Essays in Behavioural and Experimental Economics: Emotions, uncertainty and cooperation

In the present doctoral thesis I apply the experimental method in the context of two lines of research in Economics, the former investigating the role of incidental emotions in decision making under risk, whereas the latter shedding light on the impact of communication on cooperation. The dissertation consists of three sections. In the first chapter I conduct a laboratory experiment in order to study the effect of incidental sadness and happiness on risky decision making. An emotion induction procedure is the treatment variable of a between-subjects design where two sessions aim at eliciting either sadness or happiness, respectively. Two further groups are characterized by neutral conditions and serve as baseline. After a manipulation check verifies the validity of the induction procedure, I use a multiple price list `a la Holt and Laury (2002) to elicit individual risk preferences in the context of a lottery-choice task. The analysis reveals that both sadness and happiness promote greater risk aversion with respect to neutral conditions, a result which might be moderated by the risk elicitation task. Therefore, as effective explanation I propose the theory of ego depletion, whereby regulating emotions so as to subsequently process information consumes a limited self-control resource, which is needed to take risks as well. The second chapter is a meta-analysis of experimental studies on the same topic, so as to explain traditional heterogeneity of outcomes in the field. After performing an advanced search in Google Scholar and filtering out studies that do not match a list of selection criteria, I include 16 studies from which 46 observations are drawn at the treatment level. At this point, I code a set of moderator variables representing experimental protocols and calculate Cohen (1988)’s d effect size as dependent variable of a weighted least squares (WLS) regression where larger studies are given more weight. Among the results, which are robust to different techniques for computing standard errors, I find that emotions induce higher risk aversion when a multiple price list `a la Holt and Laury (2002) is used in place of stated preferences methods, as well as in case the risk elicitation task is framed as an investment decision instead of an abstract choice. Given the variety of procedures employed in this type of experiments and in the absence of a tailor-made game to answer such research questions, I recommend faithful study replication as preferential path in order to investigate the influence of emotions on risky decision making and ensure comparability. The third chapter offers evidence on the impact of communication on the provision of public goods whose quality is uncertain. I run a laboratory experiment with two treatments, where the control variable is pre-play communication in the form of unrestricted text chat. A binary threshold public goods game with four-person groups and threshold of three is at the core of the design, the main novelty lying in the provision mechanism with ambiguity. Moreover, a private signal for the actual value of the public good is provided, before the contribution decision. In accordance with related literature, I find that communication significantly increases public good provision by reducing inefficiency that comes from wasteful undercontribution. Nevertheless, the players in the chat treatment seem to neglect the free-rider issue and often end up overcontributing, in contrast with previous scientific findings. After chat analysis, I propose the pursuit of symmetric payoffs within the group as original explanation of the massive overcontribution, in addition to group identity generated by the partner matching and the common fate hypothesis. Since the players prefer to minimize ambiguity than to maximize the group earnings, I finally speculate that under uncertainty satisficing is more salient than optimizing.