School Ties and Mutual Fund Success

Among managers with the strongest connection to senior officials (same school at the same time with the same degree), the connected holdings earned an average annual 16.05 percent excess return (the total gain minus the Treasury bill return). That was more than double the fund's average 7.69 percent excess return on non-connected stocks or the fund's overall average: 7.81 percent.

"Who you know" -- often touted as the key to career success -- also may be a big part of investing success, at least for the professionals who manage mutual funds. Consider the stock-picking savvy of one Mr. Smith, portfolio manager of the Phantom Aggressive Fund (his name and the fund's name have been changed). In the six months starting in October 1995, he bought 233 thousand shares of Cummins Engine Co., which soon thereafter announced two joint ventures, record annual sales, and a new market for its engines. By June 1997, the engine maker's stock was up 72 percent and Smith, having bought additional shares, sold all of his holdings for a hefty profit. His secret?

The secret is connections -- specifically, university connections -- conclude Lauren Cohen, Andrea Frazzini, and Christopher Malloy in The Small World of Investing: Board Connections and Mutual Fund Returns (NBER Working Paper No.13121). Smith is a Harvard MBA; Cummins is what the authors call a "Harvard stock." In 1996, 62 percent of the directors on its board held Harvard degrees; 46 percent of them had Harvard MBAs.

These university connections are not only common in the investment world, they're also profitable, the authors conclude. Managers tend to place bigger bets on companies with board members who share the same college or university affiliation. And, their holdings of such "connected" companies outperform their holdings of "non-connected" stocks by up to 8.4 percent a year.

Managers earned those outsized returns largely around news events, such as mergers, that boosted the stock price, the study finds. That finding suggests "social networks may be an important mechanism for information flow into asset prices," the authors conclude.

Of course, it's a matter of endless debate, even among academics who have investigated the issue, whether portfolio managers really do earn higher-than-normal returns. In the few studies that have found a positive link, the higher returns have been related to above-average SAT scores of the managers' undergraduate institutions or the geographical proximity of the companies they invest in. Other research has shown that managers tend to make similar portfolio choices if they live in the same city or have similar educational backgrounds. This study suggests not only that such educational institution links exist, but also that they give fund managers an informational advantage over other investors.

To put their thesis to the test, the authors focused on 2,501 portfolio managers of 1,648 actively managed equity funds that specialized in aggressive growth, growth, or growth and income stocks between January 1990 and December 2006. They matched the managers' educational backgrounds -- from undergraduate and business to law and even medical degrees -- with those of 42,269 board members and 14,122 senior officials at 7,660 publicly traded companies. They found many ties.

For example: of all publicly traded firms in the United States, 12 percent have mid-level managers and/or senior officers with Harvard degrees. So do the portfolio managers at 16 percent of active equity mutual funds. The University of Pennsylvania, University of Chicago, and Columbia University also consistently rank in the top five most "connected" schools, both among officers at public companies and mutual-fund managers.

The study finds that the stronger the educational connection, the more concentrated was the bet a mutual fund manager placed on a given company. The average stock in a fund's portfolio represented 89.4 basis points of the fund's assets. Funds in the study invested an added 28.45 basis points in companies whose senior officials attended the same institution as the portfolio manager. If their time on campus overlapped and they received the same degree, the manager invested an additional 41 basis points in the firm on average.

Those decisions turned a handsome profit. Among managers with the strongest connection to senior officials (same school at the same time with the same degree), the connected holdings earned an average annual 16.05 percent excess return (the total gain minus the Treasury bill return). That was more than double the fund's average 7.69 percent excess return on non-connected stocks or the fund's overall average: 7.81 percent.

"Connected holdings outperform non-connected holdings in a statistically and economically significant way for all four degrees of connectedness," the authors conclude. Such gains in the connected stocks did not involve increased risk, as measured by their Sharpe ratios.

The authors also find a significant advantage in performance -- up to 6.32 percent per year for the strongest connection -- for connected stocks that managers owned compared with connected stocks that they chose not to own. Such results "lend support to the hypothesis that fund managers have comparative advantages in gathering information about connected firms," the study says. By contrast, managers didn't appear very good at timing the selling off of connected stocks, a pattern consistent with the notion that portfolio managers were more likely to get positive rather than negative information through their social network.

To determine the robustness of their conclusions, the authors tested alternative explanations. They looked at the characteristics of the funds, those of the firms they invested in, and the industries. They divided up the holdings by geography, examined stretches of time within their sample period, and looked at whether a handful of top schools were driving the results. In no case could they find a factor other than educational connection to explain either the managers' large bets on connected stocks or the significant abnormal returns they earned.

"[S]ocial networks are important for information flow between firms and investors," they write. "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."

-- Laurent Belsie