The good, the bad and the average
New analysis from Merrill Lynch indicates that hedge funds have never been so closely correlated with equities as they have in the past three years. It initiated another of those periodic waves of comment that pension funds are wasting their time and money going to hedge funds for "uncorrelated returns".
IPE's monthly Off the Record survey reveals mixed experiences. "High correlation with equity markets," complains one investor. "Low correlation with other asset classes," enthuses another. "I no longer believe that [they represent value for money," says one. "Yes [they do]," says another, "with good selection."
And that, surely, is the point. Hedge fund managers do a wide range of things with varying degrees of idiosyncrasy. Some of it simply cannot be meaningfully correlated with equity markets: writing insurance against earthquakes, for example, or running a portfolio that is long the volatility of volatility. Medium-term trend followers correlate, funnily enough, with anything exhibiting a medium-term trend: that may well be equities; it could just as easily be Japanese yen or Kansas wheat.
Lots of hedge funds are correlated with equities, of course. Much long/short equity is concerned with old-fashioned stockpicking. As much market risk is hedged out as possible, but that doesn't change the fact that the stocks you buy and sell will correlate with the market because they are part of that market.
But with these guys, it's not the correlation coefficient but the beta coefficient that's important. It's worth remembering why pension funds got excited about hedge funds in the first place: they preserved capital through the dotcom crash. But over that period, their correlation with equities leapt from about 60% to close to 90%. They protected their clients because their beta fell from 40% to 20%.
And we're still talking about the average fund here. It's odd that commentators who are always telling us the average active manager cannot outperform the index seem unable to apply the same logic to hedge funds. The more long/short equity managers you pull together the more the negative alpha diversifies away the positive alpha: the more representative an index gets, the more it will correlate with the market. That correlation still tells you nothing about the index's beta coefficient (the effect of funds managing their market risk), but more importantly, it tells you less and less about the characteristics of individual managers - for the same reason the Russell 3000 tells you nothing about Warren Buffet's stockpicking skills relative to mine.
Thankfully, fewer pension funds continue to search for uncorrelated returns from long/short equity. But that doesn't mean they've given up on it as a source of variable-beta equity risk. Nor have they given up on uncorrelated returns from things like global macro, managed futures, volatility and non-financial investments. When you start to think of individual hedge funds in this way those fees can start to make sense again.