Distillate Stocks Fall to 103.4m, 14% Below Average (27 August 2026): The Crack Gave Back $10 on a Hormuz Deal the Foreign Ministries Never Announced

The margin for turning a barrel of crude into diesel set an all-time record above $102 on 17 August 2026, then gave back about $10 over the following nine days as crude sold off on reports of a Strait of Hormuz agreement. On 26 August two things happened that the price move had not accounted for. The EIA reported that US distillate stocks had fallen another 2.2 million barrels to 103.4 million — about 14% below their five-year average — with refineries at a fresh high of 97.4% utilisation and distillate output falling for a third straight week. And the only party to have described the Hormuz talks as a revenue-sharing agreement turned out to be the Islamic Revolutionary Guard Corps; the two foreign ministries had announced no such thing. The price of the shortage fell on a headline. The shortage got worse on the data.

That gap between a margin and a physical gauge is the most instructive thing in the energy complex right now, and it is not a contradiction. A crack spread is a live price; distillate inventory is a stock. One of them can be repriced by a sentence attributed to a foreign ministry, and the other cannot.

What actually happened

Oil fell for three consecutive sessions into 26 August, taking Brent down roughly 9% on the week to near $86 and WTI to near $80.30. On Thursday 27 August both steadied — Brent around $88.31 and WTI around $82.34 — as, in the market's own framing, improving supply prospects through the Strait of Hormuz were weighed against growing disruption to Russian energy exports.

The catalyst for the selloff was diplomatic, and it is worth being precise about what was actually said, because the distinction is the whole trade. On Tuesday 25 August the Iranian and Omani foreign ministries issued a joint statement describing discussions of an "interim framework" for resuming ship transits; Iranian deputy foreign minister Kazem Gharibabadi had said a permanent route would be negotiated within 30 to 60 days. On Wednesday 26 August an IRGC spokesman, Hossein Mohebbi, went considerably further, telling the state-run Sepah News agency that "agreements have been reached regarding each country's share of the strait's waters as well as Iran and Oman's share of its revenues," and adding that the United States was obstructing the process. Iranian foreign minister Abbas Araghchi has not confirmed the revenue-sharing claim, and Oman's foreign ministry did not immediately comment. Talks had been reported as close to a deal earlier in the month before stalling. Nothing has been ratified, and no additional barrel has sailed on the strength of it.

Diesel went with the tape and then stopped. ULSD futures traded at $4.1954 a gallon on 27 August. Multiply by 42 and that is $176.21 a barrel; set it against WTI at $82.34 and the implied crack is $93.87. Run the identical sum on 17 August — WTI settling at $84.50, ULSD near $4.4371 — and wholesale diesel was $186.36 against a crack of $101.86. So the margin gave back roughly a tenth of itself in nine days, and has since clawed back about $2.

Crude settles and inventory data: EIA Weekly Petroleum Status Report; retail prices: EIA Gasoline and Diesel Fuel Update. Inventory, utilisation and production columns are for the weeks ending 7, 14 and 21 August respectively; retail is the survey of 17 and 24 August, with the next survey due 31 August.

Why a foreign ministry can move a refining margin at all

A crack spread contains two separable things, and only one of them is physical.

The first is the realised gap between how much distillate the world wants and how much it can make this month. That is what inventories measure, and it is why the crack was above $100 in the first place.

The second is the market's estimate of the chance that the gap gets worse. A closed or contested chokepoint is not primarily a distillate story — Hormuz mostly carries crude — but a disrupted Gulf raises the probability of the next refinery outage, the next stranded product cargo, the next re-route that adds a fortnight to a voyage. That probability has a price, and it sits inside the same number.

A credible framework for reopening the strait removes most of the second component and none of the first. That is the whole move. It is the same logic that took the risk premium out of crude, arriving at a product market that happens to quote its scarcity as a spread. Anyone reading the $10 retreat as evidence that the diesel crunch is resolving has confused a fall in the option value of things getting worse with an improvement in things as they are — and in this instance has done so on the strength of a claim that one party to the talks made and the other two have not repeated.

The physical gap got wider, not narrower

The EIA's Weekly Petroleum Status Report for the week ending 21 August, published 26 August, is the cleanest evidence available, because it reports both sides of the refinery gate in one table.

On the crude side: commercial inventories were essentially unchanged at 428.9 million barrels, which is now 1% above the five-year average for the time of year — an improvement on the deficit of a fortnight earlier and a build of 10.6 million barrels against the same week last year. By any normal reading, crude is comfortable.

On the product side, everything moved the wrong way. Distillate inventories fell 2.2 million barrels to 103.4 million, taking them to about 14% below their five-year average and 10.9 million barrels — 9.5% — below the 114.2 million of a year earlier. Gasoline stocks fell 2.5 million barrels to 206.8 million, 6% below average, so the tightness is no longer confined to the distillate half of the barrel.

And in between, the plants tried harder and delivered less for the third consecutive week. Utilisation rose from 97.2% to 97.4% of operable capacity, with crude runs at 17.4 million barrels a day — and distillate production still fell, from 5.2 to 5.1 million barrels a day. That sequence is the crunch in miniature: the constraint is no longer how much crude you can push through, it is what the configuration of the remaining plants yields at the other end. EIA capacity data puts US operable atmospheric distillation capacity at 18.2 million barrels per calendar day at the start of 2026, down more than 250,000 barrels per calendar day across 130 refineries — two fewer than in 2025.

The export pull is visible in the same report. Net imports of petroleum products ran at −6.639 million barrels a day over the four weeks to 21 August, against −5.215 a year earlier: American refiners are selling into the strongest product market in the world, which is rational, and which drains a domestic system already 14% below its inventory average.

What is still physically broken

Three supply losses built this margin, and the Hormuz framework touches none of them.

Russia is the largest, and it deteriorated further this week. On 8 July 2026 Russia banned diesel exports, with Deputy Prime Minister Alexander Novak announcing the restriction at a government meeting and saying Russia would begin importing fuel, after sustained Ukrainian drone strikes on Russian refineries cut domestic processing. Russian crude processing fell to roughly 3.91 million barrels a day in July, more than 1.4 million below the prior year's average and the lowest in about two decades on independent estimates. Strikes continued at a record pace through August, and Bloomberg reported on 25 August that Moscow was weighing a further extension of the producer restrictions. Russia supplied roughly 11% of global diesel in 2025. The return date for those barrels is a policy variable that keeps being rolled forward, not a fixture on a calendar.

Saudi Arabia is the newest and the most concrete. Aramco's Jazan refinery — 400,000 barrels a day — has been shut since 27 July after Houthi drone attacks that damaged its gasification complex and tank farm, and the restart was pushed from 15 August to 30 August. The Houthi movement claimed a further drone attack on the plant on 18 August. That restart is the single largest identifiable swing factor in the next week, and it is a date rather than a negotiation — which also makes it the cleanest test of whether adding conversion capacity moves this margin at all.

The third is American, and it is the counter-intuitive one: record distillate exports, visible in the net-import figure above, are the rational response of a refiner to the widest margin in history — and the reason a global shortage lands on a US inventory number.

The pump is a lagging price, and it just printed a four-year high

While the wholesale margin lost $10, the price a truck operator actually pays went up again. The EIA's retail survey released 25 August put on-highway diesel at $5.652 a gallon for the week of 24 August, up 19.8 cents — a second consecutive weekly rise of roughly 20 cents, following 19.7 cents the week before. That is the highest weekly reading since the week of 4 July 2022, and it sits 15.8 cents below the all-time high of $5.810 set on 20 June 2022.

Regular gasoline, in the same survey, was $4.085 — up 3.6 cents. The gap between the two fuels is now $1.567 a gallon.

There is no mystery in the divergence. Retail prices are cost-pass-through prices, set against wholesale product bought weeks earlier and against inventory already sitting at a distributor. They follow the futures tape with a lag, and the lag is longer on the way down than on the way up. The next survey lands on 31 August, and it will still be reporting the $102 margin rather than the $94 one. What the pump price is emphatically not is a live indicator of whether the shortage is easing.

Inflation: still a business input, and increasingly a demand story

US CPI rose 0.1% in July 2026 and the annual rate eased to 3.4% from 3.5%, with core CPI at 2.5% year on year and the gasoline index falling 2.9% on the month, per the Bureau of Labor Statistics. A record diesel margin and a cooling CPI print in the same fortnight is not a data error; households buy gasoline and businesses buy diesel, and diesel reaches consumer prices through haulage, food distribution and anything that moves on a truck, with a lag measured in months and only to the degree firms pass it on rather than absorb it in margin.

The demand side is now the fastest-moving part of the dataset, and it is moving in one direction. Distillate product supplied over the four weeks to 21 August averaged 3.798 million barrels a day — down 2.2% year on year, after −0.8% the week before and +1.9% the week before that. Total products supplied were down 3.0% on the year over the same period, and jet fuel down 2.3%. Three consecutive weeks of deterioration is no longer a single noisy print. It is the only mechanism that has ever ended a product spike without new capacity: a price high enough to remove discretionary freight and industrial consumption. That does more to cap the inflation channel than a $10 retreat in a wholesale margin, because it works on the physical gap rather than on the premium attached to it.

For the rate factor that sits at the centre of any currency read, none of this settles anything. A relative price shock in a business input is the category central banks have historically looked through, because tightening policy does not build refineries. What converts a look-through into a response is duration and expectations — and the duration question now has two competing answers inside the same dataset: supply that keeps getting worse, and demand that has started to fold.

The currency read flipped sign without changing its lesson

Through the record-margin phase the reflex to reach for commodity currencies misfired, because the price that rose was a processing margin. It was earned by whoever stands between crude and diesel. Canada sells the input — crude, overwhelmingly by pipeline to a single customer — and WTI ended 17 August at $84.50, roughly where it had been, while the margin above it set a record. Most of the refining capacity that captured it is not Canadian.

The last ten days inverted the move and left the lesson intact. The crude benchmark itself fell about 9% on the week for Brent before steadying near $88, and the margin fell with it. That combination is a terms-of-trade deterioration for a crude exporter, in a way the record margin never was — and a genuine relief for the euro area, a structural net importer of both crude and diesel with a far higher diesel share in its vehicle fleet. The same shock lands on the two blocs with opposite signs, which is exactly what the earlier product-led episode did not do.

The through-line, across both phases, is that "energy is up" or "energy is down" is not a directional signal for any currency until you name which energy price moved and who sells at it. The running Hormuz storyline has been demonstrating the same point in the crude market, where the loonie has repeatedly failed to move on headlines that looked decisive, and our breakdown of the oil–CAD relationship sets out the structural reasons why.

What would change the picture

The crack is a price, and prices of this kind are estimates of how long a physical gap lasts. Three things would close it, and a contested announcement in Muscat is not one of them.

Conversion capacity returning is the fastest and the most measurable: run rates cannot go far above 97.4%, so realistic movement is downward through outages, and Jazan's scheduled 30 August restart is the nearest concrete addition — with the caveat that the plant has now been the subject of a further claimed attack since the date was set. A large exporter returning is the biggest single lever, and Russia's restrictions have been rolled forward rather than lifted while its refining runs sit near two-decade lows. Demand destruction is the third, and it is the one actually moving, three weeks running — the same ceiling both the IEA and OPEC cited when they cut 2026 demand forecasts this month.

The observable worth watching is not the crack. It is distillate inventories against their five-year average, published weekly, alongside the utilisation rate that determines whether the deficit can be worked off at all. The crack is the price of that gap; the last ten days repriced the price while the gap widened by another 2.2 million barrels, and no amount of crude arriving at a refinery running at 97.4% will narrow it sooner. For anyone reading the energy complex through a currency lens, the reading order this autumn is: distillate stocks first, run rates second, the crude benchmark third — the reverse of the order most people use, and the reason this week looked like a resolution when it was a repricing. You can see how the commodity factor currently sits across all eight majors on the live meter, and the wider method behind it on our about page.

Educational macro context only — not investment advice.