Work-family policies and the effects of children on women's employment hours and wages

Abstract

Welfare state generosity around work-family policies appears to have somewhat contradictory effects, at least for some measures of gender equality. Work-family policies, in encouraging higher levels of women's labor market participation, may have also contributed to lower wage-levels for women relative to men, for instance. We consider the relationship between particular work-family policies and mothers’ employment outcomes. Analyses use data on employed women aged 25–45 from the Luxembourg Income Study for 21 countries across Eastern and Western Europe, North America, Israel, and Australia. We estimate within each country differences in employment hours and wages for women based on their number of children. Then, we examine the association of estimated per child penalties in wage and employment hours with country-level data on leaves and childcare. Work-family policies are generally associated with positive employment outcomes for mothers, relative to childless women. Work-facilitating policies such as childcare for young children have decisively positive effects on mothers’ employment hours and wages. Work-reducing policies, such as parental leave, however, can have positive effects if the leaves are moderate in length, but tradeoffs if the leaves are long.

La générosité de l'État-providence autour des politiques de conciliation travail-famille semble avoir des effets contradictoires, du moins pour certains critères concernant l'égalité des sexes. En encourageant davantage la participation des femmes dans le marché du travail, les politiques de travail-famille pourraient, par exemple, avoir contribué à la réduction du niveau des salaires des femmes relatifs à celui des hommes. Cette étude se penche sur le rapport entre des politiques particulières de travail-famille et leur impact sur le travail des mères. L'analyse se porte sur des femmes employées ayant entre 25 et 45 ans, et dont les données proviennent de l'Étude de Revenu du Luxembourg sur 21 pays en Europe de l'Est et de l'Ouest, en Amérique du Nord, Israël, et en Australie. Dans chaque pays, l’étude mesure les différences en heures d'emploi et en salaire pour des femmes, relatif au nombre d'enfants qu'elles ont. Dans un deuxième temps, nous examinons la relation entre les pénalités accrues en heures de travail et en salaire pour chaque enfant, et les données disponibles pour chaque pays sur les congés de travail et la garde des enfants. Les politiques travail-famille ont généralement des résultats positifs sur l'emploi des mères comparé aux femmes sans enfants. Les politiques qui facilitent le travail, telles que la garde d'enfants en bas âge ont des effets positifs décisifs sur les heures d'emploi et les salaires des mères. Les politiques de réduction de travail, telles que le congé parental, peuvent avoir des effets positifs, si leur durée est modérée, mais deviennent compromettantes si elles sont longues.

Keywords:

• work and family
• gender
• employment
• wages

Mots-clés:

• conciliation travail-famille
• l'égalité des sexes
• salaire
• heures de travail

Acknowledgements

We are thankful to the National Science Foundation Sociology Program (Grants #0600926 and #0751505) for their support and grateful to the comments raised by the anonymous reviewers, those attending the Sloan Network Panel meeting at Utrecht University, as well as participants in thematic session on work-family policies at the Eastern Sociological Society 2010 meetings.

Notes

1. Mandel and Shalev (2009a, 2009b) have unpacked one piece of this question – how family policies may affect women of different class positions differently. Mandel and Shalev (2009b, p. 1901) note ‘the consequences of any given role of the welfare state vary quite dramatically for women in different class positions’ – arguing that while these policies may help lower-class women, they may limit professional women's advancement.

2. We do not include the self-employed because some crucial information such as working hours are not consistently available across all of the countries in our sample, making analyses of them unreliable.

3. This includes differences in educational enrollment rates (which we are unable to measure consistently with the available data across all countries).

4. The exact formula for transforming coefficients into percentage change in a logged dependent variable given a one-unit change in independent variables is 100(e^b – 1) (Allison, 1999). However, if coefficients are relatively small, the differences between the transformation using simple multiplication times 100 and using the formula involving exponentiation are very small.

5. It is, however, possible to construct categorical measures of full-time and part-time employment status using information on the number of annual weeks worked full-time or part-time for Finland. In the Polish (2004) and Slovak (1992) datasets, categorical variables on full-time and part-time status are available. Therefore, these are included in the wage models.

6. This dataset collection and development was funded by the National Science Foundation Grants #0600926 and #0751505.

7. When we take the United States and Australia out of the picture, the relationship between maternity leave and wages increases from an R-square of 0.012–0.059 and the relationship between maternity leave and employment hours increases from 0.080 to 0.168.

8. These data refer to the ‘Family and Changing Gender Roles’ modules (1994 and 2002) of the International Social Survey Program (ISSP), available through the Leibniz Institute for the Social Sciences: http://www.gesis.org/en/services/data/survey-data/issp/modules-study-overview/family-changing-gender-roles/2002/