New Fama-French Study Puts Managers Under Microscope

April 12, 2009

Besides the latest Fama-French research, a trio of profs compare performances of managers against their most appropriate benchmarks.


A pair of recent studies on the performance of actively managed mutual funds have confirmed many earlier studies showing that indexing is a better strategy for most investors. 

On March 9, research originally published in 2007 by Eugene Fama from the University of Chicago and Kenneth French from Dartmouth was updated. The revised study is titled "Luck versus Skill in the Cross Section of Mutual Fund Alpha Estimates."

(The paper can be downloaded free here.) 

The two academics use data from the Center for Research in Security Prices, or CRISP, to look at mutual fund returns from 1962–1996.

Using regression analysis, the researchers basically conclude that when all appropriate factors are taken into account, active fund managers as a whole have added zero additional value over market returns.

They find typical testing methods that look only at persistence in return histories have weaknesses.

"Such persistence tests have a downside. They rank funds on short-term past performance, so there may be little evidence of persistence because the allocation of funds to winner and loser portfolios is largely based on noise," the authors write.

The Fama-French study takes a different approach. "We use long histories of individual fund returns and bootstrap simulations of return histories to infer the existence of superior and inferior managers. We compare the actual cross-section of fund ? estimates to the results from 10,000 bootstrap simulations of the cross-section," the report states.

At the end, the researchers conclude that "for [active] fund investors the simulation results are disheartening."

Over the longer term, the study of net returns to investors found that a dominant cross section of fund managers lacked "skill sufficient to produce expected returns that cover the costs funds impose on investors," Fama and French noted.

They added: "Thus, if there are managers with sufficient skill to cover costs, they are hidden among the mass of managers with insufficient skill."

Mutual funds look better, the report continued, "when returns are measured gross, that is, before the costs included in expense ratios."

Better Ways To Analyze Funds

Another interesting study of active mutual fund managers was published in the second half of last year by the team of: Laurent Barras with the Imperial College London; O. Scaillet with the University of Geneva; and Russ Wermers with the University of Maryland. 

The title of this study is "False Discoveries in Mutual Fund Performance: Measuring Luck in Estimated Alphas."

(This study can be downloaded for free here.)

Alpha, of course, refers to the amount of return a fund manager provides above an appropriately compared benchmark's return. Alpha adjusts for the amount of risk the manager took to score any additional performance gain. 

What makes this study especially relevant is that it examines managers' performance relative to the risks they take. Also, the study takes into account the probability that a certain number of funds outperform the market by pure chance. When evaluating for performance, of course, investors need to account for the risk managers take by investing in more or less risky areas of the market. 

For example, small company stocks tend to be more volatile and risky. As a result, it is reasonable to expect that funds focused on small-cap stocks tend to provide greater returns to investors who have longer time horizons. 

Measuring The Right Benchmarks

Experienced investors realize, therefore, that it's not appropriate to attribute managers' skill to beat the S&P 500 by simply tilting their portfolios to small company stocks. 

In addition to adjusting returns for style and other return factors when evaluating fund manager performance, the researchers in this study also adjust results to account for the fact that we should expect a certain number of market-beating managers to be present by sheer luck or chance alone. 

That's a fascinating approach and one well worth considering. With these elements included in the mix, the authors came to some quite startling conclusions. 

For the period of 1975 to 2006, they found that:

  • Some 75.4% of fund managers added zero alpha—net of their expenses.  
  • Another 24% of the managers had a negative alpha—actually taking away returns from investors.
  • Less than 1% (0.6%) of fund managers added alpha (i.e., showed skill at adding value above their most appropriate benchmarks). 

The other interesting conclusion from this study is that as time progresses, management skill in stock picking tends to get lower and lower.

In fact, the study found that virtually no skilled managers were able to consistently produce significant alpha by the final year of the study.

Kenneth Smith is chief executive of Seattle-based Empirical Wealth Management. He welcomes suggestions and comments for future columns at [email protected].


Find your next ETF