Critical-minerals supply-response equities — the allied answer to the China chokepoint
Web-verified 2026-06-11. Structured + sources: macro-critical-minerals-equities.json. The equity cohort building the allied supply response to the China rare-earth + antimony chokepoint (defense_chokepoint.py: independence infeasible until ~2028). Pairs with defense_web, allied_mincut, the chokepoint globe, macro-quantum-computing (the narrative-beta-split sibling), and commodities-metals. Overlay; not used in the proofs.
The corpus already proves the physical chokepoint the AI/defense build cannot buy past on the timeline — China's ~90% control of rare-earth processing, with defense_chokepoint UNSAT for US independence until ~2028. This block covers the equities trying to close it, and adds a mineral the map was missing: antimony.
The new chokepoint: antimony
Antimony is military-critical — ammunition primers, armor-piercing/hardening alloys, night-vision/IR optics, flame retardants — and China controls ~90% of it. In December 2024 China imposed antimony export restrictions on the US; they were suspended in November 2025 (through 27 Nov 2026), but export licensing and the military-end-user ban remain in force (fact). The US had no domestic mined antimony. That is the same auto-deny / military-end-user logic as the REE controls (Dec 2025), on a second mineral — and it is why the cohort below exists.
The state-anchored builders (the real ones)
What separates a strategic supply chain from a momentum trade is a state anchor — equity, a price floor, anchor offtake, an EXIM loan, DPA funding. The four the question named all have one:
| Co. | Asset | State anchor (the de-risking) | Online |
|---|---|---|---|
| MP Materials (MP) | Mountain Pass CA mine + Fort Worth magnets | DoD is the largest shareholder ($400M conv. preferred + warrant = ~15%, Jul 2025); 10-yr $110/kg NdPr price floor; Apple $500M for recycled magnets; "10X" 2nd magnet plant (~10kt, commissioning 2028) | producing; magnets scaling |
| Perpetua (PPTA) | Stibnite, Idaho — only mined US antimony + high-grade gold | DoD DPA Title III $59.2M + DOTC $22.4M; EXIM $2.9B loan (May 2026) fully funds construction; early works began Oct 2025; injunction denied 29 May 2026 | targeting 2029 |
| United States Antimony (UAMY) | Thompson Falls, MT smelter (+ Mexico, Alaska) | Only fully-integrated mil-spec antimony smelter in North America; ~$350M defense+industrial contracts; Pentagon supply deal (Sep 2025); DPA Title III $27M (Feb 2026); 6× smelter expansion (Jan 2026) | refining now; expanding |
| Ucore (UURAF) | Louisiana Strategic Metals Complex (Alexandria) — RapidSX separation | DoD $18.4M to commercialize; separates Pr/Nd/Sm/Gd/Tb/Dy — the six elements under Chinese export controls | full-scale H2-2026 |
The read: MP is the template (state-as-customer + price floor + anchor offtake). Perpetua is the antimony mine (gold pays for it; antimony is the point). UAMY is the midstream refiner — and allied_mincut already found the bottleneck is midstream separation, not mining, which is exactly what UAMY (antimony refining) and Ucore (REE separation) attack.
The cohort (the "etc.") and the narrative-beta tail
- USA Rare Earth (USAR) — Round Top TX (heavy REE + gallium) + Stillwater OK magnet plant.
- Energy Fuels (UUUU) — White Mesa mill (UT): pilot-scale heavy-REE oxides, commercial target late 2026; also the leading US uranium producer (ties the REE + nuclear-fuel legs).
- NioCorp (NB) — Elk Creek, Nebraska: 2nd-largest US critical-minerals source (niobium/scandium + REE); ~$1B, seeking EXIM/DoD.
- Idaho Strategic (IDR) — US antimony + gold + REE exploration.
- Critical Metals Corp (CRML) — Tanbreez (Greenland) + Wyoming; pre-revenue ~2028–29 (already a globe node).
A Trump $12B critical-minerals strategic stockpile ("Project Vault", ~Feb 2026) pumped the entire cohort: Critical Metals +241%, NioCorp / Energy Fuels / Idaho Strategic +100%+ over three months (fact). That is the narrative-beta tail — the same split as macro-quantum-computing (IonQ vs the 800×-sales cohort) and altcoin-lens: separate the state-anchored real builders from the names that "double on a headline and halve as fast."
Why it matters to the map
- Antimony is a second China chokepoint beyond REE — same auto-deny / military-end-user logic, same ~2028–29 independence window the
defense_chokepointZ3 proves infeasible to beat sooner. - The supply response is real but slow. Every builder targets 2026–2029 — exactly the window the proof says independence is infeasible. These equities are that timeline materializing (MP/UAMY now; Ucore H2-2026; Perpetua 2029) or slipping.
- The state is the funder of last resort. Equity stakes, 10-yr price floors, EXIM loans, DPA, and a $12B stockpile are a quasi-nationalization of critical-minerals supply — the mirror image of the AI-capex private circular financing this project documents. For the physical chokepoint the AI build cannot buy past, the government becomes the circular funder; for the compute build, the firms fund each other. Both are markets that won't stand on unsubsidized economics — one underwritten by the state for security, the other by each other for growth.
These relationships are now wired into the bubble-map graph as structural overlay edges (US_Government → Perpetua_Resources / United_States_Antimony / Ucore; the supply chain Perpetua/UAMY → Antimony, Ucore/MP → Rare_Earths; and the demand side China → Antimony/Rare_Earths → Defense_Primes) — joining the existing US_Government → MP_Materials → Defense_Primes state-circular structure. They are deliberately outside the AI financial SCC (the core stays 12; structural_edges_add_no_cycle stays True), so they add context without touching the proofs.
← Research index · structured data: macro-critical-minerals-equities.json · macro-critical-minerals-equities.md