Esther Duflo of the Poverty Action Lab at the Massachusetts Institute of Technology in the US is using the novel approach of randomised trials to find solutions to questions such as: “Does microfinance reduce poverty?”

She has been speaking at the TED conference and impressing the assembled experts and celebrities with this approach which is seldom used in research in development issues.

In a paper published last year, she and her colleagues assessed the impact of microfinance institutions by comparing the impact of microfinance loans on families in slums in Hyderabad, India. Her randomised trial approach compared families in the slums who received microfinance loans with those that did not.

In summary, she found that the outcomes differ for different families. Existing business owners appear to use microcredit to expand their businesses but those that do not have businesses tended to use the loans to pay off existing loans or to use it to purchase food and other consumables.

She concludes that there is no evidence that the loans have an impact on developmental indicators like education and health for households that do not have existing businesses.

My summary does not do justice to the paper which is well worth reading in full.