Robust persuasion of a privately informed receiver

Abstract

This paper studies robust Bayesian persuasion of a privately informed receiver in a binary environment, where an ambiguity-averse sender with a maxmin expected utility function has limited knowledge about the receiver’s private information source. We develop a novel method to solve the sender’s information design problem. Our main result shows that the sender’s optimal information structure can be found within the class of linear-contingent-payoff information structures. We also fully characterize the sender’s optimal linear-contingent-payoff information structure and analyze the impact of ambiguity on the sender’s payoff.