Most of the empirical literature on the relative merits of alternative exchange rate regimes uses the IMF de jure classification based on the regime that governments claim to have, abstracting from the fact that many countries that in theory follow flexible regimes intervene in the exchange market to an extent that in practice makes them indistinguishable from fixed rate regimes, and vice versa. To address this problem, in this paper we construct a de facto classification of exchange rate regimes. Using cluster analysis techniques, we group different regimes according to their behavior along three classification dimensions: the nominal exchange rate, changes in the nominal exchange rate, and international reserves. We compare our results with the IMF classification, and discuss the main discrepancies.