On Sanctions and Signals: How formal and informal mechanisms produce compliance

Abstract

Why people comply with rules, why they contribute to public goods and why they behave prosocially in general is a fundamental question of social science. In the tradition of Gary Becker and the chicago school, economists have traditionally considered punishment by the authorities as the main or sole reason why people would comply with the law and contribute to public goods. In this thesis I argue that this model is importantly incomplete and leads to lopsided or even mistaken policy advice. I stress the importance of social interactions between agents and apply game theoric examples to show how the standard model can be enriched. In the second chapter, I survey the empirical literature, both experimental and econometric, on the deterrence literature. From this review I conclude that the literature does not demonstrate a robust effect of deterrence. I then review theoretical work in which sanctions interact with social norms or long-term processes of preferences formation. In such models deterrence often does not have the straightforward effect that it has in standard theory. The chapter concludes with an example: a model of crime in neighborhoods where signaling is important. I show that in this case, the threat of police violence may be counterproductive on its own, but can be useful in combination with other, softer approaches. The third chapter departs from the fact that the population of contributors to a public good consists of a mix of reciprocal and selfish agents, an assumption borne out by much experimental evidence. I then show that if there exists a government or authority that is superiorly informed about the fractions of these types in the population, a policy of harsh sanctions may convey that there are a lot of bad types in equilibrium. As a result, equilibrium sanctions will generally be lower then they would be under symmetric information. In the fourth chapter, I report the results of a laboratory experiment aimed to test if sanctions can indeed have a signaling effect. In accordance with the signaling hypothesis I find that `endogenous sanctions' tend to make people more pessimistic, especially those who were optimistic at the start of the game. In the last chapter, I model an alternative approach to compliance. I consider the widely reported fact that the possibility to participate in a decision making procedure tends to raise voluntary compliance with authorities, even if the actual decision is not beneficial to the agent. I show that the introduction of a decision making procedure in which an agent can change a decision of the policymaker with some probability, can be a signal of altruistic motives of the policy maker towards the agent. This means that even if she does not change the outcome of the decision in practice, the agent trusts the policy maker to treat her well in the future, and will engage in more voluntary compliance. In the Epilogue I add some remarks on the potential of participatory decision making as an alternative policy tool to the standard economic command and control framework.