Download As PDF

How the pursuit of ‘energy dominance’ put the final nail in the coffin of electrification

The Honest Sorcerer

Drilling for oil, you might tempted to believe, will always pay for itself as it produces the fuels needed for its continued extraction. The more oil we lift, the more fuel (and energy) we have, right? Wrong. Let me explain. A barrel of oil would produce 5.9 Gigajoules of energy when fully burned, which might sound like a lot. Considering the fact, however, that only 27% of what we call ‘oil’ can be turned into diesel—the most vital fuel of all—on a global average¹, and that diesel engines burn fuel at an average 35% efficiency, only a tiny fraction of the energy stored in a barrel of oil can be used to drill more wells, mine minerals, harvest fish/wood/crops, build infrastructure, power militaries or to transport goods across large distances. That roughly 10% of energy derived from a barrel of crude is what keeps billions of people and a globalized world economy alive. The rest remains embedded in products (plastics, solvents, lubricants, asphalt, wax etc.) or simply get burned for our convenience in cars and jets. Unlike in the US it’s perfectly possible to live a decent life without a car or flying around with jets in many other parts of the world. The same could not be told about diesel, without which perpetuating modern, high-tech civilization would be impossible.

Trying to squeeze more diesel fuel out of a barrel of oil, on the other hand, would not only cannibalize the output of other fuels (mostly jet and fuel oil) but would also increase the energy cost of getting the raw power we need, thereby defeating the very idea of getting more useful energy out of a barrel of oil. Thus, when a study on Energy Return on Energy Invested (Delannoy et al., 2021) concluded that we use up a little more than 15% of energy stored in a barrel of oil to explore, drill, lift and deliver the next barrel of oil, they pointed out something profound. We need more energy to extract oil than what we could obtain by burning it in tractors, trailers, trucks and all kinds of heavy machinery to maintain civilization as it is. Again: no oil well was drilled using gasoline nor any mineral was extracted or fish caught with airliners—not in any meaningful quantity at least. What’s worse, as the authors found, the energy cost of oil is projected to reach 50% by 2050 as low cost conventional reserves continue to deplete, and get increasingly replaced by unconventional oil and energy intensive methods required to push what remains to the surface. That’s clearly not going to work on a global scale. But then how does the math work with even today’s numbers?

Simply put: we use the rest of the global energy system to subsidize liquid fuel production. Many wells and pumps are now running on electricity, and much of the energy used in refineries comes from natural gas. Sure, when taking all fuels together, producing oil is still net positive in energy terms but extracting petroleum long ceased to be a self-sustaining, self-perpetuating process. Half a century ago when the energy cost of oil was a mere 3% (or less), the 10% useful work derived from the diesel portion of a barrel was more than enough to power the oil business AND provide 7% “free” energy to the economy. Now its the other way around: we are using more and more electricity, natural gas, and in some cases even “renewables”, just to keep diesel flowing and the world economy ticking.

‘Then why don’t we shift to other fuels or electrify transportation?’—the question poses itself. You see, batteries and hydrogen, the most often touted “alternatives” to oil, are a way of storing energy at a loss, and not a source of energy. Thus, in order to replace oil we would not only need to dig up all the raw materials required to build these technologies—by using diesel fuel, what else?—but we would also need to multiply the electric grid’s capacity in order to cater for battery charging and hydrogen production demands. Even when considering engine inefficiencies, and calculating with the net energy portion of diesel, jet fuel and gasoline only² we would still need to produce at least 395 kilowatt-hours of power for each barrel of oil replaced. For the 103 million barrels humans burned every single day in 2025—that energy amounts to 40.7 Terawatt hours per day, or 14,850 TWh for an entire year… And we haven’t even taken electric engine and AC/DC conversion losses (10%) or battery recharge cycle inefficiencies (another 10%) into account—let alone the enormous energy cost of generating hydrogen… Not to speak of transmission losses over the grid which could be as high as 60% in the case of America or 40-50% everywhere else. All in all, as this back of an envelope calculation shows, we would need to generate around 36,600 TWhs of electricity on top of the existing 32,600 TWhs produced annually—just to electrify every machine which currently burns oil products globally. And then we haven’t even mentioned electrifying the industry which would take a similar amount of electricity. Triple that grid Mr. President, please.

In order to achieve our current grid expansion goals (which are far-far lower than we calculated above) the US alone would need to install 5,000 miles of new high-voltage lines annually, yet actual additions have collapsed from nearly 4,000 miles in 2013 to a historic low average of just 392 miles between 2022 and 2025. During the same time period China has doubled it’s electricity generation and now consumes nearly 33% of all electric power generated globally, while America’s share has fallen below 15%. Yet, expanding the grid and generation capacity is not even the biggest concern here, no matter how insurmountable a challenge it might seem in the West. It’s not even battery manufacturing capacity, much of which was going into electric vehicles already, and has surpassed 1.5 Terawatts of storage created annually. It’s the global scale of the change, combined with the rapid depletion of resources needed to make the shift, that will eventually put the energy transition into death.

“I have to finish this sweater before I run out of yarn!”

As a 2024 (but now removed) study by Rystad found: we have no more than a couple of years before we max out global oil supply, facing a long decline afterwards. And while green energy fans might rejoice, while envisioning a smooth ride into an electrified utopia, they have to be reminded that we are still mining (and thereby destroying) the planet with diesel fuel. And without mining, there is no aluminum, nickel, copper etc. needed to build batteries and to expand the electric grid with. Nor food… Or long distance transportation… Let alone construction. The coming peak in supply is not a matter of investment decisions either: we are no longer able to replace those oil reserves we have used up³ and now have to live off of our legacy fields—much of which have already passed their production peak. Making matters worse while these older larger fields deplete slowly at first, their depletion rate accelerates with time. (These results were later confirmed by the IEA as well.) The oil economy faces a double whammy from worsening energy returns on investment on one side, and an absolute depletion on the other; leading to a precipitous fall in net value delivered to society.

And it’s not just oil, but copper as well. Independent from what happens to oil, or whether we manage to electrify mining or not, the world is rapidly approaching an inflection point where mined copper supply begins to fall—irrespective of demand or investment decisions. So, even as global mined copper output reached a record 22.8 million tons in 2024, the IEA expects global supply to peak later this decade (at around 24 million tons) before falling noticeably to less than 19 million tons by 2035, as ore grades decline, reserves become depleted and mines are retired. Despite the potential contribution from African copper, new greenfield supply will struggle to make up the difference, as it takes 17 years on average till a mine starts production from discovery, and as new mines cost more and more to open. Simply put, we have run out of time, capital, reserves and energy to prevent a massive shortfall in copper production by 2030.

On the demand side, at the same time, achieving net-zero carbon emissions by 2050 would require a whopping 460% increase in copper production, which would necessitate the opening of 194 new large-scale mines over the next 32 years; tapping into imaginary reserves which are simply nowhere to be found. I think it’s not a terribly risky bet to say: it probably ain’t gonna happen… What you see here is what your ‘grandma running out of yarn before she finishes’ means in real life.