A dynamic analysis of the determinates of the US current account deficit

Abstract

This article examines the determinates of the US current account deficit. We decompose the trade balance into its import and export components. This allows us to model responses of imports and exports separately. The analysis is conducted with a semi-structural Vector Error Correction Model (VECM). Our results show that a US stock market improvement, a depreciation of the US dollar, an increase in US interest rate and economic growth of the Japanese economy all will help reduce the US trade deficit.

Notes

¹ The trade balance is by far the largest component of the US current account. Fluctuations in the trade deficit are the primary cause of fluctuations in the current account.

² We do not use data from China because quarterly Chinese data is not readily available for the time period under consideration.

³ For a survey of the literature on the topic, see Obstfeld and Rogoff (1995, 1996).

⁴ Financial variables do have an impact on real variables through the lags in the VAR.

⁵ Reordering the variables within the recursive block does not change the estimated impact of shocks to these variables on IM, EX and SP.

⁶ The RATS program used to estimate Equations 1–3, and the coefficients in the recursive block, was provided by James S. Fackler.

⁷ Expanding the lag length to six leaves only 16 degrees of freedom in each VECM equation.

⁸ IRFs for the impact of shocks to SP on EX and IM (not shown here) indicate that shocks to SP tend to have a positive impact on exports and an insignificant impact on imports.

⁹ Pesaran and Shin's (1998) generalized IRFs compute shocks with each model variable in turn placed first in a Choleski decomposition. For more details, see Estima (2002).

¹⁰ We have also estimated IRFs using a Choleski decomposition and the ordering Y, Y*, REX, R and R*, IM, EX, SP. These IRFs also produce policy implications consistent with the IRFs estimated with our semi-structural approach.