Social norms and beliefs about collective action, constitutional compliance and intra-gender productivity: An experimental investigation

Abstract

This cumulative dissertation has two goals. First, it investigates the interaction between social norms and beliefs and individual behavior. Secondly, it introduces new tools for the study of social norms and beliefs. The first chapter acts as an introduction and a summary of the methods and the findings. The second chapter contains the first paper of the dissertation. The third chapter contains the second paper and the last chapter the third paper of the dissertation. The second chapter discusses how monetary incentives influence whether citizens challenge an executive who abuses its authority for personal gain. During a lab experiment, I vary the costs and benefits of challenge and introduce a reward for loyalty to the executive. Participants who cannot monetarily profit from challenging challenge the executive who abuses its power. When the cost of challenge or the reward for loyalty increase, a participant who can monetarily profit from challenging is more likely to challenge. Contrary, the behavior of a participant who cannot monetarily profit from challenge does not change when the cost of challenge or the reward for loyalty increase. These results indicate that participants are partially driven by non-monetary motives, when they oppose an abuse of power. The third chapter uses simulations to determine how constitutional enforcement influences the probability that unamendable constitutional provisions will become unpopular with the passage of time. The agents in the model organize their financial activities based on the society's laws, which are written by elected legislators. When the legislator is not stopped from violating the constitution, the probability that the next legislator will also violate the constitution increases. In contrast, the existence of mechanisms which stop violations (e.g. judicial review) significantly reduces the probability that a legislator will come to power who will try to violate the constitution. Constitutions which start with constraints to the legislator face a lower probability that they will be violated or that a legislator who wants to violate them will come to power, even when the constraints are removed later in time. The literature divides gender discrimination into two types: taste and statistical discrimination. The third chapter disentangles the two using an online experiment. The participants are paid when they guess correctly the winner in a number of mixed-gender opponents' pairs. Before they submit their guesses, they learn the opponents' genders, ages and education. After submitting their guesses, they decide who in each pair should get a bonus knowing the opponents' genders, education and score in the competition. On the one hand, I find that most participants believe that men have higher scores on average compared to women. Yet, I find no further evidence that the participants statistically discriminate against women. On the other hand, women are less likely compared to men to get the bonus when they have a higher score than their male opponents. Further analysis shows that the women who are the most likely to win are the least likely to get the bonus. In other words, the participants taste discriminate.