Interlinking payment systems and trade flows

This paper provides the first causal estimate of the economic impact of interlinking payment systems across countries. We exploit a new dataset of payment systems interlinking initiatives, which identifies over 2,000 connections, and employ standard gravity methods to estimate their impact on trade flows. Consistent with trade costs theory, we find that inter-connected countries have around 4% higher trade volumes, roughly half the effect of a trade agreement and a quarter of the effect of a common currency area. Our results isolate the average effect on trade, of directly connecting fast payment systems, net of country pairs already accessing the correspondent banking network. The estimated impact is larger for payment systems that allow wholesale transactions, those that link small countries, which, typically, are less connected to the correspondent banking network, and for geographical areas that face high cross-border payment costs. This suggests that the benefits from interlinking are derived from reduced cross-border trade costs. Our findings are causal - proved by parametric and semi-parametric estimators - and robust to numerous additional controls, including exclusion of the largest interlinked country group, the euro area.