Zusammenfassung (übersetzt):
This article aims to test empirical implications of the classical Beckerian theory on discrimination against women in Brazil. Specifically, the contemporary relationship between the relative use of women versus men and the performance, and how this relation changes with product market power is examined. The theory is tested using a panel of longitudinal microdata linking information at firm level from a matched employer-employee data (RAIS) and a firm level survey with detailed information on firm performance (PIA) from 2002 to 2013. We claim that the result may be contaminated by elements related to occupational segregation when the test is performed using the standard measure of relative employment of women. We propose a new measure that takes this into account; and implement the test with both measures of relative employment of women. The results indicated that in most of the specifications there is a positive and significant relationship between relative employment of women and profitability for companies with high market power. In a few specifications, this relationship also appears as positive and significant for companies with lower market power. But even in such cases the ratio is always stronger for the higher level of market power we consider. Finally, we find that the relationship between relative employment of women and profit depends on market power is much more stable when the first variable is measured by the index we propose.