Carla Hayn’s research shows well-known market anomalies largely absent among the biggest stocks
There has been a meteoric rise in the volume of data provided on traded securities over the last few decades. Through use of this data, produced by corporations through their regulatory filings or voluntary disclosures or by intermediaries (e.g., stock exchanges, brokerage houses), researchers have been able to detect many market anomalies that can be potentially exploited through arbitrage. Over the years, anomalies related to attributes such as firm size, the price-to-earnings ratio, accruals and post-earnings announcement drift were discovered. Such anomalies suggest some degree of market inefficiency.
In a working paper titled "Does the Tail Wag the Dog? Small-Firm Bias in Capital Market Research," Pennsylvania State University's Dan Givoly, UCLA Anderson's Carla Hayn and UC Irvine's Ben Lourie studied a number of such anomalies and conclude that, rather than suggest widespread market inefficiencies and lack of investor sophistication, the evidence on the anomalies is driven by "small firms which collectively represent a small fraction, typically less than 10%, of the market value of the equity markets." The bulk of the capital invested in the stock market is immune to such anomalies.
One of the anomalies examined is "post-earnings-announcement drift" or PEAD. This is the well-documented phenomenon whereby stock prices react to earnings news slowly, over weeks or months. Efficient-market theory holds that investors should quickly adjust a stock's price for good or bad earnings news. PEAD seems to defy that idea.
Analyzing stock returns from 1983 to 2015, the authors found that significant PEAD occurred primarily in stocks of the very smallest companies. The larger the company, the smaller the PEAD effect — to the point where larger firms representing "around 91% of the market value [are] not affected by the PEAD anomaly," based on the authors' findings.
Among the other anomalies revisited by Givoly, Hayn and Lourie is the "limited attention" hypothesis that posits that investors do not react correctly and promptly to news because they cannot handle the huge, continuous volume of information, and thus end up neglecting important news (at least for a while). Past studies provide evidence consistent with this theory. One such finding is that investors react belatedly to earnings reports made on Fridays (when they're busy planning their weekends); as a result, firms with bad earnings news tend to "hide" this negative information by releasing it on a Friday, when fewer investors might read it. Another finding is that earnings news released during the March Madness basketball games (again, a big distraction) goes less noticed. Givoly, Hayn and Lourie confirm these prior findings. However, when they break down the data into firm-size groups, they find that these distractions affect the price of smaller companies whose aggregate market value is only about 5 percent of the total market value of the stock market.
The authors allow that "Studying [market] anomalies enhances our understanding of investor and management behavior, trading mechanisms, the functioning of financial intermediaries, and the effectiveness of regulations." Yet, they also conclude that an important caveat should be attached to any finding of a market anomaly when the economic significance of the anomalies is limited, which is almost always the case.