About Alpha has made noble attempt to paint hedge funds as not particularly
risky, and "fat tails" as not particularly worrisome. But it misses
the main point, I think.
The blog makes some good points, foremost among them being that the standard
measurement of tail fatness, kurtosis, is a really bad way of measuring how
likely a hedge fund is to encounter a disastrous fat-tail event.
The blog concludes with a provocative quote: "If you think hedge funds
are risky, try stocks." Which is a bit misleading, because what that really
means is that individual stocks are more likely to lose their value than individual
hedge funds. That’s true, which is why no one in their right mind puts all or
even most of their money into an individual stock. Instead, they put their money
into funds, which are designed to be much lower risk. (The reason why hedge
funds are so called is precisely because they’re supposed to hedge the risks
inherent in the stock market.)
What’s more, the problem with improbable but disastrous events is that they
don’t happen very often, and that therefore they’re very hard to measure with
any accuracy. A fund might have low kurtosis, while still being very highly
exposed to any number of individual events which just haven’t happened over
the course of the fund’s existence.
Stocks, for all their volatility, represent real companies in the real world,
which tend not to be wiped out overnight by improbable fluctuations in the financial
markets. In fact, companies have much lower event risk than hedge funds do.
And in any case, the problem with hedge funds is not that they have too much
event risk in aggregate. The problem is rather that there’s bound to be one
or two hedge funds out there which are too highly levered and carry too
much event risk. If those one or two hedge funds are big enough – as we
saw with LTCM – they can pose real systemic risk.