Defaults in the 90´s: Factbook and Preliminary Lessons

The literature has identified at least five approaches to the determinants of the choice of exchange rate regimes: i) optimal currency area theory; ii) exchange rate policy and the absortion of real and nominal shocks; iii) exchange rate rules as a policy crutch in credibility-challenged economies; iv) the impossible trinity in light of increasing financial globalization; and v) the balance sheet exposure to exchange rate changes in financially dollarized economies. Using both a de facto and a de jure regime classification, we test the empirical relevance of these approaches simultaneously. We find overall empirical support for all of them, although their relative relevance varies substantially between industrial and non-industrial economies. We show that regime choices, as well as deviations between actual and reported policies, can be accurately predicted by a small number of economic and political characteristics of each country. When regimes are correctly characterized, they display no time trend, suggesting that the trends typically highlighted in the exchange rate regime debate can be traced back to the evolution of their natural determinants