Devaluations and Income Distribution in Latin America

Abstract

Although the adoption of floating exchange rates by industrial countries may have reduced their foreign trade preoccupations, most developing countries (LDCs) retain some form of fixed rates, and have to confront, with seemingly increasing frequency, the dilemmas raised by large balance of payment deficits. In spite of the clamor that arises over the allegedly negative impact on real wages and income distribution of the typical stabilization package of exchange rate devaluation and domestic contraction, little rigorous study' of the problem has been attempted since Diaz Alejandro's pathbreaking work of over two decades ago [12]. In this article we present a model in which devaluations and domestic contraction2 lower the real wage and worsen the distribution of income, and test that hypothesis for nine Latin American countries. For the theoretical presentation of our argument we construct a pair of aggregate supply curves, one modeling labor supply along lines, the other utilizing what is called the new classical macroeconomics. The demand side is composed alternatively of a straightforward function representing changing domestic demand, or of formulations of devaluations utilizing either of two markedly different assumptions about what happens to the money supply. Although these models have different implications about what happens to aggregate output, [47], they all agree as to the effects on real wages. A brief argument is then presented as to why changes in real wages should determine changes in labor's share of output, or the functional distribution of income. Empirical tests support our hypotheses. We also discuss the relation between labor's share and the size distribution of income, and the broader question of intersectoral income redistribution due to these demand shifts. We end the paper with some concluding remarks on related issues.