EDITOR’S NOTE: Goizueta Newsroom regularly shares news on the school, its faculty, staff, students and alumni. But we also take pride in sharing vital pieces of research from the world-class business school faculty. Look for regular updates on published pieces of research like this one…
Those on Wall Street with skill in picking stocks tend to be compensated in one of two ways.
Some run mutual funds that actively invest in stocks and collect fees from investors. Others sell information to investors through research reports published by brokerage firms. The relationship between the two groups of investment professionals has generated considerable attention over the years in the media as well as among academics. The New York Times, for example, once called it the “dirty little secret” of Wall Street that research from sell-side analysts is meant primarily for institutional investors rather than individuals.
The idea is that mutual fund managers and other sophisticated investors best understand conflicts of interest at brokerage firms and know when reports or recommendations can’t be taken at face value. But research from a trio of finance professors at the Goizueta Business School — Jeffrey A. Busse, T. Clifton Green and Narasimhan Jegadeesh — challenges one key aspect of the “dirty little secret” theory.
Their study results suggest that institutional investors don’t, in fact, have any special skill in discerning the quality of analyst recommendations.
Buy-side trades and sell-side recommendations: Interactions and information content (Journal of Financial Markets)
The researchers began their analysis by judging the performance of both mutual fund managers and stock analysts. For the first group, they looked at data from two sources that collectively contained data on more than 5 million transactions involving more than 10,000 stocks. Together, the data sets contained trades by 908 different institutional investors over a span of 12 years beginning in 1993.
Mutual fund performance was judged in terms of both the abnormal returns from stock purchases as well as the difference between abnormal returns from purchases and sales. Researchers also considered whether funds of different sizes exhibited different levels of skill in selecting stocks. They concluded that the value of funds’ private information was roughly one percent and found no evidence that skill varied with fund size. To evaluate analyst performance, the researchers looked at 135,652 recommendation revisions covering 8,174 different stocks during a similar time period. A comparison of abnormal returns from the two groups indicated that sell-side analysts were more skilled.
Next, the researchers considered how the two groups interacted with one another.
Many institutions pay brokerages for their research, so recommendation revisions likely play a role in trading decisions, the researchers noted. Similarly, it’s possible that analysts become aware of trades by their mutual fund clients and use the information in formulating recommendations. To evaluate how the actions of each group affects the other, the researchers looked at the pattern of trades both before and after analyst recommendations. They found that while buy-side trades often followed recommendations from sell-side analysts, the dynamic didn’t seem to work the other way around. There also was evidence that analysts might tip fund managers prior to downgrades, but not upgrades.
Finally, researchers addressed the “dirty little secret” theory.
If mutual fund managers know how to selectively use good information from analysts and ignore bad recommendations, then upgrades that institutions purchase should outperform upgrades that they sell, the researchers wrote. Conversely, downgrades that institutions purchase should outperform downgrades that they sell. Yet, the study showed that purchases around recommendation revisions outperformed sales by a relatively small magnitude, suggesting that mutual fund managers don’t have private access to superior information from analysts.
“We find that whether funds trade in the same direction as recommendation revisions or in the opposite direction does not contain incremental information,” the researchers wrote. “Our findings indicate that the empirical evidence is not consistent with the notion widely expressed in the media and in some academic studies that institutional investors are able to see through any inherent biases in recommendations due to sell-side analysts’ conflicts of interest.”
– Chris Snowbeck