Fields from Afar: Evidence of Heterogeneity in United States Corn Rotational Response from Remote Sensing Data

We construct estimates of own- and cross-price corn rotation elasticities using a field-level dataset that accounts for over 83% of the US corn-producing area. We allow rotational response to vary by estimating separate models across 115 subsamples that we delineate using Major Land Resource Areas (MLRAs) and soil characteristics. The results show a high degree of rotational response heterogeneity. Across the country, we find that rotational response is elastic in some areas and near zero in others. After aggregating the results to the national level, we find that modeling rotational response without allowing for heterogeneity produces a short-run own-price elasticity of corn planting of around 0.50, which conforms to the latest estimates in the literature. When allowing heterogeneous price sensitivity, our preferred estimate of the rotation elasticity is 0.69. This is evidence that imposing a uniform rotation response could seriously bias aggregate elasticity estimates.