Data and Code for: A New Perspective on Spatial Heterogeneity in African Development

Access to basic infrastructure is a critical component of quality of life and an important measure of economic development. However, on-the-ground data about infrastructure access, especially in low-income countries, is often sparse and costly to collect. We leverage satellite imagery and survey data to train a machine learning model that predict access to infrastructure for each 6.72x6.72km area of Africa. The model achieves accuracy levels of 77.1% to 84.7%. We use a spatial regression discontinuity design to study how much of the heterogeneity in infrastructure access across countries comes from differences in institutional quality, finding a positive effect of a modest magnitude, reconciling previous contradicting results in this literature.