Jonathan Davis: Hedge fund management may be more style than substance

When I wrote in this space about hedge funds a few weeks ago, I had not yet seen an interesting recent piece of academic research that throws more light on the fascinating question of whether or not, in today's market conditions, hedge funds are the answers to every investor's prayer. It seems reasonable to return to the subject, though be warned that some of the issues raised are quite complex, in a way that only academics can make them.

When I wrote in this space about hedge funds a few weeks ago, I had not yet seen an interesting recent piece of academic research that throws more light on the fascinating question of whether or not, in today's market conditions, hedge funds are the answers to every investor's prayer. It seems reasonable to return to the subject, though be warned that some of the issues raised are quite complex, in a way that only academics can make them.

The new paper is by Guarav Amin and Harry Kat, respectively a PhD student and junior professor at the University of Reading. It is based on an analysis of the performance of 77 hedge funds and 13 hedge fund indices from1990 to 2000. The 77 funds, drawn from a leading global hedge fund database, are the only ones out of 1,476 funds that can demonstrate 10 years of continuous monthly performance data.

This narrow sample is a key pointer to one of the main problems with finding out how good hedge funds actually are. The number of funds has proliferated so rapidly in the last few years that it is difficult to establish data series of sufficient length or breadth to be confident the results can be trusted. Most performance figures bandied about by promoters are riddled with survivorship bias; funds that fail are frequently dropped from the data.

The oldest indices of hedge fund performance only date back to around 1994-95, which means that most of the funds have still to be tested across a full range of market and economic conditions. As a cautionary tale, it is worth remembering that the last time hedge funds enjoyed a period of huge popularity was in the late 1960s, and most of those were wiped out, or disappeared from sight, when markets tended to crater in the crisis conditions of the early to mid-1970s.

What is not in dispute, the academics find, is that the growth of hedge funds is well founded on the basis of conventional performance measurement statistics. The case for buying hedge funds is that you are buying into a small and flexible fund managed by a highly motivated individual or team. They will typically invest in a variety of specialist areas (which often have quite distinctive risk profiles) and look to enhance their returns with the aid of gearing and other more sophisticated techniques.

The other great leg of the hedge fund case is that, whatever their particular area of expertise (be it arbitrage, investing in junk bonds, global macro plays, or simple long/short equity investing), they are all primarily set up to pursue absolute rather than relative returns. That is to say, they hold out the promise that they will make positive returns even when markets are weak and falling.

By definition, their performance will demonstrate low correlation with that of the overall stock market, which in turn means that they have added value as a risk-reducing element in investment portfolios. If you have a big chunk of equities in your portfolio, standard finance theory demonstrates that the effect of balancing this with a chunk of uncorrelated hedge fund as well must be to reduce the overall risk per unit of return you make.

At first glance, the performance of the 77 funds studied by Messrs Amin and Kat broadly supports this conventional analysis. If you use Sharpe ratios, the most widely used industry measure of risk-adjusted returns, then virtually all the hedge fund indices and many of the individual funds produced risk-adjusted returns that were superior to that of the S&P 500 index over the period 1990-2000. An even higher proportion (71 out of 77) funds could claim a positive "alpha", the conventional measure of fund manager skill.

These calculations are all made net of fees and other costs, so, apparently, the conclusion is straightforward. Hedge funds are a good bet, though there is an interesting sidelight on another issue that occupies hedge fund investors – which is whether to invest directly in a fund yourself, or whether to do so through a so-called "fund of funds", in which an intermediary invests a pool of money for a group of investors into several types of hedge fund.

On this issue, the performance data suggests that while funds of funds are effective at reducing the risk of investing in hedge funds, the net benefit to investors may be outweighed by the burden of having to pay two sets of fees – one to the intermediary for picking and managing the investments, and a second to the managers who are investing the money for the hedge funds themselves. This, it should be said, is broadly in line with the hedge fund advice given by independent fee-based advisers.

At this point, however, the academics introduce another caveat that they say may seriously undermine the seemingly convincing case for hedge funds. Their contention is that the conventional academic measures used to calibrate risk, such as Sharpe ratios and alphas, may not work, as they are based on assumptions misrepresenting the character of the underlying data.

Specifically, the measures assume that hedge fund returns, like those of the stock market itself, fall into a "normal" distribution, when this appears not to be the case. If you dismiss this assumption and test the data using another performance metric, it produces a rather different overall picture.

The academics find that most hedge funds are nothing like as good at producing given end returns to investors as they claim to be. However they do still produce a useful function as a risk-diversifier within an overall stock market portfolio, with the optimal proportion being around 10-20 per cent of the overall total, and are still way more efficient than the average unit trust.

Now it is not for me to pass judgement on the rights and wrongs of this arcane statistical debate, but the research does throw up some interesting ideas that mesh with what many professional investors privately admit about hedge funds. I accept the average hedge fund manager is more skilful and motivated than his counterpart in mainstream fund management, and that there will always be periods when absolute return disciplines will produce superior results, as they are clearly and logically doing at the moment. But on the other side has to be set the facts that: a) hedge fund managers need to be more skilful and more motivated in order to justify investors paying the much higher fees that they are asked to pay; and b) the performance data has to be treated with some caution as it overstates returns and understates the risks involved.

There is a strong suspicion that what investors are paying for with hedge funds is a style of investing rather than the investment genius that the marketing people like to claim you are getting.

davisbiz@aol.com