Thousands of microfinance institutions around the world collectively provide financial
services to over 100 million clients through an incredible variety of programmes. Some lend
to groups of 40 women who gather each week under a tree; others offer individual loans
from air-conditioned bank branches in capital cities.
New research by Professor Dean Karlan, published in the February 2007 Economic
Journal, explores the circumstances in which ‘group lending’ – the typical arrangement in
microfinance – is most effective in terms of generating high loan repayment rates.
He finds that individuals who live closer to one another and are more culturally similar to
others in the group are more likely to repay their loans and save more. Forming groups of
clients from among geographically and culturally similar people may be both profitable for
microfinance institutions and good for borrowers.
Last year, Muhammad Yunus and the Grameen Bank of Bangladesh were awarded the
Nobel Peace Prize in recognition of the extraordinary microfinance movement they
launched to provide the poor with credit and lift themselves out of poverty. Yunus’
innovation was to harness the social bonds between groups of borrowers to overcome their
lack of collateral, by personally guaranteeing each other’s loans.
This form of ‘solidarity banking’ ensures repayment through a combination of peer selection
(in which borrowers use their knowledge of their neighbours to screen out bad credit risks)
and monitoring and enforcement (borrowers observe each other over time and exert
pressure to repay). As economist Hal Varian wrote in the New York Times, ‘Peer pressure
can be an immensely strong force, and the Grameen Bank has figured out how to make it
work in the cause of economic development.’
Though group lending is commonly considered to be the key factor responsible for the
growth of microfinance, there is surprisingly little empirical evidence to show if and how it
actually improves repayment rates. One important element may be the strength of the
social connections between group members. It is easy to imagine that groups that are more
socially connected may perform better at screening, monitoring and enforcement, as they
will have superior knowledge of their peers’ trustworthiness and activities.
But in theory, social connections could lead to worse outcomes if, for example, individuals
are unable to punish their close friends or family. Empirically, these are very difficult
questions to answer because individuals who have strong social connections tend to be
different in other respects (generally outgoing and entrepreneurial). Merely examining a
correlation between social connections and repayment in a group lending setting does not
establish a causal link from social connections to repayment.
In this study, Dean Karlan takes advantage of a unique group formation process to single
out the causal effect of social connections on repayment and savings rates among
microfinance clients. Unlike most microfinance programmes, which form groups of clients
from among the same villages or neighbourhoods, women seeking a loan from FINCA-Peru
are put on a list at the main office in the centre of Lima, Peru. Once this list contains 30
names of women from all over the city, a group is formed. Group meetings take place in the
FINCA office in the city centre and not in the various neighbourhoods. Thus the signup
process creates a natural experiment. The group composition, due to the sufficiently small
group sizes, produces some groups with greater social connections than others.
Comparing the groups, Karlan finds that individuals who live closer to one another and are
more culturally similar to others in the group are more likely to repay their loans and save
more. He also shows through qualitative data that groups with better connections have
more effective monitoring – knowing each other's default status and causes – and are more
likely to punish defaulters with social exclusion.
In addition, social connections help groups distinguish between default due to true negative
personal shocks and mere reneging on loans. Better-connected groups are thus more likely
to forgive those who have defaulted due to circumstances beyond their control, and allow
them to continue borrowing.
Why is this important? There are myriad ways to serve the poor with microfinance, but not
all methods are equally successful. The study provides important insight into how group
lending can be successful in overcoming the default problems that have historically plagued
development finance projects. The findings suggest that forming groups from among similar
geographic and cultural groups may be both profitable for microfinance institutions and
good for borrowers.
But Karlan cautions that the results in no way suggest that group lending is superior to
individual lending. In a separate study, which directly compares group liability to individual
liability, Karlan and Xavier Giné from the World Bank find that converting group liability
loans (specifically, individuals who already underwent peer screening) into individual liability
loans did not harm the repayment rates for the bank. But further study is required, in
particular, more replicable studies so we can learn which findings are robust and which not.
With this information, we can learn how and when social connections help financial
institutions provide deeper access to finance for the poor.
ENDS
Notes for editors: ‘Social Connections and Group Lending’ by Dean Karlan is published in
the February 2007 issue of the Economic Journal. Dean Karlan is assistant professor of
economics at Yale University.
For further information: contact Dean Karlan on +1-203-432-4479 (email:
dean.karlan@yale.edu); or Romesh Vaitilingam on 07768-661095 (email:
romesh@compuserve.com).