The Fertility and Women's Labor Force Participation puzzle in OECD Countries: The Role of Men's Home Production

Abstract

One effect of Southern Europe's rapid fertility decline is the emergence of a positive cross-country correlation between women's labor force participation and fertility across developed countries, despite the continuing negative correlation between these factors within countries. This study uses individual-level data for several OECD countries to examine how men's participation in home production can explain the positive relationship between fertility and women's labor force participation at the cross-country level. It finds that women living in countries where men participate more in home production are better able to combine having children with market work, leading to greater participation in the labor force at relatively high fertility levels. Within each country however, women with higher relative wages continue to have lower fertility and to participate more in the labor force than lower-paid women due to the higher opportunity cost of remaining at home. This finding on men's home production can thus explain the positive cross-country correlation between female labor force participation and fertility.

Keywords:

• Social externality
• women's labor force participation
• childcare
• fertility
• housework
• time use

Acknowledgments

We thank Brown University, the Social Science Research Council Program in Applied Economics, and the Population Council for financial support. We are grateful to very helpful comments received at the Population and Training Center at Brown University and at the meetings of the Society of Labor Economics, the European Association of Labor Economics, the European Economic and Econometric Association, the European Society of Population Economics, and the Population Association of America. All errors are ours.

Notes

¹ Not all couples consist of a man and woman, but for the purposes of the analysis that follows it is necessary to assume a heterosexual household.

² Deciphering the channels through which social networks operate is an arduous task, and it is beyond the scope of this paper. Although there is possibly an aspect of learning behind the concept of social externality proposed here, in the present context we are more inclined to think of social externality in terms of sanctions and rewards.

³ The comparative statics can be found in the working paper version of the paper at the authors' websites.

⁴ To see this we linearize a man's share of home labor in household i and write it as a function of the average share of home labor performed by men in country k, the man's and woman's wages, and gender role attitudes:

• Solving for the average share yields
  whose expectation is clearly different from zero for k = i.

⁵ The constructed fertility and real completed cohort fertility values for the European countries in our sample are respectively: 1.74 and 1.76 for Austria, 1.65 and 1.67 for Germany, 3.13 and 2.67 for Ireland, 1.68 and 1.80 for Italy, 2.04 and 2.05 for Norway, 2.00 and 2.03 for Sweden, and 1.76 and 2.02 for the UK.

⁶ Existing studies point to market substitutes as causes of reduction in women's home labor (Michael Bittman 1999; Jeremy Greenwood, Ananth Seshadri, and Mehmet Yorukoglu 2002). We do not want to take away the importance of market substitutes (such as electrodomestics, for instance). However, the fact that women in OECD countries continue to do most of the housework and childcare, and more importantly that the ability to outsource across the countries considered does not seem to be different, suggests that there are still a great many household goods that cannot be outsourced and that there are other important factors that can explain the observed differences across countries.

⁷ Earnings (from all sources) reported in Germany, Austria, and Italy are after taxes. Net earnings are constructed for these countries using personal income tax information published by the World Bank. Using the Penn World Tables 6.1, all earnings are transformed to a common scale by calculating Purchasing Power Parity (PPP) estimates using the formula: (w_currency /w _$) PPP.

⁸ It is possible that causality runs the opposite way: the higher fertility leads men to do a higher share of home production, resulting in a higher mean egalitarian index for a country. IV estimates deal with this endogeneity problem to some extent. Also notice that this effect runs from an aggregate variable (higher share of home production in a given country) to an individual variable (the number of children in a given household). It is more plausible that the externality affects individual fertility, rather than that individual fertility affects the externality.

⁹ Notice that simply including additional variables that capture institutional differences is unlikely to solve this problem, as there is likely to be a simultaneity bias arising from the fact that institutional features may themselves be responses to fertility levels. In other words, maternity benefits and flexible paid work options, for instance, may themselves be responses to fertility levels.

¹⁰ The results are available upon request to the authors.