Emotions and financial risk-taking in the lab: A meta-analysis

Abstract

This paper is a meta-analysis of experimental studies dealing with the impact of incidental emotions (happiness, sadness, fear, and anger) on financial risk-taking, so as to explain traditional heterogeneity of outcomes in the literature. After devising a standard search strategy and including studies that comply with a list of eligibility criteria, we code 114 effect sizes at the treatment level from 26 selected articles, and a battery of moderator variables representing design and sample characteristics. Meta-regressions with adjusted predictions find causal impact of fear on risk aversion, albeit to a small extent. On the contrary, average null effects characterize happiness, sadness, and anger. It also turns out that when studies provide financial incentives, country-level individualism moderates the relationship between emotions and risk-taking by increasing risk propensity. We discuss possible interpretations of our findings.