Neutrally stable outcomes in cheap-talk coordination games (Q1581903)

Recall that a strategy is evolutionarily stable if it is a best reply to itself and a better reply to all other best replies than these are to themselves. It is neutrally stable if it is a best reply to itself and a weakly better reply to all other best replies than these are to themselves. This paper examines the weaker of these two concepts, neutral stability, and compares its cutting power with that of stringent noncooperative refinements. An outcome is called neutrally stable if it is the payoff that results when a neutrally stable strategy meets itself. The authors characterize the set of neutrally stable outcomes in cheap-talk \(2\times 2\) coordination games. If the message set is finite, so is this set of outcomes. As the number of messages goes to infinity, the set of outcomes expands toward a countable limit with the Pareto efficient Nash equilibrium as its unique cluster point.