Chama Networks

1. The Question

There has been much celebration of the journey that Kenya has been through when it comes to financial inclusion: there was an increase in the number of adult Kenyans who were accessing at least one formal financial product from 41.3% in 2009 to about 75.3% in 2016, thanks to mobile money in particular. However, while there has been a formalization journey, about one in four adults in Kenya, about 5.6 million individuals in 2013 as per the last estimate on this subject, was saving and borrowing through informal groups.

This paper asks: why have chamas not been crowded out by the rapid expansion of formal financial access, and what does their persistence reveal about the specific microeconomic functions formal banking in rural Kenya still fails to perform? Answering this requires moving past a simple “access gap” model, where informal finance exists only because formal finance is unavailable, toward a model of chamas as solving distinct problems that formal products, even when accessible, do not solve as well.

2. The Microeconomic Mechanism

Three separate mechanisms explain why a chama can outperform a formal bank account or loan product for a given rural Kenyan household, even when both are available:

Commitment and self-control. Regular savings accounts are liquid since one is free to withdraw money whenever one wishes; this is actually the main issue for households facing social pressure to spend rather than save (competing demands of relatives, temptation goods, or mere present-biasedness in intertemporal decisions). A ROSCA transforms the act of saving into a social obligation where failure to make a contribution during a predetermined meeting results in being sanctioned socially by other members who are aware and observing. This is the only commitment mechanism that cannot be acquired in the formal sector since there is no regular Kenyan savings account that punishes you for failing to make voluntary contributions as happens in a chama’s meetings every week.

Social collateral substituting for missing physical collateral. Regular savings accounts are liquid since one is free to withdraw money whenever one wishes; this is actually the main issue for households facing social pressure to spend rather than save (competing demands of relatives, temptation goods, or mere present-biasedness in intertemporal decisions). A ROSCA transforms the act of saving into a social obligation where failure to make a contribution during a predetermined meeting results in being sanctioned socially by other members who are aware and observing. This is the only commitment mechanism that cannot be acquired in the formal sector since there is no regular Kenyan savings account that punishes you for failing to make voluntary contributions as happens in a chama’s meetings every week.

Risk-pooling for lumpy, unpredictable expenditures. A number of the major costs faced by rural families, such as funerals, weddings, school costs, and medical emergencies, are indivisible and highly unpredictable, precisely the type of shock that can only be met inefficiently through an accumulated savings account, either because the accumulated balance will not be adequate, or because the shock will come before the accumulation takes place. A lump sum payment through a rotational system, or an emergency reserve fund in the welfare chamas, makes sure the shock is met immediately with a large amount rather than waiting for the accumulation of the necessary saving. There are no insurance products designed for such shocks in rural Kenya.

3. Evidence

Chamas operate at a genuinely large scale, comparable to formal microfinance. Surveys indicate that the yearly intermediation from ROSCAs and ASCAs in Kenya is around KSh 60 billion (USD 860 million), which at the time of calculation was larger than the total savings in Equity Bank, then the biggest and most vibrant microfinance oriented bank in Kenya. This is not a small or insignificant channel of financial intermediation but a major one in terms of value.

Participation has been strikingly stable despite a formal-inclusion boom. Based on FinAccess surveys, almost 25% of adult Kenyans, roughly 5.6 million people, stated that they used either a ROSCA or an ASCA to save and borrow money in 2013, which is just slightly less than what was seen in 2006 and 2009 even as the use of formal financial services grew from around 41% (2009) to 75% (2016). Should the purpose of chamas been merely to plug any gap in access, the stability would have been difficult to achieve; instead, it suggests chamas do something else.

Chama membership is concentrated among exactly the populations these mechanisms would predict. Women are twice as likely as men to be part of ROSCAs in some regional studies (66% of women compared to 30% of men in one case from Central Kenya), while households headed by matriarchs are more inclined towards merry-go-round and welfare chamas, which is consistent with the fact that women experience greater commitment problem because of competing demands on household resources, in addition to being less access to collateral-based financing through formal institutions. In addition, ROSCA participants are more likely to be self-employed and informal sector workers compared to formal sector employees, which is consistent with collateral substitution theory since informal sector workers are the least likely to have access to formal financing.

Chama function varies systematically by purpose, and welfare/insurance functions dominate: The recent classification of the types of Kenyan chamas shows that the welfare chamas (which include the people who contribute to cover emergencies such as school fees or funeral expenses) form the majority of chamas (about 47%), followed by classic rotating merry-go-round ROSCAs (40%), and savings and investment clubs (13%). Thus, this distribution helps us in understanding that the risk-pooling principle is the most important factor leading to the formation of chamas, which is even more than the rotating credit principle which has always been emphasized by the academia. Geographically, the merry-go-round chamas are widespread in the agricultural regions of Kenya, namely, Nyanza, Eastern and Western provinces of Kenya.

Formal institutions have responded by trying to formalize, not replace, chamas. Instead of viewing savings groups as a market failure that should be competed out of existence, Kenyan institutions have focused their attention on bolstering them: FSD

Kenya’s savings group support initiative had engaged over 14,700 savings groups and over 384,000 members by 2015, while financial institutions have even introduced mobile products specifically tailored to exist atop chama systems (e.g. group savings mobile products). In this sense, it is clear proof that the sector sees chamas as a long-term solution whose digitalization is worth pursuing rather than a temporary stopgap measure doomed to disappear.

4. Argument

However, the framing used in the paper title, “substitutes for formal banking,” is only partially correct. In regard to asset-building credits, which is the specific use of credit where formal banks or collateral substitutes as described in the paper on the land titles are designed to provide, chamas are in fact substitutes created out of necessity: households participate in them precisely because formal credits are either inaccessible or unsuitable from the perspective of collateral availability. However, for the purposes of commitment savings and insurance against lumpy shocks, it seems that chamas are not a temporary substitute at all: rather, they represent a more sophisticated tool for these goals than anything offered by the formal sector, hence participation in chamas did not decrease despite the almost doubling of the penetration of formal accounts. Chamas' key technology, publically committed savings through social enforcement and peer-supervised trust, is a solution to issues that cannot be solved with the key technology of a bank account, which is individual liquidity.

This has a close parallel with the earlier result from this research line on M-Pesa and SME formalization: increased access via mobile money did not address the underlying institutional conflicts (regulatory compliance costs in the former, and commitment/social enforcement in the latter) that drive the behavior. In each case, the expansion of fintech access formally and informally is not a paradox that must be solved through “greater access.” Rather, this reflects the reality that formal and informal financial technologies are addressing separate issues and will continue to coexist until the formal products are re-designed to address the specific conflicts that chamas address.

5. Implications

• For fintech product design: The best way forward in terms of formal-informal sector integration is not to replace chamas, but to digitize their central idea, commitment savings through the group, as some mobile “group savings” initiatives have sought to do, not simply give people access to personal accounts.
• For microfinance and credit policy: Since chamas provide real economic leverage where formal collateral requirements (such as land titling) cannot, formal lenders working directly through the existing and established structures of chamas can lend to a greater extent among rural people than formal lenders who only lend on the basis of land titling.
• For future primary research: It is a very good choice for a primary data-based research project by a student researcher who has access to people in that community, a survey or an interview conducted with members of the chamas can help find out whether self-control, lack of credit, or risk pooling was the reason behind forming the chamas, as suggested by the evidence presented in this paper.

6. Limitations

The most cited quantitative figures in the literature, the KSh 60 billion intermediation number and the 25% participation rate among adults, are derived from FinAccess and FSD Kenya surveys carried out from the mid-2000s through 2013-2016; more up-to-date nationally representative data regarding participation in chamas specifically (as opposed to financial inclusion more broadly) is difficult to find, so the assertion of “steady” participation is based on comparisons between wave after wave of surveys conducted before the last wave of financial inclusion advances, 2021-2025. The causal factors posited above (commitment, social collateral, and risk sharing) are strongly supported by the qualitative and demographic evidence in the literature, but are not verified here using an econometric model disentangling the unique role of each factor in contributing to the decision to participate; this is the goal of the classification study of chama types but one it falls short of achieving.

References (selected)

1. Financial Sector Deepening (FSD) Kenya. The Role of Informal Financial Groups in Kenya. FSD Kenya research report.
2. FSD Kenya. Explainer: Savings groups in Kenya and Savings groups program pages, FinAccess data summaries (2006, 2009, 2013 survey rounds).
3. FSD Africa. Formalising informality: savings groups, community finance and the role of FSD Kenya.
4. Storchi, S. & Rasulova, S. (2017). Impact evaluation of FSD Kenya’s savings groups project. Nairobi: FSD Kenya.
5. Johnson, S. (2004). Gender norms in financial markets: evidence from Kenya. World Development.
6. Anderson, S., Baland, J-M., & Moene, K. O. (2004 working paper) / Gugerty, M. K. (2007). You can’t save alone: commitment in rotating savings and credit associations in Kenya. Economic Development and Cultural Change.
7. Journal of African Development (2024/2025). Households’ Decision to Join an Informal Financial Sector Social Group: Evidence from Kenya.
8. Wikipedia contributors. Chama (investment). (background/overview source)

Prepared as part of the East Africa Microeconomics Research Series. This is a student working paper that synthesises existing research and published data rather than presenting original fieldwork; feedback and data corrections are welcome via the contact page.