Wage Convergence of Agricultural Workers of Brazil: 1992 - 2009

Wage Convergence of Agricultural Workers of Brazil 1992 - 2009 The aim of this study is to examine differences of wages of temporary and permanent workers of the agricultural sector in the Brazilian states from 1992 to 2009. We used the spatial econometric models to verify the wage convergences in the states Brazilians (beta absolute and conditional beta and sigma) and we used data from the National Survey of Household Sample. For beta conditional convergences were used the average schooling and experience (capture the effects of human capital) of each segment of workers and power tractors (capture the effects of fixed capital). The Exploratory Analysis of Spatial Data revealed an inverse distribution between the growth rate of wages of permanent and temporary workers in the Brazilian states, that it show us how the forces of demand and supply local labor markets are important. The coefficient of I of Moran in the growth rate of wages / hours for permanent workers was negative while temporary workers was positive. The clusters of the temporary workers have similarity spatial data while the clusters of the permanent workers have dissimilarity for spatial growth rate of wages. The coefficient I of Moran from the estimated regression OLS shows that the presence of autocorrelation therefore is appropriate to apply the spatial econometrics; however, the Lagrange Multiplier that indicates which model is most appropriate spatial show us only significant indication of the spatial error model for beta absolute convergence of temporary workers. Thus it was chose the most appropriate model based on Akaike and Schwarz criteria which indicated that spatial lag was more appropriated to beta absolute and beta conditional of wages of permanent workers and the spatial error model was appropriated to beta conditional convergence of wages of temporary workers. The results revealed that existence of convergence absolute and conditional. The speed of conditional convergence is faster than absolute convergence both segments, also revealing that inequalities wages between states tend to decrease. The education of worker was only significant variable. The spatial effect indicated the existence of two distinct labor markets in Brazilian agriculture. The convergence sigma shows a decrease in wage inequality by the fall in the dispersion of wage levels.