Ⅰ.Introduction

The theory of optimum currency areas had been given a lukewarm reception ever since Mundell initially advanced his framework in 1961. Only recently has it been re-examined. The impetus has been provided，not by his Nobel prize nomination，but by the debate over European monetary unification and the current Asian crisis. It is no exaggeration to say that for some Asian regions and countries，the problem of whether to fix or tofloat the exchange rate is of almost life-or-death importance. Thailand is a good example. Many arguments have been made both for and against a flexible exchange rate system. The major advantages of a flexible exchange rate are a）the increased independence of government policies in stabilizing the macroeconomic fundamentals，and b）the fact that they provide a less painful adjustment mechanism for trade imbalances（Friedman，1953）. The Mundell-Fleming-Dornbush model，however，argues that fixed exchange rates are superior to floating rates because they can reduce trade and investment uncertainties. Whatever the theoretical arguments，a notable feature in this strand of literature is the lack of solid empirical work，excepting that of Flood and Rose（1995）and Baxter and Stockman（1989），which concentrate on OECD countries before and after the breakdown of the Bretton Woods system. Flood and Rose found little empirical evidence that reducing exchange rate volatility from the flexible to the fixed，or increasing volatility from the fixed to the flexible regime compromises the stability of macroeconomic variables. At the end of their paper they call for future research to place greater emphasis on microeconomic factors which may provide insights into the phenomenon of regime-indifferent volatility. Baxter and Stockm