Policy Implications of Second-Generation Crisis Models

The 1992-93 currency crises in Europe and the 1994 crisis in Mexico renewed interest in modeling speculative attacks on government-controlled exchange rates. Responding to events, economists began to develop models of currency crises with multiple solutions in which a crisis could be modeled as an economy jumping from one solution to another. In these models, a crisis need not be motivated by misaligned economic fundamentals observed before the crisis. Instead, the crisis itself becomes a source of exchange market volatility and is the event that triggers altered post-crisis policy. This paper examines one of the new models and finds it implies that raising the cost of devaluation may make a crisis more likely. Consequently, if the observed volatility in currency markets is due to the existence of multiple equilibria, slow convergence to a monetary union, which increases the cost to the government of reneging on an exchange rate peg, may be counterproductive. This conclusion is exactly opposite to that obtained from earlier models.