Mean Reversion Lesson
A reader, Chris Dugan, forwarded me this comment on Amaranth from Bernie Schaeffer:
“It’s amazing to me that Amaranth seems to have relied on the same “principle” as did the disgraced managers of Long-Term Capital, “stewards” of what was (until Amaranth) the biggest blow-up in hedge-fund history. Simply put, this “principle” holds that various historical relationships between various markets will “mean revert” once they begin to diverge significantly from the norm. The mean reversion principle has become an article of faith among many of the major hedge funds. Given the alleged sophistication of these supposedly risk-savvy players, it is shocking that a principle that thumbs its nose at the “fat tail” events that are far, far more common than any modeling ever suggests continues to have holy-grail status. And then, when funds like Long-Term Capital and Amaranth blow up, the ready excuse is “who would have thought such a ‘highly remote’ event would ever occur?”
He added:
“The lesson here is that these models that are so avidly followed by those who should know better have been dead wrong and continue to be dead wrong about the odds of so-called rare events. The equity markets have been in a derivatives-induced coma for several years now, and my sense is that the upcoming fourth quarter is about as ripe a period as I can imagine for this coma to come to an abrupt end. Regardless of which way the break goes, we will likely have derivatives pain, which almost certainly means more hedge fund pain. An upside breakout can blow away those who’ve been heavily overwriting calls; a downside break creates potentially huge liability for those who have sold a massive quantity of very low delta puts that have thus far been “free money,” month after month. In other words, market moves may begin to feed on themselves rather than mean revert, and some players will pay - big time - for this.”
Now that is a great view from Bernie Schaeffer.
