The Impact of Marriage on Women's Employment in the Middle East and North Africa

Abstract

Marriage is a central stage in the transition to adulthood in the Middle East and North Africa (MENA). This article investigates the effect of marriage on women's employment in MENA, examining how different types of work are affected by relatively early marriage, defined as marriage by the median age of marriage. An important contribution of this study is to examine the two main mechanisms by which marriage can affect work: (1) its effect on ever entering work and (2) its effect on exiting work. This study endogenizes the marriage decision using an instrumental variables approach. It finds that marriage by the median age reduces women's probability of market work by 47 percent in Jordan, 30 percent in Tunisia, and 16 percent in Egypt. Much of the effect is due to a reduction in the probability of private wage work, which women tend to leave at marriage.

HIGHLIGHTS

• Women in Egypt, Jordan, and Tunisia often leave employment at marriage.

• Marrying by the median age has varying effects on different types of employment.

• Women are particularly likely to leave private sector wage work at marriage.

• Changes are needed to reconcile private wage employment with women's domestic roles.

KEYWORDS:

• Economics of marriage
• labor markets
• employment
• age at marriage
• gender
• Middle East and North Africa

JEL Codes:

• J12
• J16
• J21

ACKNOWLEDGMENTS

The authors acknowledge funding from the Economic Research Forum. The authors appreciate the helpful research assistance of Sofia Jawad and Anna Shilongo. The authors are grateful for the feedback of colleagues, particularly discussant Paul Schultz, at the “Economics of Lifecourse Transitions” workshop held by ERF in Cairo, and discussant Kathryn Yount, at the Economic Research Forum 23rd Annual Conference.

SUPPLEMENTAL DATA

Supplemental data for this article can be accessed at https://doi.org/10.1080/13545701.2021.2007415https://doi.org/10.1080/13545701.2021.2007415.

Notes

1 We restrict our analyses of Jordan to Jordanians only in order not to confound patterns in Jordan with those of persons in Jordan displaced from other countries.

2 At www.erfdataportal.com. For more information on the three datasets see Caroline Krafft and Ragui Assaad (2021), Ragui Assaad and Caroline Krafft (2013), and Ragui Assaad et al. (2016).

3 Fathers and mothers with missing information on education levels are added to the no education category.

4 See the Online Appendix for more details on the discrete time hazard model we use.

5 For the bivariate probit model, we calculate the marginal effect as the difference between the probability of working if the individual was married by the median age and that of working if the individual was not married by the median age.

6 We additionally estimated models with interactions between education and marrying by the median age (probit models, not shown, available from the authors upon request). The interaction of education and marrying by the median age shows that women with university education in Jordan and Tunisia who married by the median age did not experience significantly reduced probabilities of work. Rather, it is women with secondary education in Jordan, and women with less than secondary and secondary education in Tunisia who witnessed a significant drop in their probabilities of work when marrying by the median age. In Egypt, women of every education level become significantly less likely to work when they married by the median age, in comparison with those with no education who experience a significant increase in their probability of work. The different result in Egypt can be explained by the tendency for uneducated women in Egypt to engage in non-wage work, which is less affected by marriage, and the increasing scarcity of public sector jobs, which are more compatible with marriage, for women with higher education. 

7 See Assaad, Krafft, and Selwaness (2017) for additional analyses of potential mediators, including childcare, maternity leave, the gender composition of workplaces, and commute times.

8 Women engaged in other than market (that is, subsistence) work have high hours of domestic work as well.

9 In these countries, employment of domestic help is very limited. Just 0.4 percent (in Egypt) and 0.5 percent (in Tunisia), and 5.6 percent of employment (in Jordan, including non-Jordanians) is in activities of households as employers (that is, domestic help).

Additional information

Funding

This work was supported by the Economic Research Forum.

Notes on contributors

Ragui Assaad

Ragui Assaad is Professor at the Humphrey School of Public Affairs at the University of Minnesota. He has been a Research Fellow of ERF since 1994 and currently serves as a member of its board of trustees. His current research focuses on labor markets in the Arab world, with a focus on youth and gender issues, transitions from school to work, education policies, and family formation. He holds a PhD in City and Regional Planning from Cornell University and a Master's degree in Mechanical Engineering from Stanford University.

Caroline Krafft

Caroline Krafft is Associate Professor of Economics at St. Catherine University. She received her Master's degree in public policy from the University of Minnesota's Humphrey School of Public Affairs and her PhD from the Department of Applied Economics at the University of Minnesota. Her research examines issues in development economics, primarily labor, education, health, and inequality in the Middle East and North Africa. Current projects include work on refugees, labor market dynamics, life course transitions, human capital accumulation, and fertility.

Irene Selwaness

Irene Selwaness is Assistant Professor of Economics at the Faculty of Economics and Political Science, Cairo University. Her current research focuses on the link between informality and social security systems, school-to-work transition, marriage, and employment issues in the Middle East. She holds a PhD in Labor Economics from University of Paris 1– Panthéon Sorbonne (France), and a Master's degree in Quantitative Economics with specialization in Labor and Demographic Economics from the same university.