| Hedge Funds: An Analytic Perspective |
| Andrew W Lo, Princeton University Press, 2008, 364 pages, £26.95. |
| Professor Andrew Lo is a major figure in finance so his new book on the fast-moving world of hedge funds ought to be in the ‘must read’ category. And so it is. However, this is not a stand-alone textbook but a reformation of previously published papers. There has been editing, updating and some re-writing but, in this writer’s opinion, not enough. There is still repetition and unevenness. The book is the authoritative distillation into an accessible form of a huge amount of academic research and practical experience that the reader would expect. Yet Chapter 3 makes a brief excursion into another world of lemmas and propositions. |
| Professor Lo gives a masterful illustration of the problems in gauging hedge fund performance with his famous fantasy fund Capital Decimation Partners. By selling out-of-the-money puts on the S&P, this strategy acquired a highly attractive track record over the period 1992-1999 despite requiring zero skill. The book does not include an update of subsequent performance (though he points out that the fund would probably have been forced to close in 1998) but we can safely assume it would have been dreadful. |
| Lo gives several examples of such ‘negative gamma’ strategies. He also offers a diagnostic test of this and other unattractive features. If returns to a fund are auto-correlated, they are likely to reflect illiquidity and understated volatility. The volatility can be corrected by simply scaling up by the degree of positive autocorrelation, which, in some cases, implies a massive change. Positively auto-correlated returns feature again and again as an undesirable feature in the first half of the book. (I quickly checked my own fund and was relieved to see that it received the all clear.) However, if a series of recent meetings with investors is anything to go by, awareness in the industry of this straightforward test is low. |
| Hedge fund replication, or cloning, was a hot topic when this book was being prepared and is the longest Chapter in the book. In essence the issue is whether statistical analysis can devise simpler – and hence cheaper – ways to deliver hedge fund-like returns. In other words, do hedge fund managers have sufficient skill to justify their fees? Lo presents results for a range of hedge fund strategies. The work is careful and detailed. The results of post-sample models are inferior to the actual results even though the clones are assumed to charge no fees. The results improve if in-sample models are used but the gap remains. Yet in-sample models have no place in this kind of analysis, for reasons that Lo gives: devising successful investment strategies with the benefit of hindsight is not difficult. On the basis of this evidence from 1986 to 2007, even the average manager seems to have skill, even after deduction of fees, in most sectors. As investors receive the dreadful news on recent performance, they may draw cold comfort. |
| Professor Lo does discuss relatively recent events, notably August 2007, but the book obviously does not cover the latest financial crisis. However, I was struck by the following passage. “Our tentative inferences suggest that the hedge fund industry may be heading into a challenging period of lower expected returns and that systemic risk has been increasing steadily over the recent past.” History will judge. |
| Steven Bell |
| Portfolio Manager, GLC Global Macro Fund, and Director GLC Ltd. |
|