Leila Safavi

A new working paper by Fiona Burlig, Christopher Knittel, David Rapson, Mar Reguant and Catherine Wolfram studies the cost-effectiveness of energy efficiency investments in over 2,000 K-12 California public schools served by the Pacific Gas and Electric Company (PG&E). The study uses high frequency electricity metering data collected by PG&E to estimate the electricity consumption savings for schools, including using a novel machine learning approach to predict counterfactual electricity consumption in treated schools.

Focusing on lighting and HVAC upgrades, the authors find that schools with energy efficiency interventions reduced electricity consumption by 2-4% on average compared to control schools. Specifically, investments in energy efficient lighting lead to 5 to 7% reductions in electricity use in daytime hours and efficient HVAC systems produced a 2-4% daytime reduction in electricity when temperatures were highest. This decrease in consumption can result in substantial savings for public schools, although findings show that savings may be smaller than projected. While actual savings from HVAC and lighting upgrades are predicted to be 70-90% of ex-ante engineering estimates, the authors estimate that actual savings are only 15% of projected savings when a wide range of upgrades are considered. Many papers cite the discrepancy between actual and projected savings as cause for doubt on the extent to which energy efficiency can lead to reductions in emissions. However, results from the experiment show that, at least for lighting, interventions translated into real energy savings for schools and delivered a substantial part of the expected energy consumption reductions.

Looking forward to future projects, in June E2e solicited Letters of Intent for innovative economic research on critical or novel energy efficiency issues. This solicitation allocates up to $250,000 in funding per proposal and targets projects using rigorous empirical techniques, such as randomized controlled trials and high quality quasi-experimental methods, to evaluate the real-world impact of existing energy efficiency policies and programs. Five proposals have been selected to move to the next round of consideration and a full list can be found on the E2e website. Proposals span a variety of energy efficiency issues ranging from the impact of smart meter engagement on household energy savings to the effect of management practices on industrial energy use. During the next phase of review, the selected researchers will be asked to submit a complete proposal which will potentially be recommended to the Alfred P. Sloan Foundation for funding by the end of 2016.