Survivorship and backfill bias in hedge fund returns have been written about extensively. A recent article in Hedge Fund Alert drew out attention to yet another problem with the reporting of hedge fund returns. It turns out that last year’s carnage has left so many hedge funds underwater that the returns posted for this year are not actually what you will earn if you are a new investor.
Hedge funds have high water marks so that managers do not receive the 20% performance fee until the fund has reached its prior high. That is one of the reasons why so many managers simply shut down their funds and launch new ones not subject to that constraint. But it also leads to difficulties with the reporting of returns. Performance fees are assessed not at the fund level but on each investor’s capital account. In other words, if you invest on December 30 you do not have to pay the performance fee for the whole year.
And that’s where the bias in return reporting comes in. Returns are shown for an investor who has been in the fund since its beginning. After last year’s carnage most hedge funds are still underwater and therefore do not charge performance fees this year to their investors who have been with them for some time. A new investor, however, who entered this year will have to pay performance fees for this year.
This now leads to the absurd situation that hedge funds generate two different after-fee returns this year, a higher one for longer investors and a lower performance for newer investors. The number reported to the major databases is, of course, the better one.
The problem is not just theoretical. Hedge Fund Alert reports that 77% of all hedge funds tracked by hedgefund.net were underwater at the end of August. As a result most hedge fund indices report inflated return numbers for this year.
Startup hedge funds are likely to suffer as a result of the bias. Typically they are above water and report their performance for this year net of performance fees. Therefore, they appear to do worse than more established managers even though the discrepancy has nothing to do with actual performance but is only due to the skew in reporting. Many investors, including supposedly sophisticated institutions, do not understand this subtlety and are likely to make decisions based on data that compares apples to oranges.
To make matters worse, the performance skew that results from the high watermark and performance fees is not the only issue. The good old survivorship and backfill biases also contribute to an overstatement of returns this year. Through August, HedgeFund.net reports an average performance of 13.7% for hedge funds, whereas funds of hedge funds returned only 6.7%. The difference can not be explained by fee layering but is an indication that many poorly performing hedge funds have simply stopped reporting their numbers to the databases. Fund of funds are still invested in them and therefore provide a better guide to the real performance of hedge funds as an asset class than the indices.
All this proves what we have been thinking for some time: data is nice to have, but its usefulness is often limited and even simple figures can have so much devil in the details that they are a clear and present danger in most investors’ hands.
Disclosure: Thomas Kirchner manages the Pennsylvania Avenue Event-Driven Fund (PAEDX), which does not charge performance fees. He is the author of the book Merger Arbitrage: How to Profit from Event-Driven Arbitrage (Wiley Finance, 2009).