brief history and explanation of the EMH. Eugene Fama, recent recipient of the Nobel Prize in economics, is considered the father of the efficient-market hypothesis (EMH). The EMH asserts that financial markets are "informationally efficient." As a consequence, we would expect that there would be no persistence of risk-adjusted outperformance by active managers beyond what would be randomly expected.

There are three major versions of the EMH. The weak form hypothesizes that prices on stocks and bonds reflect all publicly available information. That would render technical analysis inefficient as a means of forecasting price movements - it's not possible to outperform based on analysis of past prices. The semi-strong form of the EMH hypothesizes that prices both reflect all publicly available information and prices instantly change to reflect new public information - neither technical nor fundamental analysis should result in abnormal returns. The strong form of the EMH hypothesizes that prices reflect all information - even the use of insider information should not generate abnormal returns.

As anomalies to the EMH have been uncovered, there's been mounting criticism of the EMH. For example, there's an overwhelming body of evidence that momentum is a factor that's both persistent and pervasive across markets and asset classes. Even Fama himself stated that momentum is the single greatest challenge to the EMH. Another well-known anomaly is the poor risk-adjusted performance of small growth stocks.

The existence of these anomalies clearly demonstrates that the markets aren't perfectly efficient. However, what's important to understand is that the EMH is a hypothesis, not a law (such as the law of gravity). It's a model, a simplification of the world that doesn't always hold true. And like all models, by definition they are wrong (or they would be called laws, like we have in the physical sciences). As John Cochrane pointed out, Fama himself recognized this. Writing in his first paper, Fama explained that "efficiency, like all perfect-competition supply-and-demand economics, is an ideal, which real-world markets can only approach. Empirical work can find only how close to or far from the ideal a given market is."

Interestingly, Fama, along with colleague Kenneth French, wrote the 2005 paper "Disagreement, Tastes, and Asset Prices", which showed how investor preferences for assets with "lottery-like" characteristics can lead to mispricings of assets when viewed from a rational perspective (helping to explain the small growth anomaly)

With that said, the EMH is a hypothesis that provides an important organizing principle that helps us understand how markets function and prices are set. And most importantly, as we'll discuss, unless there's clear evidence of an inefficiency, investors are best served by assuming the market is efficient. However, when there's clear evidence of an inefficiency, it's not prudent to ignore it just to hold to a stubborn belief in a pure form of the EMH. For example, faced with the overwhelming evidence on a momentum premium, Dimensional Fund Advisors (Fama has played an important role in the research efforts at DFA and is a member of the board of directors) began incorporating formal momentum screens into their portfolios in 2003. Full disclosure, my firm Buckingham recommends Dimensional funds in constructing client portfolios.

In addition, for a long time DFA has screened out from their eligible universe extreme small growth stocks, IPOs, "penny" stocks, and stocks in bankruptcy. In each case, they have done so because the research showed that these stocks delivered poor risk-adjusted returns (all problems for the EMH). In other words, even the investment firm most closely associated with Fama and his research doesn't invest according to a strict, or one might say a "religious," belief in the EMH.

With that said, the historical evidence on efforts to generate alpha by exploiting market inefficiencies shows that markets behave much closer to the EMH than most investors believe, and Wall Street and the financial media want and need you to believe.

There's one more important point we need to address about market efficiency - the debate over market efficiency is clouded by what is called the "joint-hypothesis problem." Any statement about market efficiency must be accompanied by a model of expected returns that tells us how stock prices behave in a perfectly efficient world. As Fama himself pointed out in one of his early papers, when we find evidence that stock prices are not explained by our model, it's not clear if this anomalous behavior is attributable to market inefficiency or a bad model.

Over time, the asset pricing models have greatly improved in their ability to explain the variation in returns of diversified portfolios. What was once alpha (outperformance), has become beta (exposure to a factor that explains returns). The original CAPM model was a single factor (beta) model. In 1992, Fama and Kenneth French developed the three-factor model, adding the size and value factors. With that change, no longer could value fund manager John Doe, who had beaten the S&P 500 Index over a 20-year period, be hailed by observers as a stock-picking genius if all the outperformance was explained by his exposure to the value premium.

In 1997, Mark Carhart added momentum as a fourth factor. And more recently, profitability has been added as a fifth factor helping to explain returns. As these factors were added, alpha disappeared (was converted to beta), as did some anomalies. Today's improved models have been able to eliminate most, but not all, of the anomalies. In other words, the models aren't perfect.

In tomorrow's post, we'll look at the evidence from both mutual funds and pension plans to determine if their alpha-seeking efforts have been rewarded.

Source: The Efficient Market Hypothesis, Fact Or Fiction? Part 1