This paper attempts to assess the performance of alternative exchange rate regimes in Latin America relative to the benefits they are theoretically supposed to deliver. We will test empirically whether flexible systems allow for better cyclical management, more monetary autonomy and improved control of the real exchange rate. We find that flexible exchange regimes have not permitted a more stabilizing monetary policy but instead have tended to be more pro–cyclical. In addition, flexible regimes have resulted in higher real interest rates, smaller financial systems and greater sensitivity of domestic interest rates to movements in international rates. We also find that flexible regimes tend to promote wage indexation. We show that the revealed preference of Latin America is to allow very little exchange rate movement, even in periods of large real shocks such as 1998. We explain this preference as a consequence of de facto wage indexation and the high proportion of dollar–denominated financial liabilities. The paper then discusses the problems with fixed exchange rates and reviews the current interest in supra–national currencies, including full dollarization.