Why Fossil Fuel Demand Can Rise Even as Clean Energy Wins Terrain

French historian Jean-Baptiste Fressoz has offered energy-transition skeptics an appealing historical argument.

In More and More and More, he argues that humanity has never really moved from one energy source to another. Wood was joined by coal, coal by oil, oil by gas and all of them by renewables. Energy sources accumulate rather than replace one another.

Global primary-energy charts make the argument look convincing. Fossil use remains near record levels after trillions in clean-energy investment. Biomass did not disappear when coal arrived, nor coal when the oil age began.

But the chart uses the wrong scale and watches the wrong indicator. Technologies compete in particular services and markets, not one global bucket labelled “energy.” Replacement appears first in sales, then installed equipment, fuel demand and finally the global aggregate.

By the time the last chart turns downward, the transition may already be irreversible.

Global Growth Can Hide Local Destruction

Michael Liebreich’s recent rebuttal of Fressoz makes the scale problem clear.

Many once-important energy systems did not merely lose market share; they disappeared from the services they formerly supplied. Whale oil stopped lighting homes. Town gas was replaced. Pit ponies, canal horses, district hydraulic power and wind-powered grinding mills largely vanished from their original markets.

Humanity still uses animals, biomass and manual labor. That does not make horses competitive with cars for urban transportation or whale oil with LEDs for lighting.

The same happens geographically with coal. Coal generation can decline in Europe and North America while rising in India or Southeast Asia. The global number may remain flat, but individual electricity markets are clearly replacing it.

In the European Union, wind and solar produced 30.1% of electricity in 2025, compared with 29% from all fossil sources combined. A decade earlier, fossil generation was almost three times wind and solar output. That is not addition. It is a crossover.

Sales Turn Before Fleets—and Fleets Turn Before Fuel

Transport demonstrates why transitions are so easily underestimated. The first mobile phone was demonstrated in 1973, yet global landlines rose until 2006. Nobody concludes mobile telephony failed because the incumbent grew for another 33 years.

The US horse population similarly peaked in 1920, more than three decades after the first practical combustion-engine cars appeared. New technology captured sales long before the old installed base disappeared.

Cars are repeating the pattern. Global sales of combustion-engine cars peaked in 2017. By 2024, they had fallen by nearly one-third, according to IEA data, while electric-car sales exceeded 17 million and continued growing. Yet most vehicles on the road still burn petrol or diesel because a car fleet turns over slowly.

This produces three different stories depending on the chosen metric. New sales show rapid substitution. The installed fleet shows gradual change. Global oil demand shows only the first signs of pressure.

All three can be true simultaneously.

Liebreich estimates that roughly 58 million EVs represented just under 4% of the global vehicle fleet by the end of 2025, but were already displacing slightly more than 2 million barrels of oil per day. The IEA expects displacement to reach around 5 million barrels per day by 2030 under current policies and about 9 million by 2035.

Oil will not disappear. Aviation, shipping, petrochemicals and the existing vehicle fleet will preserve substantial demand. But road transport—the largest source of oil consumption—is no longer simply adding electric vehicles on top. New combustion-car sales are being squeezed. The stock takes time to follow the flow.

Efficiency Makes Primary-Energy Charts Misleading

Fressoz’s aggregate also treats units of energy as though they deliver equal services. They do not.

An internal-combustion engine discards most of its fuel energy as heat. An electric motor converts a much larger share into motion. A heat pump can deliver three or four units of heat for each unit of electricity because it moves ambient heat rather than creating all of it through combustion.

Replacing a petrol car with an EV can therefore reduce primary-energy consumption even if both travel the same distance. Replacing a gas boiler with a heat pump can do the same while maintaining a warm home.

If the transition is measured mainly through primary energy, eliminating waste can look like lost energy rather than increased productivity.

The economy needs mobility, light, heat and industrial work. It does not need barrels, tonnes or cubic metres for their own sake.

This is why electrification is the transition’s central multiplier. Liebreich estimates that electricity supplied roughly 10% of useful-energy demand in 1970 and around 30% today. As transport, heating and parts of industry electrify, fewer units of primary energy are needed to provide the same or better service.

The global economy has grown by roughly 50% over the past 15 years while emissions increased only 9%. That is not sufficient absolute decarbonization, but it is substantial evidence that economic growth and fossil-energy growth are separating.

Additive Today Does Not Mean Additive Forever

Fressoz retains one important point: a rising clean-energy share does not guarantee fossil fuels decline quickly enough for climate goals.

Electricity is easier to decarbonize than aviation, cement, steel, shipping and food production. Renewables can dominate power additions while fossil fuels persist elsewhere. EV production still consumes energy and materials. Substitution is not environmental purity.

The transition can therefore be real and still be too slow.

That distinction matters. Declaring the transition fictional encourages governments and companies to preserve incumbent assets on the assumption that clean technologies will remain permanently additive. Declaring it inevitable encourages complacency about grids, permitting, finance and hard-to-abate sectors.

Neither conclusion follows from the evidence.

What matters is simple arithmetic. If clean energy grows faster than demand for energy services, it first meets all incremental demand. Once it grows faster still, fossil supply must contract. BloombergNEF estimates that wind and solar already provided 20% of global electricity by 2025, after annual installations rose from 300 GW in 2021 to 815 GW four years later.

The threshold between addition and replacement is not ideological. It is the moment clean-energy growth exceeds total demand growth.

The Transition Is Visible Where Competition Happens

Energy systems have enormous inertia. Power stations last for decades. Cars remain on roads long after manufacturers stop selling similar models. Buildings outlive their heating systems, while fuel infrastructure survives through sunk costs and political protection.

That is why incumbents can keep growing after their eventual replacement becomes apparent.

The useful question is not whether every fossil-fuel total has already declined. It is whether clean technologies are capturing new investment, sales and energy-service demand faster than incumbents can replace their aging assets.

In power and passenger vehicles, the answer is increasingly yes. In aviation, shipping and heavy industry, it is not yet.

The energy transition is neither instant nor imaginary. It is a long sequence of competitive defeats that appears additive—right until it does not.

This post appeared first on https://oilprice.com