Two thirds of hedge funds create genuine value through contrarian strategies, choosing well among cheap and beaten-down stocks, according to a new study of their holdings.

In contrast, two thirds of mutual funds are playing the momentum game, buying what’s gone up and selling what seems to be going down. These mutual funds, in contrast to contrarian hedge funds, aren’t really earning their keep.

The study, released in January, is ground-breaking, according to its authors, because it relies on 13F filings, which are holdings disclosures that money managers above a certain size are obliged to release to the U.S. Securities and Exchange Commission.

That not only allows for funds to be sorted between styles based on what they actually do, it also eliminates many of the pitfalls of hedge fund performance management statistics, which rely on voluntary reports of returns.

“Contrarian hedge funds’ success derives from their managers’ superior stock-picking skills - the ability to pick the right losers among stocks with similar characteristics,” write Mark Grinblatt of UCLA, Gergana Jostova of George Washington University, Lubomir Petrasek of the Federal Reserve Board and Alexander Philipov of George Mason University. (http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2712050)

“We find performance persistence only among outperforming hedge funds - there is no evidence of performance persistence among underperforming hedge funds and outperforming or underperforming mutual funds and other institutional investors.”

The study examined 13F filings from a 1998-2012 sample period, and included analysis of 589 unique mutual fund advisors and 1,342 hedge fund advisors.

Contrarian hedge funds outperformed a passive benchmark by 2.4 percentage points a year using returns adjusted for size, the book-to-market of stocks involved and momentum. Two thirds of that outperformance came from buying, as opposed to selling stocks.

Turning to mutual funds, two thirds follow a momentum strategy, meaning they try to surf along with the waves of stocks’ movements. These funds tend to stay with this style. Momentum mutual funds do outperform contrarian mutual funds, while contrarian hedge funds outperform all groups including momentum hedge funds.

Momentum is a well-known and much studied anomaly in financial markets but it looks as if mutual funds following it are adding little. Once you adjust their returns for the effects of momentum, a momentum index return, if you like, momentum mutual funds do about as well as their mutual fund contrarian peers.

And momentum mutual funds themselves are doing little better than if they followed a naive index-like momentum strategy, adding just 2 basis points of extra return per quarter.

Hedge funds in the study also did a fairly good job of market timing at crucial turning points, according to the findings of the study, again in contrast to mutual funds. Having been “fairly neutral” during much of the run-up in stocks in the dotcom bubble, hedge funds in the study turned strongly contrarian in the second half of 2000. Similarly in the latter stages of the housing bubble, they became far more contrarian between the end of 2006 and end-2008. Mutual funds intensified momentum trades between mid-2007 and mid-2008.

While I am skeptical about the benefits of hedge funds, this study, in using 13F filings, avoids a lot of the common errors of studies based on reported hedge fund returns, such as survivorship, backfill, and self-selection biases.

So where, you may ask, are contrarian hedge funds finding all these bargain stocks to buy?

From mutual funds of course. One possible consequence of a momentum strategy is that you will tend to be a seller of that which is going down. It's also the case that mutual funds, which have to offer daily liquidity to their investors, are thus often forced sellers at times of stress if retail holders get jumpy and want out.

While hedge funds sometimes come undone based on similar dynamics, they will tend to have redemption gates which makes them less likely to be forced to dance at the end of their investors’ strings.

“We also find that hedge funds’ contrarian trading (buying of losers) increased during 2008-2009, while mutual funds’ selling of losers intensified,” the authors write.

They also find that hedge fund outperformance still holds when taking into account transaction costs and other expenses associated with a fund. Hedge funds have higher turnover in their portfolios, costing themselves an estimated 15 basis points a quarter against seven for mutual funds.

One important factor here is that hedge funds are not required to report short sales in the same way as purchases, a factor which may either flatter or underestimate their performance.

The study adds to the weight of evidence that mutual funds are not pulling their weight and offers encouraging hard data about hedge funds.