Microcredit

What the evidence says

In 2006 Muhammad Yunus and the Grameen Bank he founded shared the Nobel Peace Prize. The committee praised their efforts to “create economic and social development from below.”

Randomized evaluations then put microcredit to the test. Six were published together in 2015, covering six countries on four continents, cities and villages, and many kinds of lenders and loans. Banerjee, Karlan and Zinman summarized a consistent pattern: modestly positive effects that fell short of transforming borrowers’ lives. A policy bulletin covering seven evaluations reported that:

• Demand was lower than lenders expected. Take-up was 13 to 31 percent in Ethiopia, India, Mexico and Morocco.
• Six of the seven studies saw some expansion of business activity, but profits rarely rose.
• None found a significant effect on average household income, and none of the six that measured schooling found children in school more.
• Six studies found households gained more freedom in how they earned, spent, invested and managed risk.
• None found widespread harm, even with individual liability or high interest rates.

Meager (2019) pooled seven experiments and concluded that the average effect on business activity and consumption is probably small and may be negligible. Households that already ran a business gained more, but less predictably. In Hyderabad, India, six years after microfinance arrived, households that had run businesses beforehand had about 35 percent more business assets and roughly double the revenue of similar households in control areas, while others showed almost no effect.

The grade is strong for the finding stated in the summary: several rigorous studies agree that the average effect is modest. That is different from finding that microcredit does nothing, and the debate over what it is worth is in the Disagreement Register.

Where it sits on the depth scale

Charitology places microcredit at repair (depth 1), the same level as cash transfers. It works on a household’s access to money, as a loan rather than a gift, to build income or handle shocks. It doesn’t change the rules that keep people poor. Expanding lending into places that lacked it does change a local financial market, which touches structure, but the evidence suggests the average effect on poverty is modest.

Open questions

• Are short trials missing the story? In Hyderabad, gains for existing business owners persisted and grew over six years.
• Can lenders tell in advance which borrowers will gain?
• How should flexibility and risk management, which don’t show up as income, be valued against other outcomes? See the exchange rate problem.