Gender and Salaries of the Qualified
Workforce in Brazil and Mexico
Maria Cristina Cacciamali and Fábio Tatei

where Wi is the salary of the i-th worker, is the vector of individual characteristics for each worker, β is the coefficient associated with each of these characteristics and is the disturbance term.

Going back to equation (1) and the assumptions regarding the salary structure in the presence or absence of discrimination, we arrive at (3) and (4):



In this way, the salary differential is broken down and explained by the estimated effects of differences in salaries between the main group and the discriminated group in expression (3), and in another, conventionally associated with the estimated effects of discrimination, expression (4). However, this interpretation should be taken with caution, as the discrimination term includes a series of unobservable features that may or may not be linked to pure discrimination.


The model applied for employed adults shows the expected results. That is, there is a positive and decreasing relationship between salary and experience, and disaggregation by gender shows differences in the three types of samples studied and between the two countries. In Brazil, men have higher rates of return in accordance with their experience and in comparison with those for women. This may not necessarily be due to discrimination, as there are other factors that can interfere in this difference, such as personal abilities, employment category, geographic region and more.4 In Mexico there was no significant difference between the rates of return related to experience for men and women, which shows that, at least for this variable, there was no evidence of discrimination by gender (Figure 1).

Figure 1. Rate of Return on Experience. Brazil and Mexico.
Adult Population. 2008

Source: pnad and enoe 2008. Prepared by the authors.

4 The rates of return for experience were calculated using the partial derivative of the salary equation in relation to the respective variables.