This PhD thesis uses data from various experimental settings in order to analyze how incentives shape social and competitive behavior.
The first research project examines social preferences, broadly defined as the concern for payoffs of other people. Such deviations from purely selfish own-money maximizing behavior have been studied extensively in the behavioral and experimental economic literature of the last decades. More specifically, this project deals with the concept of guilt aversion, which postulates that pro-social behavior arises from a psychological cost that an agent suffers whenever he falls short of other people's expectations. The data from this project show that transfers in a dictator game are indeed driven by the incentive of meeting other people's expectations, but only up to the point where these expectations seem reasonable. Hence, pro-social incentives arising from a psychological cost of guilt are relevant, but only under certain conditions.
The second research project uses data from an artefactual field experiment conducted with a sample of professional internal auditors. The research question here is how different incentive schemes for internal auditors influence their performance as well as their objectivity. We let participants perform a real-effort task, the nature of which is akin to an auditing task, under various incentive schemes (individual, tournament, and team incentives) in order to measure the following variables: (i) performance in the task, (ii) objectivity in the way in which participants evaluate the performance of others under different payment schemes, and (iii) fraudulent behavior in the form of dishonest reporting or sabotage. We find that incentive-based compensation increases dishonest behavior: competitive incentives lead to under-reporting of other participants' performance, while collective incentives lead to over-reporting of performance.
The third and last research project deals with the problem of dual incentives, meaning that in certain tournaments higher as well as lower effort levels are rewarded. I provide evidence on behavior in such settings using data from a sports tournament, namely from the National Hockey League and its entry draft system. As I explain, this tournament provides incentives for some teams to lose their games. The data analysis reveals that indeed teams lose more in the present of such incentives. Interestingly, I also find that the response of teams to losing incentives is really systematic in the sense that more lost games cannot only be explained by teams having lower motivation or being disappointed. Instead, this paper is the first to show that there is a concrete strategy behind losing more games. Hence, as in the second research project, here I am again in a way looking at the ``dark'' side of tournaments and on circumstances under which competitive incentives can reduce overall
The different settings examined in this thesis and the different experimental methods employed can provide multi-faceted evidence on how incentives in the form of direct or indirect, monetary and non-monetary benefits and costs influence agents' behavior in individual or team decisions. In terms of the precise thematic focus, this thesis relates to the experimental literature addressing social and competitive behavior.