Cassandra has moved. Ugo Bardi publishes now on a new site called "The Seneca Effect."

Tuesday, March 5, 2019

The real Energy Return of Crude Oil: smaller than you would have imagined


A simple but important study by Luciano Celi shows what is the real energy return that oil companies manage to attain. Much smaller than you would have believed, in several cases it is today well below 10. Which means that renewable energies already produce a larger EROI than oil and gas. No more excuses for not switching to renewables as fast as possible!And we have to do it fast because, as Celi shows, we are on the edge of the Net Energy Cliff of fossil fuels.

By Luciano Celi

Researchers involved in the energy sector know very well what the EROI (acronym of Energy Return on Investment) is and how frequently is cited in scientific publications, in spite of differences of definition among researchers.
Well aware of the fact that I arrived last in the middle of a dispute that has engaged researchers for decades, I accepted the challenge and tried to understand, with Claudio Della Volpe, Luca Pardi, and Stefano Siboni, if there was a way to know something that is difficult know: the EROI of oil companies.

The method we have implemented is quite simple, even if with some useful simplification that we have discussed in details in our paper.

The denominator in the EROI value is the Energy Invested, while the numerator is the Energy Return (how much energy is gained with respect to the investment). Knowing the numerator it is quite simple (we are talking of oil companies) because this value corresponds to the production of a day (or of a year). The issue in different cases was knowing the value of energy cost of that production. It is difficult to know how much energy they use to produce what they produce in a year (or in a day), sitting at a desk without wandering the world knocking on the doors of companies. However... we have found an indirect way to have these data.

The oil companies are requested to compile a Sustainability Report (SR) yearly. Even if not mandatory many companies have accepted to prepare them probably under the pressure of the public opinion or/and the business branch named CSR (Corporate Social Responsibility), especially after the alarms launched at the last world conferences on climate.

In these reports, it is possible to find the emissions of their up- and downstream activities expressed in CO2 equivalent. This is good news because we can try to do a mental experiment: to burn all oil that a company produces and convert it in CO2 equivalent, with a simple stoichiometric ratio. In this way, we have the equivalent value in CO2 of production (numerator) and the equivalent value in CO2 of emissions (denominator).
As already mentioned this is an approximation - first we have many kinds of oil, moreover the petroleum production almost always includes a quota of natural gas that has a different composition (mostly methane) and a different HHV (Higher Heating Value) compared to oil - but we have corrected our estimation taking into account these variables. The interesting side of the issue is that the data are above criticism because they come directly from the oil companies reports (Sustainability, Annual, Commercial, and so on).

The choice of companies

I focused my research activity on SR 2015 and looked at the companies with the highest revenues in the world. I have found this list (on or Wikipedia: both these sources - the second one takes the data from the previous one - are updated to 2017) and I have chosen to make a cut off for a market share above 1%. According to this criterium the companies selected are 30:

Billions $
Market share %

Saudi Aramco

China National Petroleum Co.


Exxon Mobil

Royal Dutch Shell

Kuwait Petroleum Corporation


Total SA



Valero Energy


Chevron Corporation



National Iranian Oil



China National Offshore Oil

Marathon Petroleum



JX Holdings



Indian Oil Corporation


Reliance Industries

< 1%


Total amount >>>>

Highlighted in green is found the total amount of the market share for the first five companies: this value is quite high considering that the first thirty companies represent almost the 90% of the market share, and only the 16,6% of them hold the 34,7% of the market share.

The EROI values

Applying the method briefly described above, the estimated values of EROI are represented in the graph below:

If we consider that the total contribution of oil and gas to the world primary energy demand in 2016 was 57.41% of the total (as shown in the graph below), the weighted mean of the EROI is rather low. Why does it seem that nothing happens?
The explanation is found in the relation between EROI and Net Energy.

The net energy and the Seneca Cliff

Net energy is defined as:

Net Energy (NE) = Energy Return - Energy Investment.

If we divide all terms for a single quantity ER, the result is:

NE/ER = 1-(1/EROI)

and, under the hypothesis of the ER always equal 100, the NE value can be expressed in percentage. 

So, the equation is:
NE(%)= [1-(1/EROI)]*100.

If we graph NE vs. EROEI the result is the following:

The industrial society experiences Net Energy, not EROI, decline. As shown in the graph above, there is a strong non-linear relation between Net Energy and EROI. For a long range of EROI values, say from 100 to 10, the Net Energy is declining very slowly. Presently although the EROI decline is quite clear the Net Energy is still well above 90%. The society feels pretty safe. The problem is that we are walking along a cliff and it is increasingly urgent to make an energy transition, before finally ending up in the abyss.


  1. May be that owners of Big Oil think that it is too early to react considering that net energy is still reasonably high. But what is perplexing to me is that they still don't see the climate and environmental costs like we limits-to growth believers do.

  2. Sorry I don't agree.

    You have to calculate EROI in joule not in cash: because of cash is not energy unit of measure.
    For mouving a TI-class supertanker, it needs very few oil barrels than it brings.
    So EROI in joule is not a problem for mankind.

    The problem is everybody in the world will burn oil for long time, because there is no sign of energy migration elsewhere.

    The problem is climate change and overpopulation.

    When climate change damages will raise up in the world, nations will need more and more fossil fuels for mitigate problems.

    For example:

    +1MT of sea level rising, it will mean that lot of coast cities will sunk: because people don't breath underwater, billions of people will have to mouve away, otherwise nations will need to install pumps for sucking away the sea water from cities. If people mouve away inland, it will need more oil for people transportation and for building new cities elseware.

    +1MT of sea level rising it will also mean more sea water infiltrate in the fresh waters inland into the natural aquifers.
    So lots of fresh water will transform iteslf in bad salty water, quite bad for thirsty people and agricolture need.
    Nations will build a lot of desalination plants, they will need for more electric power, so more fossil fuels for power plants.

    +1MT of sea level rising it will also mean a decreasing production in the food sector, so prices of food, fresh water, fertile lands will raise up.

    In any case, only for sea level rising the mankind gas serra emission will raise up.

    Well, I don't want to talk about methane hydrate bomb in Siberia, because is already melting, only God know how the things will turn from bad to worse, in the next decades. In any case, those stuff is preatty dangerous bomb, it is an uncontrallable for mankind, and it is also a key variable for the mankind destiny.

    Terraforming technologies are science fiction, nobody know how to build and how to work and how to produce green energy to feed them.

    So, for sure in the next decade there will be severe weather damages, mega hurricanes and giant typhoons, fresh water lacking in large part of the Earth continents, because you can't talk with termodynamic rules. Then deaths and wild fire in bad deseases will spread away, massive migrations and huge and violent regional wars in the cities with ethnic genocide. Wars will bring fear in the financial markets, commodities gold/silver/copper/food will prices will raise up.

    Mankind passed the oil peak more or less in 1980s-1990s now there's only hard oil to mine.
    There's lot of methane in Siberia.
    There's a lot, and when I say a lot, I mean really a lot of coal in the world: coal could mean synthetic fuels, those stuffs will feed transportation sector in the future.

    A preatty bad positive feedback will rule the world: while climate damages will increase, more fear and more regional wars and more fossil fuels mankind will fight and burn, and so on and on.

    As you know, when the commodities prices raise up, the energy stock in depots have more value, so net profits raise up.
    Frankly only financial wars could slow down the oil business market, because of (ROI-i)*leverage.
    Financial wars will bring cyber-wars, cyber-wars could escalate in interstate and symmetric wars.

    The world will stay at least in 1 or 3 scenarios, and all those 3 stuffs will drive mankind to WWIII in Siberia.

    EROI is not a problem, instead as you can see, it's a preatty nice highway to hell, don't you?

    1. Look at the chart again. EROI is CLEARLY important. It's just that its importance will only violently manifest itself once we reach the point it can no longer be hidden in rich countries. Once that exponential curve can no longer be smoothed over with debt or outright monetization of costs, the [S]ystem will willingly or not make it explicit that a leisure & consumption economy is permanently off the table.

      Before the 'event horizon' of the exponential curve, we could get away with just increasing global net energy extraction.
      But now we've reached the Era of Broken Promises...

    2. Quite so. This book by Dr Tim Morgan, a specialist from the oil industry should know, and supports your contentions:

    3. "The problem is climate change and overpopulation."

      Climate change is on part of the wholesale destruction of the biosphere. We are in the deep throws of the 6th mass extinction due to habitat loss, habitat that humans also need to live. The problem is that DILUTION IS NOT THE SOLUTION TO POLLUTION.

      Also is population really the issue when 10% of the population is responsible for over half of all pollution. Include that a large % of the pollution generated by the other 90% is generated by meeting the demands of the 10% it might be even more.

    4. >> The problem is climate change and overpopulation.

      Well, if you really want to separate the symptoms from the disease, then the heart of the problem is overpopulation. EVERYTHING that can be categorized as a problem for society, except the eventual death of our sun, is a symptom of, or is exacerbated by, our overpopulation. And we can't even have a dispassionate, logical conversation about the subject because someone in the room always throws the "racism" bomb into the discussion.

  3. The hilarious wings of tragedy ... !
    Our clueless indoctrinated political leaders and 'Captains of Industry' never saw it coming because the 'value' of fossil fuels appears NOWHERE on any balance sheet. No one paid a dime for their geological formation, like no one paid a dime for biological diversity and climate stability and so on...
    We pay rents to the land owners and then labor and industrial capital to extract the resources. Plowing heaven and hell for the last drops is good for GDP !

    If you can mange to view the thing from a detached enough perspective... there is a lot of humor in the situation.

    If my remark makes you uneasy, have a glass of wine'll pass ;-)

  4. So net energy is still above 90%, but this present society is well into overshoot. Where does that leave us?

    And....that graph of declining EROEI has been around for ages. What was it based on if it wasn't the CO2 equivalent of energy like yours?

  5. Ugo
    I do not pretend to fully understand the semantics of EROI and Net Energy discussions, but Celi et al it seems to me have usefully looked at CO2 emissions as a quantitative measure of energy entering the economy. They make their estimates for oil as a source of energy.

    I found a pie chart here for EU 2014 that divides up greenhouse gas emissions according to 'sector'.
    This would need some work to extract the CO2 'measure' in order to be comparable with Celi et al, but seems useful enough just now.
    The 'IPPC' sector 'Energy supply' seems interesting at 29.3% of total EU emissions. I assume this calculation includes coal burning for electric power where a lot of energy is rejected in the cooling towers and the process could appear much less efficient than energy required for crude oil extraction and refining.
    Nevertheless, this approach might help estimate the combined 'energy cost' for future quantities of energy entering 'the rest of the economy'? [Quote from Celi, above: "The industrial society experiences Net Energy"]

  6. 3 chines companies in the top 5 but no Russian ? A bit strange, guess it the lukoil rosfnet, gazprom split, but still

  7. When wear and tear take a car's engine at its end of life to the junkyard, there is still plenty of fuel available to run the engine, but the engine is mechanically and physically dead.

    EROEI is not how much you put into a device of useful work running it, but how much energy is expended in constructing the device.

    This is a big deal, indeed, as it leads to where the 2nd Law of Thermodynamics governs:

    “No Energy store holds enough Energy to extract and collect an equal amount of the Energy it stores” (The Fifth Law).

    "They told me that the only way shale oil could work is if the company went public so it could raise money from some poor unworthy slobs they didn’t know in order to fund an uneconomic business model".

  9. I fully agree that energy transition would be welcome but I have one simple question: where we'll find sound money to finance that transition? Take Italy for example. It has crippling debt huge enough to destroy Europe's banking sector. Who will borrow money for Italy's energy transition even if there is sound money somewhere, and there is none anywhere. Other Western countries are in no (much) better position. Only USA has exorbitant privilege to print worthless fiat money and use it to buy from the whole world. But for how long?

    1. Prof Michael Pettis

      "broadly speaking, debt can be divided into two types:

      Self-liquidating debt is used to fund investment projects that increase economic productivity enough (after including all associated positive and negative externalities) to service the debt fully. In such cases, an increase in debt is used to create an equal or greater increase in assets. While this usually leaves the overall economy better off, there could still be an argument about whether it is best to fund a particular project with debt (versus equity), about the best (or least risky) way of structuring the borrowing, and about how the debt and its subsequent repayment affects income distribution.

      All other debt funds household consumption, nonproductive government activities (such as military spending, welfare programs, and other kinds of consumption on behalf of households), and nonproductive investment by either the government or businesses. In some cases, this debt can have a positive impact on economic welfare, such as when debt is used to smooth out consumption over a person’s life cycle. In other cases, it can be positive or negative for economic well-being or for overall economic growth depending on how it affects the way income is distributed. (Indeed, this is one of its least understood but most important functions.)

      Self-liquidating debt adds to the total debt in the economy, but rather than heighten the economy’s debt burden it usually reduces the burden by increasing the wealth or productive capacity created by the project by more than the cost of the project. The most common form this debt takes is business investment or government investment in infrastructure. I say that this type of debt usually reduces a country’s debt burden, rather than saying it always does, because this may not be the case if the debt is badly structured; (if, for instance, debt servicing costs are severely mismatched relative to a project’s net increase in production), such a project can raise uncertainty in ways that adversely affect the rest of the economy."

    2. Even if energy transition is financed with the best kind of self-liquidating debt, a country that already has huge debt and low capacity to repay it, is risky for borrowers. For instance, Italy has first to repay existing debts in order to ask for new. Or, alternatively, creditors first have to cancel old debts as Michael Hudson suggest in his great economics writings, which we all should read. Theoretically, Italy could finance it's energy transition with it's own currency (which it does not have) but many things needed for energy transition are not produced in Italy and in the future many foreign producers could ask for gold, not for fiat money!

  10. Yes, population is part of the problem. Poor people getting richer is the main driver of new coal plants. Poor people having children drives forest clearing and extinctions. Rich people having fewer children would help enormously. Looking at the Kaya Identity, all four right hand side terms have to play a part or it doesn't work: C = C/E*E/Y*Y/P*P (C, emissions, E, energy, Y output, P, population. An expanded and quantifiable version of the I = PAT impact equation. Population is a key multiplier--all the output is for people.

  11. Once net energy starts to drop then it becomes obvious that we have a problem. But to see the problem before that, I agree with a concept I read about which proposes the metric "net energy per capita". Net energy could theoretically be rising or holding steady, and we would not see that the amount of energy available to each citizen is actually falling due to rising population. It seems a little wrong-headed to say that things are OK because energy available to a privileged few remains steady.

  12. The fastest way to population reduction - especially for those areas and countries with high birth rates are through industrialization, stamping out corruption, rule of law, pension systems all joined with the free enterprise system. Once productivity leaps the need for large families for productivity [and as a substitute pension plan] will drop like a rock.

    I'm 64 years old there actually was a time average people thought that Japan despite losing WWII would roll over the U.S. economically plus the attitude was something akin to breading like rabbits [their words not mine!]. Today of course nothing is further from the truth as Japan's population is actually declining. As long as the birth rate is under 2.1 population will decline [Japan today is at a scant 1.4].

    You may want to entertain that declining birth rates is due to declining energy flows [speaking generally] however one would like to measure. My point whether it is what we consider a modern society or declining energy flow does it really matter? Either way world wide population is moving in the direction to decline. My proposition is through modern economies [despite the energy usage] will get you there the fastest.

  13. OT: looks as if the US fracking industry does not run into an EROI problem but plainly into an old-fashioned financial ROI trap. Could you write something about this?

  14. Good bye Cantarell:

  15. Dear Mr Celi,
    have you thought of repeating the same calculations for a series of previous years so you could plot a trend in the evolution of the EROI?
    In other words, to estimate how fast we will reach the Cliff?



Ugo Bardi is a member of the Club of Rome, faculty member of the University of Florence, and the author of "Extracted" (Chelsea Green 2014), "The Seneca Effect" (Springer 2017), and Before the Collapse (Springer 2019)