Microcredit Made to Measure: Experimental Evidence from Rural Morocco

Does matching loan repayments to expected cash flows raise take-up and the welfare impact of microcredit? We test this in a randomized controlled trial with Morocco's largest microfinance institution. The experiment introduces two new individual-liability loan products: a contract with a five-month principal grace period and a tailored contract with installments matched to expected revenues. Over 3,000 applicants are randomly assigned to an offer of one of the two new contracts or the standard loan. The new contracts attract distinct borrowers: agricultural households take up the grace-period loan, while entrepreneurs operating smaller, more capital-constrained businesses choose the tailored schedule. Most tailored-loan borrowers front-load their repayments, suggesting that demand for flexibility reflects cost minimization, not only liquidity management. The tailored loan raises sales and profits via non-agricultural business expansion and labor reallocation to self-employment. The grace-period loan shows similar but statistically insignificant effects. Neither product worsens repayment performance. We find no effects on investment or consumption. In a second stage, we randomize contract-specific information campaigns across 440 villages. Despite shifting stated preferences toward the advertised products, the campaigns do not raise take-up, consistent with supply-side frictions and a challenging macroeconomic environment.