What is Oil?

Thu May 07 2026

What is Oil?

Ideas for a Better World

No. 65 · May 2026

A typical morning. The alarm sounds on a device housed in polycarbonate. Feet touch a carpet of polypropylene. The shower runs through PVC pipes. Toothpaste comes out of a polyethylene tube onto a nylon brush. The coffee is poured into a cup lined with polyethylene and sealed with a polystyrene lid. Breakfast was grown with nitrogen fertiliser synthesised from natural gas and protected in the field by polyethylene film. The paracetamol in the medicine cabinet was synthesised from benzene. The glasses on the face, the lens in the phone camera, the fleece on the chair, the paint on the walls, the adhesive holding the shoes together, the detergent in the laundry. None of it is burned. All of it came out of a barrel.

When most people say "oil", they mean fuel. When they say "the energy transition", they mean getting off fuel. Both are correct and both are incomplete. Oil is also, and increasingly, the molecular substrate of a much larger and quieter economy. This other economy does not appear in the climate conversation because it does not look like energy. It looks like objects.

The transition that is actually happening is a transition in how we burn oil. The harder transition, away from making things out of oil, has barely started. This piece is about why the second half matters, why it is harder than the first, and why any serious strategy for decarbonisation has to confront both.

What oil actually is

Crude oil is not a substance. It is a mixture. Formed over tens of millions of years as dead organisms were compressed under sediment and heat, crude is a dense soup of hydrocarbons: molecules made of carbon and hydrogen, running from very short chains that are gases at room temperature to very long ones that are solid waxes and tars. A single barrel can contain thousands of distinct compounds. Methane and ethane at the light end. Gasoline and kerosene fractions in the middle. Benzene, toluene and xylene in the aromatics. Long paraffins, asphaltenes and resins at the heavy end.

This molecular diversity is the central fact about oil and the central reason the substitution problem is hard. Crude is not one input looking for one replacement. It is a chemical feedstock library from which the modern industrial economy has built almost all of its synthetic materials. You can electrify an engine. You cannot electrify a polymer.

It is worth noting here that crude oil and natural gas are chemically adjacent. They are the same hydrocarbon family at different molecular weights, and the petrochemical industry draws from both. Much of what is commonly called the oil economy is, strictly speaking, the oil-and-gas economy. The argument that follows holds for the whole system, not just the liquid half.

How crude becomes everything

When crude enters a refinery, it is separated by boiling point in a fractional distillation column. The lightest fractions boil off first and are drawn from the top. The heaviest residues collect at the bottom. Between them lie the mid-weight fractions that dominate the fuels economy.

The rough yield is predictable. In a typical US-refined barrel, about 45% becomes motor gasoline. About 25% becomes diesel and heating oil. About 9% becomes jet fuel. About 4% is hydrocarbon gas liquids used directly as petrochemical feedstock. The remaining 15% or so is split across asphalt, lubricants, waxes, residual fuel oils, and further petrochemicals. Globally, petrochemical feedstock today accounts for around 14% of total oil demand, and that share is rising sharply.

What comes out of a barrel Typical yield from a 42-gallon US-refined barrel of crude DISTILLATION COLUMN Petrochem feedstock · ~4% Motor gasoline ~45% Jet fuel / kerosene · ~9% Diesel & heating oil ~25% Asphalt, wax, lubricants, heavy fuel, more petrochem · ~17% LIGHT HEAVY → Plastics, solvents, synthetic fibres, pharmaceuticals → Passenger cars, light trucks, small engines → Commercial aviation, military aircraft → Heavy trucks, freight, rail, shipping, heating → Roads, roofing, machinery, more petrochemicals Source: US EIA; proportions vary by crude type and refinery configuration

Fig. 01The fuel story is most of a barrel today. That is changing fast.

What the barrel breakdown obscures is that the molecules do not stay in their boxes. The lightest fractions feed a second system, the petrochemical system, which cracks and recombines them into a handful of platform chemicals: ethylene, propylene, benzene, toluene, xylene. From these five molecules, a large fraction of the modern synthetic world is built.

The fuel story, and its real limits

Begin with what is actually changing. The fuels story is the part of the transition that is genuinely underway.

The International Energy Agency's Oil 2025 report projects that electric vehicles will displace around 5.4 million barrels of oil a day by the end of the decade, against a current total of roughly 102 million. Global EV sales exceeded 17 million in 2024 and are expected to surpass 20 million in 2025, about a quarter of all cars sold. Gasoline demand in advanced economies is already declining. Oil demand from combustible fossil fuels, the IEA says, may peak as early as 2027.

So the good news is real. Within road transport, the problem is known to be solvable. Batteries work. Charging infrastructure is expanding. Policy has teeth. Substitution is happening at scale.

Then the limits appear. Aviation consumes around 7 to 8% of global oil demand. Maritime shipping another 7%. Together they move nearly all international trade and nearly all intercontinental passenger travel. Both depend on fuels with an energy density that batteries cannot yet match. Sustainable aviation fuels currently account for less than 0.1% of aviation fuel consumed. Even with aggressive growth, SAFs are projected to meet just 2 to 4% of jet fuel demand by 2030. Green shipping fuels are a similar story. Heavy trucks, long-haul freight and high-temperature industrial heat are no easier.

None of this is intractable. But these sectors have slower clocks, more expensive alternatives and harder chemistry. The fuels transition is proceeding unevenly: the easier half is racing ahead, the harder half is crawling. That unevenness is important, because it is nothing compared to what comes next.

You can electrify an engine. You cannot electrify a polymer.

The molecular story

If every combustion vehicle on Earth vanished tomorrow, global oil demand would fall by roughly half. The other half would continue to flow from the same wells. It would flow into materials. Here is where it goes.

Plastics. Around 400 million tonnes of plastic are produced globally each year. Packaging is 36% of it. Construction is 16%. Textiles are 15%. Consumer goods, automotive and electrical applications account for much of the rest. Plastic production has nearly doubled since 2000 and is forecast to double or triple again by 2050. More than 99% of the world's plastic is made from oil and gas.

Agriculture. About half of global food production depends on synthetic nitrogen fertiliser. Nearly half the nitrogen in human tissue today arrived through the Haber-Bosch process. Ammonia, the base molecule, is manufactured from hydrogen. Hydrogen is mostly manufactured from natural gas. Somewhere between 3% and 5% of the world's gas supply is consumed in making ammonia. The pesticides and herbicides that sit alongside the fertiliser are almost without exception petrochemical products. Modern agriculture is, in chemical terms, a downstream use of the hydrocarbon economy.

Medicine. Approximately 3% of petroleum production is used in pharmaceutical manufacturing. But nearly 99% of pharmaceutical feedstocks and reagents are derived from petrochemicals. The active ingredients in most synthetic medicines, from paracetamol to antibiotics to chemotherapy agents, are built from benzene, toluene, xylene and their derivatives. The packaging is plastic. The delivery systems, from IV bags to blister packs to gel capsules, are plastic. The mRNA vaccines rely on polyethylene glycol, a petrochemical, as a delivery vehicle.

Textiles. Synthetic fibres now account for 67% of global textile production. Polyester alone is 59%. Global fibre production reached 132 million tonnes in 2024, of which the overwhelming majority is made from petrochemicals. Recycled polyester is a small and slow-growing fraction, almost all of it derived from PET bottles rather than from textile-to-textile recycling, which remains below 1% of the market.

Construction. Asphalt is oil. Roofing membranes are oil. Insulation foams are oil. PVC pipes, polyethylene vapour barriers, epoxy adhesives, sealants, paints, solvents, coatings. The built environment is saturated with petrochemical products that quietly outlast every fuel tank of petrol the occupants put through their cars.

Electronics and infrastructure. Phone casings, laptop chassis, monitor housings, wire insulation, printed-circuit-board laminates. And, notably, the polymer components of solar panels, the resins in wind turbine blades and the separators and binders in lithium-ion batteries. The low-carbon infrastructure itself is built partly from fossil feedstocks.

This is the material world. The one that does not get rebuilt by installing charge points.

Why the conversation is mis-scoped

The IEA has been saying this quietly for years. Its 2018 report The Future of Petrochemicals was subtitled, correctly, as an examination of one of the "blind spots" in the global energy debate. Its 2025 Oil report is more direct: from 2026 onwards, petrochemicals will become the dominant source of global oil demand growth, ahead of trucks, aviation and shipping. Polymers and synthetic fibres alone will require 18.4 million barrels of oil a day by 2030, or more than one in every six barrels.

In the IEA's Net Zero Emissions scenario, the most ambitious of its long-range projections, 70% of remaining oil demand in 2050 is for petrochemicals and other non-combustion uses such as asphalt and bitumen. Even if the world decarbonises transport on schedule, the remaining oil demand is still substantial, and is dominated by the molecular economy rather than the combustion economy.

The substitution spectrum How far each use of oil has a workable replacement today READY AT SCALE NO SUBSTITUTE AT SCALE Passenger cars batteries, hybrids Power generation renewables, gas Rail freight electrification Heavy trucking early, costly Aviation SAFs <0.1% today Shipping pilot scale Nitrogen fertiliser green NH₃ early Petrochemicals no substitute at scale Where the policy attention is. Electrification, EV mandates, renewables deployment, charging infrastructure. Where most of the growth is. Petrochemical demand, fertiliser-dependent food, plastics, synthetic fibres. The gap between the two is the mis-scoping.

Fig. 02The conversation is focused on the left of this scale. The demand growth is on the right.

The public conversation has not caught up. Transport electrification gets the headlines, the subsidies and the venture capital. Petrochemical decarbonisation gets thin policy, small budgets and speculative pilot plants. Bio-based feedstocks exist but are expensive and land-intensive. Chemical recycling exists but remains at small scale and is facing energy-balance problems. Green hydrogen for ammonia is making progress but remains two to three times more costly than natural-gas-based ammonia. Nothing in the materials substitution stack is close to where batteries are in the transport stack.

The easy half of the problem is getting most of the attention. The hard half is getting most of the growth.

Three things a serious conversation admits

First, the transition is not one transition. It is a portfolio of very different problems with very different physics, economics and timelines. Passenger vehicle electrification is nearly a solved problem in engineering terms and an execution problem in deployment terms. Petrochemical feedstock substitution is barely an engineering problem yet, because pathways at scale do not exist. Aggregating these into a single "net zero" narrative creates a false sense of uniform progress.

Second, reducing oil demand and replacing oil molecules are different projects. A solar panel reduces combustion. It does not reduce the need for the naphtha from which its backsheet is laminated, the ethylene vinyl acetate that encapsulates its cells, or the polyamide film that protects them. A serious transition conversation distinguishes between molecules that need to stop being burned and molecules that need to be made differently or substituted. Most current policy addresses the first. The second is where the gap sits.

Third, decarbonisation plans that do not grapple with materials are incomplete plans. Any sector whose products, inputs or logistics touch plastic, fertiliser or synthetic anything carries an oil exposure it rarely names. Consumer goods, food systems, healthcare, construction, textiles, electronics, clean-tech manufacturing itself: the list is most of the visible economy.

Six oil exposures most carbon accounting misses

These are the places where scope-3 reporting tends to bury the problem. Any serious transition strategy names them.

Plastic packaging across the inbound and outbound supply chain.Consumer goods · logistics · retail
Synthetic fibres in apparel, furnishings and technical fabrics.Fashion · interiors · industrial textiles
Nitrogen fertiliser embedded in every food input.Food & beverage · hospitality · restaurants
Petrochemical precursors in pharmaceuticals and medical plastics.Healthcare · life sciences · insurance
Asphalt, insulation, sealants and coatings in buildings and infrastructure.Real estate · construction · facilities
Polymers and resins in electronics and clean-tech hardware.Technology · renewables · automotive

§ § §

What follows

For investors, the fuels-only demand thesis understates the durability of oil demand. An oil major whose portfolio is tilted toward petrochemical integration looks materially different from one tilted toward light-vehicle fuel retail. Saudi Aramco's acquisition of SABIC and ExxonMobil's continued expansion of its chemicals business suggest the smart money has read the IEA report. A decarbonisation thesis built on forecast peak-oil dates for combustion is not the same as a thesis built on total oil demand, and the two are diverging.

For businesses, an honest carbon accounting exercise often reveals petrochemical exposures that scope-3 reporting tends to bury. Any company with a plastic supply chain, a fertiliser-dependent raw material, or a synthetic-fibre product line carries a decarbonisation problem that cannot be resolved by switching to renewable electricity. The interesting commercial work over the next decade will be in identifying and redesigning those exposures, well ahead of any regulation that forces the question.

For policymakers, carbon pricing and EV mandates address the half of the problem for which ready substitutes exist. The other half requires industrial policy for petrochemical decarbonisation: feedstock substitution targets, procurement standards for low-carbon materials, serious support for chemical recycling and bio-based chemistry at scale, and deliberate infrastructure planning for green hydrogen and green ammonia. There is currently no equivalent of the 2030 internal-combustion-engine phase-out for petrochemicals. There probably needs to be.

The short version

Oil is not, and never has been, mostly fuel. Oil is a molecular feedstock from which the modern material world has been built. The decarbonisation that is underway is focused on the fuels side, which is the easier half. The harder half, the one that touches nearly everything we make and use, has scarcely begun.

The transition will happen. The question is whether the conversation catches up to what the transition actually is. Any honest account of where we are has to start there.