Three Essays on Hedge Fund Returns, Risk and Money Flows.

摘要

The growth of the hedge fund industry has been very rapid in recent years. According to Hedge Fund Research inc., the assets under management of hedge funds grew from ;However, hedge funds did not only attract the attention from investors, but also from regulators and academics and this, for many reasons. Indeed, a hedge fund is typically defined as a pooled investment vehicle that is privately organized, administrated by professional managers and not widely available to the public. Due to their private character, hedge funds are much less regulated than mutual funds and offer limited transparency and disclosure. The recent financial crisis put them in the spotlight because governments and regulators are more than ever concerned about the impact of their trading on the global economy given their prominence in financial markets. The desire to understand their underlying risk factors and to evaluate their risk-adjusted performance led to numerous academic and practitioner studies in this area.;In the first paper, we use a Markov chain model to evaluate pure persistence in hedge fund returns. We study two forms of pure persistence:absolute persistence (positive/negative returns) and persistence with respect to the high water mark, which accounts for the size of drawdowns. In the first case, we find that hedge funds in general exhibit persistence in positive returns, but no persistence in negative returns. In contrast, the results using the high water mark criterion show the presence of both positive and negative persistence. Hedge fund managers exhibit a relatively high probability of delivering positive returns.;In the second paper, we use a binomial model to assess the impact that the level of the fund's volatility has on manager fees and on investor wealth. On a multi-period framework, our results suggest that for a given expected return, a higher level of volatility provides the manager with higher expected fees in the first period. Beyond this one, this positive relation is not always observed. However, a higher level of volatility always has a negative impact on investor wealth and this impact increases with the number of periods. This may lead to outflows when investor objectives are not accomplished and therefore may reduce the future compensation of the manager. These results are in line with the findings of Panageas and Westerfield (2009) and suggest that with an incentive contract over a time horizon superior to one year, the manager will not take excessive risk given the path-dependent nature of the payoffs. The one-period analysis also shows that the optimal volatility of the fund is related to its size, the moneyness of the option like-contract, the incentive fee and the management fee rates, the minimum net-of-fees return required by the investor and the expected return of the fund.;Finally, we explore the topics of performance, variance shifts and money fiows in hedge funds in the third paper and our results provide further insight into what has been found in previous research. First, we find that top-return funds are also the most risky funds and the presence of a considerable degree of turnover among them suggests that they are not necessarily the best choice for investors, even for strategies exhibiting a high level of relative persistence. Second, our results suggest that one reason why Brown et al. (2001) fail to find variance shifts in response to absolute performance is because they implement the analysis at the aggregate level of funds. We find the evidence of variance shifts for Macro funds and this suggests that the variance shifts in response to absolute performance may also depend on the flexibility that the manager has to alter his portfolio exposure. Finally, our results indicate that the relative performance has a more important impact on variance shifts because managers know that in the absence of a benchmark, investors evaluate them on their return, as well as their risk. Our results show a concave relationship between first-semester performance and second-semester fiows and suggest that top return funds are also penalized for their high level of risk. (Abstract shortened by UMI.).