Wednesday, October 22, 2008

Nassim Nicholas Taleb changed my life...

Last night on PBS Newshour, economics correspondent Paul Solman interviewed Nassim Nicholas Taleb and his mentor Benoit Mandelbrot (link to interview transcript). It was timely.

Taleb's second book (Fooled by Randomness) was the seventh book that I read this year, I got it last year for Christmas. I wrote back in March of how it changed my life and way of thinking. So, mark it down as one of the best Christmas gifts ever gotten.

In the book, Taleb railed at the financial economists and math wizards on Wall Street who said everything they were doing was risk free. Here's an interview with him back in April. His point was partly that the probabilities simply can't be properly determined. He has his "turkey" analogy-- a turkey is raised on a farm and fed large meals for 100 days. The turkey gets used to the routine and estimates a 100% probability of him being fed a big meal the next day. Then, on the 101st day he gets his head chopped off. It doesn't surprise anyone but the turkey.
Taleb was, of course, right.

I found this clip on YouTube of Taleb on a BBC Newsnight broadcast where he is clearly angry and quite worried about how bad the financial crisis will eventually get. Ken Rogoff, a hero of mine, was also on the show that night but unfortunately the clip edits him out (and there is no transcript or full video on the BBC site). Taleb and Rogoff together on the same show, my head would explode!!

Taleb's interview last night was equally intense as he's worried about hedge funds deleveraging. He considers this to be the worst crisis for the U.S. since the American Revolution! But, he is first to admit he doesn't know.

He got my attention.

In the fall I have to teach Risk Management, which means Portfolio Theory and things that Taleb calls unscientific bunk. I read his op-ed in the Financial Times here and cringe.

Mathematicians often hate economists for being quacks and trying to pretend to do things with mathematics that can't actually be done. Economists hate journalists and politicians. The journalists on NewsHour are the only ones I would say have a right to be called journalists, I'm definitely thankful for them.


Perc said...

Taleb is a good writer, but I am not a fan, simply because his ideas are not new. Keynesian and Post-Keynesian economists have been talking about "uncertainty" for a very long time before Taleb surfaced. So to me, his idea of "black swan" is a resurrection of an old idea, that PhD trained Economist already knew. Still, I am glad for his writings because to the general public it has value. I saw his interview on BBC with Rogoff and his overly dramatic comments against Economists, which I hope sold more copies of his books. Too bad we couldn’t hear Rogoff’s comments on youtube, which for obvious reason was edited. Since Taleb worked in Wall Street, I am sure he is familiar with Black-Scholes and Martingale models, which were mostly developed by Mathematicians (who won Nobel Prize in Economics), like Taleb, not Economists. I think the world would think of Taleb as a genius if he can develop a model that ensures financial markets consistent returns even if the underlying value of a security is negative. Securitization has made it very convenient for financial intermediaries to pass the ball, risk, on someone else's court. We know that the innovate financial products in recent years have assumptions and flaws. The question of how can a bank lend money to an unqualified borrower and be guaranteed payment, still remains. If there a model for this, perhaps, the answer is government steps in to guarantee the loan and this is what's happening right now in a form of massive bailout. Someone said the bailout can be looked at as protecting FDIC. To take my point further, what's going on today is a result of continued consumers spending beyond their means - in Economics, thinks about indifference curves above and beyond the budget line. Ramsey's paper, A Mathematical Theory of Saving, in 1928 showed that the "equilibrium would be attained by a division of society in two classes, the thrifty enjoying bliss and the improvident at the subsistence level."

JTapp said...

Perhaps. I'm reading Mandelbrot's The (Mis)behavior of Markets, Mandelbrot is Taleb's "teacher." The point is made that econometrics and much of the risk management done in finance assumes normal distributions. For example, outliers are deleted from datasets so as not to skew things. I think they have a fair criticism. Yes, there are a lot of mathematicians creating things in finance, but Mandelbrot and Taleb have issues with many of them. But, as others have said, much of what economists want to prove with econometrics cannot be proven because the mathematics simply do not exist yet.