Choosing Who Chooses: Selection-driven targeting in energy rebate programs

We develop an optimal policy assignment rule that integrates two distinctive approaches commonly used in economics—targeting by observable characteristics and targeting through self-selection . Our method uses experimental or quasi-experimental data to identify who should be treated, untreated, and who should self- select to achieve a policymaker’s objective. Applying this method to a randomized controlled trial on a residential energy rebate program, we find that targeting that leverages both observable data and self- selection outperforms conventional targeting for a standard utilitarian welfare function and welfare functions that balance the equity-efficiency trade-off. We highlight that the LATE framework (Imbens and Angrist, 1994) can be used to investigate the mechanism behind our approach. By introducing new estimators called the LATEs for takers and non-takers , we show that our method allows policymakers to identify whose self-selection would be valuable and harmful to social welfare.