Ratifiable Mechanisms: Learning from Disagreement

In a mechanism design problem, participation constraints require that all types prefer the proposed mechanism to some status quo alternative. If the payoffs in the status quo depend on strategic actions based on the players' beliefs, then the inferences players make in the event someone objects to the proposed mechanism may alter the participation constraints. We include this possibility for learning from disagreement by modeling the mechanism design problem as a ratification game in which privately informed players simultaneously vote for or against the proposed mechanism. We develop and illustrate a new concept, ratifiability, that takes account of this inferencing problem in a consistent way. Requiring a mechanism to be ratifiable can either strengthen or weaken the standard participation constraints that arise in mechanism design problems.