View Full Version : New Academic Study: Stunning revelation about hedge-fund manager behaviour

10-09-07, 03:41 AM
Another example of why I prefer the FT. Are the results of this "academic study" a surprise to anybody? :rolleyes:

From the WSJ on Oct 9 2007...
Pricing Tactics Of Hedge Funds Under Spotlight

By David Reilly and Gregory Zuckerman
New academic research suggests that some hedge-fund managers may cherry-pick flattering prices when valuing securities that don't actively trade in an effort to improve the performance of their funds.

Investors should take heed because this massaging can help make the difference between a winning or losing month, the research found. For hedge-fund managers, such statistics on the number of winning and losing months have grown increasingly significant as the number of hedge funds has exploded -- to more than 7,500 -- and managers vie to attract and retain investor capital.

"Hedge-fund managers purposefully avoid reporting losses by marking up the value of their portfolios," according to the authors of the study, Nicolas P.B. Bollen, an associate finance professor at Vanderbilt University, and Veronika K. Pool, an assistant finance professor at Indiana University. If that is the case, the authors wrote, investors may "underestimate the potential for losses in the future and may overestimate the ability of hedge-fund managers."

10-09-07, 10:15 AM
The original paper can be found here (http://papers.ssrn.com/sol3/papers.cfm?abstract_id=937990)

Conditional Return Smoothing in the Hedge Fund Industry

Vanderbilt University - Owen Graduate School of Management
Indiana University Bloomington - Department of Finance
Journal of Financial and Quantitative Analysis (JFQA), Forthcoming

We show that if true returns are independently distributed, and a manager fully reports gains but delays reporting losses, then reported returns will feature conditional serial correlation. We use conditional serial correlation as a measure of conditional return smoothing. We estimate conditional serial correlation in a large sample of hedge funds. We find that the probability of observing conditional serial correlation is related to the volatility and magnitude of investor cash flows, consistent with conditional return smoothing in response to the risk of capital flight. We also present evidence that conditional serial correlation is a leading indicator of fraud.

Keywords: Hedge funds, fraud

JEL Classifications: G23, G28