framboise mécanique: Some Musings on The Black Swan

Envoyé par guillaume e via Google Reader :

Some Musings on The Black Swan

via naked capitalism de Yves Smith le 07/03/09

I must confess to being very late to read Nassim Nicolas Taleb's The Black Swan, and frankly had considered NOT reading it. First, it has been so widely reviewed and discussed that I had assumed the additional knowledge to be gained by reading the book itself would be marginal. Second, I've read a bit of Benoit Mandelbrot, who is arguably Taleb's most important predecessor. It was Mandelbrot, a mathematician, in the 1960s, who found 100 years of cotton trading data, with daily prices. Mandelbrot cut the information every which way and found that its distribution did not at all correspond with the assumptions underpinning the new and growing school of financial economics. It was Mandelbrot that discovered "fat tails", that very extreme price movements are far more likely than the theories predict. He also found that market have memory, that their price movements do not comport with the "random walk" theory (I have assumed this pattern somehow relates to the cognitive bias of anchoring, but do not know if anyone has been able to connect the dots).

His findings were initially rejected, and continued to be resisted even when they were replicated in other markets (and then proved out in the real world: a daily price move like that of the 1987 crash was so extreme as to verge on being statistically impossible).

Put simply, computational convenience trumped empirical findings.

Even though Taleb's observations have gotten a lot of attention in the popular media, I sincerely doubt they will be internalized. In classic cognitive dissonance fashion, his views may be given more lip service, but it will not be integrated into mental models, even of those who ought to pay heed. The very fact that his construct has been reduced to the soundbite "black swan" when it is more complicated and richer is telling.

What are some of the reasons? Let me speculate.

First, Taleb goes to some length to establish that he is not the first to go down this line of thinking; he has quite a few intellectual ancestors. Yet these observations never took hold.

Of course, one reason is that the implications are pretty uncomfortable for a lot of professions (although Taleb would dispute their clams of professionalism). He contends that predictions are a fraught-to-useless exercise, and cites research that shows that lay forecasts are frequently no worse than those of experts. His book would appeal most to people who are well educated and interested in finance and economics. A fairly large subset of that group has invested in expensive educations and/or developed a lot of career experience to try to anticipate the future better than your average slob. Do you think a book, even a very persuasive book, is going to change how people operate on a day-to-day basis? Unlikely.

But second, and perhaps as important, people do not want to see the world as subject to chance to the degree that Taleb says it is. This is hugely unsettling if you really do come to terms with the implications of his argument. We like to believe we have some measure of control over our lives. And research has shown that people do pretty consistently overestimate their degree of control and influence (for instance, most people will exaggerate their contribution to the success of a project, not as a matter of PR, although that may be true too, but their private assessment). Most people (ironically those deemed psychologically healthy) have an optimistic bias and generally assign too high odds of things working out well (the mildly depressed make more accurate assessments. I have often wondered which way the causality runs: do they make better assessments BECAUSE their unhappy state strips away the rose-colored filter, or are they mildly depressed because they keep giving more realistic assessments, which makes them a drag to be around, and they are depressed because they encounter social rejection?). So if you embrace Taleb, you'd have to accept the disorienting fact that the world really is a pretty untractable place, that success had more to do with luck than application (although Taleb stresses the importance of working at being lucky, that is, accepting the opportunity to meet new people and make the most of chance encounters).

Third, if our mental construct of how the world works is off in some fundamental respects, it also calls into question our ability to make good decisions. And apart from Taleb, there are reasons to question our abilities here. It has been pretty well documented in brain research that humans can only hold so many variables in their consciousness at once. Our decision-making capabilities are more limited than we'd like to believe. And confronting every situation as if it were new would be simply exhausting, That is why we rely heavily on rules of thumb (more fancily called heuristics). Now we also have certain types of analytic processes, what I like to think of as pattern recognition, that can serve us well (this was the topic of Malcolm Gladwell's Blink). The problem is that this quick pattern recognition can work very well, or be absolutely wrong, and we have no easy way of telling which.

Essentially, Taleb paints a picture of the world and human behavior that is unflattering. So as much as his work makes a fundamentally important set of observations, its success may be largely a function of luck. It came out just when the credit markets were starting to unravel and well established practices, both among traders and the broader financial community, were being shown to have serious flaws. Had his book come out at another juncture, it probably would not have been as well received.