Selection into Credit Markets: Evidence from Agriculture in Mali

Abstract

We find that returns to capital are higher for farmers who borrow than for those who do not. We measure this using a two‐stage loan and grant experiment. In our first stage, we offer loans to some villages and not others. In the second stage, we provide cash grants to a random subset of all farmers in villages where no loans were offered, and to a random subset of farmers who chose not to borrow in villages where loans were offered. We estimate [] returns to capital for the representative sample of all farmers, whereas we find [] return for those who had recently decided not to borrow. Critical for both theory and policy, this heterogeneity persists even after conditioning on a wide range of observed characteristics.