CSOM Professor Examines Profitibility Of Stock Selection Strategies
Published: Sunday, October 6, 2013
Updated: Sunday, October 6, 2013 23:10
Finance, accounting, and economics journals have long featured studies revealing stock market anomalies, which can be exploited through money-making investment strategies that theoretically should have been arbitraged away. According to a recent study published by Boston College’s Jeffery Pontiff, a professor in the Carroll School of Management, and David McLean of the University of Alberta, however, the mere publication of these studies has consequences for their profitability. If the anomalies found are more than statistical noise—data quirks rather than strategies that could be repeated profitably—then investors should take note of the studies and take advantage of the anomalies. Accordingly, if investors began following the investment strategies outlined by the academicians, the profitability would fall.
Pontiff and McLean’s study, “Does Academic Research Destroy Stock Return Predictability?” examined 82 separate strategies that were shown to predict stock market returns in 68 papers published in finance, accounting, and economics journals between 1972 and 2011. The study measured the returns of such strategies before and after publication. It found that after the papers were published, the returns of the strategies eroded by an average of 35 percent.
After a study is published, investors will learn from it and trade on its strategy, which pushes prices toward their fundamental value. This behavior has the effect of making the strategies less profitable, as it corrects the very mispriced stocks upon which they are based. For example, if a study showed that smaller firms tended to be undervalued and produced abnormally high returns, then investors who read the article would likely buy the stock of smaller firms, pushing up demand and consequently price to its proper value.
Not all types of strategies, however, saw their returns drop by the same degree. The study found that strategies involving more liquid stocks (those that are traded more frequently) saw their profits drop the most, likely because investors have an easier time following a strategy for which trading prices are low. In Pontiff’s words, “we cannot predict how much any strategy’s returns will go down, but we did find that the easier to trade on, the more the strategy’s returns will go down.” On the other hand, strategies involving more expensive trading—that of smaller, volatile stocks—saw their profitability fall the least. Moreover, the study found that returns based on the anomalies began to rise again several years after publication.
The cost of trading on some strategies helps to explain why the anomalies failed to disappear completely. Another reason may be constraints on the ability of investors to follow certain strategies, such as going short, that pension funds and mutual funds are unable to do.
According to Pontiff, the reason for the study was “the question everyone wanted to know the answer to. What had all [these studies] amounted to?” In other words, what had been the longer-term credibility, and effects, of the past studies. “If the returns based on these strategies had all gone down 100 percent, then we would just be historians,” he said. The question driving the study was whether the strategies would work outside of their original samples, and how much use the past studies could be in predicting future behavior, rather than documenting past features of the market.