The risk associated with hedge funds and hedge fund-like mutual funds can in some cases be substantially lower than that attached to common types of mutual funds. That is the conclusion of a recent Lipper Research Series report.
In order to examine how these types of funds compare with other traditional mutual funds as portfolio diversifiers, the research focused on their risk-measurement properties rather than on their activities or investing methods. The research uses the value-at-risk (VaR) measure, rather than standard deviation, to measure risk, considering the former as a superior measure where the main objective is to minimize losses.
When deciding on a correlation measure, Andrew Clark, the senior research analyst at Lipper who prepared the research, says he based his choice
on the observation that weak or nonexistent correlations can suddenly turn into strong ones as the result of a major event.
A more appropriate calculation of correlation, which the research uses as an inverse measure of diversification, may therefore be achieved using Kendall’s tau, a measure of correlation devised by the 20th century English statistician Maurice Kendall. This is preferred to other correlation measures because in extreme situations it is seen as the most appropriate way of capturing the movement from weak to strong correlation.
Focusing on “Extreme” Events
For the 13-year period for which monthly hedge fund data was examined, from January 1994 to November 2006, the focus was on “extreme” events–that is, when there were monthly loss levels with a one-in-ten or smaller chance of occurring. For these events, the researchers found that none of the hedge fund indexes were correlated to any of the mutual fund indexes. Similar results were yielded by an examination of daily data referring to hedge fund-like mutual funds over the three-year period to the end of 2006.
Yet when all losses are considered, a number of strong correlations were revealed between certain hedge fund strategies/hedge fund-like mutual funds and traditional investing styles. For example, large caps showed a strong correlation to dedicated short bias, event-driven, multi-strategy, long/short, and distressed funds in periods in which “normal” losses occurred.
But Clark argues that the correlations identified in non-extreme circumstances are not pertinent because these are not typically the situations where diversification is most important or even necessary. “These correlations, to our minds, are spurious, since they are clearly being driven by correlated behavior in the noise zone (the 40% of returns to the left of the mean),” the research paper states. “And noise-zone correlations are best treated using trading strategies, not diversification strategies.”
These results provide a strong indication that hedge funds endow a portfolio with good protection against anomalous downturns in both equities and bond markets. In concrete terms, in extreme situations it was found that that none of the hedge fund or hedge fund-like mutual fund indexes considered were correlated to any of the mutual fund indexes. It can therefore be maintained that in such circumstances, hedge funds and hedge fund-like mutual funds have been shown to be less risky than most mutual funds, and are therefore likely to prove more effective as diversification tools.