Evolutionary Foundation for Heterogeneity in Risk Aversion

TL;DR

This paper explores the evolutionary origins of risk aversion in populations facing aggregate risks, showing how heterogeneity in risk preferences can maximize long-term growth and can be modeled with a simple distribution of risk aversion coefficients.

Contribution

It introduces an evolutionary framework explaining heterogeneity in risk aversion and demonstrates how a simple distribution of constant relative risk aversion can approximate optimal behavior.

Findings

Heterogeneous populations can maximize growth by balancing risk preferences.

Least and most risk-averse agents are indifferent at specific risk levels.

A uniform distribution of risk aversion coefficients between zero and two approximates optimal behavior.

Abstract

We examine the evolutionary basis for risk aversion with respect to aggregate risk. We study populations in which agents face choices between alternatives with different levels of aggregate risk. We show that the choices that maximize the long-run growth rate are induced by a heterogeneous population in which the least and most risk-averse agents are indifferent between facing an aggregate risk and obtaining its linear and harmonic mean for sure, respectively. Moreover, approximately optimal behavior can be induced by a simple distribution according to which all agents have constant relative risk aversion, and the coefficient of relative risk aversion is uniformly distributed between zero and two.