2

The Role of Fertility Decline and Women’s Empowerment in Economic Development and in Other Population and Societal Changes

Amy Tsui (Johns Hopkins University, emerita) served as moderator for this session and explained that the presenters were asked to consider the following questions in their presentations:

• What is the link between fertility change and economic development, especially at the macro level?
• What is the role of women’s empowerment as an intervening variable?

• What do we know about the relationship between changes in the role and status of women in developing countries and economic growth?
• Has access to family planning mattered?

REPRODUCTIVE HEALTH AND WOMEN’S ECONOMIC ACTIVITY: A NARRATIVE REVIEW

Jocelyn Finlay (Harvard University) addressed two key questions: To what extent does the nexus between fertility and women’s labor force participation (LFP) vary within and across low- and middle-income countries, and does access to family planning matter for women’s LFP? Finlay explained that her discussion of these questions was based upon a manuscript currently under review,¹ would not include discussion of gender norms, and would not assume that women’s LFP is inherently empowering.

The connection between fertility and women’s LFP varies greatly across income groups and continents, Finlay noted. Women’s LFP itself varies both in rate and in type across a country’s income levels.² Of all working women in low- and middle-income countries, approximately 90 percent and 50 percent, respectively, work in the informal sector, that is, in jobs that are unregulated by the government. The fertility rate also differs between low- and middle-income countries, with an average of 4.5 children per woman and 2.5 children per woman, respectively. The interaction of income and fertility has been illustrated in the “quality-quantity trade-off” model, in which families that earn higher incomes choose to have smaller numbers of “higher quality” children (e.g., children who are healthier and more educated).³ Finlay thus links fertility and women’s LFP by extending the quality-quantity trade-off to explain that with fewer children, women have more time available to participate in the workforce. This theory would predict an inverse relationship between the number of children and women’s LFP, yet Finlay finds that the relationship is not so straightforward in reality.

In sub-Saharan African low-income countries, the number of children and LFP can be directly related for some women: their labor participation

___________________

¹ J.E. Finlay, Women’s reproductive health and economic activity: A narrative review (in press).

² C. Goldin, The U-shaped female labor force function in economic development and economic history (pp. 61–90 in Investment in Women’s Human Capital and Economic Development, edited by T.P. Schultz, Chicago: University of Chicago Press, 1995).

³ G.S. Becker and H.G. Lewis, On the interaction between the quantity and quality of children, Journal of Political Economy 81(2), S279–S288.

increases upon the birth of a child.⁴ Finlay noted that this pattern may be explained by a lack of income sharing in the household, that is, that incomes from husbands and wives are not pooled. In low-income countries where women are responsible for all the costs associated with a child (e.g., food, clothing, health, education), they must work to cover those expenses. Because 90 percent of these working mothers are employed in the informal sector, they lack the social safety nets (e.g., maternity leave, childcare subsidies, tax breaks) that would allow a healthy balance between child-rearing and work. Therefore, women in low-income countries achieve increased LFP by relying on their older children to care for younger siblings and/or practicing long birth spacing.⁵

In South Asian low-income countries, however, a quite different relationship exists between LFP and fertility. In these countries, a nonlinear relationship between women’s LFP and fertility has been observed.⁶ Finlay highlighted gender norms that restrict women’s access to the labor market (regardless of whether or not they have children) as a central driver of this observed difference.⁷

The relationship between fertility and women’s LFP in Latin American middle-income countries is heterogeneous. Although studies have demonstrated a causal effect of fertility on women’s LFP, Finlay noted that the impact of high rates of income inequality is not considered in these investigations. She cited a study from Mayra Buvinic (United Nations Foundation)—which found that teen motherhood had no effect on LFP in teen mothers from rich households but increased LFP in teen mothers from poor households—as an example of the heterogenous relationship between fertility and LFP in middle-income countries.⁸ In this case, the finding may be explained by the increased likelihood that a teen mother would also act as the head of household in poorer families. Finlay highlighted single motherhood, income inequality, and inequality in labor market opportunity as worthy avenues for future exploration.

___________________

⁴ R. Heath and S. Jayachandran, The Causes and Consequences of Increased Female Education and Labor Force Participation in Developing Countries, NBER Working Paper No. 22766 (Cambridge, MA: National Bureau of Economic Research, 2017).

⁵ J.E. Finlay and M.A. Lee, Identifying causal effects of reproductive health improvements on women’s economic empowerment through the Population Poverty Research Initiative, The Milbank Quarterly 96(2), 300–322.

⁶ S. Joshi and T.P. Schultz, Family planning and women’s and children’s health: long-term consequence of an outreach program in Matlab, Bangladesh, Demography 50, 149–180 (doi: 10.1007/s13524-012-0172-2).

⁷ S. Jayachandran, Social norms as a barrier to women’s employment in developing countries (Northwestern University, 2020; JEL codes: O12, J16, J22, Z10).

⁸ M. Buvinic, The costs of adolescent childbearing: Evidence from Chile, Barbados, Guatemala, and Mexico, Studies in Family Planning 29(2), 201–209 (doi:10.2307/172159).

Finlay underscored that family planning and contraception do influence these trends but precisely how family planning influences fertility and women’s LFP in low- and middle-income countries could be better understood. She referenced the influential argument by Lant Pritchett⁹ that economic development drives down desired fertility, and that this decline leads to a decrease in total fertility rates with or without access to family planning. More recent analysis with greater volumes of data has somewhat supported Pritchett’s argument—particularly in sub-Saharan African countries—in that the effect of access to family planning has not overtaken the influence of economic development in declining fertility rates.¹⁰ Finlay indicated that these findings point to other factors that are likely shaping fertility decline, and that more research is needed to identify these factors.

Current studies of the impact of family planning on fertility and women’s LFP have produced a patchwork of results that could be contextualized within the framework of long-standing questions about fertility preferences and related declines. Finlay signaled that as the field begins to examine the role of family planning in women’s empowerment outcomes, study designs should carefully consider how fertility preferences are shaped by economic development, family planning, and new factors yet to be determined.

During the discussion that followed the presentation, Alex Ezeh (Drexel University) asked whether Finlay had observed any influence of marital status on women’s post-childbirth LFP. Finlay replied that marital status crystallizes differences in the relationship between LFP and fertility across continents and income groups. In sub-Saharan African countries, the birth of additional children increases LFP among women in all but the richest income groups. However, in Latin American countries this relationship was observed only in the poorest women, and in South Asian countries it was not observed at all. Finlay explained that variation in income-sharing and lone motherhood may drive these regional differences, and suggested that the effects of marital status on the relationship between LFP and fertility warrant further study. Naila Kabeer (London School of Economics and Political Science) added that women’s LFP in South Asian countries has been observed to decline at the point of marriage, whereas in other parts of the world this decreased participation often occurs after the birth of the first child.

Finlay received several questions regarding the nexus between family planning and LFP, including distinctions between organized family planning and general family planning, the influence of employment quality,

___________________

⁹ L.H. Pritchett, Desired fertility and the impact of population policies, Population and Development Review 20(1), 1–55.

¹⁰ I. Günther and K. Harttgen, Desired fertility and number of children born across time and space, Demography 53(1), 55–83 (doi: 10.1007/s13524-015-0451-9; PMID: 26786205).

and the impacts of birth spacing. She believed that while these questions were important, it was not possible to answer them based on the current peer-reviewed literature. In some cases this difficulty was attributable to a simple need for more time to accommodate the longitudinal studies capable of answering these questions (e.g., the minimum number of years required to study a birth spacing of five years is quite high).

GENDER, LIVELIHOODS, AND AGENCY: TRACKING PATHWAYS OF SOCIAL CHANGE

Naila Kabeer (London School of Economics and Political Science) began by explaining the importance of distinguishing among different domains of women’s empowerment, because change does not occur in uniform ways across them. She discussed how access to resources and opportunities translates into immediate and long-term outcomes within these domains and which outcomes allow women to challenge patriarchal structures.

Kabeer situated empowerment as one of three different kinds of agency that are promoted by access to livelihood resources:¹¹capability (i.e., expanded ways of “being and doing”), empowerment (i.e., expanded ways of “being, doing, thinking, feeling, and knowing” that seek to challenge the gender inequalities of daily life), and citizenship (i.e., actively engaging in public and political life to promote gender equality or social justice). Empowerment may be viewed as a form of capability that challenges power structures. These power structures, though variable on a global scale, share common features, such as an asymmetrical division of labor between paid and unpaid work among men and women, and norms and ideologies that position women in a lower position of power. While empowerment applies to the individual, citizenship is a form of agency that challenges power structures in a collective manner. Concepts of citizenship can further distinguish between “status” (in other words, the rights afforded to a group of citizens) and “practice” (in other words, actions taken to claim, expand, and renegotiate those rights).

Women’s access to resources such as credit and education impact the well-being of children, including their survival rates and their education. Capabilities that fall within this scope lie within women’s traditional roles as mothers and wives and enable them to perform those roles, but these capabilities do not necessarily translate to empowerment; however, they do translate to empowerment when they are used to diminish gender inequalities and challenge patriarchal power structures.

___________________

¹¹ N. Kabeer, Three faces of agency in feminist economics: capabilities, empowerment, and Citizenship (Ch. 10 in Handbook of Feminist Economics, edited by G. Berik and E. Kongar, New York: Routledge, 2020).

In the discussion, Jocelyn Finlay emphasized the significance of Kabeer’s comments regarding forms of agency that do not challenge gender inequalities and provided examples from recent work by Erica Field (Duke University) and Rohini Pande (Harvard University). Kabeer said that studies show that women’s literacy and LFP in India (when controlling for regional variation in patriarchal structures) are associated with a diminution of the gender gap in mortality as well as an increase in fertility levels, suggesting a degree of strategic agency for women. Kabeer also discussed another study conducted in India that demonstrated an increased investment in the education of girls by mothers who participated in the national employment guarantee scheme, while the participation of fathers in this same scheme had no impact on girls’ education and a negative impact on boys’ education; thus, male and female access to resources has a ripple effect on the well-being and empowerment of children.

To further evaluate the effects of empowerment on women themselves, Kabeer and colleagues conducted a survey in Ghana, Egypt, and Bangladesh using a common questionnaire that assessed indicators of different domains of empowerment: individual, family, and community.¹² In all three countries, formal employment was more consistently associated with empowerment indicators than informal employment; changes to social norms were often easier to negotiate within a household than within a wider community; and location exerted an influence on empowerment indicators; for example, there was less evidence of change in conservative areas of Egypt and Bangladesh. Kabeer further described how “customary” and “contemporary” resources¹³ could have similar outcomes: for example, in Bangladesh, women who were educated and formally employed and women who were religious and dressed conservatively were both consulted by their communities as sources of authority. Kabeer also noted that when economic interventions for women are accompanied by not only material change but also by sources of cognitive change (i.e., changes in how they viewed restrictive norms and practices), sustained evidence of empowerment is more likely.

Citizenship as a form of agency draws attention to rights and deficits of women. It ties agency to the fight for gender justice. Kabeer said that political awareness and agency rarely occur spontaneously among subordinated groups but may require external support. As far as women workers are concerned, their struggles for economic citizenship have not always been

___________________

¹² N. Kabeer et al., Paid Work, Women’s Empowerment and Inclusive Growth: Transforming the Structures of Constraint (UN Women Policy Paper, January 2013). Available: http://www.unwomen.org/~/media/headquarters/attachments/sections/library/publications/2013/1/paid-work-womens-empowerment-and-inclusive-growth2%20pdf.pdf.

¹³ S.K. Head et al., Customary and contemporary resources for women’s empowerment in Bangladesh, Development in Practice 25(3), 360-374.

supported by mainstream trade unions. It has often been women’s organizations and unions that have taken up their cause. Studies from Latin America found that women workers are more willing to join women’s organizations than trade unions, out of solidarity as well as the expectation that a women’s organization is more likely to recognize and act on gender-specific discrimination than male-dominated unions. The kinds of organizations working with women matter for outcomes. For instance, women who belonged to associations in Bangladesh experienced a greater impact on empowerment than associations in Ghana and Egypt. The reason for this might be that the associations in Bangladesh were often nongovernmental organizations (NGOs) with some degree of commitment to women’s equality, while the associations in Ghana were largely Christian organizations and more interested in women’s religious well-being, and the organizations in Egypt were largely state-sponsored and tied to state objectives.

Kabeer said that microfinance and savings groups are among the most studied for evidence of women’s empowerment. Kabeer shared her work in Bangladesh, which compared minimalist microfinance with socially-oriented microfinance and social mobilization organizations; the study revealed that minimalist microfinance had neither political nor economic effects, socially oriented microfinance had mixed results, and social mobilization organizations—which organized around savings groups and commitments to change—had the greatest effect on what she called citizenship. These impacts manifested as greater involvement in the public domain, an increased willingness to fight for justice for others, and mobilization against dowries, among other changes.¹⁴ Kabeer also shared evidence from a study in India, which showed the impact of self-help groups on the citizenship and political empowerment of women (e.g., increased voter participation and engagement with local officials) even in the absence of considerable changes in livelihood.¹⁵ Spillover effects on the empowerment of individuals who are not members of the self-help groups have also been observed.

Kabeer also noted that although they take many valuable forms, change and agency do not have equal impacts on patriarchal structures. Furthermore, these changes are not inevitably interrelated and do not display unidirectional causality; while basic capabilities may give rise to individual empowerment, collective action may lead to improvements in basic capabilities, and increases in individual empowerment may build courage to fight for broader social justice.

___________________

¹⁴ N. Kabeer, Between affiliation and autonomy: navigating pathways of women’s empowerment and gender justice in Bangladesh, Development and Change 42(2): 499–528.

¹⁵ A. Deshpande and S. Khanna, Can Weak Ties Create Social Capital? Evidence from Self-Help Groups in India, Department of Economics Discussion Paper 23 (Ashoka University, 2020).

Amy Tsui (Johns Hopkins University, emerita) inquired about the distinction between empowerment and agency, particularly whether the former is a subtype of the latter. Kabeer responded that everyone has some form of agency, or purpose of action; however, the way people use that agency has implications for power relations. Empowerment is therefore a method of using one’s agency to challenge manifestations of inequalities in daily life. By way of example, Kabeer noted that some educated women in India will use their agency to discriminate against daughters; while this action may be agency, it reinforces gender discrimination and is therefore not a form of empowerment. In contrast, Kabeer reiterated that increases in literacy and LFP close the gender gaps in survival rates, when controlling for regional patriarchies, across India, which challenges traditional cultural norms and is thus emblematic of empowerment.

Anita Raj (University of California, San Diego) asked Kabeer to elaborate further on the concept of citizenship as agency that is related to a fight for social justice. Kabeer asserted that movement against patriarchal norms takes courage and is often unsustainable when undertaken by a lone individual; for example, a woman who refuses to pay her daughter’s dowry may be empowered herself but will not change the institution of dowries. Kabeer then alluded to fellow presenter Seema Jayachandran’s (Northwestern University) work, which she said demonstrated the measurable impact of two women working together; when two women expand to a group of women, Kabeer suggested, the likelihood of structural change increases greatly. In this way, citizenship represents a progression of individual empowerment to collective action that challenges patriarchy.

SOCIAL NORMS AS A BARRIER TO FEMALE EMPLOYMENT IN DEVELOPING COUNTRIES

Seema Jayachandran (Northwestern University) discussed the role of social norms as a barrier to female employment and the resulting implications of that barrier for public policy. She noted that this presentation was based on a review article.¹⁶

A U-shaped relationship exists between a country’s economic development and rates of female employment, such that the lowest rates of female employment exist in countries at the middle of the economic development spectrum. Jayachandran explained that this initial drop in female employment as economies develop stems from jobs moving away from the home and from social norms influencing the employment of women in those

___________________

¹⁶ S. Jayachandran, Social Norms as a Barrier to Women’s Employment in Developing Countries. NBER No. w27449 (Cambridge, MA: National Bureau of Economic Research, 2020).

jobs, as well as from the rise of wealthier families that can afford to remove women from the workforce. As economies develop further, structural transformations occur in which more jobs require intellectual rather than physical ability, thereby giving women additional opportunities to re-enter the workforce. Jayachandran used this framework to contextualize the relationship between economic development, female employment, and public policy that aims to accelerate women’s ability to participate in service jobs as they emerge in an economy (e.g., through business microcredit or girls’ education).

Jayachandran called attention to the large dispersion of countries around the U-shaped curve, such that many countries exist in a state of economic development and female employment that lies well above or below the best-fit line for the data. For example, two countries can have the same gross domestic product but considerably different rates of female employment. Therefore, she argued, policy agendas that will enable women to find employment must account for these sources of variation. Patriarchal gender norms, which also vary throughout the world, are one candidate to explain these differences.

Jayachandran considered many specific norms and noted that they are not restricted to low-income countries; these norms include stigma about women serving as the family breadwinner, restrictions on women’s physical mobility and social interactions, acceptability of the harassment of women in public spaces, expectations related to control over household finances, and responsibility for household chores and child care. Jayachandran posited that to close the gap in women’s employment between countries, two potential policy approaches may be taken. The first may be to design policies that are informed by a country’s social norms (i.e., that consider what norms must be circumvented in the design of a program to increase employment), and the second may be to use policy to change the norms altogether.

Jayachandran shared an example of one program in Ahmedabad, India, that was designed to work within the social norms of this conservative region—namely, restrictions on women’s physical mobility.¹⁷ Jayachandran and colleagues worked with the Self Employed Women’s Association, a trade union that works with low-income self-employed women and offers services to increase their business knowledge and opportunities (e.g., by offering microcredit loans). The collaborative project sought to counteract norms concerning physical mobility—which restrict women’s peer networks—by allowing women to invite someone to accompany them to a business training program. When women attended the business program alone, they did not make progress in their business development (e.g.,

___________________

¹⁷ E. Field et al. Friendship at work: can peer effects catalyze female entrepreneurship? American Economic Journal: Economic Policy 8(2), 125–153.

women were no more likely to take out loans to expand their business or see increased profits than women who did not attend the training); however, when women attended the program with a friend, they were more likely to take out and repay business loans as well as earn higher business profits.

Interestingly, in that Ahmedabad example the attendance rate did not vary when women could bring a friend, nor did program comprehension seem to differ between the two social conditions. Instead, this result appeared to be driven by more ambitious goal-setting by the women who brought a friend to the program. This difference may very well be related to the norms of restricted mobility, in which women are generally less likely to have a strong network of peers with whom they can travel to business classes and collaborate on business ideas, and Jayachandran said that this case shows how policies and programs need not change a social norm or restriction in order to mitigate it.

Jayachandran shared findings from a second project that did aim to change social norms.¹⁸ In this case, high school students (grades 7–10) in Haryana, India, were exposed to new attitudes toward gender and female employment. (Jayachandran noted that this relatively small-scale project would not change the norms of an entire community, but could do so if scaled up.) Students engaged in sessions (designed by the NGO Breakthrough) that drove reflection on the gender-based norms and restrictions in their society and learned some of the human rights arguments about how those restrictions harm their society. The intervention had a large impact on gender attitudes: 16 percent of regressive attitudes converted to support for gender equality. This effect size remained similar 2.5 years after the program ended, suggesting a persistent attitude change; that persistence was somewhat more long-lasting in boys than girls, reinforcing the need to involve boys in these programs. In addition to attitudes, boys’ behavior also seemed to change more than girls’ behavior following the program, possibly because their behavior faced fewer constraints. Jayachandran noted that an opportunity exists to increase the scope of this program beyond its original two years and that she hopes to conduct follow-ups with students who completed the program to assess employment rates among the female students and the male students’ wives.

Jayachandran concluded her presentation by suggesting that, although traditional policies to improve women’s labor market outcomes (e.g., training, microcredit, education) are important, more labor market policies could be developed that specifically operate on social gender norms.

___________________

¹⁸ D. Dhar, T. Jain, and S. Jayachandran, Reshaping Adolescents’ Gender Attitudes: Evidence from a School-Based Experiment in India. Working paper (Northwestern University, 2020).

In the discussion that followed, Sunita Kishor (ICF) asked Jayachandran whether intervening with younger children may have a greater or longer-lasting effect. Jayachandran confirmed this possibility and shared plans to investigate the influence of the same intervention in children ages three to four. Mayra Buvinic asked which specific behaviors were measured in the study. Jayachandran answered that most measures were self-reported attitudes about norm-related behaviors, such as the performance of chores at home. However, the study team also measured increases in completion of a college scholarship application by girls who participated in the program, which it plans to follow up on by examining college attendance among those girls.

Jocelyn Finlay asked Jayachandran whether a current lack of income-sharing in sub-Saharan African households signals a possible divergence from the U-shaped curve of the relationship between women’s LFP and a country’s economic progress as that region develops. Jayachandran considered it possible that a lack of pooled income would mitigate the traditional decline in women’s LFP as sub-Saharan African countries develop further. However, she noted, the structural transformations that lead to jobs moving away from the agricultural sector and toward employment, a move that is more difficult to balance with child-rearing, would likely occur in these countries as well. Jayachandran suggested that it is difficult to predict whether the decrease in women’s LFP will manifest in sub-Saharan African countries because the U-shaped curve has also shifted up over time. She likened this shift to what Nancy Birdsall (Center for Global Development) described as a “globalization of norms” or “fixed time effect” as social norms have improved over time. Jayachandran added that these improvements could be driven by increased levels of education among girls.

This page intentionally left blank.