You're reading...



Antifragility is the topic of Nassim Nicholas Taleb’s last book in his trilogy on our uncertain world and how to deal with it.  The other two were Fooled by Randomness and The Black Swan.  These are both relatively narrow in their scope – though well worth their own reviews at some point – whilst Antifragility gives us the broad scope flagged by the book’s subtitle, How to Live in a World We Don’t Understand.  This puts it right at the core of what this site is interested in.  This article will remain a review.  I’ll post a ‘so-what for organisational risk management’ separately.

As the subtitle makes clear the problem is that we do not know what is going to happen.  This is not just in the sense that it might be A or it might be B with maybe some probability to reflect our belief as to which (what Taleb calls the ludic fallacy).  Instead we have to come to terms with a fundamental unknowability of what might happen: the future is populated by black swans as well as all the things on our risk register.  That’s a fact we have to embrace, not ignore.

This means that Taleb has many harsh words for prediction or forecasting, especially where it is dressed up as scientific, especially social scientific, especially where it is economics and specially especially where it is promoted by Nobel prize winners in economics.  (Though the patron saint of this column comes off relatively lightly, with a very obscure reference to his theories of general economic equilibria being based on the ludic fallacy.)  As the sort of risk modelling this author does (and sells!) falls into this category of potential charlatanism, this something we will need to come back to.

Taleb’s main theme is that our modern world has bought too far into our ability to predict and this has made us fragile.  Big government and big corporations have grossly overestimated their ability to control things for the better.  This leads them to pursue paths in which the payoff is generally a small gain but occasionally a large and fatal loss.  So the fragility of modernity leads to blow-ups such as – surprise, surprise – financial crises, and the scale of these crises is such that their instigators have to be bailed out with the consequent creation of moral hazard and a generally corrupt society in which the losses are not borne by those who are responsible for them.

Another fragility theme which runs throughout the book is iatrogenics: doctors are often bad for your health.  This happens because (a) they are under a perceived obligation to do something and (b) what they do may not be very good for you.  Taleb argues you should only see a doctor if you are seriously ill when there is a better chance that the upside will outweigh the downside.

This contains the seeds of the antidote to fragility which has two stages.  First you need to be robust or resilient.  This is the idea that, whatever happens, you can ride it out.  This is starting to rise up the risk management agenda because of the perception that this enables you to manage risk without the inconvenience of listing the risks first.  The idea of the resilient organisation is a vaguely defined article of faith in the business continuity industry, for example.  But without stopping to debate this, Taleb takes us to the next stage of antifragility, the idea that we can manage our affairs so that we can benefit without understanding the world (in detail).

The essence of antifragility is that we don’t bother with probabilities; instead we think about outcomes and, specifically, try to put ourselves into a position where there are downsides – perhaps many of them – which don’t hurt us much together with large upsides which will make us rich, happy or free.  It is an investment strategy based on this which has allowed Taleb to make the ‘fuck you’ money which subsidises his scholarly lifestyle (though, rather tweely, he has taken to asteriskising his naughty words, Daily Mail-style).

In order to make this make this work better you have to do two things: (a) focus on giving yourself future options which will tend to give you this desirable position and (b) make sure you subject yourself to volatility so that the options become apparent.  Thus Taleb recommends a volatile approach to all aspects of life: diets with severe fasting, interval training, occasional lifting of extreme weights, and the like.  This leads to the idea of tinkering or bricolage as he calls it, presumably a reflection that DIY for the French is more of a journey than for us objective-driven B&Q customers.  Try things out as long as the downside stays small.  Investigate and (though I don’t think he actually says this) fix a few things that ain’t broke.

One model for all this is life itself.  Volatility causes small mutations to arise for natural selection (or selfish genes) to take over and choose from the rich mix created.  Of course this has the slightly fragile consequence that ‘what kills me makes others stronger.’  No antifragile escape from death then, and indeed, *SPOILER ALERT* the book concludes with the demise of Fat Tony his ever-present Brooklyn wise guy who has an innate comprehension of antifragility and, in one of the book’s many humorous moments, wins a rancorous debate with Socrates.  (The point of the debate was to show that the Socratic method debased valuable notions that are hard to express: turning the unintelligible into the unintelligent.  It is an important feature of Taleb’s non-understood world that you cannot express all of what little you know about it.)

Indeed, as a contrast to modernity, Taleb appeals to the ancients for support.  He believes that a deeper wisdom is possible than that afforded by an (over-)scientised view and an overrationalised approach which leads us to think we can understand and control things better than we can.  So he leads us through a rich cast of Mediterranean, and especially Levantine, philosophers and wiseacres.  Perhaps the most praised is Seneca the Stoic; the aspiration of stoicism is to become indifferent to life’s kicks.  By doing that you adjust your preferences and with them your utility.  At a stroke the downside is reduced (though Taleb omits most mention of Seneca’s exotic sex life which he presumably regarded as generous upside) and you have made yourself antifragile.  A neat, and sensible, trick for an individual which is more difficult for corporations to pull off.

This emphasises the point that antifragility will work best for small units.  They will tend to have the agility which enables them to create options and to recognise the need for decisions.  The small is beautiful idea runs through the book as an antidote to the tendency to corruption by an overconfident and political environment in which individuals do not suffer the downside of poor decisions, or at least a downside which is commensurate with the impact of these decisions on others.  Cue, of course, state-protected bankers.

This leads to another of Taleb’s core ideas: skin in the game and it also brings us back to forecasting.  His position is that you should never believe any expert, or prophet, or forecaster unless they both live by what they are preaching and will be seriously adversely affected if it turns out to be wrong.  I touched on this in the review of Flyvbjerg’s fools and liars paper.  Once again, it strikes me that the ethics of risk analysis is quite a tricky topic given its inherent weaknesses: that it cannot be comprehensive, eg Black Swans, and that it is subjective, hence non-unique, and therefore open to pressure and gaming.

These are severe shortcomings but not obviously to the extent that you should not have a go at it.  Sketching out a view of what might  happen seems to me a useful exercise, especially of you are describing possible chains of events and worst cases and collating information about what happened to others.  And trying to give some idea of the comparative likelihood of some of these chains of events seems like a good idea too, even when these numbers are really just made up and you recognise that there is the possibility of something important you haven’t thought of turning up.

What’s important is for these problems to be recognised and not ignored.  You cannot treat the work as a scientific truth and that’s a matter for both the forecaster and the forecast to.  I have many examples from my own experience of what can happen so I’m expecting this interesting issue to get a bit more mileage on this site.  Don’t consign the risk analyst to economist hell just yet.

Leaving this on one side, what else is one to make of antifragility?  Time and again I found examples of behaviours I could recognise as my own: the avoidance of big downside risks, a lack of commitment to maximising expected utility in spite of its theoretical perfection.  As I write this I am – possibly unwisely – painfully seeing off a toothache without the potentially iatrogenic effects of antibiotics.  Even I have  a portfolio of sorts and some years ago I realised the stupidity of Markowitz-style optimisation and sacked the guy who wanted to charge me maybe 20% of the portfolio to buy and sell in line with it.  As an ex-astrophysicist I remember my incredulity at the use of the diffusion equation to value options, as in Black/Scholes/Merton (the most derided inhabitants of Taleb’s economist anti-pantheon).  So yes, I have a little skin in Taleb-world and I’m minded to take it on board in both my personal and professional life.  So in the next article we’ll look at what it might mean for organisational risk management.

Print Friendly