Iterated Revelation: How to Incentivize Experts to Reveal Novel Actions

Abstract

I examine how a decision maker can incentivize an expert to reveal novel actions, expanding the set from which he can choose, without making ex-ante commitments regarding as-of-yet unrevealed actions. The outcomes achievable by any (incentive compatible) mechanism are characterized by the iterated revelation protocol: a simple dynamic interaction where, each round, the expert reveals novel actions and the decision maker adds actions to a shortlist; when nothing novel is revealed, the mechanism ends with the expert choosing an action from the shortlist. Greedy strategies -- where the decision maker optimizes myopically -- delineate the decision maker's maximal payoff achievable by any efficient mechanism.