By Akhil Raj Gupta
Nassim Nicholas Taleb is probably one of the least misunderstood authors in the world, in spite of the incisiveness of his ideas or the depth of his knowledge. He is often pilloried by supposed stalwarts of the economics and finance establishment whom he has relentlessly attacked for their complex models based on faulty math. His magnum opus is the Black Swan, which is premised on what he calls the inductive fallacy or the turkey problem.
Consider the statement
- All swans are white
Taleb argues that the above statement is fundamentally flawed because it disregards the probability of the extremely rare event i.e. the existence of a black swan. Any conclusion drawn from past data (when all swans were white) mistakes absence of evidence with evidence of absence. Moreover, any confirmation that is positive (For eg The world is a safe place) is biased. We should rather attach more significance to what we dont know rather than carry an illusion of understanding. For instance, if we were to spot a black swan or a swan of any other colour, we can make the following statement (and only that much) with certainty
2. All swans are not white
In technical terms, he directs virulent criticism at any economic models that are based on the normal (or Gaussian) distribution. In a bell-curve stylized world, the probability of a rare event decreases at an increasing rate. For example The jump in probability from a six sigma to a seven-sigma event is significantly larger than the gap between a five and a six-sigma event (sigma standard deviation from the mean). If instead one were to assume a fractal distribution with the appropriate power law (statistical-self similarity), one would have a more realistic idea of how events in the real world work, thereby leading to better policy decisions.
However all of this boils down to a higher-order problem of metaprobability. This implies that since we have not sampled all events in the world (and never will), any model must necessarily make an a priori assumption about how these events will be distributed. Therefore, we have to assign probabilities to probabilities and that is where one can go horribly wrong, whether normal or fractal. This scope for computation demarcates the fine line between risk and uncertainty.
Taleb therefore argues that economics mostly lies in the domain of uncertainty because rare events are by definition, incomputable. He advocates not only a richer understanding of randomness, but also the need to focus more on making our institutions (financial or otherwise) robust to these Black Swans. Why? Because economic phenomena lie in Extremistan (uncontrolled randomness) v/s physical phenomena that lie in Mediocristan (controlled randomness). If we were to place 100 of the richest people in the world in the same room, a majority of the wealth would still be concentrated in a handful of the super-rich. This skewness is not observed in the distribution of weight for example, as the heaviest person cannot account for more than a small proportion of the total.
[su_pullquote]It as a must-read for anybody who wants to understand how to live in a world that we dont understand.[/su_pullquote]
As a writer, his style is extremely enjoyable and his ideas are interspersed with platitudes about his own experiences as a quantitative trader and innumerable philosophical references. I recommend it as a must-read for anybody who wants to understand how to live in a world that we dont understand.
[su_divider style=”double”]So Marcos Rojo might have some explaining to do when wife Eugenia Lusardo asks why he quite is so excited in this Instagram post.[/su_divider]
Akhil is currently in his second year at college, pursuing a Bachelor of Arts degree in Economics (Hons) at Sri Ram College of Commerce, University of Delhi.