Poor and disadvantaged people often have weak organizations and low rates of participation in community associations, potentially limiting their political representation and economic advancement. Many policy initiatives aim to build civic participation among the disadvantaged by funding local community associations. Such funding could potentially empower the disadvantaged by giving them a larger voice in community and political affairs. But funding could also shift the interests and membership of such organizations away from the grassroots. Discerning the impact of funding on community organizations is challenging, however, because it is not easy to disentangle the multiple factors that could explain connections between weak organizations and low socioeconomic status, or between funding and higher socioeconomic status if groups with higher status members are better able to attract funding.
Approximately half of all adult women in many districts in Kenya belong to a women’s group, suggesting that such groups may be the most widespread form of secular community organization that traditionally does not rely on support from the government or foreign donors. These groups conduct a range of activities, including income-generating projects, sponsorship of rotating savings and credit associations, and emergency assistance. Many of these activities have a social insurance component. For example, most groups distribute earnings equally among all members, independent of individual labor contributions, providing insurance to sick or elderly members who may not be able to work with the same capacity as others.
The research teams tackles the web of potential causal relationships between civic group participation and socioeconomic status by exploiting an opportunity created by random phase-in of a program that provided funding designed to strengthen women’s community associations in rural Kenya. The program was designed and conducted by International Child Support (ICS), in consultation with Kenyan women’s groups and local officials from the Ministry of Agriculture. ICS identified 100 women’s groups in Busia and Teso, two poor rural districts in western Kenya, through lists provided by the government and interviews with local Community Development Assistants. Eighty of those groups met project eligibility criteria, which required that groups meet regularly and already be engaged in group-based agricultural activity.
The 80 eligible groups were separated by region and half were randomly selected to receive the program immediately, while the other half were notified that they would not receive funding but that the NGO would try to fund them in the future. Prior to the beginning of the NGO program, the groups studied were largely autonomous and did not receive outside funding or government support. The program paid for leadership training, agricultural supplies such as seeds, and agricultural training at the Ministry of Agriculture’s farmer training center. The program spent $674 per group or an average of $34 per member, roughly 10% of Kenya’s per capita GDP at the time. Three sets of surveys were administered to the groups, including a baseline survey, a second survey administered 14 months after the program began, and a final survey administered six months later.
Results and Policy Implications
The researchers found little evidence that outside funding expanded organizational strength. Additionally, there were no significant differences between program and comparison groups in attendance rates at general meetings, meeting frequency, and financial support of needy members, among other indicators of group strength. However, the research team did find substantial evidence that funding changed group membership and leadership, weakening the role of the disadvantaged. The program led younger, more educated, and better-off women to enter the groups. These new entrants, along with men and more educated women, were more likely to assume leadership positions.
Additionally, there was an increase in older women leaving the groups, often due to conflict. The research team generalized the results through a formal model showing how democratic decision-making by existing members of community groups can lead to long-run outcomes where the poor and disadvantaged either do not belong to any associations or belong to weak organizations.
The researchers further point out that the entrance of higher-status individuals into groups could help groups promote their interests externally, which suggests a potential trade-off between preserving participation for disadvantaged women and allowing for greater participation in electoral politics and formal government structures. The researchers suggest a series of policy approaches that could soften this trade off, including concentrating assistance on the very weakest groups and spreading it around more generally, rather than concentrating a relatively large amount of assistance on a few groups. The results of this study suggest the need for an evidence-based approach to group funding programs. Aid programs should be phased in gradually, their results measured, failed programs abandoned, and successful programs scaled up. The researchers argue that such an approach would help ensure resources are better used and would make it possible to get a better sense of when outside funding and support is most likely to succeed in promoting equity of participation in civic life.
 Hammerslough, C. R. 1994. “Women’s Groups and Contracep- tive Use in Rural Kenya.” In The Onset of Fertility Transition in Sub-Saharan Africa, ed. T. Locoh and V. Hertrich. Liege: Derouaux Ordina Editions, 267–87.