Skin in the Game, Nassim Nicholas Taleb

Skin In The Game is a thought provoking and provocative book. Taleb’s core thesis is that skin in the game – being exposed to the possibility of loss from the consequences of one’s decisions – is vital for societies to function properly and general fairness. Throughout the book, he explores various concepts ranging from hedge funds to religion to illustrate how universal the importance of having skin in the game is.

Although the ideas in the book are very interesting, I found Taleb’s pugnacious style a bit too incoherent. This meant that quite often, Taleb’s ideas were very hard to parse amidst what seemed at times an incoherent rant. This went away as I read more of the book and got used to his writing style, but it does mean that I definitely was not able to understand him as much as I would have liked. This is a shame because the ideas that I did understand, which I will write about below, I found to be extremely thought provoking and interesting. Perhaps one day I’ll reread this book and flesh out this review with a fuller summary.

The central message is one of enforcing negative incentives. If someone makes an incorrect decision, they must be exposed to some consequences. Taleb, himself a former options trader, uses those in the financial sector as an example of why this is important. One of his favourite targets is former Citigroup director Rob Rubin, whom he repeatedly eviscerates for getting away while Citigroup had to be bailed out by US taxpayers. If Rubin had had skin in the game, he might have thought more carefully before allowing his bank to engage in the type of reckless speculation that led to the financial crisis. Instead, the banks had a sense that they were too big to fail so if something went wrong, the government would have to come and bail them out which is indeed what happened. Faced with the prospect of upside from making bets on complicated derivatives like mortgage backed securities and limited downside risk, the bankers engaged in the behaviour that they did, helping to cause the Global Financial Crisis. Taleb’s argument is that if the bankers were fully exposed to the risk they were taking, this type of reckless behaviour could have been controlled, showing how skin in the game is essential to the proper function of markets.

Markets are a fairly unsurprising application of this concept, but then Taleb extends it to other more interesting spheres. One of these is the idea of minority rule, or how a small but intransigent minority can often impose their preferences on an indifferent majority. The example he gives is kosher meat. Although only 2% of the American population is Jewish, around 40% of packaged food is certified kosher. The reason for this huge disparity can be attributed to skin in the game. The small Jewish minority has enough perceived downside risk (from divine punishment) from eating non kosher meat so as to not be willing to negotiate on this. If this minority is strongly localised, then it won’t have much of an effect. To see why, suppose all the Jewish people in America lived in New York City. In this case, food companies would have little incentive to implement special kosher certifications except in New York City. However since Jewish people are spread out all over the US and have an extremely strong belief in kosher food, it is simply easiest for food companies to mass produce their products in kosher compliant fashion. This explains how skin in the game and a minority with a critical minimum degree of spatial distribution can project its preferences onto an indifferent majority.

One very cool idea was how Taleb linked the importance of the minority being spread to a concept in Statistical Mechanics called the Renormalisation Group. To see this, suppose we split America into 100 equally sized areas. If a given area has a certain proportion of a committed minority, say 2%, then we colour it in blue, otherwise we leave it alone. If the minority is clustered in only one area, then only one area is coloured blue and requires catering to the minority rule. But if the minority is present in small but nontrivial numbers (>2% for example) in multiple areas, then there are multiple areas where minority rule will prevail. When we view this toy America at the largest scale, the minority appears irrelevant, but when viewed at more fine grained scales, the minority becomes more powerful if distributed in the right way. This is a version of the Renormalisation Group argument which is the formal way in which physicists investigate what happens when a physical system is viewed at different scales. It was definitely very cool to see something from my grad school stat mech class show up in such an unexpected context.

A fun concept that Taleb came up with is the IYI – intellectual yet idiot. The idea here is that there is a certain class of “well-pedigreed” people from elite institutions who tell people what to do, eat, speak, think etc. The core problem with intellectuals is – you guessed it – lack of skin in the game. Because they are insulated from the consequences of their advice/actions, the IYI class has been spectacularly wrong on many counts. When I read this, the first example that came to my mind was the intellectual class goading the US into war in Iraq, but just in case you need some convincing, Taleb rings up a solid list of examples on pages 125-6 which includes “Stalinism, Maosim, GMOs, Iraq, Libya, Syria, lobotomies … linear regression [lol], … Salafism, dynamic stochastic equilibrium modeling, … election forecasting, … and p values''. Phew. I’ll be honest much of this chapter just seemed like a gratuitous takedown of the intelligentsia which, while entertaining, wasn’t particularly enlightening. I think it’s reasonably apparent that many of the intellectuals, including plenty of us in academia, don’t face enough consequences when we are wrong.

Skin in the game is important because ultimately it’s tied to natural selection. Through the possibility of loss, natural selection can eliminate inefficiencies. Because of constant selection pressure, time is ultimately the gold standard of testing robustness. Taleb argues that if something has survived for a thousand generations, it’s probably not an accident. This is known as the Lindy effect. More formally, it is the effect when the future life expectancy of something is proportional to its current age. The longer something has lasted, the longer it will last, because it has been selected for by evolution. Taleb argues that it’s a good idea to listen to your grandmother’s wisdom because it’s survived through the generations and is Lindy protected. Another example where the Lindy effect manifests is in social science research. Taleb contends that “while our knowledge of physics was not availble to the ancients, human nature was. So everything … in social science and psychology has to be Lindy-proof, that is, have an antecedent in the classics”. Such concepts include cognitive dissonance, loss aversion, negative advice and many more. This was definitely a cool section and I liked the idea of social science needing to be Lindy proof.

The only thing which was a bit nauseating about the Lindy chapter was Taleb’s restriction of “the classics” or “the ancients” to Greco-Roman literature and some other French writers. This is definitely due to his own upbringing (which he is clear is “Greco-Levantine”) but it would have been nice to have at least an acknowledgement of other classical traditions that have produced multiples of the entire canons of Greek and Latin combined, such as the Indic tradition. More substantively than my little nationalist nitpick, comparing social science to other traditions might reveal that “Lindy-proof” doesn’t have a universal definition like he is making it out to be. If some modern piece of research doesn’t have an antecedent in Taleb’s artificially limited sample of “the ancients”, it doesn’t necessarily mean it’s not Lindy proof because it might have an antecedent in the broader neglected sample of the ancients. This is probably not his fault though – he grew up in a Greco-Roman tradition so his horizons are understandably oriented there. Anyway he probably has some more concrete mathematical formalism which covers these holes.

One final concept from his book that I really liked was his approach to inequality. In his characteristic aggressive style, he throws down the gauntlet to snob academics in economics faculties who focus on inequality like Piketty. In addition to accusing them of basic errors like not knowing how statistics work, Taleb accuses them of viewing the problem in a very limited light. He argues that academics and intellectuals view inequality as a static snapshot of society at one point in time which they argue must be corrected. Taleb instead advocates for a dynamic view and says that wealth inequality is in fact not a problem provided that those at the top have a nonzero probability of losing it all. In Taleb’s dynamic view, in a truly fair society everybody should have an opportunity to experience all rungs of the ladder in complete mobility, including up and down. Thus the entrepreneur who becomes rich is not to be vilified, because he took a risk and can still lose it all. The inheritance aristocrat who did not earn their wealth and has very little scope of losing it, the type that dominates lists of the wealthy in Old Europe for instsnce, is a more appropriate target.

Mathematically, he relates this to another concept which I saw first in my stat mech class – Ergodicity. This is the idea that a random process will visit all parts of the state space over time. In statmech, this means a physical system will evolve to cover the space of possible positions and momenta in configuration space. In Taleb’s example, this means a person can evolve from poor to rich and back to poor. This type of ergodic behaviour is only possible with the possibility of loss and thus skin in the game.

Overall, Skin in the Game really expanded my own intellectual horizon. While in the start I wasn’t too amazed by his example of using bankers to show the importance of skin in the game, the other extensions he made towards minority rule and the Lindy effect were very interesting. I especially love how he used concepts like renormalisation and ergodicity, which I had only seen in a physics context. This gave me insight into just how general these concepts are and helped me appreciate them from a new angle. The only downside of this book was that at times it read like the incoherent rant of a taxi driver so it was difficult to digest all of it. Regardless, what I did digest was very enlightening and I recommend this book to anyone looking to challenge their own view.