Academic journal article Economic Inquiry

Does Asymmetry of International Shocks Matter for the U.S. Business Cycle?

Academic journal article Economic Inquiry

Does Asymmetry of International Shocks Matter for the U.S. Business Cycle?

Article excerpt

This article proposes and investigates the asymmetry hypothesis, which predicts that an international asymmetric shock tends to have a stronger and longer effect on the U.S. business cycle than a symmetric shock. The hypothesis finds empirical support in the impulse responses of U.S. output and inflation to symmetric and asymmetric shocks; those responses are estimated in a four-variable structural vector autoregression. The hypothesis also finds support in stylized facts: The longest U.S. expansions have tended to occur when the rest of the world was growing below potential. (JEL E3, E5, E4)

I. INTRODUCTION

Economic expansions in the United States since World War II have tended to end with the same pattern: As the economy approaches capacity, wage and price inflation begins to rise. The Federal Reserve then tightens monetary policy to dampen inflationary pressures, thereby pushing the economy into a recession. (1) The expansion that ended in March 2001, however, appeared to deviate from this pattern: It was associated with little inflationary build-up, and consequently the Federal Reserve did not begin a string of uninterrupted tightening until late June 1999, when Consumer Price Index (CPI) inflation was running only slightly above the 2% rate. By comparison, consumer prices were rising at 6.2% at the end of the 1960s boom and over 5% at the end of the 1980s boom. Clearly, the monetary tightening of 1999 was enacted more in reaction to the threat of inflation, in light of the tight labor market, than to actual high inflation. What made the 1990s expansion different?

A number of explanations have been offered for the longevity of the latest expansion: that the credit markets have played an important role in sustaining the current expansion; (2) that greater fiscal discipline and enhanced central bank credibility have helped tame inflation expectations and sustain the boom; that the rise in the services component of gross domestic product (GDP) relative to manufacturing and more efficient inventory management have contributed to rendering the economy less susceptible to cyclical swings; that the increase in productivity growth caused by digital technology has helped meet increases in demand and labor shortages without driving up inflation; or that globalization per se has helped prolong the expansion.

This article considers another factor that is likely to have helped increase the longevity of the 1990s expansion. We call this factor the asymmetry hypothesis. The hypothesis maintains that in a flexible exchange rate regime, an asymmetric (that is, country-specific) shock will have a stronger and longer effect on output than will a shock that also hits the rest of the world. The hypothesis makes two assumptions: that the United States is sufficiently integrated with the rest of the world, or globalized, and that the Federal Reserve is forward-looking and (at least partly) effective in combating inflation and recessions. Under those assumptions, an expansion caused by a symmetric positive shock is accompanied by stronger inflationary pressures than an expansion caused by an equal-sized asymmetric shock, prompting quicker and stronger monetary tightening. Consequently, an expansion caused by a symmetric shock ends earlier than one caused by an asymmetric shock. Symmetrically, a contraction caused by a symmetric negative shock is potentially more devastating for the economy than one caused by an equal-sized asymmetric shock, prompting quicker and stronger monetary easing. Thus, a contraction caused by a symmetric shock ends earlier than a contraction caused by an asymmetric shock.

This article builds a two-country, open-economy model to illustrate why the hypothesis makes sense in an age of globalization and a forward-looking monetary regime. It then estimates a structural vector autoregression (SVAR) system using both long-run and short-run restrictions to investigate whether the hypothesis can stand the scrutiny of empirical study. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.