It's the dog days of summer, when college professors are supposed to be doing nothing but mimeographing their syllabus and mending the elbow patches on their blazers (or whatever it is that they do in the summer.)
But that's not stopping some academics from throwing shade, as the kids say, in the general direction of the hedge fund industry.
Their study shows that due to inherent biases in the way hedge-fund databases compile results, the industry's returns have been about half as strong as they appear. The average annualized return for the industry since 1996 goes from 12.6 percent to 6.3 percent when the biases are removed from the data, according to the paper.
The volatility of the funds increases along with their maximum drawdowns. Also, measures of how the returns data are distributed, known as kurtosis and skewness, also change noticeably.
(The researchers -- Mila Getmansky of the University of Massachusetts Amherst, Andrew Lo of the Massachusetts Institute of Technology and Peter Lee of Lo's firm AlphaSimplex Group -- used data from Lipper TASS, though the same biases exist in other services. Bloomberg also compiles hedge-fund results and maintains indexes of returns for various strategies.)
The biases stem from the fact that hedge funds voluntarily report results to these databases. The main reason they cough up the data is for marketing purposes, according to the paper, and funds generally begin contributing their returns once they have results worth bragging about. And since the funds include prior returns when they first enter the database, it leads to a “backfill bias” or “instant history bias” that boosts the average returns. (Hat tip to CXO Advisory for spotting the study last week.)
Since funds can stop reporting to the database at any time -- say, for example, if returns are terrible -- this can cause an "extinction bias." In 2014, they note, the attrition rate rose to 26 percent in the database studied, suggesting that either the number of hedge funds is declining or that fewer hedge funds are choosing to report their returns.
After scrubbing the data in an attempt to remove the biases, they conclude: "The historical data show that hedge funds have not, on average, meaningfully outperformed traditional portfolios of stocks and bonds after fees. On average, once returns have been adjusted for various sampling biases, hedge funds do not routinely generate double-digit returns."
At any rate, another thing this study shows is that sometimes finance professors can have sharp elbows! No wonder so many have patches on their blazers.