This article is the part 2 from Chapter 4 of Richard Heinberg’s new book The End of Growth, which is set for publication by New Society Publishers in August 2011. This chapter explores the possibilities of innovation, substitution and efficiency to maintain economic growth. 

Get the book now – Amazon, New Society, Kindle, Nook Reader

Chapter 4, Part 1
Chapter 4, Part 3
Chapter 4, Part 4
Chapter 4, Part 5
Chapter 4, Part 6

Energy Efficiency to the Rescue
The historic correlation between economic growth and increased energy consumption is controversial, and I promised in Chapter 3 to return to the question of whether and to what degree it is possible to de-link or decouple the two.
While it is undisputed that, during the past two centuries, both energy use and GDP have grown dramatically, some analysts argue that the causative correlation between energy consumption and growth is not tight, and that energy consumption and economic growth can be decoupled by increasing the efficiency with which energy is used. That is, economic growth can be achieved while using less energy.[1]
This has already happened, at least to some degree. According to the U.S. Energy Information Administration,
“From the early 1950s to the early 1970s, U.S. total primary energy consumption and real GDP increased at nearly the same annual rate. During that period, real oil prices remained virtually flat. In contrast, from the mid-1970s to 2008, the relationship between energy consumption and real GDP growth changed, with primary energy consumption growing at less than one-third the previous average rate and real GDP growth continuing to grow at its historical rate. The decoupling of real GDP growth from energy consumption growth led to a decline in energy intensity that averaged 2.8 percent per year from 1973 to 2008.”[2]
Translation: We’re saved! We just need to double down on whatever we’ve been doing since 1973 that led to this decline in the amount of energy it took to produce GDP growth.[3]
However, several analysts have pointed out that the decoupling trend of the past 40 years conceals some explanatory factors that undercut any realistic expectation that energy use and economic growth can diverge much further.[4]
One such factor is the efficiency gained through fuel switching. Not all energy is created equal, and it’s possible to derive economic benefits from changing energy sources while still using the same amount of total energy. Often energy is measured purely by its heating value, and if one considers only this metric then a British Thermal Unit (BTU) of oil is by definition equivalent to a BTU of coal, electricity, or firewood. But for practical economic purposes, every energy source has a unique profile of advantages and disadvantages based on factors like energy density, portability, and cost of production. The relative prices we pay for natural gas, coal, oil, and electricity reflect the differing economic usefulness of these sources: a BTU of coal usually costs less than a BTU of natural gas, which is cheaper than a BTU of oil, which is cheaper than a BTU of electricity.
Electricity is a relatively expensive form of energy because it is so convenient to use (try running your computer directly on coal!). Electricity can be delivered to wall outlets in billions of rooms throughout the world, enabling consumers easily to operate a fantastic array of gadgets, from toasters and blenders to iPad chargers. With electricity, factory owners can run computerized monitoring devices that maximize the efficiency of automated assembly lines. Further, electric motors can be highly efficient at translating energy into work. Compared to other energy sources, electricity gives us more economic bang for each BTU expended (for stationary as opposed to mobile applications).
As a result of technology developments and changes in energy prices, the U.S. and several other industrial nations have altered the ways they have used primary fuels over the past few decades. And, as several studies during this period have confirmed, once the relationship between GDP growth and energy consumption is corrected for energy quality, much of the historic evidence for energy-economy decoupling disappears. [5]

Figure 40, U.S. GDP, Divisia-corrected energy consumption, and non-corrected energy consumption

The Divisia index is a method for aggregating heat equivalents by their relative prices, and Figure 40 charts U.S. GDP, Divisia-corrected energy consumption, and non-corrected energy consumption. According to Cutler Cleveland of the Center for Energy and Environmental Studies at Boston University, “This quality-corrected measure of energy use shows a much stronger connection with GDP [than non-corrected measures]. This visual observation is corroborated by econometric analysis that confirms a strong connection between energy use and GDP when energy quality is accounted for.”[6]
Cleveland notes that, “Declines in the energy/GDP ratio are associated with the general shift from coal to oil, gas, and primary electricity.” This holds true for many countries Cleveland and colleagues have examined.[7] His conclusion is highly relevant to our discussion of Peak Oil and energy substitution: “The manner in which these improvements have been . . . achieved should give pause for thought. If decoupling is largely illusory, any rise in the cost of producing high quality energy vectors could have important economic impacts. . . . If the substitution process cannot continue, further reductions in the E/GDP ratio would slow.”[8]
Another often-ignored factor skewing the energy-GDP relationship is outsourcing of production. In the 1950s, the U.S. was an industrial powerhouse, exporting manufactured products to the rest of the world. By the 1970s, Japan was becoming the world’s leading manufacturer of a wide array of electronic consumer goods, and in the 1990s China became the source for an even broader basket of products, ranging from building materials to children’s toys. By 2005, the U.S. was importing a substantial proportion of its non-food consumer goods from China, running an average trade deficit with that country of close to $17 billion per month.[9] In effect, China was burning its coal to make America’s consumer goods. The U.S. derived domestic GDP growth from this commerce as WalMart sold mountains of cheap products to eager shoppers, while China expended most of the BTUs. The American economy grew without using as much energy—in America—as it would have if those goods had been manufactured domestically.
There is one more factor that helps explain historic U.S. “decoupling” of GDP growth from growth in energy consumption—the “financialization” of the economy (discussed in Chapter 2). Cutler Cleveland notes, “A dollar’s worth of steel requires 93,000 BTU to produce in the United States; a dollar’s worth of financial services uses 9,500 BTU.”[10] As the U.S. has concentrated less on manufacturing and building infrastructure, and more on lending and investing, GDP has increased with a minimum of energy consumption growth. While the statistics seem to show that we are becoming more energy efficient as a nation, to the degree that this efficiency is based on blowing credit bubbles it doesn’t have much of a future. As we saw in Chapter 2, there are limits to debt.
The actual tightness of the relationship between energy use and GDP is illustrated in the recent research of Charles Hall and David Murphy at the State University of New York at Syracuse, which shows that, since 1970, high oil prices have been strongly correlated with recessions, and low oil prices with economic expansion. Recession tends to hit when oil prices reach an inflation-adjusted range of $80 to $85 a barrel, or when the aggregate cost of oil for the nation equals 5.5 percent of GDP.[11] If America had truly decoupled its energy use from GDP growth, then there wouldn’t be such a strong correlation, and high energy prices would be a matter of little concern.
So far we’ve been considering a certain kind of energy efficiency—energy consumed per unit of GDP. But energy efficiency is more commonly thought of more narrowly as the efficiency by which energy is transformed into work. This kind of energy efficiency can be achieved in innumerable ways and instances throughout society, and it is almost invariably a very good thing. Sometimes, however, unrealistic claims are made for our potential to use energy efficiency to boost economic growth.
Since the 1970s, Amory Lovins of Rocky Mountain Institute has been advocating doing more with less and has demonstrated ingenious and inspiring ways to boost energy efficiency. His 1998 book Factor 4 argued that the U.S. could simultaneously double its total energy efficiency and halve resource use.[12]More recently, he has upped the ante with “Factor 10”—the goal of achieving ten times the productivity from half the materials and energy, or maintaining current productivity while using only 10 percent of the resources.[13]
Lovins has advocated a “negawatt revolution,” arguing that utility customers don’t want kilowatt-hours of electricity; they want energy services—and those services can often be provided in far more efficient ways than is currently done. In 1994, Lovins and his colleagues initiated the “Hypercar” project, with the goal of designing a sleek, carbon fiber-bodied hybrid that would achieve a three- to five-fold improvement in fuel economy while delivering equal or better performance, safety, amenity, and affordability as compared with conventional cars. Some innovations resulting from Hypercar research have made their way to market, though today hybrid-engine cars still make up only a small share of vehicles sold.
While his contributions are laudable, Lovins has come under criticism for certain of his forecasts regarding what efficiency would achieve. Some of those include:
  • Renewables will take huge swaths of the overall energy market (1976);
  • Electricity consumption will fall (1984);
  • Cellulosic ethanol will solve our oil import needs (repeatedly);
  • Efficiency will lower energy consumption (repeatedly).[14]
The reality is that renewables have only nibbled at the overall energy market; electricity consumption has grown; cellulosic ethanol is still in the R&D phase (and faces enormous practical hurdles to becoming a primary energy source, as discussed above); and increased energy efficiency, by itself, does not appear to lower consumption due to the well-studied rebound effect, wherein efficiency tends to make energy cheaper so that people can then afford to use more of it.[15]
Once again: energy efficiency is a worthy goal. When we exchange old incandescent light bulbs with new LED lights that use a fraction of the electricity and last far longer, we save energy and resources—and that’s a good thing. Full stop.
At the same time, it’s important to have a realistic understanding of efficiency’s limits. Boosting energy efficiency requires investment, and investments in energy efficiency eventually reach a point of diminishing returns. Just as there are limits to resources, there are also limits to efficiency. Efficiency can save money and lead to the development of new businesses and industries. But the potential for both savings and economic development is finite.
Let’s explore further the example of energy efficiency in lighting. The transition from incandescent lighting to the use of compact fluorescents is resulting in dramatic efficiency gains. A standard incandescent bulb produces about 15 lumens per Watt, while a compact fluorescent (CFL) can yield 75 l/w—a five-fold increase in efficiency. But how much more improvement is possible? LED lights currently under development should deliver about 150 l/w, twice the current efficiency of CFLs. But the theoretical maximum efficiency for producing white light from electricity is about 300 lumens per Watt, so only another doubling of efficiency is feasible once these new LEDs are in wide use.[16]
Moreover, energy efficiency is likely to look very different in a resource- and growth-constrained economy from how it does in a wealthy, growing, and resource-rich economy.
Permit me to use the example of my personal experience to illustrate the point. Over the past decade, my wife Janet and I have installed photovoltaic solar panels on our suburban house, as well as a solar hot water system. In the warmer months we often use solar cookers and a solar food dryer. We insulated our house as thoroughly as was practical, given the thickness of existing exterior walls, and replaced all our windows. We built a solar greenhouse onto the south side of the house to help collect heat. And we also bought a more fuel-efficient small car, as well as bicycles and an electric scooter.
All of this took years and lots of work and money (several tens of thousands of dollars). Fortunately, during this time we had steady incomes that enabled us to afford these energy-saving measures. But it’s fair to say that we have yet to save nearly enough energy to justify our expenditures from a dollars-and-cents point of view. Do I regret any of it? No. As energy prices rise, we’ll benefit increasingly from having invested in these highly efficient support systems.
But suppose we were just starting the project today. And let’s also assume that, like millions of Americans, we were finding our household income declining now rather than growing. Rather than buying that new fuel-efficient car, we might opt for a 10-year-old Toyota Corolla or Honda Civic. If we could afford the PV and hot water solar systems at all, we would have to settle for scaled-back versions. For the most part, we would economize on energy just by cutting corners and doing without.
The way Janet and I pursued our quest for energy efficiency helped America’s GDP: We put people to work and boosted the profits of several contractors and manufacturing companies. The way people in hard times will pursue energy efficiency will do much less to boost growth—and might actually do the opposite.
What was true for Janet and me is in many ways also true for society as a whole. America could improve its transportation energy efficiency dramatically by investing in a robust electrified rail system connecting every city in the nation.[17] Doing this would cost roughly $600 billion, but it would lead to dramatic reductions in oil consumption, thus lowering the U.S. trade deficit and saving enormous amounts in fuel bills. At the same time, the U.S. could rebuild its food system from the ground up, localizing production and eliminating fossil fuel inputs wherever possible. Doing so would increase the resilience of the system, the health of consumers, and the quality of the environment, while generating millions of employment or business opportunities.[18] The cost of this food system transition is difficult to calculate accurately, but it would no doubt be substantial.
However, the nation would be getting a late start on these efforts. We could fairly easily have afforded to do these things over the past few decades when the economy was growing. But building highways and industrializing and centralizing our food system generated profits for powerful interests, and the vulnerabilities we were creating by relying increasingly on freeways and big agribusiness were only obvious to those who were actually paying attention—a small and easily overlooked demographic category in America these days. As oil becomes more scarce and expensive, more of the real costs of our reliance on cars and industrial food will become apparent, but our ability to opt for rails and local organic food systems will be constrained by lack of investment capital. We will be forced to adapt in whatever ways we can afford. Where prior investments have been made in efficient transport infrastructure and resilient food systems, people will be better off as a result.
To the degree that energy efficiency helps us adapt to a shrinking economy and more expensive energy, it will be essential to our survival and well being. The sooner we invest in efficient ways of meeting our basic needs the better, even if it entails short-term sacrifice. However, to hope that efficiency will producea continuous reduction in energy consumption while simultaneously yielding continuous economic growth is unrealistic.
1. See for example Benjamin S. Cheng, “An Investigation of Cointegration and Causality Between Energy Consumption and Economic Growth,” Journal of Energy and Development. 21, no.1 (Autumn 1995). However, see also Jaruwan Chontanawat, Lester C. Hunt, and Richard Pierse, “Does Energy Consumption Cause Economic Growth?: Evidence From a Systematic Study of Over 100 Countries,” Journal of Policy Modeling 30, no.2 (2008).
3. In recent years, the Chinese government has set a goal of increasing the energy efficiency of the nation’s economy (producing more GDP growth per unit of energy); it has accomplished this at least partially through draconian power cuts to cities and factories. Leslie Hook, “China’s Energy Drive: Back On Track,” Beyondbrics, posted November 25, 2010,
4. See Cutler J. Cleveland, “Energy Quality,” The Encyclopedia of Earth
5. David Stern, “Energy Mix and Energy Intensity,” Stochastic Trend, posted April 17, 2010,
6. Cutler J. Cleveland, “Energy Quality, Net Energy and the Coming Energy Transition,” in Frontiers in Ecological Economic Theory and Application, Jon D. Erickson and John M. Gowdy, eds. (Cheltenham, UK: Edward Elgar, 2007), 268-284.
7. Cleveland, “Energy Quality, Net Energy, and the Coming Energy Transition,” 7; Kenneth S. Deffeyes, chapter 3 in Beyond Oil: The View From Hubbert’s Peak (New York: Hill and Wang, 2005);  David I. Stern, “Energy and Economic Growth in the USA: A Multivariate Approach,” Energy Economics 15, no.2 (1993), 137-150; Cleveland et al., 2000.
8. Cleveland, “Energy Quality,” The Encyclopedia of Earth.
9. U.S. Census Bureau, Foreign Trade Statistics, “Trade in Goods with China,” 2005, – 2005.
10. Cleveland, “Energy Quality, Net Energy, and the Coming Energy Transition.”
11. David Murphy and Charles A.S. Hall, “EROI, Insidious Feedbacks, and the End of Economic Growth,” pre-publication, 2010.
12. Ernst von Weizsacker, Amory Lovins, and L. Hunter Lovins, Factor Four: Doubling Wealth, Halving Resource Use – The New Report to Rome (Sydney, AU: Allen & Unwin, 1998).
13. Paul Hawken, Amory Lovins, and L. Hunter Lovins, Natural Capitalism: The Next Industrial Revolution (London: Earthscan, 2000).
14. Apsmith, “What’s Wrong With Amory Lovins?,” SciScoop Science, posted September 17, 2005; Robert Bryce, “Green Energy Advocate Amory Lovins: Guru or Fakir?,” Energy Tribune, posted November 12, 2007; Robert Bryce, “An Interview With Vaclav Smil,”, posted July 2007.
15. This is known as the Jevons Paradox or the Khazzoom-Brookes postulate. In one sense, the Jevons Paradox has fading relevance in a world of declining energy resource availability: in the future, efforts to increase efficiency will probably not lead to declining resource prices, merely to prices that are not rising as fast as they would without such efforts. However, another, somewhat analogous trend will come into play: In order for society to save large amounts of energy through efficiency, very largeinvestments in new and more energy efficient infrastructure are required (railroads, electric cars, etc.). This build-up of new energy efficient infrastructure requires energy for the construction and production, which will increase the need for energy. William Stanley Jevons, The Coal Question (London: Macmillan, 1866); Harry D. Saunders, “The Khazzoom-Brookes Postulate and Neoclassical Growth,” The Energy Journal (October 1, 1992).
16. Wikipedia, “Luminous Efficacy,” accessed January, 2011; Wikipedia, “Light-Emitting Diode: Efficiency and Operational Parameters,” accessed January 2011.
17. Alan S. Drake, “Electrified and Improved Railroads,” in An American Citizen’s Guide to an Oil-Free Economy: A How-To Guide For Ending Oil Dependency, posted on, October 25, 2010.
18. Richard Heinberg and Michael Bomford, The Food and Farming Transition, Post Carbon Institute, 2009, available online at


Image credit: — fire truck thegirlrg/flickr