No abstract

The Brazilian municipalities show an enormous inequality on its development level. Even within the states considered relatively prosperous, there are huge internal disparities on income levels. The richest Brazilian municipality"s GDP per capita is about 190 times greater than the poorest municipality"s, according to IBGE (2000) database. A possible explanation for this phenomenon relies on institutional theory. Many theoretical and empirical studies, mainly based on crosscountry data, emphasize the role played by institutions on the determination of long run development. Nevertheless, there still is little research concerning the income differences within the national territory and its connection to institutional quality. The literature points out that institutions matter for the level of economic development because of their effects on political power distribution, generation of economic opportunities, innovation, human capital accumulation, and so on. Based on this assumption, the present study main goal is to analyze the effects of Brazilian municipalities" institutional quality on their GDP per capita levels. The results indicate that institutions are relevant and its importance is greater for large municipalities. On the other hand, human capital human capital is more important to small municipalities. To address the endogeneity problem inherent to the relationship between institutions and development, we employ the 2SLS method.

ResumoO capital humano, a produtividade e o capital físico são considerados os principais fatores na determinação do PIB per capita das economias. Conforme a abordagem neoclássica, a acumulação de capital humano explica praticamente um terço da variação do rendimento per capita entre os países. No entanto, ainda persistem discussões sobre as formas em que esse fator afeta o PIB per capita. O objetivo do presente estudo é comparar duas formas funcionais da função de produção para os Estados brasileiros: as propostas por Solow (1956) e por Mincer (1974. Também foram feitas estimações do retorno marginal da educação, além da realização de uma análise da importância do capital humano na determinação do PIB per capita utilizando diferentes métodos de estimação, no período 1980-2002. Os resultados rejeitaram a especificação neoclássica com inclusão do capital humano em favor da minceriana. Adicionalmente, o retorno marginal estimado da educação foi de 15%, e os resultados empíricos sustentam a teoria de que o capital humano é um dos principais fatores na determinação do nível de renda. Palavras-Chavecapital humano, crescimento econômico, função de produção minceriana, retorno da educação AbstractHuman capital, productivity and physical capital are considered the main factors in the economies' GDP per capita determination. According to the neoclassical approach, human capital accumulation explains about a third of the variation in per capita income across countries. However, there is no consensus on the ways in which human capital influences GDP per capita. The present study's goal is to compare two production functions functional forms for the Brazilian States: the one developed by Solow (1956) and the one developed by Mincer (1974). The marginal return of education also has been estimated and we have analyzed the relevance of human capital on GDP per capita determination through a variety of estimation methods, for the 1980-2002 period. The empirical results rejected the neoclassical specification with human capital in favor of the mincerian's specification. The estimated marginal return of education is 15% and the empirical findings support the theory that states that human capital is one of the main factors affecting income level.

Structural changes and labor market in Brazil. In the present article, we attempt to identify the sources of the changes in the labor schooling level in the three main sectors of the Brazilian economy: manufacturing, services and agriculture. It was verified that, despite the changes in the product and employment among sectors, mainly in the 1990s, the relative demands for qualified workers has not experimented significant changes. Moreover, in the periods in which schooling has increased more, the workers' wage has decreased more. This fact suggests that the increase in labor qualification was mainly due to the increase of this factor supply. The structural changes had contributed, in general, in a marginal and negative way to labor force level of qualification demand in all the three sectors.