The conventional wisdom suggests that inflation can only be reduced at the cost of a short-term contraction in economic activity. But the last fifteen years have witnessed the emergence of a large body of evidence on the real effects of stabilization in high inflation countries, which clearly defies this conventional view. In Discussion Paper No. 1220, Research Fellow Sérgio Rebelo and Carlos Végh use a unified framework to assess, both qualitatively and quantitatively, the relevance of the different hypotheses that have been proposed to explain the real effects of exchange rate based stabilization. Using a dynamic model with rational agents who have perfect foresight, the four major hypotheses analysed are: the supply-side effects associated with an inflation decline; the perception that the exchange rate peg is temporary; the fiscal adjustments that tend to accompany the peg; and the existence of nominal rigidities in wages and prices.

Unfortunately, the performance of these different scenarios is most distinct with regard to the two facts about which least is known: the presence of a recession in the tradable sector and of a boom-recession cycle even in successful programmes. The supply-side effects prove to be an essential component in accounting for the stylized facts of exchange rate based stabilizations. This is particularly true with respect to the nominal wage rigidity hypothesis. When considered in isolation, this hypothesis produces strongly counterfactual results, but when combined with the supply-side effects of disinflation, it generates a realistic scenario. At a quantitative level, the results for the baseline parameterization fall short of explaining the orders of magnitude involved in stabilization episodes, mainly because of the insufficient information on the role played by non-tradable goods in actual economies to asses whether these configurations are empirically plausible.

Real Effects of Exchange Rate Based Stabilization: An Analysis of Competing Theories
Sérgio Rebelo and Carlos A Végh

Discussion Paper No. 1220, August 1995 (IM)