Friday, February 22, 2008

Michael Lewis in Portfolio.com on Black-Scholes (March 2008)

There are numerous mistakes of fact or reasoning in this piece, which is surprising for someone of Mr. Lewis's credibility.
Examples:
1. $412 trillion of derivatives have no satisfactory pricing model because they rely on Black-Scholes or Black-Scholes like assumptions. This is nonsense. Those mostly instruments that have no option component whatsoever (like interest swaps). Mr. Lewis might note that the largest use of options of any kind, by far, is to provide prepayable mortgages to tens of millions of households.

2. Portfolio insurance was an example of the failure of the Black-Scholes idea. It is true that Black-Scholes assumes no jumps in markets, but Wall St professionals have realised this failing and compensated for it for about 25 years. Further, portfolio insurance and the market collapse of 1987 was an example of what happens when actors FAIL to use options. They had thought they had purchased options, but hadn't. If you sell something to yourself, you can easily be fooled about the price that you're paying as this is about as far from a third-party transaction as possible. Had they actually purchased those options, they would have observed the prices of those options rising with the growing demand for these options. They wouldn't have "purchased" as much and so would not have been as exposed to equities. So perhaps there would then have been no crash. Also, had they purchased these options, some of that risk would have been redistributed to actors who, though they have sold options, don't sell when the market goes down as they aren't "delta-hedging".

3. Housing was thought not capable of declining in price because of a Black-Scholes like mentality. (I think he's claiming this.) "Black-Scholes...gave...[millions of blissfully oblivious people]...the excuse to risk the roof over their head. ... Black-Scholes is no longer just a model; it has evolved into a climate of opinion..." How about another mentality that predates Black-Scholes: Human beings appear to thing that tomorrow will be much like yesterday. So if prices haven't gone down, they assume they won't. And if internet stocks went up yesterday, they assume they will tomorrow. This claim is quite strained.

No comments: