Endogenous Lying in Economic Environments

Information economics typically assumes that agents lie when in their best interests. For instance, in mechanism design, a key part of the model description is "incentive constraints" that ensure an agent has no incentive to lie. In practice, however, people face lying costs, including the physical costs of fabricating information, psychological costs, and costs associated with the risk of being caught (e.g., penalties). This proposal aims at a better understanding of how lying costs evolve in dynamic relationships, and how relationships between economic agents are affected by lying costs that evolve with time. A key possibility is that an agent who lies is desensitized to the costs of lying and so faces lower psychological costs of lying again. The project aims at studying these questions theoretically, by developing and analysing models of dynamic relationships relevant to a range of applications (including managerial compensation, tax evasion, regulation, insurance fraud, and incentives in research). One focus is dynamic contracting models where agents' lying costs fall with past lies. Another is dynamic models of monitoring, where the monitor is an active player and the risk of being caught in a disobedient or untruthful act evolves due to the dynamic strategy of the monitor. Apart from illuminating the above economic questions, the research envisaged will aim at advances in theoretical areas such as dynamic agency and dynamic monitoring (inspection games).