Making Corruption Harder: Asymmetric Information, Collusion, and Crime
将刑事调查建模为委托人-代理人-监督者问题，研究委托人能否通过随机化监督者激励来引入信息不对称以减少合谋，并提供一个仅需未验证报告的政策评估框架。
We model criminal investigation as a principal-agent-monitor problem in which the \nagent can bribe the monitor to destroy evidence. Building on insights from Laffont and \nMartimort (1997) we study whether the principal can profitably introduce asymmetric \ninformation between agent and monitor by randomizing the monitor’s incentives. We \nshow it can be the case, but the optimality of random incentives depends on unobserved \npre-existing patterns of private information. We provide a data-driven framework for \npolicy evaluation requiring only unverified reports. A potential local policy change is \nan improvement if, everything else equal, it is associated with greater reports of crime.