Retrofits: The Key for Quantifying Energy Code SavingsGita Subramony for Zondits, January 9, 2015
Last week, Nate Silver’s blog FiveThirtyEight posted a summary of a few working papers released by the National Bureau of Economic Research. Among the papers was one entitled “How Much Energy Do Building Energy Codes Really Save? Evidence from California,” authored by Arik Levinson. The blog’s one-sentence summary for the study read, “Building codes mandating energy efficiency don’t reduce electricity use.” This conclusion, though, is incomplete, at best, because the study does not take retrofits into account.
The study focuses on residential construction in California for the past few decades to examine energy savings resulting from energy code updates. As a policy initiative, energy code updates have been touted as a useful tool for reducing energy consumption from building usage and operation. When California implemented the first statewide energy code, policy-makers and regulators asserted that buildings in compliance with the new code would be able to reduce energy usage by 80% compared to noncompliant buildings.
Quantifying energy savings resulting from implementing and enforcing energy codes by comparing housing stock across years can be challenging. Zondits has previously written on methodologies for quantifying savings resulting from these initiatives. Levinson also acknowledges the difficulties in measuring these savings. There are several confounding factors; both the nature of homes and how we use them have changed dramatically in the past few decades. In California, new homes are larger and are in hotter parts of the state. Additionally, houses contain more people and energy-using equipment now than they did in the past. Finally – and most importantly – houses are not static; a house built in the 1970s will likely have had significant upgrades over the years.
Levinson’s study uses three approaches to derive an estimate of energy savings resulting from energy code implementation. First, he uses the California Energy Commission’s Residential Appliance Saturation Study (RASS) to see if homes built under more stringent codes use less energy than similar homes built under lesser codes. This data set allows the study to control for resident demographics and other differences in building characteristics.
Levinson’s second approach examines changes in electricity use during hotter than usual conditions; houses built under more stringent codes should see electricity usage rise less steeply under these conditions than buildings built under lesser standards would. Levinson’s last approach compares houses in California to houses in other states with the hypothesis that the gap in energy usage between older and newer homes will be larger in California compared to the gap in other states where code initiatives have lagged. For this analysis, the author uses the US Department of Energy’s Residential Energy Consumption Survey (RECS).
Levinson concludes from his analysis that implementing energy codes does not result in energy savings. To be clear, he does not conclude that energy codes are bad policies, but that they should be not credited with the energy savings that they claim. As far as explaining the results, Levinson invokes the rebound effect and suggests that homeowners in newer homes use more energy since more efficient equipment makes it less costly for them to do so. However, other studies have suggested that the rebound effect is not as great as some might think and certainly does not erase all energy savings resulting from more-efficient equipment. And looking at California’s per capita energy usage compared to the rest of the country, the impact of the state’s efficiency policies is obvious.
How does Levinson explain that? Either the codes or the other state initiatives – or more likely a combination of both – have resulted in significant savings relative to the rest of the country. But if, as Levinson would suggest, the rebound effect eats up all of the savings induced by efficiency, why do we see a difference in the above graph? Levinson believes codes don’t work, so does that mean all the credit goes to efficiency programs? It certainly does seem odd that the rebound effect would apply only to codes and not to efficiency programs that pay for retrofits and so forth.
Perhaps there is an alternative explanation: retrofits. Levinson does acknowledge that the study does not take into account retrofits. The likely reason that homes built under less stringent codes perform just as well as ones built under the most recent codes is that older homes have changed. For one, any air conditioning installed in the 1970s has likely been replaced at least once, if not more times. The units simply don’t last that long. And California has been running home retrofit programs – which would improve home insulation – for decades.