College buddies are great at helping you find jobs, but it turns out that they can help you pick stocks, too.

If they happen to be corporate board members, that is.

A new study investigating the flow of information between mutual fund managers and senior company officers finds that managers put bigger bets on companies run by people who attended the same school.

Significantly, those stocks outperformed the managers' choices of stocks in companies where the manager didn't go to school with company executives. And the old-boy stocks they picked also did better than old-boy stocks that they chose not to bet on. (A free copy of the study is available here.)

"What we document using these networks is not an isolated situation or constrained to a few portfolio managers or firms, but rather a systematic effect across the entire universe of U.S. firms and portfolio managers," wrote the researchers, Lauren Cohen of Yale University, Christopher Malloy of London Business School, and Andrea Frazzini of the University of Chicago.

The researchers tested four types of links, listed here in increasing strength: Those among people who attended the same school, those among people who attended the same school and received the same degree, those among people who attended the same school at the same time, and those among people who attended the same school at the same time and received the same degree.

Not surprisingly, the strongest connection exhibited the greatest return when compared to out-of-network stocks: 2.69 percent annual gains for the weakest link versus 9.04 percent for the strongest.

Harvard University is the most connected institution, with links to 12 percent of publicly traded firms and to 16 percent of active equity mutual funds. Other schools among the most-connected include the University of Pennsylvania, Columbia University and the University of Chicago.

Perhaps counterintuitively, if you take out Ivy League schools, the gains for old-boy stocks are slightly higher.

"The intuition behind this, we think, is that for smaller schools -- although they are turning out fewer people -- it might be the exact reason ties are stronger," Cohen said in an interview. "If I meet someone else from Stanford, it's much less common, so I really have something to bond with that person over."

The study is the latest entry into the field of behavioral finance, which has traditionally focused on psychological factors in decision-making.

"Behavioral finance has done a good job incorporating psychology, but not so much on social networks," Cohen said. "Peer-effects and reactions to how others act have a huge effect on behavior."

In this particular case, the information flow seems to only work in one situation: When senior company officials have good news to share. The researchers found that when it came time to sell stocks, fund managers actually did better at timing the sale of stocks where there was no old-boy connection. This implies that board members were more loathe to reveal negative information.

What may be most interesting to regulators, however, is that price changes tied to news releases accounted for "the entire difference in return between stocks within and outside a manager's network."

Coincidence? Or something more nefarious, like insider trading?

The researchers did collect data on the most connected managers and funds, but they would not release them.

"We can't say with certainty the exact mechanism" of the information flow Cohen said. "What we can say with certainty is that these managers are taking more concentrated bets on firms in their own education networks, and that these bets are paying off."

by Zubin Jelveh