Using Targeting to Optimize Program Design: Evidence from an Energy Conservation Experiment

Abstract

We investigate the potential for targeted treatment assignment rules to improve the performance of a large-scale behavioral intervention to encourage households to conserve energy. We derive treatment rules based on observable household characteristics that maximize the expected benefits of the intervention. Targeting treatment using transparent and easily implemented rules could yield significant gains; the energy savings from optimal treatment assignments are predicted to be double those achieved by the intervention as implemented. Predicted cost savings from targeting are even larger. Our results underscore the potential for targeted treatment assignment to generate significant benefits in many domains.