Disentangling Discrimination on Spanish Boards of Directors

Abstract

The detection of discrimination on boards of directors is an interesting issue in the study of labour market inequalities, since the presence of a discriminated group would be especially scarce and tracking the possible causes of discrimination would be easier in the last steps of a professional career. Identifying the types of discrimination is a key issue, because they are usually mixed and distorted by the presence of other elements. In order to disentangle the different causes, an extreme case such as Spain, where the percentage of women on boards ranges from 6.6% to 8.6%, becomes especially relevant. In this study, discrete variable models are used to estimate the proportion of women on the boards of directors of the largest Spanish companies in 2005 and 2008. Some signals of discrimination have been identified. We have found evidence of the dynamics of Becker's model of taste-based discrimination, as time and competition can act in favour of women's presence on Spanish boards. There are clubby companies in which the homogeneity of the board prevails. This could also indicate traces of taste-based discrimination. When there is additional information about women's individual characteristics (i.e. family-based firms, cooperatives), statistical discrimination is reduced. Finally, we have detected the presence of companies that can systematically underestimate the abilities of women to fill these positions, a situation that tends to disappear when the companies already have female directors. We interpret this contagious factor as a signal of mistake-based discrimination.