energy costs were lower than the labour costs, for example. It is not surprising that this led to demand growth. But if energy efficiency is the goal, rather than the means, then the driving force is entirely different - and so are the macroeconomic implicaThe sharpening energy efficiency debate has revived claims that the apparent tions. Consider for example a sector savings from using more efficient technologies would be largely offset by the macroeconomic response - the tendency to use more energy services because they where the energy-using activity is reare made cheaper. When energy price or availability constrains demand this is latively unresponsive to price. In the correct. In more normal circumstances, it may be true for efficiency changes course of natural developments there which occur as part of general economic trends, but not for policy-driven may be little incentive for efficiency to measures aimed at bringing more efficient technologies into imperfect markets. In improve, irrespective of the technical such cases there will usually be real net energy savings only slightly smaller than potential. Yet these are precisely the sectors which are likely to offer the those suggested by a simple engineering analysis, and clear economic benefits. greatest energy savings. Conservation policy can choose to focus upon such Recent communications in Energy ment contains at least two major areas, including those dominated by Policy by Brookes 1 and Greenhalgh 2 flaws. The first, and lesser, flaw is to market failures, where the implicit have disputed the claims of those who assume that future economic reactions price falls from increased efficiency believe that improving energy efficien- will parallel those of the past. There have little impact on activity levels. cy can provide large energy savings. It are several reasons why this is not so. Macroeconomic data aggregates the is argued that the macroeconomic re- The bulk of historic data (up to 1973 behaviour of end-uses which span a sponse to improving efficiency, and the period since then has been partly wide range of elasticities and other the associated fall in energy costs, will energy-constrained and this is not the market responses. In many respects offset any apparent savings. This is not main data Brookes uses) is gathered the focus and implications of efficiena new argument, and is known as the from an era of declining real energy cy improvements which result from 'rebound effect'. prices, very rapid electrification, and policies aimed at saving energy will be Brookes offers the most cogent h~s- economic development based on the opposite of those occurring naturtorical macroeconomic analysis and heavy industry, with domestic demand ally, which tend to focus on the areas arguments. He rightly points out that growth based on satisfying fairly basic where activity levels are most responwhen energy supply or price is a con- desires such as adequate warm homes. sive to implicit price falls. Consestraint on economic activity, impro- The last two decades have shown great quently, little or nothing can be ving efficiency will not help to reduce changes in the pattern of energy and learned from general historic trends demand. In the much more usual economic development. Few people about the energy or economic implicasituation, where energy is not a con- believe that a long era of steadily tions of policies directed explicitly at straint, historical trends at least up to declining fossil fuel prices will ever saving energy. This is an important 1973 show that although energy pro- return, and there is much evidence general point: the critical distinction ductivity improved, it did not do so that some important end-uses of ener- between 'natural' and policy-driven nearly as fast as labour and capital gy are approaching saturation, 3 whilst efficiency improvements invalidates productivity, so that efficiency im- the pattern of economic development many recent attempts to use past macprovements were more than offset by is shifting towards less material- and roeconomic data to draw policy conthe new uses for energy. energy-intensive goods. 4 General clusions about the cost and impact of From this Brookes concludes that statements about the energy founda- trying to constrain energy demand. increasing energy efficiency has and Furthermore, such rebound as does tion of future economic growth, largewill simply increase the attractiveness ly based on pre-1973 data, are there- occur tends to reflect direct social and/or economic benefits, because it of energy for new uses: if it is more fore questionable. efficient, it is cheaper, so people do A much more serious flaw is to comes from uses where energy price is more with it. Consequently energy confuse the role of naturally-occurring a serious constraint on activity - for demand cannot be reduced by impro- efficiency improvements with the example, increased comfort when very ving efficiency when it is not a con- effects of deliberate attempts to mini- draughty buildings are insulated. But straint on activity. It would be harder mize energy consumption when price in many areas there is little such reto find a sharper divide between this and availablility are not constraints. In bound because energy price is a reand the claim made by others for the all the period to 1973, improved ener- latively minor factor in determining scale of savings available from impro- gy efficiency was primarily a consequ- activity levels. This is not inconsistent with the ving energy efficiency. ence of other pursuits - the drive to However, while the historical analy- colonize new markets, by producing observation that energy price is an sis is fine the final step in this argu- automated processes in which the important determinant of demand, be-
Energy efficiency and economic fallacies
ENERGY POLICY October 1990
Communication cause there is a critical distinction between activity responses and technical responses. Price increases may stimulate the takeup of more efficient technology without affecting the activity level, particularly in the long term as technologies respond. Promoting the more efficient technologies directly will similarly lead to genuine demand reductions. When the price response is based primarily on technology improvement, the implicit energy price fall from promoting more efficient technology is not equivalent to a real fall in energy prices. And there is a second asymmetry between technological and activity responses: technical improvements may stagnate, but they tend not to reverse when the price falls. Thus whilst accusing efficiency advocates of a 'fallacy of composition', Brookes himself is making a 'fallacy of aggregation' by neglecting large swathes of demand areas where improving efficiency can very effectively save energy, and by neglecting the critical distinctions between the activity and technological determinants of energy demand. There are other components to the rebound effect. Brookes cites a respending effect: 'purchasing power released by lower expenditure on existing uses finds an outlet somewhere . . . in the purchase of goods and services that require energy . . /.5 Total energy costs are generally a few per cent of G D P and this re-spending effect will, even if one assumes a linear e n e r g y - G D P relationship, be of this order. Another component is the supply rebound: if demand is lowered, the pressure on energy supply will be lowered and prices may fall, stimulating more general demand. However, this can never reverse the sign of savings and, since supply curves are often much flatter than demand curves (and for some supply industries may be negative) this effect too is usually negligible. In fact, many of these points have been made in the reclusive publication of the I A E E which Brookes himself cites most often, the Energy Journal. A n article by Khazzoom, 6 which advanced arguments related to those of Brookes and Greenhalgh, was heavily
criticized by Lovins 7 and by Henly et al. s A reply by Khazzoom 9 strongly criticized Lovins's economic theorizing and consistency, but did not rebut many of his basic arguments, and it ignored the more careful and equally devastating critique by Henly et al. Khazzoom did not, in this author's view, seriously dispute the evidence that the rebound effect is usually small, reducing the 'engineering estimate' of savings by perhaps 5-15% overall in most cases. Insofar as larger rebounds occur, this often reflects substitution rather than absolute increases (as Greenhalgh emphasizes with respect to electricity displacing other fuels): the overall energy impact is then ambiguous, and indeed such rebound may amplify the gross savings. Brookes himself seems to acknowledge some of these points indirectly, when he accuses the efficiency enthusiasts of 'wanting to jog A d a m Smith's elbow to introduce damaging reductions in energy intensity of production - since it seems that these are likely to be the only ones associated with reductions in total energy consumption'. A t the risk of repetition it therefore seems to necessary to emphasize again that the case for extensive intervention rests squarely on the evidence that A d a m Smith's elbow is suffering from crippling arthritis in this area, and jogging it to move his hand is likely to be highly beneficial. For example, a U K Parliamentary enquiry took evidence from a wide range of industry and environmental groups and concluded: 10 The most striking feature of our enquiry has been the extent to which improvements in energy efficiency - across all sectors of the economy - are almost universally seen as the most obvious and most effective response to the greenhouse e f f e c t . . , the evidence received overwhelmingly endorses the view that, for a variety of reasons, serious market imperfections pers i s t . . , widespread opportunities to invest profitably in cost-effective measures . . . are being ignored. These failures result in an inefficent use of resources and higher than needed costs to society, the consumer, and the environment. It is striking that even the analysis performed by the US Department of Energy, connected
with their attempt to make appliance efficiency standards illegal in the U S A (by setting a Federal 'zero standards standard'), nevertheless indicated that the overall macroeconomic effects of efficiency standards would be beneficial. 11 As many economists seem to have difficulty in understanding how such a state of affairs can exist, it is worth reciting at least seven reasons for it: 12
Lack of knowledge, knowhow and technical skills. Private households, car drivers, small and medium-sized companies and small public administrations generally know little about the possibilities for energy saving, or lack the skills to implement them. Architects, consulting engineers and installers, similarly often have little knowledge of or interest in energy conservation opportunities.
Separation of expenditure and benefit. The person or group owning a building or energy-consuming equipment may not be the user. Hence the owner may not receive the benefits of conservation, and/ or the user may not be allowed to make capital investments to improve the building or equipment.
Limited capital, often arising from external restrictions on capital budgets. Poor householders often have more pressing uses for their money than making investments to reduce running costs. In many small businesses, the capital budget is often quite constrained and devoted to business-related investments such as computerization. Also, non-labour running costs are often treated as unavoidable peripherals. There is thus little opportunity or incentive to invest in energy conservation to reduce overheads.
Even when those making the investment have the relevant information and expertise and receive the full benefits, they still tend to demand a much higher rate of return than that used for supply investments. 13
Impact of electricity and gas tariff
ENERGY POLICY October 1990
D o m e s t i c electricity a n d gas tariffs are usually designed in two p a r t s - a fixed charge, reflecting capital repaym e n t a n d fixed r u n n i n g costs, a n d a v a r i a b l e part reflecting v a r i a b l e o p e r a t i n g costs. M o s t e n e r g y cons e r v a t i o n m e a s u r e s will save the user only the v a r i a b l e p a r t of the tariff, o f t e n well u n d e r half t h e total charge. Yet m o s t c o n s e r v a tion m e a s u r e s do r e d u c e utility fixed c h a r g e s in t h e long r u n since they o f t e n r e d u c e the p e a k loads. 14
Lack o f interest in peripheral operating costs. E n e r g y costs are
t h e m , a n d c a n n o t be b o t h e r e d to take steps to save m o n e y in this area. Cost-effectiveness is not the only criterion; 'cost r e l e v a n c e ' is also r e q u i r e d . T h e a t t i t u d e m a y b e crudely s u m m a r i z e d as o n e of 'don't know and don't care'. This of course c o n t r a s t s sharply with e n e r g y industries.
Legal and administrative obstacles. Finally, t h e r e can b e a n u m b e r of legal a n d a d m i n i s t r a t i v e obstacles in t h e p a t h of e n d - u s e efficiency, including discriminatory tax regimes, r e n t controls, capital b u d g e t restrictions, etc.
often a small p a r t of p e r s o n a l or business e x p e n d i t u r e s . People m a y simply not b e i n t e r e s t e d in
T h e result of all t h e s e factors is to create a pervasive imbalance between i n v e s t m e n t s in supply a n d i n v e s t m e n t s
This communication is an edited extract from M.J. Grubb, ed, Energy Policies and the Greenhouse Effect, Gower, Aldershot, UK, for the Energy and Environmental Programme, Royal Institute of International Affairs, London, UK, forthcoming 1990.
the era of materials-intensive production and the beginning of a new era in which economic growth is dominated by hightechnology products having low materials content'. This should not be confused with the shift towards services on which Brookes comments. 5 Brookes, op cit, Ref 1. 6 J.D. Khazzoom, 'Energy savings resulting from the adoption of more efficient appliances', The Energy Journal, Vol 8, No 4, October, 1987, pp, 85-89. 7 A.B. Lovins, 'Energy savings . . . another view', The Energy Journal, Vol 9, No 2, April 1988, pp 155-162. 8 j . Henly, H. Ruderman and M.D. Levine, 'Energy s a v i n g s . . . A follow-up', The Energy Journal, Vol 9, No 2, April 1988, pp 163-170. 9 J.D. Khazzoom, 'Energy savings . . . a rejoinder', The Energy Journal, Vol 10, No 1, January 1989, pp 157-165. ~0 House of Commons Energy Committee,
L. Brookes, 'The greenhouse effect: the fallacies in the energy efficiency solution', Energy Policy, Vol 18, No 2, March 1990, ~p 199--201. G. Greenhalgh, 'Energy conservation policies', Energy Policy, Vol 18, No 3, April 1990, pp 293-299. 3 Home heating, and the domestic sector more generally, is an important potential case of saturation. Brookes, op cit, Ref 1, claims that 'unpublished research . . . shows that domestic consumers tend to spend a constant proportion of income on fuel and electrical energy'. The UK Family Expenditure survey shows that average household expenditure on direct energy is actually greater in the poorest homes than in middle incomes homes even in absolute terms, let alone relative to income, presumably because poorer people live in worse accommodation and are less likely to invest in insulation and more efficient equipment. The statistics make it hard to separate the energy component of other expenditures such as transport, but it is still hard to see how any of the FES data squares with Brookes's statement. 4 R.H. Williams and E.D. Larson, 'Materials, affluence, and industrial energy use', Annual Review of Energy, No 12, pp 99144, 1987. The authors amass substantial data to suggest that value added in goods occurs increasingly from fabrication and finishing, rather than from increasing bulk, and that 'while many believe this to be a temporary phenomenon . . . our analysis indicates this trend marks the passing of
ENERGY POLICY October 1990
Energy Policy Implications of the Greenhouse Effect, Vol 1, July 1989, HMSO, London, UK. 11 According to P. Rollin and Jan Beyea, 'US appliance efficiency standards', Energy Policy, Vol 13, No 5, October 1985, pp 425--436 the results suggested that with the standards examined, 'GNP would increase by billions of dollars, the net foreign trade balance would increase by a billion dollars, and around 100 000 new jobs would be generated'. The US Court of Appeals later ruled that the DoE moves to ban standards 'openly violated Congressional intent', and declared the zero-standards standard as illegal. 12 The list given here is synthesized from a
in e n d - u s e efficiency. Policies a i m e d at r e m o v i n g or c i r c u m v e n t i n g these market obstacles, a n d installing efficient, cost-effective technologies, will save energy and bring both environmental a n d e c o n o m i c benefits. This is n o t wishful t h i n k i n g b u t the almost universal conclusion from those w h o h a v e studied t h e realities of the e c o n o m i c i m b a l a n c e b e t w e e n supply a n d dem a n d . Surely it is time t h a t the d e b a t e m o v e d on to c o n s i d e r i n g the relative merits a n d impacts of different policy o p t i o n s for trying to achieve t h e s e savings?
M. J. Grubb Energy and Environmental Programme London, UK number of other discussions, notably: lEA,
Energy conservation in lEA countries, OECD, Paris, 1987; various publications from the American Council for an Energy Efficient Economy, Washington, DC; and an overview by E. Jochem and M. Brand, 'Strategic policies for improving energy efficiency and reducing greenhouse gas emissions', Fraunhogfer-Institute fur Systemtechnik, Karlsruhe, FR Germany. 13 One study of the choices made by wellinformed consumers in buying fridges which differed only in cost and efficiency concluded that only 2/5 of them bought the more efficient variety if it took more than 3 years to pay back (35% discount rate). Another 2/5 seemed to apply a discount rate above 60% (ie requiring payback in under 2 years). This contrasts with producer criteria on the order of 10% return, or 10 year payback; see A.K. Meier and J. Whittier, 'Consumer discount rates implied by consumer purchases of energy efficient refrigerators', Energy, The International Journal, Vol 8, No 12, pp 957-962, 1983. 14 Part of the fixed charge reflects connection costs, which conservation would usually not affect. The importance of the capacity element will vary greatly according to system and load conditions. The fundamental reason why it is so difficult for conservation to receive its 'capacity credit' is the impracticability of dynamic marginalcost pricing, in which very high rates at times of peak load on the system would raise the revenues for capital construction. Separation of fixed charges is really a surrogate for this. Declining block tariffs for industrial consumers can have similar implications.
The editor welcomes debate on articles published in Energy Policy. Letters and communications should be sent to the Editor at the address on the inside front cover.