Currency unions and trade: the effect is large

The impact of a common currency on trade can be grossly mismeasured if countries that belong to currency unions are systematically different from those that do not, and if the relationship between trade and its observable determinants is complex. I argue that such complications are plausible and likely to distort the empirical results of a recent Economic Policy paper by Andrew Rose (Issue 30, 2000: pp. 7–45). Using techniques designed to be robust in this situation, I find that the effects of common currency on international trade are considerably less dramatic and much less precisely estimated.
I have always maintained that the measured effect of a single currency on trade appears implausibly large, but I am not convinced by Torsten Persson's diagnosis and proposed solution. I apply a variety of estimation techniques to a new larger data set, where many more instances of currency union creation and abandonment make it possible to rely on time–series as well as cross–sectional evidence. The results are similar to my earlier ones: the effect of a single currency on trade is large.