Community groups are a cornerstone of CARE’s approach to development, and we use many permutations to get to the outcomes we seek. Women-only VLSAs, men’s care groups for children, mixed producer groups, groups for community leaders. You name it, we’ve tried it somewhere!
Recently, we took a look at our data on community groups across 3 multi-country projects, to see what was most effective for women’s economic empowerment.
The answer might surprise you: involve men, but put women in leadership.
We found that putting women in charge is good for more than women’s equality. Our results also show involvement in community groups led to a 45% improvement in women being involved in decision making, a 46% improvement in income and a 29% increase in production.
Groups that had roughly equal numbers of men and women participating, but women leaders were most successful. The next most successful was balanced participants and balanced leaders. Groups without women leaders were least successful.
Looking at the Sustainable Dairy Value Chain Project, Pathways, and Link Up projects funded by the Gates Foundation, we examined results from hundreds of collectives to see what we could learn. These projects cover 7 countries in Africa and Asia, and let us look at trends in larger contexts than just one program or geographic area.
What have collectives accomplished for women’s economic empowerment?
- Improved income: Women involved in one of CARE’s projects with collectives see an average increase in income of 9.8% per year. According to the World Bank, women in similar communities who do not have CARE projects get a typical increase of 5.6% per year. Groups that are made up of mostly women are the most successful. They see an increase in income of 28% more per year than the groups with few women.
- Doubled women’s control over income: It’s not just that women earn more money, but they are also more likely to control it. Women in collectives with women leaders are nearly twice as likely to be economically empowered and have control of their income as women without access to collectives. About 32% of women outside of collectives are able to control their income.
How? Collectives alone are not enough to solve the problem; the gender composition of both the group and its leaders matters. Only 34% of women in mostly male groups with male leaders will have control over their own income. But having more women in groups and putting women in leadership changes the game. 69% of women in collectives with women leaders, and 62% of women in collectives with mixed gender leaders can control their income.
- Improved equity in household labor: Women who are not in collectives work on average twice as many hours in a day as men do. Women in female-dominated groups get an additional 2 hours of support from men or mother’s in law for household labor every day. But the biggest gain in women’s time poverty is for women in collectives with gender balance. They get an additional 4 hours a day of help with work at home, which frees them to pursue other opportunities, including economic empowerment.
- Created equity in access to resources: A typical woman outside of a collective can only access half of the resources that men do in her community. This limits her production, her income, and her opportunities. But women in gender balanced groups that include women in leadership roles have equal access to resources as the men in their communities.
- Improved women’s production by 24%: Women in collectives with a gender balance and both male and female leaders show the highest improvement in production—up to a 29% improvement in production. The effects aren’t as strong in male-dominated groups, where production increases are only 11%. But in no cases did women’s production go as high as men’s. Women in collectives produce about 92% of what men produce, relative to 80% for women who aren’t in collectives.
- Better development outcomes: regardless of the programs’ goals, mixed gender groups with women leaders are able to achieve those outcomes more than 3 times more effectively than the average collective. By contrast, women in mostly male groups with male leaders are likely to see results nearly 4 times worse than the average group.
What have we learned?
- Put women in charge: In general, the groups that are most successful at meeting the projects’ goals—these differ for each project—are mixed gender groups with women leaders. In some instances, having men and women lead gives better outcomes, but mostly male groups with male leaders consistently score lowest on economic empowerment outcomes.
- Include Men: Women-only groups with women’s leaders do show success (like a 41% increase in income), but gender balanced groups with women leaders often outperform them (showing a 46% increase in income). While gender balanced groups with women leaders outperform the average by more than 3 times, women’s only groups are only twice as good as the average. Gender-balanced groups are especially good at improving equitable decision-making.
- Acknowledge tradeoffs and pick your goals: Different group compositions get us to different results. Women in mixed gender groups with mixed gender leaders will get 4 extra hours of help from their families a day for household chores, but they are also more likely to accept gender-based violence as the norm. Men in mostly male groups have the highest success increasing production.
- Focus on women getting equal benefits: Even in groups that are the most successful for women and women’s economic empowerment, men still get higher benefits than women do. The difference is not huge, but it is still there, and is something we need to overcome.
Methodology: To conduct this research, CARE worked with DataAssist to look across the datasets that each of the three projects assembled for their collectives. Combining these datasets allowed us to look at results in 7 countries, and examine trends beyond individual community and country contexts. The results described throughout this document are based on a series of data manipulations and statistical models that combine all available data from the three projects. The advantage to combining datasets from across projects include larger sample sizes which leads to reduced sampling errors and the ability to control for within person, within group and within country variability, which makes it more likely to accurately locate subtle effects.
Want to learn more? Look at the full report, or check out the research methodology for how we got to these conclusions.