Aga Khan Development Network

AKF home

Facts at a glance



Rural development

Civil Society


Savings Groups

Current Projects


Cross-cutting issues

Prospective grantees

Reading for Children

Other AKDN agencies


The Permanence and Value of Savings Groups in CARE Kenya’s COSAMO Programme

> View all Case Studies

Beyond Financial ServicesClick on the image to download the brochure
(PDF, 3MB)
This case study explores the commonly held belief that Savings Groups are capable of functioning successfully after external agencies have phased out their support. It examines Savings Groups in the Nyanza Province of Western Kenya two years after the CARE support programme ended to determine: i) how well groups survive when functioning independent of facilitating agency support and ii) how groups can be used as a platform for the delivery of other activities that add value to group membership.

CARE’s Community Savings Mobilisation project (COSAMO) in Kenya was an adaptation of CARE’s work with SGs in Zimbabwe. The project was designed to provide a sustainable increase in household access to savings and credit in order to strengthen livelihoods. It was part of a larger multi-service USAID-funded project. COSAMO ran from 2004 to 2008 and created 388 groups with 8,800 members, in six areas of Nyanza Province.

Few facilitating agencies promoting SGs leave their groups completely independent after the end of the training period. Most facilitating agencies maintain contact for various purposes and so occasionally visit SGs, presumably reinforcing the financial activities of the group in the process. The COSAMO programme is one of the few SG programmes that effectively severed all ties with the groups post-project. Thus, it presents a rare laboratory for examining the issue of SG sustainability. The COSAMO groups also offer an excellent opportunity to learn what value-added activities emerged after project implementation ended.

Rather than identifying a representative sample of COSAMO’s groups scattered across the six project areas, the research team completed an exhaustive study of all 50 groups that CARE reported to have trained in Rachuonyo District - a mostly agricultural area with a significant minority of inhabitants earning their living from fishing. Rachuonyo is also among the areas of Kenya most strongly affected by HIV/AIDS. The researchers discovered that six of the 50 reported COSAMO groups had never been formed and identified 37 additional SGs which had formed since the project ended. They also believe there are more new groups they did not find.

Sustainability and Replicability
This study categorizes sustainability in three ways:

Group Sustainability:
The study overwhelmingly found that groups continued to exist after the training ended. Of the original COSAMO SGs, only one group (out of 44) had stopped meeting altogether. However, even this failure was mitigated somewhat: the group maintained a bank account and intends to restart operations. The other 43 groups reported meeting regularly.

This finding of COSAMO SG continuity does not illuminate the sustainability of the group approach as much as one might expect. Many of the SGs in question had existed before the COSAMO project began, usually for several years. Twenty-five percent of all groups were existing women’s groups, self-help groups or community-based organizations (CBOs); twenty-three percent were Rotating Savings and Credit Associations (ROSCAs); and seven percent were communal businesses. Therefore, it is not surprising that groups continued to exist after the end of COSAMO. In fact, if many pre-existing groups had failed, it would have suggested that the COSAMO methodology had an adverse effect on sustainability. The prior existence of many of the Rachuonyo COSAMO groups may be a factor in their high rate of survival.

Permanence of member access
If there are sufficient numbers of groups, it is not particularly important whether an individual group continues to exist, but only whether group membership is available to anyone who would like to be a member. Since only one group in this study had failed, there was little basis to examine the issue of the permanence of access for members of failed groups.

Permanence of the concept
The study provided more conclusive evidence of the permanence of the SG concept: group members have formed a large number of new groups, independent of CARE, since the end of the project.

The study found that group members had created additional groups in at least four ways: through i) fission of large groups into smaller ones; ii) splintering, in which some members left a group to form another; iii) social entrepreneurship, in which dynamic individuals created new groups; and iv) ROSCA upgrading, in which SG members who were also members of ROSCAs, taught ROSCA members to adopt SG practices.

Value Added
SGs provide valuable savings and credit financial services for members. In Rachuonyo District, many groups have organised an additional service: informal insurance through a social fund established through member contributions. Members particularly value the social fund, which provides grants to members, often in emergency situations. The social fund has become a vital grassroots insurance and risk management system, particularly important given the high prevalence of HIV/AIDS in the community. In addition, SG members are engaged in many other activities yielding important benefits beyond financial services.

To understand the great variety of these activities, the report identified five “pathways” through which value is added to groups: i) activities initiated by an outside agency; ii) links to an outside agency initiated by the group; iii) group income-generating activities (IGAs); iv) financial assistance or other social support offered to members, often through a group-managed social fund; and v) groups reaching into the community to help others, moving members from being beneficiaries to becoming benefactors.

Thus, contrary to assumptions that most value-added activities are brought to the groups, activities and structures generated by the groups themselves often provide value that is greater and broader than those of externally-introduced components. Group-generated activities and structures provide a social safety net and expand opportunities for members to increase their income. In addition to the social fund, social safety net mechanisms include ad hoc assistance for members, their families, and in some cases, other community members, to mitigate the impact of household catastrophes, including illness and death. The groups often serve as a solidarity group, counselling in times of need.

Lessons Learnt
There is a broad gap between the popular conception of SGs and the field reality in Rachuonyo District. The study discovered that in most cases, COSAMO provided training to previously existing groups; many groups continue to have ROSCAs in addition to their SG activities, and both have their advantages. Some groups carry out activities which are counter-intuitive, including non-proportional distribution of assets at the time of share-out. The social agenda is often as important, or more important, to group members than the financial services the group provides.

These findings have implications for on-going research and for training new and existing groups. First, research and training should include more emphasis on the social aspects of the group. Second, since groups continually innovate, research and training should shift to include more emphasis on outcomes than procedures. Finally, the study recommends additional research on topics including: how the prior existence of groups affects the durability of SGs; the factors that influence security of SG assets; group creation of income generating activities and the factors that lead to their success or failure; and how best to facilitate group outreach to the community in a manner that contributes to the success of SGs.

This brief is based on research and case study sponsored by the Aga Khan Foundation’s Savings Groups Learning Initiative, written by Marcia O’Dell and Paul Rippey in June 2010. View the full case study

Return to top