| These two essays explore performance measurement issues related to hedge fund returns. Return accuracy is vital if the researcher is to understand both how hedge fords generate their performance and if their performance adds value. The first essay uses a unique panel of hedge fund debt holdings to directly measure hedge fund security pricing behavior. Using a panel regression framework, I find that the variation in bond prices assigned by different advisors increases with measures of bond distress and illiquidity. I also discover potential return smoothing behavior, as price differences between hedge funds and a third-party source increase when the bond has had poor returns. Furthermore, there is evidence that this pricing behavior is asymmetric - negative returns lead to more positive price differences, while hedge funds do not under-price following positive returns. Overall, hedge funds appear to have skill when selecting and pricing debt securities, but this skill coexists with return smoothing behavior that could bias return measurements.;The second essay examines the self-reporting bias in hedge fund data. Using holdings data from a set of limited partners, I construct a novel set of returns for hedge funds that otherwise have never reported to a commercial database. These returns allow, for the first time, a direct comparison of performance between funds that choose to report and funds that do not. Using pooled regression analysis and several different factor-models, I find evidence that estimates of managerial skill using self-reported data have an economically significant positive bias (0.45% per quarter). The nature of the data also allows me to measure the performance of funds even after they exit the databases - the so-called "dead" funds. Quarterly returns for funds that have stopped reporting are dramatically lower than returns matched to a database. |
