TITLE: LEARNING ABOUT MUTUAL FUND MANAGERS
Authors: DARWIN CHOI, BIGE KAHRAMAN, and ABHIROOP MUKHERJEE
Publication: THE JOURNAL OF FINANCE • VOL. LXXI, NO. 6 • DECEMBER 2016
#Research Question: Are investors rational in their choice of mutual funds and their managers? In a rigorously conducted study, the authors examine this question by analyzing data obtained from managers who manage two funds within a fund family. This unique data set provides a setting that permits substantial empirical controls over competing explanations, extraneous and omitted variables.
#Academic Insight: In spite of previous research indicating the opposite, this paper finds that investors are surprisingly sophisticated about their choice of mutual fund managers. For investors choosing two funds from a single manager’s offerings, the authors conclude:
(1) Investors judge managerial ability to outperform from the past returns of each fund. If one fund performs well, the other fund experiences a significant increase in fund flows. The effect is more pronounced if performance is exceptional. This result stands even when controlling for same fund family membership and the presence of a “star” fund, thus accounting for any spillover effects to other funds in the family.
(2) However, if one fund performs poorly, investors do not withdraw enough from the second fund to avoid losses. Overall, investors direct their cash infusions and withdrawals accurately, they fail to direct a sufficient response in terms of the magnitude of the directed cashflow. Further, the response is mitigated when the two funds are more similar in style, when the manager has a longer tenure, and when the underperforming fund has more volatile returns.
#Application: Even though investors recognize and direct their individual cash flows in a manner that is consistent with fund performance– both positive and negative, it turns out that the response is not always sufficient on the negative side to preclude losses. This surprising result is contrary to previously published literature about investor learning about mutual funds.
However, it does have implications for understanding why some investors continue to invest in funds that perform poorly. In terms of how investors learn about managerial skill, and in the context of the asymmetric response noted, there are at least two frictions or impediments to learning, in operation here. First, positive performing funds may have overreaching spillover effects that mask the performance of poorly performing funds, if they are the more visible fund.
As fund families increase the visibility of well-performing funds, which overshadows information on underperforming funds, the acquisition of information about the later becomes problematic. It may be very costly to move capital from one fund to another if significant front-end and back-end loads apply and if the time and effort to obtain information about performance are significant. Investors may then rely on their prior beliefs when the second fund underperforms.
Compounding this effect, investors may selectively react to positive signals from their chosen managers in order to justify their investment choices. Researchers label this phenomenon, confirmation bias, whereby investors tend to favor confirming information through biased interpretation of data and a biased search for data.