March 2009
I revisit the hotly debated topic regarding the possibility of introducing a common currency for East Asia from the point of view of shock symmetry. I first point out a serious problem of the existing studies which use the VAR method with long-run restrictions developed by Blanchard and Quah (1989) in that the signs of the impulse response functions to the same structural shock are not necessarily consistent across the countries. This means that the high (low) correlations of structural shocks do not necessarily imply low (high) costs of a common currency area. To overcome this problem, I apply the VAR method with sign restrictions developed by Uhlig (2005). I used the AD-AS model to impose sign restrictions on the responses of GDP and CPI to demand and supply shocks. One main finding is that demand shocks are significantly positively correlated among almost all East Asian countries. But overall, East Asia as a whole is not suitable for a common currency because correlations of supply shocks are low.