More Cooks Improve the Broth

Three Is Company
Team portfolio management has become quite the fashion. Whereas 67% of U.S. stock funds were run by a single manager in 1992, the figure today is less than 30%. Conversely, whereas large teams were once almost unheard of, a full one fourth of such funds currently are run by four or more managers.

Those numbers come from "To Group or Not to Group? Evidence from Mutual Funds," from Saurin Patel of University of Western Ontario and Sergei Sarkissian of McGill University. Using Morningstar data, which they cite as the reason for their breakthrough discovery (apparently CRSP, the standard academic source for fund information, has spotty manager data), the authors find evidence that U.S. stock funds with multiple managers outperform those with a single manager.

Surprisingly, the benefit comes from more returns rather than less risk. One would think that having more managers brings more diversification, which in turn brings lower volatility along with potentially lower returns. However, on average, funds run by two- and three-manager teams have slightly higher standard deviations than do those run by a sole manager. Those with a single manager have the lowest returns, with the highest numbers coming from funds with three managers and five or more managers.

These are small effects. The authors estimate the four-factor alpha of three-manager funds to be 0.43% annually, or 43 basis points, above those of funds run by a single manager. The victory margin for funds headed by five managers or more is 47 basis points. Because there are a great many funds, however, those modest amounts are statistically significant at the 5% level.

Perversely, the more managers assigned to a fund, the less that the fund costs. For funds that have five or more managers, the reason is simple; those funds are much larger than other funds, so they benefit from economies of scale. (As the authors present asset size as an average rather than as a median, I suspect that this effect is driven by several massive American Funds.) Funds with two to four managers, though, are of a similar size and age, yet are modestly cheaper than single-fund managers.

In addition to finding in favor of management teams, which is new to the literature, the professors confirm several other effects that have previously been documented. All things being equal, funds fare better if they:

1)      Have a smaller asset base,
2)      Are younger,
3)      Come from a larger fund family,
4)      Have a lower expense ratio (duh),
5)      Have superior past performance,
6)      Are run by managers who have higher SAT scores.

The latter isn't quite how it reads. I wondered how the authors found managers' SAT scores, which most assuredly are not in Morningstar's database. They didn't. They instead know where managers received their undergraduate degrees, and they plugged in the typical SAT range found at those colleges as the estimate for each manager's score. Yes, that's a stretch. However, the result is consistent with other studies that have found stronger performance for funds run by managers who attended more prestigious colleges, as well as for managers who have MBA degrees.

The authors maintain that funds run from large financial centers benefit from team structures but those run in remote areas do not. The authors believe this to be a "spillover effect" that comes from interaction among investment professionals, which aids in acquiring "knowledge, skills, and establish[ing] business connections." I don't know about all that; to me, the simpler explanation is that it's easier to find several good managers in an area that has many investment professionals. In any case, though, the claim makes some sense.

The professors also note that "female-managed funds or team-managed funds with at least one female member show much less total and idiosyncratic fund volatility, irrespective of the benchmark." Female mutual fund investors take on less risk than do male investors. Apparently, the same holds true for fund investment managers, as well. (Having watched my infant son learn about gravity by clambering over the gate to his crib, then losing his grip and falling to the floor with a loud thud, again and again, I am unsurprised.)

The authors do not explain why team-management structures have become prevalent, aside from the notion that the financial markets have become more complex. That is likely so, but there are three even-stronger reasons.

First, in the early 1990s, the SEC passed a requirement that fund companies must disclose the names of their funds' portfolio managers. Yes, before then a fund company could be mum on the subject! Fund executives quickly learned what they already suspected, that the departure of a portfolio manager who had sole responsibility for a fund led to awkward questions and, in some cases, immediate redemptions. Better to avoid those problems by having a team.

Second, American Funds and Vanguard beat Fidelity. Twenty years ago, Fidelity was not only the largest stock-fund manager in the world but also a highly visible and successful proponent of the single-manager structure. Fidelity managers were expected to stamp each fund with their individual genius, as Peter Lynch had done with flagship Magellan. In contrast, American Funds and Vanguard avoided placing all their funds' apples into one cart. American Funds split duties among several internal managers, while Vanguard increasingly doled out fund assets over multiple outside managers. Fidelity lost the performance battle--first American Funds and then Vanguard became the largest U.S. fund manager--and the competition noticed.

Finally, growth in team management is consistent with the maturation of the fund industry. The great asset boom has come and gone. Now, fund companies profit mainly by retaining their existing customers. In such an environment, the additional prudence associated with team management is attractive. 

John Rekenthaler has been researching the fund industry since 1988. He is now a columnist for Morningstar.com and a member of Morningstar's investment research department. John is quick to point out that while Morningstar typically agrees with the views of the Rekenthaler Report, his views are his own.