|
Part I
The Mechanism
|
Copper.
CNBC ran a segment Friday about "cooling demand fears" and a "data center buildout slowdown" pulling copper back from its highs. Bloomberg's dashboard shows a 6% monthly decline. The sell-side is writing notes about a surplus. If you stopped there — if you only looked at the screen — you'd think copper was loosening up.
Now call a physical trader in Shanghai. Ask them to find you 500 tonnes of cathode for next week. Listen to the silence on the other end of the line.
That number is the market screaming. Not the futures market — the one where people exchange contracts and opinions. The physical market. The one where someone actually needs copper delivered to a warehouse by Thursday or a wire-rod mill goes idle. Premiums don't triple in a fortnight because of "sentiment." They triple because the metal isn't there.
Trafigura — not exactly a company that panics — has spent the past two weeks approaching several producers and traders across Asia, bidding aggressively for November and December copper cargoes. They're willing to outbid rivals. When one of the world's largest physical commodity traders starts paying up and asking questions later, the price chart is telling you one story and the loading dock is telling you another.
The machine has split in two. There's a copper mountain in America and a copper desert in China. And the wall between them is a tariff threat that may never arrive.
Elon Musk’s Insane Projection: 7,692,207%
Where should you invest $100 right now?
Well, Elon Musk just invented and patented this new AI technology…
And he’s predicting it will launch a NEW industry that will grow more than 7 million percent in the coming years.
Think about how insane that is.
That would be enough to grow a single $100 bill into more than $7 million.
Even if Elon is only 10% right…
That would still be enough to grow $100 into more than $700,000.
That’s how big this opportunity is.
So please click here to see the details because this breakthrough could create so much wealth that Elon Musk calls it…
“An infinite money glitch.”
|
Part II
The Diagram
|
Strip the narrative. Here's the engineering.
China consumes nearly 60% of global copper. Its exchange warehouses hold enough metal for maybe 18 hours of that consumption. SHFE inventories fell more than 24% last week alone. The Yangshan import premium — what importers pay above LME to get copper into China — hit $195–210 per tonne, a record in Fastmarkets data going back to 2017. The December SHFE contract is trading at a steep premium over January. That's backwardation. In plain terms: the market is paying more to get copper now than later. That only happens when now is desperate.
Meanwhile, on the other side of the Pacific: COMEX warehouses hold a record 732,315 short tons. US warehouses account for roughly 70% of all visible global copper stocks — the highest US share since 2019. Total American copper stockpile, including off-exchange, sits around 1.6 million tonnes per Macquarie. That pile was built on a bet: traders front-ran a Section 232 tariff, importing 80% more copper than usual in 2025, stacking it in New Orleans and Baltimore.
The tariff proclamation from July 2025 imposed a 50% duty on semi-finished copper — but exempted refined cathode. The Commerce Department was supposed to provide an update by June 2026 so the President could decide on a 15% refined tariff. They haven't published a word. That copper is stranded. Not by physics — by policy uncertainty.
The COMEX-LME spread has collapsed from a peak of $2,937 per tonne in July 2025 to roughly −$85 today. The arbitrage that sucked copper into the US has flipped. But nobody's shipping it back. Traders are sitting on a mountain of metal they can't profitably export and can't legally ignore — not while there's still a chance Washington slaps a duty on it. It's like hoarding canned food for a storm that keeps getting downgraded to a drizzle, except you already bought the entire grocery store.
Two markets. One overstuffed, one starving. Same metal.
|
Part III
The Weak Link
|
Everybody's watching the refined metal. That's the wrong valve.
The weak link is one step upstream: concentrate. The stuff that comes out of the mine before a smelter turns it into cathode. Chinese smelters accepted a benchmark treatment charge of $0 per tonne for 2026. Zero. The annual benchmark — the fee miners pay smelters to process their rock — had never gone to zero in the history of the industry. By June, spot TC/RCs had fallen to negative $126.80 per tonne. Smelters were paying miners for the privilege of receiving feedstock. Let that settle for a moment.
That's what prompted Chinese smelters to pledge cuts of over 10% — though refined output has actually grown year-over-year as many plants kept running at a loss rather than cede market share. The math says they should shut down: running a smelter costs more than idling it when you have to pay for the rock going in. But even with smelters grinding through the losses, the combination of a concentrate famine and the US tariff wall diverting metal has left domestic supply overwhelmed — hence 33,848 tonnes on the SHFE and a premium that triples in a fortnight.
Now add the DRC. In June, Congo banned exports of copper and cobalt concentrates — the feedstock China's smelters desperately needed. In the first week of November, Congolese authorities seized thousands of tonnes of copper and cobalt in an illegal-mining crackdown, pulling even more supply offline. China's largest source of imported refined copper is now an active enforcement zone. Codelco quietly dropped its 1.7-million-tonne 2030 production target to 1.33 million tonnes. Antofagasta's first-half output fell 9.5%.
I've seen this pattern before — the concentrate choke leading to smelter cuts leading to refined scarcity. I traded it wrong in 2024 because I thought the smelters would eat the losses longer. They didn't. The physical layer broke first, and it broke fast. This time the squeeze is worse because the US tariff wall is trapping the release valve.
Spec positioning confirms the disconnect. CFTC net longs dropped from 48,300 contracts to 37,700 in a week. The paper market is lightening up on copper. The physical market is clawing at the walls. When I see paper and physical moving in opposite directions this aggressively, I stop listening to the paper.
|
Part IV
The Chain Reaction
|
This resolves one of two ways. Both of them move copper, but in different sequences and at different speeds.
Scenario A is the violent one. If Commerce imposes the 15% refined tariff, that 1.6 million tonnes stays right where it is. The rest of the world has to make do without it. SHFE inventories, already at less than a day's cover, draw further. Backwardation steepens. Premiums go from record to unhinged. The specs who just sold 10,600 contracts of net longs in a week would be caught offsides while the physical market rips above them.
Scenario B is the slow one. Tariff ruled out — or deferred indefinitely — and the US copper mountain starts to melt. COMEX-LME spread inverts. Metal flows back to Asia over weeks, maybe months. That provides relief, but it's measured in shipping times, not clicks. And it doesn't fix the concentrate shortage. The mine supply deficit — 0.4% refined output growth against 1.5% demand growth — persists regardless. You'd get a temporary premium collapse followed by a second, structural squeeze once the stranded inventory is absorbed.
Either way, the concentrate choke is the structural constant. No tariff decision changes the fact that smelters are paying miners $127 per tonne just to receive rock. No tariff decision brings the DRC's exports back online. No tariff decision speeds up Codelco's production timeline by a single month.
The broad copper ETFs will miss most of this. They hold smelters eating negative TC/RCs, diversified miners hedged forward, and US-exposed producers whose revenue is tied to a COMEX price that reflects surplus, not scarcity. The edge — and I use the word cautiously — is in concentrate-producing miners selling into ex-US physical markets. Companies whose revenue is linked to LME or SHFE spot. Operations in jurisdictions that haven't banned, seized, or sanctioned anything this quarter. That's a short list, which is partly the point.
The ICSG says surplus. The screen says pullback. The loading docks in Shanghai say something else entirely. In my experience — and I have the scars to prove it cuts both ways — the loading dock wins. It just wins on its own schedule.
Sources: Shanghai Metals Market (SMM), Bloomberg, Fastmarkets, SHFE, COMEX/CME Group, CFTC, Macquarie, ICSG, Crux Investor, Mining.com.au, Reuters, S&P Global
