BUDGET DEFICITS AND INTEREST RATES: A COINTEGRATION AND CAUSALITY ANALYSIS

Abstract

This paper argues that existing empirical studies of the relationship between budget deficits and interest rates are vitiated by their failure to take into account the non- stationarity, cointegration, and exogeneity features of the data. We investigate the conditions under which the single-equation framework, typically adopted in the literature, can yield valid inference. We then suggest that, in order to deal with the endogeneity problem, one should use the Phillips-Hansen fully modified estimator, which is asymptotically equivalent to full information methods as long as a single cointegrating vector exists. Furthermore, we employ some recently developed testing procedures for both short- and longrun causality in cointegrated systems, and find a positive relationship between budget deficits and interest rates both in the short and in the long run in the post World War II US economy.