The most likely explanation for this wealth-destroying behavior is that investors are unaware of the evidence. Another explanation might be that hope springs eternal, and investors do not like being “average” (meaning they confuse earning market returns with earning average returns). As Warren Buffett has said, if you choose to earn market returns, you are virtually sure to beat the great majority of investors, both individual and institutional. A third explanation may be that overconfidence is an all-too-human trait. Investors are aware of the evidence, but believe they will be among the few that can identify ex-ante the small percentage of active managers who will outperform in the future.

At least in theory, when assessing an active mutual fund manager’s skill, investors should consider all factors that explain the cross-sectional variations in fund performance. The academic literature has identified a small list of factors able to explain the vast majority of the variation in returns of diversified portfolios: market beta, size, value, momentum, profitability and investment.

Brad Barber, Xing Huang and Terrance Odean contribute to the literature on investor behavior with their study, Which Factors Matter to Investors? Evidence from Mutual Fund Flows, which appears in the October 2016 issue of The Review of Financial Studies.

The authors investigated whether investors tend to consider commonly used equity factors when assessing fund managers. In other words, do investors attempting to identify a skilled active manager strip out returns that can be traced to a fund’s exposure to investment factors known to explain cross-sectional equity returns? Fund flows should only respond to alpha, and not what is simply beta (loading on, or exposure to, a factor).

Their sample period covered the period from 1996 to 2011 and included about 4,000 equity funds. The following is a summary of the authors’ findings:

• The single-factor capital asset pricing model, with market beta as its sole explanatory factor, does the best job of predicting fund-flow relations. This result implies that investors tend to consider mutual funds’ market risk when evaluating performance, but tend to ignore other factor-related determinants of fund flows.

• Returns related to a mutual fund’s market risk positively affect fund flows.

• Investors do not completely ignore other factors that affect fund performance. However, they place less emphasis on the size and value factors than they do on market risk, and they found no evidence that investors pay attention to the momentum factor.

• Investors who buy mutual funds from the broker-sold channel respond more to factor-related returns than investors in the direct-sold channel. Such investors are attributing returns to fund managers’ skill rather than to the factors that are responsible for the returns. These results are consistent with the notion that investors in the broker-sold channel are less sophisticated in their assessment of fund performance than investors in the direct-sold channel.

• Investors buying in the direct-sold channel, as well as wealthier investors (more sophisticated investors), use more sophisticated models to assess fund manager skill, taking into account a fund’s exposure to factors (such as size and value) rather than attributing the excess returns to manager skill.

Barber, Huang and Odean concluded: “Our empirical analysis has revealed that investors behave as if they are concerned about market risk, but are largely unaware of other factors that drive equity returns. We have found some evidence that investors attend to the value, size, and industry tilts of a fund when assessing managerial skill, but these effects are much weaker than those we observed for a fund’s beta. Moreover, we have found that investors strongly respond to the factor-related return associated with a fund’s Morningstar-style category. Since the category-level return is not under the control of the manager, this result suggests some mutual fund investors confuse a fund’s category-level performance and manager skill.”

They also found that “the flows of investors who are likely more sophisticated – direct-sold fund investors, investors trading during low-sentiment periods, and wealthier investors – are generally less responsive to factor-related returns, suggesting that they are more aware that those returns are not indicative of the skills of the fund manager.”

Finally, the authors noted that to “adjust for factor-related returns when evaluating a fund, an investor needs to know the factor return. Sophisticated investors will seek out this information. But less sophisticated investors may not be aware of size, value, momentum, or industry returns. The market’s performance, however, is ubiquitously reported. This may be one reason why investors do pay attention to market risk when evaluating mutual fund managers.”

In our book, The Incredible Shrinking Alpha, my co-author, Andrew Berkin, and I present evidence demonstrating that over the past 20 years playing the game of active management has become more and more of a losing proposition, with the percentage of active funds generating statistically significant alpha falling from about 20 percent to just 2 percent. And that’s before considering the impact of taxes (for taxable investors, taxes are typically the largest expense, even greater than the fund’s expense ratio or trading costs).

Clearly, choosing to invest in actively managed mutual funds is playing a loser’s game; it’s one that’s possible to win, but the odds of doing so are so poor that it’s imprudent to try. If you are going to try and play that game anyway (by investing in managers that you believe have outperformed in the past, defying the evidence and the SEC’s warning that past outperformance isn’t predictive of future outperformance) at the very least you should know if the fund has been generating true risk-adjusted alpha, or whether its market-beating returns are simply a result of its exposure to other factors (such as size and value) that can be obtained more cheaply with passively managed funds.

Fortunately, today investors can determine the risk-adjusted alpha of any fund, whether active or passive, simply by using the multi-factor regression tool available for free on Portfolio Visualizer.