Gender and Salaries of the Qualified
Workforce in Brazil and Mexico
Maria Cristina Cacciamali and Fábio Tatei
THE OAXACA-BLINDER DECOMPOSITION

The Oaxaca-Blinder decomposition of salary regressions, previously mentioned, allows us to disaggregate the salary gap between men and women into two parts: the first determined by the differences in productive factors of the workers and the other due to factors that cannot be measured in the model and for the present study we shall consider them relevant to gender discrimination in the labor market. To facilitate this decomposition, Table 11 shows how this methodology has been applied to the sample of employed adults to determine the composition of the income differential: the percent portion due to the difference in salaries and the portion that results from unexplained factors. The values were calculated as follows:

where X represents the percentage due to discrimination, A1 is the salary/hour that the analyzed group would receive in the absence of discrimination, A2 is the observed salary/hour of the same group, B is the observed salary/hour of the reference group — men — and Y represents the percentage due to the difference between the factors.

The results indicate that, if women’s attributes were valued in the same way as those of men, in other words, if there were no discrimination in the labor market, they should receive higher salaries than men in both countries (Table 11). Brazilian female workers effectively receive an average salary/hour of R$ 3.77/h. However, if they were valued in the same way as the men are valued, they should receive R$ 5.00/h. In other words, instead of receiving a salary that is 16.6% lower than that of men, they should receive 10.6% more. In Mexico, the differential is even greater. Women should receive an income 27.3% higher than what men with similar qualifications receive. Considering the fact that men and women with similar education, experience, geographic location and type of job have different income levels, this difference must be related to discrimination against women in the labor market. It is worth mentioning that this is one of the most key scenarios where unequal social gender relations exist. These results are consistent with literature that has shown that the effect of gender discrimination on salary gap between men and women stands out above all other income determinants.5

Below, we shall present the Oaxaca-Blinder decomposition for the sub-sample of workers with higher education and those employed in jobs that require higher levels of studies, only including the categories of executives and professionals/technicians with higher education. We observe that in both countries, there is higher income/hour related to the results of the regression for the sample of employed adults as well as reduced levels of discrimination, although discrimination is still the main factor determining the income differential between men and women (Table 12). Likewise, we observe that in Mexico, the amount of discrimination in the salary gap is still higher than in Brazil. In the group of qualified employees, discrimination is relatively less important than it is for just employed adults, which indicates that when more women have access to university education and executive positions, it may help overcome gender restrictions in the work market. However, the results make this proposition problematic.

5 See, among others, Maria Cristina Cacciamali, Fábio Tatei, Jackson William Rosalino, 2009.