Oil after Hormuz
Crude round-tripped, then the July memorandum collapsed and Brent came back through a hundred dollars. Gasoline never got cheap. Refiners still sit on the spread.
Iran military action in early 2026 disrupted Strait of Hormuz transit. US retail gasoline prices rose roughly 50% year over year into the second quarter; PCE inflation reached 3.5%, a three-year high. Then crude round-tripped — Lyn Alden put WTI back at roughly $69, pre-crisis levels, by late June. The consumer did not get the same relief. Rory Johnston’s flow read explains why: “gasoline and crude oil are two different commodities.” In moments where processing is the bottleneck, gasoline runs away from crude. Gasoline crack spreads sat near $50 a barrel against a roughly $20 norm; diesel near $60 — retesting the 2022 records on hobbled Middle East refineries, Russian refined-product exports at post-COVID lows, and very low product inventories. The pump price is a product story.
Darius Dale put the constraint in plain terms in August 2026: “crude oil is not the issue. It’s the fact that we don’t have any refining capacity.” The three-to-one crack spread, he said, tells you that releasing every barrel in the Strategic Petroleum Reserve will not create gasoline and diesel. An SPR drawdown into the midterms will not fix pump prices — the constraint is conversion, not barrels. US refining runs were at 96% to 97%, with some units above 100%. Anas Alhajji noted demand destruction begins around $160 and “already happened and already passed” — the consumer squeeze may need a new chokepoint to re-arm, even while cracks stay elevated.
If and when the Strait of Hormuz reopens… for a while there’s going to be demand to store oil because global inventories are going to need to be replenished… But at some point you’re going to see a race for market share. And if we have a race for market share, then everyone wants to produce more oil, then the price of oil has to come down.
Javier Blas, May 2026
Two ways refiners win
Valero is the named expression on the product side. Crack spreads at seasonal all-time highs mean refiners with capacity capture the spread between cheap-ish crude and expensive products. Valero traded near a fresh 52-week high around $268 in July 2026; product-side positioning was not yet crowded — gasoline specs and commercials sat in the low 30s of their one-year range, per the wiki’s Ceresna read. Policy risk targets the beneficiary directly: Trump and Bessent were publicly talking price manipulation by refiners, and US product export restrictions remained a wild card into the November midterms.
A second leg runs through feedstock, not cracks. Venezuelan crude exports recovered above one million barrels per day; US imports of Venezuelan crude were up 300% since January. Heavy-sour differentials widened — Western Canadian Select at roughly $11 to $11.50 under Brent, about $4 cheaper than the fourth-quarter average; Mars discount to Brent around $1 per barrel wider than the prior quarter. Coking-capable Gulf Coast refiners buy the discounted heavy barrel while selling products priced off light benchmarks. Kpler put sour coking margins versus WCS at around $20 a barrel in the first quarter of 2026, compared with roughly $15 a year earlier — a widening it attributed to Venezuelan crude returning to markets.
Valero Q2 refining operating income
Refining operating income from Valero’s Q2 2026 results. Throughput about 3.0 million barrels per day; refining margin $23.62 per barrel.
Valero’s second-quarter refining operating income was $4.5 billion against $1.3 billion a year earlier — adjusted earnings per share $12.54 versus $2.28. Management named feedstock, not crack alone. Gary Simmons said the market structure was “resulting in an improvement in delivered crude costs relative to the benchmarks” and that Valero’s ability to process very high volumes of heavy, high-acid crude is a key competitive advantage. Valero is the largest US consumer of Venezuelan crude and expects processing rates to exceed its historical maximum. Phillips 66 quantifies differential sensitivity at roughly $140 million of annual EBITDA per $1 of WTI–WCS widening — sourced secondarily in the wiki, flagged for verification. Marathon Petroleum expects spreads to widen further but says there are “better options” than Venezuelan barrels — differential beta is broad; Venezuelan-sourcing alpha is narrow.
What crude producers face
The UAE’s exit from OPEC reframes the post-reopening crude leg. Javier Blas called it the biggest challenge OPEC has faced — the cartel “bleeding badly.” The UAE has geological endowment and capital to expand from roughly 4.5 million barrels a day toward five million and potentially higher; it was withholding proportionally more barrels than Saudi Arabia, so its exit removes the larger restraint. When Hormuz reopens, inventory replenishment demand gives way to a market-share race — and once everyone produces flat out, crude has to come down. That reopen is not dated. An AFP dispatch on August 26, reprinted by France 24, had Iran and Oman talking about a joint temporary corridor and a mine-clearing project. Deputy foreign minister Kazem Gharibabadi said the United States must first implement “all of the commitments it has violated” — end the war, lift the blockade, Yemen — before any action to reopen the strait. Oman hoped to “soon announce” a corridor. Traffic, AFP wrote, “remains largely paralysed” nearly six months in. The EIA’s August 11 Short-Term Energy Outlook assumes severe Hormuz constraints persist through August and that most regional crude returns toward pre-conflict averages in early 2027. That is a forecast assumption, not a calendar. The share-war phase stays a hypothesis. The strait has not reopened.
Rory Johnston’s July flow data complicates the sequencing: a post-ceasefire “jailbreak” surged roughly 125% to 130% of pre-war Middle East supply, juiced by floating-storage drawdown with a week or two of runway, into a China-absent market — prompt Brent contango, crude the weakest part of the complex. Saudi loadings were among the slowest to recover, raising whether an implicit tactical cut is holding despite the UAE exit. Anas Alhajji argued the UAE will act responsibly to avoid a price war and that OPEC will stay intact — a named counter to the share-war framing.
At the consumer, the fuel shock produced a K-shaped split: personal savings rate compressed from 6.2% to 4.0%; 90-day-plus credit card delinquency hit 13.12%, a 15-year high; the top 10% of households account for roughly 50% of spending. Off-price retailers TJX, Ross, and Dollar General reported comp acceleration in the first quarter — but the beneficiary leg has since mixed; Marmaxx comps slowed to 1% in the second quarter while HomeGoods accelerated to 7%.
Walmart’s second-quarter FY27 call put a first-party print on the pump-price side of that split. John Rainey said you can tell when fuel prices go above four dollars — June was a little more obvious — and customers made trade-offs. Guidance assumes today’s fuel prices persist. John Furner said rollbacks went from 7,000 to 11,000. David Guggina, who runs Walmart U.S., said grocery grew mid-single digits with unit volume and share gains, and that bettergoods is now a billion-dollar brand. The fuel-to-trade-down channel is live. The beneficiary that printed is grocery value and own-brand, not U.S. apparel off-price. The wiki leaves the contradiction open: Walmart grocery held; Marmaxx missed. Different categories. Do not collapse them. The crude round-trip weakened the fuel-as-crude story; the product-price leg stayed armed.
A September 8 NielsenIQ note restated the grocery and own-brand landing without closing that split. US consumer-packaged-goods unit volume was up 1.7 percent while dollars were down 1.4 percent. Private label sat at about 24 percent of dollar share, up 1.2 percentage points year on year. Value and premium both grew. The mid-tier shrank. Same barbell Walmart already printed first-party. It does not make Marmaxx and grocery the same trade. Do not graduate Walmart on it.
August CPI put another number on the pump. Gasoline rose 3.9 percent on the month and accounted for more than a third of August’s 0.4 percent CPI print. Other motor fuels were up 9.6 percent on the month and 44 percent on the year. Energy as a whole was up 2.1 percent and 16.3 percent. That restates the product-price leg. It does not close Marmaxx versus grocery. Do not graduate Walmart on a gasoline print.
September 16 put a later clock on the same tightness. A Columbia Energy Exchange brief dated the latest crude rally to the July collapse of the memorandum of understanding. Daniel Sternoff had Brent up about thirty dollars to around $107, a global liquids shortfall around five million barrels a day, a US heating crack to WTI over $100 a barrel with WTI itself over $100 — bulk diesel over $200 — and US refiners running at about 98 percent. The SPR, in his telling, is no longer a buffer. Jason Bordoff: Brent broke $100 for the first time since July, and US diesel climbed above $6 a gallon for the first time. Karen Young: the Saudi East-West pipeline struck; the Red Sea corridor less than two million barrels a day against seven million nameplate; weeks of repair; Yanbu storage rundown intensifies diesel tightness. JD Vance, on a different show the same week: Iranians shooting at commercial shipping even in the midst of peace negotiations. That re-tightens the pump-price and product-tightness leg after the June de-escalation the book already recorded. It does not close Marmaxx versus grocery. Do not graduate Walmart or TJX on it. The refiner sit near the highs is not a gap. Do not rank Valero, Phillips 66, or Marathon on it.
August retail sales then printed a goods rebound. Headline sales rose 1.2 percent on the month, to $773.9 billion, after July’s 0.5 percent drop. The control group was up 1.4 percent. Building materials were the only major category down, 0.2 percent. A goods rebound is not a grocery-versus-Marmaxx close. The wiki’s conviction on that split stays low. Do not graduate Walmart or TJX on it.
In early September Palacios restated the configuration fit — US refiners are “pretty well suited to that type of crude” — and noted Chevron had announced another Venezuela deal, “they’re going to invest $7 billion,” the same week as NABEP: 65 billion barrels, a 35 percent US government stake. Bob McNally, who runs Rapidan and once sat on the NSC energy desk, said having Uncle Sam at the table “creates problems while trying to solve problems,” and that durability after 2028 needs “a durable, legitimate, democratic government.” Independent color on an input step the wiki already treats as confirmed, and on a sanctions falsifier it already named. Not a new chain. Do not re-rate Valero, Phillips 66, or Marathon on it. The names still sit near 52-week highs. A sit is not a gap.
The refill is not a bid
Dale already said releasing every barrel in the reserve will not make gasoline. The other SPR argument is that once the caverns are empty, Washington has to buy the oil back. That is not how the 2026 releases were written. The Department of Energy treats them as exchanges — time-loans of similar-quality crude plus premium barrels, not cash purchases. Energy Secretary Chris Wright authorized 172 million barrels to be replaced by about 200 million, “at no cost to the taxpayer.” The fiscal-year 2026 requests for proposals put the extra oil at 8 to 24 percent, with return windows into 2027 through 2029.
Anas Alhajji said this first. The companies borrowed at $120 and will return the barrels when oil is cheaper. “It’s a loan, it’s not a purchase.” Doomberg independently called them molecular loans repaid in molecules, with no cash and no appropriations. The wiki now has the rule: 10 CFR 626.7. Cash is a close-out residual, not ordinary repayment.
Yes, it supports prices, it creates a floor, but it’s not going to raise prices substantially.
Anas Alhajji, Macro Voices, July 2026
Injection has not started. EIA weekly stocks fell from 415.441 million barrels the week ending February 20 to 284.957 million the week ending September 11 — still a draw, about 58 thousand barrels a day that week. Alhajji’s 400-thousand-barrel-a-day ceiling sits near a GAO effective-fill figure of 440 thousand and well below the Department’s design fill of 785 thousand. It is not DOE law, and it has not been tested. A July Goldman recap treats global SPR rebuilding, about a million barrels a day, as a floor that still leaves a 2027 surplus. No fetched note says XLE or USO still prices a US refill rally. The wiki leaves that as a hypothesis. Do not graduate a crude bid on it.