The Black Swan | Nassim Nicholas Taleb | Book Review

By Vijay Shah

Key arguments:

In The Black Swan, widely regarded as one of the most important books of the past few decades, Taleb makes the argument that economists use oversimplified approaches to analysis, leading to a more fragile system prone to negative shocks, by failing to appreciate that most of the world today is fundamentally more exposed to fatter tails.

He makes this point using the concepts of Mediocristan and Extremistan. Mediocristan refers to data in which one event does not significantly alter the entire set (e.g. If, while analyzing the weights of a set of 1000 individuals, you eventually find 1 really heavy outlier, it is still unlikely to significantly affect your overall analysis with regards to mean, std. dev, etc.). On the other hand, Extremistan refers to the set of events with the opposite characteristic (e.g. after years of positive stock market returns, one shock event such as GFC could have the potential to wipe out all the gains up to that point). With an increasingly interconnected world, things tend to grow more multiplicatively (e.g. Pandemics), and thus fall more into the realm of Extremistan.

This reality makes simplified statistical analyses using past data increasingly pointless — to paraphrase an interesting analogy from the book: A turkey analyzing data of the trends of its interactions with humans would become increasingly convinced in the hypothesis that each day, the human feeds it. It is ironically at the point where it would have the biggest sample of this data, and thus be the most convinced of the nature of the relation, that it ends up killed.

Taleb relates this construct to the concept of “the black swan” as being an event that is unpredictable not because it is random, but because of our narrow approach. Put simply — the absence of evidence is not the evidence of absence (just because you have not seen a black swan, does not mean none exist).

Why you should read this book:

Taleb’s point about academics and practitioners oversimplifying the world and deluding themselves into believing they are being scientific while misusing statistics*, is particularly relevant given his book was published in 2007, during GFC. However, the initial response of a lot of institutions to Covid-19 is a good example of the failure to heed his teachings continuing to plague us. Taleb advocated mandatory masks back in January — his argument being that the lack of evidence of airborne spread based on the little data available to that point, was not enough to eliminate the possibility — thus making WHO’s initial guidelines that masks were unnecessary a stupid risk, given the multiplicative nature of pandemics.

Closer to home — Taleb said of B-Schools that they are a place “where people pay tons of money to learn to run a business from people who are only there because they don’t know how to run a business”. He would probably refer to a lot of what we learn at IIM A as concepts propagated by “IYIs” (Intellectual, Yet Idiots). Taleb is harsh but extremely intelligent- Academia can often be an echo chamber, and his views are admirable for their ability to compel the reader to think from a different perspective.

Check out this Twitter thread by Amihai Glazer (and his feed in general) for an example of a Yale-trained Economist demonstrating how hilariously poor a grasp on statistics many economists have: https://twitter.com/nntaleb/status/1279954325087891464