Investing in Women Smarter: Moving from Good Intentions to Better Design

There is no shortage of commitment to women’s empowerment in Asia. What is sometimes in shorter supply is honest conversation about what actually works, why some well-funded efforts fall short, and how we — funders and non-profits together — might redirect our resources toward the design problems that matter most.

Our goals to empower women and children

Farida started making batik fabric in her kitchen in Yogyakarta when her youngest child was two years old. She had trained as a teacher, but the nearest school was an hour’s commute away and she had nobody to care for her children during the day. So she built a business around work she could do without leaving the house. By the time we heard her story, she was supplying three retailers in Jakarta and one buyer in Singapore, and employing four other women from her neighborhood — all of them with similar constraints: skilled, motivated, geographically tethered.

Farida’s story gets told at development conferences as evidence that investing in women works. And it does — in her case. But the question that actually moves things forward is not ‘does it work?’ It is: ‘what specifically worked, for her, and what would have failed?’

We ask that question not to raise doubt about the value of women-first investment — we believe in it deeply — but because the banner of ‘empowering women’ has grown wide enough to cover almost anything. Microfinance. Leadership training. Mobile banking. Scholarships. Artisan marketplaces. Digital skills courses. Some of these approaches transform lives. Some, despite good intentions and real funding, quietly miss the people they were designed for.

This post is an invitation to think through that distinction together. We are sharing what patterns we notice and what questions we are sitting with — not as a verdict on what works, but as a starting point for the kind of conversation that tends to produce better design. We would genuinely welcome your pushback, your experience, and your additions.

The goal is not less investment in women. It is smarter investment — investment designed around real lives, real constraints, and real evidence. That shift starts with asking harder questions together.

Pattern One: Women Reinvest Income — but Control Is What Unlocks the Multiplier

One of the most consistently replicated findings in development economics is that women tend to direct a substantially higher share of income toward household wellbeing — food security, children’s health, school fees — than their male counterparts. The World Bank, UN Women, and the International Finance Corporation have each documented this pattern across countries and contexts. It is the empirical foundation for the argument that investing in women creates broader social returns than equivalent investments elsewhere, and it is well-supported.

The McKinsey Global Institute has estimated that closing gender gaps in economic participation across Asia-Pacific could add trillions of dollars to regional GDP over a decade. Even allowing for the inherent uncertainty in macroeconomic modelling, the magnitude is significant. Women’s economic exclusion in Asia carries an enormous and largely uncounted cost.

Here is where it gets more interesting: the multiplier effect is real, but it is not automatic. Research on microfinance programs across South and Southeast Asia has found that when loan repayments pass through a male family member, the household wellbeing effects are noticeably smaller than when women manage repayment directly. The money was the same. The control structure was different. The outcomes diverged.

Conditional cash transfer programs — used in various forms across Indonesia, the Philippines, and Cambodia — tell a similar story. Well-designed programs that centre women’s decision-making show strong intergenerational effects on child nutrition and school enrolment. Poorly designed ones create new administrative burdens that fall disproportionately on women without meaningfully improving their agency. The difference is almost always in the design, not the budget.

What Does This Suggest We Actually Do?

The most practical shift here is surprisingly simple: ask, at the point of design, whether the woman controls the resource — or whether it passes through a structure that dilutes her agency before it reaches her.

For funders, this might mean preferring programs that disburse directly to women and track decision-making autonomy, not just income levels. It might mean asking grantees: how do you know the woman in this program is actually in control of how the money is used?

For non-profits, it might mean re-examining the logistics of your program delivery. Does the way you run repayments, savings groups, or training attendance inadvertently give male household members veto power over women’s participation? These are often unintentional design choices that can be redesigned.

We recognize this is not always straightforward. In many communities, requiring a woman to manage finances independently of her household can create conflict that undermines the program’s goals entirely. There is genuine tension between what the evidence suggests maximizes impact and what is sustainable within a given social context. We do not have a clean answer to this tension.

What we think is worth holding onto is the question itself: whose agency does this design actually serve, and how would we know?

Pattern Two: Education Has Run Ahead of Participation. The Gap Is Structural.

One of the less-discussed facts about gender inequality in Asia is this: the gender gap in primary and secondary education has largely closed across much of the region, and in countries like Vietnam, Thailand, Malaysia, and the Philippines, women now make up the majority of university graduates. By the conventional theory of change — invest in education, create opportunity — this should have produced a dramatic reduction in the gender gap in economic participation.

It has not. Women in Asia remain more likely than men to be in informal, unprotected, low-wage work regardless of their educational level. In parts of South Asia and rural Southeast Asia, highly educated women remain outside the formal economy entirely. The gap between what women know and what the economy allows them to do is wide and, in some contexts, widening.

What this pattern suggests is that the primary barrier to women’s economic participation in much of Asia is not a skills deficit. It is structural. And a significant share of women-first investment — focused on training, scholarships, and skills development — is being directed at a barrier that is not the binding constraint.

Structural Barrier	What Smarter Investment Looks Like
Mobility constraints (cultural norms, safety, transport)	Design for where women already are: home-based or community-adjacent delivery, digital tools that work on shared family devices, programs embedded in existing trusted networks like savings groups or faith communities.
Discriminatory hiring practices	Invest upstream in employer engagement and norm change, not only in candidate training. Support organisations that work with businesses to shift hiring culture, not just prepare women for a market that may still reject them.
Lack of formal identity documents	Partner with programs that help women navigate registration and documentation as a precondition, not an afterthought. A woman who cannot open a bank account cannot benefit from most financial inclusion programs.
Marriage and maternity penalties	Advocate for and fund policy-level change alongside program-level work. Where legal reform is slow, support informal alternatives: community savings systems, peer lending networks, and business structures that don’t require individual registration.
Concentration in unprotected informal sectors	Invest in cooperative models, and digital marketplaces that allow informal workers to operate with more security and market access — without requiring formalisation as a prerequisite.

We want to be careful here not to suggest that education and skills investment is wasted — it is not. However the evidence increasingly suggests that its returns are lower when it is not paired with investment in the structural conditions that determine whether those skills can be put to use. The highest-leverage investments we have seen tend to address education and structure simultaneously, rather than sequencing one before the other.

We are genuinely curious about what others are seeing in their contexts. The picture is different in urban Dhaka than in rural Sulawesi, different for garment workers than for home-based artisans, different for adolescent girls than for women returning to the workforce after a decade of caregiving. If your experience suggests a different pattern, we would like to hear it.

The most effective investments we have seen address education and structural barriers together — not one before the other. Which of those two receives the larger share of funding in the programs you know?

Pattern Three: The Care Burden Is a Problem. It Can Be Designed Around.

Across Asia, women perform roughly four to five times more unpaid care and domestic work than men. This is often framed as a cultural observation, something to be acknowledged before moving on to the ‘real’ program design. We think that framing undersells its importance.

The unpaid care burden is not context. It is the most consequential constraint for any program, tool, or investment targeting women in the region. Get the design right for this constraint, and you reach the women who most need your program. Get it wrong, and you systematically exclude them — not visibly, not all at once, but through quiet dropout rates and low uptake that rarely appear in output reports.

Programs that require women to attend multi-day workshops in city centers. Microfinance schemes with weekly morning meetings. Digital platforms that assume a personal smartphone and reliable electricity. All of these, by design, select for women who already have fewer caregiving responsibilities, more household support, or better access to resources. The women who complete them might not the women who most needed them. The gap between intended beneficiaries and actual beneficiaries is real, significant, and largely invisible.

The encouraging part of this pattern is that it is designable. The care burden is not going to disappear quickly — but many of its effects on program participation can be substantially reduced through deliberate design choices.

Practical Design Moves that Make Differences

• Schedule around caregiving reality, not organizational convenience. Ask: at what time of day can a woman with two young children and no household help actually attend? That answer — not your team’s preferred meeting time — should set the schedule.

  
  

• Build in asynchronous options. Voice messages, recorded videos, and offline mobile content allow women to engage on their own time. This is not a digital-age luxury; for many caregivers, it is the difference between participating and not.

  
  

• Work through networks women already belong to. Village savings groups, mother’s groups, faith community circles, and cooperatives already have the trust and attendance habit. Embedding programs here is almost always more effective than building new attendance infrastructure.

  
  

• Budget for childcare and transport stipends. These are often treated as secondary costs. In many contexts they are the primary barrier. A small stipend that makes it possible for a woman to attend is often higher-leverage than the program content itself.

  
  

• Involve women in the design. This is said often and done rarely with genuine depth. It means asking women who look like your intended beneficiaries to walk you through a typical day before you design the schedule, the location, the format, or the technology. Their answers will change your program.

We want to name something honestly here: redesigning for the care burden often costs more in the short term. Childcare stipends, community facilitators, asynchronous technology, flexible scheduling — these require resources. Funders who want programs to serve the hardest-to-reach women need to be willing to fund the design features that make that possible, rather than selecting for programs with the lowest cost-per-beneficiary. Cost efficiency and equity of reach are sometimes in tension, and pretending otherwise does not serve anyone well.

Redirecting Resources: What Smarter Women-First Investment Could Look Like

We want to be clear about what we are and are not saying. We are not arguing for less investment in women’s empowerment. We are arguing that a meaningful share of current investment is going to the wrong layer of the problem, and that redirecting some of it toward design quality, structural enablers, and community co-creation would produce significantly better outcomes without necessarily requiring more total funding.

Here are the specific redirections that seem most supported by the evidence and by what we have observed in the field. We offer these as starting points for conversation, not as settled conclusions.

For funders and donors:

• Fund the design, not just the program. The best programs we have encountered spend a meaningful share of their budget on community listening, iterative design, and user testing before they launch at scale. This is not overhead — it is what makes the difference between a program that reaches Farida and one that reaches a different, easier-to-serve woman instead.

• Lengthen time horizons. Structural barriers do not shift in twelve-month grant cycles. Programs addressing mobility constraints, hiring norms, or identity documentation need multi-year commitments to show results. Short cycles systematically disadvantage the highest-impact work.

• Track agency, not just activity. Output metrics (number of women trained, number of loans disbursed) can mask programs that reached participants without meaningfully improving their economic agency. Ask grantees how they measure whether women are in genuine control of the resources and decisions the program was designed to shift.

• Invest in peer learning across grantees. Some of the best design knowledge in this space exists in the heads of practitioners who have learned from failure but have no platform to share it. Funding learning networks and knowledge exchange across grantees often produces more value than another standalone program.

• Be willing to fund the enablers. Identity documentation support, community facilitator training, transport stipends, and childcare costs are often the actual binding constraint. Programs that bundle these with skills training consistently outperform programs that treat them as outside scope.

For non-profit leaders and program designers:

• Map the actual woman before you map the program. Before any design work begins: who specifically are you trying to reach? What does her day look like? What are the decisions she makes, and which decisions does she not get to make? The more specific this picture, the more useful it becomes as a design constraint.

• Conduct an access audit of your existing programs. Who is completing your programs, and who is dropping out? When in the program does dropout happen? These patterns almost always reveal a design problem — a scheduling issue, a technology barrier, a documentation requirement — that is fixable once it is visible.

• Build in structural advocacy alongside direct service. If the barrier you face is discriminatory hiring or lack of identity documents, direct service alone will not solve it at scale. Consider what role your organisation can play in the policy and norm-change work that creates better conditions for your programs to succeed.

• Be honest with funders about what the evidence says. If your theory of change relies on skills training producing economic participation, and the evidence in your context suggests the structural barriers are the binding constraint, say so — and propose what addressing those barriers would require. This is harder than it sounds when funding relationships are involved, but it is the conversation that leads to better design.

• Share what is not working. The field’s collective learning is held back by the understandable incentive to report success. Finding structured ways to share failures — with your team, your board, your peers, and where possible your funders — is one of the highest-value things an organisation can do for the broader ecosystem.

Where We Are Still Uncertain — and Why That Matters

We want to conclude this section honestly. The patterns we have described are real and reasonably well-supported, but the translation from evidence to practice is never clean.

Context matters enormously. What works in urban Bandung may not work in rural Mindanao. What serves women in their forties navigating re-entry to the workforce is different from what serves adolescent girls still in school. Cultural dynamics around women’s mobility, household decision-making, and economic participation vary not just across countries but across communities within the same city.

There are also genuine tensions that we do not think have easy resolutions:

• Between designing for women’s agency and designing for what is acceptable within a household or community. Moving too fast on agency can create backlash that harms the women you were trying to support. Moving too slowly can trap programs in accommodating norms that perpetuate the very inequalities they exist to address. How do you navigate this in your context?

• Between cost efficiency and equity of reach. The women who are hardest to reach — those with the most caregiving responsibilities, the least mobility, the least access to technology — are often the women whose lives would benefit most from the program. Reaching them costs more. How are funders and non-profits thinking about that tradeoff?

• Between local solutions and the risk of accepting local inequalities as given. Community-led design is generally better design. But communities also contain power structures that marginalise certain women — by age, by marital status, by caste or ethnicity. How do you centre community knowledge without inadvertently centring the voices of those who already have power within it?

We do not have settled answers to these questions. We are working through them in the same way we suspect many of you are. What we do think is that naming these tensions explicitly — in team conversations, in board meetings, in funder relationships — is more useful than pretending they do not exist.

An Invitation, Not a Conclusion

Farida did not succeed because someone gave her a workshop on breaking gender stereotypes. She succeeded because she had a skill, a phone, a neighbour who became her first customer, and a market that cared about the quality of the fabric more than where it came from. The investment that helped her worked because it found those specific conditions and made them slightly more possible.

That is what we mean by smarter investment: not more programs, not less funding, but a more honest reckoning with what the binding constraint actually is for the specific woman you are trying to reach. Sometimes it is skills. Sometimes it is access to finance. Often it is something more structural — the time she does not have, the mobility she is not permitted, the market that does not yet know she exists.

We'd Like to Hear from You

If you are a non-profit leader or program designer working on women’s economic empowerment in Asia, we are curious: what is the binding constraint you keep running into that your current program design does not fully address? What design move has made the most difference in your context?

If you are a funder or donor, we would love to know: what would make it easier for you to invest in the design layer — the listening, the iteration, the structural enablers — rather than just the program outputs? What makes that difficult within your own funding structures?

We are not asking because we have the answers. We are asking because the answers are distributed across the people doing this work, and better thinking tends to emerge from shared conversation than from any single organization’s perspective.