Some multi-country evidence on the effects of real exchange rates on output

BIS Working Papers  |  No 48  | 
03 September 1997
A considerable amount of theoretical literature suggests that, contrary to the predictions of textbook analysis, exchange rate devaluations may be contractionary rather than expansionary. This hypothesis has also received considerable empirical support. However, prior statistical analyses of the effects of exchange rate devaluation on output have been subject to several limitations: (i) they have failed to distinguish adequately between short and long-run effects; (ii) they have not controlled for the full range of external shocks; and (iii) they have not considered whether the effects of devaluation might differ between different regions of the world. The purpose of this paper is to estimate the impact of devaluation on output for 27 countries while attempting to address these limitations in previous empirical analyses. We find no evidence that devaluations are contractionary in the long run. Additionally, controlling for sources of spurious correlation and reverse causality appears to mute the measured contractionary effect of devaluation in the short run, although evidence that this effect exists remains even after these controls are introduced. Finally, no significant differences among the regions stand out; in fact, while the literature on contractionary devaluation has focused primarily on developing countries, we found no evidence that this effect is stronger in developing countries than in industrialised countries.