How to exit from fixed exchange rate regimes?

This paper improves upon the recently developed literature on exits from fixed exchange rate regimes in three ways: (1) It allows for two indicators for post-exit macroeconomic conditions, the change in the exchange rate and the change in the output gap; (2) it tests whether the distinction between orderly and disorderly exit is statistically justified, and concludes that it is not; (3) it deals with the sample selection problem. The results, subject to extensive sensitivity analysis, suggest that post-exits are better when depegging occurs in good macroeconomic conditions - an unnatural move for most policymakers - when world interest rates decline and in the presence of capital controls. Importantly, 'good' macroeconomic policies do not seem to help with post-exit performance.