Wednesday, March 30, 2016

Fat tails and geopolitical risk - Why should you care?



It may seem that the world is gripped by terrorism, but there is a school of researchers who suggest that the world has gotten more peaceful and that violence is trending down. There certainly has been something of a "Long Peace" with no world conflicts in more than 70 years, yet it may be hard to draw conclusions on whether the trend is toward non-violence. Leave it to Nassim Taleb and his co-ahtor Pasquale Cirillo to provide some useful perspective on this issue with a math foundation.

Employing the tools of extreme value mathematics with fat-tails distributions the authors provide some much needed analysis on this critical research topic. Their conclusion is that we just do not have enough data to draw any conclusions on the trends in violence around the world. It is fine to be hopeful, but there is little evidence to support a good news story. See their working paper, "What are the chance of a third world war?" and their paper "On the statistical properties and the tail risk of violent conflicts" in Physica A.

Ins simple terms, the tails of the distribution drive the mean and with fat-tails it is hard to make generalizations. You need more data to assert there is no black swan than to assert there is a black swan and the data are too inclusive to make any broad statements. The second paper which focuses on the math behind the fat-tailed problem of violence shows that without too much work, you can find that there is little support for any conclusions about the trends in violence.


No investor should get comfortable with any talking head about trends in violence or a chance of a major war. It is all idle speculation that may give people a false sense of calm. Similarly, we cannot say that the world is getting worse. There is not enough evidence for either case.Tail events are likely and do occur which ensure that any general conclusions are bankrupt. 

Don't go out and become a survivalist but a realist will say that bad outcomes are possible regardless of the sophistication of society and portfolio structures should include the chance for extreme dislocations. A precautionary principle suggests that you should prepare for rare negative events.

No comments: