Journul of’Internutional Monr~y md Finance (1991), zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQP 10, S90-S99 Trade balance news and exchange rates: Is there a policy signal? KED HOGAN Faculty qf Management, McGill University, Montreal, Quebec H3A lG5, Canada MICHAEL MELVIN Department of Economics, Arizona State University, Tempe, AZ 8.5287, USA AND DAN J. ROBERTS* Federal Reserve Bank of Kansas City, 925 Grand Avenue, Kansas City, MO 64198-0001, USA Contrary to earlier studies, surprisingly large US trade deficits are shown to have a significant effect on exchange rates throughout the 1980s. Three possible reasons for the time-varying effect are considered. The evidence presented yields the following inference: deficit news is likely to have changed expectations of Fed intervention that moved exchange rates; deficit news probably has an effect on revisions offuture deficit expectations and exchange rates change as a result; and deficit news may change expectations regarding US trade policy that is reflected in exchange rates. Every month the US Department of Commerce announces the US trade balance for two months prior. As with other economic data, the monthly trade balance announcements provide an opportunity to study the effect of new information or news on key macroeconomic variables. We expect the trade balance news to have an effect on exchange rates as market participants discount the new information in determining current foreign currency prices. Previous researchers have studied the effect of trade balance news on exchange rates. Hardouvelis (1988) included unexpected trade deficits among his news model of asset prices over the 1979984 period and found no statistically significant effect, although the coefficient signs suggested that unanticipated increases in the trade deficit are associated with dollar depreciation. Deravi et al. (1988) examined *We thank Ali Kutan for his assistance in estimation and data gathering. Useful comments on an earlier draft were provided by Clas Wihlborg and other participants at the conference. While we could not incorporate all of their excellent suggestions in this paper, the final product has been greatly improved by their advice. The views expressed herein are solely those of the authors and do not necessarily reflect the views of the Federal Reserve Bank of Kansas City or of the Federal Reserve System. O261~56O6~9l/Ol/OS9~lO s(_‘) 1991 Butterworth-Heinemann Ltd