Elsevier

Economics Letters

Volume 118, Issue 2, February 2013, Pages 300-303
Economics Letters

Fiscal deficits and mean reversion in real exchange rates

https://doi.org/10.1016/j.econlet.2012.11.022Get rights and content

Abstract

The mean reversion of real exchange rates in G5 countries depends on both countries’ fiscal deficits/surplus in a nonlinear way. When the fiscal policy pushes the real exchange rate to be deviated further away from the equilibrium level, the mean reversion process is faster.

Highlights

► A fiscal policy variable is allowed to act as a conditioning variable in modeling exchange rate movements. ► Fiscal deficits/surpluses impact the mean reversion of real exchange rates in a nonlinear manner. ► The additively separable response of real exchange rates to fiscal deficits is insignificant.

Introduction

The effect of monetary policy on exchange rates has been extensively explored. Early theoretical works (e.g. Frenkel and Razin, 1986) demonstrated that both monetary and fiscal policies could have substantial effects on floating exchange rates, although little serious attention has been paid to empirically explore the role of fiscal policy in foreign exchange markets. Furthermore, the existing empirical work typically views fiscal policy as a separate tool from monetary policy and consider its stand-alone effects on exchange rates as a direct information variable. More recently, the theoretical literature (e.g. Sargent, 1999, Dixit and Lambertini, 2003) has underscored the potentially complex interaction between monetary and fiscal policies and the fact that fiscal deficits can be binding constraints on the effectiveness of monetary policy.

Our research fills the gap in the empirical literature by allowing for a fiscal policy variable (i.e., the level or the change in fiscal deficits) to act as a conditioning variable in modeling exchange rate movements, therefore implicitly addressing the role of fiscal policy via its interactions with monetary policy. The flexible varying coefficient model (Cai et al., 2000, Li et al., 2002) appears to be particularly suitable for modeling fiscal deficits as a conditioning variable, while the idea that fiscal deficits may effectively work as a binding constraint on monetary policy can hardly be modeled by a parametric (linear or nonlinear) model, as no theory has made an explicit suggestion about functional forms. Our econometric method thus improves on the popular parametric nonlinear smooth transition models used in the real exchange rate mean reversion literature (e.g. Michael et al., 1997). This type of varying coefficient models allow for a much more flexible mean reverse pattern including symmetric and also all kinds of possible asymmetric adjustment processes, which extends the previous literature.

Section snippets

Empirical results

The main data source is Datastream. Given the data availability, we consider quarterly US dollar-based real exchange rates of G4 countries–UK, Canada, Italy, and Japan–in the main analysis. The additional robustness check below based on the quarterly data allows us to include Australia, which yields a similar inference. Since Germany experienced reunification in 1990 that was clearly accompanied with a substantial increase in fiscal deficits, we decide not to include Germany here because we

Conclusions

We show some evidence that the mean reversion of dollar-based bilateral real exchange rates in G5 countries depends on the fiscal policy status of both countries. When fiscal policies in both countries push the bilateral real exchange rate to be deviated further away from the equilibrium level, faster mean reversion occurs. The findings also suggest the possibility of nonlinear causality running from fiscal deficits to exchange rates in a perhaps complex manner, which has been little explored

Acknowledgments

We thank Woody Eckard, Amy Glass, Dennis Jansen, and seminar/session participants at Texas A&M University, University of Colorado Denver, the 2008 Financial Management Association and the 2010 Midwest Finance Association annual meetings for helpful comments and suggestions. Gu and Li acknowledge research support by the Private Enterprise Research Center, Texas A&M University.

References (10)

There are more references available in the full text version of this article.

Cited by (0)

View full text