Can Self-Help Groups Really Be "Self-Help"?
NBER Working Paper No. 18970
This paper examines a cost-reducing innovation to the delivery of "Self-Help Group" microfinance services. These groups typically rely on outside agents to found and administer the groups although funds are raised by the group members. The innovation is to have the agents earn their payment by charging membership fees rather than following the status quo in which the agents are paid by an outside organization and instead offer free services to clients. The theory we develop shows that such membership fees could actually improve performance without sacrificing membership, simply by mitigating an adverse selection problem. Empirically, we evaluate this innovation in East Africa using a randomized control trial. We find that privatized entrepreneurs providing the self-help group services indeed outperform their NGO-compensated counterparts along several dimensions. Over time, they cost the NGO less and lead more profitable groups; also, households with access to privately-delivered groups borrow and save more, invest more in businesses, and may have higher consumption. Consistent with the theory, these privatized groups attract wealthier, more business-oriented members, although they attract no fewer members.
Document Object Identifier (DOI): 10.3386/w18970
Published: Brian P. Greaney & Joseph P. Kaboski & Eva Van Leemput, 2016. "Can Self-Help Groups Really Be "Self-Help"?," Review of Economic Studies, Oxford University Press, vol. 83(4), pages 1614-1644. citation courtesy of
Users who downloaded this paper also downloaded* these: