Determinants and implications of fee changes in the hedge fund industry
Authors: V. Agarwal and S. Ray
Date: March 2011
According to the authors, fees are usually considered in the hedge fund literature as fixed fees once an investment is made. However, by using historical data on changes in management fees, incentive fees, and high water mark, the authors find that about 8% of the funds that they study exhibit at least one change in their fee structure. Then the paper examines two issues related to changes in hedge fund fees: first, the determinants of fee changes, second, the effects of fee changes on future performance and capital flows from investors. The database is provided by Lipper TASS and comes from April 2008 to November 2010. Fee changes are studied at a monthly frequency. The sample contains 3770 funds.
A first range of statistics is calculated. 292 funds have one change in the fee structure on the period, 11 funds have 2 changes, and 1 fund has 3 changes. About 92% of funds display no change in the fee structure. Focusing on the cases of changes in management fees, 58.1% correspond to an increase from zero to a positive number, while focusing on incentive fees, 92.7% correspond to an increase. To determine the factors of fee changes, a regression is made on the following factors: initial incentive fee, initial management fee, inception year, size, redemption notice period, lockup period, and payout period. There is a negative and significant relationship between fee changes and both initial incentive fee and initial management fee. Moreover, fee changes are positively and significantly related to the inception year, and negatively and significantly related to the fund size. Authors’ interpretation is that “smaller funds that initially start with lower incentive fee and management fee tend to increase their fees as their managers learn about their abilities.” A second regression is made on time-varying operational characteristics of funds. The time-varying variables are as follows: annual returns defined as the net returns over the last 12 months, the total risk defined as the standard deviation of monthly returns over the last 12 months, and the net inflows over the last 12 months expressed as a percentage of AUM at the beginning of the 12-month period. Other variables are the size of the fund, time that increases by 1 every month, and aggregate inflows expressed as the percentage of total AUM for all funds in the sample. There is a positive relationship between fee changes and past performance. According to the authors, it shows that “fee changes (are) motivated by adjustment of incentives and expropriation of surplus by the hedge fund manager.” The relationship between fee changes and time is positive. By replacing raw returns by risk-adjusted returns in the regression, the positive relationship between fee changes and past performance is confirmed. In additional regressions, the same factors are used, but fee changes are replaced successively by fee increases and fee decreases, and returns are replaced by excess returns over funds following the same investment style. Results exhibit that only fee increases are significantly related to past performance. Fee increases follows superior past performance. Concerning fee decreases, the authors observe that the low number of observations can explain that the absence of relation between fee decreases and past performance.
The second part of the paper deals about the impact of fee changes on future performance, and fund flows. First, it appears that there is a statistically significant negative relation between fee changes on future performance. The authors observe that “fee changes, mostly in form of fee increases, are associated with worse future performance, suggesting fund managers strategically increasing their fees after good performance only to expropriate the surplus and deliver worse performance in the future.” Second, the relation between fee changes and future fund flows is examined. It appears that increases in inflows follow fee increases. The authors suggest that “funds publicize impending fee increases that may already be reflecting in the Lipper TASS database and give investors “one last chance” to get in under the old fee regime, leading to this consistent increase in inflows.”