The Present Value Model of the Current Account Has Been Rejected: Round Up the Usual Subjects

Testable implications of the basic intertemporal model of current account determination are almost always rejected by the data. We confirm these rejections for a sample of post-war Canadian data, then account for them by calibrating and simulating a small open economy, real business cycle model. Bayesian Monte Carlo experiments reveal that several of the "usual suspects", in particular the costs of risk premia and shocks to fiscal policy and the world real interest rate, are important sources of the rejections observed in the data, while other suspects are unimportant.