Bunching of residential building energy performance certificates at threshold values

Abstract

Energy performance certificates (EPC) provide a measure of and raise awareness of the energy efficiency of homes. The Irish system of energy performance certificates comprises fifteen alphanumeric grades, which imposes fourteen discrete notches in an otherwise continuous scale. Visual evidence of bunching occurs in the distribution of EPC certificates on the favourable side of EPC grades among homes that have completed energy efficiency retrofits but not in the equivalent distribution of EPC certificates prior to completion of retrofits. Using a regression discontinuity design and estimation of a counter-factual distribution, evidence is found of bunching in the post-works distribution but not the pre-works distribution. We analyse whether statistical evidence exists of drivers of this bunching and whether sources of bunching can be identified. We find that bunching is widespread but not systemic and find discontinuities in the distribution of certain parameters of assessment (i.e. low energy lighting) which suggests their contribution toward bunching. With comparable EPC schemes across Europe the results have policy implications for other countries. Where continuous EPC scales are used, additional energy efficiency improvements may be achieved if a multi-point scale is considered. In countries where multi-point EPC scales are used, bunching of assessments is also likely to occur and quality control measures may be necessary to ensure that the bunching is due to genuine energy efficiency improvements and not a reflection of illicit activity.

Can green labels become the new normal?

“…This phenomenon was also reported in Ireland [72]. A subset of this data was studied two years later, concluding that Bunching appeared after houses had retrofits, and suggested that 'low energy lighting is the parameter most highly associated with bunching at the favourable of the BER thresholds' [71]. Such a study discarded that the bunching was caused by the behaviour of assessors or regional distribution.…”

“…It was found that thresholds can drive bunching behaviour, i.e. behavioural responses may collectively bunch on one side of the threshold due the discontinuity in the incentive [71]. However, to observe the phenomenon, large datasets are requireded because 'bunching usually occurs in close proximity to specific points' [71, p. 663].…”

Quantifying the Measurement Error on England and Wales EPC Ratings

“…The large number of dwellings attaining a D or C on second measurement is possibly explained by retrofit interventions on these properties, however, the observed step change is an unusual feature of genuine data distributions [28] and may indicate some manipulation. Discontinuities from one band to another in post-retrofit EPC results have been observed before: Collins et al [29] documented a 'bunching' discontinuity effect in Irish EPC data from a sample of dwellings having undergone grant-funded retrofit works. This 'spike' form of discontinuity differed from the step change observed in the England and Wales dataset; nevertheless, we agree with the statement by Collins et al highlighting the importance of understanding whether the effect 'is real and reflects accurate assessments of properties' energy performance or the result of illicit activity'.…”

An analysis of errors in the Energy Performance certificate database

“…Lack of reliability and accuracy in EPCs may not be only a UK phenomenon; The EPCs for Irish properties suggest a discontinuous distribution of EERs, with bunching around the beginning of a EPC band being observed. This bunching may indicate home-owners undertaking retrofits to get them just over the lines or, alternatively, assessors manipulating EPC inputs to ensure the client receives a more positive result [17].…”