Asymmetric Effects of Government Spending: Does the Level of Real Interest Rates Matter?

Abstract

This paper examines empirical issues on asymmetric effects of government spending. Increases in government spending under low real interest rates are not associated with the same increases in future tax liabilities as those under high real interest rates. Consequently, the negative impact from the Ricardian effect is smaller with lower real rates. Our empirical work on US data, using threshold regression models, provides new evidence that an expansionary government spending is more conducive to real activities when real rates are low. We also find asymmetric effects on interest rates and threshold effects associated with substitution between financing methods.