In group lending schemes of the kind pioneered by 2006 Nobel laureate Muhammad
Yunus, personal trust and group homogeneity are more important for producing good
repayment rates than societal trust or mere acquaintanceship between group members.
That is the central finding of new research by Alessandra Cassar, Luke Crowley and
Bruce Wydick, published in the February 2007 issue of the Economic Journal
Much has been made of the wonderful simplicity of the group microfinance model. But
empirical studies of group lending and its catalyst, ‘social capital’, have proven anything but
simple. This new study employs traditional laboratory experiments in non-traditional field
settings in an effort to sidestep the stumbling blocks that typically beset this type of
research.
There are many mechanisms that make group lending successful. Most of these
mechanisms exploit the one asset that the rich cannot hold over the heads of the poor –
social capital. But social capital is an umbrella term used to describe any number of
informational and relational networks between individuals. Figuring out which specific types
of social capital are most important to group lending is the key to determining where the
largest payoff to continued research and innovation lies for practitioners, academics and
policy-makers.
One of the key mechanisms of microfinance – ‘peer screening’ – makes the study of group
lending and social capital in their natural environment very difficult. By its very nature, peer
screening embeds social capital inside the microfinance model, while a researcher would
prefer to introduce social capital from outside the model.
The typical solution for this kind of dilemma is to reconstruct the model in a laboratory
setting. This methodology allows the problematic portion of the model – in this case, peer
screening – simply to be edited out. But while this approach works the vast majority of the
time, the sterile laboratory environment actually proves too sterile for a proper investigation
of group microfinance and social capital.
It is easy to imagine how the social networks or shared norms of Western university
students – the standard subjects in a laboratory experiment – might differ from those of
microfinance borrowers in a developing country. For this reason, Cassar and her
colleagues use a non-standard subject pool in their microfinance games to conduct what
are sometimes called ‘artefactual field experiments’.
Individuals that fit the profile of typical microfinance borrowers were recruited from two sites
with established microfinance markets – Nyanga, South Africa and Berd, Armenia.
Although some measures of social cohesion were site-specific (there are no clans in
Armenia and no post-perestroika generation in South Africa), the experimental
methodology used in each location was the same.
In addition to finding evidence of free-riding, the results of the study support many of the
traditional ideas concerning group microfinance and social capital. The fact that personal
trust between specific pairs of individuals is more important than societal trust bolsters the
case for informational social capital and the importance of peer screening.
At the same time, the fact that socially homogeneous groups consistently outperformed
socially heterogeneous groups speaks to the importance of relational social capital and
mechanisms such as social sanctions.
ENDS
Notes for editors: ‘The Effect of Social Capital on Group Loan Repayment:
Evidence from Field Experiments’ by Alessandra Cassar, Luke Crowley and Bruce Wydick
is published in the February 2007 issue of the Economic Journal.
Alessandra Cassar and Bruce Wydick are at the University of San Francisco. Luke Crowley
works for Innovations for Poverty Action.
For further information: contact Alessandra Cassar on +1-415-422-5351 (email:
acassar@usfca.edu; website: http://www.usfca.edu/fac-staff/acassar/); Bruce Wydick via
email: wydick@usfca.edu; Luke Crowley via email: lcrowley@poverty-action.org; or
Romesh Vaitilingam on 07768-661095 (email: romesh@compuserve.com).