Practitioners Discuss: Tweaking for the Urban Savings Group

Bruce Galloway:

How should an urban savings group differ from a group located in a rural village?  Apparently, the bonds between the urban savings group members are not as strong as the ancestral ties that bond members of the traditional villages.  What do you think of the idea of suggesting that they reduce the amount that can be loaned to a group member from three times the amount saved in the fund, to two times?  Do you know of any written discussions or research that has been performed on the urban versus rural savings group question?

Paul Rippey:

A recent study in Kenya showed the strongest growth among savings groups in urban and peri-urban areas.  People like the groups because of the social support for savings, the ease and convenience, and the high returns, even when they have access to formal services. Groups were originally marketed as “the rural alternative”. That was probably under-selling their virtues. They are equally a complement to people who have a bank account that they use occasionally. 

As for your suggestion, if the loan leverage ratio is too small, then group membership might not be perceived as quite so valuable, and repayment might decrease rather than increase. I don’t know that would happen, but there are often perverse outcomes to good ideas that are hard to predict. 

 Perhaps challenge the group to work it out. You could have the trainers say, “Okay, folks, I want you to think about this: in the village, everyone knows everyone else, and it’s hard for anyone to leave, and you have the traditional power structures. That helps keep people honest, right? But here in the city, people can run away, and they don’t answer to anyone, and no one will stop them. How are you going to make sure that anyone who borrows repays?” They’ll figure it out. (A friend recently steered me to this quote from Henry David Thoreau: “If ever I was sure that someone was coming to help me, I should run like hell.” ) 

 Hugh Allen:

The key differences I have found between urban and rural are:

1.  Urban groups tend to be smaller. It’s not just a question of social capital being harder to come by in the city, but at least as much determined by how many people can crowd into a small room in a shanty.

2.  Frequently (but not always) urban groups meet less frequently because in the city there isn’t time to attend weekly meetings

3. Urban groups that are made up of people who work in seasonal industries and therefore migrate in and out are very dodgy. 

4.  I can’t really comment on leverage ratios needing to be lower in the city.  We need to remember that at the start the loans will, by virtue of the limited pool of capital be a) in demand and therefore b) quite small (especially when the first few months’ savings is also not amounting to much).  So the risks are pretty low and manageable and I can’t say that I’ve heard reports about any real difference in repayment rates, so maybe it’s not an issue.  I’m not too inclined to worry about this when I look at competing methodologies such as SHGs where the leverage ratio is as much as 10:1.  I am also not inclined to just let it go, as the sharks of the group will push for higher leverage and increase the risk exposure of the poorer members.  But I don’t really think it needs to be changed: current ratios of savings to loans on the website (reporting on 39,000 groups and 870,000 members in Africa) are 223%, indicating a strong savings preference and capital adequacy.  I think this is a case of if it ain’t broke, don’t fix it, especially when lower leverages reduces intermediation efficiency.

Reader Comments (2) 

Interesting discussion and good question. At Mpendulo Savings, in the Eastern Cape of South Africa, all of our groups are urban, peri-urban. I agree with what Hugh says about urban group tending to be small.....but in our case, that was only at first. When we started, we had a heckuva time getting anyone interested (see my post in "What if my bank made me join a group"), and then when we did we were lucky if there were 6 people in the group. 

But now, ALL of our graduated groups and alot of our new groups have up to 22 members, most being around 12 - 15. So it isn't a foregone conclusion that groups tend to be small in urban areas and large in rural. For example, in Zimbabwe, many of CARE's VSL rural groups were 6 to 10, but that was because of population density (lack thereof) and distance between where people lived. For us, we just needed to hit a tipping point and that came when a critical mass of people demonstrated - in very concrete ways - to their friends and neighbors that "hey, this thing really works".

What has been very different though from the rural scene is that we had to get extremely creative in the way we marketed the project and mobilized groups. The other really different element is the issue of time and livelihoods. None of our groups wants to meet weekly, and only a few meet fortnightly. The large majority meet monthly. Also, none of our group members rely on farming for their living. Only 30% use loans for business purposes, 60% rely on government grants and some patch together a living from their grant, a marginal business, part time and/or casual jobs. Only a few have full time jobs. Many use the share-out lump sums for building/repairing houses, buying household assets, business, paying off debt, or a really, really good time at Christmas.

So for me, the answer to whether urban groups, once mobilized, should be treated differently than rural groups is no. And not just on the savings/loan ratios. There, I would agree totally with Paul on the approach he suggested, even putting the issue in the form of a story for the group members to discuss. Some of our groups have put in their constitution that until a member proves him/herself, they can't get the full 3x's savings loan. Others don't seem to have a problem with lending out the maximum....the most important thing is for the group to work it out for themselves.

Wed, March 16, 2011 | Jill Thompson

Jill's post is interesting. The average size of an SG in Africa is 22 members (, from the 7 countries and 85 projects studied. Averages will vary, as Jills says, according to population density. Kupfuma Ishungu, for example, in Zimbabwe has an average of 8 members in a low-density rural area. In the large Kibera slum in Nairobi the average is 10-12 but in moderate density rural areas the number can be as high as 29-30 (large Niger villages). But In Kibera, where CARE has had an urban programme for several years and is reaching about 10,000 people, the meetings are monthly because people have to be much more tied to the money economy to survive and can't be forever attending group meetings.

Thu, March 17, 2011 | Hugh Allen

The Sour Spot

The Sour Spot

Please 'Splain the Difference »