Iran Oil Supply Shock Is Tightening Global Crude Markets

I keep coming back to a number that should not be this quiet: a supplier that was still moving well over a million barrels a day into its only real customer has, for practical purposes, stopped showing up. Not in a press conference. On the water. If you trade crude, run a refinery, or simply pay for diesel, that absence is already someone else’s problem, and it is drifting toward yours.

Iranian barrels used to sit in the footnotes of the global balance. Useful, discounted, politically awkward, and easy to underestimate. That habit is getting expensive. China’s independent plants are hunting replacements just as their own runs start to recover, freight is punishing long-haul grades, and a narrow strait is carrying a surprisingly heavy flow of everyone else’s oil. I’ve found that markets rarely price a missing barrel until the spare one is already spoken for. We may be closer to that moment than the calm headlines suggest.

Why A Missing Supplier Suddenly Matters To Everyone

For years, sanctioned crude worked like a pressure valve. It did not set the official benchmark, yet it kept a slice of Asian refining alive at a discount. Remove the valve and the rest of the system has to breathe harder. That is the plain version of what is happening now.

After the political link that once moved Iranian cargoes toward Syria broke in late 2024, China became the only meaningful buyer. Through 2025 that flow averaged about 1.4 million barrels a day. Not a rounding error. A structural input for a particular kind of refinery, mostly the smaller independent plants clustered in Shandong, the ones traders still call teapots even when the hardware is anything but small.

Then the war that opened in late February changed the geometry. While other tankers were being held back from the Gulf, Iranian cargoes still moved, and Chinese intake climbed to roughly 1.76 million barrels a day in April. A strange competitive edge: the producer under fire was, for a few weeks, the one that could still deliver. That edge died with the blockade announced on 13 April. Loaded ships could not leave. Empty ships could not come in. Loadings at Kharg Island, the main export terminal, collapsed from about 1.8 million barrels a day in March to 260,000 in May.

A barrel that cannot leave the terminal is not a supply. It is inventory with a flag on it.

A memorandum on 17 June opened a 60-day window and loadings recovered, to about 740,000 barrels a day in June and 890,000 in July. The window closed. August slumped back toward 250,000. By September, trackers were not seeing fresh Iranian loadings in the Gulf at all. Perhaps the most interesting part is not the terminal. It is what happened outside the strait, where a floating cushion kept China supplied long after the tap had narrowed.

The Floating Cushion That Bought Time

In mid-April, Iran had something like 160 million barrels sitting on ships scattered through South, Southeast and East Asian waters. That stockpile is why May still looked almost normal in Chinese import data: about 1.37 million barrels a day of Iranian crude arrived, only about 10 percent below February. The ships were spending a buffer that had been built when the politics were ugly but the sea lanes were still usable.

Buffers shrink. By mid-June the floating pile was down near 106 million barrels. The temporary reopening pushed it back toward 128 million by mid-July. Then the replenishment stopped. China still took about 980,000 barrels a day in August. September fell to roughly 475,000, and arrivals effectively ceased from 26 September. The last cargoes to show up had been loaded back in June. Read that twice. The oil reaching port in early autumn was old oil.

What remains on the water is about 86 million barrels, the lowest since January 2025. More than a quarter of that, some 23 million barrels, is trapped inside the Gulf. The total has barely moved in recent weeks, which tells you the system is no longer circulating. Onshore tanks are filling too. Storage looks about 60 percent full, holding something near 70 million barrels. When both the sea and the shore stop absorbing crude, production is the only lever left.

Domestic refineries can still absorb roughly 2.2 million barrels a day. Pre-war output near 3.2 million does not look achievable while exports are blocked. In my experience, producers hate shutting in wells they may struggle to restart. They do it anyway when the alternative is oil with nowhere to go.

How The Export Window Opened And Closed

It helps to see the year as a sequence of doors, not a single policy. Each door changed who could lift a cargo and who had to wait.

• Late 2024 into 2025: China is effectively the sole crude buyer, averaging about 1.4 million barrels a day.
• April: intake jumps toward 1.76 million barrels a day while other Gulf traffic is constrained.
• 13 April blockade: Kharg loadings fall from 1.8 million barrels a day in March to 260,000 in May.
• 17 June memorandum: a 60-day passage for Iranian cargoes lifts June and July loadings.
• August onward: the reprieve ends, shipments slump, and September loadings in the Gulf go dark.

Short version? The market had a supplier, then a stockpile, then a brief restart, then silence. Silence at the loading buoy is what forces buyers onto the open market.

What The Numbers Look Like Side By Side

Tables are dull until the columns start arguing with each other. This one does.

Notice the gap between the strait and the terminal. Neighbors are moving a heavy stream. The producer at the center of the political fight is not. That imbalance is the story’s hinge.

Shandong’s Ports Are Already Feeling It

Teapots built a business model around discounted sanctioned barrels, Iranian and Russian above all. They account for something like a fifth of China’s crude imports. Lose the discount and you do not just pay more. You sometimes cannot find the molecule in the size and timing you need.

Qingdao is the clearest tell. A pipeline links the port to a dozen independent refineries. In 2025, Iran supplied about 40 percent of incoming flows near 690,000 barrels a day. Recent months brought more Brazilian Tupi and Buzios, and even Guyana’s Golden Arrow from July, alongside Saudi and Russian grades. Intake still sank to a record low around 150,000 barrels a day over the past three months. Buying different oil is not the same as buying enough oil.

Dongying, up on the Bohai coast near 32 independent plants, was even more concentrated. Last year Russia and Iran supplied virtually all of a 330,000 barrel-a-day intake, roughly two-thirds and one-third. Iranian deliveries faded through the summer: two cargoes in August, one in September. Total intake dropped to about 220,000 barrels a day in September as plants cut runs. Crude-deprived is not a metaphor when the crude unit is turned down.

In mid-September, ten independent refiners reportedly sent traders to Singapore to lock available barrels from the Middle East, West Africa and South America. That is what a scramble looks like when it still wears a suit. Guyanese crude, in particular, gets expensive fast when the voyage is long and very large crude carriers are scarce. Freight at record levels can erase a grade’s appeal before the ship leaves the berth.

Quotas Do Not Create Barrels

Beijing tried to give the independents room. An extra 28.05 million tonnes of crude import quota landed in late September, lifting the annual non-state allocation to a record 257 million tonnes. Quotas decide how much a refiner is allowed to import. They do not put a ship on the berth, and they do not cheapen a barrel that three other buyers also want.

I keep seeing this confusion in market notes. Permission is treated as supply. It isn’t. A quota is a license to compete. Right now the competition is ugly.

1. Russian ESPO has been bid to an all-time premium near 28 dollars a barrel versus ICE Brent, by some trading accounts.
2. Urals has traded 7 to 8 dollars above the same benchmark, a reversal of the old discount culture.
3. State-owned buyers now take roughly half of China’s seaborne crude imports, up from about 45 percent in February.
4. Independents are fighting both the state firms and each other for what used to be a private aisle of the market.

Premiums like that do not last forever. They last long enough to hurt margins, delay runs, and spill into product prices. Diesel and gasoline do not care which flag was on the crude.

China’s Recovery Is Still Early, Which Is The Awkward Part

Seaborne crude imports rose from 7.24 million barrels a day in August to 7.5 million in September. Better. Still a long way from February’s 11.5 million. From April through July, imports sat near half the pre-crisis pace, weighed down by a mandated ban on product exports, lower refinery runs, and a slow draw on strategic stocks.

Strategic reserves, state and private together, sit near 1.12 billion barrels, down from 1.25 billion in April. That draw bought time. Rebuilding commercial imports while Iranian supplies disappear means China has to pull harder on barrels that other buyers also need. Early-stage recoveries are when people get complacent. The volume is not back, so the stress looks optional. It isn’t optional if the missing supplier was the marginal cheap barrel.

Replacing a discounted barrel with a full-price barrel is not a one-for-one swap. It is a tax on every subsequent buyer.

Market desk observation, paraphrased

Every replacement cargo tightens the pool. West African grades, Brazilian subsalt, Middle East term volumes, Latin American spot: none of these were idle waiting for Shandong. They had homes. Shift them east and someone in Europe, India, or the US Gulf has to pay up or switch slate. That is how a regional sanctions story becomes a global crude story without a single new outage.

The Strait Is Busy, And That Is The Risk

Here is the part that keeps risk desks awake. Oil from neighboring producers is moving through Hormuz at about 13 million barrels a day, only around 5 million below the pre-crisis level. A relief for buyers. Also a vulnerability. As long as Iran cannot export, it has little economic reason to protect an arrangement that lets everyone else through.

Peace talks continue without a breakthrough. Export revenue is squeezed. Onshore tanks are filling. Floating oil inside the Gulf cannot leave. Financial pressure and a busy strait are a bad pair. No one needs a lecture on geography to see the leverage. A renewed disruption would not just remove Iranian barrels that are already gone. It would remove the replacements.

Would Tehran actually go further than it has? I don’t know, and anyone who claims certainty is selling something. What I do know is the incentive has shifted. When your own oil is trapped and your neighbor’s oil is sailing, patience is a cost, not a virtue. Markets have a habit of treating that cost as theoretical until a tanker turns around.

Two risks, one balance:
  China must replace Iranian oil as demand heals.
  Iran may lose patience with a strait open to everyone else.
  Either risk tightens supply. Together they reprice it.

Who Pays If The Cheap Barrel Stays Gone

Follow the money without the drama. Independent Chinese refiners pay first, through higher feedstock and lower runs. State firms pay next, because they are now competing for the same seaborne pool they used to leave partly to the teapots. Term buyers in Asia pay if Middle East producers have less spot flexibility. Atlantic Basin refiners pay if West African and Brazilian cargoes are pulled east. Freight owners get paid, which is nice for them and rotten for everyone lifting a long-haul barrel.

Consumers pay last and least visibly, which is why the story feels abstract until the pump or the factory invoice moves. A few dollars on the benchmark does not sound like a crisis. Stack freight, a quality mismatch, and a refinery that cannot run the new slate at full rate, and the product crack does the talking.

There’s another payer people skip: Iran itself. Shut-in production is lost revenue and, sometimes, lost reservoir performance. A country that cannot sell crude still has bills. That is not a moral point. It is a cash-flow point, and cash flow is what usually decides whether a standoff stays tidy.

What A Normal Month Would Have Looked Like

Imagine September without the blockade’s second act. Kharg would still be loading. Floaters outside the Gulf would be turning over, not stagnating. Qingdao would not be at a three-month low. Dongying would not be explaining a one-cargo month. ESPO would not be wearing a record premium. The extra import quota would be a growth tool, not a permission slip to hunt scarce oil.

That counterfactual matters because some commentary still treats Iranian crude as optional. It was optional for buyers who never touched it. It was not optional for the plants that designed their margins around it. When those plants step into the unsanctioned market, they do not add demand. They relocate it.

Recent flow tracking shows the relocation already happened in the data, even if the benchmark has not fully admitted it. Arrivals winding down to a halt over a fortnight is not a forecast. It is a receipt.

Freight, Quality, And The Hidden Cost Of Substitution

Substitution sounds clean in a spreadsheet. In a refinery it is messy. Iranian grades that teapots knew how to run are not identical to a Brazilian subsalt barrel or a Guyanese cargo that spent weeks at sea. Yields shift. Hydrogen balance shifts. The coker and the reformer do not care about your geopolitics.

Add a shortage of VLCCs and record freight, and the delivered cost can jump even when the flat price looks calm. I’ve watched desks celebrate a “flat” benchmark while their netback collapsed on shipping. This is that kind of tape. The barrel is available somewhere. The affordable barrel is not.

• Long-haul Latin American grades carry a freight penalty that can dwarf a small quality discount.
• Middle East alternatives may be tighter on term allocations once Chinese state buyers lean in.
• Russian barrels are no longer the easy backup; premiums say the aisle is crowded.
• West African cargoes redirected east leave a hole in other regions’ slates.

So the question is not only “where does China buy?” It is “whose barrel does China take, and what does that buyer do next?” Chains like that are how a Shandong throughput cut becomes a European diesel story two months later.

Storage As A Clock, Not A Comfort

People treat storage like a safety blanket. Sometimes it is. Right now it is a clock. Offshore Iranian oil at 86 million barrels, with 23 million stuck in the Gulf, is not a strategic reserve you can aim. It is oil already committed to a journey that stalled. Onshore fill near 60 percent means the next cargo that cannot sail pushes the producer toward shut-ins.

China’s own strategic stocks tell a different clock. Down about 130 million barrels from the April level, they absorbed part of the spring and summer shock. Drawing reserves while imports are depressed is sensible. Drawing them while you also lose your discounted supplier, then trying to rebuild both commercial cover and strategic cover, is how you bid the market up without meaning to.

If I were sketching a risk memo, I would separate three stocks and refuse to add them together: Iranian oil trapped in the Gulf, Iranian oil afloat outside it, and Chinese reserves already partly spent. Different owners, different politics, different ability to hit a refinery gate. Lumping them into “plenty of oil around” is how surprises happen.

Production Math Tehran Cannot Ignore

Roughly 2.2 million barrels a day can still find a home in domestic refining. That is the floor that does not require a foreign buyer. Everything above it needs either an export outlet or a tank. Tanks are filling. The outlet is shut. The pre-war 3.2 million barrel pace starts to look like a memory rather than a target.

Shutting in a million barrels a day is not a switch. It is a sequence of field decisions, gas balance issues, and revenue holes. It also sets the tone for any later restart. Wells that sit idle do not always return at the old rate. Buyers who spent a season learning new grades do not always rush back to the old discount, even if politics thaw. Path dependence is an unglamorous phrase for a very real commercial scar.

That scar runs both ways. A prolonged absence trains the market to live without the barrel. A sudden return, if it ever comes, can crash differentials. Neither path is stable. The unstable bit is the transition, and we are in it.

How Traders Are Likely To Misread The Next Print

The next few data points will be easy to misread. A small rise in Chinese seaborne imports will be called a recovery. It can be a recovery and still be short of barrels. A quiet week in the strait will be called de-escalation. It can be quiet and still carry 13 million barrels a day of other people’s oil past a producer that cannot sell. A premium on ESPO will be called a one-off squeeze. It can be a one-off and still reveal how little spare discounted crude is left.

Watch loadings, not speeches. Watch whether floating volumes outside the Gulf start to fall again without new arrivals in Shandong. Watch whether Qingdao and Dongying intake stabilize or keep bleeding. Watch freight. If VLCC rates stay elevated while Chinese runs creep up, the substitution is not getting cheaper.

And watch the political clock against the storage clock. Talks without a breakthrough plus tanks at 60 percent is a combination that does not require a dramatic headline to become a supply event. The supply event may already be the missing September loading.

A Practical Way To Think About The Balance

Strip the flags off and the balance is simple enough to write on a notepad.

Missing Iranian exports
+ Chinese demand healing from a deep trough
+ reserves already drawn
- no fresh Gulf loadings
= tighter spot barrels, higher freight sensitivity, fatter Hormuz risk premium

None of those lines needs a price target to be useful. They describe pressure. Price is how the pressure eventually prints. Sometimes it prints in the benchmark. Sometimes it prints in a differential, a crack, or a freight invoice that never makes the front page.

For anyone allocating risk, the honest position is not a heroic call on the next ten dollars. It is respect for a barrel that left the visible market and has not been replaced one-for-one. Respect is cheaper than surprise.

What Buyers Can Actually Do

Refiners cannot conjure Kharg back open. They can stop pretending the old slate is coming back this quarter. That means term discussions earlier, freight cover that is not an afterthought, and a clear view of which units hate the replacement grade. It also means admitting that a quota is not a cargo.

Importers outside China should assume a bit more competition for the same Atlantic and Middle East molecules, not less. The teapots are not a side market anymore. They are in the room in Singapore, asking for barrels that used to be someone else’s base load.

Governments watching inflation should resist the idea that this is contained because the missing oil was sanctioned. Sanctioned oil was still oil. When it fed refineries, it kept products in the system. Its absence does not stay inside a sanctions list.

The Incentive Problem In The Strait

Return to the incentive, because it is cleaner than the rumors. A producer that can sell has a reason to keep the lane predictable. A producer that cannot sell watches 13 million barrels a day of regional crude earn revenue for someone else. Negotiations without a breakthrough do not pay invoices. Empty jetties do not either.

That does not mean a closure is destined. It means the cost of restraint has risen, and markets are bad at pricing a rising cost until it spikes. A partial disruption, a slowdown, a new inspection regime, a few turned-away ships: any of those would hit a flow that has only partly healed. The strait is not back to its old self. It is back enough to matter, and not back enough to be comfortable.

If you want a single sentence for the risk section of a note, use this one. The disappearance of Iranian barrels is already tightening supply, and a fresh hit to Hormuz would make the replacement far more expensive than the barrel it replaces.

Why This Does Not Stay An Asia Story

Crude is a pool with thick walls, not sealed rooms. Pull a million barrels toward Shandong and the pool level drops everywhere the pipeline and the ship can reach. Europe’s refiners, India’s spot buyers, US Gulf exporters balancing heavy and light: all of them live in that pool. Product markets then carry the adjustment inland, into trucking, farming, and petrochemicals.

There is a temptation to file this under geopolitics and move on. Fair, until your feedstock slate changes. The commercial translation is already visible at two Chinese ports and in two Russian differentials. That is enough evidence for anyone who prefers receipts to narratives.

I’ll say the quiet part. Underestimating Iranian crude was comfortable when the barrel was discounted and someone else was running it. It is less comfortable when that someone else shows up in the same tender you hoped to win.

Signals Worth Tracking Through The Next Quarter

You do not need a war room. You need a short list and the discipline to update it.

• Kharg and other Gulf loadings: any sustained return, or a continued zero.
• Floating storage inside versus outside the Gulf: trapped oil is not available oil.
• Shandong port intake, especially Qingdao and Dongying, against last year’s baselines.
• ESPO and Urals differentials versus Brent: the crowded-aisle gauge.
• VLCC freight: the hidden tax on Latin American and West African substitution.
• Chinese seaborne imports versus the February pace, not versus last month alone.
• Hormuz transit volumes: strength is good for supply and bad for leverage.

If loadings stay dark, floaters outside Asia keep draining, and Chinese imports grind higher, the squeeze is not a headline. It is arithmetic. If talks suddenly free cargoes, differentials can gap the other way, and anyone who chased ESPO at a record premium will feel it. Both tails are live. The base case in the middle is already tighter than the old balance sheet assumed.

A Note On Certainty

Flow data gets revised. Tanker tracking misses a ship, then finds it. Quotas get used slowly. Premiums mean-revert. I would rather be plain about that than pretend a single September print is destiny. The direction, though, has been consistent for months: a supplier pushed off the water, a buyer drawing down a cushion, a recovery that is real but incomplete, and a strait that is busy for everyone except the country that sits on it.

That pattern does not require you to pick a political winner. It requires you to stop treating Iranian oil as a footnote. Footnotes do not move 1.4 million barrels a day into the world’s largest importing country. When they vanish, the footnote becomes the paragraph.

So the practical conclusion is unromantic. Budget for harder feedstock. Do not confuse an import license with a cargo. Treat Hormuz strength as both a supply relief and a political temptation. And if a desk tells you the missing barrels do not matter because they were never in the official balance, ask them who is now paying 28 dollars over Brent for a Russian grade that used to be the bargain. The answer is sitting in Shandong, and it is everyone’s problem whether the screen admits it yet or not.