Predictable Investment Horizons and Wealth Transfers among Mutual Fund Shareholders



    Search for more papers by this author
    • Johnson is from the University of Oregon. This article is based on the third chapter of my dissertation at Columbia University. I thank my dissertation committee for helpful guidance and constructive criticism: Charles Calomiris (my advisor), Stephen Cameron, Dwight Dennison, Lawrence Glosten, and Charles Jones. I have also benefitted from interactions with John Chalmers, Larry Dann, Richard Green (the editor), Charles Himmelberg, Laurie Hodrick, Susan Ji, Val Larsen, Megan Partch, Lenore Robbins, Joseph Stiglitz, Xiaoyan Zhang, an anonymous referee, and seminar participants at many schools during winter 2002. Special thanks go to the anonymous mutual fund family that provided proprietary data for this study and to its employees who helped in its interpretation. Financial support from the Center for the Marketing of Financial Services at Columbia University is acknowledged. This study was presented at the 2004 American Finance Association meetings in San Diego after being accepted for publication. An earlier version of this study was entitled “Pooling Externalities in Mutual Funds.”


This study analyzes the distribution of investment horizons in a large, proprietary panel of all shareholders in one no-load mutual fund family. A proportional hazards model shows that there are observable shareholder characteristics that enable the fund to predict reliably on the day each account is opened whether the account will be short term or long term. Simulations show that the liquidity costs imposed on the fund by the expected short-term shareholders are significantly greater than those imposed by the expected long-term shareholders. Combining these results, the analysis argues that mutual funds do not provide equitable liquidity-risk insurance.