This group brings together the best thinkers on energy and climate. Join us for smart, insightful posts and conversations about where the energy industry is and where it is going.

Post

Is Energy Efficiency a Good Thing Even with Rebound?

Gernot Wagner's picture
Environmental Defense Fund
  • Member since 2018
  • 83 items added with 20,363 views
  • Oct 28, 2014 8:00 pm GMT
  • 507 views

Your access to Member Features is limited.

By Inês Azevedo, Kenneth GillinghamDavid Rapson, and Gernot Wagner.

ceiling-163866_640Lighting is critical to our livelihoods. Humans have used lighting technology since long before industrialization. For many centuries, this lighting was extremely inefficient, with over 95% of the energy consumed wasted as heat. Recently, the Nobel Prize in Physics was awarded to Isamu Akasaki, Hiroshi Amano and Shuji Nakamura for their remarkable contributions towards highly efficient light emitting diode (LED) technology. A day later, Michael Shellenberger and Ted Nordhaus reignited a long standing debate with an Op-Ed inThe New York Times claiming that these developments are not likely to save energy and instead may backfire. (TheTimes has since corrected a crucial point of the article, and it has published three letters to the editor, including one by a subset of co-authors here.)

As evidence for these claims, Shellenberger and Nordhaus cite research that observes the vast improvements in the efficiency of lighting over the past two centuries having resulted in “more and more of the planet [being] dotted with clusters of lights.” They take this as evidence of how newer and ever more efficient lighting technologies have led to demand increases and, thus, have “led to more overall energy consumption.” Further, they refer to “recent estimates and case studies” that suggest “energy-saving technologies may backfire, meaning that increased energy consumption associated with lower energy costs because of higher efficiency may in fact result in higher energy consumption than there would have been without those technologies.”

First off, yes, it is likely that many efficiency improvements are associated with some rebound effect. It’s been with us forever, and it’s been known for over a century. More efficient lighting leads to people using more light. Key here is “leads to.” Causality matters. More on that in a minute.

For now, a quick look at the actual technology in question. It turns out the technology developments for LED lighting are, in fact, much greater than previous advances in lighting. Figure 1 [see the pdf] shows the dramatic pace of technology change in LED efficacy. The Nobel Prize was well-deserved: LEDs provide a major energy-saving innovation.

But what about the claim that this efficiency improvement will only lead to more energy use? This claim is simply not justified. Noting that lighting dots the globe at night today when it did not in the 19th century may be confounding correlation with causation. The world is also much wealthier today and the service provided by light from electricity is very different than candlelight. Perhaps earlier lighting would have dotted the globe at night in 1850 too had we been as wealthy as today and had consistent lighting. We cannot say without looking at the evidence.

The evidence we have is quite clear. Shellenberger and Nordhaus say “The I.E.A. and I.P.C.C. estimate that the rebound could be over 50 percent globally,” and they then proceed to talk about “backfire,” a rebound effect of over 100 percent. That’s quite a jump from 50 to 100. What’s missing here is that most studies, including the IEA’s and their own(!), take 60% as an upper bound. The IPCC summarizes the evidence as thus:

“A comprehensive review of 500 studies suggests that direct rebounds are likely to be over 10% and could be considerably higher (i.e., 10% less savings than the projected saving from engineering principles). Other reviews have shown larger ranges with Thomas and Azevedo (Thomas and Azevedo, 2013) suggesting between 0 and 60%. For household‐efficiency measures, the majority of studies show rebounds in developed countries in the region of 20-45% (the sum of direct and indirect rebound effects), meaning that efficiency measures achieve 65-80% of their original purposes.”

We have each performed our own detailed surveys of the literature (Azevedo 2014Thomas & Azevedo, 2013Gillingham et al. 2013;Gillingham et al. 2014) and largely agree with these statements from the I.P.C.C. The bottom-line: the evidence for a “backfire” is weak. The rebound effect is clearly there, but first it’s generally relatively small—especially in developed countries. Perhaps most importantly, where it does exist—and it does—it’s good.

Energy inefficiency can’t be good. That doesn’t yet mean that efficiency alone is sufficient. Every economist worth his or her degree would conclude that we need a price on carbon or a similar instrument. Bonus fact: there’s no direct rebound effect with pricing mechanisms.

As the Nobel Committee notes in its press release: “The LED lamp holds great promise for increasing the quality of life for over 1.5 billion people around the world who lack access to electricity grids.” In short, and as two of us say in a shorter letter to the editor, LEDs alone clearly won’t solve global warming, nor will they solve global poverty. But they are a step in the right direction for both. Thank you, Isamu Akasaki, Hiroshi Amano, and Shuji Nakamura, and to the Nobel Committee for recognizing their work.

Published in full as part of a broader post on “Is There Room for Agreement on the Merits and Limits of Efficient Lighting” by Andrew Revkin on the DotEarth blog of The New York Times. For a shorter take, see our letter to the editor of The New York Times. For a longer take, see “The Rebound Effect and Energy Efficiency Policy.”

Gernot Wagner's picture
Thank Gernot for the Post!
Energy Central contributors share their experience and insights for the benefit of other Members (like you). Please show them your appreciation by leaving a comment, 'liking' this post, or following this Member.
More posts from this member
Discussions
Spell checking: Press the CTRL or COMMAND key then click on the underlined misspelled word.
Keith Pickering's picture
Keith Pickering on Oct 28, 2014

Hi,

You write:

Perhaps earlier lighting would have dotted the globe at night in 1850 too had we been as wealthy as today and had consistent lighting.

Which is precisely the point. The principal effect of lighting efficiency is to reduce the cost of lighting. In other words, one of the reasons we are wealthier today is because of increased energy efficiency and the economic wealth it provides. So, saying that we would have had more lighting in 1850 if we had been wealthier in 1850 is equivalent to saying we would have had more lighting in 1850 if we had been more energy efficient then. You’re saying exactly what Nordhaus and Schellenberger are saying, without realizing it.

“We have each performed our own detailed surveys of the literature … “

It’s more than a little disappointing, then, that none of your detailed surveys (nor IPCC, nor IEA, for that matter) has cited the most important paper on the rebound effect since Jevons. I refer, of course, to Garrett 2011:

Garrett, T. J. (2011). Are there basic physical constraints on future anthropogenic emissions of carbon dioxide?Climatic change104(3-4), 437-455.

Garrett considers civilization as a whole to be a thermodynamic engine, which can be modeled with thermodynanamic equations. Civilization grows and maintains itself with a feedback loop (the economy) that can also be modelled in thermodynamic terms through (a) a variable that defines the growth rate; (b) a variable that defines the total amount of created wealth since the beginning of time; and (c) a constant that defines the amount of energy needed to maintain created wealth without degradation. 

The key point is found in equation 2, not fully reproducable here, but part of it is: 

αw = αεa ≡ ηa

… where w is work done by the system (civilization), ε is energy efficiency, α is a system-wide availability constant, η is the growth rate, and a is the rate of energy consumption.

Garrett comments, “Note that, perhaps counter-intuitively, higher energy efficiency ε corresponds to higher values of η, and therefore more rapidly exponential evolution of energy consumption a and heat production a − w”.

In other words, the total rebound, including secondary rebounds, cannot be less than 100%. The key point is that the monetary gains from efficiency, even if not spent in that same activity, will be spent somewhere in the economy that will require more energy for the economy as a whole. If the primary rebound doesn’t kill you, the secondary rebounds will.

Energy efficiency does have economic benefits, and should be pursued. But economics is no match for physics. Those who push energy efficiency as a means to solve the climate crisis are taking us in the wrong direction, and relying on a pipe dream. Energy efficiency will not, and cannot, solve even the smallest part of the climate crisis. There is only one viable solution: we must decarbonize our economy, and rapidly.

 

 

douglas card's picture
douglas card on Oct 28, 2014

Keith,

Well thought out and presented, but I disagree.  Early adapters all want 1 or 2 things.  Lower cost and more evironmentally helpful.  Auto’s for example: No one who has pruchased a Prius has even considered it as exceptablt to drive twice as many miles simple becuase it is twice as efficient.  More miles – maybe. But never more than 50% more.  Those of us who converted our lights:  I leave night lights on 24 hours instead of 12.  BUT, they use 1/4 watt instead of 7 watts.  why would you leave lights on longer becuase they are more efficient?  A fridge that is twice as efficient?  Would anyone leave the door open for hours because its not as bad as before?  CEO’s don’t leave the lights on at night because they are CFL’s or LED’s – they put the savings in their pockets.  

Your logic fails when faced with real world actions, IMHO

Keith Pickering's picture
Keith Pickering on Oct 29, 2014

Douglas,

You’re missing the point. Like the economists, you’re looking so closely at the tree that you’re missing the forest. 

So you buy a Prius and save on gas. What do you do with the money you save? If you don’t buy more gasoline with it, you buy more something with it, and that something takes energy to make and to use. And the energy for making and using that something would not have been used, were it not for your savings on gasoline. The same is true of lights, refrigerators, and everything else.

 

 

Keith Pickering's picture
Keith Pickering on Oct 29, 2014

But what is productivity growth, in a thermodynamic sense, other than increases energy efficiency? It’s pretty hard to claim that Garrett “ignores” that, when it’s right there in his equations.

Unless, of course, you have an axe to grind and are reduced to slinging around ad-hominems because you don’t have any actual evidence to support your arguments.

 

Jeffrey Miller's picture
Jeffrey Miller on Oct 29, 2014

Keith,

“If you don’t buy more gasoline with it, you buy more something with it, and that something takes energy to make and to use.”

This sentence could be elaborated to prove that the rebound effect is greater than zero. It doesn’t imply that the rebound effect is 100%. To prove the second claim, you’d have to prove that the “something” that you spend your money on in lieu of burning gasoline is on average as energy intensive as burning gasoline. But the average unit of economic activity is much less energy intensive than burning gasoline, so this seems implausible. 

I am persuaded by the empirical evidence that people have presented on this site and elsewhere that the rebound effect is significant in wealthy countries and very large in the poorer countries. Developing countries contain most of the world’s population and will account for most of the world’s energy use in the future, so we should expect a strong worldwide rebound effect in the global economy in the years ahead. This means that relying on energy efficiency to save us from a climate disaster, as much of self described ‘environmental community’  (to which, to my dismay, I once belonged) wants desperately to believe, is delusional. We need to supply the world much more energy, not less, and we need it quickly – not in 2100, but in the next few decades. For the sake of our descendants and the rest of the current biosphere,  we had better be sure of two additional things. One is that it is carbon free and the other is that it is affordable. If you impose these four strong constraints (massively scalable, can be done on a decadal scale, affordable, and carbon free), we have, as far as I can tell, narrowed our primary options to one.  Needless to say, this is the source of energy that the ‘environmentalists‘ spend a lot of resources fighting.

Keith Pickering's picture
Keith Pickering on Oct 29, 2014

“This sentence could be elaborated to prove that the rebound effect is greater than zero. It doesn’t imply that the rebound effect is 100%. To prove the second claim, you’d have to prove that the “something” that you spend your money on in lieu of burning gasoline is on average as energy intensive as burning gasoline.”

Exactly right, Jeffery, and Garrett does indeed show that in his paper. It turns out, from empirical evidence, that the constant dollar cost to maintain the total wealth of civilization is itself a constant through time, amounting to 10 mW per 1990 $US. Therefore increasing the wealth of civilization through economic growth adds to the total energy required to sustain civilization, and that total energy load is independent of energy efficiency.

“But the average unit of economic activity is much less energy intensive than burning gasoline, so this seems implausible.”

That’s not really true, if you think about it. Let’s say you spend that extra cash going to the movies more often. So (a) you DO drive more, to get to the movies more; (b) the movie theaters feel greater demand, which means they cook more popcorn, and maybe they add an additional matinee; (c) the movie studios see their profits rise, so they make and distribute more movies; (d) the studio employs their actors and employees more, who see more in their paychecks, so THEY go out and spend more too. The economic effects ripple and rebound throughout the entire economy. And pretty much every one of those activities uses electricity, which is roughly as energy intense as driving.

And speaking of empirical evidence: US cars keep getting more and more energy efficient. And yet we keep using more and more gasoline. Why? It’s because we have more cars now. It’s because of economic growth. And that economic growth is spurred on by greater efficiency.

Keith Pickering's picture
Keith Pickering on Oct 29, 2014

Job,

Here’s global primary energy use since 1965, a period that includes vast innovation in things like computers, knowlege, transistors, solar power, and allegedly never-bottoming learning curves. Perhaps you can point out for us, on the graph, where all those innovations have had their effect in reducing global energy demand.

Global energy use, 1965-2012

Or maybe you can’t.

Jeffrey Miller's picture
Jeffrey Miller on Oct 29, 2014

Regardless of the exact value of the rebound effect, which I agree is not easy to measure and depends on many variables like per capita gdp, I think we both agree on the big picture: energy efficiency is not going to solve our carbon problem. Only plentiful, reasonably priced, carbon free energy can do that. 

First off, I should say that I think it’s great that Garrett is looking at these questions and throwing out some interesting observations. It’s always good to have different perspectives on hard questions. I also appreciate your bringing the paper to the attention of people on this site. It was interesting to read. That said, I find his model unconvincing. 

For example, do you think that Garrett’s relation between energy and what he calls “wealth” can be true in general? What he’s saying is that global energy use next year will equal the energy use this year plus some constant times next year’s inflation adjusted global gdp. This means that, according to Garrett, energy use can never decline since gdp is always positive. Even if the population of the world plummets, for example because of a serious global pandemic that kills off a significant fraction of the population, Garrett says we are bound to use more energy every single year. Doesn’t this strike you as extremely unlikely to be true? Don’t you think that if GDP falls a lot for whatever reason, say a world wide depression, energy use will also fall and not increase, contrary to Garrett’s prediction? Doesn’t it also strike you as extremely unnatural that our energy use this year depends just as much on a dollar of GDP produced during the Roman Empire as it depends on a dollar produced last year? Or, if technology stagnates and we manage to stabilize world population so that GDP stops growing, do you expect, in that scenario, that annual energy use  – in this stagnant economy – use would grow linearly with time, as Garrett predicts? While I am naturally sympathetic to the physicists, I have to go with the economists on this one.  

Regarding gasoline in the US use, I think it has been roughly flat for over a decade, despite a 10% increase in population. I don’t think this is a point for or against Garrett though, since he is focused on global energy use. 

 

Bob Meinetz's picture
Bob Meinetz on Oct 30, 2014

Keith, regarding

And pretty much every one of those activities uses electricity, which is roughly as energy intense as driving.

That’s completely dependent on the activity.

Perhaps Garret is right or wrong about the proportion of the rebound effect, but in relation to global warming it’s irrelevant – unless we assume decarbonizing is impossible. In any case, Garret doesn’t have enough variables – by many orders of magnitude – to make a purely physical argument out of how society will respond to the unprecedented circumstances it will face in the next century.

It’s an appealing but simplistic theory based on faulty assumptions.


Get Published - Build a Following

The Energy Central Power Industry Network is based on one core idea - power industry professionals helping each other and advancing the industry by sharing and learning from each other.

If you have an experience or insight to share or have learned something from a conference or seminar, your peers and colleagues on Energy Central want to hear about it. It's also easy to share a link to an article you've liked or an industry resource that you think would be helpful.

                 Learn more about posting on Energy Central »