Utilization of modern contraceptive methods has increased significantly among women around the world of reproductive age (15-49) in the past 50 years, from 28% in 1970 to 48% in 2019 ( Coulson, Sharma, and Wen 2023). However, this progress has not been evenly distributed globally. It is estimated that 218 million women of reproductive age in low- and middle-income countries (LMICs) have unmet need for contraception ( Pillai and Nagoshi 2023; Sully et al. 2020), with nearly a third of those women residing in sub-Saharan Africa (SSA) ( Haakenstad et al. 2022). Previous research has shown that increased access to family planning services leads to significant improvements in maternal and child health through reductions in early and unintended pregnancies, reducing the need for unsafe abortion, increasing birth spacing, and limiting family size ( Cleland et al. 2012; Dehingia et al. 2020; Marston and Cleland 2003; Molitoris 2017; Rau, Sarzosa, and Urzúa 2021). However, there is less evidence on the direct linkages between access to family planning services and other social and economic benefits, such as economic empowerment.

This project leverages randomized controlled trials (RCTs) and longitudinal data collection involving adolescents and women of reproductive age living in rural and urban areas in three LMICs in SSA with the ultimate goal of estimating the causal impact of family planning on economic empowerment at different points across the life course. We draw on data from three ongoing studies, each intervening at different ages and decision points in family planning: the evaluation of Act With Her - Older Adolescents (AWH-OA) as part of the Gender and Adolescence: Global Evidence study (GAGE) in Ethiopia, Connect in Tanzania, and Increasing Uptake of Modern Contraceptives (IUMC) in Cameroon. This paper outlines the conceptual framing that links these studies, the planned methodology for developing contextually relevant and age-appropriate measures of family planning and economic empowerment, and the analysis plan for establishing a causal link between contraceptive use and female empowerment.

We define economic empowerment following the “Can-Act-Resist” framework ( Raj 2020), which characterizes the empowerment process as a series of shifts from “critical consciousness” of their options to aspiration for change to the ability to take action to achieve those goals and ability to resist backlash for taking that action. This conceptualization provides a helpful framework for identifying and classifying indicators that capture the components of this empowerment process. We focus on economic empowerment across four overarching domains identified in the literature ( Glennerster, Walsh, and Diaz-Martin 2018): human capital development; participation in labor markets; financial literacy, inclusion, and control; and time use autonomy.

The existing literature has linked the consequences of increased modern contraceptive use—i.e. delays in first birth, increased spacing between first and second births, and ability to achieve optimal family size—to improvements in female economic participation through increased school attainment, experience gains in the labor market, and more control over their time ( Bailey 2006; Baird et al. 2016; Bloom et al. 2009; Canning and Schultz 2012; Herrera Almanza and Sahn 2018; de Jong, Smits, and Longwe 2017). However, this relationship has primarily been established through a set of piecemeal linkages (e.g., contraceptive use increases birth spacing, birth spacing increases experience attained and work-life continuity) where each link in the chain is drawn from a different study population and context ( J. E. Finlay et al. 2024). Furthermore, these studies have not drawn a link between family planning and broader impacts on female economic empowerment (e.g., decision-making over earnings and household expenses, access to financial markets).

We start by providing a framework for thinking about how to conceptualize family planning and economic empowerment across the life course for adolescents (ages 10-19) compared to young women (ages 20-25) and adults (ages 25+). Figure 1 presents a summary of key components of economic empowerment (panel D) and family planning (panel C) and the age ranges (panel B) at which they are most relevant and thus are the most likely to be responsive to intervention. We provide a discussion of these key components of economic empowerment and family planning and their relevance over the life course in the rest of this section.

The overarching goal of the research agenda we outline in this paper is to assess how access to family planning causally impacts female economic empowerment across the life course in SSA. To achieve this, we will leverage data from three ongoing RCTs in diverse settings: urban and rural Ethiopia (Act With Her - Older Adolescents (AWH-OA)), urban and rural Tanzania (Connect), and urban Cameroon (Increasing Uptake of Modern Contraceptives (IUMC)). These settings, each with its unique characteristics, target girls and women at different points in their reproductive and economic lives. The evidence from these studies, as depicted in Figure 1, will be used to illustrate the relationship between family planning and economic empowerment across diverse settings in terms of life stage and economic status. Table 1 provides a summary of selected country characteristics.

Figure 1 overlays the three interventions we study (panel A) with the age groups targeted (panel B) and the expected family planning and economic empowerment outcomes affected (panels C and D, respectively) to summarize the life stage targeted by each intervention. Figure 2 provides more details of each program and then maps them to the specific family planning and economic empowerment outcomes that they are hypothesized to affect. The GAGE study in Ethiopia evaluates AWH-OA, holistic programming targeting adolescents aged 14-18, as well as the people and systems that surround them, among a sample of 3,312 female adolescents who are largely unmarried and have not yet experienced their sexual debut. Connect in Tanzania evaluates programming aimed at increasing adoption of modern contraceptive methods in the first 12 months after giving birth among a sample of 1,134 first-time mothers aged 15-25. IUMC in Cameroon evaluates an intervention aimed at improving the quality of family planning counseling and product adoption to reduce unintended pregnancies among a sample of 784 women aged 15-49 who received family planning counseling at a hospital in the capital city. Table 2 provides an overview of the three studies, and Table 3 provides baseline summary statistics for each. We briefly summarize each sample and intervention below, with additional detail found at their respective trial registrations (AWH-OA ( Baird et al. 2023); Connect ( Baird and Seager 2023; Baird 2024); IUMC ( Ozler et al. 2021).

All trials received appropriate ethics approval from the relevant country ethics board, among others. Consent or assent was obtained as appropriate.

The Ethiopia research design and tools were approved by the George Washington University Committee on Human Research Institutional Review Board (071721, received 28 September 2017), the Overseas Development Institute Research Ethics Committee (02438, received 11 May 2017), the Ethiopian Policy Studies Institute (EDRI/DP/00689/10, 18 September 2017), the Addis Ababa University College of Health Sciences Institutional Review Board (113/17/Ext, 24 December 2017) and the Amhara and Oromia regional Bureaus of Health ethics committees. Verbal consent was obtained from caregivers and married adolescents; assent was obtained for all unmarried adolescents under the age of 18. There was also a robust protocol for referral to services tailored to the different realities of the diverse research sites.

The Connect study design was approved by the George Washington University Committee on Human Research, Institutional Review Board (NCR203091, approved on 4 October 2022), The Tanzania National Institute for Medical Research (NIMR/HQ/R.8a/Vol.IX/4174, approved on 19 December 2022) and COSTECH (approved on 15 February 2023). Following approval from NIMR and COSTECH in Tanzania, approval from the PO-RALG (President’s Office - Regional Administration and Local Government) was obtained prior to data collection. All survey participants (first-time mothers between the ages of 14 and 17) were considered emancipated, as they already have a child of their own and/or are married. Because first-time mothers who are married often reside outside their parents’ household and often in a different village or geographic area, obtaining informed consent from the parents of first-time mothers aged 15-17 is unfeasible in most cases, and would thus result in their perspectives not being represented in the design of the intervention. In the case that written consent is not possible, given a lack of literacy, consent will be obtained via thumbprint.

The IUMC study protocols were approved by Cameroon’s national ethics committee for human subjects research, the Comite National d’Ethique de la Recherche pour la Sante Humaine (CNERSH) on 12 ^th of July 2019 – decision No. 2019/08/1183/CE/CNERSH/SP – and received administrative authorization from the Ministry of Health’s (MinSante) Division of Health Operations Research (DROS) on 30 ^th of October 2019 – decision No. D30-760/L/MINSANTE/SG/DROS. These approvals were renewed yearly until the end of the experiment in 2022. The protocols were also approved by the implementing hospital’s own IRB on the 7 ^th of November 2018 – decision No. 780/CIERSH/DM/2018. Consent to participate in the follow-up surveys was obtained by the research team at the very end of the counseling session, to ensure that the study did not interfere with the counseling session between the health provider and the client nor influence the client’s actions and decisions regarding the adoption of contraception in any way. Assent was obtained for non-emancipated minors along with the consent of their caregivers. When written consent could not be obtained immediately from the caregivers of assenting adolescents, verbal consent was obtained through the phone.

We will use the developed measures of family planning and economic empowerment to estimate the causal impact of family planning on economic empowerment for each study separately, given the variation in study design and outcome measures described. For each study, we will first estimate reduced form models to provide causal estimates of the individual family planning interventions directly on economic empowerment, estimating models of the form: EE ic = α + β T ic + X ic ′ γ + ε ic , where EE ic is the measure of economic empowerment for individual i in cluster c (where appropriate), T ic is the relevant binary treatment indicator for the specific study, X ic ′ is a vector of baseline controls (including randomization strata, as appropriate), and the error term ε ic is clustered as appropriate for the study design.

Then, we will employ a two-stage least squares approach, using treatment status as an instrumental variable (IV) to leverage the experimental variation in access to family planning to estimate a local average treatment effect (LATE) of the causal impact of family planning on economic empowerment, using the following model: (1)

F P ic = θ 0 + θ 1 T ic + X ic ′ φ + v ic

Identification of the LATE requires additional assumptions: instrument exogeneity, relevance, exclusion, and monotonicity ( Imbens and Angrist 1994). Instrument exogeneity is trivially satisfied given random assignment in all three settings. Monotonicity requires that treatment influences take-up only in one direction – i.e., offering family planning counseling influences take-up by only increasing the likelihood of adopting a contraceptive for every individual, never reducing it. This is also likely satisfied. Instrument relevance (the first stage) will be shown for each study. We expect this condition to be met as the included interventions specifically target increasing knowledge and use of family planning services. We will use Anderson-Rubin confidence intervals and t-tests ( Keane and Neal 2023). Finally, the exclusion restriction requires that the treatment has no independent impact on economic empowerment other than its influence through improvements in family planning. This condition is theorized to hold for Connect and IUMC. Additional assumptions will be needed for AWH-OA.

The LATE will only provide a causal estimate for the subset of women whose family planning outcomes change due to the treatment. For example, if using contraceptive use as the family planning measure, the LATE is estimating the impact of contraceptive use on economic empowerment among women who were induced to adopt a method of contraception as a result of the treatment. In our settings, each of our programs addresses constraints to contraceptive adoption in ways that could be implemented through policy. In Ethiopia, AWH-OA programming focuses on increasing knowledge of sexual reproductive health, including contraceptive use and how to access services. In Tanzania, Connect focuses on increasing knowledge of postpartum family planning and contraceptive options and how to access services, as well as providing increased access to contraceptive methods through home visits. In Cameroon, programming offers discounts on contraceptives to make them more affordable. Thus, this is likely the exact group of people of interest to policy-makers, providing a policy-relevant estimate of how female economic empowerment would evolve through such instruments.

This paper develops a framework for estimating the causal impact of family planning on female economic empowerment across the life course, from early adolescence (10-14 years old) to adulthood (25 years and older). This framework highlights that, over the life course, the goals of family planning evolve from knowledge and awareness at younger ages to contraceptive utilization and reproductive agency at older ages. For economic empowerment, the framework highlights the importance of education and aspirations at younger ages in contrast to work and financial inclusion and agency at older ages. This framework can be used to bridge previous evidence that has focused on either broad age ranges (e.g., women of reproductive age, 15-49 years ( James-Hawkins et al. 2018)) or specific populations (e.g., postpartum mothers ( Dev et al. 2019; Mehare et al. 2020)).

In applying this framework, we develop age- and life-stage-specific scales to more accurately capture experiences of family planning and economic empowerment over the life course. Our proposed measure of family planning expands beyond contraceptive utilization ( Blackstone, Nwaozuru, and Iwelunmor 2017; James-Hawkins et al. 2018) to capture nuances around family planning preferences alongside reproductive agency and knowledge. Limiting the measure of family planning to the more traditional measures of contraceptive use and unmet need ignores other ways women may express and exert their reproductive agency to meet reproductive goals ( Potter et al. 2019; Senderowicz 2020). In addition, such measures may not be appropriate for all age ranges, as highlighted by the framework developed in this paper. Utilization of age-specific measures of family planning expands our ability to understand the role of family planning interventions at younger ages, particularly in adolescence when stigma and cultural norms make contraceptive adoption difficult ( Hall et al. 2018; Morris and Rushwan 2015). Similarly, the proposed development of age- and life-stage-relevant economic empowerment measures contributes to literature that has primarily focused on the relationship between family planning and specific indicators related to economic empowerment (e.g., increasing schooling, participation in paid work) ( J. E. Finlay et al. 2024; J. Finlay and Lee 2018); such indicators do not capture economic empowerment as a process ( Kabeer 1999; Raj 2020) that includes not only the act of engaging in activities but also the aspirations and ability to do so, even in the face of resistance ( Raj 2020).

Finally, this research proposes an analysis framework—applied to three RCTs targeting girls and women at different age ranges and life stages—to assess the causal link between family planning and female economic empowerment. These three studies evaluate family planning programming with distinct populations in terms of life stages in three LMICs in SSA. Specifically, AWH-OA provides insights into the impact of increased family planning knowledge during adolescence before sexual initiation and childbearing; Connect focuses on intervening after first birth with older adolescents and young women; and IMUC provides evidence from concentrating on a broader range of ages, mostly among women who are later in their reproductive life cycle. The interventions under study also vary in breadth, with AWH-OA having the broadest focus across sexual and reproductive health and family planning topics, Connect narrowing in on post-partum family planning and contraceptive adoption, and IUMC targeting the role of cost in deterring contraceptive adoption. This variation provides insights into the potential returns to different types of family planning programs and policy instruments in terms of female economic empowerment.

There are some limitations to the framework and its application, both within our own planned research agenda and in future research on this topic. First, as discussed above, nuances in the applicability of some indicators beyond age differences (e.g., living arrangements, marital status) can make the framework challenging to apply uniformly across studies. Moreover, certain constructs and measures may be more or less relevant in a given context, thus local adaptation of the framework is still likely necessary. Second, rich data is needed to fully apply this framework to generate one standard measure across time points. This may limit its applicability to many studies, including those that are the focus of this analysis. That said, even a partial application of the framework can shed important light on this issue. Third, our discussion of the framework focuses on a causal relationship between family planning and economic empowerment, which becomes challenging without the RCT structure laid out here; even within the RCT structure, causal identification is not guaranteed. While not the focus of the analytic set-up described in this paper, this framework can also help explore associations when causal analysis is not possible.

This paper presents a framework that can be applied to past and future studies of family planning and economic empowerment to synthesize evidence on the causal impact of family planning on economic empowerment. The framework highlights the need for age-disaggregated data for adolescents and women of reproductive age in order to assess the benefits of investment in family planning, given the nuanced differences in women’s needs and experiences at various ages. Moreover, this framework advocates for including knowledge, agency, and desired fertility alongside contraceptive use to measure achievement in family planning—a more person-centered approach, building on recent literature. Ultimately, the goal of this framework and its application in LMICs is to provide increased evidence on the role that family planning access and control play in driving female economic empowerment, which in turn offers a case for increased investment in family planning programming by implementers and policymakers.

The Ethiopia research design and tools were approved by the George Washington University Committee on Human Research Institutional Review Board (071721, received 28 September 2017), the Overseas Development Institute Research Ethics Committee (02438, received 11 May 2017), the Ethiopian Policy Studies Institute (EDRI/DP/00689/10, 18 September 2017), the Addis Ababa University College of Health Sciences Institutional Review Board (113/17/Ext, 24 December 2017) and the Amhara and Oromia regional Bureaus of Health ethics committees. Verbal consent was obtained from caregivers and married adolescents; assent was obtained for all unmarried adolescents under the age of 18. There was also a robust protocol for referral to services tailored to the different realities of the diverse research sites.

The Connect study design was approved by the George Washington University Committee on Human Research, Institutional Review Board (NCR203091, approved on 4 October 2022), The Tanzania National Institute for Medical Research (NIMR/HQ/R.8a/Vol.IX/4174, approved on 19 December 2022) and COSTECH (approved on 15 February 2023). Following approval from NIMR and COSTECH in Tanzania, approval from the PO-RALG (President’s Office - Regional Administration and Local Government) was obtained prior to data collection. All survey participants (first-time mothers between the ages of 14 and 17) were considered emancipated, as they already have a child of their own and/or are married. Because first-time mothers who are married often reside outside their parents’ household and often in a different village or geographic area, obtaining informed consent from the parents of first-time mothers aged 15-17 is unfeasible in most cases, and would thus result in their perspectives not being represented in the design of the intervention. In the case that written consent is not possible, given a lack of literacy, consent will be obtained via thumbprint.

The IUMC study protocols were approved by Cameroon’s national ethics committee for human subjects research, the Comite National d’Ethique de la Recherche pour la Sante Humaine (CNERSH) on 12 ^th of July 2019 – decision No. 2019/08/1183/CE/CNERSH/SP – and received administrative authorization from the Ministry of Health’s (MinSante) Division of Health Operations Research (DROS) on 30 ^th of October 2019 – decision No. D30-760/L/MINSANTE/SG/DROS. These approvals were renewed yearly until the end of the experiment in 2022. The protocols were also approved by the implementing hospital’s own IRB on the 7 ^th of November 2018 – decision No. 780/CIERSH/DM/2018. Consent to participate in the follow-up surveys was obtained by the research team at the very end of the counseling session, to ensure that the study did not interfere with the counseling session between the health provider and the client nor influence the client’s actions and decisions regarding the adoption of contraception in any way. Assent was obtained for non-emancipated minors along with the consent of their caregivers. When written consent could not be obtained immediately from the caregivers of assenting adolescents, verbal consent was obtained through the phone.

Jennifer Seager: Conceptualization, Data Curation, Formal Analysis, Funding Acquisition, Investigation, Methodology, Project Administration, Supervision, Visualization, Writing-Original Draft Preparation; Joan Hamory: Data Curation, Investigation, Methodology, Writing-Original Draft; Luca Parisotto: Data curation, Investigation, Methodology, Writing-Original Draft Preparation; Sarah Baird: Conceptualization, Data Curation, Funding Acquisition, Investigation, Methodology, Project Administration, Writing-Review & Editing.