Capital mobility and sustainability: Evidence from U.S. current account data

We propose an empirical framework that allows us to jointly test for the sustainability of the current account deficit and evaluate the capital mobility thesis by examining the time series properties of the current account. We argue that this approach is more useful than the Feldstein-Horioka (1980) cross-section regression because of its firm basis on the long run inter-temporal budget constraint and of its richer dynamics that allow for a more useful method to evaluate the capital mobility thesis. Based on a century and half of U.S. current account data, we find evidence of current account sustainability and major breaks in the current account dynamics such that adjustment in the current account ‘switches off’ allowing the current account to accumulate at a non-stationary rate. We assess whether periods in which the current account accumulates in a non-stationary way correspond to historical periods believed to have witnessed high degree of capital mobility.