Cet article a été publié dans la série des documents de travail du DIAL (DT/2006-03) sous le titre "Theory and evidence on the glass ceiling effect using matched worker firm data".
Dewey
Sociologie économique
Sujet
Données appariées employeurs-employés; Régressions de quantiles; Plafond de verre; Écart salarial selon le genre; Matched worker-firm data; Quantile regressions; Glass ceiling; Gender wage gap
In this article, we investigate the relevance of the glass ceiling hypothesis in France, according to which there exist larger gender wage gaps at the upper tail of the wage distribution. Using a matched worker-firm data set of about 1 30 000 employees and 14 000 employers, we estimate quantile regressions and rely on a principal component analysis to summarize information specific to the firms. Our different results show that accounting for firm-related characteristics reduces the gender earnings gap at the top of the distribution, but the latter still remains much higher at the top than at the bottom. Furthermore, a quantile decomposition shows that the gender wage gap is mainly due to differences in the returns to observed characteristics rather than in differences in characteristics between men and women.
