This paper provides evidence of heterogeneous treatment effects on trade for six different types of exchange rate regime transitions, utilizing data on 218,643 country-pair-year observations all together. Previous research mainly focused on the currency union effect on trade and widely assumed that country-pairs with identical observable characteristics receive the same effect. We test the hypothesis of a homogeneous treatment effect empirically and find substantive statistical support for the presence of effect heterogeneity. Moreover, we characterize the probability distribution of effect heterogeneity among observationally equivalent country-pairs by way of a random coefficients model. We estimate selected features of the empirical distributions of the treatment effects for each of the six different types of exchange rate regime transitions under two different sets of assumptions about the probability distribution of effect heterogeneity. Our findings suggest that the immediate effects on trade from switching to an exchange rate regime different from the one in the outset have wide waists and exhibit large positive extreme values. We estimate the probability of two countries to experience an immediate positive effect on the growth of bilateral trade due to the adoption of any new exchange rate regime (tighter or not) to be approximately one-half.