Do Fixed Exchange Rates Cause Greater Integration?

Abstract

A classic argument in favor of a fixed exchange rate regime (ERR) has been the promotion of international trade between the pegging country and its base country. Results from previous literature point to a significant and highly positive effect of adopting a fixed ERR on bilateral trade between any given country pair. In this paper, it is argued that these results should not be interpreted as causal effects, since countries do not typically choose their ERR independently of their trade flows. This source of selection bias can be greatly reduced by using the matching approach in estimating treatment effects. Estimates of the effect of fixed ERR using this procedure are close to the ordinary least squares estimates reported in the literature, suggesting that there is little bias in these conventional estimates. These findings are robust to using different propensity score matching methods.