Behavioral Interventions in Energy Consumption

This field experiment monitored plug-level dorm-room energy usage for 90 students, including a control and a treatment group. The primary hypothesis was that the incentive based on potential losses, manifested through an Energy Efficiency Escrow (EEE), will result in lower usage for the same potential savings offered by a pay-for-performance (P4P) program. The unit of analysis are individual students, each of whom live in on-campus residence hall rooms. Wi-fi connected power strips were assigned to each student and transmitted the entire plug-load demand, with individual plug and total power recorded every second. A consulting energy company (GOEFER, Inc.) developed a user-app for students to control plug loads and to monitor their usage. At the beginning of the treatment period a new version of the GOEFER app added features to allow the control and treatment group to assess their financial incentives. Pre/post-survey data provided controls for sociodemographic differences and other factors. A combination of difference-in-differences methods and clustered OLS modeling of pre and post treatment periods was used for all analyses.