Gender and Unconditional Cash Transfers in Africa
•Download as PPTX, PDF•
1 like•851 views
Amber Peterman presents case studies from government-run social cash transfers for a socialprotection.org webinar
Recommended
More Related Content
What's hot
What's hot (20)
Similar to Gender and Unconditional Cash Transfers in Africa
Similar to Gender and Unconditional Cash Transfers in Africa (20)
Recently uploaded
Recently uploaded (20)
Gender and Unconditional Cash Transfers in Africa
- 1. Gender & Unconditional Cash Transfers in Africa: Case Studies from Government programs supported by the Transfer Project Amber Peterman on behalf of the UNICEF Office of Research—Innocenti & FAO Transfer Project Teams
- 2. Van den Bold et al. (2013). Women’s empowerment and nutrition: An evidence review. IFPRI Discussion Paper 01294. Washington DC.: International Food Policy Research Institute. Research gaps in gender & social protection (following from Quisumbing) 1. Programs often target women as a means to achieve positive outcomes (particularly for children) -- women are perceived as spending cash in a more ‘family responsive’ way Literature supporting this claim is dated, taken mostly from studies on intra-household consumption/expenditure – rather than gender- randomized experiments Where rigorous studies exist, findings are mixed (Yoong et al. 2012) 2. Under conditions of (1), it is assumed programs will ‘empower’ women beneficiaries We see large potential in this possibility – but current evidence is mixed Part of the lack of consensus stems from multitude of indicators utilized, as well as large variation in gendered context Empowerment outcomes are often linked to effectiveness in program implementation/operations Source: Yoong et al. (2012). The impact of economic resource transfers to women versus men: A systematic review (Technical report). London, UK: EPPI-Centre, Social Science Research Unit, Institute of Education, University of London.
- 3. Review: Programming and impacts on women’s empowerment in LMIC “While many development initiatives seem to target women specifically, or have women’s empowerment as one of their objectives, no sufficient body of evidence overwhelmingly points to success … (p. 29, van den Bold et al. 2013)” Intervention Quantitative evidence Qualitative evidence Conditional cash transfers (CCTs) +/- + Unconditional cash transfers (UCTs) +/- More needed Micro-finance +/- +/- Agricultural interventions +/- More needed +/- More needed Van den Bold et al. (2013). Women’s empowerment and nutrition: An evidence review. IFPRI Discussion Paper 01294. Washington DC.: International Food Policy Research Institute. Source: van den Bold et al. (2013). Women’s empowerment and nutrition: An evidence review. IFPRI Discussion Paper 01294. Washington DC.: International Food Policy Research Institute.
- 4. Rise of social protection in Africa: Non-contributory Govt programming triples over last 15 years Source: Cirillo & Tebaldi 2016 (Social Protection in Africa: Inventory of Non-Contributory Programmes): www.ipc- undp.org/pub/eng/Social_Protection_in_Africa.pdf
- 5. Source: Cirillo & Tebaldi 2016 (Social Protection in Africa: Inventory of Non-Contributory Programmes): www.ipc- undp.org/pub/eng/Social_Protection_in_Africa.pdf Typologies of programs & target groups
- 6. Deep dive: Case study examples from the Transfer Project 1. Gender Targeting (Lesotho Child Grants Program) – Sebastian et al. forthcoming What are gender-differentiated impacts on child level outcomes (schooling, labor, time use)? Does gender of recipient & HH structure affect outcomes? 2. Women’s Empowerment (Zambia’s Child Grant Program) – Bonilla et al. 2016; Natali et al. 2016 Does receipt of benefits translate into higher intra- household decision-making, women’s savings and non- farm small businesses?
- 7. The Transfer Project Who: Community of research, donor and implementing partners – focus on coordination in efforts and uptake of results UNICEF, FAO, UNC, Save the Children, National Governments Mission: Provide rigorous evidence on of government- run large-scale (largely unconditional) SCTs Motivation: Income poverty has highly damaging impacts on human development Cash empowers people living in poverty to make their own decisions on how to improve their lives Where: Ethiopia, Ghana, Kenya, Lesotho, Malawi, South Africa, Tanzania, Zambia and Zimbabwe
- 8. Overview of programs & evaluations • All programs unconditional, with exception of Tanzania (schooling, health) • Longitudinal qualitative studies in Ghana, Malawi, Tanzania, Zimbabwe Country (program) Targeting (in addition to poverty, ultra- poor) Transfer size (% of baseline consumption) Methodology Years of data collection Ghana (LEAP) Elderly, disabled or OVC 7 Longitudinal PSM 2010, 2012 Ghana (LEAP 1000) Pregnant women, child<2 16 RDD 2015, 2017 Kenya (CT-OVC) OVC <18 22 RCT 2007, 2009, 2011 Lesotho (CGP) OVC <18 20 RCT 2011, 2013 Malawi (SCTP) Labour-constrained 18 RCT 2011, 2013, 2015 Tanzania (PSSN) Food poor ~ RCT 2015, 2017 Zambia (CGP) Child 0-5 27 RCT 2010, 2012, 2013, 2014 Zambia (MCTG) Female, elderly, disabled, OVC 21 RCT 2011, 2013, 2014 Zimbabwe (HSCT) Food poor, labour- constrained 20 Longitudinal matched case-control 2013, 2014, 2016
- 9. Zoom in on Child Grant models Lesotho Zambia Ministry of Social Development (MoSD) Ministry of Community Development & Social Services (MCDSS) Poor HH with child 0-17 – 5 districts HH with child 0-5 – 3 districts ~67% female beneficiaries ~99% female beneficiaries USD 36 (quarterly, 20% of exp) + messaging USD 24 (bi-monthly, 27% of exp) RCT (96 clusters) RCT (90 clusters) 2011, 2013 2010, 2012, 2013, 2014 1006 HH (agricultural sample) 2500 HH
- 10. Lesotho: Impacts on children’s outcomes • Improved outcomes for secondary school aged children (13-17): schooling, time-use, labor • 12 pp more likely to be enrolled • 20% less likely to have missed school in last 30 days • One extra hour spent at school • 45 min/day reduction on household chores • 0.9 fewer days/week on the farm • Results driven by girls, with the exception of reduction in farm labor, driven by boys
- 11. Lesotho impacts: Gender matters! • Results maintained for dual adult MHHs, however: • In de jure FHH, outcomes improved among secondary school-aged boys relative to secondary school-aged girls • Within dual adult MHHs, cash in the hand of the mothers may not always result in ‘better’ results: • Receipt by the father may be more likely to have positive impacts on girl’s schooling, as well as decreasing boy’s labor in farming while simultaneously increasing boy’s labor input in household chores (!)
- 12. Lesotho: Discussion points • The CGP has been successful in terms of improving schooling, time use and labour outcomes for secondary school aged children in agricultural households • If concerned about reducing gender inequalities, an undifferentiated cash transfer could include gender- specific messaging to promote boys' and girls' equal benefit in schooling • Child welfare may not be driven by the gender of the transfer recipient (e.g. mother or father) contrary to common belief • HH structure may partially explain some of these differences – children in MHH are more likely to be biological children as compared to FHH – FHH may be more labour constrained
- 13. What about Zambia? ©FAO/Ivan Grifi
- 14. Total consumption pc Food consumption pc Non-food consumption pc Food security scale (HFIAS) Does not worry about food Does not go to sleep hungry at night Does not go whole day w/o eating Domestic asset index Livestock index Productive asset index Does not consider hh very poor Hh better off compared to 12 months ago Life will be better in the future (women only) Shoes Two sets of clothes Blanket Currently enrolled Days in attendance prior week Not stunted Not wasted Not underweight Consumption Food security Assets Relative Poverty Material needs (children 5-17) Schooling (children 11-17) Nutrition (Young children 0-59m) -.2 0 .2 .4 .6 .8 1 1.2 Effect size in SDs of the control group Intent-to-Treat effects (CGP, 48-months)
- 15. Impact on intra-household decision-making • Question: “Who in your household typically decides XX” • Code indicator = 1 if women reports sole and/or joint decision-making • Impacts on 5 out of 9 domains – child schooling, own income, partners income, children’s cloths and shoes, family visits • No impact on child health, major or daily purchases and own health • BUT total is qualitatively small (0.34 additional decisions) Source: Bonilla et al. 2016 6.96 6.34 0 1 2 3 4 5 6 7 8 9 Count of sole/joint decisions Treat Control 0.34 impact*** Note: Results from adjusted ANCOVA OLS models ***1% significance.
- 16. Qualitative findings support the story • CGP has not led to massive change in relations or dynamics: “Even in the laws of Zambia, a woman is like a steering wheel, and us (the men) are the ones to drive them in everything.” ~Male, age 53 (beneficiary) • Yet, there is subtle change: transfer income is under control of women, and women equate empowerment = financial standing: “I am very happy because I don’t have to wait for him to make enough money as he puts it. I am able to suggest anything for the children now. He is in charge, but at least the money is in my hands.” ~Female, married, age 24 (beneficiary)
- 17. Examining financial standing directly?: Savings and non-farm enterprises (NFE) Evidence from LAC of CCTs on savings/investment mixed Evidence on UCTs in SSA scant, but promising Micro-credit and other savings programs have not delivered impacts as strongly as previously assumed – special issue of AEJ: Applied (Banerjee et al. 2015) Measures: • Savings: if woman is currently saving in cash • Non-farm enterprises: if the household has operated any non- farm enterprises (NFE) or provided any services (store, transport, home brewing, trade, or others) in the last 12 months Van den Bold et al. (2013). Women’s empowerment and nutrition: An evidence review. IFPRI Discussion Paper 01294. Washington DC.: International Food Policy Research Institute. Source: Banerjee et al. (2015). Six Randomized Evaluations of Microcredit: Introduction and Further Steps. American Economic Journal: Applied Economics, 7(1): 1-21
- 18. 47% 36% 47% 45% 22% 23% 30% 31% 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% Any savings (24-months) Any savings (36-months) Operates NFE (24- months) Operates NFE (36- months) Treat Control 10 pp impact** 23 pp impact** Impacts on saving and small businesses 17 pp impact** 15 pp impact** Increase in savings does not seem to crowd out other forms of household savings (livestock, assets) Note: Results from multivariate adjusted models difference-in-difference LPM *10% significance, **5% significance; ***1% significance.
- 19. Has Zambia’s CGP ‘empowered’ women? Yes, women have more capital (cash) in their control – and are using it for income generation However, few meaningful impacts on classic ‘bargaining power’ measures (decision-making) Entrenched gender norms limit transformative shifts (even in medium term – 4 years) Important, as CGP had no specific gender components (beyond targeting) Zambia, credit: Amber Peterman
- 20. • Interviewer: “What does it mean to you to be empowered? For example, if you were to describe a woman in your community who is empowered, what would she be like?” • Respondent: “Yes, there is a certain woman called Mary. She buys fish and sells . . . before that she never used to do anything. She was also receiving the CWAC money. Her husband had two wives . . .he never paid attention to the CWAC money. She saved some money and started buying fish and give her friends to sell for her in Mansa. She was giving her friends because she didn’t have enough money for transport costs. . . she made some good money and started going to sell herself. She has changed; her children look very clean and they eat well. She buys new clothes for herself and she looks nice.” ~female beneficiary (Kaputa district) In their own words. . .
- 21. Conclusions and what’s next? • SCTs have potential to decrease gendered inequalities – both for adults and children • Cross-country comparison among multiple settings in SSA allows a more comprehensive picture (across outcomes and program design) – external validity (programs at scale) • Child welfare outcomes may not be driven by transfer recipient in as conventionally assumed – HH structure and context are also important • Much more work is needed! • Still no consensus on how to measure empowerment or in what contexts cash can ‘empower women/girls’ • Missed opportunity for small add ons – such as messaging?: UNICEF engaged in >100 countries on social protection systems • Next frontier: Cash ‘plus’ programming (agriculture/livelihoods or human capital)
- 22. Lesotho and Zambia CGP references: Sebastian, S., A. P. De la O Campos; S. Daidone, B. Davis, O. Niangz and L. Pellerano. (2016). Gender Differences in Child Investment Behavior among Agricultural Households: Evidence from the Lesotho Child Grants Programme. From Protection to Production. FAO, Rome. Forthcoming. Bonilla J, Castro R, Handa S, Nowlin C, Peterman A, Reeves H, & D Seidenfeld on behalf of the CGP Evaluation Team (2016). Cash for women’s empowerment? A mixed methods evaluation of the Zambian Child Grant Program [Innocenti Working Paper 2016-01] Natali L, Handa S, Peterman A, Seidenfeld D, Tembo G on behalf of the Zambia Cash Transfer Evaluation Team (2016). Making money work: Unconditional cash transfers allow women to save and re-invest in rural Zambia [Innocenti Working Paper 2016-02] Further reading: • Transfer Project website: www.cpc.unc.edu/projects/transfer • FAO. 2015. Empowering rural women through social protection. FAO Rural Transformations, Technical Papers Series No. 2 (http://www.fao.org/3/a-i4696e.pdf ). • FAO. 2015. Qualitative research on women’s economic empowerment and social protection. PtoP research guide (http://www.fao.org/3/a-i4420e.pdf ). • FAO. 2015. The impact of cash transfers on the economic advancement and decision- making capacity of rural women. PtoP policy brief (http://www.fao.org/3/a-i4726e.pdf). Works cited
- 23. Transfer Project is a multi-organizational initiative of the United Nations Children’s Fund (UNICEF) the UN Food and Agricultural Organization (FAO), Save the Children-United Kingdom (SC-UK), and the University of North Carolina at Chapel Hill (UNC-CH) in collaboration with national governments, and other national and international researchers. Current core funding for the Transfer Project comes from the Swedish International Development Cooperation Agency (Sida), as well as from staff time provided by UNICEF, FAO, SC-UK and UNC-CH. Evaluation design, implementations and analysis are all funded in country by government and development partners. Top-up funds for extra survey rounds have been provided by: 3IE - International Initiative for Impact Evaluation (Ghana, Malawi, Zimbabwe); DFID - UK Department of International Development (Ghana, Lesotho, Ethiopia, Malawi, Kenya, Zambia, Zimbabwe); EU - European Union (Lesotho, Malawi, Zimbabwe); Irish Aid (Malawi, Zambia); KfW Development Bank (Malawi); NIH - The United States National Institute of Health (Kenya); Sida (Zimbabwe); and the SDC - Swiss Development Cooperation (Zimbabwe); USAID – United States Agency for International Development (Ghana, Malawi); US Department of Labor (Malawi, Zambia). The body of research here has benefited from the intellectual input of a large number of individuals. For full research teams by country, see: https://transfer.cpc.unc.edu/ Acknowledgements
Editor's Notes
- In the first webinar in this series, which I encourage everyone to watch -- Agnes Quisumbing laid out two broad areas with research gaps around gender & social protection. So I am going to pick up on those two themes again here to frame my case studies The first issue has to do with the assumption that women should be targeted as beneficiaries as they are perceived to invest resources in a more family responsive way – essentially that by engaging women (as opposed to men) might be able to help us reach our broader program goals. And while there is a lot of literature which supports this claim, it is often taken from studies on intra-household resource allocation, rather than impact evaluations were benefits are randomized to women and men. Where rigorous studies do exist, we find that the evidence is actually mixed and so I think this is an area we need to continue innovating in terms of research design and thinking through if and how targeting matters. The second issue follows from the first – is essentially interrogating the assumption that if we do target women, that giving them benefits will lead to their empowerment. And I think there is more and more evidence showing that this is possible, however there is a general lack of consensus on what type of programs and in what settings can actually lead to meaningful change. One of the issues here is that we tend to look at a wide range of outcomes, from intra-household decision-making to land ownership or earnings – and label them as empowerment, when in fact they are very different things. Last point: Obviously nearly all impacts could be linked to successful program implementation, but in the cases of several of the CTs in Africa that a lot of the "empowerment" effect is lost because of delays in payments and lack of individual bank accounts (in the case of Rwanda) or control on who gets the cash (the elderly get fooled in some cases when they have to delegate to youth in their HH go to collect the cash)
- So to drive this point home – here is a summary from a review done by our IFPRI colleagues that tried to catalogue across four types of interventions, the evidence for impacts on women’s empowerment, broadly defined. In only one case – that upper right hand corner (qualitative evidence from CCTs, and this comes largely from LAC region) – the review finds sufficient evidence to conclude there is promise of impacts. In the remaining cases, either the evidence is mixed or we need more research, and particularly because program designs can be quite heterogeneous. *CCTs have also been criticized for increasing women’s time burden and reinforcing traditional gender roles.
- So before I dive into the case studies – I think it is useful to say a few words about SP in Africa. This chart is taken from a recent inventory of non-contributory programs done in collaboration with IPC/SP.org identified 127 government implemented programs in 39 African countries. And other kind of flagship publications like the WB SSN document a similar trend. . . Compare to WB SSN - Approximately half (21) SSA countries had an unconditional cash transfer (UCT) in 2010 -- this doubled (40) by 2014 (Honorati et al. 2015).
- When we look at program typologies – we can see a large number of unconditional cash transfer programs, as well as cash for work or conditional cash and fewer of the in-kind or training programs – although these are certainly being run by NGO and other actors. The targeting ranges quite significantly, we see demographic targeting (children/elderly) but also this dual issue of poverty plus some type of vulnerability (OVC, people living with aids and labour-constrained households). But overall, I think the rise and range of institutional (government programming) really speaking to the opportunities available when thinking of utilizing SP for bridging gender inequalities.
- So I am going to spend the rest of the time talking about two case studies – which are both unconditional child grant programs run by governments. The first is the Lesotho Child Grants Program, and here I will talk about work by the FAO team which sought to unpack the issue of gender targeting and HH structure in relation to gendered benefits of children (looking at outcomes of schooling, labor and time use). The second is the Zambian Child Grant Program, and here I will talk about work done by UNICEF and partners AIR which tries to unpack the impacts on women’s empowerment through looking at women’s decision-making, savings and non-farm business participation.
- Both of these case studies, come from a community of research and practice called the Transfer Project, which focuses on providing evidence on government run unconditional cash transfers on a broad range of outcomes – last I counted there were somewhere around a dozen impact evaluations either finished or somewhere in progress, in the countries listed there.
- Here is just a snap shot of some of the evaluations we have ongoing – and some of the details – you can see from the column on the right some of the data collections are still ongoing and many of these also have longitudinal qualitative studies embedded, and there is a lot more information about these on our website.
- So a few more details about the two programs in Lesotho and Zambia. In Lesotho the target group was poor households with a child 0-17, while in Zambia benefits were given to any household with a child under the age of 5 within a poor rural area. In Lesotho, about 2/3 of the beneficiaries were women, whereas in Zambia nearly all beneficiaries women as they targeted the primary caregiver of the young child. On face value, these programs were very simple – unconditional cash transferred either quarterly or bi-monthly. In Lesotho there was messaging included that the transfer should be used for child investments – Both evaluations were RCTs and I am going to be showing you results for two years after program start in Lesotho and 3-4 years after program start in Zambia. Importantly, both of these programs had poverty-related goals – and were not meant to address gender inequalities.
- So first looking at the overall impacts among agricultural households, CGP had strong impacts on schooling enrolment, attendance, time spent in school, and reductions in chores and working on farm. However, these are seen among older, secondary school children aged 13—17 – which are shown here. The impact on primary school children aged 6—12 were small and mostly statistically insignificant, likely because since primary schooling is compulsory in Lesotho, nearly all children 6-12 were already enrolled in school (99-100% likely to be enrolled in school, although 20—27% in this cohort missed school in the last 30 days, and 42—55% were likely to have ever repeated school). The impacts are driven by girls, which is quite interesting because in contrast to many countries in Africa, in Lesotho, boys are actually disadvantaged in terms of schooling (39 percent of girls were in secondary school, compared to 22 percent of boys. At baseline, 77 percent of boys aged 13 to 17 ever-repeated school (20 percent more than girls) and 37 percent of them missed school in the 30 days prior to the baseline survey (10 percent more than girls) – and this is hypothesized to be the case, particularly due to the HIV epidemic, boys are relied on for labor, and are leaving school at higher rates to work in crop and livestock activities. Results driven by girls: Older girls in beneficiary households were 24 percentage points more likely to be enrolled in school in the current year, and 32 percentage points less likely to miss school in the past 30 days compared to older girls in the control group. Older girls also spend 140 minutes more on a typical day in school and almost one hour less on household chores compared to their peers, as a result of the cash transfer. Reduction in on-farm labour among older children driven, however, by boys in CGP households, who worked 1.23 fewer days than boys in the control group. This result is not surprising in the context of rural Lesotho, since older boys are frequently engaged in livestock herding and crop production, while girls typically spend more time on household chores. In addition, boys are disadvantaged when it comes to schooling in Lesotho as compared to girls.
- So, these are interesting findings, but what happens when we look at M versus FHH? Well in MHH, results are generally maintained, while in FHH outcomes actually favored boys – so it could be that female heads were relying more on boy's labor, and that constraint was loosened by receipt of the cash transfer. But going back to look at the MHHs, when men (or the father) got transfers, girls schooling increased as compared to when women (or mothers) got the transfers – at the same time – boys labor on farm decreased and boys input into HH chores increased. So we see these interesting dynamics playing out depending on HH structure and who receives the transfer.
- Overall, evidence like Lesotho is quite interesting, because although there is a promising picture in terms of child and human capital investment, there are also a lot of gendered differences in terms of impacts – it seems to matter who gets funds, but not necessarily in the way we might expect. There is also perhaps an opportunity here, for example the messaging could be slightly tweaked to try to influence gender inequalities between boys and girls to see if this made a difference – if we agree that closing these gaps should be a program goal.
- So just to give you an idea of how the program functioned in relation to the key poverty related goals, the program did a pretty good job. This figure gives a snap shot of program effect sizes in SDs of the control group – so if the confidence interval does not cross zero (the red line) there is a significant program impact. We see increased consumption, food and non-food, Increased assets, relative poverty, etc. The only two that it didn’t impact was schooling & nutrition, though some heterogeneous impacts (positive on those 11-14, just not 11-17 overall).
- So the first thing we did was examine program impacts on intra-household decision making indicators – these are the standard question types that are collected in the DHS and other surveys and ask some version of the question “who in your household typically decides about a number of domains. So we ask 9 different questions and create measures of decision making power if the woman says yes, she makes the decision by herself (sole DM) or with another HH member (joint DM). And you can see that after four years, we found impacts on five out of nine domains. On the bar graph you see the women receiving the transfer says she makes sole or joint decisions on about 7 domains, while women in the comparison group say they make 6.3 decisions. The impact is statistically significant, however in real terms it is small – meaning that women only made and additional about 0.34 decisions over the four years. Collecting DM questions in these types of evaluations is relatively rare – and many evaluations collect these measures because their use is widespread. However, they are by no means perfect and there has recently been more discussion around their weaknesses, particularly in understanding how to interpret them. Range: 0-9
- Although not big impact in DM scale overall, there is a subtle change because when we ask women in their own words, what is empowerment, they equate it with economic empowerment. 30 IDI among women, 10 with partners – stratified on changes in DM, marital status and program participation.
- So then the next step was to look at economic empowerment directly, instead of just decision-making. Banjerjee: 6 countries microfinance and found programmes didn’t deliver on main objectives (savings and economic impacts) – not even on main outcomes, let alone gender. People aren’t able to save, etc.
- Large impacts on savings and non-farm enterprise (small business) – here not specifically women, but in paper we should that these are largely women-operated businesses, and a lot is due to savings accumulated through CGP. Savings larger in 24 month, less in 36 month, but still there. Consistent with story of investing in business. The increase in women’s cash savings appears to be linked to greater involvement in NFEs: The CGP has an impact on NFEs (17 and 15 pp at 24 and 36-months) This impact is partly caused by women’s cash savings accumulated due to CGP (14-21% of total effect of CGP on NFE is due to women’s savings))
- This narrative is important because there were no program goals of empowering women.
- This sums up that story. Enumerator asking a woman what it means to be empowered in her community. So she might not have gained DM/bargaining power in the household, but she has higher economic empowerment. Yet status in HH might not have changed much.
- So in conclusion, we feel there is a lot of promise for cash transfer in particular to decrease gendered inequalities, and in particular the types of cross-country evidence from programs at scale is encouraging. However there is a lot of work still to be done – particularly on randomization benefits to get better information on how targeting matters, getting empowerment measures right and manageable so that they can be added and replicated in other evaluations. And I think here, there is an obvious need to move beyond cash to think about cash plus and other addon programming which can facilitate impacts on women’s empowerment – which I don’t think we are doing enough of at the moment in ways which are testable. Particularly for large organizations like UNICEF and FAO who work in many different countries. Design improvements, could be as simple as location/timing of services or utilizing technology (mobile services) in case women are not able to travel or due to family or other pressures. WBER Mayra Buvinic & Rebecca Furst-Nichols on promoting Women’s Economic Empowerment (productivity and earnings) - a lot of variation depending on the client (very poor, marital status, age). Social constraints.