Using Satellite Imagery and Deep Learning to Evaluate the Impact of Anti-Poverty Programs
The rigorous evaluation of anti-poverty programs is key to the fight against global poverty. Traditional approaches rely heavily on repeated in-person field surveys to measure program effects. However, this is costly, time-consuming, and often logistically challenging. Here we provide the first evidence that we can conduct such program evaluations based solely on high-resolution satellite imagery and deep learning methods. Our application estimates changes in household welfare in a recent anti-poverty program in rural Kenya. Leveraging a large literature documenting a reliable relationship between housing quality and household wealth, we infer changes in household wealth based on satellite-derived changes in housing quality and obtain consistent results with the traditional field-survey based approach. Our approach generates inexpensive and timely insights on program effectiveness in international development programs.
We thank Edward Miguel, Jeremy Magruder, Ben Faber, Marshall Burke, Joshua Blumenstock, Supreet Kaur, Ethan Ligon, Elisabeth Sadoulet, Alain De Janvry, Aprajit Mahajan, the participants in the AGU Fall Meeting 2019 (Session GC34C), the UC Berkeley Trade Lunch, Development Workshop, Development Lunch, Good Data Seminar, and the FAO Technical Network on Poverty Analysis for feedback. We thank Edward Miguel, Michael Walker, Dennis Egger, Johannes Haushofer, Paul Niehaus, and the rest of the GiveDirectly team, for generously sharing the dataset with us and responding to our inquiries. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research.