Do hedge funds provide a free lunch?
Try asking this question of a friend who has some idea about hedge funds and free luncheons: hardly anybody will respond with a stunning ‘yes’. “There Ain’t No Such Thing As A Free Lunch”, or TANSTAAFL, as the famous saying goes. The concept was first used in Robert Heinlein’s fiction ‘The Moon is a Harsh Mistress’ in 1969. In the novel, there is always a price to pay for anything worthwhile, including the air you breathe: well, actually the book is about self-reliant people casting off their repressive masters on Earth and establishing themselves as a free society… on the Moon.
In economics and finance, the absence of free lunch is deeply rooted among the many believers that there are no ways of increasing returns without having to accept higher risk. The idea that returns go hand-in-hand with risk was hammered out by Markowitz in 1952 when he developed the concept of efficient frontier in what is still referred to as Modern Portfolio Theory. If the ‘no free lunch’ story is to hold true with plain vanilla traditional investments, how could it be otherwise with hedge funds, given that they make use of potentially risky tools such as leverage and derivatives?
Rather than fighting what seems to be ‘prima facie’ a lost war, ask your friend the following question: do you believe that adding a poorly correlated asset to a set of traditional investments helps shift the efficient frontier to the north-west, in other words a direction where you both increase performance and reduce risk? Hardly anybody would wish to refute such evidence, as it has solid mathematical underpinnings. To show this, start building an efficient frontier linking all possible combinations of two traditional asset classes, say equities (MSCI World in the chart) and bonds (Lehman Global Bond Composite Index). Using the period 1996 to date, the set of optimal combinations is depicted by the blue line. Then bring in hedge funds as an additional asset: using the CSFB Tremont Global Index, you will get a new frontier (in red) that lies to the left of the previous one, thus embodying a set of new combinations all of which have higher risk-adjusted returns. Indeed, the magnitude of the shift has a direct link to the degree of correlation between the additional asset class and the ones it is being added to: the poorer the correlation, the greater the magnitude of the shift.
The next question to ask is pretty obvious: do you believe that Hedge Funds are poorly correlated with traditional asset classes? Almost a rhetorical one: would they be labelled ‘alternative’ if they were highly correlated to traditional assets? By now, your friend should be ready to reconsider his or her reservations. Or possibly not. He might actually tell you that standard deviation may not be the best measure of hedge fund risk. One possible reason arises from the fact that short-term volatility may not affect hedge funds that typically have monthly liquidity. Thus, if a market index drops in the middle of the month and recovers thereafter, a traditional fund with daily liquidity will experience a rise in its volatility, whilst a hedge fund will typically be able to even out the problem, from one net asset value to the next. But the major problem of using standard deviations when dealing with efficient frontiers is that this measure implicitly relies on the symmetry of the returns distribution around a stable mean. And hedge funds generally fail to have a normal distribution of their returns. The reason is that hedge fund managers aim at absolute returns, namely a risk-free rate plus a premium no matter what market conditions prevail. Thus, the payoff diagram of several hedge funds’ returns will be close to replicating that of some option strategies with positive skewness: the maximum loss is capped at the initial premium paid, whilst the potential gain is virtually unlimited. The resulting convex payoff clearly appeals to private investors for it best reflects their asymmetrical tolerance to risk: high in bull markets, low in bear ones.
Let us then take for granted that the standard mean-variance approach is inadequate when dealing with alternative investments. Instead, let’s use what is thought to be one of the best measures of hedge fund risk: downside deviation. By concentrating on the sub-mean part of the distribution, this measure best accounts for the kind of volatility and outliers that most affect us as investors: that akin to negative returns. As the table shows, downside deviations speak even more in favour of hedge funds. Indeed, at 6.41%, this truncated volatility measure accounts for only 53% of the MSCI downside risk, which is lower than the 59% recorded using standard deviations. Hence, the efficient frontier shift to the north-west is even more impressive when using a measure that is more accurate to account for alternative Investments risk.
The icing on the cake
Your friend may still be reluctant to accept the idea that hedge funds offer a lunch for free. He could point to the existence of survivorship bias, which arises from the fact that the performance of indices such as the CSFB Tremont may be over-stated because their universe excludes funds which go out of business, typically for performance reasons. Depending on the source, this bias is estimated at around 1% to 2% per annum. OK then: let us take 2% from the annual 11.5% return which hedge funds have delivered since 1996. Comparing the resulting number to the performance of both bonds and equities does not dent an inch the supremacy of alternative investments. Your friend still mumbles that the whole story seems too good to be true? Try this killer question: would you say that fund-of-funds may also be subject to survivorship bias? The answer is no: there is no way you may adjust an NAV backwards if you happen to have one fund that goes into disarray. As UBP’s flagship fund of funds investing in long-short and macro strategies, ‘Dinvest Total Return’ does indeed provide the icing on the hedge fund cake, beating after fees, both traditional assets and the Tremont index. And tell your friend he can get it for free. Well... almost.