Monitoring versus Gatekeeping

We study alternative mechanisms facing adverse selection and moral hazard, as well as the problems of collusion and free-riding, which are often ignored in the literature. We derive the optimal monitoring mechanism and show that it solves free riding and collusion problems. However, with different types of agents, the optimal mechanism needs to also solve an "assignment problem", which, coupled with the need to generate incentive for monitoring, prevents the optimal monitoring mechanism from attaining full second best efficiency. The paper then considers an alternative mechanism in which some agents are simply given gatekeeping powers: they can either allow or block any investment project. The mechanism allows rent extraction through side payments from investors to the gatekeepers. A gatekeeping mechanism with competing gatekeepers attains first best efficiency, and is also proof against collusion between investors and gatekeepers by construction. We show that the crucial issue for the success of monitoring is whether monitors can be penalized for false reporting. Without this assumption monitoring reduces to gatekeeping. Further, the crucial assumption for gatekeeping to succeed is that gatekeepers behave in a competitive manner. The results provide an explanation for the observed institutional choices: monitoring is typical in informal collectives, whereas government regulation of investment (licensing, issuing permits etc) leads naturally to gatekeeping.