Energy Research & Social Science 96 (2023) 102918 Contents lists available at ScienceDirect Energy Research & Social Science journal homepage: www.elsevier.com/locate/erss Original research article The politics of measurement and the case of energy efficiency policy in the European Union Tessa Dunlop a, *, Thomas Völker b a Institut de Ciència i Tecnologia Ambientals (ICTA - UAB), Edifici ICTA-ICP, Carrer de les Columnes s/n, Campus de la UAB, 08193 Cerdanyola del Vallès, Barcelona, Spain b Centre for the Study of the Sciences and the Humanities, University of Bergen, 5007 Bergen, Norway A R T I C L E I N F O A B S T R A C T Keywords: Energy efficiency is a policy strategy to make energy savings and reduce greenhouse gas emissions in Europe. Energy efficiency This paper investigates the passage of amendments to the Energy Efficiency Directive during 2016–2018 as a case Climate study. It focuses on an incident, in which the rapporteur in charge of the energy efficiency file in the European Energy Parliament proposed to re-think the way energy efficiency is defined and measured thus initiating a struggle over Policy the meaning of energy efficiency itself. Drawing on insights from the field of Science and Technology Studies Social sciences (STS) and work on indicator politics, this paper explores how the meaning of energy efficiency was negotiated through the way it is measured. These political negotiations are used as a case to empirically explore politics of measurement, i.e. practices and procedures through which the conceptualization and measurement of energy efficiency get entwined, and in which a seemingly “technical” discussion weaves together scientific and meth­ odological aspects with political, societal and environmental issues. In this way, the work presented here ex­ plores how processes of politicization and depoliticization in the formulation of energy efficiency indicators contribute to bringing about the very governance object that is energy efficiency. 1. Introduction economic growth and energy consumption (i.e., energy intensity), or energy savings and overall lower energy consumption. This conceptual Energy efficiency is a major part of EU energy policy, seen as a key ambiguity is a problem that has been highlighted in both the scientific ingredient to reduce energy consumption and greenhouse gas emissions literature and the policy arena – especially the interchangeability be­ in Europe [1]. Between 2016 and 2018, a process took place to amend tween energy savings and energy efficiency. Various scholars [2–4] have the Energy Efficiency Directive 2012/27/EU (EED). The EED sets the sought to clarify the definition of energy efficiency to avoid this ambi­ rules and obligations for the EU’s 2020 and 2030 energy efficiency guity, explaining that different ways of conceptualizing and calculating headline targets. Before the file reached the European Parliament (EP), energy efficiency indicators have different implications for the effec­ there was an expectation that the political debates were likely to focus tiveness and outcomes in policy. more on the level of the targets, i.e., the ‘ambition’ in percentage terms, This moment in the amendment process is a fascinating case for rather than the methodologies used to calculate the headline target in­ observing the interplay of political actors, their conceptualizations of dicators. In an unusual move, however, the rapporteur in charge of the energy efficiency, and the processes through which they sought to reach energy efficiency file in the EP proposed an overhaul in the way that a compromise on issues of energy policy through debating the seemingly energy efficiency was defined and measured. The action led to a struggle technical issue of how to measure energy efficiency. Using insights from over the meaning of energy efficiency itself – including the way it is Science and Technology Studies (STS) we show that ostensibly “tech­ measured, described, conceptualized and applied in practice. In doing nical” and apolitical scientific objects such as energy efficiency defini­ so, it opened a space up where indicators and measurement temporarily tions and indicators are in fact spaces where the negotiation of interests became an object of debate. Different versions of energy efficiency were and the different visions and valuations of what is regarded important brought forward by different actors. These include an understanding of for future energy consumption are enacted. Or, in other words, this is an energy efficiency as a ratio between either primary and final energy or example of the temporary opening-up of a “governable space which lent * Corresponding author. E-mail address: teresaeleanor.dunlop@e-campus.uab.cat (T. Dunlop). https://doi.org/10.1016/j.erss.2022.102918 Received 16 June 2022; Received in revised form 7 December 2022; Accepted 9 December 2022 Available online 21 December 2022 2214-6296/© 2022 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC license (http://creativecommons.org/licenses/by- nc/4.0/). T. Dunlop and T. Völker Energy Research & Social Science 96 (2023) 102918 itself to political intervention” ( [5] pp.125) and struggles for “calcu­ solve. The argument goes that, by focusing on relative rather than ab­ lative power” [6]. solute levels of consumption, energy efficiency supports the continua­ In doing so, this article aims to contribute to the energy efficiency tion of high levels of energy services and comfort rather than literature by bringing together work at the nexus between the social encouraging change in energy production and consumption systems or sciences literature and the natural and engineering sciences that domi­ more sustainable energy-use practices [16,17]. nate the field of energy efficiency. As Dunlop [7] argues, such inter­ Many of the methodological and theoretical challenges and alter­ disciplinary work is needed to better understand sociological issues natives outlined above, however, are rarely cited or publicly acknowl­ embedded in quantitative representations such as energy efficiency edged in EU energy efficiency policy. Much of the literature on energy definitions and indicators. efficiency policy is focused on improving policy design and reducing “barriers to energy efficiency improvements” [18], rather than exam­ 2. The challenges of implementing the energy efficiency concept ining methodological limitations of the concept itself, and how these are in policy manifest in quantitative representations such as indicators.4 A shortcoming identified by scholars in the energy literature is that it Conceptual ambiguity and methodological challenges affect the is dominated by a techno-economic energy imaginary which impedes design and implementation of energy efficiency policy in practice. For efforts to find solutions that can properly address sociological energy example, authors have highlighted the confusion in the academic, policy consumption problems. Genus et al. [19], for example, point to the need and grey literature among energy efficiency and other related terms, for a more situated approach that puts emphasis on social practices including energy conservation [8] energy demand [9] and energy sav­ relevant to energy use, interdisciplinarity and the coproduction of ings [3]. A distinction is often made between energy efficiency and knowledge with diverse actors. Similarly, Nikas et al. [20] argue that energy conservation, as these terms are used interchangeably. If energy there should be a more holistic and transdisciplinary perspective on the efficiency is reducing energy input while keeping energy services con­ role of human choices and behaviors in influencing the low-carbon stant, energy conservation is thought of as reducing energy input by transition in energy research, including how the desires of individuals reducing energy services [10].1 and communities interact with the energy and economic landscape and Given the conceptual ambiguity in the energy efficiency literature, lead to systemic change at the macro-level. The issue of energy transi­ calls for more precise definitions of concepts such as energy efficiency tions is furthermore brought together with questions of distributive are quite common. Clearer definitions, so the argument goes, can justice by von Platten et al. [21], who argue that the choice of energy improve the field’s understanding of what is at stake and what the trade- performance metrics influence which buildings are considered energy offs are for any given indicator. Authors have sought to explain what efficient and call for open discussion on these metrics. In a similar energy efficiency is and why the implementation of the concept of en­ manner Park et al. [22] call for open debate and improvement of per­ ergy efficiency is so problematic in practice [2,3]. Energy efficiency is formance metrics. generally conceptualized as a ratio of energy input (e.g. kwH of elec­ There is also a growing body of empirical work that utilize social tricity) to the output of useful energy services (e.g., light or heating). An sciences concepts to better understand how discussions on energy effi­ ‘improvement’ in energy efficiency thus implies that one is using less ciency indicators shape policy. Applying discourse analysis to historical energy for the same level of service or output, resulting in “energy energy efficiency policy, Dunlop [23] found that certain values are re­ savings”. This method, however, does not necessarily imply an overall flected in how energy efficiency is defined and measured over time and reduction in energy consumption, just that one is using energy at a more how some potentially relevant concepts and issues get sidelined in favor efficient rate, due, especially, to technical change and innovation of others. Niskanen and Rohracher [6] show that while the definition of [11,12] or consumer behavioral change [13]. Energy efficiency is energy targets, standards and calculation tools are largely presented as a therefore a ratio, not a quantity. For example, a clothes dryer that is technocratic and expert-driven process, they are actually highly political labelled energy efficient may use a more efficient work rate of energy discussions involving different visions and valuations of various lobby input to energy service output (i.e., clothes drying). However, if the user groups which fight for “calculative power” to shape calculation pro­ decides to use the dryer on a more frequent basis, it may consume a cedures and housing codes regarding energy use. higher overall level of energy. This problem relates directly to one This paper aims to contribute to this growing strand of research that particular challenge that has been discussed for decades in the energy combines a study of the methodological challenges of energy efficiency efficiency literature: the “rebound effect”.2 The rebound effect can be implementation with political dynamics to ask questions about what is defined as a mechanism whereby expected savings through energy ef­ quantified, what is left out and what issues co-emerge with the ways that ficiency actions are reduced [14]. This means that we may be making energy efficiency is discussed as a quantifiable and governable object. In less savings from energy efficiency actions than we thought. Our current particular, we look at a case in which a space of negotiation was opened understanding of the rebound effect entails a degree of uncertainty given up through the initiative of one EP rapporteur and subsequently closed a lack of consensus on the nature and extent of the phenomena and down again through processes of depoliticization. To do so we draw difficulties in accurately measuring it.3 inspiration from STS concepts that sensitize us to the practices and Another major problem highlighted in the literature is that energy processes though which certain versions of energy efficiency are efficiency is framed too narrowly and, therefore, leaves out variables defined, constructed, maintained, and shaped, and what impact these that are important to energy conversion processes. Among these are e.g., visions have on social arrangements, social structures, and the co- energy qualities and their compositions, environmental sustainability, production of social and technical orders. pollution displacement and the durability of materials and power [2,15]. There is a lively debate on these issues in the social sciences and 3. Conceptualizing energy efficiency measurement humanities (SSH) field that suggests that the concept itself is problem­ atic, as it contributes to perpetuating the problems that it is meant to We draw on research on the politics of measurement by exploring the indicator politics at play in discussions over energy efficiency 1 For example, turning lights off and heating down is sometimes viewed as an 4 “energy efficiency measure”, however it could also be technically classified as For example, research on energy efficiency policy has focused on the level “energy conservation” or “reducing energy demand”. of energy efficiency targets [46], National Energy Efficiency Action Plans 2 A related, but distinct, term is known as the “Jevon’s Paradox”. (NEEAPS) [47] and energy efficiency governance, which examines predomi­ 3 For more on the rebound effect see for example [3] nantly institutional and organizational considerations in efficiency policy [48]. 2 T. Dunlop and T. Völker Energy Research & Social Science 96 (2023) 102918 legislation, together with the technical and methodological aspects 4. Methods highlighted in the energy efficiency literature. The basic assumption is that indicators are necessarily political as they involve decisions that go This paper applies a case study approach combining document beyond the technical level [24,25]. As Merry [27,pp.21] puts it: “The analysis with expert interviews. Data were collected from three sources. technical is always political because there is always interpretation and Firstly, legislative documents relating to the Energy Efficiency Directive judgement in systems of classification, in the choice of things to mea­ (2012/27/EU) were identified and collated using a saturation approach sure.” Furthermore, the development of indicators is not merely about [31], from June 2016 to December 2018. These included the European the neutral measurement of a pre-existing object “out there”, but rather Commission (‘Commission’) impact assessments, parliamentary reports, involves the production of an legible and governable object [26]. The public debate transcripts and working documents relating to the EED. way in which an entity such as energy efficiency is quantified and Following a literature review, 50 documents were initially selected to measured is not self-evident, but is often a matter of political interests as identify major themes. These were pared down to a total of 20 using a well as practical considerations such as data availability, methodolog­ list of codes that were identified through a literature review. A thematic ical preferences and skills. Turnhout et al. [25] show how a governance analysis was applied to the 20 documents [32]. This analysis was guided logic espousing neoliberal ideals of transparency, effectiveness, and ef­ by the analytical framework, the themes identified in the literature re­ ficiency – or “measurementality” - creates a particular understanding of view and the first round of analysis [33]. ecosystems that may lead to an impoverished understanding of biodi­ Secondly, as a supplement to this document analysis, twenty-one versity. In a similar manner, Asdal [5] - in her work on the European semi-structured expert interviews were conducted with policymakers, “critical load project” that aimed at quantifying how much pollution is policy officers and independent academics to validate and deepen the still acceptable - asks how a new ‘Nature-whole’ is enacted through results. Interviewees include parliamentarians and/or their assistants quantification and how it becomes consequential in politics. This who worked directly on the EED file from the Socialists and Democrats argument points not only to the performative features of method and (S&D), European People’s Party (EPP) and Alliance of Liberals and measurement, but also highlights the importance of political work and Democrats for Europe (ALDE) parties. Two representatives from the “the little tools of governance” [27], that are the “mundane epistemic Commission were interviewed. From this initial sample, policy analysts, and administrative tools and practices” aimed at “producing controlled, academics and energy efficiency specialists who had experience and objective and non-politicized situations and objects.” (ibid. pp. 16). Such knowledge of the file were identified through a sampling technique. The “little tools” can be certain reports, graphs, or visualizations that are interviews lasted 60 min on average. They were designed to gauge the used and become influential in the machinations of bureaucracies. reasoning and rationales on which certain decisions were based, and to Focusing on them means to trace how they are created and revised, how elicit the lines of argument used to support these decisions. In particular, they travel and how they are used by different actors. However, also the these conversations focused on the methodological approaches to the latter part of the quote above is crucial to keep in mind for our work: the measurement of the headline energy efficiency target and how and why default mode of bureaucracies is to create non-political objects and sit­ energy efficiency was defined in certain ways. These interviews were uations and thus a “politics of no politics” [27,28]. The case presented in intended to do two things: first, to better understand the methodologies the empirical part of the section provides an example of an actor who of measurement and indicators, including the effects of measuring en­ attempts of open up a space where indicators for the measurement of ergy efficiency in different ways and the trade-offs involved; and second, energy efficiency could be debated and contested. explore that rationales and lines of argument used in these expert ac­ We draw inspiration from these analytical concepts taken from STS counts [34]. The interviews were used as a base from which to explore to zoom in on the practices through which different objects of gover­ relevant topics and provide context for the main themes found in the nance are enacted [5,29,30]. The starting point for this inquiry is the primary data sources. assumption that such objects are not simply “out there”, but rather the Finally, in addition to this core material, 60 news items directly outcome of these practices. Procedures and methodologies of quantifi­ relating to the energy efficiency policy-file were collected from news cation and measurement play an important role. Understanding the websites (Euractiv, Political Pro) during 2016–2018. The news stories debates surrounding the initiative of rapporteur Gierek through this lens complemented the institutional texts with additional themes, context allows us to zoom in on how the technical gets woven together with and timelines. The articles underwent an initial scan to crosscheck key political and organizational orderings as well as with implicit normative dates and details of the file. Then, using themes identified in the texts assumptions about the common good and desirable futures. Energy ef­ and literature review, the articles were used to triangulate information ficiency as an object that is measured and governed is thus the tempo­ and provide additional context where necessary.5 rally stabilized outcome of this ongoing debate, not its starting point. In this paper, we dismantle a controversy in the EP that went beyond the 5. Politicizing the way energy efficiency is defined and usual topics – level of ambition of targets – and focused instead on the measured ways energy efficiency should be measured (and consequentially also defined). We use this case to explore how debates about the way energy In November 2016, the Commission presented a proposal to revise efficiency is measured contribute to bringing about the legible and the 2012 EED as part of the ‘Clean Energy for all Europeans’ package. governable object that is energy efficiency. In doing so, we aim to carve According to the Commission, the EED was amended to meet EU climate out an understanding of how a certain version of energy efficiency was stabilized through establishing a particular assemblage of the technical- methodological with the social, organizational, legal, and normative. 5 Building on this conceptual framing, we ask how the concrete po­ A limitation of our research relates to the actual size and complexity of the litical procedures and work with and through policy documents energy efficiency policy-file. Focusing on documents and supplemental in­ contribute to establishing and stabilizing a certain version of energy terviews meant that it was impossible to equally integrate all perspectives and efficiency as a governable and legible policy-object. By addressing this arguments into our account. The selection of documents and interviewees fol­ lowed the logic of one particular case as it unfolded and focuses on main ar­ question, we aim to contribute to the debate on energy efficiency guments and turning points. In addition, while we focused on how the “little quantification with an empirical perspective that shows how certain tools” of policymaking become crucial in this case, future research is needed procedures mechanisms of EU policymaking become instrumental in de- that delves deeper into the actual methods of calculation and the implications politicizing indicators and thus stabilizing a particular version of energy of the choices made there. And finally, a broader perspective and a research efficiency. design centred more around interviews and focus group discussions might yield important insights into the struggles for “calculative power”. 3 T. Dunlop and T. Völker Energy Research & Social Science 96 (2023) 102918 and energy targets for 2030 and align it with other aspects of the Clean production and conversion (extraction and transportation of raw ma­ Energy package, including renewable energy and carbon emissions. terials prior to processing into electricity and/or heat). This would be Following the EU’s Ordinary Legislative Procedure (OLP),6 in 2016 the added to the sectors already included in measures such as the electricity, Commission presented its proposal for amendments together with an housing and consumption sectors.7 Thus, ITRE’s enactment of energy impact assessment that were then debated by Energy Ministers in the efficiency focused more on production, rather than consumption sys­ Council of the European Union (‘the Council’). Once a proposal reaches tems. The ITRE report maintained that the aim was to put more onus on the EP, its role is to form consensus among parliamentarians and present energy producers, generators and distributors to make energy efficiency proposals as a plenary resolution to the Council for negotiation/agree­ improvements, rather than concentrating the burden on energy con­ ment. The EP’s Committee on Industry, Research and Energy (ITRE) was sumers at the end of the chain, as the Commission had proposed. The responsible for managing the energy efficiency directive, including ITRE report had argued that this “would not put enough responsibility writing the parliamentary report and managing the interinstitutional on energy providers to modernize their energy production systems with negotiations. The ITRE energy efficiency file was led by a Polish a mind to deliver energy that is efficient in conversion and transfer ef­ rapporteur from the S&D group, Adam Gierek, who also represented the ficiency” [35]. Polish Democratic Left Alliance-Labour Union (Unia Pracy) Party. Prior Energy poverty was a central theme in the parliamentary discussions. to his parliamentary term, Gierek was an engineering professor of The ITRE report argued that focusing on the upstream energy processes smelting and material science. He had been on the ITRE committee since would take the burden off consumers at the final stage to make energy 2004 and had experience in parliamentary energy matters. Gierek’s savings, thus helping to alleviate energy poverty. Here, a certain polit­ expectation was that he would follow his Party’s line to support a 40 % ical geography was enacted by pointing to the division of political energy efficiency target by 2030. This was an ambitious stance landscapes between rich and poor and “Eastern” and “Western” Member compared to Commission (30 %) and Council (30 %) proposals. As the States. In Gierek’s native Poland, there was reportedly “little patience rapporteur, he was tasked with securing the highest possible target as an with expensive energy efficiency solutions that will hurt poorer people”8 outcome of the trilogue negotiations, which are interinstitutional ne­ – i.e., energy efficiency improvements at the consumer end. Gierek gotiations to form a consensus on legislation between the Commission, believed that obliging consumers in poorer countries to reduce their Council and EP. energy consumption was unfair because “people in poorer countries like The ITRE draft report [35] presented to the EP in May 2017 was Poland and Romania use half the energy of Germans or the French and a unusual in that it proposed to overhaul the directive, seeking to detail third of what Swedes consume”.9 and explain problems with the calculation of current indicators and offer In contrast, the Commission’s argument was that a focus on end-use alternative solutions. In doing so, the ITRE report created a version of consumption was precisely the key to reduce energy poverty. Its pro­ energy efficiency that was distinct from that of the Commission. Spe­ posal included targets for energy consumption reduction and energy cifically, it proposed to calculate energy efficiency through the indicator savings in end-use sectors, (known as Article 7). The rationale for tar­ of cumulative primary energy consumption, rather than the separate geting the end-use consumer sector was that by reducing energy con­ primary and final energy consumption indicators that the Commission sumption, consumers would save money in energy bills: “High energy had proposed, and that the directive originally stipulated in 2012. The bills and energy poverty are a major problem in many Member States. In argument for this change was that the current indicators did not measure particular, 11% of European citizen cannot keep their houses warm in energy efficiency comprehensively enough across the economy – i.e., the winter. Also, companies suffer from high energy bills due to that parts of industry were omitted from measurement but needed to be competitive disadvantages, in particular energy intensive industries measured to make them more efficient. These gaps, the report main­ (large and small) but also those in the service sector. Therefore, energy tained, were notably the energy conversion processes between the pri­ efficiency policies should in particular target end consumer sectors to mary and final energy stages, “precisely where the links in the energy improve their situation.” [40,pp. 73].10 flow chain that have huge efficiency increase potential and reserves” [35]. 5.1. Reaction to the ITRE report The proposed methodology therefore included what it saw as the overlooked parts of the energy chain, including primary energy The report caused upset in Parliament following its release, with MEPs submitting some 600 amendments.11 Politico reported that the message from S&D parliamentarians was confusion over the report. 6 Many were having difficulty understanding it, describing it as “very The Ordinary Legislative Procedure is the standard decision-making process in the EU. In the system, the directly elected EP has to approve EU legislation academic”. “Frankly speaking, politicians don’t understand technicali­ together with the Council (the governments of the 27 EU countries). The pro­ ties,” Gierek responded.12 He went on to explain that he was cedure starts with a legislative proposal from the Commission. Before its pro­ posal, the Commission prepares impact assessments that assess the potential 7 economic, social and environmental consequences of possible policy options. The report argues that “Using only primary energy would pave the way for a The Commission formulates a proposal and addresses it to the EP first, which better focus on the entire energy supply chain, including upstream processes can adopt, reject or amend the proposition. A Member of the EP, working in one (extraction and transportation of raw materials prior to processing into elec­ of the parliamentary committees, draws up a report on a proposal for a ‘legis­ tricity and/or heat). It would also take full account of all final consumers’ lative text’ presented by the Commission. The parliamentary committee votes savings through the use of primary energy factors.” 8 on this report and, possibly, amends it. When the text has been revised and News item: “Polish MEP’s quixotic quest to upend the energy efficiency adopted in plenary, Parliament has adopted its position. Then, the EP’s opinion debate,” Politico Pro Morning Energy and Climate, 5th December 2017. 9 is transmitted to the Council that can reject/accept or put forward its own News item: Morning Energy and Climate,” Politico Pro, 27th September modifications. Once the text is agreed upon, the two co-legislators adopt 2017. 10 legislation jointly, having equal rights and obligations. Agreements between the In addition to the headline indicator, the ITRE report also proposed that co-legislators are often reached through tripartite interinstitutional negotiations methodologies capture pollution displacement (e.g. through lifecycle analysis) (‘trilogues’) between the EP, Council and the Commission. and new, more precise formulas for calculating the Primary Energy Factor (PEF), which captures the rate at which energy conversions are calculated. 11 News item: “Morning Energy and Environment,” Politico Pro, 5th September 2017. 12 News item: “Morning Energy and Climate,” Politico Pro, 27th September 2017. 4 T. Dunlop and T. Völker Energy Research & Social Science 96 (2023) 102918 approaching the file from a technical, not political, point of view, Politico ensure a holistic approach to energy efficiency investments and services further reported. One expert interviewee said “his ideas didn’t cut and provide a legal basis to the promotion of both end-use and gener­ through the political discussion at all”. The move that caused the ation and distribution efficiency” [40]. greatest rift within his Party, however, was that Gierek, in an unusual Under pressure from his Party, the rapporteur changed the proposed move, had not followed his S&D group’s line in support of a 40 % target to 40 % in the ITRE report. A number of amendments were tabled headline target. Instead, he supported 35 % in his report, claiming that, and negotiated, and many of the original proposals were omitted, as the rapporteur, he could achieve wider support “for a more realistic including the calculation of cumulative energy and the energy efficiency target” (i.e., one that is easier to negotiate with the EP and EU in­ definition. Unable to support the new report that conflicted with his stitutions). Gierek clarified that he would not rule out supporting the 40 position, Gierek abstained from the vote to approve it, which the ITRE % energy efficiency target on the condition that all three stages of the committee narrowly endorsed on 28th November 2017 (33 votes in energy flow (i.e., those according to his proposed methodology) were favor, 30 against, 2 abstentions). The rapporteur stepped down on included [35]. Applying his measure across the whole energy chain with December 5th to be replaced by another rapporteur. The report to be a 35 % target, he argued, would be more effective in improving energy presented to the wider parliament for debate, therefore, was much less efficiency than by using the current methodology with a 40 % target. radical, and less complex than that Gierek had originally presented. This point, however, was lost amid the debates that centered on the numbers. The cost-benefit analysis in the Commission’s 2016 impact 5.3. Methodological discussions re-emerge in parliamentary debates assessment [36] was central to discussions. The analysis considered different headline indicator options, including 30 %, 35 % and 40 %, Methodological issues remained a concern for some parliamentar­ considering the cost, social and environmental benefits of each. The ians. One concerned the calculation of the rebound effect. On 29th conservative parties (European People’s Party (EPP) and European November 2017, Philippe Lamberts (Verts/ALE) asked the Commission Conservatives and Reformists (ECR)) tended to support the lower target how it factored in the rebound effect when drawing up its objectives and of 30 %, claiming that a higher target was too expensive. Gierek argued proposals relating to energy efficiency.14 The Commission posted its that a 35 % target was the optimal level because the impact assessment reply almost two months later, on January 17th, 2018 - the same day showed that “the closer we get to 40%, the more costs rise significantly” that the parliamentary report was adopted as a plenary resolution (i.e., [35]. The parliamentarians supporting a higher target, including the at a stage that was too late for the EP to make changes). The Commission S&D party, argued that the social and environmental benefits out­ responded15 that the rebound effect was included in modelling for the weighed the costs. Some NGOs and industry groups used the impact impact assessment, using two values based on a literature review: 21 % assessment to argue for a highly target. WWF, for example, released an and 43 %. The significant range of these two figures highlights the extent infographic showing the trade-offs in implementing a 30 % versus 40 % of uncertainty in the energy efficiency literature regarding the rebound target, based on the impact assessment (Fig. 1). effect. The Commission explained that when modelling the first scenario of 21 % rebound, sufficient savings would be delivered to reach a 30 % 5.2. NGOs and stakeholders weigh in energy efficiency target in 2030. The second value of 43 % rebound effect, however, involved more uncertainty in that it “would require Gierek was weakened further politically when NGOs and industry additional measures (such as the recently proposed post-2020 CO2 groups criticized the ITRE report. Many viewed Gierek’s moves as an standards for cars and vans)”. Despite this clarification, however, the attempt to subsidize coal by driving innovation investment in energy complexity of the rebound effect appeared to hamper decision-making, generation, and that the potential for households, industry, the service as one key informer noted: “It was such a difficult concept […] that not a sector, and small businesses to make energy savings would be lost by lot of people were mastering. So, as a result, I think it is also something drawing the focus away from end-use consumption. The European that was a bit pushed to the background, because people didn’t really Environmental Bureau maintained that: “Changes proposed by the MEP want to take a decision on it”. go beyond the scope of the Directive, even attempting to redefine the entire concept of energy efficiency […] The report’s focus on primary 5.4. Focus on achieving consensus energy savings in power plants, transmission and distribution loses sight of the potential for citizens and final users to make gains” [38]. Friends Methodological issues were raised during two parliamentary debates of the Earth Europe stated that: “According to Gierek, energy poverty on the energy efficiency file.16 Parliamentarians, however, attempted to can only be fought by making power plants more efficient, de facto firmly steer discussions back to the negotiation of the headline target. supporting so-called cleaner coal. This focus on primary energy disre­ Gierek had been replaced as rapporteur by the time the first debate had gards the opportunity to act on one of the root causes of energy poverty: begun in January 2018, but he continued to argue his position with leaky, inefficient homes that waste energy and drive up bills”.13 Climate regards to his methodology. During the debate, Gierek clashed with his Action Network Europe asserted that: “[Gierek], proposes to dismantle fellow S&D Party member, Kathleen Van Brempt, on how the headline future EU energy efficiency policy and promotes measures that would target was calculated. The confrontation revealed that the two politi­ extend the life of coal power plants” [39]. cians were supporting different methods of calculation. Gierek was Industry criticism came from those set to be negatively affected by defining energy efficiency as a ratio, explaining that “Energy efficiency the proposed changes. EURELECTRIC, a federation representing the is the percentage of energy obtained each time it changes from one state electricity industry in Europe, said the proposals for a new methodology would hamper electrification efforts and undermine efficiency im­ provements and decarbonisation in Europe. Over 50 companies and 14 https://www.europarl.europa.eu/doceo/document/P-8-2017-007339_EN. business organisations in industries that would most likely benefit from html. a focus on end-use efficiency, such as glass manufacturing, lighting, 15 The debates were held on 15 January 2018 and 12 November 2018, see: insulation, electricity and other similar industries, wrote to Gierek https://www.europarl.europa.eu/doceo/document/P-8-2017-007339-AS asking him to support the Commission’s original proposal for energy W_EN.html. 16 efficiency to be expressed both in primary and final energy terms, “to The debates in January and November 2018 also included the reports on proposals for Governance on the Energy Union and renewable energy). 13 News item: “How MEPs misuse energy poverty to water down efficiency legislation,” . Energy Post, 15th November 2017. 5 T. Dunlop and T. Völker Energy Research & Social Science 96 (2023) 102918 Fig. 1. A flyer published by NGO WWF [37] detailing the differences between a 30 and 40 % target. See https://www.wwf.eu/?300131/Joint%20position%20on%20the%20Energy%20Efficiency%20Directive%20review. to another.17 It can be a percentage of final energy in terms of primary “Mr. Gierek, I know that you can scientifically substantiate that we are energy, or a percentage of usable energy in relation to final energy.” Van doing a lot of stupid stuff here. But we are politicians, and politicians Brempt, on the other hand, was supporting the Commission’s proposal, must also be able to ultimately reach compromises.” Similarly, the which is based on reductions in energy consumption (i.e. whole Parliamentary Vice-President, ALDE’s Pavel Telička, said that it was numbers), not a ratio.18 Amid the controversy, other parliamentarians important to move on from the methodological issues and achieve a attempted to focus the discussion on forming consensus on the headline clear majority vote to strengthen the EP’s negotiating position in the target. The Shadow Rapporteur for the PPE group, Markus Pieper, said: trilogues: “I regret on some of the issues we were unable to maintain a broader coalition due to issues which are difficult to comprehend. So let us do a serious vote and let us send out a message and create a good basis 17 for negotiations, because this is key and we do have to be ambitious and Gierek responded by explaining what he called the “technical” definition of realistic”. energy efficiency. He did this by treating energy efficiency as a ratio between primary energy and final energy, which determines the efficiency of the system. The parliamentary report was adopted with a clear majority on He argued that under this scenario, reaching 40 % was a technical impossibility January 17th, 2018 as a plenary resolution (485 votes in favor, 132 if the system efficiency was already 63 % (i.e. 40 + 63 % ≥100). against, 58 abstentions). Gierek’s ideas had gained the support of 51 18 The target (including the 2020 and 2030 targets) is based on an estimation (largely Polish but also cross-party) MEPs. His proposals, however, were of future energy use (in this case, 2030 consumption) and the target is the not adopted, and he voted against key amendments put forward by his ambition to lower that target by a percentage amount [49]. Thus, the target S&D group in the committee vote. The plenary resolution supported a cannot be compared to Gierek’s definition, because it is not a ratio and there­ binding EU target of at least 35 % efficiency improvements by 2030, to fore cannot exceed 100 %. 6 T. Dunlop and T. Völker Energy Research & Social Science 96 (2023) 102918 be calculated using the Commission’s original proposal, in primary and February 2018. Major changes were negotiated, effectively watering- final energy consumption, and achieved by means of indicative national down the proposed headline target. One major change was to loosen targets.19 the requirement for EU countries to make energy efficiency improve­ ments, by removing the obligation to meet both primary and final energy 5.5. Controversy about what energy efficiency means consumption targets. This went against what the Commission and the EP had wanted. Instead, an option was written into the text to achieve either Gierek was again vociferous in the second debate in November 2018, primary or final energy as the Council had wanted.21 This meant that arguing that the directive did not measure “real” energy efficiency as he countries could focus on only one part of the target rather than both, i.e., saw it: “Commissioner, the efficiency directive is, in my opinion, a leg­ that some industries could potentially be left out of the measurement. islative failure, because it concerns restrictions on the use of energy and According to one informant, this issue had been a major sticking point not real efficiency […] today’s vote doesn’t show the real solution. The since 2012, when France and Finland argued to focus only on final en­ people who voted for the amendment don’t understand what it’s about. ergy consumption because they were concerned that a focus on primary They voted for savings, but that’s not efficiency.” What he meant was energy would incentivize reductions in nuclear power in their domestic that limits should be placed on primary energy rather than on final debates. The Commission had explained in its impact assessment why it energy. In particular, he opposed Article 7 of the directive, which set was important to include both primary and final energy consumption in targets for energy savings. Fellow Pole, Jadwiga Wiśniewska from the the indicator [36]. It said that if only final energy is counted, then it ECR supported Gierek in arguing that Article 7’s focus on final energy would generate changes in industry, residential, transport, services and consumption was “confusing energy efficiency with saving energy [...] other final sectors, and not in the generation sector and energy networks this will threaten economic growth and increase energy poverty”. Others where “huge energy efficiency potentials exist.” [36]. The opposite from the conservative parties (EPP and ECR) agreed that energy effi­ would be true if the focus was only on primary energy: “Member States ciency should be distinguished from energy savings. One informant said: could achieve their target without generating any savings in the final “Within our group we had a very strong opinion that energy efficiency energy consumption sectors (e.g. in buildings) if the target is expressed doesn’t mean energy saving; energy efficiency is input of coal and in primary energy consumption only. [If this were the case] Member output of one unit of GDP”. (i.e. energy intensity). The EPP and ECR also States could e.g. focus on a shift towards more renewable energies only, put forward a proposal in the compromise amendments [41] which instead of aiming for energy efficient housing.” [36,pp.75]. stipulated that Member States could discount extra energy consumption Towards the end of the trilogue negotiations, there were calls among that was consumed in the case where economic growth was higher than industry campaigners and the Commission to keep both indicators in the expected.20 This proposal was voted down. target. Paul Hodson, Head of Unit for energy efficiency at the Com­ Energy efficiency, however, meant something different to other mission, said that keeping with the original “and” formulation “could be policymakers. Those from the ENV working group and left-leaning 1% different in ambition from having a target that is expressed as being parties tended to describe energy efficiency as a strategy to limit con­ so much in primary or so much in final. That’s really quite important and sumption and achieve energy savings, sometimes mixing the ideas of not at all obvious,” Hodson told delegates at the annual conference of energy efficiency and savings together. The S&D’s Jytte Guteland, who COGEN Europe (European Association for the Promotion of Cogenera­ was energy efficiency rapporteur for the Committee on Environment, tion Europe) on 5th June.22 COGEN Europe also lobbied for a focus on Public and food Safety (ENVI), said: “I was delighted to get the ambition primary energy, arguing that more emphasis should be placed on of the Committee on the Environment to 40 percent energy savings by making savings in the production and distribution of electricity because 2030 […] With a goal of 32.5 percent more energy efficiency by 2030, of average losses of 60 %, a figure which can reach 75 % for power plants hundreds of thousands of new jobs will be created in Europe, while running on biomass, they said. “As long as the electricity sector is not climate emissions will decrease, our fossil fuel imports will decrease and 100% efficient, measuring energy efficiency in primary energy should be health will improve.” Pavel Telička (ALDE) also aligned energy effi­ considered first”.23 These calls were not applied, however, and as a ciency with reduced consumption: “We need to focus on the other side of result of the trilogues, a provisional agreement was reached on 19th the coin: reducing consumption […] with an emphasis not just on the June 2018 that set an indicative EU target of 32.5 % efficiency im­ opportunities in terms of technology, renewables, etc. but also on provements by 2030, and the final text was signed into law on 10th lowering consumption”. The Commission had stated its definition in its December 2018. impact assessment, that energy efficiency was not about conservation, In April 2018, the European Council for an Energy-Efficient Economy and should not prohibitively limit consumption. “What the energy effi­ released a report [42] that found that the Commission’s 2016 impact ciency policies aim at is not to cap the absolute amount of energy assessment miscalculated the estimated costs required to meet the en­ consumed in a manner that could turn out to be prohibitive in times of ergy efficiency targets.24 The study stated that if the Commission had high economic growth or a very lax target in times of economic slow­ used the member state average discount rate, costs would have been down. Instead, energy should be used in a more efficient way (i.e., using lower and negotiators could have been freer to push for higher targets less energy per unit of economic output)” [36,pp.75]. and annual savings. 5.6. Parliament negotiates a deal with the Council Following the first debate and the EP’s plenary resolution, trilogue negotiations between the EP, Council and Commission began in 21 The final text read: “has to be no >1321 Mtoe of primary energy and/or no >987 Mtoe of final energy. 22 News item: “Primary’ energy efficiency in focus as EU talks near finishing 19 Strengthened provisions on annual (end-use) energy savings obligations line” Euractiv, 8th June 2018. 23 were added, and included a broader range of sectors than in the original News item: “Primary’ energy efficiency in focus as EU talks near finishing Commission proposal, including buildings, transport, heating and cooling. line” Euractiv, 8th June 2018. 20 24 “Where a Member State has not based its contribution on energy intensity, it When the Commission produced its initial energy efficiency proposal, it may detail in its integrated national energy and climate plan […] its expected used a discount rate of 10 % to come up with its 30 % binding target, judging it level of industrial output in 2030 and may subsequently exclude from counting to be the most cost-effective option. However, an April 2018 study by the Eu­ towards its contribution for 2030 any energy consumption resulting from excess ropean Council for an Energy-Efficient Economy (ECEEE) insisted that 10 % industrial output.” (European Parliament, 2017 pp. 5). was too high compared to a member state average of just 5.7 % for buildings. 7 T. Dunlop and T. Völker Energy Research & Social Science 96 (2023) 102918 6. Discussion effect. Another problem was that the Commission’s response to a par­ liamentarian’s question with a typically cut-and-dried explanation of This case study offers insight into a political struggle that revolved how the rebound effect was factored into modelling – with the response around the definition and quantitative articulation of energy efficiency. arriving too late for the parliament to take the answer into account in The case is unique because Gierek’s initiative created a temporary their conclusions. opening for a space where definitions of energy efficiency and the way The debates were in theory a place to openly discuss disagreements, energy efficiency is measured became negotiable. Similar to Asdal’s [5] but they became a place where the conversation was shut down and findings, the “numerical technologies” of defining and calculating en­ energy efficiency was turned back into an apolitical object. Whenever ergy efficiency help produce “a governable space which then lends itself points were raised about methods, the conversation was swiftly turned to political intervention” (pp. 125). The creation of this space was back to the priority for parliamentarians as they saw it – reaching a enabled by the normal parliamentary negotiation proceedings over the consensus on the headline target. There was simply no more time nor draft ITRE report. Importantly, however, the rapporteur - defying pro­ space for in-depth discussions. Gierek, on the other hand, had argued tocol – used another document and attempted to rewrite the Commis­ that the way energy efficiency was measured would have more effect on sion’s amendment proposals, putting forward radical new ways of energy savings than the level of the targets – but his arguments were lost defining and measuring the headline indicator. Gierek’s actions resulted in the louder conversation that proceeded it about the headline target in a politicization of energy efficiency indicators, because he helped level and assumptions that the higher the better. reveal conflicts and trade-offs inherent in a their expression. We can see Finally, the process of negotiation in favor of the Council during the in this case that there were a clash of diverse party interests, lobbies, trilogues meant that EU member states had significantly more flexibility industries and countries in the way that they conceptualized energy in whether to focus their progress towards the target on primary or final efficiency and also in the way that they constructed energy efficiency energy efficiency. This means, on the one hand, that countries may measures and indicators for their own perceived gains. choose to ignore the sectors precisely where Gierek wanted change in Paradoxically, however, a number of political processes and prac­ primary energy. Another downside of this scenario, according to the tices of the EU’s OLP resulted in the opposite process, that actually impact assessment, is that member states could potentially shift atten­ depoliticized the energy efficiency issues. A significant part of this tion away from end-use consumption and “focus on a shift towards more depoliticization was the ambition of many policymakers to reach a renewable energies only, instead of aiming for energy efficient housing” compromise for political gains, especially under time pressure. The [36,pp.75]. This latter example illustrates how energy efficiency and downside of forming compromise under such conditions was that poli­ renewable energy objectives can conflict in practice.25 For example, to cymakers were effectively shutting down and taming the issues that avoid the complexities and policy challenges presented by Gierek’s others, especially Gierek, had raised. The sticky issues included where to interpretation of efficiency and energy sources, primary energy con­ draw the boundary of measures, whether more onus should be placed on sumption could be instead measured and captured by the Emissions industry to make energy efficiency gains, and whether more attention Trading System rather than through the energy efficiency directive. This should be paid to the rebound effect and tackling scientific uncertainty. would also avoid uncomfortable tradeoffs of energy efficiency in dis­ These controversies reveal that policymakers had different epistemo­ incentivizing decarbonisation by reducing the price on carbon in carbon logical backgrounds and different areas of expertise, which clashed trading. when it came to legislating energy efficiency policy. For example, in the view of his parliamentary and Commission colleagues, while Gierek had a strong technical engineering background, he lacked communication 6.2. What do we learn from studying the politics of energy efficiency skills, including the language necessary to speak with colleagues (both indicators? English itself and expressing ideas in a way that colleagues could un­ derstand). His parliamentary colleagues, on the other hand, had exper­ The nature of energy efficiency is that, on the one hand, it can bring tise in political negotiation and compromise, but lacked the skills actors together in agreement to support energy efficiency policy, necessary to understand trade-offs in the formulation of energy effi­ because the broad goal and outcomes appear to mean the same thing: ciency measures and indicators. These disparities resulted in less- using energy less wastefully; and more recently in EU policy, it means effective policy that would otherwise have occurred with practitioners benefits for society in the forms of poverty alleviation, jobs and a better who had a better grasp of all aspects of the process. environment.26 But what is apparent when it is broken down is that energy efficiency can mean very different things when it comes to the way it is implemented through indicators. This finding is reflected in 6.1. How the political process depoliticized the issue: the technologies of historical EU policy [23]. Table 1 below outlines these differences in the politics and “thing-work” epistemic basis of different conceptualizations and what the various enactments of energy efficiency when attached to these meanings. The Commission’s impact assessment helped shape the political ne­ To Gierek, energy efficiency concerned the “technical” ratio rather gotiations because it was the go-to reference for policymakers to argue than energy savings, focused on systems efficiency with the headline how high or low the headline target should be set. It broke policy choices indicator expressed as cumulated primary energy. To some, energy ef­ down into crisp, clean, understandable scenarios measured in dollar ficiency meant energy intensity, and not savings. Others however, terms, jobs and greenhouse gas emissions, thus acting as a useful tool to wanted to focus on a reduction in overall energy consumption. What was defer to when difficult and complex choices needed to be made. being played out in parliamentary discussions was a common conflict Different actors used the impact assessment for their own gain, to and confusion over whether energy efficiency meant relative efficiency, strengthen arguments both in favor of, or against, a higher or lower target. In this way, the impact assessment kept discussions centered on the target level rather than the methodology used to calculate it. While 25 For example, in the energy efficiency legislation, renewables are given an the appeal of this document was its clarity, on the flip side it tended to artificially high PEF factor in order to incentivize their use. What this means is deal with complex issues in a fairly simplistic way. For example, that renewables are considered 100 % efficient when in fact they are often regarding the rebound effect, the options were neatly presented with an comparatively inefficient. Solar panels, for example, are comparatively ineffi­ explanation that the lower estimate (21 %) was favored over the higher cient to other technologies, at between approximately 20–25 % efficiency [50]. estimate (43 %) because the lower estimate was the most recent finding. 26 The typically broad definition of energy efficiency that has been agreed on But this type of simplistic treatment belies the fact that science still since 2012 (energy efficiency means the ratio of output of performance, service, cannot accurately determine the true extent and nature of the rebound goods or energy, to input of energy) reflects this ambiguity. 8 T. Dunlop and T. Völker Energy Research & Social Science 96 (2023) 102918 Table 1 this should not overlook the fact that richer nuclear-dominant countries Different enactments of energy efficiency among policymakers and institutions. advocate for measures that shift the focus away from making nuclear Group/ Preferred Narrative meaning/ Enactment energy generation more efficient. Nuclear is arguably inefficient when institution headline definition of energy taking costs, slow construction times and nuclear waste into consider­ indicator efficiency ation. Despite its downsides, nuclear energy receives favorable treat­ Commission Primary and final Energy should be A mixture of ment when comparing its efficiency to other energy sources, in the form energy used in a more relative efficiency, of the primary energy factor.27 This special treatment puts into question consumption efficient way (i.e., where overall the utility of applying the energy efficiency concept if indicators and less energy per unit savings are not of economic output) necessarily measures are used to prop up wasteful and expensive industries. This achieved, and issue is more relevant given the recent public outcry over the Commis­ absolute energy sion decision to include nuclear on the list of “sustainable investments” savings (Article 7). in its recently created EU taxonomy.28 Gierek and Cumulated Energy efficiency is Relative efficiency, Thus, in this case, it is not necessarily energy efficiency that counts, ITRE primary energy not about energy where overall (including pre- savings but about savings are not but how energy savings are made that is important. This can be seen in conversion phase) being efficient, necessarily the case of Gierek’s proposed methodology to measure energy effi­ especially energy achieved ciency. He was criticized for supporting coal because of his focus on generation and generating innovation in primary energy efficiency. Given the reliance conversion processes Conservatives Energy intensity Energy efficiency Relative efficiency, of Poland on coal for citizen’s livelihoods, it makes sense that Gierek and does not mean where overall his Polish parliamentary colleagues would advocate for efficiency savings savings are not measures in coal. There are potential energy efficiency gains to be made necessarily in coal plants, however, this option was not politically desirable to achieved others who saw the move as supporting the fossil fuels industry. S&D and Primary and final Energy savings, Absolute energy Greens energy reducing/lowering savings Although coal may be a ‘dirty’ fuel, supporting it may arguably sustain consumption energy consumption Poland’s energy poor. Thus, the question arises: how do we deal with energy inequalities – in terms not only of who is richer or poorer, but also of who consumes more per capita, and how does their energy mix where overall savings are not necessarily achieved, OR, whether it affect energy poverty? This is a familiar conundrum at the crux of global means a reduction in overall consumption and energy savings in and of climate negotiations – who should bear the burden of emissions re­ itself, more akin to energy sufficiency or conservation principles. The ductions? The answer is not simple and requires more than the current differences in meaning highlight the role of ambiguity in energy effi­ policy process to solve. Unfortunately, the format of EU policymaking ciency politics. According to political scientists, ambiguity is an essential involving the use of impact assessments and technical documents tends ingredient in policymaking: “Ambiguity enables the transformation of to favor a technocratic type of governance tht doesn’t leave room for individual intentions and actions into collective results and purposes. more diverse viewpoints and perspectives to be heard. More wider cit­ Without it, cooperation and compromise would be far more difficult, if izen involvement in energy efficiency policy could help to weigh up not impossible” [43,pp.178]. complex issues to do with values, identities, culture and inequality with It is, nevertheless, important to analyze the underlying differences of regard to energy use. opinion in the meaning of energy efficiency, to understand what are the While the Commission’s proposal focused more on energy savings various effects of policy implementation and different enactments of made at the end of the energy chain, it is largely consumers and tax­ energy efficiency. In doing so, we aim to contribute to the very context- payers who foot the bill for many of these initiatives, unless funding specific and situated analyses that energy social scientists have called for comes from the public purse. Improvements, for example, include [19,20]. By focusing on primary, rather than final energy, Gierek was housing insulation, buying more efficient appliances and more trans­ enacting an energy efficiency that was distinct from that of the Com­ parent billing practices by energy companies. These strategies are seen mission, i.e., one that focused more on production than consumption as ‘win-win’ because they support growth in industry, for example in systems. While Gierek argued that targeting primary energy would help glass, insulation and electricity companies, while at the same time to reduce energy losses and alleviate poverty by taking the burden off helping consumers save money from the energy they save. There still consumers to reduce energy at the final end of the chain, the Commis­ exist, however, perverse incentives that, for example, reward energy sion made the opposite argument, that the best way to tackle energy companies for the energy they sell rather than for cutting bills; that pay poverty is to improve energy efficiency at the final energy consumption architects and engineers for what they spend, not what they save; and stage. Who is right? A pertinent query here is to what extent consumers that put the burden for renovation on renters, rather than on apartment would benefit from more targeted measures to improve energy effi­ or building owners. Furthermore, as the rebound effect literature shows, ciency on the supply side, where significant savings can still be made. for e.g. [44], citizens may not necessarily be incentivized to reduce Comparatively richer EU countries may argue that there are few eco­ nomic gains to be made in primary energy industries (supply side), but 27 The Primary Energy Factor (PEF) is the rate to which electricity is converted between different stages in the energy chain. It should reflect how much pri­ mary energy is used to generate a unit of final energy (e.g. electricity or thermal energy) through the use of consumption indicators. The current PEF for elec­ tricity that is prescribed within the European Union, as detailed in the Energy Efficiency Directive 27/2012/EU, is 2.5. That implies that each unit of elec­ tricity requires an input of 2.5 units of primary energy to produce it. So, for every unit of electricity consumed in our homes, 2.5 times more energy was needed to generate it. Therefore, the greater the difference between primary energy consumption and final energy consumption, the larger the implied losses throughout the energy system. For nuclear, this figure is placed at 3/1 which is more generous than conventional power stations which have a PEF ratio of 2.5. 28 https://www.euractiv.com/section/energy-environment/news/investors- warn-green-label-for-gas-undermines-eu-taxonomy/ 9 T. Dunlop and T. Völker Energy Research & Social Science 96 (2023) 102918 energy overall, but rather to use it at a more efficient rate, for example [2] M. Patterson, What is energy efficiency? Concepts, indicators and methodological issues, Energy Policy 24 (5) (1996) 377–390, https://doi.org/10.1016/0301-4215 through more efficient appliances. A nuanced policy to reduce energy (96)00017-1. use and energy poverty may be more effective in this regard, for [3] T. Goh, B. Ang, Four reasons why there is so much confusion about energy example, policies that focus on absolute reduction in energy such as zero efficiency, Energy Policy 146 (August) (2020), https://doi.org/10.1016/j. energy buildings (focusing on a mix of renewables, energy sufficiency, enpol.2020.111832. [4] L. Pérez-Lombard, J. Ortiz, D. Velázquez, Revisiting energy efficiency and energy efficiency) whereby buildings generate more energy than fundamentals, Energy Effic. 6 (2) (2013) 239–254, https://doi.org/10.1007/ they consume [45]. s12053-012-9180-8. Overall, this research highlights how a certain way of measuring [5] K. Asdal, Enacting things through numbers: taking nature into account/ing, Geoforum 39 (1) (2008) 123–132, https://doi.org/10.1016/j. energy efficiency also contributes to stabilizing a certain imaginary of geoforum.2006.11.004. energy production and consumption together with an imaginary of [6] J. Niskanen, H. Rohracher, A politics of calculation: negotiating pathways to zero- science and measurement that goes with it – and that this imaginary energy buildings in Sweden, Technol. Forecast. Soc. Chang. 179 (October) (2020) 2022, https://doi.org/10.1016/j.techfore.2022.121630. could be very different depending on the various interests and actors at [7] T. Dunlop, Mind the gap: a social sciences review of energy efficiency, Energy Res. play. Soc. Sci. 56 (Oct. 2019), https://doi.org/10.1016/j.erss.2019.05.026. [8] G. Owen, Energy efficiency and energy conservation: policies, programmes and their effectiveness, Energy Environ. 11 (5) (2000) 553–564. 7. Conclusion [9] N. Bergman, Energy efficiency or energy demand?, in: ECEEE 2007 Summer Study, 2019, pp. 351–360. This study has revealed some of the disputes over political priorities, [10] N. Labanca, P. Bertoldi, Beyond energy efficiency and individual behaviours: policy insights from social practice theories, Energy Policy 115 (April 2018) (2018) values and trade-offs by applying a sociological lens to energy efficiency 494–502, https://doi.org/10.1016/j.enpol.2018.01.027. indicators. This study builds upon the so far sparse literature on the [11] M. Moezzi, Decoupling energy efficiency from energy consumption, Energy politics of energy efficiency measurement. By treating energy efficiency Environ. 11 (6) (2000) 521–537. [12] C. Calwell, Is Efficient Sufficient?, 2010. Stockholm. as a governance object, we have shown how the dichotomic processes of [13] A. Barbu, N. Griffiths, G. Morton, Achieving Energy Efficiency Through Behaviour politicization and depoliticization occur in energy efficiency policy - Change: What Does It Take?, 2013. Luxembourg. through so-called ‘technical’ methodological discussions and policy [14] P. Brockway, S. Sorrell, G. Semieniuk, M.K. Heun, V. Court, Energy efficiency and economy-wide rebound effects: a review of the evidence and its implications, processes that use their own quantitative language - which may be Renew. Sust. Energ. Rev. 141 (January) (2021), https://doi.org/10.1016/j. difficult for non-experts to understand. Through a politization of the rser.2021.110781. issues we were able to glimpse some of the sticky issues that impair the [15] R. Velasco-Fernández, T. Dunlop, M. Giampietro, Fallacies of energy efficiency indicators: recognizing the complexity of the metabolic pattern of the economy, energy system, including special treatment for certain industries, loop­ Energy Policy 137 (2020), https://doi.org/10.1016/j.enpol.2019.111089. holes for EU countries and a lack of focus on overall consumption [16] E. Shove, What is wrong with energy efficiency? Build. Res. Inf. (2017) 1–11, reduction. Opening up these governance issues and in tandem under­ https://doi.org/10.1080/09613218.2017.1361746. standing how they become depoliticized through bureaucratic processes [17] T. Princen, The Logic of Sufficiency, The MIT Press, Cambridge, 2005. [18] A. Reddy, Barriers to improvements in energy efficiency, Energy Policy (December) is key to understanding how to improve policy. A first step would be to (1991) 953–961, https://doi.org/10.1016/0301-4215(91)90115-5. openly discuss the tension between relative and absolute efficiency. The [19] A. Genus, M. Iskandarova, G. Goggins, F. Fahy, S. Laakso, Alternative energy application of “energy sufficiency” principles has already gone some imaginaries: implications for energy research, policy integration and the transformation of energy systems, Energy Res. Soc. Sci. 73 (January) (2021), way to doing this, by pushing to openly discuss the subjectivities https://doi.org/10.1016/j.erss.2020.101898. involved in placing absolute limits on energy consumption. For example, [20] A. Nikas, et al., The desirability of transitions in demand: incorporating an emphasis is placed on better understanding what constitutes ‘needs’ behavioural and societal transformations into energy modelling, Energy Res.Soc. Sci. 70 (September) (2020), https://doi.org/10.1016/j.erss.2020.101780. and ‘enough’ in terms of consumption: are two cars per family too [21] J. von Platten, M. Mangold, K. Mjörnell, A matter of metrics? How analysing per much? How much living space is enough? Is air conditioning set at 18 capita energy use changes the face of energy efficient housing in Sweden and degrees Celsius necessary? Meaningful change will only occur when we reveals injustices in the energy transition, Energy Res.Soc. Sci. 70 (February) (2020), https://doi.org/10.1016/j.erss.2020.101807. openly address these societal issues and alter our energy supply and [22] W. Park, N. Shah, K. Shiraishi, E. Vine, Improving energy performance metrics to consumption systems, and the systems that govern them, to live maximize the benefits of disruptive technologies, Energy Res.Soc. Sci. 89 (May) sustainably. (2022), 102678, https://doi.org/10.1016/j.erss.2022.102678. [23] T. Dunlop, Energy efficiency: the evolution of a motherhood concept, Soc. Stud. Sci. (2022), https://doi.org/10.1177/03063127221096171. [24] S. Merry, The Seductions of Quantification Measuring Human Rights, Gender Declaration of competing interest Violence, and Sex Trafficking, University Chicago Press, 2016. [25] E. Turnhout, K. Neves, E. de Lijster, in: ’Measurementality’ in Biodiversity Tessa Dunlop reports the following financial support for the research, Governance: Knowledge, Transparency , and the Intergovernmental Science – authorship, and/or publication of this article: This research was sup­ Policy Platform on Biodiversity and Ecosystem Services (IPBES) vol. 46, 2014, pp. 581–597, https://doi.org/10.1068/a4629. ported by the European Union’s Horizon 2020 research and innovation [26] T. Völker, Z. Kovacic, R. Strand, Indicator development as a site of collective programme under Grant Agreement No. 689669 (MAGIC) and the imagination? The case of European Commission policies on the circular economy, Spanish Ministry of Science and Innovation (MICINN), through the Cult. Organ. 26 (2) (2020) 103–120, https://doi.org/10.1080/ 14759551.2019.1699092. ‘María de Maeztu’ program for Units of Excellence (CEX2019-000940- [27] K. Asdal, On politics and the little tools of democracy: a down-to-earth approach, M). This work reflects the authors’ view only; the funding agencies are Distinktion J. Soc. Theory 9 (1) (2008) 11–26, https://doi.org/10.1080/ not responsible for any use that may be made of the information it 1600910x.2008.9672953. [28] A. Barry, Political Machines: Governing a Technological Society, Bloomsbury contains. The first author was a student in the PhD program of the Publishing, London, 2001. Institute of Environmental Science and Technology (ICTA) at the Uni­ [29] M. Callon, Markets in the Making: Rethinking Competition, Goods, and Innovation, versitat Autònoma de Barcelona. Princeton University Press, 2021. [30] J. Law, Seeing like a survey, Cult. Sociol. 3 (2009) 239, https://doi.org/10.1177/ 1749975509105533. Data availability [31] B. Saunders, et al., Saturation in qualitative research: exploring its conceptualization and operationalization, Qual. Quant. 52 (4) (2018) 1893–1907, https://doi.org/10.1007/s11135-017-0574-8. The data that has been used is confidential. [32] A. Bogner, B. Littig, W. Menz, Interviewing Experts, Palgrave Macmillan, London, 2009. References [33] A. Srivastava, S.B. Thomson, Framework analysis: research note, J. Adm. Gov. 4 (2) (2009) 72–79. [34] M. Lamont, A. Swidler, Methodological pluralism and the possibilities and limits of [1] European Commission, Proposal for a Directive of the European Parliament and of interviewing, Qual. Sociol. 37 (2) (2014) 153–171, https://doi.org/10.1007/ the Council Amending Directive 2012/27/EU on Energy Efficiency, 2016. COM s11133-014-9274-z. (2016) 761 Final. 10 T. Dunlop and T. Völker Energy Research & Social Science 96 (2023) 102918 [35] ITRE, Draft Report on the Proposal for a Directive of the European Parliament and [40] European Alliance to Save Energy (EU-ASE), Deutsche Unternehmensinitiative of the Council Amending Directive 2012/27/EU on Energy Efficiency (COM(2016) Energieeffizienz (DENEFF), Letter to Adam Gierek [Online]. Available: https:// 0761 - C8-0498/2016/0376(COD)), 2017. mehi.hu/sites/default/files/business_letter_eed_final-1.pdf, 2017 [Accessed: 01- [36] European Commission, Impact Assessment (Accompanying the Document Proposal Jan-2021]. for a Directive of the European Parliament and of the Council Amending Directive [41] European Parliament, Compromise Amendments: Proposal for a Directive of the 2012/27/EU on Energy Efficiency), 2016. Brussels, Belgium, SWD(2016) 405 European Parliament and of the Council Amending Directive 2012/27/EU on final/2. Energy Efficiency, European Parliament, Brussels, Belgium, 2017. [37] WWF, JQ Visual Communications, Stop Missing Opportunities: Energy Efficiency [42] ECEEE, Between a Rock and a Hard Place, 2018. Delivers, 2017. [43] D. Stone, Policy Paradox: The Art of Political Decision Making, Third, Norton, New [38] European Environmental Bureau, “MEPs must support, not sabotage, the Energy York, London, 2012. Efficiency Directive,” Energy Savings [Online]. Available:, 2017 [Accessed: 01-Jan- [44] J. Freire-gonzález, Evidence of direct and indirect rebound effect in households in 2021], https://eeb.org/meps-must-support-not-sabotage-the-energy-efficiency-d EU-27 countries [Online]. Available: Energy Policy 102 (November 2016) (2017) irective/. 270–276, https://doi.org/10.1016/j.enpol.2016.12.002 [Accessed: 12-Jun-2017]. [39] Climate Action Network Europe, Polish MEP misuses energy efficiency file to [45] Y. Saheb, S. Shnapp, C. Johnson, The zero energy concept: making the whole promote coal [Online]. Available: http://www.caneurope.org/publications/press greater than the sum of the parts to meet the Paris Climate Agreement’s objectives, -releases/1422-polish-mep-misuses-energy-efficiency-file-to-promote-coal, 2017 Curr. Opin. Environ. Sustain. 30 (2018) 138–150, https://doi.org/10.1016/j. [Accessed: 12-Jun-2017]. cosust.2018.04.014. 11