Motivating Loyal Bureaucrats in Sequential Agency
Dongsoo Shin (Santa Clara University); Alexander Rodivilov (Stevens Institute of Technology)
Abstract
We study a two-stage adverse selection model in which a legislature and a bureaucrat make their decisions sequentially. In the first stage, the legislature chooses an amount of transfer to the bureaucrat for setting up and implementing a public project, and also collects information on the public's valuation of the project. In the second stage, using the information from the legislature, the bureaucrat chooses the project size and implements the project. We show that the legislature's transfer to the bureaucrat trades off using the collected information and exploiting the bureaucrat's loyalty to the public. Our result suggests that when the bureaucrat is loyal enough, it is optimal to induce him to make his decision independent of the public's valuation of the project.