Clement Research Analyzes

the Stock Analysts

By Tracy Mueller

“Many analysts at big brokerage firms pull down six or seven figure incomes and can make a stock’s price dance by uttering a few magic words on the morning call to a firm’s brokerage force.”
 
Michael Clement, associate professor of accounting, read this quote from a 1994 Wall Street Journal article, and was struck by one question: Who are those high-paid stock pickers?
 
That was the subject of his Feb. 12 lecture, part of the Faculty Research Presentation Speaker Series.

Video: Watch a clip from Prof. Clement's talk.
 
Clement researches analysts’ behavior and performance, and examines which characteristics are shared by the most accurate earnings forecasters.
 
“Anytime I see someone making that much money, I wonder if they really are that good, and if some are actually better than others,” Clement said.
 
He found that an analyst’s forecast accuracy improves with general experience in the industry and with firm-specific forecasting experience.
 
“So if you see 20 analysts following Intel and you're trying to figure out which one to listen to, look at the one who has the most experience, in general and with Intel," Clement said.
 
Clement’s research also shows analysts at larger firms such as Morgan Stanley or Merrill Lynch supply more accurate forecasts, and that an analyst’s accuracy declines as he or she follows more firms and industries, a concept students should relate to.
 
"If you take too many classes, then your grades start to suffer, and you can't stay on top of the homework,” Clement said. “So as an analyst starts to follow a whole lot of companies, they can’t do as good of a job on each one."
 
Does Accuracy Matter?
 
Clement’s second research question asks how important forecast accuracy is to investors. He found investors often value timeliness over accuracy.
 
“A weatherman could tell you on Tuesday that a big storm is coming the next day, at exactly 10 o’clock,” Clement explained. “Or he could tell you right now that a storm is coming sometime next week, but he can’t say precisely when. Sometimes there is a trade-off between accuracy and timeliness, and the stock markets seem to value both.”
 
Analyst Bias
 
Clement’s most publicized research examines how an analyst’s relationship with a firm manager affects his or her decisions.
 
He found managers are more likely to grant favors to analysts after poor earnings reports. Favors could include connecting the analyst with a manager at another firm or meeting with an analyst’s clients.
 
“It’s kind of like taking your professor out to lunch after you receive a bad grade,” Clement joked.
 
He also found analysts who received favors are less likely to downgrade stocks after poor earnings reports. In fact, in a 2003 investigation, regulators found analysts recommending stocks to their clients, but in private emails to friends, declaring “I would never buy this stock,” Clement said.
 
“These people are clearly operating with a set of biases, so we must be aware of that if we’re going to make decisions based on their information.”