Academics Prof. Jonathan B. Berk from Stanford University Business School and Associate Prof. Jules H. van Binsbergen of the Wharton School at the University of Pennsylvania have undertaken a study that points out that the “search for alpha” mantra that hedge fund investors have long followed in assessing active fund managers is conceptually flawed. Their research, published as Stanford University Graduate School of Business Research Paper No. 15-37 shows that active managers do have skill in picking investments and create value in doing so.
Berk and van Binsbergen note: “Active fund managers are skilled and, on average, have used their skill to generate about $3.2 million per year. Large cross-sectional differences in skill persist for as long as ten years. Investors recognize this skill and reward it by investing more capital in funds managed by better managers. These funds earn higher aggregate fees, and a strong positive correlation exists between current compensation and future performance.”
Value added is a more accurate measure of active managers skill than alpha
In their research, Berk and van Binsbergen make it clear that “the correct measure” of managerial skill is value added, that is, the total amount of dollars the manager extracts from markets. The study shows that the average mutual fund manager adds value of greater than $3 million per year. Moreover, this value added is very predictable and superior managerial performance is likely to continue for a decade into the future.
Interestingly, the data also showed that current compensation more accurately predicts future returns than past performance: “Perhaps most surprisingly, investors are aware of this skill and reward it by compensating managers accordingly. Consequently, current compensation better predicts future performance than past performance.”
Active managers: Gross alpha does not reflect managerial skill
Keep in mind that gross alpha is equal to the alpha investors earn plus fees. However, based on theory, if capital markets are truly competitive, then investors will earn a zero alpha. This situation boils down to the gross alpha equaling the fee charged, and given managers choose their own fees, they in effect also choose their own gross alphas.
That is why Berk and van Binsbergen note: “Consequently gross alpha is not informative about skill differences.”
However, although managers get to pick their fees, they do not get to choose their size because the size of a fund changes based on the fee charged.
Therefore: “All else equal, funds that charge higher fees will be smaller because investors understand that fees impact performance. To correctly account for size, the gross alpha must be multiplied by the fund’s assets under management (AUM) to get the total dollar value added by the manager.”
Our sample consists of 5974 funds, considerably larger than comparable samples used by other researchers. There are a number of reasons for this. Moreover, the sample was not limited to funds that hold only U.S. stocks. Clearly, managerial skill, if it exists, could be used to pick non-U.S. stocks. Also, by eliminating any fund that at any point holds a single non-U.S. stock from their sample, past researchers have been eliminating managers who could have skilfully moved capital to and from the U.S. at opportune times.
Berk and van Binsbergen explain: “Expanding the sample to all equity funds is not innocuous: The statistical power of our tests is greatly increased by including all funds, and, more important, managerial skill is positively correlated to the fraction of capital held in non-U.S. stocks.”
The statistical methid used in the study involves collapsing all subfunds into a single fund by adding all the subfund AUMs together and calculating the weighted average return based on AUM.
See full PDF below.