Measuring exchange rate flexibility by regression methods

A new and easily implemented regression method is proposed for generating an index of exchange rate flexibility, whilst simultaneously identifying anchors of pegged currencies. The method can distinguish floats from pegs, including those with occasional devaluations. An annual index is calculated that can be compared with other regime classification schemes, or used directly in empirical research as a measure of exchange rate flexibility. Different categories in the IMF’s de facto classification, and also in the Reinhart-Rogoff classification, are associated with significantly different average values of the index. Further analysis of managed floats shows that they have a strong tendency to track the US dollar