There's No Good Reason to Mimic a Hedge Fund
Everyone seems to be talking about this paper by Mikhail Tupitsyn and Paul Lajbcygier of Monash University in Australia. The paper is about hedge funds. Gawker writer Hamilton Nolan and Vox writer Matt Yglesias both say that the upshot of the paper is that hedge funds are a scam.
What is this paper about? It's about whether hedge funds, as an asset class, can be replicated with a simple passive strategy. To make an analogy, suppose your money manager put all of your money into the Vanguard 500 Index Fund (which tracks the S&P 500), and then charged you a yearly fee for doing so. You could fire your money manager, buy the Vanguard 500 yourself, and keep the fee that you otherwise would have paid. The Tupitsyn and Lajbcygier paper is about whether you can do the same thing with hedge funds, cutting out the hedge-fund middleman and saving yourself some big fat fees. Hedge fund fees may have come down from the days of "2 and 20," but they are still very high compared with most of the finance industry, so investing in some kind of passive strategy that imitated hedge funds would potentially save investors a lot of money.
This isn't the first time someone has asked that question. Legendary financial economist Andrew Lo published a paper in 2007 with Jasmina Hasanhodzic that concluded that you could get pretty close to hedge fund returns with a passive strategy, but not quite. Lo and Hasanhodzic's strategy looked only at linear models. A team of economists at EDHEC in France looked at nonlinear replication strategies in 2009, and found that these models actually do worse than the simple ones. That's very close to the finding of the Tupitsyn and Lajbcygier paper that everyone is talking about.
So it turns out that you can get pretty close to hedge fund returns with a passive strategy -- as long as you have cheap plentiful access to exotic options like "lookback straddles" that are required to construct the strategy. But the real question, as I see it, isn't whether you can replicate hedge fund returns -- it's whether you would want to in the first place.
See, first of all, hedge funds aren't an actual asset class. Anyone can start one. There is no barrier to entry. So to invest in hedge funds as an asset class is exactly the same, mathematically, as throwing money at every person who stands up and says "Hey, I'm a hedge fund!" That doesn't strike me as a wise investment strategy. But somehow, even though finance writers keep repeating ad infinitum that hedge funds are not an actual asset class, people all over the world keep treating them as if they are one.
But there's a second reason you wouldn't want to imitate hedge funds as an asset class, even if you could -- misaligned incentives. Hedge funds get money two ways. First, they receive a management fee of as much as 2 percent of your total investment every year, no matter how well or badly they do. Second, they receive a performance fee of as much as 20 percent of the profits they earn for you. Performance fees are inherently asymmetric between gains and losses -- the percentage of gains that fund managers get to keep is typically a lot larger than the percentage of losses that they are forced to share. Although "clawback" provisions allow you to recoup some previous fees if the hedge fund's performance goes south, clawback can't completely remove the asymmetry. In general, hedge funds still have limited liability if they fail.
This provides an incentive for hedge funds to engage in a popular strategy called "picking up nickels in front of a steamroller." This means making bets that pay off a small amount most of the time, then catastrophically fail and explode once in a great while. Long-Term Capital Management, the famous super-giant hedge fund that blew up in 1998, used strategies like this.
Nickel-and-steamroller strategies attract lots of investors because of their highly consistent returns -- it really looks from the outside like they have some secret sauce (it also enhances Sharpe ratios, which are commonly used to measure performance). Every year they turn in consistent wins. They look like they've beaten the system. So money flows into these funds, and then when the black swan hits -- when Russian bonds default, or when there's a financial crisis, or when the tech bubble bursts -- the whole thing goes to zero overnight. Poof, there goes all your money!
Since hedge funds are so opaque -- investors never really know what any fund is doing, otherwise they could just do it themselves -- they often have little choice but to use nickel-and-steamroller strategies. They may even do it without knowing it -- their first realization that they haven't found an unbeatable strategy comes when the strategy blows up. But it's pretty clear that there is lots of nickel-and-steamroller going around. A 2004 paper by Vikas Agarwal and Narayan Naik found that a large proportion of hedge-fund strategies end up looking like this.
So why would you want to replicate the average return of an industry with this kind of fundamental problem? I'm not saying you can't pick a winning hedge fund -- maybe you can. And some funds, such as Renaissance Technologies, have certainly made their investors very rich. But because of fee problems, as well as the fact that hedge funds aren't a coherent asset class, investing in the average hedge fund sounds like a bad idea to me.
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
To contact the author on this story:
Noah Smith at firstname.lastname@example.org
To contact the editor on this story:
James Greiff at email@example.com