Testing Microfinance Theories: Field Experiments in Developing Countries (Princeton University)
CAREER: Field Experiments in Credit, Insurance, and Behavioral Economics (Yale University)
NSF-funded research on microcredit lending shows it's possible to make money fighting poverty. Economist Dean Karlan of Yale University examined a wide variety of policy interventions in several developing countries using randomized trials. His work provides solid scientific data for policies to fight poverty using microcredit loans.
Research results from South Africa show that microcredit is both effective and profitable. Together with his collaborator Jonathan Zinman of Dartmouth College, Karlan worked with a local bank to make small loans to working, poor individuals. The recipients all applied for regular bank loans and were randomly selected from the group of applicants who barely missed qualifying under the usual policy. The study compared the borrowers to a control group drawn from the unsuccessful applicants. The research team followed both groups for two years, with data gathered on a range of social, economic, health and mental health measures. Results demonstrate that the loans resulted in significant benefits. Borrowers are more likely than others to retain wage employment, less likely to experience severe hunger, and less likely to be impoverished. What's more, the bank made money on these loans. The loans were made at the bank's standard interest rate and the poor borrowers repaid at almost the same rate as the bank's richer standard customers.
Karlan is conducting other studies to determine how to maximize the effectiveness of microcredit lending. In Peru, another randomized control experiment measured the effect of adding business training to an existing Peruvian microcredit program for low-income female entrepreneurs. Some borrowers received weekly or monthly entrepreneurship training, while the control group received the same loans without the training. The training program improved business knowledge, practice, revenues and repayment rates. Borrowers who were least interested in the training at the beginning of the experiment actually benefited the most from it.
Human brains organize visual object information in a unique wayResearch Areas: People & Society Locations: New Jersey
Some interview methods produce less reliable testimony from child witnessesResearch Areas: People & Society Locations: Michigan, New Jersey