Exchange rate regimes, capital controls, and currency crises: Does the bipolar view hold?

This paper empirically examines the link between de facto exchange rate regimes and the incidence of currency crises in 84 countries from 1980 to 2001 using probit models. We employ the de facto classification of Reinhart and Rogoff (2004) that allows us to estimate the impact of relatively long-lived exchange rate regimes on currency crises with much greater precision. We find no evidence that, as the bipolar view argues, intermediate regimes have a significantly higher probability of currency crises than both hard pegs and free floats. Using the combined data of exchange rate regimes and the existence of capital controls, we also find that hard pegs with capital account liberalization have a significantly lower probability of currency crises than intermediate regimes with capital controls and free floats with capital controls. Hence, the bipolar view does not strictly hold in the sense that intermediate regimes are significantly more prone to currency crises than the two extreme regimes. However, the fact that hard pegs with capital account liberalization are substantially less prone to currency crises is worthy of note.