Capital humano, trabalho e P&D no Brasil: análise de uma rede complexa: 2006-2016

Abstract

The main objective of this study is to analyze the relationship between the accumulation of human capital and the wage variations of the professions considered as "fully allocated in research, development and innovation (PD & I)", according to the classification of Gusso (2006). According to Tchernis (2010), a more in-depth investigation of the effects of human capital on wage growth is required, as well as the construction of new human capital measures that reflect transferable skills between some, but not all, jobs. In this sense, we use the method of statistical mechanics of complex networks, from which we analyze this relationship over the period 2006-2016. Since this method is non-parametric and strongly influenced by graph theory and topology, we did not test the causality between these variables and chose to find evidence of significant topological relationships and compute complex network statistics. Thus, this work develop an analysis based on two specific objectives: (i) test the hypothesis that there is a relationship between the educational level and the wage variation in the occupations of industrial companies with greater innovative capacity and that invest in PD & I; and ii) verify if there is evidence to support the hypotheses formulated by Acemoglu and Autor (2014) and Alon (2018). Acemoglu and Autor (2014) suggest that individuals who attend the same amount of years in the same school may have asymmetric levels of human capital due to wage incentives that improve the learning by doing process. Alon (2018) considers that the increase in the degree of specialization of American workers accounted for about 29% of the increase in the premium paid for the skill and 25-30% of the increase in educational level in the period 1965-2005. To deal with such hypotheses, we developed statistical measures derived from four complex networks based on the Yifan Hu Proportional (Hu, 2006) proportional algorithm for the 2006-08, 2009-2011 and 2012-2014 triennials and one for the 2015-2016 biennium. Using this data, we compare the wage variations resulting from advanced human capital investment and learning-by-doing between occupations.