People often say in this literature that people overestimate thethe riskiness of some process. And what they mean by that, i think, is that they don't follow the expected value. This reminds me of another conflation that people make in their minds, that improving returns with lower risk. So if you worry just about the risk of rui ruin improvement, any improvement that, because it doesn't have a track record, brings some uncertainty ntot them, increases that risk of ruin. But we have to be very careful well. The left tail reacts vastly more to the scale of the distribution,. which for a galcian or log normal would be the sigma, the standard
Nassim Nicholas Taleb, author of Antifragile, Black Swan, and Fooled by Randomness, talks with EconTalk host Russ Roberts about a recent co-authored paper on the risks of genetically modified organisms (GMOs) and the use of the Precautionary Principle. Taleb contrasts harm with ruin and explains how the differences imply different rules of behavior when dealing with the risk of each. Taleb argues that when considering the riskiness of GMOs, the right understanding of statistics is more valuable than expertise in biology or genetics. The central issue that pervades the conversation is how to cope with a small non-negligible risk of catastrophe.