Collaborative Research: Sequentially Optimal Mechanism Design

California Institute Of Technology, Pasadena CA

Investigators

Abstract

This award funds research in the economic theory of mechanism design. This area of economics focuses on understanding how incentives affect economic outcomes, with the goal of designing effective and efficient payment and reward plans. The project aims to expand mechanism design theory by providing a new tool to study the problems caused when long-term commitment to a contract is unrealistic. For example, online retailers, insurance companies, and banks interact with their customers over time. As these firms learn about their customers, they may want to change contract terms by making personalized offers to some customers. In the same way, governments may sign agreements about sovereign debt, but bailouts and renegotiation may change the terms of the original agreement. The new method has promise for new applications of this important area of economic theory, including a range of applications from debt and mortgage contracts, monetary policy, the design of online platforms, online privacy, and auctions for advertising sales. As a result, the research could result in more effective management methods for businesses and government policy makers. Progress on optimal mechanism design under limited commitment has been hindered by a lack of a tractable methodology. The team will develop a new methodology based on the idea that a mechanism should encode not only the rules to determine the allocation, but also the information the designer obtains from the interaction. This means that how much the designer learns becomes an explicit part of the design. The project has three key parts. First, the team will develop this new tool, which will be akin to the use of the revelation principle in classical mechanism design. Second, the team will use the method to characterize optimal trading mechanisms in infinite horizon settings. The characterization provides a solid foundation to Coase's conjecture and the solution method can serve as a prototype to characterize optimal fiscal policy or social insurance when governments have limited commitment ability in infinite horizon settings. Third, the team will examine the optimal design of information collection policies (for example, the use of cookies) and the resulting issues of transparency and privacy that arise when firms collect information about consumers. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.