Gender discrimination in promotion: the case of Spanish labor market

This paper tests the hypothesis that a glass ceiling phenomenon exists in the Spanish labor market -that as women rise in the professional hierarchy, they face increasing obstacles for promotion. Taking data from the Spanish Survey on Quality of Life at Work (ECVT) for 2001, a number of indicators were selected for promotion: “number of promotions”, “supervision levels”, “number of people supervised” and “net wage”. The relative gaps were calculated for each of the intervals (categories) for these variables, in order of size, starting with the smallest. These measure the percentage by which women’s participation would have to increase in order to reach the level that would exist if there were no discrimination (the latter being calculated, in line with the Oaxaca decomposition, by evaluating women’s endowments in the model estimated for men, for whom it is assumed there is no gender discrimination). In order to calculate these relative gaps, for each of the indicators one model was estimated for women and one for men. The ordered probit model was used to calculate the probabilities (or theoretic frequencies) for the presence of women/men at each of the intervals considered. The results indicated that, in all cases, the relative gap tends to increase as the intervals rise. This might indicate the presence of a glass ceiling problem. The paper has four sections. The first covers the existing literature on the glass ceiling phenomenon. The second examines the main characteristics of the data and shows some preliminary results. The third presents the ordered probit model used to estimate the women’s and men’s equations. The fourth shows the main results. The closing section gives some conclusions.