Genetic Dilution Erodes Productivity : Exploring Farmers' Low Adoption Levels of Improved Maize In Ethiopia

Household surveys often rely on self-reports of hybrid seed use, which are prone to measurement error. This context exhibits considerable heterogeneity when estimating productivity impacts. Using a three-year panel, we estimate heterogeneous returns to adoption from self-reports of hybrid seed use and find results driven by non-classical measurement error due to seed misclassification. We leverage information on DNA-fingerprinting collected over the same areas in 2018 and find that heterogeneity in the quality of the genetic material masks positive returns to adoption, with the highest returns found for farmers using higher-purity, drought-tolerant, and newly released varieties. When we account for misclassification in our initial heterogeneity analysis, we find that farmers that have the largest potential to benefit from adoption are constrained in their ability to do so. Our research points to the need for policies that expand the accessibility of improved seed and the need for mechanisms to verify its quality