Title: Do Mutual Fund Managers Take More Risk toward Yearend
Abstract: INTRODUCTION Most mutual fund managers receive compensation in proportion of the assets under their management. Unlike hedge funds or pension funds, the mutual fund industry seldom uses rewards based on investment performance (Elton et al., 2003; Golec, 2003). According to Moody's Manual on Bank and Financial Companies (1996), among 2351 actively managed equity mutual funds, 2190 used asset-based management fees whereas only 39 used performance fees. Thus, mutual fund managers have incentive to attract more new investment inflows because larger assets under their management increase their income. Researchers have found that mutual funds earning the highest returns receive greater increase in new capital (Ippolito, 1992; Sirri & Tufano, 1998). Sirri and Tufano (1998) also find that funds performing poorly do not have a large outflow of capital. This asymmetry between reward and penalty may provide mutual fund managers with an incentive to take greater risk for their funds in order to maximize their own benefits. Since mutual fund managers' behavior is not directly observable and may conflict with the goal of investors, such excessive risk taking aggravates the agency problem between mutual fund companies and investors. How fund managers adapt their investment behavior to the incentives has been of interest to researchers. One theory views the mutual funds as in a tournament in which the funds outperforming the peers win and receive higher rewards. Brown et al. (1996) argues that investors respond to the annual rankings of funds published by business magazines and services at the end of the calendar year. To win the annual tournament, managers with poor relative year-to-date returns tend to change the risk of their funds before the end of the year. Why would the managers gamble? Grinblatt and Titman (1989) argues that the long-run income of the fund manager is a convex function of his current performance. The gains from outperforming the peers exceed the losses from performing poorly, especially for new managers with smaller assets under management. In this study we apply Arrow's (1965) theory of risk aversion and utility values to examine the fund managers' risk-taking behavior. Arrow (1965) advances the hypothesis that the absolute risk aversion is a decreasing function of wealth. He argues that a risky asset is a normal good and the willingness to engage in small bets of fixed size increases with wealth, in the sense that the odds demanded diminish. In other words, risk aversion is decreasing with wealth. This implies that funds with higher year-to-date returns may have stronger incentives to increase risk at the end of the year. The theory of decreasing risk aversion offers the opposite prediction about the relation between risk adjustment and interim returns. The empirical evidence of managers' risk-changing behavior is mixed in the past studies. Using monthly returns of growth-oriented mutual funds from 1976 to 1991, Brown et al. (1996) find support for the tournament hypothesis that mid-year losers tend to increase fund volatility in the latter part of a year more than mid-year winners. Similarly, Koski and Pontiff (1999) find a negative relationship between a fund's performance in the first half of a year and its change in risk in the second half of the year using monthly data. However, Busse (2001) does not find the same behavioral pattern using daily return data. Chevalier and Ellison (1997) finds that funds that perform best have the strongest incentive to gamble, using monthly data of a sample of mutual funds from 1982 to 1992. The goal of this paper is to provide new evidence of fund managers' risk-adjusting behavior. We test the hypotheses of tournament theory and Arrow's utility theory using monthly returns of 438 growth-oriented mutual funds over the period of 1990-2000. We analyze the relation between interim performance and fund risk changes using both contingency tables and regression analysis. …
Publication Year: 2011
Publication Date: 2011-01-01
Language: en
Type: article
Access and Citation
Cited By Count: 1
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot