The Missing Barrel: Why Energy Infrastructure Is the Blind Spot in the Oil Shock

An OilWatch strategic report. This report is the infrastructure-level application of the Compound Cascade Systems Modelling framework — and several of the throughput indicators it calls for in Section 10 are already tracked live across the OilWatch network (see the cross-links in that section).

**A note on sourcing.** This report distinguishes between confirmed official figures, attributed market estimates and interpretive risk analysis. Where infrastructure-damage or outage figures are uncertain, they are presented as time-bound estimates rather than permanent capacity losses. All major factual claims were verified against primary sources in two source audits; a condensed sourcing note appears at the end, with the single residual caution (Baltic GNSS-interference counts) flagged there.

Executive summary

When conflict flares in the Gulf, the world asks one question: can the oil still flow? Analysts count tankers, watch the Strait of Hormuz, track Brent, and price the risk of a blockade. It is the right question to ask first. It is the wrong question to stop at.

Oil does not flow by itself. It flows through infrastructure — a long, interdependent machine of wells, pipelines, pumping stations, ports, tankers, insurers, refineries, gas systems, power grids, control software, storage tanks and distribution networks. Crude that reaches open water is still many steps away from being usable energy. The world does not consume crude oil. It consumes diesel, gasoline, jet fuel, LPG, heating oil, lubricants and petrochemical feedstocks — products that exist only after crude has been refined, powered, transported, insured, stored and delivered as the right grade, in the right place, at the right time.

This report argues that the central risk in modern oil security has shifted. It is no longer primarily a question of whether a single volume of crude can be supplied. It is a question of system throughput — whether the whole chain can keep moving usable fuel to the economy.

The next oil shock, on this argument, may not arrive as a clean shortage of crude. It may arrive as diesel scarcity, jet-fuel stress, refinery outages, port congestion, a gas-supply squeeze, a grid failure near a refinery, or a cyberattack on a pipeline operator. It may be a compound infrastructure failure: crude available but not refined; gas available but not liquefied; products refined but not distributed; tankers loaded but uninsured; refineries operable but without power; inventories present but trapped in the wrong region.

The market counts barrels. Societies depend on usable fuels. The gap between those two things is the missing barrel — and it is the blind spot in most oil-shock commentary.

Key findings

• Hormuz is real but partial. Around 20 million barrels per day of oil — roughly one-fifth of global petroleum liquids consumption and more than a quarter of seaborne oil trade — transited the Strait of Hormuz in 2024 (EIA, 2025). It is the obvious pressure point. It is not the whole system.
• Bypass capacity cannot replace Hormuz. Combined spare pipeline capacity that avoids the Strait is roughly 2.6 million b/d on EIA's estimate, or 3.5–5.5 million b/d on the IEA's — a fraction of what transits (EIA; IEA). There is no bypass at all for LNG.
• The world runs on products, not crude. Even with crude flowing, shortages can appear first in diesel, jet fuel and feedstocks. Refining margins and middle-distillate cracks hit multi-year highs in late 2025 (IEA OMR Oct 2025; EIA).
• Refining capacity is concentrating and ageing in the West. Around 400,000 b/d of European refining capacity was slated to close in 2025, and roughly 30 European refineries have closed since 2000, with new capacity shifting east of Suez (Argus; IEA Oil 2025).
• The oil system runs on electricity — and the grid is straining. The IEA says the world must add or replace 80 million km of grid by 2040 and nearly double grid investment to over $600bn a year by 2030 (IEA, 2023).
• The system has digital chokepoints too. The 2021 Colonial Pipeline ransomware attack shut roughly 5,500 miles of pipeline carrying ~40–45% of US East Coast fuel for about six days (DOE/CESER; EIA).
• Weather already does what war threatens. Hurricane Ida shut in 96% of Gulf of Mexico crude production at its peak in 2021 (EIA).
• Gas failure becomes a food problem. Natural gas is the feedstock for over 70% of ammonia production; in 2022, high gas prices curtailed roughly 70% of EU ammonia capacity (IEA; Fertilizers Europe).

1. The market is asking the wrong question

When a missile lands near a Gulf terminal, the first question on every trading desk and news bulletin is familiar: will the war block the flow of oil?

It is an understandable instinct. But oil does not flow simply because a sea lane is open.

Markets focus on visible chokepoints because they are easy to track. You can count tankers on satellite. You can read freight rates and war-risk insurance premiums off a screen. You can watch Brent, inventories and headline production numbers tick by the second. These are liquid, measurable, tradeable signals, and they cluster around the parts of the system that are easiest to see — the open-water transit of crude.

That focus misses a deeper issue: the functioning of the whole delivery chain. A barrel that clears the Strait of Hormuz has not arrived anywhere useful. It still has to be received at a port, stored, moved by pipeline, refined into products that match local specifications, distributed by truck, rail, barge and product pipeline, and delivered to a petrol station, an airport, a power plant or a factory. Every one of those steps depends on physical assets, skilled people, electricity, digital control systems, insurance and legal compliance. Any of them can fail.

This is the argument of the report in a sentence: the next oil shock may not appear as a clean shortage of crude. It may appear as diesel scarcity, jet-fuel stress, refinery outages, port disruption, a gas shortage, a power failure, a cyber disruption or a distribution bottleneck. The crude can be there and the economy can still run short of fuel.

To see why, it helps to start with the chokepoint everyone already watches — and then keep walking down the chain that begins where the chokepoint ends.

2. The visible chokepoint: why Hormuz still matters

The Strait of Hormuz is the most important oil transit chokepoint in the world, and nothing in this report should be read as downplaying it.

In 2024, around 20 million barrels per day of crude oil, condensate and refined products passed through the Strait — equal to roughly one-fifth of global petroleum liquids consumption and more than a quarter of all seaborne oil trade (EIA, Today in Energy, June 2025). The International Energy Agency puts the 2025 figure at a similar ~20 mb/d, of which roughly 15 mb/d is crude and 5 mb/d products, and notes that crude alone represents about a third of global crude trade (IEA, Strait of Hormuz). OilWatch tracks the live transit picture on its Chokepoint Transit Monitor.

The Strait is also a gas chokepoint. About one-fifth of global LNG trade transited Hormuz in 2024, almost entirely from Qatar (around 9.3 billion cubic feet per day) plus a smaller volume from the UAE (EIA, June 2025). Qatar accounted for roughly one-fifth of global LNG exports in 2024 — the International Gas Union puts global LNG trade at 411.24 million tonnes that year, of which Qatar supplied 77.23 Mt, or 18.8%, making it the third-largest exporter behind the United States and Australia (IGU 2025 World LNG Report).

Crucially, these flows are heavily directional. Around 84% of Hormuz crude and 83% of its LNG went to Asia in 2024, with China, India, Japan and South Korea taking the bulk (EIA). A disruption at Hormuz is therefore, in the first instance, an Asian energy-security event: the importers most exposed are in Asia, and China and India in particular would feel a closure first. But oil is a globally priced commodity, so a shock that begins as an Asian supply problem transmits to every consumer through the price — which is precisely why a chokepoint thousands of miles away moves the cost of diesel in Europe or petrol in the United States.

What history shows: stress short of closure

Hormuz has never been fully closed, despite decades of threats. But the history of Gulf energy security is not a history of nothing happening — it is a history of repeated stress that fell short of closure and still moved markets. That pattern is the key to reading the risk correctly.

During the "Tanker War" of the 1980s, the closing years of the Iran–Iraq War, more than 400 ships were attacked in the Gulf (US Naval History and Heritage Command, cited via Lloyd's List). Yet oil largely kept flowing: one analysis found the attacks disrupted less than 2% of ships passing through the Gulf, and most attacked tankers were damaged rather than sunk (Strauss Center, University of Texas). What the Tanker War actually produced was not a supply cut-off but a transformation of the cost and risk of moving oil: war-risk insurance premiums soared, and the United States ended up reflagging Kuwaiti tankers and escorting them through the Gulf under "Operation Earnest Will" (1987–88). The lesson, often misremembered, is that the binding constraint was insurance, naval risk and convoying — the financial and logistical machinery of shipping — not the physical availability of crude.

The pattern repeated in 2019. In May, four tankers were sabotaged off Fujairah, just outside the Strait; in June, two more were attacked in the Gulf of Oman, and per-voyage war-risk premiums for a supertanker jumped from around $50,000 to at least $185,000 — nearly a fourfold rise — in the space of a month (Insurance Journal / Bloomberg, June 2019). In July, Iran's Revolutionary Guard seized the British-flagged Stena Impero in the Strait and held it for roughly ten weeks (Al Jazeera). None of these events closed Hormuz; all of them raised the cost and uncertainty of using it. (Attribution was contested throughout: the US blamed Iran for the tanker attacks, and Iran denied it.)

But the single most instructive event of that year did not happen at the Strait at all. On 14 September 2019, drones and missiles struck Saudi Aramco's Abqaiq processing facility and the Khurais field, knocking out around 5.7 million barrels per day — over half of Saudi output and roughly 5% of global supply — in a matter of minutes (US Congressional Research Service; EIA). Brent jumped about 20% at one point the next trading day — its largest intraday move since the 1991 Gulf War — and closed roughly 15% higher, up about $9 a barrel, the biggest single-day settlement rise in a decade (EIA). Aramco restored output within weeks, faster than many expected. The lesson for this report is exact: the most damaging blow to Gulf oil supply in years was not a blockade of the sea lane but a strike on a single piece of processing infrastructure onshore. The chokepoint stayed open; the machine behind it was hit. That is the blind spot this report is about, visible in miniature.

Why bypass routes cannot save the day

The standard reassurance is that Saudi Arabia and the UAE can route oil around the Strait. They can — but only a fraction of it.

• Saudi Arabia's East–West (Petroline) pipeline has a design capacity of about 5 million b/d, temporarily expanded toward 7 million b/d, running to the Red Sea at Yanbu (EIA).
• The UAE's Habshan–Fujairah pipeline can carry about 1.8 million b/d to the Gulf of Oman, bypassing the Strait (EIA).

But design capacity is not available spare capacity. The EIA estimates that only about 2.6 million b/d of combined bypass capacity could realistically be available; the IEA gives a wider range of roughly 3.5–5.5 million b/d, partly because it assumes a higher, untested Petroline ceiling (IEA). Either way, the gap is stark: against ~20 mb/d of normal throughput, usable bypass capacity covers well under a third, and possibly far less. And there is no pipeline alternative whatsoever for the LNG — Qatari and UAE gas must leave by sea, through the Strait.

The capacity gap is not the only limitation. Bypass pipelines have fixed endpoints — Petroline runs to Yanbu on the Red Sea, Habshan–Fujairah to the Gulf of Oman — so they can only relieve the producers physically connected to them. They do nothing for Iraqi, Kuwaiti or Qatari volumes, which have no such westward outlet and depend on Hormuz absolutely. Rerouting through the Red Sea also exposes cargoes to a second chokepoint, the Bab-el-Mandeb strait, which has itself seen sustained attacks on shipping. And a pipeline moves a particular slate of crude grades; it cannot magically deliver the specific grade a distant refinery is configured to run (Section 3). So even the spare capacity that exists may not be the right oil reaching the right plant.

So Hormuz matters enormously. But notice what the bypass debate already reveals: even the "solution" to a chokepoint is itself a question of infrastructure — pipeline capacity, pump stations, terminals at Yanbu and Fujairah, the grades the lines carry, and the tankers and insurers to lift the oil at the other end.

Even partial disruption is costly

A full closure of Hormuz is the tail risk, and a low-probability one: closing it would choke the very Asian customers — above all China — on whom the Gulf producers, and Iran in particular, depend for revenue. The far more probable scenario is partial, intermittent disruption — and, as the 2019 episodes showed, that is damaging in its own right and does not require a single barrel to be physically blocked. Delays force rerouting around longer voyages, tightening tanker availability and pushing up freight rates. War-risk insurance premiums spike, and some owners decline the route at any price. Naval escorts and convoying slow transits. Port congestion builds downstream as schedules break. Strategic uncertainty alone widens price volatility and raises the cost of hedging. (How long the system could absorb a sustained squeeze is the question behind the OilWatch Hormuz Inventory Runway.)

Market analysts make the same point in reverse: in a Gulf crisis short of closure, the impact tends to show up first as higher freight and insurance costs rather than an outright supply shock, unless tensions escalate into a formal and sustained closure (S&P Global Commodity Insights, June 2025). That is exactly the report's argument: the cost of oil security is set not only by whether crude is available, but by the state of the financial, logistical and physical machinery that moves it.

### Why this matters for insurance and freight markets War-risk insurance and tanker freight are the **earliest and most liquid signals** of chokepoint stress, and they move before a single barrel is blocked. They are leading indicators that headline oil prices tend to lag — yet markets often treat them as background noise. (OilWatch surfaces the Lloyd's Joint War Committee listings and premium readings on its War-Risk Watch panel at [/supply](/supply).) - **War-risk premiums.** Cover for transits through high-risk waters is priced per voyage and can multiply in days. In June 2019, the war-risk premium for a single supertanker voyage jumped from roughly $50,000 to at least $185,000 — nearly fourfold — after two tanker attacks ([Insurance Journal / Bloomberg](https://www.insurancejournal.com/news/international/2019/06/17/529537.htm)). In the 1980s Tanker War, insurance, not crude availability, was the binding constraint. The mechanism is the marine insurers' Joint War Committee "Listed Areas," which mark zones of enhanced risk where cover can be suspended or surcharged (Section 3). - **Freight rates.** When owners avoid or slow-steam around a risk zone, effective tanker supply tightens and rates spike. During the mid-2025 Gulf tensions, S&P Global Platts reported Gulf-to-China crude freight rising more than 80% ([S&P Global](https://www.spglobal.com/commodity-insights/en/news-research/latest-news/crude-oil/062325-factbox-middle-east-conflict-raises-shipping-risks-as-some-tankers-avoid-hormuz)). - **Why it matters.** A crude price can stay deceptively calm while insurance and freight quietly reprice the *delivered* cost of oil. An analyst — or a policymaker — watching only Brent is watching the slowest gauge on the dashboard.

In other words, Hormuz is the entrance to a larger machine. It is the obvious pressure point. But the rest of this report is about the machine behind it — because that is where the blind spot lies.

3. The hidden chain: from underground barrel to usable fuel

Crude oil passes through many layers before it becomes usable energy. A failure at any layer can trap value even when the rest of the chain is intact. It is worth walking the whole chain, because each link is a point of potential failure that headline barrel-counts ignore.

Upstream

Production begins with wells, offshore platforms, drilling rigs and the unglamorous systems that keep them running: gathering pipelines, produced-water handling, gas reinjection, field power supply, process chemicals, spare parts, specialist crews and scheduled maintenance windows.

Production is not "turning a tap". A modern field is a continuously managed system, and shutting it in — whether for a hurricane, an attack or a maintenance failure — is not symmetric with starting it back up. Restarting shut-in or damaged production can be slow, technically delicate and occasionally dangerous: reservoirs behave differently once flow stops, equipment must be inspected and recommissioned, and damaged facilities may need parts and crews that are themselves in short supply during a crisis. The lesson of every major outage is that the clock to restart is longer than the clock to shut down.

The asymmetry is worse offshore and in technically demanding fields. Deepwater platforms must be evacuated ahead of a storm and re-manned and re-pressurised afterwards; the Gulf of Mexico took weeks, not days, to recover full output after Hurricane Ida (Section 8). Much of the equipment that matters — large gas compressors, specialist turbines, subsea components, control modules — is built to order, with lead times measured in months or years, not held on a shelf. A single damaged item can therefore hold a field down long after the headline threat has passed. And many fields are not standalone wells but complex flow systems that depend on continuous gas reinjection to maintain reservoir pressure, on water handling and disposal, and on chemicals to prevent corrosion and hydrate formation. Lose the supporting systems and the wells themselves may be sound but unproducible.

There is also a human and contractual layer. Drilling rigs, specialist crews, well-intervention vessels and inspection teams are globally mobile and often booked far in advance. In a wide regional crisis, the same scarce crews and rigs are needed everywhere at once, and the queue for them lengthens precisely when speed matters most.

Midstream

Between the wellhead and the refinery sits the midstream: trunk pipelines, compressor stations (for gas), pumping stations (for liquids), storage tanks, metering systems, blending facilities, terminals, and the SCADA and digital control systems that run all of it.

The midstream is where oil and gas can be physically stranded. A compressor station offline, a pumping station without power, a metering failure, or a control-system shutdown can trap hydrocarbons that have already been produced. Production can exist, and demand can exist, and the two can still fail to connect because a single link in the middle has gone down. The Colonial Pipeline episode (Section 7) is the textbook case: nothing was destroyed underground, yet fuel stopped moving.

Pipelines are also less flexible than they look. A long-haul line is a scheduled, batched system that moves specific products in sequence; it cannot instantly reverse direction, switch grades or re-route around a damaged segment. Reversing a major pipeline's flow can take months of engineering and regulatory work — as Europe discovered when it tried to re-plumb gas and product flows after 2022. Many lines have few or no parallel alternatives, so a single rupture, fire or cyber shutdown creates a genuine bottleneck rather than an inconvenience.

Storage is the system's shock absorber, but it only helps where it physically sits. Tankage at the wrong end of a broken pipeline, or full of the wrong grade, does nothing for a shortage downstream. And the entire midstream now runs on supervisory control and data acquisition (SCADA) systems and digital metering — which means the midstream's vulnerabilities are no longer only mechanical. A control-system failure, accidental or malicious, can halt a physically intact pipeline in minutes (Section 7).

Maritime and export infrastructure

Getting oil onto the water depends on far more than open sea. It needs ports, berths, loading arms, harbour pilots, tugs and marine services. It needs tankers in the right place and the right size. And it needs an entire legal-financial layer: marine insurance and war-risk cover, vessel classification, and sanctions and compliance checks that determine whether a cargo can legally and practically move at all.

Shipping, in short, runs on legal, financial, human and physical systems — not only on water. When war-risk premiums spike or insurers withdraw, a sea lane can be physically open and commercially closed.

The financial layer is more concentrated, and more fragile, than the size of the tanker fleet suggests. Marine hull cover, protection-and-indemnity (P&I) cover and war-risk cover are provided by a relatively small set of insurers and mutual clubs. The marine insurers' Joint War Committee publishes "Listed Areas" judged to present enhanced war and related risks; insurers and clubs may treat such zones as Additional Premium areas, where standard cover can be suspended or supplemented with bespoke, short-term war-risk cover priced per voyage. The Committee does not itself set the price of a transit, but its listings help define where that enhanced-risk pricing applies. A handful of classification societies certify that vessels are seaworthy; without class, a ship struggles to trade or to be insured at all. Sanctions and compliance screening add another gate: a cargo can be physically loadable but legally un-movable if its origin, ownership or destination falls foul of restrictions.

Two trends have made this layer more brittle. First, the rise of a "shadow fleet" of older, opaquely owned, often under-insured tankers carrying sanctioned crude has pushed a meaningful share of seaborne oil outside the mainstream insurance and classification system — raising the risk of accidents and spills that no reputable insurer stands behind. Second, the human element is thinner than it appears: harbour pilots, tug crews and specialist marine personnel are limited in number at any given port, and a labour dispute, evacuation or security incident at a major terminal can stop loadings even when ships and berths are available.

Downstream

Then comes the layer the market most often takes for granted: the refinery. Crude oil is not gasoline, diesel or jet fuel. It becomes those products only after passing through crude distillation units, hydrocrackers, cokers, reformers and desulphurisation units — a complex, tightly integrated plant that depends on hydrogen supply, natural gas, cooling water, electricity, continuous maintenance, skilled workers and specialised spare parts.

Refineries are also configured for particular crude grades. A plant built to run heavy, sour crude — with cokers and deep conversion units to break down dense molecules — cannot simply switch to light, sweet crude and produce the same slate at the same rate, and a simple plant built for light crude cannot suddenly absorb heavy barrels. So a crude shock is not just about volume; it is about whether the right crude reaches a refinery configured to process it. This is why the loss of a particular grade — a specific heavy or sour stream — can hurt even when total crude supply looks adequate: the barrels that remain may not match the plants that need feeding.

A refinery is also not one machine but a tightly coupled chain of them — distillation, then conversion (hydrocrackers, cokers, catalytic crackers), then treating (desulphurisation) and blending. A unit taken down for maintenance or by a fault can constrain the whole plant, because the units feed one another. Major maintenance "turnarounds" are planned years ahead, last weeks, and require specialist contractor crews and long-lead spare parts; deferring a turnaround to keep running during a price spike raises the risk of an unplanned, larger outage later. And like upstream production, a refinery does not restart instantly: bringing a complex plant back to full, on-spec output after a trip or a storm is measured in days and weeks, not hours.

Finally, restarting after a shock is itself a sequencing problem. A refinery needs power, cooling water, hydrogen and gas before it can run; if those are disrupted (Sections 5 and 6), the plant cannot recover even if crude is waiting at the gate. The result is that a refining system stripped of spare capacity has no margin to absorb the loss of a single large plant — and, as the next paragraphs show, the consuming West has been steadily stripping out exactly that margin.

Final distribution

Even finished products are not yet delivered energy. They must move through product pipelines, depots, rail, barges and truck fleets to reach airports, petrol stations, military allocation systems and emergency services. This is the layer closest to the public — and often the first place a shortage becomes visible. Panic-buying, queues at the pump and rationing typically appear at the distribution layer well before there is any literal absence of crude.

The final mile is deceptively fragile. Fuel often reaches its market in the right total quantity but the wrong specification: gasoline and diesel are blended to regional and seasonal standards, so a surplus of one grade in one region does not necessarily relieve a shortage of another grade elsewhere. Delivery depends on tanker-truck fleets and the drivers to crew them — a chronic shortage in several countries — and on functioning payment, dispatch and terminal-automation systems. A depot that cannot authorise loadings, a card-payment outage at the forecourt, or a driver shortfall can empty pumps in a region that is, on paper, well supplied. The UK's 2021 fuel-station crisis is a clean illustration: national fuel stocks at refineries and terminals remained adequate, but a forecourt delivery problem — a shortage of tanker drivers, compounded by localised demand spikes and panic-buying — emptied pumps anyway. The government put military tanker drivers on standby and authorised temporary visas for foreign drivers. The barrels existed; the means to deliver them did not.

The chain, then, is only as strong as its weakest link — and most of those links are invisible to the barrel-counting that dominates oil-shock commentary.

4. The refinery blind spot

Here is the single most under-appreciated fact in oil-shock analysis: the world does not run on crude oil. It runs on refined products.

Even if crude flows resume after a disruption, societies can still face shortages of diesel, jet fuel, gasoline, LPG, heating oil, lubricants and petrochemical feedstocks. A crude shock and a product shock are different events, and the second can arrive without the first.

Diesel deserves special attention, because it is the workhorse fuel of the physical economy. It moves freight by road, rail and sea; it powers tractors, harvesters and irrigation pumps (Section 5); it runs construction and mining equipment and backup generators; and in many countries it heats homes. When diesel tightens, the cost of moving and making almost everything rises at once. The middle distillates — diesel, gasoil, jet fuel and heating oil — are produced from the same part of the barrel, so stress in one tends to spill into the others. This is why a "diesel crisis" can be more economically damaging than a gasoline price spike of the same size: gasoline is largely a consumer fuel, but diesel is an input to the whole supply chain.

There is also a petrochemical dimension that rarely features in oil-shock commentary. Refineries and associated crackers supply the naphtha, LPG and other feedstocks from which plastics, solvents, packaging, textiles and countless industrial inputs are made. A refining disruption is therefore not only an energy event but a manufacturing-input event, with effects that surface weeks later and far from the petrol station.

Refining is fragile for structural reasons:

• Refineries are large, complex, ageing assets that require near-constant maintenance and periodic multi-week turnarounds. A deferred or failed turnaround can take capacity offline unexpectedly.
• They depend on other systems to function at all — electricity, cooling water, natural gas and hydrogen. Nearly all commercially produced hydrogen in the US comes from steam-methane reforming of natural gas, and refineries use that hydrogen to remove sulphur and upgrade heavy feedstocks into clean fuels (EIA, Energy Explained). Lose reliable gas, hydrogen, water or power, and a refinery cannot turn crude into clean products at normal rates.
• They are geographically concentrated. In the US, the Gulf Coast (PADD 3) held about 54.6% of national operable crude distillation capacity as of 1 January 2025 — 10.06 million b/cd out of 18.42 million b/cd nationally (EIA) — and it is also the most hurricane-exposed region of the country (Section 8).
• They are configured for specific crude grades, so the available crude and the available refinery must match.

On top of this sits a structural shift: refining capacity is closing in the West and growing in the East. Around 400,000 b/d of European capacity — including Scotland's Grangemouth, which stopped processing crude in April 2025 — was slated to close in 2025, and roughly 30 European refineries have shut since 2000 (Argus, Dec 2024; Argus, April 2025). The UK consequences of that shrinking domestic base are the subject of a separate OilWatch report, The Fall of the UK?. US operable capacity stood at about 18.4 mb/d at the start of 2025 with no major additions (EIA, June 2025). The IEA projects net global additions of around 2.6 mb/d to 2030, but concentrated in Asia — especially China and India — with high-cost plants in Europe and on the US West Coast hardest hit (IEA, Oil 2025). The result is a refining system with less slack and less redundancy in the consuming West, and more distance between where crude is refined and where products are used.

The evidence that this matters is already on the tape. In late 2025, refining margins and middle-distillate cracks reached multi-year highs. The IEA reported that light-sweet refining margins hit two-year highs in Europe and 18-month highs on the US Gulf Coast and in Singapore in September 2025, "led by improved diesel and jet fuel cracks following the disruption to Russian refining and exports" (IEA, Oil Market Report, Oct 2025). The EIA reported product crack spreads above $1/gallon at multiple hubs in late 2025 — three-year highs — driven by refinery outages and trade restrictions (EIA, Dec 2025). Persistent attacks on Russian refineries cut Russian crude processing, producing domestic fuel shortages and squeezing global diesel and jet markets as buyers scrambled for alternatives (IEA, Oct 2025) — a live case OilWatch examines in Russia's Fuel Shortage Is Becoming a Food-Logistics Warning.

Note what that last example shows: crude was not the binding constraint. Russia had crude. What it lost was refining throughput — and the shock propagated worldwide as a product shock. The same logic runs through the EU's 2023 ban on Russian seaborne diesel: the crude was still being pumped, but the product trade had to be re-plumbed around longer-haul routes, tightening diesel for everyone and lifting cracks.

The strategic worry is that the consuming West has been quietly removing its own buffer. Every closed European or US West Coast refinery is not just lost capacity but lost redundancy — fewer plants means each remaining one is more critical, and an unplanned outage at a large refinery has a bigger system-wide effect than it would have had twenty years ago. At the same time, shifting refining east of Suez lengthens the supply lines for finished products, so the West increasingly imports diesel and jet fuel over long distances from a smaller number of mega-refineries. That arrangement is efficient in calm conditions and brittle in a crisis: it concentrates risk in a few large assets and a few long sea routes — several of which pass back through the very chokepoints, such as Hormuz and the Red Sea, that the system was trying to hedge against.

The oil crisis may not appear first as "no crude". It may appear as no diesel, no jet fuel, no feedstock and no spare refining flexibility.

5. Gas is not separate from oil

It is tempting to treat oil and gas as separate markets with separate risks. For infrastructure security, that separation is artificial.

Gas is woven through the oil-products chain. LNG transits the same Strait of Hormuz as crude, with Qatar alone supplying roughly a fifth of global LNG (Section 2). Gas generates a large share of the electricity that runs pumps, ports, pipelines and refineries. Gas is the feedstock for the hydrogen that refineries use to desulphurise fuels. Gas drives compressors and supplies industrial heat. And gas is the basis of the petrochemical and fertiliser industries.

The key argument is simple: a gas disruption can become an oil-products disruption. If a refinery loses reliable gas, hydrogen or electricity, it cannot turn crude into clean fuels at normal rates — so a gas-supply problem shows up, one step later, as a diesel or jet-fuel problem. The systems are coupled, and a shock injected into one surfaces in the other.

This coupling also runs through the wider economy. Which brings us to two systems that conventional oil commentary rarely treats as part of "oil security" at all: electricity and food.

Why this matters for food and fertiliser

The gas–food link is direct and quantifiable. Natural gas is both the energy source and the feedstock for ammonia, the building block of nitrogen fertiliser. Just over 70% of ammonia production uses natural-gas-based steam reforming, and around 70% of ammonia goes into fertilisers (IEA, Ammonia Technology Roadmap). Because gas is such a large share of the cost, a gas-price spike can shut the industry down: in summer 2022, gas accounted for up to 90% of the variable production cost of European ammonia, and roughly 70% of EU ammonia capacity was curtailed or shut (Fertilizers Europe; Yara announced cuts to ~35% of European capacity, Aug 2022). OilWatch tracks the live urea/ammonia/gas readings behind this link on its Fertilizer Watch page.

The diesel–food link is just as real. Diesel powers the tractors, harvesters, irrigation pumps and freight fleets that plant, water, gather and move food. US agriculture remains heavily diesel-dependent: USDA ERS reports that diesel accounted for 44% of direct US farm energy consumption in 2016, while USDA/NASS's 2024 Farm Production Expenditures Summary shows US farms spent $15.4 billion on fuel in 2024, of which diesel was $9.9 billion — 64.3% (USDA ERS; USDA NASS, 2025). (The often-quoted "96%" figure applies to the heavy trucks that move agricultural commodities to market, not to farm equipment itself.) A diesel shock is therefore a food-cost shock with a lag. Energy infrastructure failure does not stay in the energy sector — it reaches the dinner table through fertiliser and through farm diesel. This fuel-to-food cascade is the subject of the companion OilWatch report From Hormuz to Hunger.

6. The electricity dependency

Here the report widens beyond standard oil commentary, because the oil-and-gas system runs on electricity — and the electricity system itself is under increasing strain.

Pumps, compressors, terminals, refineries, metering systems, pipelines, ports, control rooms and communications all depend on power. US policy treats energy and communications as uniquely critical precisely because they enable every other critical sector: virtually all other infrastructure depends on electricity to operate (CISA, Energy Sector). The Colonial Pipeline shutdown demonstrated the dependency in reverse: when the operator disconnected its control systems, "all pipeline operations" halted (US GAO, 2021).

It is worth being concrete about where the dependency bites. Liquid pipelines are driven by large electric pumps; gas pipelines by compressors, many of them electrically driven. Ports run on electrified cranes, pumps and control systems. Refineries draw electricity for the motors, pumps and compressors that move fluids through every unit, and for the instrumentation and control rooms that keep the plant within safe limits. Offshore platforms and onshore fields depend on generated power for pumping, separation and safety systems. Tank farms and terminals need power to gauge, pump and load. In almost every case, losing grid power does not merely slow operations — for safety reasons it often forces an orderly but complete shutdown, exactly as Colonial chose. And recovery is not instant: bringing this equipment back, and in some cases helping "black-start" a collapsed local grid, takes time and sequencing.

The dependency is also circular. The electricity system itself leans heavily on gas and oil: gas-fired plants are a major source of power in many grids, and oil products fuel the backup generators that data centres, hospitals and control rooms rely on when the grid fails. So a fuel disruption can become a power disruption, which then deepens the fuel disruption — the coupling that Winter Storm Uri exposed in Texas, where frozen gas wellheads and gathering lines cut gas supply to power plants at the very moment power was needed to keep the gas system and everything else running (Section 8).

Meanwhile the grid that all of this depends on is straining:

• It needs enormous, urgent investment. The IEA estimates that reaching national energy and climate goals means adding or refurbishing over 80 million kilometres of grids by 2040 — an amount it describes as equivalent to the entire existing global grid — and that grid investment must nearly double to over $600 billion a year by 2030 after a decade of stagnation (IEA, Electricity Grids and Secure Energy Transitions, 2023).
• It is congested. At least 3,000 GW of power projects, 1,500 GW of them well advanced, are stuck in grid connection queues worldwide — about five times the wind and solar added globally in 2022 (IEA, same report). Grid-related outages already cost around $100 billion a year, roughly 0.1% of global GDP (IEA, same report).
• Demand is surging. Global electricity demand rose 4.3% in 2024 and is forecast to keep growing near 4% a year to 2027, driven by industry, air-conditioning, electrification and data centres (IEA, Electricity 2025). Data-centre electricity use alone could roughly double — the IEA's Energy and AI report projects a rise from about 415 TWh in 2024 to around 945 TWh by 2030 (IEA, 2025).
• It is weather-exposed. In the US, around 80% of major power outages from 2000–2023 were weather-related, with roughly twice as many in 2014–2023 as in 2000–2009 (Climate Central, 2024). Heat reduces the carrying capacity of transmission lines and the efficiency of thermal plants; drought cuts hydropower and cooling water (IEA commentary).
• Its assets are ageing. Much of the grid in advanced economies was built decades ago and is now near or beyond its design life. As an illustrative US example, the EIA cites a 2015 DOE report finding that 70% of power transformers were 25 years or older, 60% of circuit breakers 30 years or older, and 70% of transmission lines 25 years or older (US DOE, Quadrennial Technology Review 2015). This should be read as historical US evidence of the ageing trend, not as a current global estimate. Older networks are more failure-prone and slower to repair.

There is a further twist: the new demand and the old grid are competing for the same scarce thing — capacity to connect. The surge in data centres, electrification and clean-energy manufacturing is arriving faster than transmission can be built, which is why thousands of gigawatts sit in connection queues. New industrial and energy infrastructure — including the electrified parts of the oil-and-gas system itself — may find that the binding constraint is not generation but a grid connection that takes years to obtain.

The oil system is increasingly dependent on an electricity system that is itself becoming more congested, more weather-exposed and more politically important.

This sets up a vicious loop. An oil disruption raises fuel costs. Higher fuel costs stress economies and power systems. Stressed, congested grids and high power prices reduce industrial resilience. And a fragile electricity system makes oil-and-gas recovery harder — because the pumps, refineries and terminals needed to recover all need power. The two systems can drag each other down.

7. Cyber, hybrid threats and digital chokepoints

The energy system has digital chokepoints as well as maritime ones — and they can be just as disruptive.

The clearest case study is the Colonial Pipeline ransomware attack of May 2021. On 7 May 2021, the operator of roughly 5,500 miles of pipeline carrying about 40–45% of the US East Coast's fuel shut the system down after a ransomware intrusion by the criminal group DarkSide (DOE/CESER; EIA, May 2021). Notably, the malware hit the company's business IT systems, but the operator proactively disconnected its operational control systems to contain it — and that precaution alone halted all pipeline operations (US GAO). The pipeline was down for roughly six days, triggering panic-buying and fuel shortages across the south-eastern US. Colonial paid a ransom of 75 bitcoin; the company's chief executive testified to a payment of about $4.4 million, and the US Department of Justice separately said it recovered 63.7 bitcoin — then valued at about $2.3 million — from the ransom proceeds (DOJ, June 2021).

The lesson is precise: a cyberattack created a real, physical fuel shortage without destroying a single physical asset. Nothing was bombed; a criminal group encrypted business data, the operator shut its own control systems as a precaution, and a fifth of a continent's fuel supply seized up for the better part of a week. Energy trading, scheduling, metering, payments and dispatch are now so thoroughly digitised that the boundary between an "IT" problem and a physical-supply problem has largely dissolved.

Western agencies have warned repeatedly about this exposure, and the nature of the warnings has shifted from criminal extortion toward state-level pre-positioning. In April 2022, a joint CISA/DOE/NSA/FBI advisory disclosed custom-built tools designed to scan for, compromise and control industrial control and SCADA devices once an attacker had reached the operational-technology network — capabilities aimed squarely at the systems that run pipelines, plants and grids (CISA AA22-103A). Then in February 2024, NSA, CISA, the FBI and partners said the state-linked actor known as Volt Typhoon had targeted the IT networks of communications, energy, transportation, and water and wastewater organisations, and that in some cases the actors had maintained access for years to pre-position for disruptive or destructive cyberattacks against operational-technology systems (CISA AA24-038A). The significance is the change of intent: this was not data theft or ransom, but the quiet establishment of footholds that could be activated to cause physical disruption at a moment of geopolitical choosing. The NSA also describes Volt Typhoon's use of "living off the land" — relying on legitimate built-in tools rather than malware — which makes such intrusions far harder to detect and means defenders cannot assume they would see an attack coming.

Europe's cybersecurity agency ENISA frames the underlying trend bluntly: "energy infrastructures are increasingly exposed to cyber threats, with the attack surface increasing due to the massive use of ICT" (ENISA). Every smart meter, remote sensor and digital control interface that makes the energy system more efficient also offers a new point of entry. The same digitisation that lets an operator run a pipeline from a control room hundreds of miles away lets an intruder, in principle, halt it from the other side of the world.

Beyond cyber sit broader hybrid threats — sabotage, undersea infrastructure attacks, GPS interference and drone strikes:

• Pipelines: The Nord Stream gas pipelines were severed by underwater explosions on 26 September 2022; Denmark, Germany and Sweden told the UN Security Council the damage was caused by "powerful explosions due to sabotage" (UNSC, 2023).
• Undersea cables and links: A string of Baltic incidents since 2023 has shown how exposed subsea infrastructure is — and how hard attribution is. In October 2023, the Balticconnector gas pipeline and a parallel telecoms cable between Finland and Estonia were damaged by a dragging anchor from the Chinese-owned vessel NewNew Polar Bear; the pipeline was out of service for around six months, resuming in April 2024, and China later acknowledged its ship had caused the damage, describing it as accidental. In December 2024, the Estlink 2 power cable between Finland and Estonia was severed by the anchor of the Eagle S, a tanker linked to the sanctioned-oil "shadow fleet" that dragged its anchor for roughly 90 km; transmission capacity on the link fell sharply, and Finland's grid operator returned it to service only in June 2025 at a repair cost estimated at €50–60 million (Fingrid). In November 2024, the C-Lion1 (Finland–Germany) and an East–West (Lithuania–Sweden) telecoms cable were cut, with a Chinese bulk carrier suspected. NATO launched a maritime patrol mission, "Baltic Sentry," in January 2025 in response. Caveat: attribution is genuinely unsettled. Several of these cases remain legally unresolved; the Estlink 2 charges were dismissed in late 2025 on jurisdictional grounds, and at least one 2025 cable cut (the Vezhen case) was ruled accidental by prosecutors. Only Nord Stream carries a formal multi-government sabotage finding. The pattern is striking, but the report does not treat every incident as proven sabotage.
• GPS and navigation interference: European aviation authorities have warned of a notable increase in GNSS jamming and spoofing since February 2022, including over the Baltic Sea, Black Sea, Middle East, Mediterranean and Arctic (EASA). Several states attribute the Baltic interference to Russia as hybrid pressure; Russia denies it. The wider point: navigation and timing signals are themselves a shared dependency for shipping, ports and energy logistics, and they can be degraded cheaply and deniably. (Specific incident counts circulating in the press — for example a "~46,000" regional figure — derive from open-source flight-tracking analysis rather than a primary aviation authority, and should be cited as such.)
• Drone and missile strikes: The war in Ukraine has seen sustained attacks on energy infrastructure on both sides — Russian strikes have repeatedly targeted Ukraine's grid and generation, while Ukrainian long-range drones have struck Russian refineries. At points in 2025, estimates relayed by the Atlantic Council suggested Ukrainian strikes had knocked out around one-tenth of Russian refining capacity, though later reporting indicated Russia limited the actual processing decline by drawing on spare capacity (Atlantic Council). Caveat: higher "38%" figures circulating refer to the nameplate capacity of targeted plants, not actual lost output, and should not be stated as fact. The strategic lesson is that refineries and grids are now treated as legitimate military targets — and that comparatively cheap drones can damage assets worth billions and take months to repair.

Nord Stream also marked a turning point in how Europe thinks about its infrastructure — and that shift in mindset has proved as consequential as the physical damage. The explosions did not, by themselves, cause the acute gas shortage of 2022: Russia had already throttled Nord Stream 1 to around a fifth of capacity over the summer and halted it entirely by late August, and Nord Stream 2 had never entered commercial service. What the blasts destroyed was the option of resumption, and with it the lingering assumption that critical energy infrastructure was safe simply because it lay deep underwater and was costly to attack. The recognition that pipelines, cables and terminals are now plausible targets has reshaped European policy on infrastructure protection, naval patrols ("Baltic Sentry") and supply diversification.

The incident also polarised opinion in a way that itself became a strategic factor. With no public attribution from the Danish and Swedish investigations — both closed in early 2024 without naming a culprit — competing narratives blaming Russia, Ukraine, Western states or unknown actors hardened into entrenched positions. That ambiguity is not incidental. Uncertain attribution is a feature of hybrid threats, not a flaw in the response to them: an attack that cannot be cleanly pinned on a perpetrator divides publics, complicates any collective response, and lets deterrence erode. The polarisation is part of the payload.

The energy system now has digital chokepoints as well as maritime ones.

8. Weather, climate and concentration risk

Infrastructure risk is not only caused by war. Weather already does, routinely, what a hostile state can only threaten.

The clearest example is Hurricane Ida (August–September 2021). At its peak, around 96% of US Gulf of Mexico crude oil production and 94% of natural gas production were shut in — roughly 1.7 million b/d of crude offline (EIA, Sept 2021; BSEE). Onshore, at least nine refineries were shut or curtailed, with about 75% of Louisiana's refining capacity (~2.2 million b/d) offline at one point (S&P Global). Crucially, the recovery was slow, not switch-like: in its final report more than three weeks after landfall, BSEE still recorded around 16% of Gulf oil production shut in and dozens of platforms evacuated (BSEE). One refinery — Phillips 66's Alliance plant — never reopened. This is the concentration problem in a single storm: the Gulf of Mexico provides about 15% of US crude production, and the Gulf Coast hosts more than half (about 55%) of US refining capacity (Section 4), so one hurricane track can hit production, refining and export at once. It is also the restart-asymmetry problem of Section 3 made concrete: shutting in took hours, but restoring full output took weeks, and some capacity never came back.

Winter Storm Uri (February 2021) is the cleanest illustration of compound, self-reinforcing failure — and of the oil–gas–power coupling described in Sections 5 and 6. Freezing temperatures across Texas froze wellheads and gathering lines, cutting state gas production by around 45% and driving the largest single-month decline in US natural gas output on record (EIA). That gas shortfall starved gas-fired power plants of fuel at the very moment heating demand peaked; the resulting power cuts then knocked out the electrically driven equipment that the gas system itself needs to keep producing and moving gas — a downward spiral. At least 4.5 million customers lost power, some for days, and the federal FERC/NERC review concluded that inadequate winterisation of both the gas and power systems was central to the cascade (FERC/NERC final report). Uri is the warning case: not one system failing, but gas and power dragging each other down together.

Drought and heat attack the system differently — slowly, and through the power side. In 2022, drought severely reduced European hydropower (Italy ran more than 30% below its 2017–21 average, France around 20% below its previous five-year average) and forced cooling-constrained cutbacks at some thermal and nuclear plants, because there was not enough cool river water to run them safely (S&P Global, 2022). Heat also directly degrades the grid: high temperatures cut the carrying capacity of transmission lines and the efficiency of thermal generation, so demand peaks for cooling exactly when supply capability dips. The trend is measurable. In the US, around 80% of major power outages from 2000 to 2023 were weather-related, with roughly twice as many in 2014–2023 as in 2000–2009 (Climate Central, 2024); analysis by Oak Ridge National Laboratory put the average annual cost of major US outages at roughly $67 billion over 2018–2024, reaching about $121 billion in 2024 alone (ORNL, 2026).

The deeper lesson is compound risk. Geopolitical conflict, extreme weather, grid stress, cyber vulnerability and ageing, concentrated infrastructure do not occur in isolation — they can interact and stack, as Uri showed when cold weather, gas freeze-offs and power loss reinforced one another. A heatwave that strains the grid while a hurricane threatens the Gulf while a cyber actor probes a pipeline is not a far-fetched scenario; it is the logical extension of trends already visible. And because these triggers are independent, the probability that at least one of them strikes the system in any given year is far higher than the probability of any single one.

Modern oil security is not just about hostile states. It is also about storms, floods, heat, grid collapse and infrastructure concentration.

9. The market's accounting problem

Why does commentary keep asking the narrow question? Because markets are very good at counting some things and structurally bad at counting others.

Markets price what is visible and liquid: crude production, OPEC spare capacity, tanker flows, inventories, futures prices and benchmark spreads. These are continuously quoted, standardised and tradeable.

Markets systematically undercount what is illiquid and operational: refinery operability, pipeline redundancy, port resilience, power dependency, cyber exposure, crew and pilot availability, deferred maintenance, crude-grade mismatch, product-specific shortages, regional distribution bottlenecks, restart time after an outage, and insurance and legal constraints. These are real, but they are hard to express as a single number on a screen — so they tend to be ignored until they break.

The consequence is that the market can misprice risk by treating every barrel as equal. A barrel is not a barrel. Its value to the economy depends entirely on whether the chain behind and ahead of it is intact.

A barrel in the wrong place, in the wrong grade, without refinery capacity, without power, without insurance or without distribution is not the same as a barrel available to the economy.

This is not a criticism of markets so much as a description of their blind spot. The signals that would reveal infrastructure stress earliest — product cracks, refinery outage reports, war-risk premiums, freight rates, grid congestion near energy assets — are often treated as secondary "noise" around the main event of crude prices. The argument of this report is that they are not secondary. They are frequently the leading indicators.

10. What policymakers and markets should measure instead

If barrels are the wrong unit, what is the right one? Throughput. The table below sets out a set of oil-system throughput indicators — signals that, taken together, describe whether the system can actually deliver usable fuel, not merely whether crude exists.

Oil-system throughput indicators

The point of such a dashboard is not to predict the next shock precisely. It is to stop watching only the sea lanes when the system can break in a dozen other places.

This is not a hypothetical wish-list — OilWatch already runs parts of it. The chokepoint, port and Oil Route Stress indicators are live on the Chokepoint Transit Monitor; the insurance layer is tracked through the War-Risk Watch panel on the same page; the gas-to-fertiliser link is monitored on Fertilizer Watch; refinery thermal anomalies are picked up via the Refinery Health Watch (NASA FIRMS); and the storage/runway question — how long the accessible buffer lasts under a sustained squeeze — is modelled in the Hormuz Inventory Runway. The throughput dashboard is the operating philosophy behind the whole network, set out here in full.

11. Strategic implications

For governments

Energy-security planning should include infrastructure resilience, not only reserves. Strategic petroleum reserves are valuable, but they are mostly crude — and crude is useless to a society without refineries, power and distribution to turn it into delivered fuel. Emergency planning should therefore model product-level shortages (diesel, jet, heating oil), not just crude shortages, and weigh whether to hold or contract for finished-product reserves and distribution capacity, not only crude in salt caverns. The IEA's emergency-response framework, which obliges member states to hold stocks equivalent to 90 days of net imports, is a genuine strength — but a stockpile only delivers security if it can be moved, refined and distributed under stress.

Two domains that have traditionally sat outside energy policy now belong inside it. Grid hardening becomes part of oil security: governments should map where refineries, terminals, ports and major pipelines depend on the grid, and treat those nodes as priorities for resilience, backup power and connection. Cyber defence becomes part of fuel security: after the Colonial attack, the US Transportation Security Administration issued Security Directive Pipeline-2021-01 (27 May 2021) — requiring covered pipeline operators to report cyber incidents to CISA, designate a 24/7 cyber coordinator and conduct vulnerability assessments — followed by Pipeline-2021-02 (19 July 2021) imposing further cybersecurity requirements, an attempt to make critical-infrastructure cyber standards mandatory rather than voluntary. Other jurisdictions face the same choice. More broadly, governments should treat domestic refining capacity as a strategic asset rather than a purely commercial one, and stress-test their systems against compound scenarios — a storm plus a cyber incident plus a chokepoint disruption — rather than single, isolated shocks. A government that holds 90 days of crude but has lost its refineries' power supply or its largest product pipeline to a cyberattack is not as secure as its reserve figure suggests.

For markets

Oil-price models should incorporate infrastructure risk, not treat it as residual noise. Product cracks and refinery-outage data may reveal system stress earlier than crude prices — as the late-2025 diesel and jet-fuel cracks did. Insurance and freight signals are not side issues; they are part of the price of moving a barrel and can move faster than the barrel itself. An analyst who watches only Brent and OPEC spare capacity is watching the most visible layer of a much larger machine.

In practice, that means treating refinery turnaround schedules, unplanned outages, war-risk premiums, freight rates and even electricity-grid stress near refining hubs as part of the core dataset rather than as colour. It also means resisting the instinct to read a single number — a crude price, a headline production figure — as a sufficient summary of risk. The market's habit of treating every barrel as fungible is precisely what makes it slow to price the missing barrel: the one that exists somewhere as crude but cannot reach the economy as usable fuel.

For businesses

Supply-chain planning should treat fuel availability as a distinct risk from fuel price. Firms — especially in logistics, agriculture, aviation and energy-intensive manufacturing — should monitor regional product shortages, not just headline crude. Backup power, on-site fuel storage, firm supply contracts and alternative logistics routes may become strategic assets rather than overheads. The question is no longer only "what will energy cost?" but "will the right fuel be physically available where we operate?"

The same logic applies to electricity. As grid connection becomes a binding constraint (Section 6), energy-intensive businesses should regard secured power capacity, on-site generation and demand flexibility as part of their resilience planning, not just their cost base. The firms that fared best in recent disruptions were generally those that had mapped their dependencies one or two layers deeper than their immediate suppliers — and knew, in advance, where the single points of failure lay.

For the public

The most counter-intuitive implication is the most important: petrol, diesel, heating-oil and jet-fuel shortages can occur even when crude oil still exists. Energy security is about delivery systems, not only production. The queues at the pump during the Colonial Pipeline outage were not caused by an absence of oil. They were caused by a failure of the machine that delivers it.

Why this matters for Europe and the UK

Europe and the UK are unusually exposed to exactly the risks this report describes. The region has been closing refineries for two decades — roughly 30 since 2000, with Grangemouth the latest — leaving fewer, larger, more critical plants and greater dependence on imported products that must travel further (Argus). The closure of Grangemouth leaves the UK with a shrinking domestic refining base and growing reliance on imported diesel and jet fuel, much of it arriving by sea from the Middle East, the US and Asia along the very routes a Gulf crisis would disrupt — the argument developed in The Fall of the UK?.

It is import-dependent for both gas and refined products, and the loss of Russian diesel after the 2023 ban forced a costly, longer-haul reorganisation of supply that left European diesel structurally tighter and more sensitive to any further outage. The gas dependency feeds directly into the food chain: Europe's heavy reliance on imported gas was what curtailed roughly 70% of its ammonia capacity in 2022 (Section 5), with knock-on effects on fertiliser prices and farm costs. Its undersea infrastructure in the Baltic and North Seas — gas pipelines, power interconnectors and data cables — has become a repeated hybrid-threat target (Section 7). And its grids face the same congestion and investment gap the IEA describes globally, with large connection queues in markets including Great Britain. For Europe and the UK, "oil security" that ignores refining, products, gas, grids and undersea cables is security in name only.

12. Conclusion: the next oil shock may be an infrastructure shock

The next oil shock will not necessarily look like 1973. It may not be a clean embargo, a single blockade or a single missing volume of crude. It may be a compound infrastructure failure — harder to see, harder to price and harder to fix.

The next oil shock may be crude available but not refined; gas available but not liquefied; products refined but not distributed; tankers loaded but uninsured; refineries operable but without power; inventories present but trapped in the wrong region. The world is still watching the sea lanes. It should also be watching the machinery behind them.

The barrel that clears the Strait of Hormuz is not the end of the story. It is the beginning of a long, fragile journey through pipelines, ports, insurers, refineries, gas systems, power grids, control software and delivery networks — any link of which can break. To count barrels and stop there is to mistake the entrance of the machine for the machine itself.

Oil security is no longer just a question of barrels. It is a question of throughput, resilience and delivery.

This report is the infrastructure-level companion to the broader OilWatch body of work: the Compound Cascade Systems Modelling framework and its companion Institutional Failure Mode Typology, which together explain how shocks propagate through coupled systems and why institutions so often fail to see them coming. The missing barrel is what that cascade looks like when it runs through the physical machinery of energy.

Appendix A — infrastructure layers and vulnerabilities

Sourcing and method. This report draws only on primary and named sources: EIA, IEA, IGU, USDA ERS, USDA NASS, US DOE/CESER, US GAO, US DOJ, US CRS, TSA, BSEE, FERC/NERC, CISA, NSA, EASA, ENISA, NATO, Fingrid, Climate Central, ORNL, the Strauss Center, Argus, S&P Global, Insurance Journal/Bloomberg and Fertilizers Europe — all linked inline. Infrastructure-damage and outage figures are presented as time-bound estimates, not permanent capacity losses. Three cautions are worth keeping in view: the Baltic GNSS-interference "~46,000" figure is an open-source flight-tracking estimate, not a primary aviation-authority count; the Abqaiq market move is split into a directly EIA-verified close (~15% / about $9, the largest single-day settlement rise in a decade) and a wider-reported ~20% intraday peak (Reuters wire), which are frequently conflated; and the figures for Ukrainian strikes on Russian refining describe the nameplate capacity of targeted plants, not confirmed lost output. This is strategic analysis, not financial advice.