Seeking Alpha? Less Competition Helps Beat the Market

SMITH BRAIN TRUST -- One of the most closely scrutinized questions in finance is whether any fund managers can beat the market, year after year. The evidence is far more mixed than you might think from reading stories in the business media about investment "geniuses" of various stripes. New research from the University of Maryland's Robert H. Smith School of Business adds to the evidence that some fund managers indeed can beat the market, but with a caveat: Whether they can do so depends heavily on how much competition their funds have.

To compare fund performance, N. R. Prabhala, a Smith School professor writing with Gerard Hoberg, of the University of Southern California, and Nitin Kumar, of the Indian School of Business, introduce a new concept called "customized peer alpha." ("Alpha" refers to risk-adjusted outperformance by a fund of recognized benchmarks.) The approach relies on identifying the peers of each fund, based on how similar their holdings are. This requires categorizing stocks based on such metrics as company size and book-to-market value.  The researchers examined 3,390 unique funds from 1980 through 2010.

Prabhala, Hoberg and Kumar re-analyzed peer groups quarterly, meaning that a given fund's peers were constantly shifting (as fund size or strategy changed). Peers from the new methods overlap only partly with categories that are widely known to consumers, such as those used by Lipper.  

Some funds had just a few dozen rivals while others had hundreds. Depending on the number of peers the funds had, the authors placed funds into high, medium and low competition categories.

When fund performance was examined, it turned out that a fund's past performance relative to its peers served as a predictor of its future performance. That suggests that some fund managers did, in fact, possess a measurable ability to spot valuable investments, whether they achieved this through skill, access to better information or some other means. But the more competition a fund faced, the less likely it was to maintain an advantage — over both the short and the long term.

In the high-competition space, alpha persistence was close to zero. In the low-competition category, however, the funds in the top decile of performers maintained an advantage of 185 basis points (1.85 percentage points) over those in the lowest decile, even after controlling for variables such as fund size or turnover.

"What we are saying is that if a given manager does well, the other funds are watching, and possibly copying successful strategies," Prabhala says. "Competition therefore helps determine whether the fund will continue to outperform its peers."

The paper hardly provides a road map for beating the market. It's far from clear that a fund moving from a more competitive space to a less competitive space would automatically attain better returns. For instance, to find alpha, "you need to generate ideas," Prabhala says, "and it's very hard for you to generate new ideas if the field is new to you." Other institutional constraints also limit the extent to which a fund can change a longstanding investment strategy.

An alternative theory for why it is so hard for funds to beat the market has to do with "diseconomies of scale." When investors see a fund manager is doing well, money flocks to the fund, and sheer size reduces its nimbleness. The authors found evidence that size limits alpha, but the competition effect was stronger.

Competition seemingly shaped other aspects of fund performance, including momentum. For example, a fund on a strong upward trajectory was far more likely to continue on that path if it had few rivals than if it had many. 

"Mutual Fund Competition, Managerial Skill, and Alpha Persistence," by Gerard Hoberg, Nitin Kumar and N.R. Prabhala, is a working paper.

SMITH BRAIN TRUST HOME | ABOUT | ARTICLES | HAPPENING NOW | VIDEOS | WEEKLY NEWSLETTER | SUBSCRIBE