There is a single sentence in a 2025 defense-industry study that ought to end a certain kind of complacency in Washington. Researchers at Govini, a firm that maps the Pentagon’s supply chains, found that more than 80,000 parts across roughly 1,900 weapon systems incorporate just five processed minerals — and that the global supply of all five is dominated by China. The conclusion, in the report’s own words: nearly 78% of all Department of War (DoW) weapon systems are potentially affected. Renaming the department the “Department of War” changes the letterhead, not the arithmetic. Nearly four out of five American weapon systems depend, somewhere in their supply chain, on materials controlled by the country they are most likely to be used against.
The five elements are antimony, gallium, germanium, tungsten, and tellurium. None is a household name. Together they form a chokepoint on American military power more absolute than any adversary’s missile, and the uncomfortable part of the story is not that China built it. It is that the United States handed it over — deliberately, incrementally, and with ample warning, across decades of procurement decisions that treated the cheapest supplier as the best one.
The five minerals at a glance
The table makes the shape of the problem visible: on the two elements the arsenal can least do without — gallium and tungsten — the United States has essentially no domestic production at all, and China’s grip is near-total. This is not a portfolio of diversified risk. It is five single points of failure with the same address.
The number, and what it really measures
The 78% figure originates from one specific source — the Govini report From Rock to Rocket: Critical Minerals and the Trade War for National Security — and has since been cited by Defense One and the US-China Economic and Security Review Commission’s 2025 annual report. Honesty about what it means is what gives it force. The claim is not that 78% of weapons would stop working the day China cut supply. It is that 78% of DoW weapon systems contain at least one part whose supply chain runs through these five Chinese-controlled minerals — a measure of exposure, not of instant paralysis.
That caveat is a reason to take the number seriously, not to wave it away. The same analysis traced over 43,000 supply chains down six tiers of suppliers and found 88% of them contain at least one Chinese vendor. Some services are far worse than the average: the Navy sits at roughly 91–92% of weapon systems exposed — Arleigh Burke destroyers, Nimitz carriers, America-class amphibs — with the Marine Corps near 62%. The Minuteman III nuclear missile, the F-16, and the Virginia-class submarine all appear on the affected list. And the exposure has been growing: DoW contracts for parts containing these minerals rose more than 23% annually since 2010, with gallium-related contracts climbing nearly 42% a year. The dependency did not sneak up on anyone. It was procured, year after year, on purpose.
Whose choice was this?
Here is where accountability has to move from the abstract to the specific, because “the supply chain runs through China” is a passive construction that hides a chain of decisions made by people with authority.
For thirty years, American defense procurement optimized for cost and efficiency in a globalized market that assumed supply would always be there. Refining these minerals is dirty, energy-intensive, low-margin work, and the United States was content to let China absorb the environmental cost and the capital expense while American primes bought the finished material cheaply. Every acquisition officer who picked the lowest-cost qualified part, every program that declined to fund a redundant domestic source because it blew the budget, every administration that treated critical-minerals warnings as a problem for the next one — each made a locally rational decision that added up to a strategic catastrophe. The Govini data showing 23% annual growth in these contracts since 2010 is the paper trail. This dependency was deepened on the watch of multiple administrations of both parties, long after the risk was understood.
And it was understood. The vulnerability of rare-earth and critical-mineral supply chains has been the subject of government reports, think-tank warnings, and congressional testimony for well over a decade. China telegraphed its intentions too — it demonstrated mineral coercion against Japan as far back as 2010. The information was never the missing ingredient. Political will was.
China pulled the trigger — and Washington called the flinch a victory
This is not a latent risk. Beijing has already run the play, in deliberate stages, and the American response exposed how little slack the system had.
China started with licensing controls on gallium and germanium in August 2023, added antimony in 2024, and in December 2024 escalated to an outright ban on exporting gallium, germanium, antimony, and superhard materials to the United States — plus a blanket prohibition on any dual-use shipment to US military users. The controls deliberately extended to the processing technology itself, to keep rivals from building an exit. The cost landed fast: the US Geological Survey estimated the gallium-and-germanium ban alone could cost the US economy $3.4 billion, and prices for gallium-containing defense parts jumped about 6% within three months while all other DoW parts rose 1.4%. China had proved it could slow American weapons production and raise its cost without firing a shot.
Then came November 2025, and this is where policy accountability gets sharp. After a Trump–Xi meeting, Beijing suspended the ban on gallium, germanium, and antimony until November 27, 2026 — and the suspension was widely received as a de-escalation, a win, a thaw. Read the fine print and it is nothing of the sort. The minerals stayed on China’s dual-use control list, so every shipment still needs a Beijing-issued license that can be slow-walked or denied case by case. The ban on exports to US military end-users was never lifted. The whole arrangement expires in late 2026, at Beijing’s discretion. And in March 2026, China went the other direction, promulgating a dedicated supply-chain security framework that fuses its export controls into a unified national-security regime purpose-built for this leverage.
To treat that as a problem solved is precisely the complacency that created the dependency in the first place. The lever is still in Beijing’s hand. It has been eased off, temporarily, on terms China wrote and can revoke — and mistaking a suspension for a repeal is how a country talks itself out of the urgency required to actually fix the underlying exposure.
The 2027 deadline nobody believes
The government does have a hard target: Defense Federal Acquisition Regulation Supplement rules are meant to bar Chinese-origin critical minerals from key weapons supply chains by January 1, 2027. I flagged this in my previous article on the PAC-3 missile. On paper, accountability at last. In practice, the deadline illustrates the gap between declaring a policy and being able to execute it.
Industry says openly it cannot be met. Standing up mineral separation and refining is a five-to-ten-year undertaking — permitting, chemistry, environmental review, and the qualification cycles defense-grade material must pass before it enters a weapon. The US has no domestic primary source for gallium, germanium, or tungsten at all. Tungsten producers racing toward the deadline warn flatly that reshoring will take years and that contractors will have to keep buying Chinese-linked material in the meantime. A deadline that the people responsible for meeting it publicly call impossible is not a plan; it is an aspiration with a date attached, and setting it without funding the multi-year industrial base to hit it is its own form of unaccountability.
Money is not the binding constraint here, which is what makes the policy failure so stark: the constraints are refining capacity and qualification time, and those were knowable — and buildable — years ago, had the will existed. Recent federal investment, Defense Production Act actions, streamlined permitting, and even equity stakes in miners are real and overdue. But they are the response you launch a decade early, not the one you launch after the adversary has already demonstrated the weapon.
How did this happen? Post-Cold-War globalization made Chinese processing genuinely cheaper and more reliable for a generation; building redundant domestic capacity would have meant higher weapons costs and real environmental and permitting fights at home that communities resisted. Diversification is not free, and some of the dependency reflects market logic that looked sound at the time. Reasonable officials made defensible individual calls.
But that is an explanation, not an absolution. The whole purpose of strategic planning in a defense establishment is to override locally rational cost decisions when they create a single foreign chokepoint on national survival. That is the job. The warnings were on the record, the adversary’s willingness was demonstrated, and the trend line ran the wrong way for fifteen straight years. “It was cheaper” is exactly the reasoning a serious defense policy exists to overrule.
The 78% figure endures because it names something the usual debate about carriers and budgets obscures: American military power is assembled from parts, and a startling share of those parts trace back through five obscure minerals to a single rival supplier that has already shown, more than once, that it will use that position as leverage.
No amount of money fixes this on a wartime schedule; the constraint is refining capacity and qualification time, not funding. The suspension of China’s ban does not resolve the dependency — it postpones the reckoning to a date Beijing controls. And the 2027 deadline meant to end the exposure is one the responsible industries say they cannot meet. The accountable parties are not hard to identify: the procurement system that rewarded the cheapest supplier, the successive administrations that filed the warnings under “later,” and the reflex — visible again in the applause for a temporary suspension — to mistake a pause in the pressure for the end of it. Until the United States can separate and process these five materials at home and at scale, the arsenal of the world’s foremost military will keep running, in critical part, through the refineries of its foremost adversary — and everyone with the authority to change that has known so for years.
My sources for this article include the Govini report From Rock to Rocket (2025), Defense One, the U.S.-China Economic and Security Review Commission 2025 Annual Report, the South China Morning Post, MINING.COM, Fastmarkets, GLOBSEC, the U.S. Geological Survey, and PBS NewsHour (2025–2026). The 78% figure measures weapon-system exposure — systems containing at least one affected part — not systems that would immediately fail; control percentages vary by mineral and by whether one counts mining or refined output. China’s export ban was suspended, not repealed, as of the November 2025 announcement.
Here's my latest Counter Currents with Andrei Martyanov:
Nima and I spent some time discussing the new Houthi offensive in southwest Yemen:
A great INTEL Roundtable today with Scott Ritter, Joe Kent and me:
I did an interview last week with India & Global Left host Jyotishman Mudiar:
China has 56 Universities where you can major in rare earth refining. The US has zero.
We need to adopt a stronger will to work hard mentality in the US if we want to maintain our country’s infrastructures. We can’t keep taking the easy way out (out sourcing everything including work). I’m tired of people here being scared of hard work, this mentality obviously led to this supply chain screw up. Also capitalism is decent sometimes but we can’t keep putting profits over sense.