Antifragile: Things That Gain from Disorder
These notes are old and were written while reading — they don’t necessarily reflect my current views.
Anti-fragility is the concept that is the opposite of fragility. Fragility is something that is negatively impacted by volatility. The opposite of this would be something that is positively impacted by volatility not something unaffected (robustness). There are many of these systems that we dont think of as such because we are unable to express the thought (there was no word for anti-fragility). The evolutionary process, markets and humans are examples of antifragile systems.
From his talk on the topic:
Taleb critiques Greenspan for micromanagement the economy. If we care about every Forrest fire, there will accumulate a lot of dead wood and the one big fire will be worse. He says that people (Keynesians) make the mistakes of confusing the market ,a system that benefits from volatility, with a system that does not. Driving a car in the wall 100 times with 1 mph is much better than doing it once with 100 mph. This is also an argument for decentralised government. We are generally in a political climate where we try to save us from small problems and therefore create large ones. Exposure is important.
Systems inside the larger antifragile system are often fragile. The market is antifragile because the firms inside it are. Because the mechanism can select for appropriate firms, the market as a whole is more resistant that if there was no selection. Because the firms that survive struggles the best are continued while others are not, the overall quality and reliability of the supply increases.
Taleb is very critical of forecasting and says that black swan events are per definition not forecastable. Therefore, he advocates for antifragile systems. He makes an even better example than the black swan story on how this can go wrong. He pictures a turkey that grows up at a farm an observes the framer feeding it every day. At first it cant be certain of the good intentions of the farmer as it could be statistical coincidence that every day there is the same behaviour (feeding the turkey). After a couple of months the turkey has achieved certainty that the farmer is its friend, as the data on this is not only overwhelming but contradictory data is not even existent. Then comes thanksgiving.
Taleb explains that fragility and anti-fragility stem from non-linearity. Non-linearity might be convex or concave.
Convex non-linear systems gain from volatility, while concave systems behave contradictory.
He continues to write:
We can see applications of the point across economic domains: central banks can print money; they print and print with no effect (and claim the “safety” of such a measure), then, “unexpectedly,” the printing causes a jump in inflation. Many economic results are completely canceled by convexity effects—and the happy news is that we know why. Alas, the tools (and culture) of policy makers are based on the overly linear, ignoring these hidden effects. They call it “approximation.”
Likewise, government deficits are particularly concave to changes in economic conditions. Every additional deviation in, say, the unemployment rate—particularly when the government has debt—makes deficits incrementally worse. And financial leverage for a company has the same effect: you need to borrow more and more to get the same effect. Just as in a Ponzi scheme.
As an example for (c), which is a more complicated version of the bias, assume that the function under question is the squaring function (multiply a number by itself). This is a convex function. Take a conventional die (six sides) and consider a payoff equal to the number it lands on, that is, you get paid a number equivalent to what the die shows —1 if it lands on 1, 2 if it lands on 2, up to 6 if it lands on 6. The square of the expected (average) payoff is then (1+2+3+4+5+6 divided by 6)2, equals 3.52, here 12.25. So the function of the average equals 12.25.But the average of the function is as follows. Take the square of every payoff, 12+22+32+42+52+62 divided by 6, that is, the average square payoff, and you can see that the average of the function equals 15.17.So, since squaring is a convex function, the average of the square payoff is higherthan the square of the average payoff. The difference here between 15.17 and 12.25 is what I call the hidden benefit of antifragility—here, a 24 percent “edge.”As an example for (c), which is a more complicated version of the bias, assume that the function under question is the squaring function (multiply a number by itself). This is a convex function. Take a conventional die (six sides) and consider a payoff equal to the number it lands on, that is, you get paid a number equivalent to what the die shows —1 if it lands on 1, 2 if it lands on 2, up to 6 if it lands on 6. The square of the expected (average) payoff is then (1+2+3+4+5+6 divided by 6)2, equals 3.52, here 12.25. So the function of the average equals 12.25.But the average of the function is as follows. Take the square of every payoff, 12+22+32+42+52+62 divided by 6, that is, the average square payoff, and you can see that the average of the function equals 15.17.So, since squaring is a convex function, the average of the square payoff is higher than the square of the average payoff. The difference here between 15.17 and 12.25 is what I call the hidden benefit of antifragility—here, a 24 percent “edge.”
Taleb shows that larger enteties are more negatively affected by volatility than smaller ones.
He explains the Lindsay principle, which states that the expected lifetime of something is approximately its current age, as it is most probable that its in the middle of its lifetime.
He connects the concept of anti-fragility (optionality)to his concept of skin in the game by stating that gaining optionality at the expense of others is the essence of ethical misconduct. He uses bankers that use the Greenspan put to hedge their risks and academics that advance their careers by cherrypicking data as examples of such behaviour.