Certification by the Leadership in Energy and Environmental Design (LEED) program is often proposed as a potential method to improve building energy efficiencies. This is despite the general lack of data regarding the efficacy of LEED certification and inconsistent results from past studies that often focused on a few buildings from a single city. Using recently available building energy use data from a nationwide set of 10 cities, we studied the effects of LEED certification on building energy efficiency measured by site energy use intensity, source energy use intensity, and greenhouse gas emissions. In addition, we used natural language processing methods to study patterns in the acquisition of specific credits by LEED-certified buildings. We find that LEED-certified buildings are not more energy-efficient by any measure except in a single city. In addition, neither the total amount of credits nor the number of Energy and Environment credits achieved correlate with building energy efficiency measures. Finally, buildings with high Energy and Atmosphere credits corresponding to renewables are not more energy-efficient in many cases. These conclusions call into question the use of LEED certification as a policy metric for improving the energy efficiency of buildings.
Keywords Energy Efficiency, LEED, Building