The distinctive feature of an open economy is its ability to borrow and lend in international capital markets in response to cyclical disturbances. As a result, domestic income and spending can be different as countries run current account surpluses and deficits. This behaviour is the result of consumption smoothing and capital flows to those countries where investment opportunities exist. These two forces shape the dynamics of saving and investment rates, and thus the response of the current account.

In Discussion Paper No. 1280, Graham Elliott and Research Affiliate Antonio Fatás analyse the transmission of productivity shocks across countries and how the responses of investment and the current account differ depending on the degree of propagation of the shocks. The authors explore both issues by estimating a structural model for Japan, the United States and Europe. They postulate, as an identifying assumption, that the propagation of shocks is proportional to trade. The paper finds that there is a strong asymmetry in that shocks to the United States propagate quickly to the other two economies, while European and Japanese shocks have little impact on other countries' productivity. Moreover, it is found that productivity increases lead to domestic investment booms and current account deficits. Investment in other countries tends to react positively to productivity shocks, even when the shock is purely national. This second result contradicts the predictions of a standard open-economy model with perfect capital mobility where, in response to country-specific shocks, domestic and foreign investment should move in opposite directions. The paper also finds quantitative differences among the three countries in the response of the current account. These differences are not related to the global or idiosyncratic nature of the shocks.