What Price Hedge Fund Lock-Ups? 130 Basis Points Per Year |
Date: Wednesday, December 13, 2006
Author: Martin de Sa'Pinto, Senior Financial Correspondent, Hedgeworld.com
GENEVA (HedgeWorld.com)—You want to invest in a particular hedge fund, but are concerned by its three-year lock-up. "Do they really need to tie my money up for three years," you ask yourself, "and if they do, what do I get in return?" "There is an increasing number of hedge funds asking for lock-ups, and lock-up periods are getting increasingly longer," said Emanuel Derman, addressing delegates at ICBI's "Ri$k Minds" conference, taking place this week in Geneva's President Wilson hotel. Mr. Derman is director of Columbia University's financial engineering program and head of risk at Prisma Capital Partners, a fund of hedge funds launched by former Goldman Sachs executives at the start of 2005. "But what excess return should you ask in return for allowing your money to be locked up?" he asked. One of the main effects of a lock-up, the professor noted, is that it deprives investors of their periodic option to redeem from one fund and invest in another. So the excess return investors should expect equates to the opportunity cost that is lost by renouncing this periodic option. "But is this option worth anything?" he asked, and he himself provided the answer: "Only if you can get out of a predictably bad fund into a predictably good one." All right up to here, but at this stage a number of debatable points begin to surface. While on the one hand even the most patient investor will eventually redeem from a fund with consistently sub-par performance, what is the probability that jumping ship at the first sign of trouble will prove the optimal decision? After all, investors are told often enough that past performance is no guide to future performance. Even so, said Mr. Derman, evidence suggests that there is some degree of predictability and persistence in the performance of hedge fund managers. "The hypothesis is basically that hedge fund alpha takes skill, and skill persists," he explained, offering some graphs demonstrating return patterns in various hedge fund strategies to illustrate this. The statistics he offered seemed to support the hypothesis. Classifying hedge fund managers as "good" (those who match or outperform peers using the same strategy), "sick" (those who regularly underperform the strategic benchmark) and "dead" (the 3% to 5% of funds that close each year owing to consistently substandard performance) he noted that a fund that was "good" one year had an 85% probability of being "good" the next year. Similarly, a fund that was "sick" one year had a 65% probability of remaining sick the next, and just a 35% probability of getting better. The probability of a sick fund ending up among the dead was negligible, he said … but this already begs the question, "Where do the 3% to 5% of dead funds come from?" If we put our faith in these statistics—and even if we do not—it is reasonably logical that investors in a "good" fund will tend to stay there, lock-up or no. Investors in a "dead" fund will get the residue of their money back once the fund has been liquidated and redeploy it. Here, too, the presence or otherwise of a lock-up makes no difference. So there is no optionality in either of the first two cases. The only optionality, then, lies in the case of a "sick" fund where, if an investor can redeploy to a better performing fund, he or she will do so, unless prevented from doing so by the terms of a lock-up. In this case, the effect of the lock-up "is to deprive the fund of funds manager [or any other investor] of this periodic option to redeem from one fund and invest in another," said Mr. Derman. Modeling the value of this optionality proved to be complex. Also, some of the assumptions used in the model were debatable, principal among them the assumption that performance, whether above or below the benchmark, would tend to regress toward the mean. In a nutshell, the model assumes that a fund which produces excess returns of Ü last year is likely to produce Ü/2 excess returns next year and Ü/4 excess returns the year after. Likewise, a fund's underperformance this year will be half of what it was last year, and next year it will be half of what it is this year. Crunching the numbers concerning sick funds (bearing in mind that there is no "optionality" in the case of either "good" funds or "dead" ones) and the probabilities that they will remain sick, get better, or die and factoring in the likelihood of performance regressing toward the mean, Mr. Derman has come up with a result: An investor agreeing to a three-year lock-up should expect returns of 130 basis points per year over and above what would be expected if there were no lock-up. So at the end of the day, what does it mean for investors? One logical conclusion would be for investors to require compensation for the opportunity costs of agreeing to a lock-up. For example, investors in funds wishing to apply a three-year lock-up might expect the fund's high water mark to be raised by 130 basis points with respect to funds using the same strategy but with no lock-up. One delegate pointed out that in general, emerging managers from smaller funds tend to perform better than established managers at larger funds, an observation that, Mr. Derman agreed, seemed to be borne out by the statistics. On the other hand, ceteris paribus, emerging managers were less likely to look for a lock-up—and generally had less chance of investors agreeing to one—than well-established managers. It therefore stands to reason, statistically, that funds with lock-ups should tend to underperform. "You can argue that if they're so good, they don't need lock-ups," Mr. Derman offered as a parting shot.
Reproduction in whole or in part without permission is prohibited.
