More money or better procedures? Evidence from an energy efficiency assistance program

We provide the first estimate of the money equivalent of procedural changes in public assistance programs. Observing 77,305 appliance replacement decisions in an energy efficiency assistance program targeting low-income households, our RDD design exploits two quasi-exogenous temporal discontinuities in voucher value and procedures. Procedural changes from automatic to elective enrollment and more rigid voucher terms raise the replacement rate, consistent with psychologicaltheories of goal setting and time management. The effects are equivalent to an additional e44 in voucher value. Low-cost procedural changes that target the behavioral responses of low-income households represent plausible areas of unexploited economies in program design