Energy Products: Return on Investment Is Already Too Low

My major point when I gave my talk at the Fifth Biophysical Economics Conference at the University of Vermont was that our economy’s overall energy return on investment is already too low to maintain the economic system we are accustomed to. That is why the US economy, and the economies of other developed nations, are showing signs of heading toward financial collapse. Both a PDF of my presentation and a podcast of the talk are available on Our Finite World, on a new page called Presentations/Podcasts.

My analysis is with respect to the feasibility of keeping our current economic system operating. It seems to me that the problems we are experiencing today–governments with inadequate funding, low economic growth, a financial system that cannot operate with “normal” interest rates, and stagnant to falling wages–are precisely the kinds of effects we might expect, if energy sources are providing an inadequate energy return for today’s economy.

Commenters frequently remark that such-and-such an energy source has an Energy Return on Energy Invested (EROI) ratio of greater than 5:1, so must be a helpful addition to our current energy supply. My finding that the overall energy return is already too low seems to run counter to this belief. In this post, I will try to explain why this difference occurs. Part of the difference is that I am looking at what our current economy requires, not some theoretical low-level economy. Also, I don’t think that it is really feasible to create a new economic system, based on lower EROI resources, because today’s renewables are fossil-fuel based, and initially tend to add to fossil fuel use.

Adequate Return for All Elements Required for Energy Investment

In order to extract oil or create biofuels, or to make any other type of energy investment, at least four distinct elements described in Figure 1: (1) adequate payback on energy invested, (2) sufficient wages for humans, (3) sufficient credit availability and (4) sufficient funds for government services. If any of these is lacking, the whole system has a tendency to seize up.

Figure 1. One sheet from Biophysical Economics Conference Presentation

EROI analyses tend to look primarily at the first item on the list, comparing “energy available to society” as the result of a given process to “energy required for extraction” (all in units of energy). While this comparison can be helpful for some purposes, it seems to me that we should also be looking at whether the dollars collected at the end-product level are sufficient to provide an adequate financial return to meet the financial needs of all four areas simultaneously.

My list of the four distinct elements necessary to enable energy extraction and to keep the economy functioning is really an abbreviated list. Clearly one needs other items, such as profits for businesses. In a sense, the whole world economy is an energy delivery system. This is why it is important to understand what the system needs to function properly.

What Happens as Oil Prices Rise

When oil prices rise, wages for humans seem to fall, or at least stagnate (Figure 2, below). The comparison shown uses US per capita wages, so takes into account changes in the proportion of people with jobs as well as the level of wages.

Figure 2. High oil prices are associated with depressed wages. Oil price through 2011 from BP’s 2012 Statistical Review of World Energy, updated to 2012 using EIA data and CPI-Urban from BLS. Average wages calculated by dividing Private Industry wages from US BEA Table 2.1 by US population, and bringing to 2012 cost level using CPI-Urban.

In fact, if we analyze Figure 2, we see that virtually all of the rise in US wages came in periods when oil prices were below $30 per barrel, in inflation-adjusted terms. The reason why the drop in wages happens at higher per-barrel levels is related to the drop in corporate profits that can be expected if oil prices rise, and businesses fail to respond. Let me explain this further with Figure 3, below.

Figure 3. Illustration by author of ways oil price rise could squeeze wages. Amounts illustrative, not based on averages.

Figure 3 is a bit complicated. What happens initially when oil prices rise, is illustrated in the black box at the left. What happens is that the business’ profits fall, because oil is used as one of the inputs used in manufacturing and transportation. If the cost of oil rises and the sales price of the product remains unchanged, the company’s profits are likely to fall. Additionally, there may be some reduction in demand for the product, because the discretionary income of consumers is reduced because of rising oil prices. Clearly, the business will want to fix its business model, so that it can again make an adequate profit.

There are three ways that a business can bring its profits back to a satisfactory level, illustrated in the last three columns of Figure 3. They are

  • Automation. Human energy is the most expensive type of energy a business can employ, because wages to paid to humans to do a given process (such as putting a label on a jar) are far higher than the cost of an electricity-based process to perform the same procedure. Thus, if a firm can substitute electrical or oil energy for human energy, its cost of production will be lower, and profits can be improved. Of course, workers will be laid off in the process, reducing total wages paid.
  • Outsourcing to a Country with Lower Costs. If part of the production cost can be moved to a country where wage costs are lower, this will reduce the cost of manufacturing the product, and allow the business to offset (partially or fully) the impact of rising oil prices. Of course, this will again lead to less US employment of workers.
  • Make a Smaller Batch. If neither of the above options work, another possibility is to cut back production across the board. Even if oil prices rise, there are still some consumers who can afford the higher prices. If a business can cut back in the size of its operations (for example, close unprofitable branches or fly fewer airplanes), it can cut back on outgo of many types: rent, energy products used, and wages. With reduced output, the company may be able to make an adequate profit by selling only to those who can afford the higher price.

In all three instances, an attempt to fix corporate profits leads to a squeeze on human wages–the highest cost source of energy services that there is. This seems to be Nature’s attempt way of rebalancing the system, toward lower-cost energy sources.

If we look at the other elements shown in Figure 1, we see that they have been under pressure recently as well. The availability of credit to fund new energy investment is enabled by profits that are sufficiently high that they can withstand interest charges incurred in the payback of debt. Debt use is also enabled by growth, since if profits will be higher in the future, it makes sense to delay funding until the future. In recent years, central governments have seen a need to put interest rates at artificially low levels, in order to encourage borrowing. To me, this is a sign that the credit portion of the system is also under pressure.

Government’s ability to fund its own needs has been under severe stress as well. Part of the problem comes from the inability of workers to pay adequate taxes, because their wages are lower. Part of the problem comes from a need for governments to pay out more in benefits, such as disability income, unemployment, and food stamps. The part that gets most stressed is the debt portion of government funding. This really represents the intersection of two different areas mentioned in Figure 1: (3) Adequacy of credit availability and (4) Funding for government services.

The constellation of energy problems we are now experiencing seems to me to be precisely what might be expected, if energy return is now, on average, already too low.

The Role of Energy Extraction in this Squeeze

When any energy producer decides to produce energy of a given type (say oil or uranium), the energy producer will look for the resource that can be extracted at lowest cost to the producer.

Figure 4. Resource triangle, with dotted line indicating uncertain financial cut-off.

Initially, production starts where costs are most affordable–not much energy is required for extraction; governments involved do not require too high taxes; and the cost of human labor is not too high. The producer may need debt financing, and this must also be available, at an affordable cost.

For example, easy-to-extract oil located in the US that could be extracted very simply in the early days of extraction (say before 1900), was very inexpensive to extract, and would be near the top of the triangle. Tight oil from the Bakken and bitumen from Canada would be examples of higher cost types of oil, located lower in the triangle.

As the least expensive energy is extracted, later producers wishing to extract energy must often settle for higher cost extraction. In some cases, technology advancements can help bring costs back down again. In others, such as recent oil extraction, the higher costs are firmly in place. Higher sales prices available in the market place enable production “lower in the triangle.” The catch is that these higher oil prices lead to stresses in other systems: human employment, government funding, and ability for credit markets to work normally.

What Is Happening on an Overall Basis

Man has used external energy for a very long time, to raise his standard of living. Man started over 1,000,000 years ago with the burning of biomass, to keep himself warm, to cook food, and for use in hunting. Gradually, man added other sources of energy. All of these sources of energy allowed man to accomplish more in a given day. As a result of these greater accomplishments, man’s standard of living rose–he could have clothes, food which had been cooked, sharper tools, and heat when it was cold.

Over time, man added additional sources of energy, eventually including coal and oil. These additional sources of energy allowed man to leverage his own limited ability to do work, using his own energy. Goods created using external energy tended to be less expensive than those made with only human energy, allowing prices to drop, and wages to go farther. Food became more available and cheaper, allowing population to rise. Money was also available for public health, allowing more babies to live to maturity.

What happened in the early 2000s was a sharp “bend” in the system. Instead of goods becoming increasingly inexpensive, they started becoming relatively more expensive relative to the earnings of the common man. For example, the price of metals, used in many kinds of goods started becoming more expensive.

Figure 5. Commodity Metals Price Index from the International Monetary Fund, adjusted by the US CPI-Urban to 2012 price levels. Commodity Metals include Copper, Aluminum, Iron Ore, Tin, Nickel, Zinc, Lead, and Uranium.

There seem to be two reasons for this: (1) In the early 2000s, oil prices started rising (Figure 2, above), and these higher prices started exerting an upward force on the price of goods. At the same time, (2) globalization took off, providing downward pressure on wages. The result was that suddenly, workers found it harder to keep a job, and even when they were working, wages were stagnant.

It seems to me that prior to the early 2000s, part of what buoyed up the system was the large difference between:

A. The cost of extracting a barrel of oil

B. The value of that barrel of oil to society as a whole, in terms of additional human productivity, and hence additional goods and services that barrel of oil could provide.

As oil prices rose, this difference started disappearing, and its benefit to the world economy started going away. The government became increasingly stressed, trying to provide for the many people without jobs while tax revenue lagged. Slower economic growth made the debt system increasingly fragile. The economy was gradually transformed from one which provided perpetual growth, to one where citizens were becoming poorer and poorer. This pushed the economy in the direction of collapse. Research documented in the book Secular Cycles by Turchin and Nefedov shows that in past collapses, the inability of governments to collect sufficient taxes from populations that were becoming increasingly poor (due to more population relative to resources) was a primary contributing factor in these collapses. The problems that the US and other developed countries are having in collecting enough taxes to balance their budgets, without continuing to add debt, are documentation that this issue is again a problem today. Greece and Spain are having particular problems in this regard.

A More Complete List of Inputs that Need Adequate Returns

My original list was

  1. Energy counted in EROI calculation–mostly fossil fuels, sometimes biomass used as a fuel
  2. Human labor
  3. Credit system
  4. Cost of government

To this we probably need to add:

  1. Profits for corporations involved in these processes
  2. Rent for land used in the process – this cost would be highest in biofuel operations.
  3. Costs to prevent pollution, and mitigate its effects – not charged currently, except as mandated by law
  4. Compensation for mineral depletion and degradation of soil. Degradation of soil would likely be an issue for biofuels.
  5. Energy not counted in EROI calculations. This is mostly “free energy” such as solar, wind, and wave energy, but can include energy which is of limited quantity, such as biomass energy.

Given the diversity of items in this list, it is not clear that simply keeping EROI above some specified target such as 5:1 is likely to provide enough “margin” to cover the financial return needed to properly fund all of these elements. Also, because the need for government services tends to increase over time as the system gets more stressed, if there is an EROI threshold, it needs to increase over time.

It might also be noted that the amounts paid for government services are surprisingly high for fossil fuels. Barry Rodgers gave some figures regarding “government take” (including lease fees as well as other taxes and fees) in the May 2013 Oil and Gas Journal. According to his figures, the average government take associated with an $80 barrel of US tight oil is $33.29 per barrel. This compares to capital expenditures of $22.60 a barrel, and operating expenditures of $7.50 a barrel. If we are to leave fossil fuels, we would need to get along without the government services funded by these fees, or we would need to find a different source of government funding.

Source of the EROI 5:1 Threshold

To my knowledge, no one has directly proven that a 5:1 threshold is sufficient for an energy source to be helpful to an economy. The study that is often referred to is the 2009 paper, What is the Minimum EROI that a Sustainable Society Must Have? (Free for download), by Charles A. S. Hall, Steven Balogh, and David Murphy. This paper analyzes how much energy needs to provided by oil and coal, if the energy provided by those fuels is to be sufficient to pay not just for the energy used in its own extraction, but also for the energy required for pipeline and truck or train transportation to its destination of use. The conclusion of that paper was that in order to include these energy transportation costs for oil or coal, an EROI of at least 3:1 was needed.

Clearly this figure is not high enough to cover all costs of using the fuels, including the energy costs to build devices that actually use the fuels, such as private passenger cars, electrical power plants and transmission lines, and devices to use electricity, such as refrigerators. The ratio required would probably need to be higher for harder-to-transport fuels, such as natural gas and ethanol. The ratio would also need to include the energy cost of schools, if there are to be engineers to design all of these devices, and factory workers who can read basic instructions. If the cost of government in general were added, the cost would be higher yet. One could theoretically add other systems as well, such as the cost of maintaining the financial system.

The way I understood the 5:1 ratio was that it was more or less a lower bound, below which even looking at an energy product did not make sense. Given the diversity of what is needed to support the current economy, the small increment between 3 and 5 is probably not enough–the minimum ratio probably needs to be much higher. The ratio also seems to need to change for different fuels, with many quite a bit higher.

The Add-On Problem for Fossil Fuel Based Renewables

With renewables made using fossil fuels, such as hydroelectric, wind turbines, solar PV, and ethanol, the only way anyone can calculate EROI factors is as add-ons to our current fossil fuel system. These renewables depend on the fossil fuel system for their initial manufacture, for their maintenance, and for the upkeep of all the systems that allow the economy to function. There is no way that these fuels can power the whole system, based on what we know today, within the next hundred years. Thus, any EROI factor is misleading if viewed as the possibility what might happen if these fuels were to attempt to operate on a stand-alone basis. The system simply wouldn’t work–it would collapse.

A related issue is the front-ended nature of the fossil fuels used in creating most of today’s renewables. People today think of “financing” any new investment, with easy payments over a period of years. The catch (as Tom Murphy pointed out in his BPE talk) is that Nature Doesn’t Do Financing. Nature demands up-front payment in terms of any fossil fuels used. Thus, if we build a huge new hydroelectric dam, such as the Three Gorges Dam in China, the fossil fuels required to make the concrete and to move huge amounts of soil come at the beginning of the project. This is also true if we make a huge number of solar panels. The saving we get are all only theoretical, and will take place only if we are actually able reduce the use of other fossil fuel energy sources in the future, because of the energy from the PV panels or other new renewable.

In nearly all cases, adding renewables requires increasing fossil fuel use for this reason. We could, in theory, reduce fossil fuel use elsewhere, to try to cover the greater fossil fuel use to add renewables, but this would mean cutting industries and jobs currently using the fuel, something that many find objectionable. Several readers have suggested that we could greatly ramp-up solar PV. Yes, we could, but we would have to greatly ramp up fossil fuel usage (mostly coal in China, if current manufacturing approaches are used) to create these panels. Any future savings would be theoretical, depending on how long we keep the new system operating, and how much fossil fuel energy consumption is actually reduced as a result of the new panels.


At this point, the foregoing analysis suggests that products created using today’s oil and other energy products are not producing an adequate financial return to cover wages, interest expense, and necessary taxes. If EROI plays a major role in determining financial returns, EROI on average is already too low for many developed economies.

It is convenient to think that an economy can keep adding lower and lower EROI resources, but at some point, a “stop” signal starts appearing. I would argue that the issues we are seeing in many sectors of the economy are clear indicators that such a threshold is already being reached. An economy in which the wages of the common worker are buying less and less is an economy in trouble. I talk in another post (Energy and the Economy–Basic Principles and Feedback Loops) about the fact that economic growth seems to be the result of one set of feedbacks. As the price of oil rises and related changes take place, these feedbacks change from economic growth to economic contraction. It is these feedbacks that we are already having problems with.

One can argue that EROI has nothing to do with these issues. But if this is the case, what is the point it analyzing it in the first place? We clearly need to understand when an economy is giving us “stop” signals with respect to increasingly low quality energy inputs. If EROI is not helpful in this regard, perhaps we need to be looking at other indicators.

Originally posted at

Also, I don’t think that it is really feasible to create a new economic system, based on lower EROI resources, because today’s renewables are fossil-fuel based, and initially tend to add to fossil fuel use.

Hi Gail, just curious, are you suggesting that this is the end of all economic systems, FULL STOP?

People will always be trading goods and services, and converting the "cost" of goods to a common denominator--whether it is bushels of wheat or US dollars. People did this 5,000 years ago, with accounts kept on tablets of clay.

What will change is the proportion of goods coming available from "promises" in advance--I will work for 40 years, and you will promise to pay me a pension when I turn 65. These promises just don't work. Or I will make you a loan of $100,000, and you will pay it back in 10 years. If things are going downhill quickly, a high interest rate is needed to make such a loan viable.

Another thing that will change is the extent of International trade. There will be an increasing number of local currencies, that are not really convertible internationally, because things are changing quickly. (These will include countries like Syria and Egypt, with internal conflict situations, or Greece and Spain with changing financial situations.) Countries will tend to set up bilateral trade agreements with trusted friends, rather than "just anyone".

What will change is the ability of any kind of currency to "hold value" over time. Even if I own gold, it will buy fewer and fewer bushels of wheat, because society's ability to grow bushels of wheat is declining over time. Money of any kind that is saved for the future will be worth less and less over time.

So I guess it is a matter of definition. New systems will be much different, and much more limited.

A lot of small currencies will decrease the need to move. Currency value will be adjusted so that it is cheaper to manufacture goods in a country with demand for imported goods.

First order of business for any successor to TOD:

Stop focusing on absolute EROI and start talking about RATE of energy return. I find this entire blog to be of very little practical use because it discusses absolute returns and not rate of return.

RATE of return is what matters, along with resource limitations and production constraints.

An oil sands operation could have an EROI of 1:1.1 - but it wouldn't matter if it was making that return on a daily basis. It would be returning 3,750% of the initial energy invested every year.

A solar PV installation might have an absolute EROI of 10:1. But if that 10 fold return happens over the course of 100 years then it is only returning 10% per year.

As someone who has made my share of pitches for project investment, I can promise that no financial investor would give you the time of day if you told him the absolute return without knowing the rate return (i.e. IRR). The principles are exactly the same here, and we need to start talking the right language if we are going to start understanding these issues properly.

Sorry Gail, other than that, thank you very much for your effort with TOD over the years. #legend.

The fact that EROI ignores timing is a major deficiency of this measure in my view. Early oil was pulled out, almost as investment was made. Intermittent renewables usually require huge up-front investments, which as you point out, is very different. This is an apples to oranges comparison.

There are major differences in types of energy as well. Intermittent electricity is not equivalent to "dispatchable" electricity, which is not equivalent to coal energy. These differences, plus "boundary issues" make comparisons from one type of energy to another very difficult.

Thanks, and thanks for not taking it personally! I agree with you on the other 2 points as well but I tend to see them as rather fudgy problems that can always be improved upon but never perfectly solved, whereas rate versus absolute returns is a relatively easy problem to solve and absolutely fundamental to understanding how useful energy can flow to society.


It is easy not to take it personally, because EROI is not my subject. I got drawn into it because others are using the measure.

"Intermittent renewables usually require huge up-front investments..."

I agree this is necessary if one is to maintain one's current economic level. In my case, I intentionally crashed my personal energy economy and created a new base to work from. The modular nature of PV allowed me to build up a new energy economic system over time (new growth along a different path) while making more investments in types of efficiency that don't sacrifice resilience, both in infrastructure and in our behavior. Why I agree this is unlikely to work at scale:

1. I couldn't have done this without the macro, mostly fossil fuel based economy. My choices would have been much more primitive. Our current economy spawned the choices available.

2. I began planning and implementation ahead of the affective decline curve. I chose to divest in previous investments; emphasis on choice.

3. I had the willingness to reset my expectations.

4. The only people I had to convince were my immediate family (tribe) who were fully on board with the idea that I could, and would, accomplish my objectives; unity of purpose. They, too, gained a willingness to invest, and had time to reset their own expectations. They became convinced that the path I set was beneficial, if not necessary (over time).

5. I took care to convert surpluses into lasting investments, limiting waste.

6. This simply doesn't work at scale. Too many widely varying objectives; too many well-established competing investments limit the efficiency of process; our collective decision-making processes are inefficient and dysfunctional (political stalemate). Far too much denial and delusion begets inefficiencies that hinder affordability. Our current economies are largely waste-based. We are in the habit of squandering surplus. We, collectively, are too busy infighting to see who/what the real enemy is. Too late - the enemy is at our door.

This process will continue to be forced and chaotic with high levels of resentment, resistance and conflict. Too bad for those unwilling or unable to see this coming, many (most?) with few choices. Great opportunity for those who would exploit increasing vulnerabilities. In this age of triage, divisions, rather than unity will become the norm. The inevitable simplification from our hyper-complex state will be messy indeed.

I sometimes talk about solutions for a small percentage of people who are able and willing to do a lot of up-front preparations vs solutions for society as a whole. Solar PV may be at least temporary (as long as the panels last, and the things they run last) solution for some small percentage of the population. They really are not a solution for society as a whole, for the reasons you mention. Once the panels cease to work, or the things that they run cease to work, there may be a need to downshift further.

"They really are not a solution for society as a whole..."

...because, IMO, there aren't any. It's the nature of predicaments, and I'm largely in agreement with Orlov as he lays it out in his "Five Stages of Collapse", discussed here recently. Societies collapse and reform; a long, usually messy process. It comes down to buying time, creating buffers, especially in regard to basic necessities; slowing the process locally so that one's group has the oportunity (perhaps just a chance) to adapt rather than panic and react. Education, hope, and a proactive, early response is the best I could come up with under the circumstances.

I think many central bankers are aware of the direction that our society is headed in, and their response is to also buy time at a macro level. EOR in oil extraction is having the same effect, though, IMO, there's a much more costly downside to centralised, top-down responses; makes matters worse. Our resources should be applied to enabling less centralised adaptations, which would, of course, require an acceptance that BAU and growth are no longer supportable. This won't happen, as that would stampede the herd. Meanwhile, society cooks away in its own detritus, suffering the consequences and latching on to whatever false hopes come along. Hard to watch, even from the cheap seats.

Rocks and hard places...


You're obviously an intelligent person who is very knowledgeable, but you're really doing a disservice by repeatedly talking down renewables. I don't mean to suggest renewables will be able to compensate for the decline in fossil fuels entirely, or that Humanity doesn't face daunting challenges, but by repeatedly slamming renewables you're just helping to perpetuate the current fossil fuel dominated system.

Gail, if Wind Power is such an inadequate energy source, why do I pay 9 cents per KWh for Wind Power that covers over 80% of my electricity consumption?

Again, I don't want people to take away the idea renewables will allow for the continuation of business as usual (BAU), but adding to the negativity regarding renewables is really counter-productive.

If renewables could really replace oil, they would be producing enough profit that they could be paying high taxes like the oil and gas industry has been over the years, and still does now. The fact that wind energy still needs subsidies (and this is without considering the amount of grid upgrades needed) is verification that wind energy is a real problem for replacing oil. The fact that wind energy still needs subsidies means it isn't even very good for extending electricity--although if wind energy it is onshore, near where the wind energy is to be used, and the total penetration of the grid is kept low, it may be a passable "extender" for other electricity.

How long wind energy lasts is really a function of how long we can keep the electric grid working. It also depends on how long we can keep the wind turbines repaired. How long they can be used is the lower of these two lengths of time. All of the EROI studies that have been done assume that they will last their full planned working life, so are biased on the high side. The front-ended nature of the investment also makes comparability with other EROIs very "iffy."

I don't know how you get the wind rate you get for 80% of your electricity consumption, but I would be willing to bet that (1) it doesn't come from a grid system that in fact is generated 80% by wind and (2) the 9 cents per KWh rate reflects more than one kind of subsidy.

You clearly have some axe to grind because no where did I say renewables would "replace oil." This is an apples to oranges comparison in the first place. As for Wind Energy "needing" subsidies, if I correctly recall the Oil and Gas Industry, in which you put so much faith in, receives considerable federal tax relief.

I'm going to ignore your last paragraph.

I'm going to ignore your last paragraph.


Remarkable post Gail. I have not read this slant on EROEI before and wonder if this is original with you? I will need to mull it over at any rate and read the comments to see if I can understand it better and possibly refute some points. If I can't, then you have really drawn attention to a looming very serious problem. I also guess I don't understand the IRR comment including the (on the surface) absurd comment that an EROEI of 1.1 could return 3750% depending upon the rate of return. The math is unexplained and I suspect doesn't reflect a real world rate of return of the fossil based industry. People are going to be upset with their renewable strategies being minimized, probably because that was their last best hope. You pointed out some obvious oft repeated points about renewables needing subsidy (hidden or otherwise) from fossil fuels to exist,but even these comments upset folks. Too bad. What I would like to see emphasized is that solar and wind have a great future as well as a great past, and essential applicability going forward, but not for generating electricity but for doing work directly, grinding grain, pumping water and the like. Wind and water are extremely efficient driving dynamo generators yielding over 90% efficiency but once they jump on the grid you face line losses depending upon the distance from the source. Solar electricity generation is very inefficient and faces long term issues you covered. Solar is very diffuse but it works fabulously well for heating structures, water, and for drying. The final point of emphasis is the low efficiency of fossil fuels especially in transportation engines which perform in the 30% range. Burning coal is better in the best combined cycle plants, approaching 40%. Gas plants almost 50% better and not spewing heavy metals and leaving contaminated waste.If this isn't a reason to end coal powered generation, I don't know what is. I am from Wyoming and could get shot for that comment. Gail, you are so polite upsetting people's applecarts. Yesterday, I was on a long hike with some very fine, intelligent and oh so wealthy folks high above Jackson Hole. I tried to point out the risks of living in a remote mountain town devoid of resources, unable to feed itself and at the tail end of supply chains. I don't think I persuaded anyone. Once they have wrapped themselves in their Lexus Suv on their way to their Gulfstream where Ben Bernanke and his financial cronies are climbing off theirs, they just don't perceive the risk to continuing their energy extravagant lifestyles.The Bourbons didn't see the risk either.
In conclusion, it is going to be painful to lose posts and comments such as yours with such a high level of value as we put TOD to bed. Thank you again , Gail.

"The math is unexplained"

Try this:

In this example The EROI is 1:1.1 because every time you put 1 unit of energy you get 1.1 units back. That is how EROI is measured. In this example the length of time of this cycle is 1 day.*

At the start of the year the oil sands company (SandTarInc)has 100 units of energy. This is the "energy investment" that the company uses to start up the operation for the whole year. This energy was lent to it by "society" and "society" had to forego some quality of life in order to give it to SandTarInc so they could start up the tar sands mining operation.

Importantly, all future energy required to keep the operation running will be taken from energy-flow that is generated as they goes along. Although all of the energy ever spent is acouunted for in the EROI accounts (to give us our 1:1.1m ratio) - only the first 100 units was ever put at risk and temporarily foregone by "society" the rest of it was just re-invested over and over again.

SandTarInc uses all 100 units on the first day, and at the end of the day it has produced 110 units of energy. It gives 10 units of energy back to "society" which has been waiting all day to have it's hot bath, and keeps 100 units back so it can operate again the next day.

The next day it again uses the entire 100 units that it saved from the last day but again produces 110 units. Again it gives 10 units back to society for its hot bath, and keeps 100 units back for the next day's operations.

He does this every day of the year except on December 31st when the CEO decides to quit to become a blogger about peak oil and the company folds. On this day it does not need to keep 100 units back for the next day so it gives that back to "society" as well.

So, "society" invested 100 units of its precious energy in SandTarInc at 12:01 on January 1st. It was then repaid 10 units of energy every day of the year. Therefore, over the course of the year received:

365 * 10 units = 3,650 units.

On the last day "society" was also given back an additional 100 units as it wasn't needed any more.

3,650 + 100 = 3,750

So, "society" invested 100 unit in OilTarInc and after one year had been given back 3,750 units, so after one year, its annual return was:

3,750 / 100 * 100(%) = 3,750%

And that's why the annual "internal" rate of return was 3,750% when the EROI was only 1:1.1

I hope that isn't patronising but you did say you didn't understand the maths. I had previously assumed it was obvious.

*Note: this is an oversimplified example that assumes 100% operational energy costs and 0% capital costs, but in a surface mining oil sands operation this is not such a terrible approximation - the vast majority of the energy costs are "variable" cost used throughout the life of the operation to power the machines that extract and transport the sands and then separate the bitumen from the sand. Relative to that the energy sunk in capital machinery is tiny over the lifetime of the operation.

The math is technically correct. However in the real world such examples might be phony due to unaccounted monetary and energetic costs.

EROEI accounts for the energetic costs - that's the point. The monetary costs - I think that's why he picked out the examples he did. If a project doesn't make a monetary return on a scale we believe is appropriate we scoff at it, although it might be entirely suitable on an energy basis.

The biggest fault, by far, has been the lack of taking externalities into account. Pollution, displacement, etc. With biofuels it comes as soil mining and displacing of fields that could be used for food. There may also be pesticide pollution involved. Tar sands use a lot of water, fracking pollutes the water table.

Does ERORI account for the massive amounts of energy from nuclear radiation and gravitational collapse that originally allowed for the epithermal concentration of minerals? Does it account for the energy used to mine minerals essential to the energy industry or the energy consumed by workers (and their families) that commute to coal mines, stripper wells and wind farms?

Does EROEI account for Kim Kardishian ?

Without her to "entertain us", perhaps the workers and their spouses might not be motivated enough to not call in sick one day ?

The boundaries can be very wide in accounting for EROEI. Kim apparently (empirical evidence) performs some function in our overall economy. In theory, I personally cannot describe exactly what, but there is ample empirical evidence that she has some very indirect effect on EROEI.


The EROI is calculated daily, but the IRR was calculated annually? No wonder the IRR looks so good compared to the EROI.

I guess I am thinking that a 10% interest on my savings ( EROI 1.1 for every 1 I had the day before) , as a daily rate of return, would make me a rich man pretty fast, but on an annual basis, its less than a wash with inflation.

They burned up an enormous amount of energy to extract that 3,750 units which means society probably had to pay for that in other ways, i.e. heat pollution, greenhouse gas emissions, whatever.

Its not a pretty picture however you want to look at it. But if its all you have...

Hi Hardhat

You are right - the EROI in the first example is terrible, a horrible amount of energy has to be invested in order to keep supplying some net energy.

But as far as society is concerned it really doesn't matter. The net energy produced each year is still huge relative to the initial investment in energy. We are rich!! (in that example anyway, and at least until the tar-sands run out or the planet over-heats).

My point is that if you just focus on energy returns, and don't worry about environmental impacts or any constraints or limitations on the resource (which are not captured by EROI either) then some energy production operations will provide easily enough annual net energy returns on investment - despite having a very low absolute EROI - to keep society chugging along very happily. Meanwhile other types of primary energy production ("capital intensive" ones) might have quite a good EROI, but actually supply from those sources might not keep up with demand, giving us a big problem. Rate of net energy return (or Energy IRR) is much more important for society to know than absolute returns on energy.

Actually having looked at recent energy costs for solar PV, I am already getting more and more comfortable about energy returns. Without having applied enough academic rigour to strongly stand behind this, my own desktop-research is showing an Energy IRR of about 20% for good PV, supplied to a near-by grid with demand matching supply profile. That is very encouraging as it shows that PV - the least constrained of all energy resources and one with huge potential for further efficiency/cost-reduction gains - is already well ahead of the kind of growth levels that make society function well (which it has to be to drive available wealth, but growth is exponential so any difference at all will scale quickly).

Unfortunately I suspect that the amount of PV that can be supplied in this 20% scenario is relatively limited - the next stage would be to add bigger and bigger transmission lines and more and more overnight storage plus new electric vehicles - and that would incrementally take that E-IRR down to a much less comfortable level and eventually negative. The challenge for society is to keep improving the technology at a rate that is faster than the rate at which we will have to move to add capacity in the less atractive locations. I am quite optimistic about that, and I think we have quite a good timeframe to achieve it especially relative to what has been achieved already in the last 8 years alone.

I would guess that a 100% renewable grid would be roughly very, very roughly 30% solar, 50% wind, and 20% other. Solar doesn't really need to provide anything like 100%.

Wind is around 50:1...

It very much depends upon the location. Solar in Iceland - not. In Saudi Arabia, potentially quite a bit.

Hydro and geothermal take as large a % as the local resource allows. HV DC (to trade renewables) and pumped storage will have to figure in most renewable grids.

Brazil could be 85% hydro, 18% solar and 2% wind for example. Iceland is about 70% hydro, 30% geothermal. etc.


Math was obvious to me.

I also think Hoover is onto something important. EROI has to add up in the positive to even have a chance for an energy source to be viable. But the rate of return is also very important, economically and otherwise. Perhaps more so than total EROI.

Looking at nature, haven't you wondered how it functions at such a low EROI? Plants and things are perhaps 2% efficient in converting solar to energy. Yet nearly all life is built upon this base of photosynthesis. Some of that is possible due to the rate and not just the total return. A small EROI per batch can be a large resource indeed if the batches are turned very quickly. Just like in Hoover's example.

That immediately points to one of the reasons solar hasn't taken off I think. The rate of turnover and payback is too slow and drawn out. Even though the total EROI is not too bad it just doesn't quite add up in the sense of rates of energy and rates of investment. It also points to maybe a different strategy for replacing fossil fuels. Things that can turn over quickly are more valuable to the flow of energy. It is the total flow our culture and economy are built upon. Yes, fossil fuels once had an EROI approaching 100, but they also had a very high rate of energy flow. Drilling wells didn't take long. Now with offshore deepwater oil drilling even if the EROI is positive look at the lag in flow how it impacts the rate at which energy is available.

Energy return rates on energy invested is a much more meaningful measure to our energy needs. Economies, civilizations, life itself are about energy flow rates. Feedbacks are about flow rates not total amounts. Money in terms of wealth is about money flowing in faster than it goes out more than about totals.

Looking at nature, haven't you wondered how it functions at such a low EROI?

Slowly. Very, very, slowly. And that's the problem. Nobody is saying that humans can not function at those energy rates. We did for million years, after all. It's just that those energy rates are too slow for our current civilization and our current population numbers.

Well, plants are pretty inefficient: overall, well below 1%. PV does a lot better. And, of course, wind is still at 50:1 EROEI.

The sun drops 100,000TW continuously on the earth - we're surrounded by vast, high quality energy - it's just a matter of figuring out how to capture it. And, we've now done that.

Yes, but why are plants so slow? Because they depend on the cycle of water, nitrogen, phosphorus, etc... And they had billions of years to optimize those processes to run for another billions of years. We may have figured out how to outsmart them, but we did it by depending on external sources for metals and raw materials, sources that are just as limited as fossil fuels. The plants' solutions may be very slow, but they are infinitely scalable. Ours aren't.

Plants have grown just as far as they can.

No question humans have limits too, but we haven't really come to those imposed by energy supply. Pollution, on the other hand...

We need to eliminate FFs ASAP.

That immediately points to one of the reasons solar hasn't taken off I think. The rate of turnover and payback is too slow and drawn out. Even though the total EROI is not too bad it just doesn't quite add up in the sense of rates of energy and rates of investment. It also points to maybe a different strategy for replacing fossil fuels.

Perhaps not...

All of the US energy capacity added in March was solar—all 44 megawatts

For all the talk of a natural gas boom, it takes time to build multibillion-dollar power plants to take advantage of the current glut of cheap fuel. Case in point: In March, the US added a whopping—wait for it— 44 megawatts (MW) of new electricity generating capacity, according to the Federal Energy Regulatory Commission (FERC).

That’s enough to keep espresso makers humming in about 30,000 American homes. Not exactly China-scale. But here’s what’s really interesting about that number: March marked the first time that 100% of new generating capacity in the US came from photovoltaic, or solar, power plants. (FERC doesn’t count the electricity generated by solar panels installed on residential rooftops.)

...Why? It’s simply much quicker and cheaper to install thousands of solar panels or erect wind turbines than build a complicated and capital-intensive natural gas power plant.

Part of they dynamic here is that electricity demand is pretty flat, so new capacity isn't really needed. But...consumers still want solar, so they'll keep installing, and taking market share away from FFs.

I disagree with a number of your conclusions and the points made by Gail.

Renewables are viable sources for a high fraction of the electrical grid.

Unfortunately, I will be busy for a couple of days to properly refute her points.


on this post.


Gail, do you really feel that renewables are subsidized more heavily than fossil fuels??

What about the $2T oil war recently fought by the US to ensure the stability of oil supplies? That investment would have bought a lot EVs.

What about mountain top removal, coal ash, mercury, and CO2?

What about the vast subsidies to fossil fuel provided by governments around the world? $.10 per liter for gas in Venezuela, or Saudi Arabia isn't a subsidy?

Do you really see the oil industry as generating a lot of taxes? Remember, fuel taxes don't go into the general fund, they simply pay for road maintenance. Further, both the Federal and local governments subsidize road consruction and maintenance heavily.

Finally, do you really not believe in Climate Change, and the cost of all that CO2 pouring into the atmosphere?

Why, oh why, do you defend fossil fuels so much?

To use wood as a heat source money need to be invested in a heating system and maybe also some forest.

An intersting thing with forests are even though it takes 70 to 80 years for the trees to grow large some large trees could be harvested every year provided that there are trees in all ages available. It is also possible to deplete the forest totally so no large trees could be harvested for 70 to 80 years.

Peat (turf) are continuosly replenished so even if takes thousands of years for a thick layer to develop some could be harvested every year but it could also be totally depleted so that it will take thousands or at least hundreds of years before any ny peat (turf) could be harvested.

I never heard any numbers of how fast other fossil fuels buried deeper in the earth like: natural gas, oil and coal replenish. Under the right conditions I guess it would possible to build a trap of silt or clay above peat (turf) so an oil accumulation is formed over time but in such case the peat (turf) can't be used today.

Historically, there has been a lot of problem with deforestation, apparently even as far back as 4,000BCE, according to Sing Chew, "The Recurring Dark Ages". It is very tempting to cut down too many trees. For one thing, this can provide more arable land, at least until erosion depletes soil level. Even more importantly, forests can be used for heating homes and for making charcoal, to be used in making metals and glass. It is this use that tends to quickly lead to deforestation.

Peat in remade at a very slow rate. According to Wikipedia, "Peat usually accumulates slowly, at the rate of about a millimetre per year."

Wikipedia also says, ". . . peat is not generally regarded as a renewable source of energy, due to its extraction rate in industrialized countries far exceeding its slow regrowth rate of 1mm per year,[6] and as it is also reported that peat regrowth takes place only in 30-40% of peatlands." I don't think we should count on much from peat.

It comes down to population and extraction rate. Avoiding overshoot is the key, something we'll come to terms with soon enough.

There's plenty of free energy if one just learns to utilize it. Passive solar/passive cooling are grossly under-utilized, and much higher levels of insulation and better construction techniques are examples of where efficiency and resilience coexist nicely. Many of our structures are much too big. Simply wearing the right clothing for the seasons and letting our bodies adapt to seasonal changes can reduce the need for external energy sources.

Isn't the problem more that we do not, conventionally, include all energy inputs when calculating EROEI. That is at least in part because they are so difficult to quantify. For instance, oil includes cost of producing pipes, cost of rigs, transportation of labor daily to and from rigs, storage facility manufacture, and later demolition, etc., etc. ad nauseam. Consequently, the input value is almost certain to be in error.

Then there are inputs later on in the cycle. Deliver vehicles (pipelines, railroads, rolling stock,) roads, more pipe manufacture, ships, disposal and clean up, and so forth.

Then there is the time value of money if there is a long time involved between start and delivery (by burning, or transforming into useful things).

I think it is more difficult to quantify the input than the output by far. The energy value of an oil or any hydrocarbon product can be measured easily in caloric terms.

Add to that the fact that some products are valuable not as energy delivery, but rather in some other sense, and can be used to create profit, even if at a net energy loss. Oil and gas would be extracted at negative EROEI for use in production of certain pharmaceuticals, and gas for use in production of plastics, as a few easy examples.

What we are left with is some generalities, and a "feeling" that at some point our economy will no longer be able to sustain if EROEI is too low.


The EROEI calculations have been "sold" as a way of telling whether an energy product is "working" in terms of providing adequate energy for investment. Yet our economy is now presenting the classic symptoms of not enough net energy, and EROEI doesn't have a way of really showing this. EROEI values have generally been moving down over time. They also vary greatly around the world--something I doubt academic studies have been able to investigate in any meaningful way. The high EROEI values of the past have subsidized our economy, but there is no from the EROEI values of knowing when this subsidy is now inadequate.

Hoover: "An oil sands operation could have an EROI of 1:1.1 - but it wouldn't matter if it was making that return on a daily basis. It would be returning 3,750% of the initial energy invested every year.
A solar PV installation might have an absolute EROI of 10:1. But if that 10 fold return happens over the course of 100 years then it is only returning 10% per year."

I think you are looking at it wrong. EROEI is supposed to measure the TOTAL investment and TOTAL return over the entire lifecycle of a resource. Of course, that means that it is a very abstract measure, and a very hard (or nearly impossible) to calculate properly. Nevertheless, when you calculate the totals, then your examples are not correct, the first one would have a much higher EROEI and the second one much lower.

Hi Strummer, I hear what you are saying but I am already doing that. Obviously I have over-simplified in both cases and just used illustrative numbers, but that is just to demonstrate a point about rate of return versus absolute return it's not supposed to be a comment about either technology.

The PV example already uses the full lifetime of the PV panel - it has to because it is 100% upfront Capital investment and 0% ongoing operational investment. The tar sands example was the exact other way around 0% capital and 100% operational cost. Because it was 100% operational costs and the operational cycle in my example was 1 day, I just worked on that basis, but if the resource lasted 100 days (or years) both the inputs and the returns would be multiplied by 100 and you would still get the same EROI.

Of course in real life there will always be some capital investment so you will always have to work on the basis of the full lifetime of the investment (and likewise there will always be some ongoing operational costs as well). But none of that changes the point that EROI does not account for time and hence tells us next to no useful information about how free energy will flow to society or how that energy could be re-invested to make sure that even more free energy flows at some point in the future when we might need it.

Let's take it to the far extreme:

I find a resource into which, when I put 1 joule of energy it will pay me back 1.01 joule after 1 second. It doesn't matter how may joules I put in, I always get 1% more back in return, so it has an EROI of 1:0.01. If it was a limitless resource and I started the day with enough energy to power an LED lightbulb for 1 second, and kept on reinvesting my 1% gains every second, by the end of the day I would be producing - with my energy source EROI of 1% - far far more energy every minute than the entire human race could possibly consume in its history before our Sun goes super nova.

But if I had an energy source of EROI 1:1,000,000 that had a 100 year production cycle with all of the energy returns being generated in one go at the end... well it might have been some use to my grandkids, but they were never born because my own kids and I starved to death first.

They are stupid examples, that obviously don't exist, I am just trying to point out that to measure energy returns without knowing the rate of return is of purely academic interest and of no use to society.

This is a very good point, but more importantly: no financial investor would give you the time of day if you told him either the EROEI *or* the energy rate of return, without knowing the financial IRR.

EROEI is really only useful for analysis of unusual cases, like biofuels, where subsidies are large or the business case is still very theoretical. In the case of almost everything else, like PV and wind, the EROEI is "good enough", and the financial case (combined, of course, with all of the other important factors, like internalization of pollution and security) is the most important thing.

So, for instance, PV is now nearing and surpassing grid parity in many places, as the Germans, good engineers that they are, knew that it would.

"An oil sands operation could have an EROI of 1:1.1 - but it wouldn't matter if it was making that return on a daily basis. It would be returning 3,750% of the initial energy invested every year."

I point that out every time I talk about EROI, and I used an example like the one you did above. The other major caveat is that unless you are talking about fungible energy inputs you could have a situation with a poor EROI and great economics. The EROI in that case just means you are accelerating the depletion of the input energy, even though it may be cheap input energy.

Fungibility of energy inputs and outputs is such a basic problem.

What if the inputs are cheap natural gas, and the outputs are peak electricity, which is worth 3-10x as much per joule (to the grid and it's consumers) as the NG?

Or cheap NG, vs expensive liquid fuel?

Surely EROI (or E-IRR!) is only applicable to situations where primary energy is being produced?

As a power station developer I can honestly say that of the couple dozen or so projects I have worked on over the years not a single one of them had an EROI above 1:1 or a positive E-IRR.

At this point this is where the good old financial returns help us out - and as you said earlier Nick, we can never forget about those.

For energy accounting purposes though, I think some simple principles can give us a good idea:

gas to power: based on the efficiency of a good CCGT
gas to liquids: volume weighted average of (a) efficiency of a GTL plant and (b) small scale liquefaction plant (volume weighting of LNG to depend on amount of LNG used in transport in that country, changing over time)
power to liquids: arbitrage through gas
liquids to other liquids: crack efficiencies
anytime power to peak power: pumped storage efficiency
etc etc etc

I think some simple principles can give us a good idea

I take it you mean that these are good benchmarks for comparison & evaluation of a proposed project?

One quibble: for "anytime power to peak power": efficiency may be only a secondary criterion. In particular, if we're shifting surplus wind power from one part of the year to another, capital cost will be the overriding factor, not efficiency. So, a plant that uses surplus wind power to create hydrogen or methane (to be stored underground, or in the NG pipeline network) needs to have capex costs that are much, much lower than pumped storage, and can handle relatively low input./output efficiency.

Surely EROI (or E-IRR!) is only applicable to situations where primary energy is being produced?

Well, both wind and PV produce electricity, which is 3x as valuable as FF primary energy, joule for joule. If PV has a EROEI of 10, but uses NG to heat it's polysilicon, and outputs peak electricity, it's EROEI is arguable 3-10x higher, as a practical matter.

Most rate of energy return issues can be analyzed into resource cost issues. For example bio-energy crops require a growing season, so there appears to a time issue involved. But if we had enough fallow land available we could get very high yearly return rates. The fundamental economic issue is the opportunity cost of land use and not the rate at which crops can be produced. Of course energy balance matters since you need to know how much useful energy was produced for a given resource input.

Similarly how fast we can extract oil from the Canadian tar sands depends on how much capital and labor we are willing to invest. Again the basic issue is the opportunity cost of investing such resources rather than a time issue per se.

The use of EROEI as an economic parameter can charitably be interpreted as an attempt to make the input energy serve as a marker for the total resource opportunity cost of producing energy. But the idea that dealing with dimensionless energy ratios frees us from the complex process by which value is placed on various production resources is pseudo-intellectual delusion.

it's most useful for unusual situations, where EROEI is very low, or negative.

A classic example is Soviet toilet paper: if you subsidize it too much, the toilet paper makers start buying it from the stores, and using it to make new toilet paper, because it's cheaper than the normal wood feedstock!

I think high EROEI is necessary based on a couple of informal questions
1) what distinguishes us from hunter gatherers?
2) why can't we shake off coal?

We are different to Neolithic people because we have cars, computers, electric toothbrushes, air conditioning and effective medicine. These and a myriad other things make up the obligatory energy load for modern society. There's no EROEI for Medicare but the system needs high EROEI to pay for it. We used to get by burning wood grown in realtime then we changed to accumulated fossil fuels like coal, oil and gas. Now some think we can go back to realtime energy fluxes. Maybe not.

The coal and gas fudge. We look at PV and wind turbines and think ain't they sweet but without the despised fossil fuels we wouldn't have the necessary steel, silicon and cement nor most backup power. As Gail says things will be tough without fossil fuels unless wind and solar get to run their own mines, furnaces and factories.

Germany has an aggressive renewables program but is building new coal plants. Same goes for China who are now starting to publicly worry about extreme weather. There was an excellent show 'Ten Bucks A Litre' on Australian TV tonight hosted by businessman Dick Smith. He flew his own exec jet, helicopter, prop plane and ultralight to various energy installations. He covered nukes, wind farms, off grid PV, EVs, conventional and coal seam and shale gas, solar thermal and efficiency. There waiting on the coastline were all the coal ships going to China, India and elsewhere. For now some are saying we don't need no coal others can burn it for us. Maybe all the low EROEI stuff is a big delusion for domestic consumption so long as we can import manufactured goods made with coal.

I think trying to discourage coal usage internally while allowing importation of goods made with coal from the rest of the world is counterproductive. What happens is we send manufacture of goods overseas, where they are made with coal. Sending manufacturing of goods overseas has a multiplier effect, because those countries suddenly need all of the services, roads, and new homes that go with new prosperity. These are all created using coal. At the same time, the US and other developed goods get the "reverse multiplier" effect of losing the jobs that would have gone with manufacturing if it had stayed here. Think of Detroit, with its lower population, and less need for school teachers, police (?), grocery stores, and many other things.

Most of the time humans and pre-humans have lived on earth, we were hunter-gatherers. Even then, we did not live very sustainably, because we killed off the large animals and not-so-smart animals, such as the auk, as soon as we moved to new territory. We also burned down woods, in attempts to force animals out to where we could reach them. Even back then, we had a problem with rising population, so there was no doubt fighting among groups. I think the main thing that distinguishes us from hunter gatherers is that there are more of us now, so we have a need to use more resources than what can be obtained from simple hunting and gathering.

We can't shake off coal because in any economic contest, "cheap wins." Coal comes out at the bottom, or near the bottom, in almost any cost comparison.

Well perhaps what we need is a 'coal tax' put on imports from any nation that uses more than X amount of coal per capita. I agree eliminating CO2 output here only to have it increased overseas does not solve anything.

And if you look at the pollution in some coal-heavy areas, you'd think that they also realize that they need to reduce coal usage as well.

I don't know whether we can get off coal, but accepting Gail's thesis, it doesn't sound like we get off fossil fuels in general until we become unable to extract them. But since, apparently, renewables are dependent upon fossil fuels, we eventually end up in a hunter, gatherer situation assuming there is anything left to hunt and gather.

There is still a bit of an unknown for me as I see the production of solar, for example, becoming increasingly efficient. Will there ever come a day where solar electricity can be produced without any or significant fossil fuel input. I know this is a trivial example but I see some of the ff input being reduced locally by the fact that a local PV installer brings their panels and tools to the site via a bike hauler. Trivial but maybe it is the beginning to breaking down the whole process and trying to attack each step by figuring out a way to reduce or eliminate ff input.

Well, at least the bears and the ground squirrels here locally have figured out a way to deal with seasonal heat and cold with respect to their housing needs. But with 7 billion people and counting I think we will be overrun with global warming before we make a dent in this problem.

"Will there ever come a day where solar electricity can be produced without any or significant fossil fuel input."?

Latching onto the idea that fossil fuels are a direct requirement for the production of solar panels misses a broader point, one I think Gail makes and many are missing:

Can the complex systems of exchange, manufacturing, marketing, etc., required to produce and deploy PV at scale continue to function as they are? It's a hyper-complex system that requires many inputs, mostly subsidized by fossil fuels. Many things have to go right. The big question is; can a broader economy dependent on fossil fuels and many other inputs (mostly dependent on fossil fuels to be brought to market economically) continue to function at a level that allows things like PV panels to be produced and deployed,, and will we be able to afford such without the fossil fuel energy subsidy? With so many things pointing in the wrong direction, I have serious doubts. I think we'll be too busy feeding ourselves and trying to not kill each other; all competing for a declining, more expensive energy base.

The availability of fossil fuels is not an "On/Off" switch but a complex progression that varies by fuel type, fuel demand and likely locality (see LNG trade by tanker vs. US NG prices and availability).

I can see a European nation where people move with their feet, bicycles, trams and trains but ambulances, garbage trucks and fire engines move with ICEs - powered with either FF or renewables. Almost all cars are EVs and 18% of the population owns one. (18 cars per 100 population). Goods move mainly with electrified trains, ICE barges & coastal ships, EV delivery trucks with some ICE trucks for longer distances. Farms use ICE tractors and farm equipment. ICE is powered by a combination of FF and bio-fuels.

Only 1% of the population flies in any given year, using FF. Sort of like 1965 in that way. Trips up to 800 km (500 miles) are routinely made with high speed, or just fast (150 kph or so), trains. Some trips longer than 800 km are also made by train, usually with a change of trains in a major city.

Per capita electrical demand is down by half (say 1/4th of current US electrical demand) and the capitas are down too, perhaps -10% from the peak population.

Electrical production is mostly renewables (hydro being a good part) with lots of pumped storage, perhaps some nuclear and the balance coal to fill the gaps. HV DC to transmit surplus renewables and import hydro & other renewables as needed.

Such a nation could produce solar PV when electricity is cheap i.e. when renewables are in surplus. They could also use electric arc furnaces to recycle steel - again when renewables are in surplus. Steel that could be used for wind turbines and their towers. This is exactly the sort of scheduled demand needed for a high renewable grid, EV charging is another.

And this is clearly the trend for France and Denmark 2040+.

And it could be the United States as well. The French effort could be duplicated with a quarter of our subsidy for cheap gasoline and diesel - about $25 billion/year (adjusting for population and currency).

Both nations are net food exporters BTW.

Best Hopes for Those That Prepare,


For Denmark, hydro and pumped storage are in Norway. Today, four HV DC lines are in operation between Norway and Denmark. The basic trade is Danish wind (when in surplus) for Norwegian hydro (as needed). Somewhat similar trade between France (excess nuke power late at night) and Switzerland.

Actually, I think the limiting factor on fossil fuels will be sales price dropping too low, to keep up production. This limit will be reached because debt for goods like cars and homes doesn't keep growing, and because wages don't rise by much, so consumers cannot afford cars and new homes. Unwinding quantitative easing can also expected to have a downward impact on commodity prices, including those for energy products. So we indeed will get off fossil fuels, quite possibly in the very near future, because of prices too low to sustain production. This will not be a nice way of doing it, however.

I need to comment on the rigor of this post. Analysis is not a proof-by-picture process. The association of oil price and wages in fig 2 is suspect. Such a claim would produce...eyeballing the picture.... a poor correlation. However, that is my point. The sweeping claims of this post are not convincing when built on poor analysis. Fig 2 and subsequent claims is a classic case of cherry picking. Shame on ToD for posting this. The quality used to be higher.

Rigor? Hell the Term EROI was used, what more do you want? Running with the Red Queen? BAU? Oldivi?

"Shame on ToD for posting this."

Shame on you for not being more specific in your counter-points. Pot calls kettle black.

What is not specific about my claiming that Fig 2 should be a regression analysis to have any merit?

Specifically: a lagged regression model comparing change of wage with oil price.

It's not hard. A one-liner in any stat language given the dataset.

If it's not hard, perhaps you can offer a more accurate analysis; fill the void you have created. Criticism is not a solution.

It is called peer review. Something the editors should have done.

What I am saying is that the absolute level of oil price is important, relative to the change in wages. This is different from what you are trying to test. The reason is the one I gave--When oil prices are low, there is a large difference between:

A. The cost of extracting a barrel of oil

B. The value of that barrel of oil to society as a whole, in terms of additional human productivity, and hence additional goods and services that barrel of oil could provide.

This difference can, over time, feed into wages, because it looks like increased efficiency for humans, together with increased profits for a company. When oil prices are high, this difference disappears, and this difference can no longer act to support wages.

If that is what you are arguing then the figure should be plotted as the relationship between those variables over the entire dataset. Otherwise, you are hand-picking data to support your argument and providing no statistical evidence of its likelihood. Plot: X-axis = absolute price of oil and Y-axis = change in wages (i.e., wage(t)-wage(t-1)). Perform regression (I would suggest at least a lag of 1-year) and post the r^2 values at least. I'm not saying you're wrong, I'm saying that Figure 2 is not evidence to support your argument; therefore, you have no argument.

Edit: As I posted below, if you post the raw data I will do the plot. I will plot: panel(a) the raw data as a time-series to recreate your figure (so everybody sees that it is the same dataset) and panel(b) the regression described above. I will even search over all possible lags to find the highest r^2 value to best support your claim. However, I am not satisfied with the evidence provided and it is below the lowest level I have yet seen on this site.

I attached he dataset, which can be found at this link.

I converted Gail's Excel file to csv format, removed header information, and rescaled wages to match her figure exactly, Panel (a). Panel (b) depicts the relative wage dependent on oil price. Note, r^2 of 0.086 means that only 8.6% of the variance is explained by the model Panel (c) depicts r^2 values over various lagged regression models. The best explanation is a 7 year wage lag (r^2=.33 is still pretty weak evidence but moving in the right direction).

Here is the link to the figure.

Here is the source code Matlab version: (R2012b)

% Source: test_oil_wage.m
% Author: bristlecone
% Date: Aug 1, 2013
% Purpose: Validate/refute hypothesis of the
% relationship between abs oil price and
% relative wage
clear all; close all;

%Format wages into figure units
data = importdata('oil_wage.csv',',');
data_v2 = data;
data_v2(:,3) = data(:,3)*120/25;

hold on;
hold off;
title('Per capita non-gov wages compared to oil prices');

Nyears = 30;
r2vec = zeros(1,Nyears);

for(i =1:Nyears)

lag = i;

regressX = data_v2((lag+1):end,2);
regressY = data_v2((lag+1):end,3)-data_v2(1:(end-lag),3);

%Perform Regression
p = polyfit(regressX,regressY,1);

%Compute r-squared
f = polyval(p,regressX);
yfit = p(1)*regressX+p(2);
yresid = regressY-yfit;
SSresid = sum(yresid.^2);
SStotal = (length(regressY)-1)*var(regressY);
r2 = 1-SSresid/SStotal;

r2vec(i) = r2;

[ysort isort] = sort(regressX);
hold on;
hold off;
r2string = ['r^2=',num2str(r2)];
xlabel('Absolute oil price(year=t)');
ylabel('Change of wage: wage(year=t)-wage(year=t-1)');


xlabel('Wage lag');
ylabel('r^2 of model fit');

I would observe that any mathematical model you put together has a number of built in assumptions. If a particular model doesn't fit, all that proves is that the particular model isn't right.

The way I look at things is to see what patterns I would expect to observe, based on the underlying interactions of the data. I want to understand how the system works first. Then I look at the data, to see whether it in fact fits the general shape I would expect. When I do the analysis this way, the data "makes sense." In general, growth in wages takes places when oil prices are low, but not otherwise.

Economics has developed an addiction to fancy models and expected changes in one variable based on small changes in other variables. These models may have nothing at all to do with underlying reality. The economy doesn't necessarily work in the way the model assumes. If a person can put together a sufficiently complex model that it somehow agrees with reality, there is a remote chance that it will have predictive power. But if it really doesn't incorporate the actual dynamics of how the "system works," the likelihood of this is very low. Starting from the assumption that year to year fluctuations changes are meaningful indicators of long term trends handicaps a person significantly in understanding the real interactions.

In my view, the best model is the simplest one. In fact, I talk about some indications of simple models in Energy and the Economy-Basic Principles and Feedback Loops.

"However, I am not satisfied with the evidence provided and it is below the lowest level I have yet seen on this site."

Jeez, you are a bristly cone. Casting stones such as the above isn't improving the conversation nor the data. Best hopes for more constructive responses on all levels.

bristlecone you're detonating, which just comes across as arrogant since you are in this case merely a couch potato passing judgment on someone's else's effort. Why don't you contribute what you think would be a better fig. 2. Gail's given you the data link and you are apparently clear on what it should include. Have it done and posted by 5:00pm ET.

Yair . . . I am an old uneducated bushman and I tend to look at things through different eyes.

It won't happen but it is possible for humankind to live happy fulfilling lives without laying waste to all the wonder and beauty of this planet.

You have to get back to the basics. We need food, fresh water, a comfortable place to live and raise kids and creative things like art and storytelling and music to occupy our minds.

From what I have seen of the Pacific Islands, PNG and our own aboriginal cultures it seems to me populations were kept in check by the availability of local recourses and for the most part the people lived as described above.

Just as much pleasure can be obtained by gliding down a creek in a tin canoe as doing the same with a jetski . . . you just need the right mindset.

I too watched that Dick Smith program and found it quite disturbing to actually see the extent of the bull-shit going on to sustain an unsustainable way of life.


A good example of what Gail writes about how companies try to survive:

Holden workers are facing a three-year pay freeze and cuts to conditions to keep the company's plant at Elizabeth in Adelaide's north open.

The car maker is looking to save $15 million annually in labour costs to stay viable.

John Camillo from the Federation of Vehicle Industry Unions says the future of 1,700 Holden workers hangs on their vote on Friday next week.

"Holden has made it quite clear that if the workers reject the variation they have decided to close the operation down in 2016," he said.

"There's no plan B in this one. The workers will make a decision on August 9 and if it's a no the company will close."

More than 50 per cent of workers must support the secret ballot for the changes to be accepted.

Mr Camillo says Holden's commitment to Australia is also dependent on continuing taxpayer support.

"If the workers say yes and the Federal Government is prepared to support GM there will be replacement of the Cruze and Commodore," he said.

Mr Camillo says a proposed 10 per cent pay cut has been taken off the table.

Holden's managing director Mike Devereux will be in Adelaide on Monday to speak to workers about the changes

THis really should be a comment in response to Bristlecone's post above. It is an example of why what I am saying is true about the connection of salaries with high energy costs. High cost feed through to lower worker wages.

Gail, thanks for this interesting posting. However I think that this might be a typical case where the rule applies that correlation does not necessarily mean causation:
It is true that wages went down when the oil prices were high - and the same is also true for many other industrialised countries. And I also can imagine that the oil price played some role But at least for the period after 2000 there were also other important factors that "helped" to bring wages down in these countries:

- I think that after the breakdown of the communist block there was a general shift in mentality to think that the western, capitalist system has shown to be superior - so the world would become even better the more capitalistic it gets. Thus, in many countries there was a strong tendency towards fixing or lowering wages (especially those of low-payed employees).

(Curiosly, as TOD readers know, the oil price also was one cause for the breakdown of the Soviet Union, which had problems with too little oil export revenues when the oil price was low.)

- Furthermore western enterprises felt challenged by the growing competition from China (especially in the area of low qualified jobs), which even more lead to bringing wages down to levels deemed "competitive" with China.

So, the oil price may have had its effect, but it is probably not the only cause. And it is probably almost impossible to disentangle this mess of causalities and effects.

I am not trying to troll this article today. I've been a daily reader of TOD for almost four years. However, this post crossed a line with me. Gail does not get a pass just because she is well known; she should be held to high standards, as non-technical people may take her at her word.

Correlation is not causation. This is true. However, she is claiming causal structure which is always correlated but may be so through a temporal lag (hence my requirement of lagged regression), or through nonlinearity (which is harder to show).

Gail's argument becomes invalid at the point of Figure 2. She has the data and as an actuary she should have the skills to produce a figure that satisfies my comment. Either she retracts the figure or she produces a new figure that overcomes my challenge. This is how quantitative arguments are made and supported. I'm not going to let TOD go to archive and mislead future readers and researchers without giving them pause as to the validity of the content . Either the authors step up or the editors step up, but this is not acceptable.

Can you disprove her figure 2? I believe the onus is on you...

Without the dataset, no. If she posts it...I would be happy to run a figure and post the result with associated code., the burden of proof is always on the author in the peer-review process...kind of like the "beyond a reasonable doubt" standard in law. Peer-review prevents tunnel vision and group think.

Figure 2 disproved. See my figure of the regression analysis in the post above.

I disagree. The fact that a particular mathematical model doesn't fit just proves that this is the wrong mathematical model. It is not proving what I set out to show, which is verification that in general, the data is working in the direction one would expect it to work, from the underlying principles of the system. See my comment above.

And my argument is that you are responsible to show the correct model before making claims. I refuted the hypothesis class that you were using in your post.

Edit: I want to be very clear for 3rd party observers. Figure 2 is put forth as evidence of a correlation between absolute oil price and relative wage. I have shown via my analysis of the data that any such linear correlation (the common benchmark for such a claim) is refuted. Figure 2 should be retracted (or removed from the argument) unless a model of the data can be supplied that shows the claimed relationship.


One of the finest aspects of TOD comes to the fore. Learned, intelligent, focused and hard criticism of ideas, rather than ad hominem, bafflegab and nonsense that usually pervades discussion and debate essentially everywhere else on the web.

It will be a travesty great proportions to me when this place is gone.

I have been writing about globalization having an impact on wages, and indeed, this is a factor as well. In fact, I mention outsourcing to lower wage countries as being one of the issues.

Oil use and human wages are very closely related. The purchasers of human wages are also the purchasers of oil for making and transporting goods. If the cost of oil goes up, it disturbs the whole relationship. It raises the cost of the finished product. Higher oil prices also disturb demand for finished products of all types, because wage earners (including those not working for the company) find that their wages go less far, and must cut back on discretionary purchases. Hence, manufacturers are caught in a double-bind: less demand and higher costs. They must cut in one way or another, or go out of business. Wages are always affected. Note that in my calculation of wages, dropping out of the workforce or unemployment also counts as a reduction in wages--not just a cut in hourly pay.

The fact that other industrialized countries are seeing the same effects is showing that my theory is true, not refutation of it. They are big oil users as well, and have the same problems that we do.

I am an Economics PhD student and I'm working on incorporating energy into macro models - and specifically hoping to show a relationship between EROI and wages. I wrote a short post about my model, which is definitely in the early stages, and I would love to hear feedback from TOD readers!

This comment had been hidden, and I unhid it. The model proposed is incomplete, but seems to need a lot of work. I am sorry I don't have time to help with this, but perhaps someone else wants to look at this.

Of course Gail is an exceptional contributor, but with this kind of brilliant analysis appearing almost daily on TOD, analysis which is just not available on other sites, how can we imagine that TOD will go silent in just one month? Please say it ain't so.

I continue to write on Our Finite World. In fact, Our Finite World carries many of my posts that are not on TOD. I often get 200+ comments on my posts on OFW. There are button on the front of the site for e-mail subscriptions (you may need to sign in to see it) and a place for RSS subscriptions. I also Tweet new posts, when I put them up. My Twitter address is @gailtheactuary . It is also possible to get notifications on Facebook and on LinkedIn of my posts.

The notion of momentum may be applied to oil extraction. This is closely linked to the price of a marginal barrel. The world is running on the momentum of past discoveries in Saudi Arabia Iraq and Russia.

If every oil of barrel used today required the same extraction price as current production in the Baaken or Albertan oil sands then the world's economy would be seriously in decline. But this is where we are headed, no doubt.

There is a solution - and both France and Denmark are going that way.

Per capita carbon emissions, 2007-2012, for Denmark are down -26.5%, France -14.8%.

Both are moving towards highly efficient oil free transportation. Both bicycling and electrified rail.

This is, perhaps, a "different economy", but I see it as a shift in priorities and mix.

Paris announced last March that they will double the Paris Metro, and make other major investments, so this transition should just speed up.

Best Hopes for Wise Public Policies,


Is any of this reduction in carbon emissions attributable to importing more goods from China?

Not materially. the EU imports less than the USA. I think Chinese imports may have dropped, rather than increased since the crisis of 2008.

Neither Denmark nor France has pushed solar PV (unlike Germany) so not much there.


Doesn't globalisation increase wages? The effect you show of depressed US wages must be tiny compared to the boost in Chinese wages over the same period. If you want to claim that the net effect is depressing global wages, at the very least you have to include a decently representative sample rather than cherry picking US data to represent the global effect.


The idea that there is something magically better about energy from fossil fuel because it happens to be most of the current supply is a fantasy. Neither renewables nor fossil fuels care whether the energy used to extract them is renewable or fossil. Fossil fuels were originally extracted with renewable energy. That didn't prevent them being extracted with fossil energy later and the same goes for renewables.


EROI is a number that depends on arbitrary assumptions in its calculation to such as degree that the only useful information is whether its greater or less than 1. The exercise of calculating it can provide interesting information about a system, but its wholly inadequate for the sort of "society requires EROI of 5" type of argument. If you give someone an incentive to calculate an EROI of over 5, or 50, or 5,000,000 they can do so, as long as the system actually returns an excess of energy. And the converse is true. If you want to show that EROI is under 5, you can always do so. Its simply a matter of choosing where you define the system boundary. The analysis can be informative, but the only meaningful information in an EROI number that is divorced from the analysis that produced it, is whether its greater or less than 1.

Society needs more than 1, but EROI is not capable of measuring how much more than 1, because any number greater than 1 is interchangeable with any other at the whim of the analyst calculating it.

The financing effect Tom Murphy describes is real, but EROI is not up to the job of describing it. It requires using an energy analogue of the sort of tool that is actually used for evaluating financial investments instead.

Thank you
Reds wp

See my figure in the post above. The effect is very weak.

Also see my comment. Failure to prove, using a particular model, simply proves that chosen model is wrong.

Gail's post makes a claim that my analysis, using standard methods on her dataset, refutes. It is the author's responsibility to provide some mathematical model that demonstrates the validity of the core statement of their argument. Otherwise, the argument is baseless. This problem should have been picked up by the editors of this site before posting. Again, I appeal to the editors of TOD to retract Figure 2 unless Gail can provide a model that supports her claim.

I am uncomfortable with some off the figures used; for a different reason though.

It seems to me that there could be other factors(s) underlying both. There is so little time, and there are so many possible facts.

That I agree with the proposition that declining resources necessarily cause declining economies gives me pause as my skeptical nature says, "Show me." On the other hand, I am not sure Gail did any more than to state that the figure indicated a correlation. That she believes it causative is clear; whether she proved it not so much.

As stated elsewhere, some of the data are not clear either, such as, what does "energy invested" include? Whose wages are to be used? China's? India's? Or, those in the USA?

Further, we are all used to loose interpretations and assertions from some of our "holy books." (e.g., The Long Emergency, The Long Descent, etc.) where we accept almost without question that we have insufficient resources to retool, revise our electric grid, create sufficient solar arrays, etc.

Like most, we at TOD have a bit of a confirmation bias when searching through facts. More the pity since the situation may be as grim as the worst doomers on-site believe, and is at least not likely to be as trivial as others would like it.

The best we can do is to read, make public our criticisms, and try to do better in the future.

Which is all to say that you made your point, and quite strongly. We all understand, and agree or not, it is time to move along.


Failure to prove, using a particular model, simply proves that chosen model is wrong.

Or that the claim being asserted is not actually true.

That you don't seem to consider that possibility is telling.

"...compared to the boost in Chinese wages over the same period..."

It's an apples/oranges comparison. The Chinese economy is different in many ways than the US economy. Fuel in China is more subsidized and they have a history of getting more utility from their fuel. Of course, if they aspire to a more western lifestyle, this may change in time. Much of fuel use in the US is discretionary, supporting a discretionary economy which provides many jobs. When fuel prices rise, many of these discretionary uses of fuel are discarded, along with the jobs they provide. Replacement jobs are often part-time, lower paying, with fewer benefits.

While the claim that rising energy prices cut into wages may seem simplistic, saying it doesn't, or discussing lags, etc, is also simplistic. It's more complex than that, and there are other inputs to the system (QE, fiat capital, debt) that mask the role of energy in the system, especially when energy costs take their toll. Gail's, and/or Bristle's charts are thin slices of a broad set of metrics. It's hard to ascertain the real effects of rising energy costs in a system that is busy robbing Peter to pay Paul.

Whether or not globalization helps world wages, it doesn't help the financial condition of the countries whose wages decrease or fail to increase. If we look at historical collapses, the reason they have happened has been closely tied to governments not being able to collect adequate taxes from stagnating wages of workers (according to the research of Turchin and Nefedov in Secular Cycles). Stagnating wages also makes it very hard to repay debt.

Fossil fuels have been extracted in low quantity with renewable energy. At this point, I would expect that the vast majority of the amounts that can be extracted with renewable energy have already been extracted. Coal that is close to the surface, in readily accessible locations, and that does not need modern oil-based transportation to places where it can be used, is very limited in supply. Oil, natural gas, and uranium all need high tech methods for extraction now, that are only available with fossil fuels. Rare earth minerals extracted for use in "renewables" are terribly polluting. Whether they can be extracted with human and animal labor alone in such a polluted atmosphere seems doubtful.

What makes you think that in the absence of globalisation they would have done better, rather than losing big time instead of small time? I am pretty sure it is possible to come up with a group of people that definitely lost from globalisation, but what gives them the right to keep all the ones that gained in poverty?

Which is better a world with one person at 10000$ and ninety nine at $1 each, or a world with one person at 9000$ and ninety nine at $100 each? Turn the clock in the opposite direction. What would you regard as the moral status of a person that would extract 99% of the income of 99 people already significantly poorer than himself in order to mildly enrich their lifestyle?

I'd say its not so much the inability to collect taxes from the stagnating wages as the unwillingness of the elite to forgo their exemptions as they take a steadily larger slice of the economy.

If it can be extracted with fossil energy it can be extracted with renewable electricity. I doubt there's much coal that would support a 21st century lifestyle when using 16th century extraction technology, but 16th century windmills weren't as effective at generating electricity from wind as 21st century ones are. The machinery doesn't care whether its electrons come from a wind turbine or a gas turbine, and it doesn't care if the gas in the turbine came from a biomass digester or a fracked well. There's good reason to believe that wind turbines and biomass generators aren't up to providing current levels of energy supply, because there just isn't enough wind or biomass, but there's good reason to believe that fossil fuels won't last forever at current rates of use too. However, there is enough sunlight.

intuition. Globilisation should over time even out wages with EROI and per capita energy UTILITY dictating the maximum permissible level of this global average per capita wage.

That said local and short term effects from changing oil prices are likely to slosh the wage allocation from expensive workers to cheaper and vice versa. There is going to be a fair amount of noise but perhaps less than one might imagine.

This is the type of article why TOD is shutting down.

No causality, no firm, well reasoned out logic. Very light on facts, heavy on unsubstantiated causal relationships. Cherry picking with respect to time frames and variables.
EROI is one of the key concepts in this post yet not examined or questioned by either the OP or readers/commenters.
Putting a couple of data series on graphs and suggesting causality without any solid logical basis. Having two data series trending in the same direction is meaningless without solid reasoning behind it.
No review of peer reviewed research of any of the underlying concepts or data.

The very basis of this piece is BAU rather than out of the box thinking, which is what society so desperately needs

Sad, but it makes sense.
Rgds wp

Seriously? You think this is the reason TOD is closing down? Please refer me to a better site.

This seems to be a discussion, which is in the sub heading of The Oil Drum

I quite like to see the debate and it helps when people argue. We are all blind to some of our own thinking and seeing this kind of argument helps jolt these things into the open

Funny. I thought I had commented about the EROEI concept at 5:05pm.

Guess not, though, since you said there were no such comments at 11:20pm.

Maybe we should be checking out the logs in our eyes, eh?


you're right, there was/were some discussions about it, but there is still a complete lack of thorough analysis of that concept, and other ones. The structure of TOD makes it hard to really flesh out an issue in depth over a period of time, but obviously there are many advantages to the current format.

Although clearly there are some really, really bright people on TOD who comment they are the ones who quietly step aside when poor articles come along so what happens is that the bad drives out the good.

I often don't comment on comments/statements are either false or completely void of significance. Mainly that is because when I do rebut/comment/point out particular issues with an argument or line of "reasoning" there is resounding silence. Also, unlike some on this site, like ELM Jeff I don't have the stamina to keep pointing out the same thing again and again. Sadly, although theoretically repetition does not strengthen an argument in reality one often has to beat people over the head with the same message over and over (and over and over) again to get it to sink in. Good for Jeff (and Alan, and a number of others), I wish I had the stomach for it too.

(to some other poster): Clearly this is not the cause of TOD shutting down but it is a good example of a poorly reasoned post. As such it is indicative of the lack of good quality articles that Leanan (and others?) have referred to. I am glad too see that some commenters are applying some solid critical thinking skills.


I have to agree. I think bristle has struck that nail hard.

Gail, I've been a reader here for years, but this is my first time to comment. I have been in awe of your site, including the commenters. I've copied this article onto a file and stored it so it will be handy as friends and family go through the accusational stage of the hysteria that is collapse. Thank you for all these years of a wonderfully informational site. I will miss you.

"Given the diversity of what is needed to support the current economy, the small increment between 3 and 5 is probably not enough–the minimum ratio probably needs to be much higher."

I don't see where this was addressed - but thinking the difference between 3:1 and 5:1 is small is a serious blunder. When you get down into the lower numbers the difference in returned energy is huge.

I can't seem to find where the original is but that's the "net energy cliff" graph that Euan made - an excellent visual representation of EROEI in context to the return.

With renewables made using fossil fuels, such as hydroelectric, wind turbines, solar PV, and ethanol, the only way anyone can calculate EROI factors is as add-ons to our current fossil fuel system. These renewables depend on the fossil fuel system for their initial manufacture, for their maintenance, and for the upkeep of all the systems that allow the economy to function. There is no way that these fuels can power the whole system, based on what we know today, within the next hundred years. Thus, any EROI factor is misleading if viewed as the possibility what might happen if these fuels were to attempt to operate on a stand-alone basis. The system simply wouldn’t work–it would collapse.

I LOVE the "Just a Fossil Fuel Extender" arguments. If the approach being taken that the current system needs to keep running then the JaFFE argument will slap you in the face because the instant you start that argument you've already lost.

If you agree that you get more energy out than is put in you've just conceded that it's worth it to do it.

The great overarching idea is that if fossil fuel is the only thing that works for modern civilization, and wind turbines, PV, wave, etc, are "just fossil fuel extenders" then shouldn't we be extending the hell out of fossil fuel? The alternative to fossil fuel extension is to just run out quickly and collapse.

I'm all for this argument because as alternatives are built out - wind and PV in particular - we're likely to find that they're perfectly adequate for keeping modernity intact and we'll find ourselves saying "Why did we need fossil fuels?"

I'm all for this argument because as alternatives are built out - wind and PV in particular - we're likely to find that they're perfectly adequate for keeping modernity intact and we'll find ourselves saying "Why did we need fossil fuels?"

Will 7 billion and rising say "Why did we need fossil fuels?" or will modern civilization be reserved for a select few? Maybe we can continue up to 12-15 billion and build more "wind and PV", we need to run out of FF fast so we can get on with the job of "modernizing" the world. We can then all have electric cars and trains, eat jelly blubbers and cockroaches.

I usually don't say anything until someone pushes the button but no - the Earth is incapable of sustaining even 7 billion people. It can be seen all around - aquifer depletion, fish populations in massive decline, coral destruction, soil mining, habitat destruction. 7 billion is too destructive, even for lower impact things like PV and electric transportation. A population of half that size might be able to live well if they do it purposefully low impact. Somewhere around 2 - 3 billion could likely continue to advance science and have a low enough impact to live "sustainably" on Earth.

I think 2-3 billion is optimistic for a world with large fauna and intact ecosystems.
It would be a sad impoverished place.

A parable on EROI

A sower went out to sow, and as he sowed some fell by the wayside and it was trodden down and the birds of the air came and devoured it, and some fell on stony ground, and when the sun came it was scorched for it had no depth of root and it withered away, and some fell among thorns and the thorns choked it and it bore no fruit and some fell on good ground and yielded fruit that sprang up and increased and brought forth thirty-fold. and sixty-fold, and an hundred-fold. And he said to them that had ears to hear,

What is the EROI of this sower?

And some said 100, for it is recorded that some seed brought forth 100-fold.
And some said a googol, for a sower starts sowing at 20 lives for three score years and 10, and each of those 50 years returns an hundredfold.
And some said 2, for half of the harvest is taken in tax to power the rest of society and after the seed is sown the remainder is a bare sufficiency to feed the sower and his family until the next harvest.
And a scribe calculated it thus
50 seeds sown, of which the birds devoured 10, 10 withered in stony ground, 10 were choked by thorns, 15 brought forth 450, 4 brought 240, and 1 brought 100. The EROI is thus 790/50 = 15.8


Two scribes from the same school are likely to make the same assumptions in analysis and come up with the same answer, but the same types of argument that lead to 1e100 and 2 are also used, particularly when the analyst has an axe to grind. An analyst that wants to show how good renewables are will use methods that spin out the time and spin up the EROI. An analyst that wants to show they are hopeless will look to charge every conceivable externality against the process.


Say solar PV has an EROI of 20, is it better than than this biomass which the scribe says has an EROI of 15.8? You don't know. If you had the detail on the energy flows for the solar PV you could make a judgement on whether solar PV was a better investment than biomass, but critical information on whether its a good investment is not included in the EROI.

How rapidly could it be scaled up? You know from the data on the sower that reducing the tax rate from 340 per sower to 290 per sower would allow it to be doubled every year (as long as land remained and there were PV salesmen available for redeployment to sowing) There's nothing in the EROI that allows this judgement to be made though. The necessary information has not been included and you have no idea of how rapidly the solar with EROI of 20 can be scaled up.

EROI is a rubbish tool for investment judging and an economist ought to be able to come up with something better. How do solar, wind, biomass, coal, nuclear etc stack up on a net present energy basis? I don't know, but thats the sort of thing I'd expect someone with economic expertise to use. Something that is robust to whether a recycle is treated as internal to the process or a matching input and output, and something that takes time seriously.

It's IRR.

One quote I very much disagree with.

Also, I don’t think that it is really feasible to create a new economic system, based on lower EROI resources, because today’s renewables are fossil-fuel based, and initially tend to add to fossil fuel use.

The solution is extraordinarily simple - divert consumption to investment. The energy and other resources dissipated today on consumption (usually quite unnecessary) could instead be devoted to long term renewable energy and efficiency investments.

In a relevant example, someone paid for their solar panels by forgoing two vacation trips (staying @ home) and from the energy savings from the first six or seven years of solar generation. Energy that could have been used on an overseas flight, and hotel & ground transportation, was "diverted" to solar panel production and installation. Plus a fraction of the future energy production (likely 50+ years) was borrowed from the future.

Even a few % of GDP wisely invested can make the difference.

The second solution is a bit more complex, but it is being done in Denmark, France and other nations, is a quick and dramatic drop in energy density in the economy. Denmark has reduced their per capita carbon emissions by -26.5% in just five years (2007-2012) and -14.8% for France.

The United States could achieve comparable results with better policies.

Best Hopes for Better Resource Allocation,


This concern about EROEI is highly unrealistic.

Wind is the most important renewable at the moment, and even Hall shows 18:1. 2nd, Hall's data is very old: wind is around 50:1, and solar (CSP and PV) is at least 20:1.

Once EROEI is above, say 10:1, it really becomes irrelevant - it's just not the basis for competitive decisions. It's useful for analyzing bio-fuels, which are below 5:1, but not for wind and solar vs coal or NG.

Solar costs and manufacturing energy inputs are dropping fast, and wind's E-ROI is probably around 50.

Cutler Cleveland's summary of the literature (posted in the next comment, to avoid the delay of moderation) showed that wind's E-ROI was around 19. If you study his sources, you'll see that that most of the studies are quite old. If you look at the turbines analyzed in those studies, you'll see that they were much smaller than those in use today - look at Figure 2, and read the discussion. If you study that chart, you'll see a very clear correlation between turbine size and E-ROI. Given the recent very large increase in turbine size, it's perfectly clear that Vestas' claim for a current E-ROI of around 50 is entirely credible.

Again, an E-ROI of 19 is more than enough. There isn't an important difference between an E-ROI of 20 and an E-ROI of 50. It's like miles per gallon: we're confused by the fact that we're dividing output into input, when we should be doing the reverse, and thinking in terms of net energy. An E-ROI of 20 means a net energy of 95%, while an E-ROI of 50 means a net energy of 98%: there really isn't a significant difference.

Cutler Cleveland's summary:

Have oil shocks caused recessions?

Correlation is not causation.

A vast array of commodity prices peaked at the height of the economic bubble, and crashed afterwards, oil included.

The high price of oil was more important than other commodities, because oil is a large percentage of the overall commodity basket, but it was not the most important factor to the Great Recession. A contributor, no doubt, but a bubble is a bubble, and has to pop sooner or later.

There is just no way world economies can keep from shrinking if the oil supply starts to shrink.

I just don't understand why oil is considered by some to be so magical.

Sure, increasing import bills are a problem, but please note that OPEC countries are working very hard to spend their new income - that will tend to raise their non-oil imports, and raise oil-importer exports.

Yes, the cost of capex for substitutes will slow the economy down, but that's not overwhelming or permanent.

Yes, an oil shock can cause FUD, which slows down capex, but that's not permanent either. At a certain point EVs will gear up, and people will spend their money on them. In fact, car sales may go up, as there's actually a reason to buy a new car.

The US (and most of the OECD) has plenty of electricity, and plenty of time to transition from fossil fuel sources of electricity to renewable sources (not that we shouldn't transition away from FF much more quickly to reduce CO2 emissions...).

Our current operational problem is liquid fuels, and there are plenty of good substitutes for liquid fuels: electric vehicles (and their variants: hybrids, plug-in hybrids, extended range EVs, etc); freight reducing fuel consumption by 2/3 by moving from trucks to diesel trains, and then electric trains; heat pumps; and for the small percentage of energy that's needed for long-distance transportation, synthetic liquid fuels will work just fine.

At some point all the EVs etc will have been bought, and we'll not be worrying about oil any more.

I just don't understand why oil is considered by some to be so magical.

I guess to the historical fact that it was/is "magical" in building the current industrial civilization? Your examples and optimistic visions are very nice, but do you realize how extremely US-centric they are? There is a world beyond the 300 million USA, you know? It would be great if you could supply some more data on the costs of transitioning the global economy and infrastructure off of fossil fuels, not just the tiny sectors of US manufacturing and US personal transport. Those are negligible in the global view.

Oil wasn't uniquely essential. Many other things built the industrial revolution.

Actually, it's this article that's US centric: the only data shown is for the US - why not use world data? We would see that world GDP is still growing, while crude oil is pretty flat.

Costs: renewables vary around the world, but they're much more widely distributed than FF. EVs are much cheaper to operate than ICE vehicles everywhere: that's why they outsell them in China (in the form of e-bikes).

Again, there is this puzzling assumption that oil can't be replaced, that it is somehow magically necessary for industrial/modern civilization. Oil has been cheap and convenient for the last 100 years, but the industrial revolution started without it, and modern civilization certainly will continue without it. The idea that oil is necessary is an argument against solutions to Climate Change, and an argument for "drill, baby, drill".

• 130 years ago, kerosene was needed for illumination, and then electric lighting made it obsolete. The whole oil industry was in trouble for a little while, until someone (Benz) came up the infernal combustion engine-powered horseless carriage. EVs were still better than these noisy, dirty contraptions, which were difficult and dangerous to start. Sadly, someone came up with the first step towards electrifying the ICE vehicle, the electric starter, and that managed to temporarily kill the EV.

Now, of course, oil has become more expensive than it's worth, what with it's various kinds of pollution, and it's enormous security and supply problems.

• 40 years ago oil was 20% of US electrical generation, and now it's less than .8%.

• 40 years ago many homes in the US were heated with heating oil - the number has fallen by 75% since then.

• US cars increased their MPG by 60% from about 1976 to about 1991.

• 50% of oil consumption is for personal transportation - this could be reduced by 60% by moving from the average US vehicle to something Prius-like. It could be reduced by 90% by going to something Volt-like. It could be reduced 100% by going to something Leaf-like. These are all cost effective, scalable, and here right now.

I personally prefer bikes and electric trains. But, hybrids, EREVs and EVs are cost effective, quickly scalable, and usable by almost everyone.

Sensible people won't move to a new home to reduce commuting fuel consumption. That would be far, far more expensive than replacing the car. It makes far more sense to buy an EV and amortize the premium over 10 years at a cost of about $1,000 per year (much less than their fuel savings), versus moving to a much higher cost environment (either higher rent or higher mortgage).

• As Alan Drake has shown, freight transportation can kick the oil-addiction habit relatively easily.

We don't need oil (or FF), and we should kick our addiction to it ASAP.

The only reason we haven't yet is the desperate resistance from the minority of workers and investors who would lose careers and investments if we made oil and other FFs obsolete.

All of the various kinds of EVs (hybrids, PHEVs and pure EVS) would be much farther advanced if it weren't for resistance from the automotive and oil industries. The first PHEV was demonstrated more than 100 years ago. Very large and reliable EREVs were developed 100 years ago in the form of diesel submarines. This isn't new stuff, and it would be far more useful and cheaper if we had started to really push them 40 years ago, when US oil fields clearly showed their limits.

Gas should be priced at European levels (say, around $7 per gallon), to reflect it's real costs. If it were, EVs in their various incarnations would be obviously cost effective, and consumers would have demanded them long ago.

Some might ask, what about our current debt problems?

Debt is a symbol, a marker - what matters is the underlying productive capability of our economy, which will be just fine. Could we screw up the management of our economy, and go into a depression? Sure. But it's not likely.

Don't these transitions take 50 years?

The transition from kerosene to electricity for illumination took roughly 30 years. The US transition away from oil-fired generation took very roughly 20 years. The transition away from home-heating oil was also faster than 50 years (though uneven).

The fast transition from steam to diesel locomotive engines is illustrative. There were a few diesel locomotives in use in the U.S. during World War II but steam dominated in 1945. However, the steam locomotives had been very heavily used during World War II, and they all wore out at approximately the same time the first few years after 1945. When steam locomotives wore out, they were invariably replaced by diesel in the mid 1940s. By 1949, almost all steam locomotives were gone. There were still some steam locos made in the late 40's, and they were still in service in the 50's but dwindling. The RR's also relegated the steamers to branch line and switcher use - replacing the most used lines with diesel first as you would expect. Cn rail retired its last steam engine in 1959.

Other, very slow transitions are not a good guide to the future. For instance, the transition from coal to oil could be very slow, because there was no pressure - it was a trade up, not a replacement of a scarce resource. Many transitions occurred because something new & better came along - but the older system was still available and worked just fine. Oil may become very expensive very fast and that would provide us an incentive to switch over much more quickly.

On the other hand, we can point to many energy transitions that were sideways or down. The early transition from wood to coal in the UK was a big step down: harder to find and transport, dirtier - a pain in every way. Coal's only virtue was it's abundance. The transition from EVs to ICEs took a while - only when ICEs started to electrify did they become competitive. And, of course, we hid the external costs of oil from consumers: freeways (built by "engine" Charley Wilson after he went from President of GM to Secretary of Defense), pollution, overseas wars, etc. I'd argue that ICEs were never better than EVs - they just appeared that way.

On the other hand, EVs are better right now. They have better driving performance (better acceleration, better handling), and lower total lifecycle costs.

Unfortunately, we have more than 50 years worth of things we can burn for electricity. Fortunately, it doesn't look like we will. For instance, coal consumption in the US dropped 9% last year, about half of that due to loss of market share.

The transition from heating with wood to heating with coal took a lot more than fifty years. Electrification of the U.S. from small beginnings in the late nineteenth century to finishing rural electrification during the Great Depression took at least forty years.

Sure. These involved an enormous amount of infrastructure. On the other hand, EV/EREV/HEVs are manufactured on the same assembly lines as ICE vehicles, and roughly 75% drivers in the US have access to an electrical plug where they park.

Alan Drake would tell you: We transformed transportation before, in just twenty years. From 1897 to 1916, over 500 cities, towns and villages built streetcar lines. In several richer rural areas, vast networks of interurban rail lines were built. This was a nation with very limited "advanced technology", a half rural, half urban population and 3% to 4% of the real GDP of today.

so what is powering the majority of the worlds electrical generation... coal. and its use is increasing world wide. yep. was there before oil.


Time to eliminate both oil and coal...

Nick, it’s a great post and timely. My thought process is very similar (at a high level anyway, we disagree on a lot of the detail) - peak oil is a technical, political and economic challenge that is there to be solved and for which some of the tools to solve it already exist. There is absolutely no need for it to be the end of civilisation as we know it.

That is not to deny the drama and history behind this transition which will bring about huge geo-political and social volatility while it is on-going, a process that has already started IMO in ways that are small and difficult to identify in most places except MENA where the Arab Spring is rather obvious.

Anyway, one piece of feedback, please take it constructively since I get the feeling you are making these same points elsewhere.

When you're make this comment:

"Gas should be priced at European levels (say, around $7 per gallon), to reflect its real costs. If it were, EVs in their various incarnations would be obviously cost effective, and consumers would have demanded them long ago."

I think you need to explain why a transiton to EVs hasn't happened yet in Europe despite the high gas prices. It may seem minor in relation to your overall argument but it rather attacks the strength of your EV argument which is pretty central to rest of it.

That's a great question.

First, EVs are now growing pretty fast in Europe, as car makers begin to hit diminishing returns on ICE efficiency. 2nd, Europe took a different path, and so they have a different set of choices. Keep in mind that the average European only uses 18% as much fuel as the US, per capita.

There are a number of factors:

1) A different capital cost to operating cost picture.

EVs and PHEVs trade a higher purchase price for lower fuel consumption.

In Europe, fuel prices are 2-3 times as high as in the US, but due to historical factors (shorter distances, higher fuel taxes due to the high % of imports), the average car in Europe uses about 1/3 as much fuel as one in the US, due in roughly equal parts to lower kilometers driven per vehicle, and lower fuel consumption per km. Further, European taxes on new cars are generally much higher in the US.

Thus, the economic case for EVs and PHEVs is worse in Europe, and the lack of EVs and PHEVs in Europe really doesn't add any useful information to the question of how competitive electric powertrains really are with oil in the US.

2) Pure EV's still can't compete on convenience with ICE vehicles. Even in Europe, fuel costs are only a part of driving costs, and the lower cost of an EV weren't quite worth the inconvenience. The logical transition from an ICE to an EV is the PHEV, which for some reason wasn't explored seriously until very recently when GM took that path. Now that GM is pursuing PHEV extremely seriously, they've got an Opel version for Europe.

3) Europeans have fewer garages, as their housing is much older.

4) Tax preferenced diesel occupies the high-MPG niche.

and perhaps most importantly, there were large barriers to entry (billions in R&D and retooling, as well as resistance from ICE oriented manufacturers) for PHEV's, and there wasn't an obvious need for them. There was resistance from people in the industry who's careers would be hurt. This ranges from assembly line workers and roughnecks to automotive and chemical engineers.

In the US, the big low hanging fruit is personal transportation. In Europe, it's not personal cars, it's freight trucking.


If we mobilized all our resources as we did in World War II with the single objective of getting off fossil fuels as fast as possible, wouldn't the transition still take at least twenty years, and probably longer than that?

Some things much easier than that. A transition to EVs requires only a change within the automotive industry (for most drivers). Slashing coal consumption involves pretty straightforward ramping up of wind energy. 75% reductions in fuel consumption by road transportation and coal consumption for electrical generation would be ambitious, but doable.

But are we actually seeing any replacements of oil?

Consumption in the US has fallen by more than 15% since it's recent peak in 2007 (while GDP has risen by 3%), and it continues to fall. Production has risen (both C&C and all liquids), and net imports have fallen by 38% since their peak in 2005.

Didn't past transitions occur in a environment of growth, when making new investments was a good idea, and banks would lend?

The transition from horses to rail occurred mostly during the Long Depression from 1873-1890. The move from horses to tractors and automobiles continued at a very good speed during the depression, as did general electrification and business investment. The transition away from oil for electrical generation accelerated during the 1979-1981 recession(s), and CAFE standards rose.

Even at the depth of the Great Recession car sales were at least 60% of normal. Even with currently high oil prices car sales have recovered to about 14M per year, which is pretty strong. And finally, used cars were and are still turning over very 3 years, giving high-mileage/low income drivers an opportunity to switch to a more efficient vehicle.

Isn't this expensive?

EVs and their cousins (hybrids, plug-ins, EREVs, etc) don't require any more steel than ICE's, and they already have overall Total Cost of Ownership equal to or lower than ICE vehicles. We're making ICE's without a problem, and EVs aren't any harder. Wind turbines and solar panels really don't consume that much in the way of resources. Making long-haul trucks and coal plants prematurely obsolete is, of course, somewhat expensive, but the US has a big output gap (IOW, we have a lot of unemployed manufacturing and construction workers and empty manufacturing plants, waiting for something to do), and really, it would cost a lot less than another oil war.

Isn't "wasted" use of fuel is someones job providing a good or service? won't reducing fuel consumption cost jobs?

I'm thinking of the 50% of overall liquid fuel consumption that goes to personal transportation. That could be reduced easily without anyone losing their job.

Chevy Volts take as much labor to manufacture as vehicles that use 10x as much fuel. No problem there.

The average vehicle gets resold every 3 years: there's plenty of opportunity for higher mileage drivers to move to high MPG vehicles, even if they drive used.

Doesn't expanded rail mean wasteful & expensive extra handling?

Inter-modal container handling is well tested and is pretty efficient. More importantly, current distribution patterns were shaped under cheap oil. With higher oil prices the optimal mix of rail & truck has shifted sharply towards rail.

Alan Drake indicates that the clearest indicator of this is that Class I RRs are investing 18% of their GROSS revenues into capital projects. This is far higher than any other industry. The number of multi-modal transfer projects are exploding. Just 7 years ago, no Walmart distribution center was served by rail. Several new ones are. The number of factories and warehouses served by rail are expanding.

What about an emergency loss of oil supplies?

Carpooling works nicely: about 10% of all commuting is done via carpooling, more than mass transit and 3x as much as is done via commuter rail. Commuting is free, fast, and highly scalable, given that the average car only has about 1.15 passengers. Double that, and reduce overall fuel consumption by 25%. It could be done in weeks or months.

Isn't carpooling inconvenient and slow?

Yes, it's not an ideal long-term strategy. OTOH, it would work; it's bigger than bus & rail already; it's really cheap; it would eliminate congestion, which is why there are HOV lanes; and smart phones and modern telecom are making carpooling much easier.

The point is that we could reduce oil consumption very quickly, if we wanted to. If the alternative were really economic doom, carpooling wouldn't seem so bad, would it?

You also forgot that mostly all energy extraction requires uses subsidies from various sources as funding etc. These tend to be left out of EROIs. For instant low taxing, low gov leases on lands, low gov interest loans, massive gov infrastructure spending and investment to support extraction projects such as roads, railways, power lines etc.

So you see EROI is low ball by a very big amount. The feedback is also subtle on the infrastructure investment because obviously these require energy in the first place which is becoming more and more expensive with time.

Basically we had a party on easy fossil fuels because of the convenience of access to such a concentrated form of energy. We have failed to find a replacement and now we have also found out that nothing comes for free aka in this case we also get with it global warming and ocean acidification.

I also forgot a biggy. And that is the fact that the energy extraction and processing prices or costing currently does not reflect the cost of adequate pollution mitigation. So we get mines/well abandoned in such a state that tax payers usually have to foot the bill to right them and industrial accidents occurring which again much of the costs are not applied to the energy product but covered by tax payers. In the case of climate change we have a vague figure of around $120 / ton and rising for CO2.

so the EROI does not tend to take these into account. And I am sure there are some other ones by and by that we have missed.

Let me make the point above a little finer:

Businesses don't operate the in the way described above. Their first option when dealing with rising commodity prices isn't to lay off staff. Their first option is to use less of the commodity.

They might streamline their trucks, and them switch them to LNG. Their taxi's might move to hybrids. Their rail might electrify. Their plastic containers will be redesigned to remove mass, then be switched to glass or metal.

Companies don't take it lying down when oil/FF prices rise - they reduce their consumption, and if necessary they eventually eliminate it altogether.


Your argumentation was somewhat the same in your post on the 4th March 2011:

With the benefit of hindsight we can now answer your questions:

1. Is this really a new drilling technique?
No. But does that matter?

2. How likely is the 2 million barrels a day of new production, and the 20% increase in US production, by 2015?
Already achieved by 2013. In fact there has been an overall increase of 2 million barrels per day - an increase of 35%.

3. Can this additional oil supply really reduce the US’s imports by over half?
Yes. Easily. Imports are already reduced by 50% from peak and are still falling rapidly.

4. How much of a difference will this oil make to “peak oil”?
Oil will peak as energy demand switches to other sources. Many countries are already far down the transition.

My point being: Isn't your argumentation just plain wrong?