Published indexes generally show hedge funds beating out mainstream mutual funds in recent years, but a prominent scholar says the hedge fund data can't really be trusted.

Part of the problem, says Princeton University economist Burton G. Malkiel, is that the sampling of funds used for hedge fund indexes are skewed in favor of positive results.

"The reporting is purely voluntary," Malkiel said during a presentation at the annual national meeting of the National Association of Personal Financial Advisors (NAPFA), held last week in Toronto.
The voluntary nature of the reporting, by itself, tends to inflate performance, he says.

"If it's voluntary, when do you report? When things are good," says Malkiel, author of "A Random Walk Down Wall Street," which is now in its 8th edition.
That's not the only problem. Malkiel, who is in the midst of a major study of hedge fund performance, says published hedge fund returns are biased in at least two other ways.

The "end-of-life" bias, as he calls it, exists because investors rarely see the bad returns generated by hedge funds that are near collapse. There is also a survivorship bias, whereby published results don't include the returns of hedge funds that don't exist anymore.

He notes that of the 604 hedge funds that existed in 1996, only 141 exist today. The other 463 died along the way, and their returns are not typically included in hedge fund indexes.

"Hedge fund databases only report on hedge funds that exist today," he says.
Malkiel says that his study has shown that if you included dead funds in hedge fund reporting, their performance would be about 850 basis points lower.

"The data is severely biased upwards," he says.
Malkiel says the new fund of hedge funds products don't help matters. So far, he says, his study shows that all these funds do is charge multiple layers of fees without mitigating risk. "You're paying fees on top of fees," he says. "They're a total disaster."