Response: Writing the wrongs of energy efficiency

A response to Jan Rosenow and Tina Fawcett’s Commentary: ‘What is right with energy efficiency?’

Elizabeth Shove, December 2017

In writing about what’s wrong with energy efficiency my aim was to bring a body of established social theoretical ideas to bear on a concept that has taken hold – and become taken for granted – in energy research and policy.  I wanted to do this in order to provoke considered reflection on the conceptual foundations of the very idea of energy efficiency.

This is an unusual undertaking and misunderstandings are to be expected.  Some are a product of the very issue that I seek to address:  that is the extent to which certain lines of thought have become so institutionalised and so ‘normal’ that there is no space left for critical reflection. Other potential hiccups relate to the fact that I am approaching the topic as a social scientist – and what’s worse, as one interested in the philosophy of knowledge, and in the performativity of discourses, ideas and practices.  This is not at all the same position as that occupied by readers who are in the business of ‘doing efficiency’, or energy policy, or who have no background in the ideas on which I draw.

That is not a problem. It is my job, in writing the article, to convey enough of these intellectual traditions to enable people immersed in the field of ‘energy efficiency’ to see what they are caught up in.  There are some signs that my article has had this effect.  For instance, Gerrit Jan Schaeffer, Director of Dutch Solar Energy thanks me for taking the effort to think through the concept of energy efficiency ‘in a thoroughly intellectual way’ (a tweet on 6th September 2017). Others  have acknowledged the importance of articulating some of the social, historical and political assumptions that are embedded in ‘efficiency’ as it is mobilised in contemporary policy. But for Jan Rosenow, for Tina Fawcett, and quite likely for other readers too, I have evidently not got through.  I am not about to repeat my entire argument (so please read this response alongside the original article), but their  reaction is a useful diagnostic tool, signalling what I take to be four main ‘error messages’.  In responding to these alerts my aim is to counter readings of the article that remain rooted within the ‘efficiency paradigm’.

Error message 1. The article is about whether energy efficiency ‘works’ in its own terms or not. 

In their response, Rosenow and Fawcett include a graph that they use to demonstrate the value of energy efficiency. There are, of course, many other such graphs that show the same.  My question is not whether the figures on which these graphs depend are right or wrong.  Instead, and as my extensive discussion of the ‘purifying’ work involved shows,  my aim is to  highlight the taken-for-granted assumptions and the work on which all such data is based.  Presenting evidence that demonstrates the benefits of energy efficiency, in its own terms, is in no way a counter claim, or a rebuttal of my position.  If anything it serves to illustrate just how deeply embedded and how uncritical efficiency discourses have become.

Error message 2. The discussion is about whether efficiency measures, and other energy policies, can combine to combat climate change. 

Quite early on in the article I make reference to some of those who claim that an effective response to climate change – including meeting future targets – is not only about finding more efficient ways of meeting present ‘needs’: as Obama said in 2016, ‘more will be needed’. In the UK, the Committee on Climate Change is also not sure how future targets will be met without the government applying ‘more challenging measures’, perhaps calling present standards of living into question. Rosenow and Fawcett want  ‘evidence’ that efficiency measures (alongside other technical fixes) might not be up to the task ahead.  In reality it is impossible to know how future ‘demands’ might evolve and what new ways people might be using energy in two or ten or a thousand years’ time.  Accordingly there can never be ‘evidence’ to show that efficiency will be enough, or to show the reverse either.

This is not where the argument lies. Instead, and as I sought to explain, the much more important point is to recognise that societal needs and demands are not given: they are negotiable, dynamic, and in part constituted by technologies and policies, including those of efficiency.  In this more metaphorical sense, efficiency arguments are not enough because they disguise and do not engage with the ways in which present and future demands are made, or that efficiency measures are themselves a crucial part of this process.  In other words, the potential to respond to climate change depends on what energy is used for and how societal ‘needs’ evolve. My case is that the pursuit of energy efficiency has many unintended consequences, within and as part of this story.

Error message 3. There are societally or universally agreed needs and demands and energy efficient systems and technologies help meet these needs with fewer resources.

I completely agree that efficiency measures  help meet pre-specified needs with fewer resources.  But in taking ‘needed’ services for granted such measures also ‘fix’ historically and culturally specific interpretations of comfort, light, travel, etc.  For a time, and for certain groups of people, these interpretations constitute ‘real’ needs.  But as most historians would explain, these are not universal,  nor are they always agreed.  The error is to suppose that there are societally or universally agreed needs.  In response, my article is designed to help proponents and defenders of efficiency discourses see just how much political and cultural baggage is attached to the seemingly neutral concepts with which they work.

Error message 4. In being critical of efficiency discourses I am at risk of raising politically awkward questions and inadvertently lending support to those in favour of boosting energy supply.

In the article I imply that discourses of efficiency are politically convenient because they disguise their own highly political role in sustaining specific interpretations of need and service.  On this point  I do want to rock the boat.   One reason for writing the article was to draw attention to just these hidden features.  I think this is important in that identifying what I call ‘good’ forms of efficiency depends on carefully articulating the meanings of ‘need’ and service that they sustain.  And as I argue, concepts of purification and entanglement help in this endeavour.   In my view, engaging with what is wrong with discourses of efficiency is a necessary step in working towards a more considered response to climate change: it is not a sign of weakness.  Far from it. Being critical of the intellectual foundations of policy discourses of efficiency – or for that matter of other related concepts like the energy ‘trilemma’ – is part of a necessary process of conceptualising and interrogating the dark sides of conventional wisdom.

To reiterate, the article is not about  whether efficiency should get more attention in policy spheres than additional supply.  The question is what commitments and assumptions do discourses of efficiency entail, and what perverse, unwitting and unintended effects might these have?  In particular, how do references to efficiency reproduce specific interpretations of ‘service’?

Rosenow and Fawcett’s discussion of ‘what’s right with energy efficiency’ represents a simple defence of the ambition of energy efficiency and because it remains at this level it does not offer a theoretically informed critique of ‘what’s wrong with energy efficiency’.  This is a shame in that there is plenty of scope for taking issue with various features of the argument that I make.   For example, a more interesting charge, and one that political scientists might reasonably make, is that I am myself beguiled by the dominant discourse of energy efficiency and have made the mistake of taking it at face value.  In other words, I fail to reference some pretty basic ideas about the rhetoric of policy making and politics and fail to recognise that concepts of efficiency are significant not in their own right but as a symptom of much bigger and deeper political and economic battles.

Another credible challenge is that I confuse or at least conflate ‘real’ forms of efficiency (based in physics and engineering) with the very much fuzzier versions that abound in policy documents.  For example, some purist engineers agree with aspects of my position, feeling that the notion of ‘efficiency’ has been repeatedly hijacked and misused in policy circles in which it is taken to be a generalised ambition, not a context specific measure of input and output.  My mistake is to treat ‘efficiency’ discourses as one.

Third, and for those who know the literature, my attempt to apply and work with notions of purification and entanglement skates over some rather important differences between Hodder and Latour. Similarly, I do not really engage with Hodder’s ideas about the path-dependence and directionality of entanglement.  This is quite significant, especially when dealing with long term issues like those of climate change.

Others might reasonably take me to task over the somewhat slippery solution of distinguishing between good and bad forms of efficiency, and they might press me on how notions of ‘good’ efficiency do or do not relate to debates about need and sufficiency.

I’ll resist the temptation to respond to these self-set challenges.  The point is not to resolve these questions now, but to show that there are important debates still to be had about the theoretical foundations of efficiency policy, and what these mean for  how questions of energy are measured, managed, understood and known. Sliding these topics under the carpet, insisting that efficiency ‘works’, and contending that it is politically important to defend efficiency programmes come what may is not a long term solution.

Further reading: Not everyone will be interested in  thinking about energy efficiency in these terms, but for those who are curious about how  scientific ideas and policy discourses  ‘work’ I’d suggest a few key readings.  One is The Structure of Scientific Revolutions, by Thomas Kuhn, first published in 1962.  Another would be Don Mackenzie’s excellent book ‘Inventing accuracy’ (1990, MIT Press). Mackenzie writes about nuclear missiles but many of the ideas he explores apply as well to energy efficiency. A third suggestion would be to read more of Latour.  The book to which I refer in the article, Bruno Latour’s We Have Never Been Modern, (1993, Havard University Press) is not an easy starting point, but some of Latour’s earlier work, like Laboratory Life: The Construction of Scientific Facts, with Steve Woolgar (1979, Sage) would be a good place to begin.


Elizabeth Shove is Professor of Sociology at Lancaster University and PI of the DEMAND research centre (Dynamics of energy, mobility and demand). She is known for her work on social practice theory, including The Dynamics of Social Practice (Sage 2012) with Mika Pantzar and Matt Watson, and more recently, Nexus of Practices (Routledge 2017) edited with Allison Hui and Theodore Schatzki.  She has also written about different aspects of energy and demand in everyday life.

ORCID number: 0000-0002-4792-5479

Energy Performance Gaps: Promises, People, Practices

Special issue

Energy Performance Gaps: Promises, People, Practices

Kirsten Gram-Hanssen and Susse Georg introduce their forthcoming Special Issue

The building sector has been identified as the sector with the largest potential for energy savings. Yet in spite of great technical progress, a large part of the savings have yet to be realized. If the performance gaps are to be lessened, then the potential of energy savings from the residents and  building users and their practices must be considered. For example, building occupants develop  higher expectations of comfort and they often lack an understanding of the control of building technologies. But it is also true that many technologies are not adapted to user needs: they lack user friendliness and can lead to unnecessary and unwanted consumption. However, the practices of the building professionals, their considerations regarding the users and how these considerations are treated in the design, construction and installation processes, or the operation of the buildings, also must be included.

To achieve energy savings, future energy efficient technologies have to meet users’ needs to a greater extent and support their practices in a sustainable direction, and this counts for residents as well as building professionals and their practices. Pursuing this requires deeper understanding of differentiated user practices and their relations to energy consumption. Our recent Building Research & Information special issue on “Energy Performance gaps: Promises, People, Practices” with a curated collection of papers, is dedicated to understanding this and it proposes new directions for building research and practice to actually be able to realise the needed savings.

Policy has, hitherto, been effective in promoting the development of more energy efficient buildings. But policy has not curbed the continuously  growing energy consumption.  Technological efficiency on its own is unlikely to provide the dramatic and rapid reductions needed to meet climate targets. It is therefore also debatable whether the usual economic arguments underlying McKinsey’s ”global GHG abatement cost curve”, establishing the building sector as the most cost effective field for reducing consumption, are still valid. In principle, it is easy to reduce energy consumption in buildings, but the invol

vement of people changes this simple picture, as documented in the articles in the Special Issue.

The conclusions from the special issue include at least three points of relevance for public policy:

  • First, policy must move beyond just focusing on efficiency of buildings and theoretical energy reductions. Policy needs to be more firmly based on understandings of how new technologies will also introduce new practices and new norms of what home, comfort and a good everyday life is.
  • Second, it is necessary to consider how to develop new (smart) technologies which people can understand, domesticate and use, while at the same time consideration must be given to whether these new technologies induce more consumption rather than help consumers to save energy.
  • Third, if policy is to deliver the needed radical CO2 and energy reductions in buildings, then a shift of focus is needed away from regulation as a one-time gateway which only assesses the design and material aspects of a building. Instead, policies and regulations should consider alternatives which include an ongoing assessment of both the material and social aspects of building operation, along with guidance and support to occupants.

There is much to be gained from a policy and regulatory regime that takes a ‘both-and’ approach to technical and social issues. The research in our special issue collectively makes a strong case for focusing more actions on user practices in energy.

About the authors

Seniorforsker Kirsten Gram-Hanssen, SBI 

Kirsten Gram-Hanssen is a Professor at the Danish Building Research Institute (SBI), Aalborg University Copenhagen. ORCiD: 0000-0002-8543-2501


Susse Georg is a Professor in the Department of Planning, Aalborg University Copenhagen. ORCiD: 0000-0001-9981-2444

Commentary: What is right with energy efficiency? A response to Elizabeth Shove

By Tina Fawcett and Jan Rosenow


Elizabeth Shove’s recent paper ‘What is wrong with energy efficiency’ (Shove, 2017) set out three key challenges to energy policy researchers (and others) – to take stock, to reflect on the consequences of our work, and ‘to develop strategies and solutions that challenge rather than reproduce increasingly problematic assumptions about present and future ways of life’. Here, we respond to the first two challenges to present a positive and nuanced account of the role of energy efficiency, based on experience in the UK and EU. We question the assumption underlying the third challenge, that current levels of energy services are incompatible with a low carbon future.

Our aim is to add to a constructive debate on energy efficiency, its role to date, and its place in the low carbon transition.

Firmly in the mainstream or Cinderella?

The paper characterises energy efficiency as being firmly in the mainstream of energy policy making.  Citing an IEA report, the paper claims that “[i]n the energy world, efficiency is treated as a ‘fuel’”. But is it? The IMF (2015) estimated that global energy supply subsidies amounted to $5.3 trillion in 2015. This compares to just $221 billion invested in energy efficiency (IEA 2016) which is approximately 4% of the total global energy subsidies. It is a large amount of money but compared to subsidies for energy supply it is very small. Further, investments in efficient solutions are impeded by numerous market barriers to individual action; and on the supply side, industry traditions, business models and regulatory practices have always favoured, and continue to favour, fossil fuel based energy infrastructure and sales over lower sales and energy saving technologies.

Characterising energy efficiency as “politically attractive” and “uncontroversial” is contrary to the experience of those working on energy efficiency policy. As Nick Butler (2014), former Group Vice President for Policy and Strategy Development at BP, puts it “[e]fficiency is the neglected Cinderella of the [energy] policy world”. Current debates around energy efficiency at European level are highly political and far from being uncontroversial (see for example Rosenow 2017).

Challenging current levels of energy services would undoubtedly be more politically sensitive and controversial than promoting energy efficiency. However, the many political, market design and financial challenges facing efficiency should also be acknowledged.

How effective are energy efficiency policies?

The paper’s fundamental argument is that “the problem with efficiency policies is that they are much too effective, not in reducing demand but in reproducing and stabilizing essentially unsustainable concepts of service”. The first part of this statement suggests that efficiency policies do not reduce demand.

However, there is strong evidence for demand reduction due to efficiency policy.  For example, in the UK’s residential sector, total energy use fell by one fifth between 2004 and 2015, despite a 12% increase in the number of households – as shown in the graph below.

Energy consumption 2000-2015 Sheet1 Chart 1.jpg


Index of total residential electricity and gas use, where gas figures are weather corrected, UK, 2004-2015 (2004 = 100). (Source: BEIS (2017): Energy Consumption in the UK)

Average household energy use fell by 27% over this time, despite a proliferation of household appliances and lamp fittings per household and higher in-home temperatures (BEIS 2016). Sceptics will argue that this was driven by an increase in energy prices, not policy. However, price elasticities of energy demand are very low – a review of the literature concluded that “[p]rice increases […] do not lead to strong demand reductions” (Jamasb and Meier 2010:3). In addition, several thorough quantitative studies show that efficiency improvements were by far the most important driver of the reduction in consumption (CCC 2017; CEBR 2011; Odyssee 2017).

Levels of service – does efficiency promote an unsustainable status quo?

The second part of the argument is that energy efficiency maintains ‘unsustainable concepts of service’. Further, these service demands are presented as a key barrier to a low carbon future, but without specifying supporting evidence.  By contrast, many detailed studies and scenarios show it is possible to have lower carbon futures through varied mixtures of energy efficiency, fuel switching, better control, demand response and renewable energy, while maintaining current service standards (Boardman, 2012; CCC, 2015; IEA, 2013; National Grid, 2017). If current service expectations really are unsustainable – strong evidence needs to be put forward to make the case. Not least because of the very high public and political priority afforded to key energy services such as universal minimum levels of household warmth (Public Health England, 2014).

Good and bad energy efficiency

Energy efficiency is, or should be, part of the whole systems energy policy mix (including renewable and low carbon energy sources, demand response, smart systems) which delivers low carbon, secure and affordable energy. We agree with Wilhite and Nørgård (2004) that energy efficiency programmes should to be designed with the aim to reduce total energy consumption levels rather than focusing solely on relative efficiency improvements. For us, this is ‘good’ energy efficiency, with further ideas on how energy conservation can be designed into efficiency policy to be explored (e.g. Harris et al. 2008).

Shove is right that we should reflect on the effects of energy efficiency – to be sure it continues to meet socially agreed goals, and does not have adverse consequences. We must recognise that energy efficiency can facilitate increasing demands for energy service. We should also acknowledge that this can sometimes be a good thing:  it can deliver important energy services to people who could not access them otherwise. In developed countries, energy efficiency can deliver affordable warmth to people living in fuel poverty (Boardman 2010). In developing countries,  efficient lighting, often in combination with small scale renewable energy supply, and more efficient cook stoves can be transformative (UNEP 2016).  In reaching a low carbon future, both challenging (increasing) energy service standards and delivering effective energy efficiency will be important.


We think there is a lot right with energy efficiency. There is robust evidence that it has reduced total energy demand and consequent carbon emissions, despite many other drivers of growth. We suggest that rather than being a barrier to the low carbon transition, energy efficiency is a vital part of it.

However, energy efficiency is still not sufficiently in the mainstream of energy policy. By asserting that energy efficiency does not work, Shove risks taking the same side as those who defend the status quo and want to marginalise the demand side. We do not believe this is her intention, and invite her to reconsider her view of energy efficiency and its place in delivering a radically lower carbon future.

About the authors

Tina Fawcett

Dr Tina Fawcett is a Senior Researcher and Acting Deputy Leader at the  Environmental Change Institute, University of Oxford. ORCID: 0000-0003-3953-3675


Jan Rosenow

Dr Jan Rosenow is a Senior Research Fellow for the Centre on Innovation and Energy Demand at the Sussex Energy Group in the Science Policy Research Unit (SPRU), University of Sussex and Senior Associate of the Regulatory Assistance Project. ORCID:  0000-0002-5227-3129

This work was undertaken as part of the research programme of the UK Energy Research Centre, supported by the UK Research Councils under EPSRC award EP/L024756/1.