Tail-Risk Perspectives

Abstract

For purposes of risk management, typically used return distributions are asymptotic to zero, the most popular being the Gaussian and power functions. Yet this asymptotic property of tail risk is relatively unexploited in interpreting the most vexing of managerial behaviors, the tendency to obliviousness regarding the most threatening hazards facing the organization. In this article, I explore implications of vanishingly small probabilities in the tail, relate these to the writings of Kahneman and Taleb, and present a novel interpretation of existential risks. This new discernibility interpretation of tail risk is then related to selected cognitive biases—overconfidence, anchoring, and groupthink—widely discussed in behavioral economics and finance. Cognitive biases are seen as diminishing the discernibility of remote probabilities confronted in the negative tail of the returns distribution. In contrast, enterprise risk management (ERM) practices are shown to be discernibility enhancers. Both cognitive biases and ERM, therefore, are consonant with and support the discernibility interpretation of tail risk, thus providing an explanation of some of the most puzzling and egregious errors of omission.