Sunday, April 05, 2015

The hidden reasons behind slow economic growth: Declining EROI, constrained net energy

It should seem obvious that it takes energy to get energy. And, when it takes more energy to get the energy we want, this usually spells higher prices since the energy inputs used cost more. Under such circumstances there is less energy left over for the rest of society to use, that is, for the non-energy gathering parts--the industrial, commercial and residential consumers of energy--than would otherwise be the case.

It shouldn't be surprising then that as fossil fuels, which provide more than 80 percent of the power modern society uses, become more energy intensive to extract and refine, there is a growing drag on economic activity as more and more of the economy's resources are devoted simply to getting the energy we want.

A more formal way of talking about this is Energy Return on Investment or EROI. The "energy return" is the energy we get for a particular "investment" of a unit of energy. The higher the EROI of an energy source, the cheaper it will be in both energy and financial terms--and the more energy that will be left over for the rest of society to use.

But we've seen a persistent decline in the EROI of U.S. oil and natural gas in the past century, a trend that is likely to be reflected elsewhere in the world as well. Here's a summary from the abstract of a 2011 study:

We found two general patterns in the relation of energy gains compared to energy costs: a gradual secular decrease in EROI and an inverse relation to drilling effort. EROI for finding oil and gas decreased exponentially from 1200:1 in 1919 to 5:1 in 2007. The EROI for production of the oil and gas industry was about 20:1 from 1919 to 1972, declined to about 8:1 in 1982 when peak drilling occurred, recovered to about 17:1 from 1986–2002 and declined sharply to about 11:1 in the mid to late 2000s. The slowly declining secular trend has been partly masked by changing effort: the lower the intensity of drilling, the higher the EROI compared to the secular trend. Fuel consumption within the oil and gas industry grew continuously from 1919 through the early 1980s, declined in the mid-1990s, and has increased recently, not surprisingly linked to the increased cost of finding and extracting oil.

We rarely think of the energy it takes to get the energy we need because the processes are hidden from most of us. For example, when we drill for oil, there is energy expended to build the rigs, make the pipes, move and deliver them, drill the well, complete the well and pump the oil. The people involved all require energy in the form of food to live and tools and transportation to do their work. The oil is then transported by pipeline or tanker to refineries which use yet more energy to make the final products such as the diesel and gasoline we use. These products are transported to distributors and finally to retail service stations or large end users. This list is actually cursory, but it illustrates the scope of the activities involved.

A similar series of energy expenditures could be adduced for natural gas, coal, uranium, biofuels, solar power, wind and, in fact, any energy source available to us.

The methods for assessing energy consumed in obtaining energy are not universally consistent. But no matter what methods are used, they point to one fact, fossil fuel EROI including coal has been declining. This is entirely consistent with the observation that we have extracted the easy-to-get resources first and are now going after oil and natural gas deposits that are progressively more difficult to extract--in deep shale deposits requiring extensive hydraulic fracturing or fracking, in deep ocean waters and in the Arctic. For coal this is reflected in the declining heat value per unit of coal that is now being mined.

So, if EROI has generally been declining for decades, why has the economy grown consistently? The answer comes from one more piece of the puzzle: net energy. Net energy is the energy left over for the rest of society after we expend the necessary energy to extract, refine and deliver it. That sounds like EROI, but it is an absolute number, not a ratio.

It turns out that we have greatly expanded the gross amount of energy we are extracting from all sources in the past century. This vast increase in gross extractions of energy has masked falling EROI by giving us consistently more net energy for society.

However, the growth in net energy appears to have slowed while EROI of fossil fuels continues to fall. That has led to greater competition for the available net energy and a general rise in fossil fuel prices from 2000 onward. There have been fluctuations, sometimes violent ones, tied to the so-called Great Recession of 2008 and 2009 and to the softening of the world economy in the past year which led to steep declines in oil prices (something which may be telling us there is another recession in the offing).

If the composition of our energy resources weren't so skewed toward finite fossil fuels which supply more than 80 percent of all energy to human society, then the question of net energy might be less important. The vast amount of solar energy available on the Earth's surface might be available to us with a relatively low EROI, but the gross amount available is orders of magnitude greater than the amount we are using today. As solar becomes a larger and larger part of world energy production and as the technology becomes more efficient at converting sunlight to useable energy, we may see the EROI of our total energy mix turn up.

But it's doubtful that solar and other renewable alternatives can make up for the vast energy contribution of fossil fuels anytime soon. This means that we may be facing a secular slowdown in net energy growth or even stagnation or decline in the net energy available to society. As our major energy sources, fossil fuels, continue their downward EROI trajectory, it is getting harder and harder for gross extractions to compensate.

This suggests that the net energy available to society might actually peak and decline even as gross energy extractions continue rising. No doubt many experts will cite the rising trend as reason not to be concerned about energy supplies--even though, on a net basis, energy available to society might actually be shrinking.

Kurt Cobb is an author, speaker, and columnist focusing on energy and the environment. He is a regular contributor to the Energy Voices section of The Christian Science Monitor and author of the peak-oil-themed novel Prelude. In addition, he has written columns for the Paris-based science news site Scitizen, and his work has been featured on Energy Bulletin (now, The Oil Drum,, Econ Matters, Peak Oil Review, 321energy, Common Dreams, Le Monde Diplomatique and many other sites. He maintains a blog called Resource Insights and can be contacted at


BW Hill said...

Dear Mr. Cobb,

I read your article with keen interest, and it sums up much of what we have been saying. It requires energy to produce energy, and because of the entropy production that is associated with any process that obviously must be increasing. As the process continues that leaves less, and less energy per unit for end consumer. At some point that goes to zero, and the process stops. The question remains as to where we are in the process?

As far as petroleum is concerned (which is the only area that we address) we have concluded that it is further than most assume. The reason for that is that most who attempt to analysis petroleum apply a volumetric relationship, when in actuality, it is the energy values of the process that need to be considered. This leaves most of them in a perpetual state of confusion?

One area that I did not see in your article relates to the value of the energy supplied by petroleum to end consumer. Obviously, the end consumer can pay no more for the product that they buy than the value it returns to them. We address that issue here:

This implies that the price of petroleum has an upper limit, and once that limit is reached production will stagnate, and then decline. We believe we have reached that limit.

Thank you for your effort,

Anonymous said...

Hi Kurt,

I've been reading your very useful blog for a long time but never had reason to comment. This post, however, has me very confused.

I don't understand this statement in your post:

"Net energy is the energy left over for the rest of society after we expend the necessary energy to extract, refine and deliver it. That sounds like EROI, but it is an absolute number, not a ratio."

I don't know what you mean by 'absolute number'. I know what a pure number is (dimensionless) and also an absolute value, but not an absolute number.

In any case, net energy is certainly a ratio and can be derived easily from EROI. For instance, if you have an EROI of 100:1 then the net energy from that source is 99%. (You used 1 unit energy to obtain 100 leaving 99 to use elsewhere in society; your net energy is 99 units out of the 100 or 99%, a ratio.)

"This vast increase in gross extractions of energy has masked falling EROI by giving us consistently more net energy for society."

No, it has given us more gross energy even as net energy has slowly and inexorably declined. The important thing here is that technological improvements have masked falling EROI and falling net energy because it's happened slowly over decades and because even large changes in EROI only result in small changes in the percentage of net energy, AT FIRST.

For instance, if you halve the EROI above from 100:1 to 50:1 you only reduce the net energy from 99% to 98%. However, once you start exploiting energy sources with an EROI below 8:1 then you fall off an energy cliff as the second graph in this post at The Oil Drum ('The Energy Return on Investment Threshold') shows:

Now that we are exploiting energy sources with EROI less than 10:1 (or with ethanol from American corn closer to 1:1) we are starting to experience what Joseph Tainter calls 'decreasing marginal returns on increasing complexity' that better technology can no longer keep up with and mask.

Kurt Cobb said...


Part of the problem in talking about these issues is that the terminology is not fixed. We hear people talking about EROI and EROEI which are used interchangeably. The original term is EROI and is actually derived from fisheries research.

As for "absolute number", I suppose I should have said "absolute value", meaning, of course, not a relative value.

It is certainly possible to derive a ratio using gross and net energy values, but that just demonstrates why I say that net energy is almost always expressed as an absolute rather than relative value.

So, let's say that the 524 quads of gross primary energy consumed in 2012 actually provided net energy of about 458.5 quads. That is, of course, what is available to society for pursuits other than gathering and delivering energy. We could, of course, get a world EROI from these two numbers and it would be 8 to 1. (I'm not proposing that this is correct. This is just an illustration.)

In 1980 human society consumed 283 quads of gross primary energy. Let's assume this actually provided 273 quads of net energy. The world EROI under these assumptions would be 40 to 1.

I think something like this trend has taken place. But even with a drastically reduced EROI, society has significantly more net energy available to it because the gross amount has increased so much.

Of course, I think that from here on that increases in gross production with become ever more problematic, especially with regard to oil. That implies that worldwide net energy might start to fall even if gross production increases IF THE DECLINE IN EROI IS GREAT ENOUGH. Of course, if gross production levels off and the trend in EROI remains intact, then that would almost certainly signal declining net energy.

One way humans could adapt is to become much more efficient, that is, obtain more work per unit of energy consumed. We are already doing that every year in small increments. But there are limits to how efficient one can become. Each increment of efficiency becomes more costly until it reaches a stage where new increments are not worth seeking for financial or energetic reasons or both.

I hope this helps clarify how I'm using the terms net energy and EROI.

Anonymous said...

Thanks for the clarification and explanation. I did know that EROI has its origin in fisheries research.

Since "absolute value" has a specific technical definition in mathematics, I'd rephrase thusly:

Despite the secular trend of declining EROI (and thus declining net energy when expressed as a ratio), the absolute amount of net energy available to society has continued to increase because the increase in gross production has outpaced the decline in EROI.

"Of course, if gross production levels off and the trend in EROI remains intact, then that would almost certainly signal declining net energy."

We could be in trouble even if net energy isn't yet declining in absolute terms. If the growth rate in net energy slows (which you point out appears to be the case), then we face difficulties in maintaining the economic growth necessary to pay debts and keep up with population growth. That's certainly the view of Gail Tverberg (and the main point of your post):

"Each increment of efficiency becomes more costly until it reaches a stage where new increments are not worth seeking for financial or energetic reasons or both." That's a good paraphrase of Tainter's thesis applied to energy.

I run into two kinds of techno-optimists. There's the "they'll think of something" school (one interlocutor told me, "There are unimaginable technologies on the horizon!").

The other kind of techno-optimist seizes on any good news about renewables, without noting that because we're starting from such a small base even huge increases in renewables still don't amount to a hill of beans. ("But it's doubtful that solar and other renewable alternatives can make up for the vast energy contribution of fossil fuels anytime soon.") I find the energy flow charts from Lawrence Livermore National Laboratory useful in visually demonstrating the scale of the problem:

Anonymous said...

Sounds a lot like the old "net energy analysis" of the 1970s, and input-output analysis grids I used to see. The amount of real GDP we get out of each unit of energy has been falling for decades. Does this effect any of your analysis? Thanks

Kurt Cobb said...

I'm not sure that real GDP per unit of energy has been falling. The EIA reports that U.S. energy intensity has been falling for decades which suggests that we've been getting more GDP per unit of energy over time. This is probably true in other industrialized countries as well.

What isn't reflected in these numbers is the energy intensity based on net energy delivered to the economy. That number, I believe, would be flatter, but probably still have a significant downward slope.

Perhaps more relevant and something alluded to by EnonZ, we have a financial, social and physical infrastructure addicted to economic growth. When that growth slows down as it has recently, there is trouble, trouble with banks, trouble with employment, trouble with government revenues.

Yes, the economy is probably considerably more energy efficient. But, as you can see, efficiency doesn't necessarily translate into less overall energy use. In fact, Jevons Paradox practically guarantees that the efficiency will boost growth (until there are absolute limits on that growth).

johnwerneken said...

Obvious but irrelevant. Fuel is still both abundant and cheap.

Both tyranny and other more corrupt governments both now bribe citizens to endure their mismanagement, with a bewildering array of anti-growth spending, devaluation, taxing, regulation, behavior modification, subsidy, and violence directed, allegedly, against others.

All of this slows growth. Any one of these, is more important even than the availability and/or the cost of energy.

Energy is in principle in super-abundance, from any of a wide variety of sources. There are fossil fuels. There is solar radiation. There are fissionable materials. There are thermonuclear-fusion supporting fuels. There are effects of natural sources like those: the heat in the Earth's core, the effects of the weather including hydroelectric power (all driven by solar energy). There is even gravity, now harvested primarily through hydraulic type facilities powered by lunar tides.

Kurt Cobb said...

John Werneken may be right that many of the fuels he lists are superabundant (except for fossil fuels). But that is not the same thing as available to society at a cost society can bear.

I find it is easy for people to make highly generalized and misleading assertions and get a hearing especially when it is what most want to believe. But if believing trumped the laws of physics and the dictates of human economic and political systems, then I wouldn't even be writing about energy and neither would anyone else.

Now, if energy from fusion were cheap, we would long ago have built fusion reactors. In fact, to date experimental fusion reactors require more energy inputs than the energy they generate. Net energy actually does matter contrary to his unsupported assertion.

If geothermal energy from the Earth's heat were available cheaply everywhere (instead of selectively in a few places), it would be in much more widespread use.

Most of the good hydroelectric sites on rivers have been taken already in industrialized countries though many developing countries have good sites left. But even in the thoroughly dammed up United States conventional hydroelectricity provided only about 2.8 percent of the country's total primary production of energy in 2014 and 2.5 percent of it consumption of primary energy.

There is, of course, enormous energy involved in tidal forces from the Moon. But we've been able to harvest it in only a few places where bays have been narrow enough and tides high enough that tidal hydroelectric makes economic sense.

I would agree that nuclear technology, especially breeder reactors such as the molten salt reactor, might be able to provide enormous amounts of energy by converting thorium into fissionable fuel. But deployment on a broad scale of such reactors is decades away, if it is going to happen at all.

I would agree with John that humans are their own worst enemies in many cases, and often make poor decisions. But in our era this is partly the product of living in a society that is so complex that it is impossible for anyone to understand exactly how it works.

I might add that some of the ideas John recommends are now receiving hefty subsidies in terms of government-funded research and development and in the way of tax credits. And, yet he seems to think subsidies should be eliminated. If so, he should cross thermonuclear fusion off the list of possible energy sources in the world he would rule since almost all research on fusion is government-funded.

Solar would not have the foothold it has, nor developed as fast as it has without government research and tax subsidies.

The great dams of the world are almost all public works projects. Indeed, the world of cheap energy that humans have created in the past century and a half has depended heavily on government intervention and direction.

I wonder how much longer it would have taken to bring electricity to rural areas of America if it had not been for Roosevelt's rural electrification program.