Research
Information Alliance in Common Agency
Rafael Lima, Humberto Moreira, Alexander Rodivilov, and Dongsoo Shin
Abstract
We study an agency model with two independent projects each of which is owned by
a principal. Under exclusive agency, each principal uses a different agent, whereas
under common agency, all principals use the same agent. Aproject’s outcome depends
on the agent’s hidden effort. We show that, even in the absence of physical linkages
between the projects, a principal benefits from providing information on her project
outcome to the other principal under common agency (rather than benefiting from
receiving the information), whereas no such benefits arise under exclusive agency.
That is, under common agency, an information sharing principal can take advantage
of the other principal’s using the shared information. As a result, sharing information
becomes each principal’s dominant strategy, provided that the other principal uses the
shared information in a Pareto-optimal way. We show that using a common agent is
optimal when information can be shared.