The Black Swan: The Impact of the Highly Improbable
Nassim Nicholas Taleb
I really wanted to like this book. Although I am among a class of people who come in for the greatest criticism from Taleb (central bank technocrats), I often enjoy reading critical perspectives (Graeber, Meltzer). But not this one, for a variety of reasons. The blogger Eric Falkenstein covers most of these (http://www.efalken.com/papers/Taleb2....). Taleb’s writing style is rude and he often goes for the cheap laugh with stereotypes about “Frenchmen” and “guys from Brooklyn.” He tends to attack a straw man, as evidenced by his frequent use of fictional characters and his superficial treatments of the views of real people.
At base, I really do agree with one of Taleb’s main premises: that we as humans are subject to occasional very severe, very unpredictable events. This premise, and views about how to deal with it, is in fact one of the main themes of one of my favorite books, “Watership Down.” But I disagree strongly with his views on how to deal with it. I think it’s fair to characterize his main piece of advice as, “avoid exposure to negative Black Swans and maximize exposure to positive ones.” My problems with this are both practical and theoretical.
First, the practical. Taleb’s advice is glib because, by the very definition of a Black Swan, we are not likely to foresee even its possibility in advance. In many cases, the same phenomenon can be subject to both positive and negative Black Swans. Take the U.S. housing market, for example. In the early 2000’s, most people would have seen investing in housing as a way to expose yourself to positive Black Swans (the possibility of an unprecedented run-up in house prices). Later, it turned out to be the opposite (an unexpected decline). A related point is that increasing your own exposure to positive Black Swans often requires increasing someone else’s exposure to negative Black Swans, so that the game is zero-sum. For example, a good way of increasing your exposure to positive Black Swans is to buy risky stocks with borrowed money. But of course, the person who lent you the money is symmetrically exposed to a negative Black Swan.
The practical difficulties become even more apparent when Taleb tries to get specific with his advice. For example, he states that a good anti-Black-Swan investing strategy is to put 90% of your money in Treasury bills and 10% of your money in speculative bets like venture capital. He even says that through this strategy, “no Black Swan can hurt you at all.” But of course, the mother of all Black Swans would be the default of the U.S. Treasury! Treasury bills are exactly an instrument that is robust in almost all conditions but subject to risk from a Black Swan. Pre-crisis, someone might well have replaced “Treasury bills” with “super-senior MBS.” Second, both exposing yourself to positive Black Swans and protecting yourself against negative Black Swans (assuming it is possible) have costs. The question is whether those exposures are underpriced, overpriced, or correctly priced on average. Of course, many of the things Taleb talks about, like going to cocktail parties because of the chance of making felicitous connections, are difficult to quantify. But in the one area where the costs are easily quantified–financial markets–the evidence indicates that the opposite of Taleb’s assertion is true. Deeply out of the money options (which can either protect you from negative Black Swans or expose you to positive ones) are in general overpriced. Taleb might argue that this is only based on the historical data, and ignores the possibility of an even bigger future crash that would reverse the sign, but this would seem like special pleading. Taleb denigrates the financial strategy of “picking up pennies in front of a steamroller” (i.e. getting paid to insure others against catastrophes), but it turns out to be a basically profitable strategy. Furthermore, there are so many possible catastrophes that insuring yourself against “as many as possible” seems almost certain to be prohibitively expensive.
In addition, I have theoretical problems with Taleb’s advice–in the realm of psychology. Taleb actually dwells on this point in some detail, yet seems to draw a completely invalid conclusion from it. Human happiness is primarily based on the frequency of good or bad outcomes, and not their magnitudes. So in terms of rewards we are happier with frequent small rewards than with infrequent huge rewards, but in terms of punishments we are happier with infrequent huge punishments than with frequent small punishments. More technically, we are risk-averse in gains and risk-seeking in losses. This may be seen as “irrational” but it is an intrinsic feature of humanity’s relationship to its environment. It is well-documented empirically and the basis of Kahneman’s prospect theory. But what is the implication of this finding? That we shouldn’t waste time trying to gain exposure to positive Black Swans (because they won’t make us that happy anyway), and we shouldn’t worry too much about negative Black Swans (because we’d recover from them pretty quickly anyway, and we’d have to pay money every day to insure against them). In other words, pick up pennies in front of a steamroller! Taleb goes off on a fictional tangent immediately after describing this set of facts, so he fails to draw the clear conclusion from it.
Let me return to “Watership Down” (mild spoilers ahead, if you haven’t read it). The fundamental problem for rabbits is that they are exposed to a wide variety of extremely dangerous risks, basically from humans and predators. Over the course of the book, we meet three different warrens of rabbits who deal with this problem in three different ways. There is Cowslip’s warren, which is optimized to gain frequent, small rewards, at the cost of being heavily exposed to negative Black Swans. (The rabbits are fed by a farmer every day, but the cost is that every once in a while a rabbit is slaughtered.) There is Efrafa, General Woundwort’s warren, which is optimized to minimize exposure to negative Black Swans at the cost of frequent, small sacrifices. (Woundwort runs the warren as a police state that is heavily disciplined in order to maximize vigilance against predators.) And finally, there is Hazel’s warren, our protagonists, who are not really optimized for either of these. Rather, they just try to remain flexible and cohesive, willing to take risks but not doubling down on their understanding of the world. Clearly, this is a fictional book about rabbits, but I think Adams portrays very skillfully the problems with the Cowslip and Woundwort approaches to the world in terms of the fulfillment of human (rabbit) potential and happiness.