Essays in information economics: optimal regulation and the fight against corruption.

Authors
Publication date
1996
Publication type
Thesis
Summary The first part of the thesis focuses on the normative aspects of regulation. We analyze the gains from coordinating decisions in the presence of sovereign agents when transfers are socially costly. We consider a general framework with incomplete information in which multiple agents each induce externalities on the activity of others and have reservation utilities that depend on the private information parameter. When the second-order optimum is such that the regulator grants informational rents to the most efficient agents, centralized coordination reduces aggregate activity. However, coordination induces more activity when rents are (optimally) allocated to the most inefficient agents. We study several applications: the optimal structure of an industry, the coordination of r&d activities, and the allocation of non-pecuniary benefits within a firm. The second part deals with the problem of corruption. First, we study the effectiveness of measures to combat corruption when it can spread within an organization and when individuals have private information about their cost of being corrupted. It is familiar to think that the level of corruption decreases if wages are increased and control over individuals is strengthened. However, we show that in a system of internal promotions, these measures can have perverse effects on the incentives of some individuals to remain honest and can be socially harmful. We then characterize the optimal bribe offered by a client to an agent when the former ignores the latter's propensity to be corrupt. In a dynamic setting, individuals who plan to be honest in the future have a different valuation of bribes than those who plan to be corrupt. For this reason, there is an interval of stationary equilibria over which the bribe takes its values. This set shrinks when corrupt agents have a lower active life expectancy in the organization than honest agents.
Topics of the publication
  • ...
  • No themes identified
Themes detected by scanR from retrieved publications. For more information, see https://scanr.enseignementsup-recherche.gouv.fr