Why The Critical-Minerals Race Is Won In The Midstream, Not The Mine: The US Supply-Chain Stack Explained, With $UAMY As The Worked Example

01The Reframe: It’s Not About The Rocks

When a headline says a company “controls a critical-minerals deposit” or “will mine rare earths in America,” the instinct is to think the bottleneck is geology — that whoever owns the rocks wins. That instinct is wrong, and getting it wrong is how investors misread this entire sector.

The United States has plenty of rock. It has rare earths in Texas and California, antimony-bearing ore, and the technical ability to dig. What it does not have — what China spent two decades building and now uses as leverage — is the midstream: the unglamorous chemistry of separating, refining, smelting and turning raw ore into a usable, spec-grade material, and then into the alloys, magnets and components that defense and industry actually consume. China’s dominance is not mainly in mines; it is in processing capacity. That is the choke point, and that is where the money, the moats and the government intervention are concentrated.

This article does three things. First, it hands you a layered map of the critical-minerals chain, so you can see where value and vulnerability actually sit. Second, it gives you a reusable, four-question triage tool for reading any critical-minerals news item — so the next time a stock jumps on a “contract” or a “$1.6 billion” headline, you can tell in thirty seconds whether it’s substance or noise. Third, it uses United States Antimony ($UAMY) as a worked example of one specific layer, with MP Materials ($MP), USA Rare Earth ($USAR) and TMC the metals company ($TMC) mapped onto the others. The goal is not to tell you what any of these are worth — it is to leave you able to read the sector on your own.

02The Real Problem, In Plain Terms

Start with why any of this matters. A handful of obscure elements sit underneath modern defense and industry, and they are non-substitutable. Antimony goes into ammunition primers, flame retardants, night-vision and hardened equipment. Rare earths — neodymium, praseodymium and their cousins — go into the permanent magnets inside precision-guided weapons, fighter jets, drones, EV motors and wind turbines. Nickel, cobalt and manganese go into batteries. You cannot build an F-35, a Tomahawk or a large battery pack without them, and there is no easy swap.

Now the vulnerability. Take antimony as the cleanest illustration, because the numbers are stark. China accounts for roughly 48% of the world’s mined antimony, and the United States is about 91% net-import-reliant on the metal, per the U.S. Geological Survey. In 2024 China moved from “supplier” to “weapon”: it announced export restrictions on antimony products (effective September 2024) and by December 2024 escalated to an outright ban on exports of antimony, gallium and germanium to the United States, explicitly as retaliation for U.S. semiconductor controls. Western antimony prices spiked to a record of roughly $59,750 per tonne in July 2025 before a November 2025 diplomatic thaw softened the controls into a licensing regime that currently runs to November 27, 2026 — with military end-users still restricted.

The rare-earth story rhymes: China dominates not just mining but the separation and magnet-making that turn oxide into a finished magnet, and it has repeatedly threatened or throttled exports. The lesson America drew is that “we can buy it cheaply from China” is not a strategy when China can turn the tap off at will. That realization — not a mining shortage — is what launched the current wave of contracts, grants and equity stakes. It is a midstream problem dressed up in mining language.

None of this is new; it is the second act of a warning fired more than a decade ago. In 2010, in the middle of a territorial dispute, China briefly restricted rare-earth exports to Japan and prices of some elements spiked many-fold overnight. The West took the hint, panicked, funded a few projects — and then, when prices normalized and China flooded the market again, quietly let most of them wither. Mountain Pass, the U.S. rare-earth mine, actually went bankrupt in that lull. The uncomfortable truth the current cycle has to reckon with is that the free market, left alone, keeps rebuilding dependence on the cheapest processor — which is China — every time the crisis fades. That is precisely why this wave is being built on government balance sheets rather than market prices: policymakers concluded that only guaranteed demand and subsidized construction can keep a domestic midstream alive through the next price trough.

03The Three Constraints That Make It A Real Market

A geopolitical worry only becomes an investable market when structural constraints put durable money behind it. Three do exactly that here.

1. Concentration in the midstream. The bottleneck is not the ore; it is refining and processing capacity, which is overwhelmingly Chinese. Rebuilding it in the West is slow and specialized, so whoever owns working Western processing has scarcity value that a new mine does not.
2. Non-substitutable defense demand. These inputs sit inside weapons systems with no substitute, so the buyer of last resort is the government itself — through stockpiles, offtake contracts and price floors. That converts a volatile commodity into something closer to contracted demand.
3. Capital and time intensity. A smelter, a separation plant or a magnet line costs hundreds of millions and takes years to permit, build and qualify. That is a moat for incumbents and a reason Washington is willing to co-fund construction rather than wait for the market.

There is a quieter fourth constraint that compounds the first three: qualification time. A defense or aerospace buyer cannot simply switch to a new supplier because it exists; the material has to be tested, certified and locked into a qualified supply chain, a process that can take years and, once completed, is sticky. That cuts both ways. It slows how fast new domestic capacity can actually displace Chinese imports — but it also means that whoever gets qualified first enjoys a durable, hard-to-dislodge position. In a sector where everyone announces capacity, the companies that have actually been designed into a government or OEM supply chain own something the announcements cannot buy.

04The Stack, Layer By Layer

Here is the map. Every critical-minerals company lives at one or more of these layers. Knowing which layer a company occupies tells you more about its risk and value than almost anything else.

It is worth dwelling on why L2 is uniquely hard, because that difficulty is the entire investment case for the layer. Separating rare earths, or refining antimony to defense grade, is not one reaction but a long chain of them — dozens of stages of solvent extraction, precipitation and purification, each producing waste streams that are toxic, radioactive or both. It is process chemistry that has to be tuned to a specific orebody, it is environmentally fraught to permit, and the operational know-how largely lives in China after two decades of practice. You cannot simply buy a plant off the shelf and switch it on. That combination — hard chemistry, hard permitting, scarce expertise, years of ramp — is why a working Western processing asset is worth so much more than a deposit, and why governments would rather subsidize an existing plant than wait for a new one.

05Where $UAMY Sits On The Map

Now the worked example. United States Antimony ($UAMY) is useful precisely because it occupies the layer that matters most — L2, processing — in the one mineral where the U.S. midstream gap is most exposed.

UAMY runs the only significant antimony smelter in the United States, at Thompson Falls, Montana, where it processes ore into antimony oxide, metal ingots and trisulfide (and recovers some gold and silver as a by-product), with additional processing at two facilities in Mexico. The USGS notes that U.S. primary antimony metal and oxide is produced by a single company in Montana using imported and recycled feedstock — that company is UAMY. In the layer language above, UAMY is not primarily a miner sitting on a big deposit; it is a processor, the domestic node that can take antimony feedstock and turn it into usable, defense-grade material. In a mineral that is ~91% import-reliant and that China just weaponized, being the country’s only working L2 node is the entire point.

It also touches L4 (offtake) directly, which is what elevates it from “interesting processor” to “policy asset”: it holds a real contract to sell to the government (below), and it is receiving federal money to expand its plant. That combination — owning the scarce processing layer and having government demand behind it — is rare for a company this small, and it is why the stock re-rated.

It is worth noting that the word “only” is both UAMY’s moat and its burden. Being the country’s sole antimony smelter means no domestic rival is being qualified in your place — but it also means carrying the cost of strategic redundancy alone, in a small mineral where the free market, left to itself, would import more cheaply from China. That is why this kind of asset tends to exist only with a government leg under it: without the offtake and the grant, market economics alone would struggle to justify the plant. Keep that in mind as you read the traction below — it explains both why the stock re-rated and why it stays structurally delicate.

06The Traction — Read Inside The Framework

Here is where a scorecard-style article would dump numbers. Instead, read each figure through the layer map, because that is what makes them mean something.

The offtake (L4), and its fine print. In September 2025 UAMY secured a five-year, sole-source contract with the U.S. Defense Logistics Agency’s Strategic Materials arm — the body that runs the National Defense Stockpile — to supply 99.65%-purity antimony metal ingots through September 2030, with a maximum value of $245 million. That is a genuine L4 anchor. But apply the framework: a ceiling is not revenue. Per the company, cumulative orders under the contract grew from about $12 million (as of the March 31, 2026 10-Q) to roughly $57.3 million (a July 1, 2026 company update) — real acceleration — yet only the first two ingot shipments, about 82,000 lbs generating roughly $2.6 million in invoices, had actually been delivered and billed by June 2026. So the headline is $245M, the booked orders are ~$57.3M, and the delivered revenue is ~$2.6M. All three are true; they mean very different things.

Government co-funding the L2 buildout. UAMY also disclosed a $27 million grant from the Department of War under the Defense Production Act to expand the Thompson Falls smelter (about $16.2M obligated, $10.8M subject to future authorization), and it already carried a $12.8 million grant receivable at March 31, 2026 for milestones achieved. This is the government paying to enlarge the scarce processing layer — the “build it” half of a “build it and buy it” arrangement.

The business underneath. Revenue is growing fast — full-year 2025 sales of $39.3 million, up ~163% from $14.9 million in 2024 — but the operation is still small and unprofitable: Q1 2026 revenue was $6.8 million (slightly below the prior year) and the company posted an $11.3 million net loss, driven mostly by a jump in salaries and stock-based costs and a $4.1 million non-cash mark-to-market loss on an equity investment, not by the smelter collapsing. The balance sheet is clean — roughly $23.8 million in cash and Treasuries plus a $36.4 million equity portfolio, essentially debt-free — but the share count has climbed from about 121 million to ~148 million in under a year, and shareholders approved more authorized shares in June 2026.

For the full company-level breakdown — segment detail, management, governance, analyst coverage and the complete risk map — see our dedicated United States Antimony ($UAMY) Stock Hub. This article is about the sector lens; the hub is the reference sheet.

07The Paradox That Makes You Think

Here is the tension that separates a thoughtful read of this sector from a naive one: a company can be strategically indispensable and a mediocre business at the same time — and the market keeps confusing the two.

UAMY is, on paper, exactly what U.S. policy wants to exist: the only domestic antimony processor, with a defense contract and federal funding. It is a national asset. But being a national asset does not automatically make it a good stock. The very things that make it strategically important — running a costly, sub-scale processing plant for a mineral the free market would rather import — are also what make it structurally hard to earn a strong profit on. Strategic redundancy is expensive, and the market does not always pay for it. So you get a company that is simultaneously “the only US antimony smelter” and a business that lost $11 million last quarter and keeps issuing shares.

The paradox has a second edge, and it is the sharpest one for reading the tape: the number that moves the stock is often the number least connected to cash. The market reacts to the $245 million ceiling and to policy headlines; it reacts far less to the $2.6 million actually delivered. The strategic narrative — real, verifiable, important — can pull a valuation far ahead of the cash the business is producing. That gap is not proof the story is fake; it is a reminder that “strategically vital” and “currently profitable” are different axes, and that a stock can be rich on one while thin on the other.

The contrast with MP Materials sharpens the point. MP faced the same paradox — a strategic asset that the market would not reliably pay for — and the resolution was not a better quarter; it was the government changing the economics by decree: a price floor under its key product, a guaranteed ten-year buyer for its magnets, and the Pentagon taking an equity stake. In one stroke, “strategically necessary but hard to profit from” became “strategically necessary and contractually profitable.” That is the template the whole sector is implicitly betting on — that Washington will backstop the economics of the assets it deems essential. UAMY has a piece of that (a stockpile offtake and a construction grant) but not the full package (no price floor, no equity backstop, no multi-year volume guarantee). Reading the sector well means watching, for each name, how far the government has gone to close the gap between “necessary” and “profitable” — because that distance, more than the ore or the technology, is what determines whether the strategic story ever reaches the income statement.

08The Triage Tool: How To Read Any Critical-Minerals Headline

This is the part you keep. Next time a critical-minerals stock moves on news, run the headline through these four questions in order. It works for any name in the space, not just the ones here.

1. Which layer is this? Is the news about a deposit (L0), mining (L1), processing/refining (L2), magnet/component making (L3) or offtake (L4)? A processing or offtake milestone is worth far more than a deposit headline, because L2 and L4 are the scarce, defensible parts.
2. Is it an announcement or delivered revenue? Distinguish a ceiling, an order, a shipment and recognized revenue. “$245M contract,” “$57.3M orders,” and “$2.6M delivered” are three different things. Announcements set the mood; delivered revenue sets the value.
3. Who pays, and is it binding? A DoD equity stake, a signed offtake or a price floor is real money and real commitment. A “letter of intent,” an “MOU,” or a “framework” can be non-binding intention. Rank the commitment: disbursed cash and signed offtake > obligated grant > letter of intent > press release.
4. Does the thesis survive China reopening? Much of this sector’s value is an urgency premium created by Chinese restriction. Ask: if China normalized exports tomorrow and prices fell, would this asset still matter? If yes (a genuine, contracted domestic supplier), it is sturdier; if the whole case is “China cut us off,” it is more fragile than it looks.

09The Neighborhood: Where The Peers Sit

Put three well-known names on the same map and the framework immediately sorts them. Note the range — from a producing, government-backed integrator to a pre-revenue permitting story.

Read down the table with the triage tool and the sorting is obvious. MP is the furthest along and the most bindingly backed — it hits the scarce processing and magnet layers and has disbursed government capital plus signed offtake, which is why it is treated as the sector’s anchor. USAR is building the magnet layer (valuable) but leans on a pre-production deposit and a non-binding LOI, so it scores strong on layer and weaker on “is it binding yet.” TMC is a pure L0 resource story whose entire near-term value is regulatory: a permitting milestone, not a shipment, and zero revenue — the highest optionality and the highest binary risk. And UAMY, our worked example, is the L2 processing node with an offtake, in a different mineral entirely. Four names, one map, very different risk shapes.

Notice what the map does that a “critical minerals basket” cannot: it stops you treating these as interchangeable. They ripple together on the same policy headlines and the same retail enthusiasm, which is why they often move as a pack — but their underlying risk is not correlated at all. MP’s risk is execution and valuation on a producing asset; USAR’s is turning intention into financed construction; TMC’s is a single regulatory decision that is closer to a courtroom verdict than a business metric; UAMY’s is converting a contract ceiling into delivered, profitable tonnes. When a sector-wide “critical minerals” rally lifts all four at once, the map is what tells you that the same percentage move means four completely different things. That is the practical payoff of thinking in layers rather than tickers.

10Risks — What The Whole Sector Has To Prove

The framework cuts both ways: it clarifies the opportunity and it exposes the shared risks. Anyone reading this space should hold these in view.

• Announcements outrunning revenue. Across the sector, ceilings, orders, MOUs and LOIs move stocks faster than delivered product moves cash. The gap between the two is where disappointment lives.
• Policy is the tailwind and the risk. Government urgency created this trade; a change in administration, budget or the U.S.–China relationship could soften it. A thesis built on policy is exposed to policy.
• China can reopen. Much of the value is an urgency premium. The antimony licensing regime runs to late 2026; broader normalization would pressure prices and margins across the group.
• Capital intensity and dilution. Processing and magnet plants are expensive and slow, and most of these companies fund themselves with equity — so share counts rise, sometimes faster than revenue.
• Execution and profitability. Strategic importance does not guarantee a good business. Several of these names are pre-revenue or unprofitable, and being “the only” or “the first” is a burden as well as a moat.
• Valuation. When the strategic story is strong, prices can run well ahead of the cash the businesses actually produce, leaving little room for stumbles.

11Bottom Line

The critical-minerals story is real, and it is more durable than a typical thematic trade, because it rests on a genuine structural gap: the West can mine, but it cannot yet process and manufacture at the scale China commands, and rebuilding that midstream is slow, expensive and strategically urgent enough that governments are paying for it directly. That is why the layer that matters is not the mine but the middle — refining, separation, smelting, magnets — and why offtake, the commercial guarantee to buy the output, is what turns any of it into a business.

Use the map and the four questions and you can read the next headline yourself. Place the company on the stack. Separate the announcement from the delivered revenue. Rank how bindingly someone has committed to pay. And ask whether the thesis survives China turning the tap back on. Run $UAMY — the domestic processing node with a defense offtake and a still-unprofitable P&L — or $MP, $USAR and $TMC through that lens, and their very different risk shapes come into focus without anyone having to tell you what to think.

If you want a single date to anchor the sector’s next chapter, watch the antimony licensing regime’s November 27, 2026 expiry and, more broadly, the U.S.–China posture around it: a tightening would re-inflate the urgency premium, a normalization would deflate it. And for each individual name, watch the one number the framework says matters most — not the next headline contract, but the trend in delivered, recognized revenue and whether margins move with it. The theme will keep generating announcements; the companies that convert them into shipped product and defensible profit are the ones the strategic story was supposed to be about.