Quantification of energy savings from Ireland’s Home Energy Saving scheme: an ex post billing analysis

Abstract

This paper quantifies the energy savings realised by a sample of participants in the Sustainable Energy Authority of Ireland’s Home Energy Saving (HES) residential retrofit scheme (currently branded as the Better Energy Homes scheme), through an ex post billing analysis. The billing data are used to evaluate: (1) the reduction in gas consumption of the sample between pre- (2008) and post- (2010) scheme participation when compared to the gas consumption of a control group, (2) an estimate of the shortfall when this result is compared to engineering-type ex ante savings estimates and (3) the degree to which these results may apply to the wider population. All dwellings in the study underwent energy efficiency improvements, including insulation upgrades (wall and/or roof), installation of high-efficiency boilers and/or improved heating controls, as part of the HES scheme. Metered gas use data for the 210 households were obtained from meter operators for a number of years preceding dwelling upgrades and for a post-intervention period of 1 year. Dwelling characteristics and some household behavioural data were obtained through a survey of the sample. The gas network operator provided anonymised data on gas usage for 640,000 customers collected over the same period as the HES sample. Dwelling type data provided with the population dataset enabled matching with the HES sample to increase the internal validity of the comparison between the control (matched population data) and the treatment (HES sample). Using a difference-in-difference methodology, the change in demand of the sample was compared with that of the matched population subset of gas-using customers in Ireland over the same time period. The mean reduction in gas demand as a result of energy efficiency upgrades for the HES sample is estimated as 21 % or 3,664 ± 603 kWh between 2008 and 2010. An ex ante estimate of average energy savings, based on engineering calculations (u value reductions and improved boiler efficiency and use through heating controls), suggests a technical reduction potential of 5,676 kWh per dwelling. Equating this with the gas reduction in the sample suggests a shortfall of approximately 36 ± 8 % between technical potential and measured savings. This shortfall includes the effects of direct and indirect rebound effects, variations in ex ante assumptions and achieved u values and efficiencies for upgraded dwellings. The profile of household characteristics in the HES sample is influenced by the self-selected nature of scheme participants. Self-selection bias and other possible biases in the sample data impact on the validity of the comparison. Data limitations for individual households across explanatory variables in the control and treatment groups precluded corrections for these biases in the sample; however, the profiles of separate comparable data sets were used where possible to quantify the differences in the explanatory variables and how these might impact on the measured energy saving with reference to the relevant effects identified in the literature.