The Design and Price of Influence

Abstract

A sender with private preferences would like to influence a receiver's action by providing information through a statistical test. The technology for information production is controlled by a monopolist intermediary, who offers a menu of tests and prices to screen the sender's type. We characterize the intermediary's optimal screening menu and the associated distortions, which may benefit the receiver by increasing test informativeness. Because of these distortions, seemingly unfavorable changes in the prior belief may actually benefit the sender. In extensions, we study (i) a stronger intermediary who can commit to a ``coercive'' test to punish non-participation and (ii) a weaker intermediary who cannot control test design but can charge for access.

Tasks

Reproductions