Gumball Capital

The evidence supporting microfinance has long been critiqued as anecdotal or not scientific enough, but last week, the first randomized-controlled trial of microfinance was released as a working paper. I won’t go into too much detail here, as the Center for Global Development has an excellent summary of the article, but the findings are less-than encouraging.

Abhijit Banerjee, Esther Duflo, Rachel Glennerster and Cynthia Kinnan did a study of around 100 slums in Hyderabad, India, randomly assigning half a new MFI branch. They studied the communities for about a year and a half to determine what impacts the MFIs would have on the business and individuals they served.

They found that existing business will use microfinance to expand their inventories and, therefore increase profits. No surprises there. The households without an existing business, but with a “high predicted propensity to start a business,” were more likely to cut back on “temptation goods”–alcohol, lottery tickets, etc–assumingly to save up for a large initial investment in a business of some kind that can be covered by a microloan. Still no surprises.

But when looking at the larger picture–microfinance’s effects on social improvements like women’s empowerment, health and education–the MFIs had no significant effect. Quite surprising, and a little disheartening for all of us who have worked to expand microfinance as a poverty alleviation tool.

However, Banerjee et al. do note that their study is relatively short (15-18 months) and the effects of microfinance could be farther-reaching. Over a longer time frame, the effects on social factors could be more pronounced. David Roodman, in the summary linked above, says that these findings (like in the findings in all random-controlled trials) only describe these specific communities studied in India and cannot be extrapolated to all MFIs everywhere.

So the results are interesting, its good to see more scientific study of microfinance, but the debate on its efficacy is far from over.