Making a Market

Making a Market

Past the gem-green banana trees cropped on dark, chocolaty soil and well beyond vast expanses of foddering brush, the land below the Merarani Hills stretches finally into plains of stone and dust.

Here in this district of northern Tanzania the ground is not good for farming. It is difficult to imagine markets coaxed from such withered scrub. Yet, that is precisely what is happening - farmers are demanding and paying for the services of savings groups.

Though livelihoods in this landscape produce meager incomes, savings and even fees can be wrested from them. Households grow and sell onions, mango, pigeon pea, sunflower and corn - tough work and not very profitable. Some members are casual laborers in the Tanzanite mines and most raise and fatten small livestock. The lucky ones, those with prime land watered by a local river, are able to farm bananas.  

Even in harsh conditions members earn enough to pay an agent to help manage their meetings and their money. Certified agents are selected, trained, tested and certified from a pool of field agents receiving modest pay from the Archdiocese of Arusha, under the technical direction of Catholic Relief Services. Prior to taking a series of tests the field agent works under a supervisor for nearly a year, forming her own groups in the process. The final testing is serious and features three parts:


  • An oral exam that tests knowledge of group formation, conflict resolution, bookkeeping and financial management, and problem solving.
  • A portfolio review examining efficiency and productivity ratios of the field agent’s groups.
  • A customer satisfaction survey of the field agent’s groups. 


Agents may market their services to any village within their territory, one largely defined by efficient bicycle travel.  Though groups meet weekly, agents visit them monthly, charging about 200 Tanzanian shillings per member per meeting. That’s about 12 cents US. Most groups have fifteen or more members, which yields about $2 per meeting per agent. While a small amount by any standard, income generated from assisting groups supplements income that an agent will earn from usual farming or trading activities.

Said one agent, “I do my regular chores in the morning and visit groups in the evening. Since I have a cell phone as do group officers, I can easily arrange meetings to suit everyone’s schedule. It’s hard work but I manage to assist seventeen groups.”

Here, members save between USD$0.30 and 1.50 per meeting reaching about $6 per month per member. The $0.12 per member monthly fee seems reasonable. In a given month, individual agents have the time to assist fifteen to twenty groups. During the annual distribution of funds, members will pay their agent an additional sum or offer valuable gifts like livestock, roof sheeting or fencing materials.

When asked what they liked best about being an agent, the seven people interviewed shared similar responses. They appreciated the test and the prestige of certification. Their groups were proud of their completion of the oral exam and in many instances held celebrations when the agent was awarded a certificate.

The fee-for-service approach is not yet ready for replication. Catholic Relief Services and its local partners are taking stock of model’s pros and cons. The pros are impressive: the model is a good way to sustain service to groups beyond the project; it ensures that services provided are the ones demanded by members; and it supports a new kind of livelihood for the agent.

The cons are also worth noting. Agents may not be able to reach the poorest members of a village who cannot pay fees, or the poorest hamlets that lack sufficient density to make the business model work; the agent-approach could stifle self-replication of groups or tamp down innovations that speed along tasks like record keeping. Agents have no incentive to foster breakthroughs that might ease them out of a job.

Despite the challenges, CRS, its local implementing partners, and the agents themselves are creating a volunteer network that will continue the process of certification and support.

One area that the network might explore is expanding agent income from other kinds of services. Now, there is a pronounced asymmetry between the time a group has the most capacity to pay for services and the time those services are most needed. Members need support most in the early stage of their group’s development, a time when they are least likely to afford fees.  Inversely, as a group matures and its members are more able to afford fees, services are less needed. Were an agent to market other services or products, she might be able to “leverage” visits to an area as well as the goodwill of maturing groups. Selling airtime minutes, money transfer services or even useful items like solar lamps are just some ideas. The emerging agent network might find and negotiate for better ones.  

Whether a fee-for service approach is more powerful or more sustainable than traditional approaches to hiring staff or motivating volunteers to form and service groups remains to be seen. Regardless of the final verdict, the attempt at engaging demand to sustain support is refreshing indeed. 


View Printer Friendly Version

Email Article to Friend

Reader Comments (5)

Thanks Kim. I have been leading a developmental course - by which I mean, first time anyone has ever led it, and we're making it up as we go along - with eleven outstanding practitioners from eleven countries - I could not have found a better group. In the process of talking about adding other activities to savings groups, we kept stumbling upon the issue of fee-for-service v. voluntarism. This seems to me much more of an open question that many of my colleagues apparently consider it.

The class divided in two on the issue, so I had us divide into two groups and get our arguments together.

There were some good points made. Here are some arguments for the two sides I had not seen before...

For Fee for service:

"Being absent from home has its social and cultural cost. In Africa, if you leave the village, you must have something to bring home. This is a financial burden for trainers."

"Voluntarism works in small villages, where social norms lead people to freely share services and knowledge in a web of reciprocity. But we're moving away from that socialist model. It is not always appropriate to see communities as a single social unit any more, and if you don't have the social motivation, people need to be paid."

For Voluntarism

"In Zimbabwe, when we asked volunteer Village Agents if they wanted to be paid, they refused. They said it wasn't consistent with their work."

"People interested in volunteerism are different from those who want to be paid. Those who want to be paid want to move on, whereas the volunteer is probably going to stay in the village."

To that list, I would add the possible disincentive to form community driven groups in a fee-for-service environment. As in medicine, law and other professions, there is a risk of adding enough complexity that groups will not feel empowered to train others in what really is should be simple to learn.

Wed, October 12, 2011 | Paul Rippey

Hi Kim:

Are there people seriously proposing a volunteer model capable of not only ensuring sustainability, but also ensuring expansion in the numbers of associations over time?

Realistically however, a fee-for-service model involves some constraints. If the agent is going to successfully serve mature associations, s/he must not only know the rules but also be an experienced person -- especially in private and/or public business. For example, dispute resolution is a critical role. Most people fitting this profile are unlikely to bicycle 20 kms for $2!

Having said that, I've seen this work on many occasions, and when it does, I have usually had cause to suspect that there were other income activities going on behind the scenes. For example, the agent may be investing her neighbours' money in a group, and splitting the profits. Or she may be borrowing from the group and on-lending to non-members she is close to. So a sustainable balance is achieved.

My suspicion is that NGOs should be paying close attention to the kinds of informal links such as this that are taking place. It is not good enough to simply ban them. All that does is ensure the NGO will never know about them, and will never be able to factor their existence into planning. We need to research them and reflect carefully on their implications.


Sun, October 16, 2011 | Brett Matthews

Brett ask provocatively, "Are there people seriously proposing a volunteer model capable of not only ensuring sustainability, but also ensuring expansion in the numbers of associations over time?"

Well, uh, yes, I'm one of those people. I've interviewed fee-for-service agents in Kenya and Niger, asking them if they would do what they were doing, even if they weren't paid. A woman trainer in Niger looked at me like I was nuts for asking the question: "I am a daughter of this village. Do you think I would say No to my parents? Do you think I would fail to share what I know with the people who live here?"

I think we make a mistake by drawing too heavy a line between fee-for-service and volunteer models. Everyone reading this post probably does what they do professionally for a mixture of intrinsic and extrinsic motivations. A friend of mine - a household name to many people who read this site - challenged the idea that people would work without pay. I pointed out that she was putting in hours every week as the unpaid president of the international school board. "Well, okay", she replied, "but my children go to the school."

Well, okay - but the children of the trainers live in the village too, and in most cases they live in areas with a bit more social glue than most Americans or even Canadians. They have a complex set of motivations, including no doubt getting some money, but including increasing their social networks, gaining prestige, and information seeking. Plus fulfilling their duty to share what they know.

A recent quick study in western Kenya by FSD showed an incredible rate of community-driven post-project group formation. I'll leave the details for a separate post, but there is little doubt that as a general rule, People Help People Form Groups. Why this is even remotely surprising is beyond me.

I'm amazed at the rush to fee-for-service, an untested idea whose theoretical underpinnings strike me as fragile indeed. It is very probable that introducing a fee-for-service model will create some not-very-good jobs for trainers, but with several dangerous incentives, including the incentive to discourage group members from forming their own groups, and the incentive to draw out trainings if they are paid on a per-visit basis.

Until these things are handled, I want to paraphrase Brett's question: "Are there people seriously proposing the widespread adoption of an unproven fee-for-service model with known and unaddressed perverse incentives built in?"

Mon, October 17, 2011 | Paul Rippey

Paul -- thanks for responding to my provocation! Before you respond further, I offer you a quote from a paper I wrote a few years that should help clarify the direction from which I asked this question.

"Many in the micro-credit sector today object to the word ‘movement’ and say micro-credit should be seen as an ‘industry’. In Canada credit unions are today overwhelmingly a ‘system’ that competes within the financial services industry.

But, when credit unions were emerging there was a credit union movement, and the power of this movement lay in its ability to inspire a generation of Canadians to donate millions of hours of their time and their energy in the well-placed conviction that they were creating better economic opportunities for their children and grandchildren. No ‘industry’ relying entirely on more direct economic incentives could have achieved this result. And without the movement that helped develop Canada’s credit unions, our nation would have a far less balanced, less inclusive and less equitable financial industry today." (Compounding Community Capital)

In short, you are preaching to the choir!

But I asked the question for a reason -- and it not just that we perversely insist, in spite of all historical evidence, in framing our work with savings associations as an 'industry' instead of a 'movement'. (If it were a 'movement' the ASGS would have been held where people live, not where rich tourists and jet-setters vacation.)

Credit unions received much of those millions of hours of donated time from educated people who had independent incomes -- in particular teachers, priests and successful entrepreneurs. In the world of savings associations, who are we talking about? Teachers in Cambodia make $20 a month; in Tanzania possible as much as $50 a month -- and in these and other countries survive by gardening. In spite of their low incomes, they are as active in volunteering their time as we can realistically expect, and they are motivated by improving the quality of life in their villages for their children and grand children. Nevertheless, there are few people in the village with sufficient independent income to donate time on the scale that triggered modern credit unions -- though they will do the best they can and that will get better over time if we keep the incentives working properly.

This is why I believe we should be asking: "what income generating activities are they performing as they visit distant groups, years after phase-out, in their role as 'factilitator'?" And once we know that "can any of these activities be encouraged during group incubation?"

This strikes me as potentially offer a practical solution to what is -- as everyone who reads this blog knows -- a very challenging practical problem.


Tue, October 18, 2011 | Brett Matthews

In our support for both savings- and credit-based activities, are we not pursuing the goals of providing the means whereby participants can become more integrated into a cash-based market economy? “Inclusive finance”?

Despite my complete lack of training in formal economics, it’s clear to me that any market activity today, other than barter, relies upon the monetisation of, well, just about everything: commodities, labour, time, sport, leisure, creativity, intellectual property, etc.

So it seems to me that fee-for-service behaviour amongst those who supply services to savings groups is entirely consistent with the principles of the macro model within which the promotion of grass-roots savings methodologies takes place.

I too grew up in the credit union movement (as it most definitely was back then), and was told in no uncertain terms that the 50-60 hour week expected from me for my 40-hour salary was in recognition of the voluntary inputs of the credit union leaders I was serving ― all of whom undertook their voluntary roles on top of either full-time employment, well-pensioned retirement, accommodating employers, powerful trade unions, or providing spouses. The sacrifice considered inherent in being a volunteer was primarily of family time, rather than putting food in front of family.

There will always be the motivation of contributing to one’s community and social group without expectation of monetary payment but I imagine the drawing of the line between the two models will be a fluid process as behaviours adapt to the growth of the cash-based portion of the local economy. In the Solomon Islands, a historically successful use of young women trained to provide malaria testing within their home villages is being eroded as they forgo the nominal payment per test for higher paying urban jobs, their aspirations stimulated by an imported lifestyle model entirely dependant upon complete monetisation.

We can deplore the trend, but I don’t know how to turn my credit union system back into a movement either.

Tue, October 18, 2011 | Greg Pirie

 Staked to the Ground in a Savings Junkyard

Staked to the Ground in a Savings Junkyard

Voices from the Summit