Distribuição e demanda agregada na visão neo-estruturalista: teoria e aplicações

Abstract

This thesis analyzed the relation between functional income distribution and aggregate demand according to the neo-structuralist view, which involved a theoretical overview and empirical exercises using annual data for Brazil (from 1953 to 2003) and quarterly data for the United States and Japan (from 1970 and 1980, respectively, to the third quarter of 2009). On the theoretical ground, the neo-structuralist literature was developed mainly during the 1980s, being formulated as alternative closures to an underdetermined model which aimed at depicting especially the neo-Keynesian, neomarxian and Steindlian views regarding the operation of the economy. The basic hypothesis was an equilibrium relationship between, on the one hand, the real wage and the profit rate, and, on the other, consumption per worker and capital accumulation. Each configuration provided a sign for the correlation and a logical causality chain involving those variables. Staring in the mid 1990s, a series of applied work was undertaken with reference to the Bhaduri and Marglins (1990) model, which comprises the neomarxian and Steindlian outcomes as special cases assuming variable capacity utilization. The empirical analyses are based on autoregressive distributed lag models, cointegration or vector autorregressions, and most often indicate a small or zero impact of distribution on aggregate demand. Moreover, qualitatively distinct results for a determined economy are found in the literature. The empirical exercises performed in this work corroborate this first impression, since it was not possible to clearly identify the direction of the impact of a marginal change in the wage share on the aggregate demand in neither one of the three cases analyzed.