Graphite.

Every lithium-ion battery on the planet — your EV, your phone, the grid-scale pack keeping the lights on — contains more graphite by weight than any other material. A standard EV battery uses roughly 50 to 100 kilograms of graphite. It uses about 10 kilograms of lithium. Guess which one gets a Bloomberg segment and which one gets silence.

The financial press is running the "EV oversupply" story again. Graphite prices are down. Natural anode material that peaked at $8,000–$12,000 per tonne in 2022 now trades at $3,000–$6,000. The headline writes itself: surplus, glut, problem solved. Move along.

The price collapse is real. But it's not a surplus. It's a weapon. China deliberately flooded the global market with graphite anode material priced below Western production costs — a playbook they've run before with solar panels, with steel, with rare earths. The goal isn't market share. It's market elimination. Kill every Western alternative before it reaches commercial scale, then control the tap.

And the tap has an expiration date. Sixty-six days from today, China's suspension of graphite export controls expires. I traded the "China would never actually restrict supply" thesis once before, with gallium. I was wrong. They did.

Story off. Diagram on.

The raw graphite that comes out of the ground is useless for batteries. It has to be purified to 99.95% carbon, spheronised into uniform micron-size particles, graphitized at temperatures above 2,500°C, then coated. That processing chain — not mining — is the real chokepoint. And China owns nearly all of it.

China produced an estimated 1,312,000 tonnes of the world's 1,600,000-tonne natural graphite output in 2024. But mining is the easy part. China controls roughly 93% of global active anode material production and 98% of graphitization capacity — the high-temperature furnace step that no Western facility has replicated at scale. Benchmark Mineral Intelligence puts it plainly: Chinese firms produce 99% of the world's spherical graphite.

Here's the demand picture. Flake graphite demand is forecast to rise from roughly 1.3 million tonnes in 2026 to 2.7 million tonnes by 2036 — more than doubling in a decade. Battery anodes currently represent 28% of graphite consumption. By 2036, that share hits 62%. Over 300 gigafactories are under construction or planned globally. North American battery manufacturing capacity is targeted at 1 terawatt-hour by 2030 — a 1,718% increase from 2021.

With essentially zero domestic anode material to feed them.

Here's the chain that matters:

Syrah Resources started active anode material production at its Vidalia, Louisiana plant — 11,250 tonnes per year. It's the first commercial-scale, vertically integrated natural graphite AAM supplier outside China. NOVONIX received $103 million in IRA 48C tax credits for its synthetic graphite facility in Chattanooga. These are real facilities producing real material.

Together they represent a rounding error against Chinese capacity. The hole in the supply chain isn't at the mine. It's at the furnace. And you can't build a 2,500-degree graphitization plant on a policy announcement.

Three deadlines are converging within sixty-six days. Nobody is looking at all three simultaneously. That's the gap.

First: November 27, 2026. China's Ministry of Commerce suspended its enhanced export permit requirements for graphite shipments to the United States as part of the trade truce negotiated at the APEC summit in late 2025. That suspension is temporary. The word "temporary" is doing all the work in that sentence. When it expires, every tonne of natural flake graphite, spherical graphite, and artificial anode material heading to the U.S. goes back through licensing. China already demonstrated what that looks like — in early 2024, export licensing delays cut Chinese spherical graphite exports 65% in the first two months.

Second: December 31, 2026. The IRA's Foreign Entity of Concern rules currently exempt graphite as an "impracticable-to-trace" battery material. OEMs don't have to prove their anode graphite isn't Chinese. That exemption dies on January 1, 2027. After that date, any EV battery containing Chinese-sourced graphite loses eligibility for the $7,500 consumer tax credit. Given that Chinese material runs through roughly 93% of global anode production, the math gets ugly fast.

Third: already here. In February 2026, the Department of Commerce finalized combined anti-dumping and countervailing duties of approximately 220% on Chinese natural graphite anode material — 153.32% anti-dumping plus 66.68% countervailing. The ITC injury determination followed in March. Those duties are in effect right now. Chinese anode material entering the U.S. is already economically unviable at the border. The market just hasn't fully digested what happens when the other two deadlines land on top of it.

And here's the part that should make procurement officers sweat: qualifying a new graphite anode supplier takes two to three years of testing. Charge-discharge cycles. Purity validation. Cell-level performance. You cannot accelerate materials science with a purchase order. A senior battery manufacturer executive described their inventory position as roughly six to eight weeks of anode material on hand. That's it. That's the buffer between normal operations and a factory shutdown.

This is the same playbook we've watched with gallium, germanium, antimony, and rare earths. Suspend, negotiate, let the market relax, then reimpose. China already flipped the graphite switch once in December 2023. They turned it off as a trade gesture. They can turn it back on in a single Ministry of Commerce announcement. The machines pricing anode material futures have not, to my knowledge, modeled what happens when all three deadlines hit simultaneously.

Walk through the sequence. It isn't complicated. It just hasn't happened yet.

The trigger is the licensing reimposition. In early 2024, when China first implemented export permits on graphite, Chinese spherical graphite exports dropped 65% in the first two months and 28% across eight months. That wasn't an embargo — that was bureaucratic friction. Permit processing, end-user verification, military end-use screening. If November 27 passes without renewal, that friction returns immediately.

Five weeks later, the FEOC exemption dies. Now every EV manufacturer selling into the U.S. market faces a binary choice: source FEOC-compliant anode material or forfeit $7,500 per vehicle. At an industry level, that's billions of dollars in tax credits at stake. The procurement urgency will be immediate and intense — directed at the handful of non-Chinese producers who've already qualified their material.

That handful is very small. Syrah's Vidalia plant at 11,250 tonnes per year. NOVONIX's Riverside facility ramping synthetic production. Northern Graphite's Quebec mine — North America's only producing graphite mine. Nouveau Monde Graphite advancing its integrated mine-to-anode project in Quebec. Westwater Resources building out in Alabama. That's the entire non-Chinese Western supply base. Against 300-plus gigafactories worth of demand.

Where the capital doesn't go: broad battery ETFs diluted with cathode makers, cell assemblers, and Chinese-supply-chain-dependent integrators who lose their tax credit eligibility the moment the FEOC hammer falls. Where it does go: vertically integrated ex-China anode producers with qualified material, IRA-eligible facilities, and offtake agreements already in motion. Companies whose revenue flips from commodity pricing to qualification-premium pricing the moment the regulatory deadlines land.

I'll be honest about the risk. If the truce extends — if China rolls the suspension past November — the urgency deflates. If the FEOC exemption gets pushed another year (lobbying is heavy), the timeline stretches. Both are possible. Neither changes the structural dependency. It just delays the repricing.

Graphite anode material is priced like there's a surplus. It's structured like a hostage situation. Those two facts cannot coexist indefinitely. In sixty-six days, the market finds out which one was real.