There is no question the stock market has had its ups and downs, but have you ever wondered what drives people to pull their money out of or pour money into a company? Certainly investors pay attention to analyst forecasts, but to which analysts are they paying the closest attention? Beverly Walther (Kellogg School’s Department of Accounting and Information Management), Sarah E. Bonner (University of Southern California) and Artur Hugon (Georgia State University) explored whether investors react stronger to “celebrity” analysts and why.
Past research has shown mass media can influence people’s beliefs or behavior in general. Such studies are at least partly behind advertisers’ willingness to pay higher rates to ensure their spots appear in popular newspapers and magazines and on air during time slots when the largest audience is believed to be watching or listening. Over the years, the media has devoted more and more attention to the stock market and its key players, such as analysts. Recent research shows the media plays an important role both in the stock price formation process and in accounting settings. Such research, however, focuses primarily on firms and not analysts. Given the analyst’s pivotal role in incorporating accounting information into stock prices and the media’s role in disseminating information they issue, Walther and her co-authors wanted to know how the media might also serve as an information intermediary for investors, specifically with regard to higher-visibility celebrity analysts and their earnings forecast revisions.
“Given the criticism recently levied at financial analysts,” says Walther, “we wanted to investigate how much of a factor name recognition plays in how investors react to the information analysts disseminate.”
Walther and her team found that stock prices react more strongly to earnings forecast revisions (the change in the analyst’s expectations of the accounting earnings number the company will report in the upcoming quarter) issued by celebrity analysts. They defined celebrity as a famous or well-publicized person, or in Daniel Boorstin’s (1962) words, someone “well known for his well-knownness,” in addition to performance-related qualities. Specifically, the researchers selected a random sample of analysts and measured celebrity by the amount of media coverage an analyst received between 1997 and 1999 in sources included in the Dow Jones Interactive Database.
Walther and her co-authors wanted to know if this stronger reaction occurred because celebrity analysts were superior performers or because their names were more recognizable to investors. Walther says there was no compelling evidence that celebrity analysts issued more accurate earnings forecasts. Moreover, although investors may react more strongly to these analysts, “there appears to be a price correction when the actual earnings are announced. If the price after the earnings forecast was too high or too low it is corrected.” Second, they tested whether stronger reactions occurred because the earnings forecast itself had gotten more media attention, but there was no evidence to support this alternative explanation. In the end, the researchers concluded the more an analyst’s name appeared in the media, the more likely investors were to pay attention.
“The central feature of this work is that in contrast to my prior work, we examine a characteristic of the analyst—his/her media coverage—that is not solely related to the performance of the analyst,” says Walther. “Unlike the previous findings in this area, our work suggests that a non-performance factor can affect prices.”
While Walther’s findings may not pertain to companies per se, the findings are important for investors who may want to heed this advice: “Don’t assume that just because an analyst has gotten more press that his or her forecast revisions are more accurate. Investors should also look at the historical performance of the analyst before deciding on whom to rely.”
Boorstin, Daniel J. (1962). The Image: A Guide to Pseudo-Events in America. New York: Random House, 25th edition, 1987.