Why do investors buy actively managed mutual funds even though the funds, on average, underperform indexes? Gruber suggests that management ability may not be incorporated in the prices of mutual funds. He provides empirical evidence that supports this hypothesis, and he speculates that if this hypothesis is true, then fund performance and cash flows in and out of the funds should be predictable and the return on new cash flows should be better than the average for all investors in these funds.
Although actively managed mutual funds have grown rapidly (both in sheer numbers and in assets under management), their average performance has been inferior to that of index funds. In Gruber's sample, mutual funds underperformed their appropriate indexes by about 65 basis points (bps) a year. If index funds provide many of the same services to investors as actively managed funds but with higher average returns, the puzzle is why investors buy actively managed funds. A possible explanation is that because open-end mutual funds trade at net asset value, management ability may not be priced in those funds. Hence, a well-managed fund sells for the same price as one with inferior management. In contrast, closed-end funds appear to include the value of management. That is, the author finds that closed-end funds have systematic and nonsystematic risks that are different from those of the underlying assets.
If management ability exists and is not included in the price of open-end funds, then the performance of the funds should be predictable. Indeed, Gruber shows that simple past returns and historical alphas from factor models do a good job of forecasting future returns. Furthermore, superior performance is not reflected in high management fees. Fees for top-performing funds are about average and below those of bottom-performing funds. Expenses for superior funds also go up more slowly over time.
If investors are aware that performance is, to some extent, predictable, then the same predictors of fund performance should also predict cash flows in and out of funds. The author finds that the association between those variables that predict performance and subsequent cash flow is strong and as expected. At least some investors appear to act rationally in allocating money among mutual funds. Furthermore, by following these measures of performance and moving their cash in and out of mutual funds each year, investors did enhance their performance during the sample period. The negative risk-adjusted return saved on the cash disinvested and the positive return earned on the reinvestment of those funds averaged 99 bps a year.
The empirical evidence is consistent with the author's hypothesis. The prices of open-end active funds, which sell for net asset value, do not incorporate management ability, and hence, sophisticated investors can enhance performance by following the predictors of future performance. Why then does money remain in funds that are predicted to do poorly? Gruber conjectures that a disadvantaged clientele composed of unsophisticated, restricted, or tax-disadvantaged investors continues to remain invested in these poorly performing funds.