The relentless — and ever accelerating — rise in energy demand of oil extraction threatens to upend centuries of economic growth; with or without wind, solar and nuclear. Together with a decline in conventional oil production it will eventually cap off the net energy returned to the economy, making any further expansion to our material world impossible. In fact, there is a good case to be made that we have already passed this point, and that the economic pains we endure at the moment are just a quiet prelude to the massive crash to come as a result.
Alligators are nasty creatures. They crawl into the muddy waters from far away — something which could only be seen by the most sharp eyed spectators. Then they sneak up on you, lurking under the surface, only to tip over the boat at the worst possible moment. And while everybody was partying onboard and enjoying themselves, those who kept warning their fellow travelers were duly pushed aside… Which makes me wonder what the narrative will be when the nose of the boat will suddenly point upwards, as the giant alligator — known as the energy demand of oil extraction — bursts out of the water.
But why talk peak net energy from oil? Haven’t we got alternatives already? All our energy production — without a single exception — depends on oil in general, and diesel fuel in particular. From wind and solar, or hydro and nuclear to biomass, all of our current ways of producing electricity, and any type of useful work also known as economic activity, ultimately depends on oil. So when net energy from oil peaks then starts to decline there will be less and less fuel remaining to build those wonderful technologies for the future, let alone bringing about real economic growth.
Wind towers for example are not only delivered by diesel trucks on site, and erected by diesel cranes, but their very material (steel) is produced from iron ore and coal, which is in turn mined and delivered to a smelter by heavy duty machinery… Fueled by diesel. The same goes to mining minerals for solar panels, making rebars and pouring concrete to build dams, nuclear reactors and wind tower foundations (among many, many other things). When it comes to energy production, and not only that, steel and reinforced concrete is everywhere.
No diesel, no mining, no metals, no construction. At least not beyond any scale what we had during the Middle Ages.
In case of oil extraction more and more drill pipe is needed per well year after year, because the boreholes are just getting longer and longer. That also means that ever more concrete, sand and water needs to be brought on site to cast those well casings (cementing the drill pipes in place), and to frack and flush wells. All this means increased traffic from trucks in the entire supply chain: from coal and iron ore mines to smelters, or from steel works making pipes and from sand mines loading fracking sand to drilling pads. And we are talking about no small quantities here: a thousand truckloads of pipe, sand and water delivered on site — not to mention the many truckloads of coal and iron ore finding their way to a smelter — or the untold amount of diesel burned by drilling and fracking equipment... And when we start to use electricity from “renewables” or nuclear to aid this process, all we will do is to just cannibalize those terawatts, i.e. diverting them from other economic uses. So yes, net energy from oil is extremely important.
In fact, there is a good case to be made that we are indeed past peak net energy from oil. Despite all the gains in “liquids” production we haven’t managed to surpass the latest peak in real crude oil extraction ever since November, 2018. Quite to the contrary: we have lost millions of barrels per day in conventional (easy to get) petroleum production, and have replaced it mostly with more fracked tight oil from the Permian, and heavy synthetic crude from Alberta (with a little help from offshore Guyana and Brazil).
I really don’t mean to rub salt in this wound, but it gets worse. You see, the situation with net energy returned from oil is far from being a static, one off problem to be solved by engineers. In reality energy demand is like an alligator lurking in the swamp. Engineers are working overtime to keep that animal well fed and under the water, but they are fighting against a predicament not a technical problem with a solution. Simply put: the issue sooner or later will catch up with them… And that’s when the gator will tip over the boat, throwing everyone onboard into the murky waters below.
Up until 2005 (when conventional oil peaked) petroleum extraction was on a slow but still exponential curve upwards. Everything looked fine: depleted wells could have been replaced with new easy to drill ones; no fracking, no 20000 feet long laterals needed. By drilling ever more holes into the same reservoir, oil production could be grown larger and larger. In 2005, however, as predicted by M. King Hubbert and confirmed by Colin J. Campbell and Jean H. Laherrère later, we run out of new easy to tap fields. It wasn’t the end of the world, just the end of exponential growth in oil production.
“It is important to realize that spending more money on oil exploration will not change this situation.”
Colin J. Campbell and Jean H. Laherrère
The realization, that oil production growth will not be able to keep up with the demands of economic growth, however, has kicked the biggest price rally this commodity has ever seen into motion. A rally, which has brought along many other commodities as well. What they all had in common that each and every one of them was produced by using oil products… Then in 2008 credit liberalization and a ‘let a thousand flowers bloom’ attitude met the physical reality of hard limits in a spectacular way. The financial economy of fictitious wealth, built entirely on the hope that infinite growth can go on forever with all its derivatives, futures, stocks and bonds, duly crashed in a matter of months. Despite all the claims to the contrary, the financial economy had just proved itself to be a derivative of the real economy of goods and services, which in turn was, and still is, entirely dependent on oil.
As a side-effect to the combination of high Brent prices and quantitative easing (as known as money printing elsewhere) fracking the source rock to produce more oil started to look like a really good idea. The many financial blunders aside, and the fact that this new extraction method failed to become economically viable for much of the 2010’s, there was another problem. Oil from shale could not possibly offset the rapidly worsening energy return from depleting conventional fields on a global average. Thus, for much of the previous decade, conventional oil production remained flat, fighting an uphill battle with depletion and increasing energy costs. The only major additions to world supply came from shale and tar sands mined in Canada; with an even more terrible energy return on investment than depleting traditional wells, and at an even greater cost of environmental destruction. The growth in energy return (or net energy) from oil on a world wide average has started to deteriorate markedly. The alligator has quietly crawled into the water.
Then in November 2018 that recent bout of production growth has stalled as well, and produced a peak in world petroleum production; not surpassed ever since. In terms of the peak oil debate, the question remained open: will we ever surpass this limit? Perhaps yes. But it will not matter. As conventional fields deplete they will continue to require ever more energy to maintain production: more wells will have to be drilled more frequently (with each well producing less than the previous one), and more CO2 will have to be pumped underground to force more oil onto the surface. As their depletion reaches a certain point, however, output will inevitably start to decline, no matter how hard we try to keep it from falling. These wells will have to be replaced with even more unconventional sources (shale, tar sands, ultra deep water), which also suffer from having an ever lower EROEI as sweet spots run out, and as even the best shale plays start to decline. As oilfield veteran David Messler wrote: “Shale well decline rates are accelerating, and production is expected to plateau or even decline soon.” The result? An even harsher (exponential) rise in energy demand needed to maintain remaining global oil production. The Journal of Petroleum Technology, the Society of Petroleum Engineers’ flagship magazine has published an article in 2023 saying just that:
“Energy necessary for the production of oil liquids is growing at an exponential rate, representing 15.5% of the energy production of oil liquids today and projected to reach a proportion equivalent to half of the gross energy output by 2050 (Delannoy et al. 2021).”
According to the study cited above: “by 2024, the production of oil liquids will require an amount of energy equal to 25% of its energy production.” That is one quarter of the energy produced in the world’s oil fields, which I find rather sobering. What is even more problematic, however, that these calculations use the total energy content of oil as a reference, not the net energy provided in the form of petroleum’s most economically useful constituent: diesel.
The problem is, that a significant portion of oil produced in the US for example distills into motor gasoline, which cannot be used to drive heavy machinery or turned into any other fuel. Another good chunk of a barrel of oil goes into making plastics, lubricants, asphalt and a countless other chemicals… So if you consider that only 12 gallons of diesel (corresponding to 1.66 million Btu) can made from a 42 gallon barrel (containing 5.8 million Btu), then you realize that only 29% of the energy contained in a barrel of oil can be actually used to power the economy of goods and services. And now, with oil extraction taking 25% of that barrel’s energy to maintain, the economic gain of adding an additional barrel of capacity shrinks to a mere 4% — entirely eaten up by refineries distilling oil into fuels.
So, purely in terms of mathematics, we are already in a dead state, “where the [global energy] system is in complete equilibrium with its surroundings, making it incapable of performing work” — i.e. no amount of surplus energy can be extracted from it. In case you were wondering why the economy appears to be at a standstill (or in an outright decline in some places), look no further for an answer. We are already producing just enough diesel to keep replacing existing infrastructure and lost oil and mineral extraction to depletion, with barely anything left to expand the production of commodities with.
To be fair, not all energy required to access the next barrel comes from diesel: electricity, coal and gas also plays a major role. Otherwise there would be no diesel fuel left available for sale, as all of it would be used up in drilling more wells, making the pipes and the necessary cement. However, we are already in a phase where petroleum engineers need all their ingenuity and engineering skills to save as much diesel for the rest of the economy as possible, turning to all sorts of other energy sources to power the extraction and the making of this fuel. In this light it doesn’t sound all that crazy to drive oilfield activities with nuclear power after all…
Unless you consider, that all such efforts are in vain. A single glimpse on that exponential chart above should inform any sane person that we are in a red queen race when it comes to fuel production. Soon neither traditional methods (drilling then distilling oil), nor synthetic or bio-fuel manufacturing, nor battery electric trucks will be able to save us, as all of these methods will require more energy to make than what they return to the economy; not to mention their inability to scale up to necessary levels.
Thus the question of peak oil production will soon become entirely moot. As soon as we start to consume more energy to produce this vital fuel (and no it really doesn’t matter how much cheaper gasoline gets in the meantime) it will become unimportant whether we produce 84, 100 or 200 million barrels a day. Eventually all of it will come at a net loss when it comes to diesel; a fuel powering all those trucks, excavators, dumpers, cement trucks etc. needed to maintain modernity. (This is not to mention the similar exponential rise in the actual cost of recovering oil; preventing petroleum companies from trying to exceed this net energy limit.)
From this point on diesel can be expected to be constantly in short supply — barring periods of contraction in the real economy of goods and services… Like the one experienced by the overdeveloped world at the moment. Going forward more and more non-petroleum energy sources will be cannibalized to aid oil production leaving less and less electricity, coal, natural gas, steel, cement etc for the rest of the economy… At least until the diesel demand required to keep producing these inputs starts to overwhelm supply. I’m not talking about this year or the next, perhaps not even this decade, but sooner or later something will got to give: either oil production, or alternative energy sources and other inputs. As the aggregate energy demand of maintaining the economy will slowly but steadily surpass what could be supported by the available diesel supply, though, the ship of over-financialized economies will turn belly up.
The alligator is already under the boat, rocking it from beneath. Some say, it’s just the waves. Others, who paid attention, know that something more sinister is at play.
The all too familiar question poses itself: can AI save us then? It depends. I mean, from what, and for how long? If your answer is saving us from an economic, societal and ultimately civilizational decline resulting from a peak and fall in net energy from oil, made all the worse by a wrecked climate and ecosystem, then the answer is a clear no. If the question relates to kicking the can down the road a few more years, then the answer is: definitely yes. (Presuming that AI doesn’t kill us all in the process, but that’s a different story for a different day.)
AI is a massively complex operation, with a huge energy and material footprint to match; adding further terawatts to the overall energy demand. Data centers’ electricity consumption in 2026, for example, is projected to reach 1,000 terawatts, roughly Japan’s total consumption. AI can give back some of that by increasing productivity in oil recovery, or finding new sweet spots on existing fields (by running through data previously collected by seismic sensors).
“AI and robotic solutions can help us create models that will predict behavior or outcomes more accurately, like improving rig safety, dispatching crews faster, and identifying systems failures even before they arise”
The one thing AI definitely cannot do is to refill depleted easy to access fields. In other words: while temporarily it can improve energy returns, on the long term geology has the upper hand. As petroleum geologist Art Berman observed, and proved it with data, most productivity gains come from an increase in oil production, and not from incremental advancements in technology. “That doesn’t mean that technology doesn’t matter — just that it’s probably less important than energy.”
Indeed, unless geologists are all wrong about predicting a decline in well productivity and net energy, there is no bright future for AI either. Chip manufacturing also requires complex supply chains, spanning the entire globe. It takes many gallons of diesel to produce and transport the raw materials and technologies required for their making, and no, solar power and sailboats won’t cut it. What AI does in this sense is increasing the complexity of the human enterprise even further, while still remaining prone to hitting diminishing returns (where productivity gains will be eventually exceeded by its ever increasing energy demand). And with electric cars — replacing frivolous gasoline use only, but not diesel — or millions of students wanting have AI to write their homework, we will see a continued increase in electricity demand. Something which will ultimately require even more mining, construction and transport; all performed by diesel engines. Catch 22, anyone?
“The world will face supply crunches in electricity and transformers next year… However much electricity you think you need, more than that is needed.”
Elon Musk
The relentless rise in energy demand to maintain oil production in general, and making diesel fuel in particular, is an alligator lurking in the muddy waters below. Engineers and geologists are all doing their best to keep it at bay, but they are running up against an exponential curve with no relief in sight. Absent of an energy miracle — an energy source wholly independent from petroleum, which we had no luck finding so far — the fate of this high tech civilization is sealed. Overshoot, and the many side effects of burning oil and other fossil fuels (including climate change and ecosystem collapse, together with the depletion of mineral resources), are making our situation even more dire. Technology proved to be the perfect monkey trap for our species, unleashing the beast of an exponentially growing energy demand. Something, which now threatens to turn economic stagnation into a steep decline, much like the great depression of the 1930's.
Feeding an alligator with an insatiable hunger never ought to have been considered a good idea — it would always come back for more. Instead, finding one’s way back ashore could’ve proven to be a much better plan…
But do we have a choice now?
Until next time,
B
Thank you for reading The Honest Sorcerer. If you would like to support my work, please subscribe for free and consider leaving a tip. Every donation helps, no matter how small. Thank you in advance!
