Underemployment of seniors and discrimination: an empirical contribution.

Summary The ageing of populations in developed economies appears to be a major issue in which employment is at the center of all concerns. It also influences the balance of their social protection system. This thesis focuses on the low employment rate of French seniors, represented by the over-50s, compared to other European countries. The shift in public policy towards seniors from a logic of exclusion to one of inclusion in the labor market in the 2000s has led to a slow and limited improvement in their employment rate. While other European countries have also been affected by the change in policies, it is legitimate to question the reasons for their moderate impact in France.This thesis is organized around two complementary parts. The first part states the weakness of the employment rate of seniors in France compared to its European neighbors (chapter one). This weakness is reflected in an occupational segregation that increases with age, which leads to difficulties in the orientation of older men in certain sectors and certain occupations within these sectors. Other avenues of reflection on the supposed reasons for this low employment rate of older men are also considered in a review of the literature (chapter two). These reasons come from several fields: the structural field and the characteristics of labor supply and demand, which illustrates the complexity of the problem for public authorities. The second part of the paper attempts to measure a persistent factor in the underemployment of seniors, even when all the objective reasons for which recruiters' preferences are less inclined to hire seniors are neutralized: age discrimination in hiring (and gender discrimination in some specific cases). Two types of methods and data are used to measure the residual part of the differences in the probability of being employed or of access to employment (of which discrimination is one of the components): decomposition methods on survey data, giving a measure qualified as objective (chapter three) with an extension analyzing the link between this residual part and the economic situation, and the methodology of correspondence tests, or testing, on experimental data collected on a selection of occupations in tension (chapter four). The first method shows a large residual part of the probability gap, suggesting high risks of age and gender discrimination. The comparison of the job access rates of the second method conducted on nearly 6,000 applications illustrates these risks according to which seniors have less chance of accessing a job interview compared to younger people with similar characteristics. We will see the different explanatory hypotheses of the underemployment of seniors, documented by the literature, such as the short distance to retirement, the obsolescence of skills in a situation of technological shocks, the social norms of gender (through professional reconversion) and age (through the exogenous preferences of recruiters).One of the guiding threads of the thesis is to distinguish the situation of senior men and women in terms of employment. There are clear differences throughout this thesis, with men's situation more problematic than that of their female counterparts in terms of the employment rate gap, occupational segregation, and hiring discrimination.