Validating rebound-effect estimation for thermally retrofitted non-residential public service buildings: theory, measurement, benchmarks, and long-run dynamics

Grossmann, Galvin, Weiss, Madlener and Hirschl (2016) developed the first interdisciplinary methodology for estimating space-heating rebound effects after thermal retrofits of non-residential public service buildings and applied it to four German case studies, obtaining near-zero rebounds for three office/administration buildings but a deeply negative rebound of −113% for a music academy, while cross-sectional data implied positive rebounds of 17–55%. They left open a set of unresolved issues that bear on what is needed for scientifically dependable rebound estimation in this building class: whether the classical price-mediated rebound concept is even applicable where occupants do not pay for heating fuel; how measurement uncertainty in standardized U-values of older buildings distorts rebound estimates; the need for larger weather-verified cross-sectional benchmarks with confidence intervals; the unexplained −113% anomaly; unknown long-run post-retrofit consumption dynamics; untested behavioral and organizational interventions; and unverified transferability beyond German public service buildings.

Background

When a building undergoes a thermal retrofit — insulation of roofs, façades and basements, window replacement, heating-system upgrades, and hydraulic equalization of the heating circuit — the engineering calculation predicts a certain reduction in space heating demand. In reality the realized saving is usually smaller, and sometimes larger, than predicted. The rebound effect is the standard concept for the take-back part of this shortfall. In its classical microeconomic formulation, rebound arises because an efficiency improvement lowers the effective price of the energy service (useful warmth), and the consumer responds to this endogenous price signal by demanding more of the service — higher indoor temperatures, longer heating hours, more rooms heated (Sorrell & Dimitropoulos, 2008). The standard quantitative measure is the energy efficiency elasticity of energy services, $\eta_{\varepsilon S}=(\partial S/\partial\varepsilon)(\varepsilon/S)$, where $S$ is the energy service consumed and $\varepsilon$ is energy efficiency; for a discrete retrofit step it is computed as $\eta_{\varepsilon S}=1+B$ with $B=\ln(E_1/E_2)/\ln(\varepsilon_1/\varepsilon_2)$, where $E$ is actual metered consumption and $\varepsilon=1/C$ with $C$ the calculated energy intensity from an engineering standard such as Germany's DIN V 4108-6. A complementary diagnostic is the energy performance gap, $\mathrm{EPG}=(E-C)/C$, the percentage deviation of actual from calculated consumption. For German dwellings, large datasets show that pre-retrofit buildings typically consume less than calculated — the prebound effect (Sunikka-Blank & Galvin, 2012) — which inflates apparent rebound if ignored, since the retrofit lifts actual consumption toward the calculated value while the calculation itself was too pessimistic.

Grossmann, Galvin, Weiss, Madlener and Hirschl (2016) transferred this apparatus, well established for residential buildings, to non-residential public service buildings, where almost nothing was known: no rebound estimate existed for this building class. Their interdisciplinary methodology combined (i) the elasticity rebound effect and EPG computed from on-site-verified, weather-adjusted metered consumption and calculated intensities for four recently retrofitted German public service buildings (a music academy in Dresden, the Dresden police headquarters, and customs offices in Flensburg and Bremen); (ii) contextualization against 176 non-residential buildings from three German cross-sectional datasets (51 public service buildings from Lammers, 32 office and administration buildings from Oschatz et al., and 93 heterogeneous non-residential buildings from IWU), with implied rebound effects of 55%, 29% and 17% respectively derived from best-fit power curves of actual versus calculated consumption; and (iii) semi-structured interviews with 21 building users, managers and caretakers. The case-study rebounds came out near zero for the three office/administration buildings (−5.5% to +4.9%) but at −113% for the music academy, whose post-retrofit consumption fell far below the calculated value; in three buildings consumption kept declining by 6–9% per year with no stabilization over up to five years. The near-zero case-study rebounds clashed with the strongly positive implied rebounds from the cross-sectional data.

A U-value — the thermal transmittance of a building element in W/(m²·K) — is the core input to the calculated intensity $C$; the authors shared the concern among building energy specialists that standardized U-values for older buildings may be up to twice the true values, which would understate pre-retrofit efficiency and overstate rebound. This concern is corroborated for residential solid walls by in-situ heat-flux measurements (Li et al., 2015, measured a mean of 1.3 ± 0.4 W/m²K against the standard 2.1 W/m²K for UK solid walls), and Swiss certificate data show envelope U-values have only minor explanatory power for actual consumption (Cozza et al., 2020) — but neither result covers older German public service buildings, nor has the U-value uncertainty been propagated through a non-residential rebound estimate. Adjacent literatures have grown without answering the present question: reviews of the building performance gap (Shi et al., 2019; Bai, Yu & Pan, 2024) and of occupants' role in it (Mahdavi et al., 2021) concern mostly residential and low-energy buildings; non-residential benchmark datasets are emerging (Droutsa et al., 2018, for Hellenic buildings; Jurišević et al., 2018, for Serbian educational buildings) but are not rebound-oriented; econometric rebound work for non-residential energy use remains price-mediated and sector-level (Zhao et al., 2025); social-theoretic reappraisals of rebound (Galvin & Gubernat, 2016) have not produced a validated framework for institutional, non-price-mediated consumption; and the source authors' own follow-up agenda moved to domestic retrofit policy (Galvin & Sunikka-Blank, 2017).

Problem Statement

The open question posed by Grossmann et al. (2016) is what it takes to make rebound-effect estimation scientifically dependable for thermally retrofitted non-residential public service buildings. The source records the following unresolved issues, all of which remain open for this building class:

1. Theoretical adequacy of the rebound concept. Is the classical rebound effect — defined as a consumer's response to the endogenously induced price signal of cheaper energy services — an appropriate instrument for analyzing consumption in public service buildings, where the occupants do not pay for heating fuel? The observed discrepancy between the near-zero case-study rebounds (−5.5% to +4.9%) and the positive implied cross-sectional rebounds (17–55%) may partly reflect this structural difference; a rebound framework tailored to non-price-mediated consumption contexts remains to be developed.

2. Unexplained anomaly. What explains the deeply negative rebound effect of −113% in the music academy, given the candidate mechanisms the source identifies but cannot confirm: changed user groups after the retrofit, an exceptionally active caretaker who closes windows and adjusts heating, and thick acoustic insulation in practice rooms acting as unaccounted thermal retentiveness?

3. Measurement uncertainty in the fabric baseline. Standardized U-values used to estimate pre-retrofit energy intensity of older buildings may be up to twice the true values; how much does this bias rebound estimates, and how should measured rather than standardized fabric performance enter the estimation?

4. Adequate statistical benchmarks. Larger, more representative cross-sectional datasets of non-residential buildings — with verified energy intensities, weather-adjusted actual consumption, and categorization by building type and usage — are needed to provide a statistically dependable background context with known confidence intervals against which individual case studies can be compared; the combined 176-building basis is insufficient.

5. Long-run dynamics. Post-retrofit consumption in three buildings kept falling by 6–9% per year for up to five years with no sign of stabilization: when, and at what level, does consumption stabilize in the long run?

6. Behavioral and organizational levers. Interviews identified untapped saving potential through improved ventilation practices, cooperative temperature setting, management of common areas, and organizational energy-saving initiatives; the effectiveness of such behavioral and organizational interventions in reducing energy performance gaps has not been empirically tested.

7. Transferability. Does the methodology and do its findings extend to private-sector non-residential buildings, to other building types such as schools and hospitals, and to other countries with different climates, building standards, and organizational cultures?

None of these issues has been resolved for non-residential public service buildings. Each remains an open question in its own right, and each may be attacked on its own. A scientifically dependable rebound estimate for this building class, then, is one that is reproducible from documented, independently measured, weather-adjusted consumption with a quantified uncertainty, and whose assumptions and limits are stated explicitly.

References

1. Grossmann, Doreen; Galvin, Ray; Weiss, Julika; Madlener, Reinhard; Hirschl, Bernd (2016). A methodology for estimating rebound effects in non-residential public service buildings: Case study of four buildings in Germany. Energy and Buildings 111:455–467. DOI: 10.1016/j.enbuild.2015.11.063. https://doi.org/10.1016/j.enbuild.2015.11.063 (primary source)
2. Sorrell, Steve; Dimitropoulos, John (2008). The rebound effect: Microeconomic definitions, limitations and extensions. Ecological Economics 65(3):636–649. DOI: 10.1016/j.ecolecon.2007.08.013. https://doi.org/10.1016/j.ecolecon.2007.08.013
3. Sunikka-Blank, Minna; Galvin, Ray (2012). Introducing the prebound effect: the gap between performance and actual energy consumption. Building Research & Information 40(3):260–273. DOI: 10.1080/09613218.2012.690952. https://doi.org/10.1080/09613218.2012.690952
4. Li, Francis G. N.; Smith, A.Z.P.; Biddulph, Phillip; Hamilton, Ian G.; Lowe, Robert; Mavrogianni, Anna; Oikonomou, Eleni; Raslan, Rokia; Stamp, Samuel; Stone, Andrew; Summerfield, A.J.; Veitch, David; Gori, Virginia; Oreszczyn, Tadj (2015). Solid-wall U-values: heat flux measurements compared with standard assumptions. Building Research & Information 43(2):238–252. DOI: 10.1080/09613218.2014.967977. https://doi.org/10.1080/09613218.2014.967977
5. Cozza, Stefano; Chambers, Jonathan; Deb, Chirag; Scartezzini, Jean-Louis; Schlüter, Arno; Patel, Martin K. (2020). Do energy performance certificates allow reliable predictions of actual energy consumption and savings? Learning from the Swiss national database. Energy and Buildings 224:110235. DOI: 10.1016/j.enbuild.2020.110235. https://doi.org/10.1016/j.enbuild.2020.110235
6. Shi, Xing; Si, Binghui; Zhao, Jiangshan; Tian, Zhichao; Wang, Chao; Jin, Xing; Zhou, Xin (2019). Magnitude, Causes, and Solutions of the Performance Gap of Buildings: A Review. Sustainability 11(3):937. DOI: 10.3390/su11030937. https://doi.org/10.3390/su11030937
7. Bai, Yefei; Yu, Cong; Pan, Wei (2024). Systematic examination of energy performance gap in low-energy buildings. Renewable and Sustainable Energy Reviews 202:114701. DOI: 10.1016/j.rser.2024.114701. https://doi.org/10.1016/j.rser.2024.114701
8. Mahdavi, Ardeshir; Berger, Christiane; Amin, Hadeer; Ampatzi, Eleni; Andersen, Rune Korsholm; Azar, Elie; Barthelmes, Verena M.; Favero, Matteo; Hahn, Jakob; Khovalyg, Dolaana; Knudsen, Henrik N.; Luna-Navarro, Alessandra; Roetzer, Astrid; Sangogboye, Fisayo C.; Schweiker, Marcel; Taheri, Mahnameh; Teli, Despoina; Touchie, Marianne; Verbruggen, Silke (2021). The Role of Occupants in Buildings' Energy Performance Gap: Myth or Reality? Sustainability 13(6):3146. DOI: 10.3390/su13063146. https://doi.org/10.3390/su13063146
9. Zou, Patrick X.W.; Xu, Xiaoxiao; Sanjayan, Jay; Wang, Jiayuan (2018). A mixed methods design for building occupants' energy behavior research. Energy and Buildings 166:239–249. DOI: 10.1016/j.enbuild.2018.01.068. https://doi.org/10.1016/j.enbuild.2018.01.068
10. Droutsa, Kalliopi G.; Balaras, Constantinos A.; Dascalaki, Elena G.; Kontoyiannidis, Simon; Argiriou, Athanassios A. (2018). Energy Use Intensities for Asset Rating of Hellenic Non-Residential Buildings. Global Journal of Energy Technology Research Updates 5:19–36. DOI: 10.15377/2409-5818.2018.05.3. https://doi.org/10.15377/2409-5818.2018.05.3
11. Jurišević, Nebojša; Gordić, Dušan; Lukić, Nebojša; Josijević, Mladen (2018). Benchmarking heat consumption in educational buildings in the city of Kragujevac (Serbia). Energy Efficiency 11(4):1023–1039. DOI: 10.1007/s12053-018-9631-y. https://doi.org/10.1007/s12053-018-9631-y
12. Jurišević, Nebojša; Gordić, Dušan; Vukićević, Arso (2022). Assessment of predictive models for the estimation of heat consumption in kindergartens. Thermal Science 26(1 Part B):503–516 (online first 2021). DOI: 10.2298/TSCI201026084J. https://doi.org/10.2298/TSCI201026084J
13. Mashhadi Rajabi, Mona (2022). Dilemmas of energy efficiency: A systematic review of the rebound effect and attempts to curb energy consumption. Energy Research & Social Science 89:102661. DOI: 10.1016/j.erss.2022.102661. https://doi.org/10.1016/j.erss.2022.102661
14. Zhao, Yijie; Yang, Ying; Jia, Ruining; Chai, Jian (2025). Analysis of the rebound effect and induced effect of non-residential natural gas consumption: Empirical evidence from China. Journal of Management Science and Engineering 10(1):54–67. DOI: 10.1016/j.jmse.2024.08.002. https://doi.org/10.1016/j.jmse.2024.08.002
15. Galvin, Ray; Gubernat, Andreas (2016). The rebound effect and Schatzki's social theory: Reassessing the socio-materiality of energy consumption via a German case study. Energy Research & Social Science 22:183–193. DOI: 10.1016/j.erss.2016.08.024. https://doi.org/10.1016/j.erss.2016.08.024
16. Galvin, Ray; Sunikka-Blank, Minna (2017). Ten questions concerning sustainable domestic thermal retrofit policy research. Building and Environment 118:377–388. DOI: 10.1016/j.buildenv.2017.03.007. https://doi.org/10.1016/j.buildenv.2017.03.007