Stock buy recommendations from brokers who seldom give such advice fare better

ANN ARBOR—Pessimistic or optimistic? It may be best to have two different stockbrokers.

Brokers who infrequently make recommendations to buy stock tend to give good advice when they do say to buy. But these same investment banks and brokerage firms are not as successful when they advise to hold or sell, says a University of Michigan business professor.

New research by Reuven Lehavy of Michigan’s Stephen M. Ross School of Business and colleagues show that a broker’s stock ratings distribution can predict the profitability of its recommendations.

"Pessimistic" brokers—those who issue the smallest percentage of buy recommendations—outperform those brokers who issue the greatest percentage of buy ratings ("optimistic" brokers) by an average of 50 basis points per month. Conversely, downgrades to hold or sell issued by optimistic brokers outperform those of pessimistic brokers by an average of 46 basis points per month.

In their study, Lehavy and colleagues Maureen McNichols of Stanford University and Brad Barber and Brett Trueman of the University of California analyzed 438,000 stock recommendations issued on more than 12,000 firms by 463 investment banks and brokerage firms between
They examined the distribution of brokers’ stock ratings across buys, holds and sells, and determined what effect stricter securities rules—requiring investment banks and brokerage firms to publicly disclose the distribution of their stock ratings—have had on the observed tendency of analysts to issue many more buy than sell recommendations.

The National Association of Securities Dealers (NASD) proposed Rule 2711 in early 2002 as part of an effort to regulate the provision of research on Wall Street. Among other requirements, Rule 2711 requires all analyst-research reports to display the percentage of the issuing brokerage firm’s recommendations that are buys, holds and sells. This disclosure requirement is intended to provide investors with useful information to evaluate the quality of brokerage firms’ recommendations and, implicitly, to pressure brokers who consistently issue a relatively high percentage of buy recommendations to adopt a more balanced ratings distribution.

Lehavy and colleagues report that the percentage of buy recommendations increased substantially from 1996 to 2000, at one point exceeding the number of sell ratings by a ratio of more than 35 to 1 (74 percent buys vs. 2 percent sells). This practice, they say, appears to have been an industry-wide phenomenon that was used by the 10 largest investment banks, which later were singled out by the Securities and Exchange Commission for sanction in the 2003 Global Analyst Research Settlement, as well as by non-sanctioned brokers.

Beginning in mid-2000, the percentage of buys in the sample decreased steadily, and by the end of June 2003, buys exceeded sells by less than a 3-to-1 ratio (or 42 percent vs. 17 percent).

"This decrease probably was due, in part, to a worsening economy and a declining stock market," says Lehavy, assistant professor of accounting at U-M’s Ross School. "However, our findings strongly suggest that the implementation of NASD Rule 2711 also played an important role."

The reduction in the percentage of buy recommendations was most pronounced during the last half of 2002 when the new disclosure requirements became effective. During that time, the percentage of buy recommendations decreased from 60 percent to 45 percent, while the percentage of sells rose from 5 percent to 14 percent.

For more information on Lehavy, visit http://webuser.bus.umich.edu/rlehavy/