This paper uses matching econometrics to extend the literature investigating the impact of preferential trading arrangements (PTAs) on goods trade flows. Heterogeneity in PTAs is accounted for through a “provision count index” derived from data provided in a recent World Bank study. PTA formation now involves two separate, sequential decisions—first, whether two trading partners should form a PTA, and second, if they do, how broad that agreement should be. We find our explanatory variables are significant for both decisions, but often have opposing effects on each. Using matched PTA and non-PTA groups of country pairs, we estimate a dose–response function which indicates that arrangements with few provisions and arrangements with many provisions do not appear to have a significant impact on goods trade flows between their members. PTAs in an intermediate range are shown to have a significant positive effect. We then relate these outcomes to the actual content of the PTAs using the concept of “provision intensity.”.