Current account dynamics: A SVAR analysis when the country-specific shocks are correlated at leads

Abstract

The assumption of no correlation of the structural shocks as Blanchard and Quah's identification (BQ) assumes is not suitable when central banks, through discretionary policies or inflation targeting regimes, affect the output growth among other goals. In this study, we analyze the present value model of the current account (PVM) using a three-SVAR specification and a modified BQ to identify three structural shocks: country-specific permanent, country-specific temporary, and global, but allowing the correlation of the domestic ones. Using Australia, Canada, Norway, and the UK, we find that those shocks are correlated and are less volatile than BQ would assume they are. The PVM predictions that hold are: (i) a positive (no) response in the current account to a country-specific temporary (global) shock; and (ii) with the exception of Australia, there is no response in the current account to a country-specific permanent shock. In addition, for all countries, the country-specific temporary shock dominates current account changes but does not dominate net output growth fluctuations, which was a puzzle identified by a prior study. The role of the shock is enhanced by the modified BQ, but even with this enhancement, it still does not hold the most significant role in output variations, as indicated by PVM.