Reconsidering the Trade-Creating Effects of a Currency Union

This paper reconsiders recent empirical evidence found by Andrew Rose that countries adopting a common currency will triple their bilateral trade. We find that this large estimated effect is due to estimation bias arising from missing and/or misspecified time-invariant factors, rather than to the adoption of a common currency. Using the most general specification of time-invariant factors, our results indicate that a common currency actually leads to a small reduction in trade over a five-year period, although this result is not statistically different from zero. We also find that over ten- and twenty-year periods, trade volumes are more than halved by the adoption of a common currency