The role of emotions in individual and strategic choice situations

It has often been argued that emotions play an important role in several manifestations of an individual’s life. Testing whether this is actually true and, if so, identifying how exactly emotions shape decision making is therefore crucial for economists who want to get a better understanding of behavior. This thesis aims to demonstrate that emotions matter in explaining both individual behavior under risk and strategic behavior in bargaining situations. In the first chapter, I experimentally investigate the impact of four discrete emotions (joviality, sadness, fear, and anger) on risk preferences. In order to do so, I fit two models of behavior under risk: the Expected Utility model and the Rank Dependent Expected Utility model, assuming several functional forms of the weighting function. I prove that all four emotional states instigate more risk-seeking behavior than a neutral emotional state. In the second chapter, I extend this field of research by shedding new light on the impact of self-reported happiness on the willingness to invest in risky assets, which can be considered as a proxy for financial risk taking, using data on Italian households for 2008. Considering both a self-assessed measure of financial risk and several objective measures that are inferred from observed portfolio shares, I show that the happier individuals are, the more willing they are to invest in risky assets, i.e., to take financial risks. Finally, moving from individual to strategic decision making, the third chapter proposes a novel 3-person ultimatum game experiment whose purpose is to explore whether proposers’ behavior depends on the way in which the responders’ decisions are elicited (via the play method or via the strategy method), and to shed light, in a straightforward fashion, on the role of emotions in explaining such differences. I prove that, even though the play and the strategy methods make responders feel different emotional states and proposers precisely anticipate it, the latter' behavior is not affected by their beliefs. This provides good news for experimental economists: the two response methods do not significantly bias proposers' behavior.