Should NGOs love groups?

Three years ago, in rural Liberia, I sat down with a young man who had started a farming cooperative. “Why did you start this group?” I asked. “Well,” he replied “these NGOs seem to like to help groups. I tried to get help from them alone, but they ignored me. So I started this group and now they are giving us assistance.”

I’ve since heard the same thing in more than one country and community, and it’s made me cynical about the NGO love affair with groups. Even so, some new evidence suggests there might be good reason for the group lovin’:

We exploit random variation in the meeting frequency of microfinance groups during their first loan cycle to show that more frequent meeting is associated with long-run increases in social contact and lower default. Relative to clients who met on a monthly basis during their first loan, those who met weekly were four times less likely to default on their subsequent loan. We provide experimental and survey evidence that the decline is driven by improvements in informal risk-sharing that result from greater social contact.

That is Feigenberg, Field and Pande in a new paper.

I’m running a similar experiment with training and grants, where half the recipients get formed into groups. This paper has given me many interesting measurement ideas, especially the need for more data on within-group transfers, advice-giving, learning, and mimicry.

But perhaps more interesting that the economic impacts of group aid are the possible effects on politics and collective action. There are many micro-level papers (and more than a few dissertations) to be written on why some people and groups engage in collective action, and whether aid has any impact. Readers: I welcome pointers to interesting research I may have missed.