The Structuring of Air Source Heat Pumps' Prices in a Retrofitting
Total Page:16
File Type:pdf, Size:1020Kb
The structuring of air source heat pumps’ prices in a retrofitting residential buildings market: what did Ipay for? Dominique Osso, Stanislas Nösperger, Maxime Raynaud, Marie-Hélène Laurent, Catherine Grandclément, Aurelie Tricoire To cite this version: Dominique Osso, Stanislas Nösperger, Maxime Raynaud, Marie-Hélène Laurent, Catherine Grandclé- ment, et al.. The structuring of air source heat pumps’ prices in a retrofitting residential buildings market: what did I pay for?. ECEEE 2017 Summer Study, ECEEE, May 2017, Presqu’île de Giens, France. pp.1289-1299. hal-02153845 HAL Id: hal-02153845 https://hal-edf.archives-ouvertes.fr/hal-02153845 Submitted on 12 Jun 2019 HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. The structuring of air source heat pumps’ prices in a retrofitting residential buildings market: what did I pay for? Dominique Osso, Stanislas Nösperger, Aurelie Tricoire Raynaud Maxime & Marie-Hélène Laurent CSTB EDF R&D 84 Avenue Jean Jaurès Avenue des Renardières 77420 Champs-sur-Marne 77818 Moret-sur-Loing France France Catherine Grandclément EDF R&D 7 Boulevard Gaspard Monge 91120 Palaiseau France Keywords refurbishment, prices, residential buildings, heat pump, abate Our results confirm the importance of economic variables (such ment as brand or sales network) beside the technical variables in the explanation of prices. Our results also quantify the relative role of each variable. Half of the prices’ variation is explained by Abstract the models and it is a huge step in the understanding of retrofit The recent European energy proposals for the revision of the prices in order to better orientate the public subsidies. Energy Efficiency and the Energy Performance of Buildings Directives emphasize the importance to drive investments into the renovation of building stocks and to stimulate the Introduction refurbishment demand. Moreover, the challenge of acquiring The renovation of the French housing stock is one of the objec data about retrofitting is reasserted because the lack of reliable tives of the public authorities which has introduced a number data is detrimental to the perception of costeffectiveness. of incentives (tax credit, Energy Performance Certificate, lower Especially it is well known that refurbishment prices are, rate of VAT (Value Added Tax), Energy Efficiency Obligation according to various papers, subject to large uncertainty and (EEO), etc.) that are intended to make it possible to achieve can sometimes be controversial even if public subsidies are these aims in terms of reducing carbon dioxide (CO2) emis available. sions through market forces rather than through the impo In this paper, we evaluate the main determinants of prices. sition of obligations (République Française, 2016; MEDDE, Their structuring is a complex phenomenon blending technical, 2015). economical and organizational sides. For such purpose, we Nevertheless, in order to have an effective retrofit market, it analyzed hundreds of invoices concerning the installation of is necessary that the prices (representing an investment costs heat pumps in existing buildings. for the household) are understandable by the decisionmaker, In order to model the influence of the different variables on which is not the case today because of the differences observed the upfront cost paid by the households, we developed general in renovation prices (Laurent et al. 2011). The reasons for the linear statistical models (ANCOVA) blending qualitative and differences in prices between quotations remain somewhat un quantitative variables. The variables taken into account are: clear to private individuals and sometimes even for profession als in the sector. However, a particular link has been observed • Technical: living area, type of building (multi or single fam between a high price paid for renovation work and the quality ily), coefficient of performance, installed power; perceived by households meaning that a low price is associated • and economic: company description (number of employees, with a poor quality (Stolyarova 2016) but the prices’ difference main activity and sales network), average household’s income between quotations cannot be helpful for the household to make linked to location, brand of equipment installed. a decision. Although this deviation may be due in part to techni ECEEE SUMMER STUDY PROCEEDINGS 1289 Contents Keywords Authors 6-090-17 OSSO ET AL 6. BUILDINGS POLICIES, DIRECTIVES AND PROGRAMMES Table 1. Characterisation of the two samples of ASHP studied. absolute Median price of maximum relative interquartile Type of work Sample size work (in € ex. difference coefficient (RIC) VAT) (AMD) main 7,181 3,733 1.11 a-ASHP 10.9 % subsample 192 4,141 1.12 main 1,720 13,211 0.49 w-ASHP 0.3 % subsample 167 13,165 0.49 AMD=|(P_main-P_subsample)|/Min(P_main,P_subsample). The calculation of the maximum absolute difference maximises the error between the two estimates. RIC: Since the majority of the observed prices had a log-normal distribution with positive asymmetry, the measure of the relative dispersion is based on the relative interquartile coefficient (RIC), which is more robust in the case of an asymmetric distribution. cal reasons (complexity of a site) or economic reasons (quality The work analysed in the present study was commissioned of the products, company structure, etc.), it is currently poorly by residential customers, most of whom were owneroccupiers understood. There are therefore objective reasons (technical, who financed the installation of an ASHP. It should be noted etc.) and other more subjective reasons (brand image, etc.) that that the differences in the estimates (maximum absolute dif may be able to explain these differences. However, no study has ference) of the median price (P) between the two samples of yet made it possible to quantify the respective impacts of these data used was less than approximately 10 %. The Euro prices ex. two groups of causes. The public authorities have insisted on the VAT 1 given in this document are for the supplied and fitted sys need to study the level of knowledge concerning the cost of en tem (including labour and ancillary costs (Laurent et al. 2009)) ergy renovation work (Couriol & Fuk Chun Wing 2015). and excluding any eventual trade discount (which is used as an As one area of study, we have analysed the installation of explanatory variable). airair source heat pumps (aASHP) and airwater source heat For the overall analysis, the information collected, irrespec pumps (wASHP) during renovation work. As far as ASHP sys tive of the nature of the work, was: the amount in € ex. VAT of tems are concerned, the literature often analyses their techni the work without distinguishing between the work eligible for cal and economic performance (Asaee 2017, Kelly et al. 2016, EEO and any other associated work not admissible for EEO, lo Torekov et al. 2007) but, to our knowledge, has devoted little at cation (zip code), type of dwelling (singlefamily house – SFH, tention to the observed price structures. The value chain for w multifamily house – MFH), company’s SIREN2 code and the ASHP systems was analysed by (In Numeri, 2014) who found membership in a commercial network (here, we considered that in 2012, the observed price breakdown was approximately company networks, not networks of energy providers). Despite 29 % manufacturing cost, 23 % for the distributor, 22 % for this, filtering on the number of operations conferring entitle the cost of installation, and a remaining 25 % of “unexplained” ment to EEO does not guarantee that the overall cost of any costs. Furthermore, recent ASHPrelated studies have rather given operation conferring entitlement to EEO might not con examined the difference between the estimated and real costs tain other work not eligible for EEO but which might be listed (Kelly & Cockroft 2011, Raynaud et al. 2016) or the potential in the invoice and therefore included in the analysis. This partly for ASHP installation (Szekeres & Jeswiet 2016, Petrović et al. explains the presence of work with very high costs and the re 2016, Gupta 2014). moval of these values from the analysis. The “technical” approach (subsample) permitted a detailed analysis of the work performed in combination with the addi Methodology tional technical data collected, such as the living area covered Based on invoices for renovation work in several residential (m²), the performance (COP3), the installed power (kW) and areas, we produced statistical models combining both qualita the brand of the installed heating system. To provide further tive and quantitative variables in order to estimate the effect of input into the priceofwork databases, we used also other different types of variables: technical data (COP, type of ASHP, sources of information which were joined together on the basis m²…), macroeconomic data (regional added value, median of the INSEE4 code, the SIREN code or the region (8 areas): income in the town …), and microeconomic data (brand of • Number of RGE (Guaranteed