Income inequalities for recently graduated French workers: a multilevel modeling approach

This paper presents a simultaneous study of the impact of gender, occupational and localization inequalities on the earnings of higher education graduates. The framework draws on both individual level (i.e., pertaining to the individual elements of groups) and aggregate level (i.e., pertaining to the group as a whole) data under a single specification. To take into account the selection process for employment, our multilevel model uses the Heckman two-step procedure. Occupational Groups (OG) are found to capture around 40 % of the wage heterogeneity, whereas Employment Area (EA) nests capture less than 10 %. Higher wages are offered to young workers in (1) OG dominated by seniors and (2) OG dominated by men. These group characteristics also influence gender inequalities: there is a higher wage penalty for women in (1) OG dominated by men and (2) OG dominated by senior workers. In contrast to gender inequality, immigrant inequalities manifest closer links to EA.