If You Can’t Stand the Heat, Design a Better Study: Insights into Residential Energy Conservation

Space heating is the single largest end-use consumer of US residential energy, representing 41.5 percent of consumption. For many homes, there are relatively simple and cheap changes available that could lead to major savings in energy expenditures and reductions in associated environmental impacts. From a policy perspective, it is tempting to incentivize these conservation efforts in order to realize the benefits, but measuring and comparing the efficacy of different energy efficiency measures is rarely straightforward. For example, compact fluorescent light bulbs may offer considerable reductions in energy usage but not if users leave the energy-saving lights on longer or only install them in rarely used areas.

Such complicated end-use behavior can interfere with savings predictions across energy efficiency programs, but randomized controlled trials could help to isolate impacts. In “Returns to Residential Energy Efficiency and Conservation Measures: A Field Experiment,” Jordan F. Suter and Rumi Shammin compare the relative effects of three heating-related energy efficiency treatments: installing attic insulation, installing a programmable thermostat, and implementing a financial incentive-based conservation measure. The results of the two-year study show that financial incentives and attic insulation on their own are enough to reduce natural gas consumption, but when combined, or when the incentives are combined with a programmable thermostat, the savings are especially prominent.

The sample in the study consisted of 24 single-family homes all rented to undergraduate students at Oberlin College in Ohio. Energy consumption data was gathered for each of the homes in the three years prior to the study in order to compare the treatments to the baseline usage. The 24 homes were divided into four groups of six. Each group was assigned a specific condition of either installing programmable thermostats, installing attic insulation, participating in an incentive-based conservation program, or serving as the control. Beginning in the second year, three of the six thermostat homes were also selected to receive the financial incentives in order to see if the combination had any measurable effect.

The students were not made aware of the study but were trained in how to operate the programmable thermostats and how to participate in the financial incentive program. Temperature sensors were installed in each of the homes to compare indoor ambient temperature with energy consumption, allowing Suter et al. to differentiate between structural energy reductions and behavioral energy reductions.

The results of the study show that all of the treatments reduce heating-related energy consumption, but whether those reductions are due to structural or behavioral changes varies across treatments. Homes in just the incentive program reduce consumption by 16 to 20 percent, insulation alone reduces consumption by 19 percent, and combining the incentives with a programmable thermostat leads to more than a 30 percent reduction. The mean hourly indoor ambient temperature, averaged across the heating season, was not significantly different between the control and insulation only treatment, but was significantly lower in the incentive treatment and lowest in the treatment combining incentives and programmable thermostats. These results suggest that incentive programs and programmable thermostats, especially when coupled, can lead to behavioral reductions in energy use. Energy savings associated with just insulation may reduce consumption due to the structural changes but have no effect on behavior when compared to the control group.

This study has policy implications in both its results and employed methodologies. Simple and cheap incentives could be enough to significantly influence energy use behavior, but given the homogeneity of this study in terms of geography, residence types, and tenant characteristics, further studies should be conducted to see if these trends hold across more diverse and larger-scale populations. Universities interested in reducing on-campus energy consumption may be especially interested in these results since the sample consisted entirely of undergraduate tenants in university-owned residences. Arguably more notable than the results is the novel research design, which allows the behavioral and structural factors leading to energy reductions to be isolated. Similar experimental designs in future studies could bolster the predictive power and efficacy of demand-side energy policies.

Feature Photo: cc/(stellardot)