Waleska de Fátima Monteiro

The objective of this research is to investigate the existence of increasing returns to formal education in Brazilian states and, by extension, Brazil, between the years 2003 and 2007. We also explore by gender. The theoretical and empirical model used allows us to tackle two problems present in the literature. The first is the non-linearity of Mincerian coefficients. The second is the potential sampling selection bias. The use of this model and technique lead us to obtain robust results for the presence of increasing returns to education at marginal level for Brazil and its states. For instance, at the aggregate level, the increasing returns starts at 4.0 years of formal education and, considering each gender separately, this return reducing this value. In most states of the North and Northeast, the female has yet to cross the threshold barrier of increasing returns. But these features are also present, nearly all the regions of the Center-East. The State of São Paulo and Paraná had a better rate of return for males. As for females, the best results were in the State of Sao Paulo and Rio Grande do Sul. As a general result, the estimates allow for future construction of human capital functions for each state to test economic policies.