Hedge Fund Performance: Are Stylized Facts Sensitive to Which Database One Uses?

Abstract

This paper proposes a novel database merging approach and re-examines the fundamental questions regarding hedge fund performance. Before drawing conclusions about fund performance, we form an aggregate database by exploiting all available information across and within seven commercial databases so that the widest possible data coverage is obtained and the effect of data biases is mitigated. Average performance is significantly lower but more persistent when these conclusions are inferred from the aggregate database than from some of the individual commercial databases. Although hedge funds deliver performance persistence, the average fund does not deliver significant risk-adjusted net-of-fee returns while the gross-of-fee returns remain significantly positive. Consistent with previous literature, we find a significant association between fund characteristics related to share restrictions as well as compensation structure and risk-adjusted returns.

The Performance of Hedge Fund Performance Fees

“…We can attempt to estimate the amount of incentive fees paid in excess of the nominal fee for the entire hedge fund industry. To do so, we extrapolate from our sample to the sample used by Joenväärä et al (2019), which is, to the best of our knowledge, the most exhaustive hedge fund sample used in an academic study to date. 15 In their study, the authors include funds from seven separate data providers, including the U.S.-dollar funds from the two data providers studied here, as well as non-U.S. dollar funds.…”

“…Third, we consider a terminal return of −10% applied to all delistings. The latter adjustment is motivated by the recent findings of Joenväärä et al (2019), who use funds cross-listed on multiple data vendor platforms to augment delisting returns (see Panel C of Table 3 of that paper). In short, they find that some of the funds that delist from BarclayHedge and TASS keep reporting to other data vendors for a few additional quarters.…”

“…funds holdings (Aiken, Clifford, and Ellis, 2013), and cross-listings in multiple databases (Jorion and Schwarz, 2013;Joenväärä et al, 2019), prior studies have found that fund performance after delisting is on average poor. Given the uncertainty regarding how to control for delisting bias, we follow prior literature and adjust the last reported return by cumulatively adding a terminal return.…”

“…The results presented in the book, however, have been criticized by both academics (e.g., seeSchneeweis and Kazemi, 2012) and practitioners (see Andrew D. Beer, "A Lack of Rigor in 'The Hedge Fund Mirage'," AllAboutAlpha.com, November 15 2012). A crucial criticism is that the HFRX index is not representative and has an abnormally low AUM-weighted internal rate of return due to an unusual computation method.15 The back-of-the-envelope calculations reported here are based on the pre-publication version ofJoenväärä et al (2019). At the time of this writing, it appears that the information regarding AUM that we use here has been removed from the version ofJoenväärä et al (2019) that was accepted for publication.…”

“…A crucial criticism is that the HFRX index is not representative and has an abnormally low AUM-weighted internal rate of return due to an unusual computation method.15 The back-of-the-envelope calculations reported here are based on the pre-publication version ofJoenväärä et al (2019). At the time of this writing, it appears that the information regarding AUM that we use here has been removed from the version ofJoenväärä et al (2019) that was accepted for publication.…”