Co-auteur
  • PÉRIVIER Hélène (3)
Type de Document
  • Working paper (4)
This paper aims to explain why the division of domestic labor within couples differs according to their marital status. We analyze the evolution of the gender division of labor in France using time-use surveys (1985, 1999 and 2009). In 1985 and in 1998, married women were performing a larger share of domestic labor than cohabiting women. In 1985 this gap is explained by differences in the observed characteristics of married and cohabiting couples, whereas by the late 1990s cohabiting couples were opting for an organization that was less unequal than that of married couples, all else being equal. In 2009, the average share of domestic labor performed by women was about the same whether they were cohabiting or married (72% and 73.5%), but it was significantly lower for women in civil union, which was introduced in 1999 (65.1%). The self-selection process of couples regarding their gender ideology might explain this result: the civil partnership attracts more egalitarian couples.

L’objet de cet article est d’analyser le partage des tâches domestiques dans les couples selon leur statut marital, et son évolution depuis les années 1980 à partir de trois enquêtes emploi du temps de l’Insee (1985-86, 1998-99 et 2009-10). La méthode des MCO est complétée par la méthode d’appariement qui permet de tenir compte de l’auto-sélection des couples au regard de leurs caractéristiques observables dans les différentes formes d’union (mariage, union libre et pacs pour l’année 2009-10). En 1985-86 ainsi qu’en 1998-99, le degré de division sexuée du travail était plus important dans les couples mariés que dans les couples vivant en union libre. Pour l’année 1985-86, cet écart s’explique par les différences de caractéristiques des couples vivant en union libre. En revanche, à la fin des années 1990, les couples en union libre optaient pour une organisation moins inégalitaire que celle des couples mariés toutes choses égales par ailleurs. Pour l’année 2009-10, en moyenne la part de travail domestique réalisée par les femmes est sensiblement la même qu’elles vivent en union libre ou qu’elles soient mariées (respectivement 72% et 73,5%), mais elle est significativement plus faible pour les femmes pacsées (65,1%). Ces écarts ne sont pas dus aux différences de caractéristiques observables des couples selon le type d’union. L’article montre qu’une auto-sélection des couples qui s’opèrerait sur les valeurs expliquerait cet écart : en 2009-10, le pacs attirerait les couples les plus « égalitaires », qui, avant l’introduction de l’union civile, optaient pour l’union libre.

The purpose of this article is to analyse the division of domestic tasks within the couple according to their marital status as well as how this has changed since the 1980s based on three INSEE time-use surveys (1985-86, 1998-99 and 2009-10). The ordinary least squares (OLS) method is complemented by the matching method, which is used to account for the self-selection of the couples in terms of their observable characteristics in different forms of union (marriage, cohabitation and civil partnerships for 2009-10). In 1985-86 and in 1998-99, the degree of the sexual division of labour was higher for married couples than for cohabiting couples. For 1985-1986, this difference is explained by differences in the characteristics of the couples who were cohabiting. However, by the late 1990s cohabiting couples had opted for an organization that was less unequal than that of married couples, all else being equal. For 2009-10, the average amount of domestic work performed by women was about the same whether they were cohabiting or married (72% and 73.5%), but the level was significantly lower for women in civil partnerships (65.1%). These differences are not due to differences in the observable characteristics of the couples based on the type of union. The article shows that this difference is due to a process of the couples’ self-selection based on their values: in 2009-10, civil partnerships attracted more “egalitarian” couples who, prior to the introduction of civil partnerships, had opted to cohabit.

Publié en 2015-01 Collection Working Paper de l'OFCE : 2015-02
8
vues

8
téléchargements
This paper proposes a dynamic statistical-discrimination model of job assignment and promotion which takes into account the endogeneity of human-capital investment and where the employer’s prior beliefs are self-fulfilling in equilibrium. The model shows that the equilibrium results from standard statistical-discrimination models may change when we account for discrimination/self-selection in hiring via the employer’s beliefs about worker expected quit rates and ability. The model is estimated on the Egyptian labour market using a multivariate simulated maximum likelihood model, and the results confirm the model’s predictions. When women face significant adversity in hiring, those women who overcome this initial discrimination are as likely to be promoted as their male counterparts with similar characteristics.