Nassim Taleb, The Black Swan

I just finished listening to Nassim Taleb's The Black Swan (2006), and I liked it. This book made Taleb famous for a while, because it seemed to predict the financial crisis of 2008. Actually Taleb would deny that he predicted anything, in a strict sense, since his main argument in the book is that the future is not predictable. What he did say was that financial crises happen from time to time, that changes in the financial industry (fewer, bigger banks, more global interconnectedness) made it likely that the next crises would be very bad, that such a worldwide crisis might be triggered by the failure of any of the big banks, and that the whole apparatus of "risk management" and "portfolio science" would not help anybody when the crisis came. Since he was right on every one of these points, he earned a hearing for the arguments in this book.

Taleb argues that the future is not predictable because events that are too rare for us to be able to model their probability have a large impact on how things turn out. He calls these events "black swans." He spends most of the book attacking economists and financial gurus whose models discount the possibility of large, rare events. Most models in economics and finance, he says, are based at some level on the bell curve or "normal distribution" of events. In a bell curve world, events far away from the norm are vanishingly rare, and in a large data set they have very little impact on the average. This is a dangerous illusion, says Taleb. In the real world we have no idea what the likelihood of dramatic, rare events might be, and when they do come they are so significant that they overwhelm years or even centuries of normal progress. In the banking crisis of 1982-1983, American banks lost more money than they had earned in the previous century. (This started the chain of events that led to the savings and loan bailout and the mass mergers of American banks.) Because their work is based on a false model of the world, the prescriptions of economists and financial "experts" are useless. Taleb is particularly savage about the Nobel Prize committee, which gives its prizes to economists whose models are so mathematically pristine that they have no relationship whatsoever to reality.

You might think that, well, sure, but nobody takes economists' models seriously enough for them to do any real damage. Not so. In 1982, American banks had models showing that their exposure to Latin American debt was a manageable risk. In 1998, Long Term Capital Management, run by a staff of luminaries that included two Nobel Prize winning economists, was driven into bankruptcy by a Russian financial crisis that lay outside the models they used to "manage" their risk. Without massive intervention by European and American central bankers, this crash might have brought down the world financial system. People who should have learned from this near disaster bet massively on complicated financial instruments, many of them tied to mortgages, and used bell curve-based models to convince themselves that were running essentially no risk. Those models were wrong, and we are paying the price.

Taleb relates this mathematical hubris to an old philosophical quandary, the Problem of Induction. Just because something has happened the same way every day for your whole life does not mean that it will happen that way tomorrow. As David Hume put it, a chicken fed every day by the farmer becomes convinced that this is the universal order of things and is shocked on the day the ax falls. Banks and financial "experts" convince themselves that what has been happening predictably for a few years will continue to happen. The Greek government has been making regular payments on its debt, so it will keep paying. Mortgage-based securities have held their value, so they will keep holding it. Housing prices will keep going up. And so on. When the future turned out to be different from the past, the models failed. The chicken got eaten, and we ended up with a financial crisis and a worldwide recession.

I agree with Taleb about all of this, and with his withering attacks on the professions of academic philosophy and economics. His book suffers from some bad choices in terminology, repetition stemming from poor organization, and too much self-congratulation, but it is a very valuable attack on some of the biggest categories of self-important nonsense in our world. I wish it might have some impact on how we value economists, financial "experts", and other charlatans. But I doubt it.