Which Mutual Fund Managers Exploit Investors and Which Investors Get Exploited?

Abstract

The purpose of this paper is to relate mutual fund expenses and other costs to the characteristics of both fund managers and fund investors. Specifically, we test whether some types of fund managers are more likely to manipulate fees, and some types of fund investors are more susceptible to such exploitation. We use a variety of fund manager characteristics ranging from demographics to fund style and performance. We measure investor behavior both implicitly (with mutual fund inflows and outflows, and with ownership of specially-themed funds) and, for a reduced sample period, with individual investor characteristics including behavioral biases following the empirical behavioral finance literature. The specific innovations in the paper include incorporating mutual fund manager characteristics, in addition to mutual fund investor characteristics as we have previously used (Bailey, Kumar, and Ng, 2010), to understand the forces behind mutual fund costs. We envision benefits to practitioners and policy-makers. We explicitly extend the burgeoning empirical literature on behavioral finance to mutual fund manager characteristics, the interactions of those characteristics with investor characteristics, and the implications for the structure of mutual fund fees and other costs. Practitioners ranging from mutual fund companies that hire fund managers to retail investors who rely on them will have a greater understanding of how individual fund manager characteristics affect fees and, ultimately, performance. Policy-makers will have more evidence with which to ponder changes in the way that funds are regulated and, in particular, how they can disclosure fees to maximize value for even the most unsophisticated or inattentive retail investors. We will employ several data bases, ranging from the well-known CRSP mutual fund and Morningstar sources, the record of 1991 to 1996 individual brokerage records used by Bailey, Kumar, and Ng (2010) and many previous authors, to more novel databases of political donations and voting patterns by zip code.