Is Risk Aversion Really Correlated With Wealth? How Estimated Probabilities Introduce Spurious Correlation

Abstract

Economists frequently focus on correlations between wealth and risk preferences but rarely observe the probabilities needed to test this relationship empirically. These unobserved probabilities are typically estimated via profit or production functions conditioned on wealth correlates, which may leave statistical fingerprints on subsequently-estimated risk aversion coefficients and confound correlations between wealth and risk preferences. Using data from an experiment with observable probabilities, we compare risk aversion coefficients based on true probabilities with those based on probabilities estimated using standard approaches and show how estimated probabilities can change risk aversion coefficients substantially and introduce spurious correlation between risk aversion and wealth.