eThesis - Kauppakorkeakoulun sähköiset gradut

OBJECTIVES OF THE STUDY:

In this thesis I study the relation between hedge fund alpha and hedge fund fees. Alpha refers to the unique return component, which cannot be attributed to any source of systematic risk. Therefore, alpha may be considered as a noisy signal of manager skill. Beta on the other hand, is seen as compensation for being exposed to various sources of less unique underlying risk. In that sense, the unique alpha should be the rationale for the fees. In this paper I decompose hedge fund returns into alpha and beta, and compare the alpha component to the fees. My main objective is to quantify to what extent the fees are compensation for alpha and improve the understanding of hedge fund investing. The study comprises analyses of hedge funds as an aggregate group and across different hedge fund categories.

DATA AND METHODOLOGY:

My main sample consists of 1491 hedge funds from the TASS Lipper hedge fund database. I use monthly net and gross returns from January 1994 to August 2013 where the gross returns are derived using the reported net returns and fee levels. Gross returns are essential for two reasons. Firstly, the difference between gross and net returns effectively represents the fees. Secondly, gross returns enable a more accurate analysis of hedge fund performance and decomposition of returns into alpha and beta. Alpha is derived econometrically through linear regression analysis where the gross return is regressed against selected risk factors. The regression analysis is done over the entire period and in 24-month rolling windows. In addition, I conduct a robustness check with 11 hedge fund indices.

FINDINGS OF THE STUDY:

Investors in 41% of hedge funds do not get alpha on average. Alpha is also time-dependent and there are extended periods during which hedge funds do not produce alpha for their investors. In general, hedge funds produce relatively more alpha when the economic environment is turbulent, however alpha falls during strong equity market performance. Across hedge fund categories the average share of investors who do not get alpha varies significantly between 17% and 58%. Moreover, the results indicate that also funds within the same categories differ considerably from each other. Finally, redoing the analysis with hedge fund indices suggests that alpha exceeds the fees even more rarely. In total, there seem to be many skilled hedge fund managers, but the skill to produce alpha is not universal nor is it uniform.