The Paris Agreement targets rely on energy efficiency measures to reduce energy demand and so deliver the bulk of emissions reductions. The effects of the global pandemic have thrown these expectations into disarray, with the deployment of efficiency measures faltering across the world.
Still worse, academic analysts are now concluding that W. S. Jevons was right and that even if deployed successfully efficiency improvements will not deliver as much energy conservation as is required, and may actually increase demand.
Global emissions reduction policies aimed at delivering the Parish Agreement rely on the timely arrival of a very large number of highly improbable things, for example, lower-cost renewable energy, cheaper electricity storage, globally synchronised carbon pricing, a viable “hydrogen economy”, carbon capture and sequestration that works, and practical and affordable electric vehicles, to name a few of the better known. Behind all these, though much less salient to the general public, is the central enabling assumption that gives the aura of reality to the ambitions in all other areas. Namely that improvements in the energy efficiency of end-use conversion processes, such as buildings and industry, will greatly reduce energy demand, bringing it down to manageable proportions. I last wrote here about this assumption in August 2018 (Energy Efficiency, Smart Meters and Climate Policy), illustrating the point with a chart from the International Energy Agency (IEA) that summarised the aggregated ambition of the “stated policies” of the various nations.
The chart, slightly revised, was reissued in December last year when the IEA published its annual study of progress on the implementation of energy efficiency measures in the global economy, Energy Efficiency 2020.
Figure 1: International Energy Agency projection of Energy Conservation to Deliver Climate Goals. Source: IEA.
The fundamental assumptions in the figure remain the same, with a reduction in energy consumption delivering over 40% of the emissions reduction required by 2040, a larger share even than renewables.
However, the prospects of this historically unprecedented reduction in demand materialising were already poor, but are evaporating due to the effects of the global pandemic. The IEA itself comments: “Energy efficiency’s weakest progress in a decade threatens international climate goals.” Grave and realistic though that may seem, the chart illustrating this point reveals it as an understatement.
Figure 2. Primary energy intensity improvement rate, 2015–2020. Source: IEA.
It is obviously true that Coronavirus has not helped, but it is equally obvious that the strong downward trend in the rate of energy intensity improvement predates the pandemic. Indeed, even in 2015, the improvement rate was barely adequate to deliver the 3% per year that would, in the IEA’s view, reduce demand to levels where decarbonisation policies realisable in practice and tolerable in their cost. Energy efficiency uptake wasn’t working to plan; coronavirus is simply the coup de grâce.
But could it ever have worked, or was the whole endeavour mistaken? Was the IEA guilty of what W. S. Jevons notoriously referred to in 1865 as a naïve “confusion of ideas” that misleads one into supposing that “the economical use of fuel is equivalent to a diminished consumption” when, in fact, “The very contrary is the truth.”? If you had not examined Jevons for yourself and had only read or heard about his criticism, you might be inclined to give the IEA the benefit of the doubt and imagine that Jevons was speculating and that he overestimated the actual strength of “rebound” effects, whereby an improved process is used more because it is cheaper. But those who have engaged directly with Jevon’s steel-trap logic will know firstly that he is not principally referring to parochial phenomena, he is much more concerned with the entire economic system, and secondly that his “paradox” is not an empty speculation awaiting empirical testing, but itself an empirical historical observation.
What Jevons claims is nothing less than that “the whole of our present vast industrial system […] has chiefly arisen from successive measures of economy” of fuel. Jevons was not speculating, but making an historical observation, namely that improvements in energy efficiency are seen to be the cause of economic growth and thus of increased energy consumption:
No one must suppose that [energy] saved is spared – it is only saved from one use to be employed in others, and the profits gained soon lead to extended employment in many new forms. The several branches of industry are closely interdependent, and the progress of any one leads to the progress of nearly all.
Academics and others have been slow to follow Jevons’ lead, which is to a degree understandable, the remarks in the Coal Question being telegrammatic in their brevity. But in the last few months, several researchers well-known in the field, Paul Brockway of the University of Leeds, and Steve Sorrell, at the University of Sussex, Gregor Semieniuk at the University of Massachusetts, Matthew Kuperus Heun of Calvin University, and Victor Court at the Institut Louis Bachelier have published a substantial joint paper, “Energy efficiency and economy-wide rebound effects: a review of the evidence and its implications”, in Renewable and Sustainable Energy Reviews that this is changing. The abstract of the paper deserves full quotation:
The majority of global energy scenarios anticipate a structural break in the relationship between energy consumption and gross domestic product (GDP), with several scenarios projecting absolute decoupling, where energy use falls while GDP continues to grow.
However, there are few precedents for absolute decoupling, and current global trends are in the opposite direction. This paper explores one possible explanation for the historical close relationship between energy consumption and GDP, namely that the economy-wide rebound effects from improved energy efficiency are larger than is commonly assumed.
We review the evidence on the size of economy-wide rebound effects and explore whether and how such effects are taken into account within the models used to produce global energy scenarios. We find the evidence base to be growing in size and quality, but remarkably diverse in terms of the methodologies employed, assumptions used, and rebound mechanisms included.
Despite this diversity, the results are broadly consistent and suggest that economy-wide rebound effects may erode more than half of the expected energy savings from improved energy efficiency.
We also find that many of the mechanisms driving rebound effects are overlooked by integrated assessment and global energy models. We therefore conclude that global energy scenarios may underestimate the future rate of growth of global energy demand.
These conclusions are derived from several substantial bodies of information, and the analysis itself is as laboriously technical and explicit in detail as that of Jevons is condensed and elliptical. But the convergence of views is unmistakable.
In illustration of the first point made by the authors, that the global trends show no evidence of decoupling energy consumption and growth, we can refer to the paper’s figure 1:
Figure 3: Historical trends and future scenarios for global final energy use and GDP (1971–2050). Notes: Scenario plots are in four groups: orange (IEA models); green (1.5 ◦C IAMs); purple (2.0 ◦C IAMs) and blue (other models). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.). Source: Brockway et al. 2021.
This is an intricate chart, and readers are referred to the original for confirmation, but the main contrast is sufficiently clear even at a glance. All the modelled policy scenarios represented by the various coloured curves are strongly divergent from the trend of the empirically observed relationship between GDP and Final Energy Demand (it is, incidentally, a merit of this study that it concentrates on Final Energy Demand and not Total Primary Energy Supply). The authors quite properly raise the question as to whether the “radical departure from the historical trend” proposed by the policy models is plausible. To provide an interim answer the remainder of the paper surveys twelve technical models in published studies estimating economy-wide rebound effects:
However, the most notable finding from this review of other methods is that the studies consistently estimate large economy-wide rebound effects. Specifically, 10 of the 12 studies in Table 3 provide baseline estimates of ~50% or more, and three estimate >100% rebound.
As a crude indicator, the mean estimate of economy-wide rebound effects from the 12 studies is 71% – _with a mean of 62%from the macroeconomic models, 104% from the econometric studies, and 46% from the growth accounting studies.
As the paper observes, this degree of consistency is especially striking since such divergent methods are used in each of the studies, and thus lead the authors to conclude that “the results broadly reinforce the conclusion […] that economy-wide rebound effects may erode more than half of the energy savings from improved energy efficiency.”
But with the logic of Jevons in mind one may suspect that the economy-wide effects of energy efficiency improvements will as a rule always tend to increase overall energy consumption, which indeed some of these studies suggest. The resulting tight relationship between energy consumption and GDP thus becomes a deeply intractable problem for climate policy, and even the provisional conclusions offered by the Brockway et al. paper have extremely significant implications for the Paris Agreement policy agenda as summarised in the IEA figure cited above. If efficiency does not deliver as much conservation as is assumed in the global policy model, then the emissions savings will have to be supplied by other means if GDP and GDP growth is to be maintained. If that means more renewable energy, cost control will become a still greater concern than it is today.
Dr John Constable: GWPF Energy Editor
 W. S. Jevons, The Coal Question (1865), p.103.
 Coal Question, p. 105.
 Coal Question. p. 115.
 Paul E. Brockway, Steve Sorrell, Gregor Semieniuk, Matthew Kuperus Heun, Victor Court, “Energy efficiency and economy-wide rebound effects: a review of the evidence and its implications”, Renewable and Sustainable Energy Reviews (In press). https://www.sciencedirect.com/science/article/pii/S1364032121000769?via%3Dihub.