Antifragile, pro and anti

[amazon_link id=”0141038225″ target=”_blank” ]Antifragile: Things that Gain from Disorder[/amazon_link], the latest tract from Nassim Nicholas Taleb, left me with mixed feelings. It’s interesting, and I find a lot of his argument intuitively appealing. On the other hand, it really needed a thorough edit – the train of the argument is convoluted and there are horribly self-indulgent passages. It could have been significantly shorter, too. On balance, it’s worth a go but lacks the punch of [amazon_link id=”0141034599″ target=”_blank” ]The Black Swan[/amazon_link] or [amazon_link id=”0141031484″ target=”_blank” ]Fooled by Randomness[/amazon_link].

[amazon_image id=”0141038225″ link=”true” target=”_blank” size=”medium” ]Antifragile: Things that Gain from Disorder[/amazon_image]

Which is a pity, because it would be good for the key ideas in [amazon_link id=”0141038225″ target=”_blank” ]Antifragile[/amazon_link] acquired the same traction in popular thought as the importance of fat tailed distributions (hence the frequency of Black Swans) and the fact that basic probability means luck plays a much bigger role in life than we think. Taleb’s main argument is that just as small tremors release tension along a geological faultline, averting a big earthquake, small setbacks play a useful role in economic and social contexts. It is a good thing for the economy as a whole that some firms fail; the policy manipulation that gave us the Great Moderation (the Greenspan ‘put’ of cutting interest rates whenever the markets declined) built up the imbalances that led to the Great Crash. The book gives many examples of contexts in which small stresses play a healthy, error-correcting role, and over-regulating creates the conditions for big errors.

Along the way, Taleb has many swipes at economics, mainly for its insistence on linear models and normal distributions – and I’m someone who thinks that’s a fair cop. The world is self-evidently non-linear, and it’s alarming that so many policymakers cling to the illusion of control they get from linear thinking – pull this policy lever, and that desirable consequence will follow.

There is one cracking story in the book, where Taleb recounts giving a lecture to Société Générale’s top executives on risk, warning them that the bank was taking massively greater risks than they imagined. The reception was hostile, he reports. Weeks later, SocGen had to liquidate $70bn of assets in a fire sale to cover the losses caused by the trades of Jerome Kerviel. At the talk, Taleb says, he had been “heckled relentlessly by Kerviell’s boss and his colleague, the head of risk management.”

There are some general lessons from observing anti-fragility, Taleb concludes. There are three kinds of context – the fragile (negative feedbacks or concavity), robustness (no feedbacks), and antifragility (positive feedbacks or convexity). Be aware of whether you are in a situation where the distribution of outcomes is curtailed at one end – travel times, for example, have little scope to be shorter than expected and much scope to be far, far longer than expected. Is there more upside than downside? How quickly do the outcomes change – does the time taken to drive from A to B increase by a lot more when you add a second hundred extra vehicles on the road than it did for the first hundred extra vehicles? Guard yourself by following what Taleb calls ‘barbell’ strategies – he means dual strategies avoiding the middle way – put your money 90% into safe assets and 10% into very risky ones, so limit your downside risk and create large upside, but don’t put 100% into medium risk assets which could lose you everything if you miscalculated the risk. But he extends the idea. So also, if you want to be a writer, work in a boring job and leave your free hours for writing, rather than taking a job as a creative writing academic, which will suck out your creative marrow with teaching and admin.

This extension points to one of the book’s weaknesses, which is that it extends arguments that make complete sense at the level of probability and asset prices to other areas, from macroeconomics to the rest of life, where they have intuitive appeal but would probably not convince an unsympathetic reader. So I agree with Taleb that there is too much effort to squeeze the variability out of various contexts, with ultimately damaging consequences, but then I thought that anyway without the paraphernalia of anti-fragility to get me there. I enjoyed reading [amazon_link id=”0141038225″ target=”_blank” ]Antifragile[/amazon_link] despite its bagginess, and other Taleb fans will enjoy it too, but I doubt it will have the wider impact of his previous books.

5 thoughts on “Antifragile, pro and anti

  1. Thanks for this, Diane. I followed Taleb’s arguments for this book through his facebook page. He published some passages and related papers as he wrote it.
    I like Taleb – he says things that reassure an outsider (like me). But I think I’ll bump this one down my ‘to read’ list. I could do without the bloat and I think I have the flavour of his arguments already.

  2. I actually had the opposite reaction to the book. His first two books I felt were rather abstract—I left them feeling that I had been presented some interesting ideas but wasn’t really sure what to do with them.

    Antifragile is the only one I walked away from feeling like there were concrete lessons from the smallest (personal, individual) level of analysis to the largest (global, human social system).

    • That’s interesting. I’m glad to have read Antifragile and enjoyed the process. But my taste obviously inclines more toward the very analytical.

      • Mine usually do as well, so I was surprised to find myself enjoying it for those particular reasons. But did you check out the technical notes at the end afterwards? That’s about as analytical as it gets (and frankly beyond my abilities as a reader).

  3. Pingback: Antifragile – an update | The Enlightened Economist

Comments are closed.