Residential Building Codes Do Save Energy

Building Codes Do Save EnergyLinks:   paper    EI blog

In 1978, California adopted building codes designed to reduce the energy used for temperature control. The codes specified minimum standards for wall, ceiling, and raised-floor insulation, allowable heat loss through windows, and the efficiency of climate control systems in new buildings. This was one of several policies in the U.S. designed to reduce fossil fuel use following the 1973 energy crisis. In recent years, the interest in reducing fossil fuel consumption has intensified due to concerns about climate change. 

Prior studies found that residential building codes may not even reduce energy use, let alone pass a cost-benefit test (e.g., Levinson (2016), Chong (2012), Jacobsen and Kotchen (2013), Kotchen (2017)). 

We use a rich dataset of hourly electricity consumption during 2012 and 2013 in 158,112 Sacramento houses. We ask whether houses built just after 1978 use less electricity when it gets hot than similar houses built just before 1978. 

The first figure shows how much more electricity a house uses when the average daily temperature deviates from 62ºF. In Sacramento, an average daily temperature of 62ºF implies a high of about 75ºF and a typical day with an average of 80ºF will reach 100ºF.  

We estimate that the average house built just after 1978 uses 8% to 13% less electricity for cooling than a similar house built just before 1978. The second figure shows that these reductions occur in the late afternoon and evening, especially on hot days.

Comparing the estimated savings to the policy’s projected cost, our results suggest the policy passes a cost-benefit test. In settings where market failures prevent energy costs from being completely passed through to home prices, building codes can serve as a cost-effective tool for improving energy efficiency.

Citation: Residential Building Codes Do Save Energy: Evidence From Hourly Smart-Meter Data. Novan, K.; Smith, A.; and Zhou, T. Review of Economics and Statistics. forthcoming