Credibility and Stabilization

Abstract

The paper deals with stabilization as a one-shot problem to ask what “credibility” means in a world where programs will succeed with probability one. A model is spelled out where the equilibrium program has some ex ante probability of failure. The model draws attention to the factors that raise or lower the probability of success of a stabilization program and thus offers a positive theory of stabilization. Reputation that emerges as a central element in repeated games has no role in this model.