Standard, widely-used measures of mutual fund performance are inaccurate and unreliable, and can lead to faulty conclusions by investors, a new study shows. The study, conducted by Professors S.P. Kothari and Jerold B. Warner of the William E. Simon Graduate School of Business Administration and funded by the Association for Investment Management and Research, has important implications not only for fund managers who claim they can outperform the market, but also for ratings services such as Morningstar which track mutual fund performance.

The study, Evaluating Mutual Fund Performance, empirically analyzes the measures used to evaluate mutual fund performance. This research examines commonly used measures of performance like risk-reward ratios, market timing, and benchmarking by investment styles. Kothari and Warner find that it is easy to erroneously conclude that a mutual fund has abnormal (superior or inferior) performance of more than three percent per year when in fact no abnormal performance exists.

The study demonstrates that investors should have less faith in the ratings services because performance measurement techniques are crude. Since current methods that measure and adjust for risk are flawed, consumers may be investing in funds that actually take on more risk than is apparent. Moreover, drawing inferences from performance data about stock-picking ability of fund managers is not necessarily accurate.