This is a CFA Institute summary of "The Efficient Market Hypothesis and Its Critics," by Burton G. Malkiel, from the Journal of Economic Perspectives, Winter 2003.
The efficient market hypothesis states that when new information comes into the market, it is immediately reflected in stock prices and thus neither technical nor fundamental analysis can generate excess returns. The author examines recent research related to behavioral finance, momentum investing, and popular fundamental ratios that purports to contradict the theory and concludes that it is not significant in the long run. Therefore, in his view, the efficient market hypothesis remains valid.
The efficient market hypothesis holds that when new information comes into the market, it is immediately reflected in stock prices; neither technical analysis (the study of past stock prices in an attempt to predict future prices) nor fundamental analysis (the study of financial information) can help an investor generate returns greater than those of a portfolio of randomly selected stocks. The author reviews the recent findings of three schools of thought that challenge the efficient market hypothesis based on their claims that evidence of predictable patterns in stock prices exists.
One school of thought challenging the efficient market hypothesis is momentum investing, a combination of technical and fundamental analysis that claims that certain price patterns persist over time. The second is behavioral finance, which maintains that investors are guided by psychology more than by rationality and efficiency. And the third is fundamental analysis, which holds that certain valuation ratios predict outperformance and underperformance in future periods.
Momentum investors base their argument against the efficient market hypothesis on the following. In a truly efficient market, the short-term serial correlations among stock prices should be zero, but several studies have shown examples of short-term serial correlations that are not zero, thus indicating the possibility of a discoverable pattern. The author, however, shows that although these findings are statistically significant, they may not be economically significant. For example, as soon as evidence of the so-called January effect was made public, investors incorporated the information into their investment decisions and the effect disappeared. Furthermore, momentum strategies do not perform well in all markets. Although they led to excess performance in the late 1990s, they generated underperformance relative to the poorly performing market of the early 2000s.
The author then addresses the findings of behavioral finance, which indicate that investors overreact to some events and underreact to others. He cites research indicating that underreaction is as common as overreaction and that postevent continuation of abnormal returns is as common as postevent reversals. In other words, what appears to be a trend according to the tenets of behavioral finance may merely be a random event.
With regard to fundamental analysis, many believe that initial dividend yield and price-to-earnings multiples can be used to predict future stock results. The author points out, however, that these measures do not consistently predict stock performance in all time periods, which means that they do not contradict the efficient market hypothesis. The author concludes that occasional anomalies do not violate the efficient market hypothesis; they lose their predictive power when they are discovered and do not hold true in the long run.
The author, a well-known proponent of the efficient market hypothesis, refutes the claims of all these schools of thought currently challenging the efficient market hypothesis. He notes, however, that a difference between market efficiency and perfect pricing exists; the market often misprices securities, at least in the short run, but an investor cannot know before the fact when mispricing will occur.