Tearing apart the Gulf aluminum restart illusion

Aluminum.

Bloomberg ran the headline Friday: "Aluminum Heads for Fourth Weekly Drop on Mideast Supply Return." CNBC followed with a panel discussing how the US-Iran deal would normalize commodity prices by Q3. If you read that and felt relieved, I understand. If you read that and felt suspicious, keep going.

LME aluminum has dropped roughly 15% since June 2, falling from $3,752 to around $3,165 per tonne. Fourth consecutive weekly decline — the longest losing streak since April 2025. The story everyone is telling: Gulf smelters are coming back online, Hormuz is reopening, the supply crisis is over. The story the warehouse data is telling: something very different.

The market is pricing in the return of Middle Eastern aluminum shipments because two governments shook hands on a framework MoU. The smelters that produce that metal are either missile-damaged, force-majeure'd, or running at 60% capacity. You cannot restart an aluminum smelter with a diplomatic communiqué. Pot-line rehabilitation alone takes six to twelve weeks. Full restoration of a missile-struck facility? Up to twelve months.

The price is falling. The metal is disappearing. One of those trends is real.

Story off. Diagram on.

The Gulf produces roughly 6 million tonnes of primary aluminum annually — about 9% of the 74-million-tonne global market. A few key facilities dominate: EGA's Al Taweelah smelter in Abu Dhabi (1.6 million tonnes/year) and its Jebel Ali smelter in Dubai (1.1 million tonnes/year), Alba in Bahrain (1.6 million tonnes/year, the largest single-site smelter on earth), and Qatalum in Qatar. Combined, EGA and Alba alone account for 4.3 million tonnes — roughly 72% of the region's identified output.

Here's the current status of every major Gulf smelter. Al Taweelah was hit by Iranian missiles and drones on March 28, forcing an uncontrolled shutdown. EGA says full restoration could take up to 12 months — putting the latest return at April 2027. Wood Mackenzie flagged before the peace deal that offline Gulf capacity was unlikely to normalize before 2028. Alba cut 19% of its capacity on March 15 to preserve raw material inventory after Hormuz choked off alumina shipments. Alumina imports into the Middle East fell 63% year-on-year in March. Qatalum is operating at approximately 60%, constrained by gas supply interruptions, with force majeure declared on aluminum sales contracts.

Between March and April 2026, Gulf output dropped by an annualized 2 million metric tons. ING projects a 600,000-tonne global supply deficit for the full year. That's roughly 50,000 tonnes of unmet demand per month hitting a warehouse system that's already draining at 8% in four weeks.

Meanwhile, the Asian reference price — which incorporates actual physical premiums and freight, not just financial sentiment — climbed to $3,164.50 on June 25. Up 1.35% while LME spot fell. The physical buyers in Asia-Pacific are not buying the peace narrative. They're buying metal.

Here's the part nobody on the peace-deal panels mentions: you cannot cold-start an aluminum smelter the way you restart a refinery or flip a pipeline valve. An aluminum smelter runs electrolytic reduction cells — pot lines — at roughly 960°C, continuously, consuming 13 to 14 megawatt-hours of electricity per tonne of metal produced. When a smelter goes into emergency shutdown, the molten bath inside each pot solidifies. The cells have to be broken out, relined, and restarted one by one.

Pot-line rehabilitation runs six to twelve weeks per line for a clean shutdown. Al Taweelah didn't get a clean shutdown. It got hit by missiles. EGA's own statement: complete restoration "could take up to 12 months." And that's the optimistic reading — the kind a company gives its board when the insurers are watching.

I've seen this movie before. In 2022, when the LME briefly halted nickel trading after a short squeeze, the assumption was that prices would normalize once the exchange reopened. They didn't — not for months. The physical market had its own timeline, and it didn't care about exchange rules. Aluminum's physical timeline doesn't care about peace frameworks either.

Even the Strait itself isn't clear. Mine-clearing operations were scheduled to begin after the formal agreement signing on June 19. Five Western maritime security sources told Reuters the operation could run 40 to 50 days before insurers feel confident enough to reinstate coverage. Without war-risk insurance, ships don't sail. Without ships, alumina doesn't reach the smelters that are still partially operational. It's a nested dependency chain, and the market is treating it like a light switch.

And then there's the US exposure. The United States imported 21% of its primary aluminum from the UAE and Bahrain in 2025 — 584,000 tonnes in eleven months. That pipeline is severed. The Midwest premium, which peaked at $2,645/tonne in May, sits around $2,579 today. Still historically enormous. Still screaming that physical metal is scarce on the ground where it matters.

China can't save this. Chinese smelter output is structurally capped near 45 million tonnes per year by Beijing's own energy policy. There is no spare global capacity waiting in the wings. The deficit is real, the drawdowns are accelerating, and the only thing that's dropped is the price.

The trigger isn't hard to find. It's a calendar.

Mine-clearing operations have a 40-to-50-day window. Sometime in late July or early August, the market will get its answer: either the Strait is certified safe and insurers reinstate coverage, or it isn't and the peace-deal narrative collapses. Either way, the smelter timeline doesn't change. Al Taweelah is still twelve months from full production. Alba is still running with three lines dark. No headline accelerates metallurgy.

At the current drawdown rate — roughly 27,500 tonnes in June alone — LME registered stocks will breach 300,000 tonnes within days and could approach 275,000 by mid-July. In March 2026, live warrants dropped by 98,300 tonnes while cancelled warrants surged 96,050 tonnes. That kind of reclassification precedes a wave of physical metal being pulled out of the exchange system entirely. The June data is following the same directional pattern, just at a slower pace.

When the premiums catch, the short covering follows. Citi maintains a $3,600/tonne near-term target for 2026; Goldman Sachs holds a $3,150/tonne base case, noting prices could reach $3,600 only under extended Gulf production halts. Citi's bull case is $4,000; Goldman's disruption scenario peaks at $3,600. To get there, you'd need exactly what's already happening: extended Gulf production losses, low exchange inventories, and no new capacity. The thesis doesn't require a surprise. It just requires the market to stop pricing in a supply return that physically cannot arrive for months.

Where capital doesn't go: the broad commodity ETFs that dilute aluminum exposure across a basket of metals with different supply dynamics. And not into the Gulf producers themselves — not yet. EGA and Alba are rebuilding, not producing. You don't buy the patient during surgery.

The paper market says the Gulf is coming back. The physical market says it can't — not this year, maybe not next. I know which one I trust. But I've also been early before, and early in commodities feels exactly like wrong until it doesn't. The warehouse data will settle this. Watch the stocks, not the headlines.