Overlooked Tungsten Play

🌾 Welcome to StableBread's Newsletter!

Take a look at the price of tungsten year-to-date:

Ammonium paratungstate (APT), the traded form of tungsten, sat around $340 per metric tonne unit before China's export controls in early 2025. It's north of $3,000 now, the highest on record!

Tungsten is one of the hardest and strongest metals in use. It melts at 3,422°C, the highest of any metal, and it's about as dense as gold.

It's found as tungsten trioxide, or WO₃, in two ore minerals, scheelite and wolframite. Grades get quoted as % WO₃, and APT trades in dollars per metric tonne unit (MTU), where one MTU is 10 kg of contained WO₃.

Looking at the tungsten price chart above, you may conclude that you've missed the ride. But what matters is the supply and demand imbalance.

Demand is structural and growing, supply is concentrated in China and shrinking everywhere else, and the problem now is availability more than price.

Tungsten is hard to replace in most of its uses, which is why demand comes from a few places that aren't going anywhere:

The bottom line is that tungsten is positioned upstream of the two demand stories the market cares about most—defense and AI—right as its supply tightens at the source.

Below, I recap the supply shock and why it looks structural, then propose why the trade isn't the metal or the miners. It's the hardest inputs to secure, high-purity tungsten powder and increasingly molybdenum, which Samsung and SK Hynix are switching to inside their memory chips.

I cover seven names below, plus Japan's workaround to the China cutoff. Six mine tungsten or are developing mines, most have already run, and all carry the same tail risk of China flooding the market in 2027 or beyond.

The seventh (my personal favorite) is more protected because it sells finished parts made from both metals rather than producing the metal itself. Plus the company is already profitable and trades below 3x against 19x+ for the producing miners.

China mines and refines ~80% of the world's tungsten and consumes over half of it.

For years, China's low-cost supply made it hard for mines anywhere else to compete. That's why there's so little Western supply to talk about.

But in February 2025 China placed export controls on tungsten and a list of other critical metals, requiring licenses to ship them. It then published a 2026-2027 whitelist that cut legal tungsten exports to just 15 approved companies.

Chinese APT exports fell ~70% over 2025, and the Western benchmark price ran from the $340 range past $3,000/MTU.

So the 7-9x move happened because the marginal Western buyer suddenly had to source from a market that was short without China.

Two forces make the shortage structural, not temporary.

The first is the WF₆ choke point.

In mid-2026, two Japanese chemical makers, Kanto Denka and Central Glass, said they would halt WF₆ production from July 1. Between them they made about a quarter of the world's WF₆ (more on them later).

They ran out of the Chinese high-purity tungsten that feeds the process. Roughly 60-70% of WF₆'s cost is that tungsten powder, and Japan has none of its own.

WF₆ prices more than doubled. Qualifying a new gas supplier takes 18-24 months, so the squeeze runs into 2027.

The second is U.S. defense procurement.

From January 1, 2027, the U.S. Department of Defense will bar tungsten powder and heavy alloy mined or refined in China, Russia, Iran, or North Korea from its supply chain.

Additionally, NATO members agreed to lift defense spending toward 5% of GDP, more than double the old 2% guideline, and the US is stockpiling the metal.

So Western governments now have a structural reason to pay up for non-Chinese tungsten, and a premium for non-Chinese supply is becoming the going rate.

The whole thesis still comes down to one question: what would make China ease its grip on supply?

There are two ways the squeeze ends, and they point in very different directions.

The first is a US-China deal, where China eases its mineral controls in exchange for the US easing chip and tech restrictions.

It's possible, but tungsten is one of China's strongest points of leverage, and the export-license rules it set up in 2025 are easy to keep on the books while appearing to ease them.

Almonty's CEO, Lewis Black, put it bluntly:

Tungsten is too useful as leverage, and China is running the same playbook across rare earths, gallium, germanium, and antimony. Handing tungsten back cheaply would undercut all of it.

The second is worse, and more likely. China floods the market to crush the price and bankrupt Western supply before it scales.

China has done this before. After rare earth prices spiked in 2010 and 2011 and the West started building mines, China reopened supply and let prices collapse. Most of those projects died, and Molycorp, the biggest US rare earth producer, went bankrupt.

China controls ~80% of production, so it can sell at a loss long enough to make a high-cost Western mine uneconomic, then tighten once the competition is gone.

This is where cost structure decides everything. A price war guts the high-cost developers but barely dents the lowest-cost producer.

It also explains the price floors. Almonty (one of our tungsten miners discussed below) has signed offtakes that guarantee a minimum price with no cap, and the US and EU are building similar support for domestic mines. Both exist to keep Western mines alive through exactly this kind of price war.

So when I say "if China reverses," I don't mean a friendly policy U-turn. More than likely it's a deliberate price war, which is why cost and price floors matter more than a developer’s ore grade.

Watch what China does, not what it says. Four signals tell you whether the squeeze is tightening or about to break:

China's mining quota

The Ministry of Natural Resources sets a tungsten mining quota twice a year, capping how much tungsten China mines.

A sharp increase would be the first sign of loosening. Instead the 2025 quota fell ~6% to 58,000 tonnes, and the second batch was cut a further 8-10%.

Monthly export volume

Chinese APT exports fell about 70% in 2025 and hit zero in January and February 2026. The China Tungsten Industry Association reports this monthly.

If China floods the market, it shows up here first, as monthly exports jumping back toward pre-2025 levels.

China-to-West price gap

With exports choked, Rotterdam trades far above China's domestic price. If China floods, exports reopen and the Rotterdam price falls back toward the Chinese price.

The warning sign is the gap closing because Rotterdam is falling, not because China's price is rising.

China's domestic APT price (the red line in the CTIA chart below) swinging on its own, like its 50% slide since March, just means the two markets are disconnecting further.

Tracking it takes two numbers. CTIA publishes China's domestic APT price weekly in RMB per tonne. Divide by 88.5 (APT is ~88.5% WO₃), then by the yuan rate, for US$/MTU.

Right now that's ~US$1,100 against ~US$3,050, a 2.7x gap. It was ~1.3x at the March peak and near parity before the controls.

Licensed exporters

China funneled all 2026-2027 tungsten exports through 15 approved firms under a one-order, one-certificate system. Widening that list, or easing the licenses, would signal the shift.

The rare earth playbook says the flood comes after Western supply is built, not before, so the window to watch most closely is 2027 and beyond.

Let's first cover the tungsten miners, then move down the chain to the powder problem and the two ways to own it.

Almonty Industries (NASDAQ: ALM)

Almonty is the highest-quality producer in the group.

Sangdong in South Korea mines at ~US$127/MTU, low enough to survive any price war, Sangdong's output is sold under price floors with uncapped upside, and an oversubscribed US$700M convertible funds its expansions.

However, at ~128x trailing sales, the market’s expecting a flawless Sangdong ramp, with full Phase I throughput expected in Q3 2026.

Almonty is also building a tungsten-oxide plant in South Korea aimed at the WF₆ supply chain, which makes it one of the few miners moving toward the powder problem.

EQ Resources (ASX: EQR)

EQ Resources is the closest thing to owning the tungsten price itself.

It ships today from Mt. Carbine in Australia and Barruecopardo in Spain, it sells unhedged, and its costs back out to ~US$500/MTU, which is why gross margin only turned positive in late 2025 once APT had already run.

If APT stays high, EQ Resources compounds fastest of the group. However, if prices normalize/crash, its margins are also the first to go, because its costs run ~4x Almonty's.

EQ Resources is also targeting 175,000 MTU a year at Mt. Carbine and over 160,000 at Barruecopardo, which together would rival Almonty's current capacity. But those are aspirational rather than guidance, and Almonty's 2027 expansion would pass them at a fraction of the cost.

Tungsten Mining (ASX: TGN)

Tungsten Mining has no mine yet.

Watershed in Australia is permitted, and its study models an AUD 1.3B pre-tax NPV and a 198% IRR on AUD 274M of capex, but there's no revenue and first concentrate would arrive in 2027 at the earliest.

The AUD 274M build isn't funded (it holds about AUD 54M of cash) and the $207M market cap is a claim on the study's numbers holding at actual market prices. The September 2026 final investment decision is what investors are watching.

Guardian Metal Resources (LON: GMET)

Guardian Metal Resources is the furthest along of the US developers.

Its just-released Pilot Mountain study shows a US$660M after-tax NPV and a 59.6% IRR on US$289M of capex, it holds US$6.2M of Defense Production Act funding, and it added a NYSE American listing in March.

At ~US$490M, the market already pays ~0.75x the modeled value of a mine that isn't built, isn't funded, and is years from production. For comparison, Tungsten Mining trades at ~0.25x its own study's NPV.

The government money is validation, but you're paying up for it.

American Tungsten (CSE: TUNG)

American Tungsten is a well-funded drilling story.

It holds ~$37M of cash against an $87M market cap and plans to restart the IMA mine in Idaho, which produced ~199,449 MTU between 1945 and 1957.

But there's no defined resource, no economic study, and no production date yet, so you're buying management and drill results. Until a maiden resource lands, this is an option on US tungsten policy.

Viking Mines (ASX: VKA)

Viking Mines is the smallest and earliest name here.

It carries an AUD 24M (~US$16M) market cap, AUD 4.7M of cash, 2.4 billion shares, and no resource yet at Linka in Nevada. Linka is a past producer, though, at a 0.54% WO₃ grade that beats everything else in this write-up.

The first assays (lab tests showing how much tungsten the drilled rock actually holds) land this quarter, and American Tungsten is an early strategic investor, a modest vote of confidence. If the assays disappoint, there's nothing underneath the stock.

None of those mines resolves the shortage inside the chip.

WF₆, tungsten hexafluoride, is the gas that deposits tungsten inside a semiconductor. That tungsten becomes the microscopic wiring inside DRAM, HBM, 3D NAND, and logic chips. As NAND stacks climb past 200 layers, the tungsten pulled per wafer rises steeply, so every step up in AI memory demand pulls more tungsten with it.

In July 2026 the two Japanese makers I mentioned earlier, Kanto Denka (TSE: 4047) and Central Glass (TSE: 4044), stopped making WF₆ for good and sent final-delivery notices to their Korean chip customers.

The trigger wasn't demand, which is clearly booming. They could no longer get the high-purity tungsten powder that feeds the process, because China's powder shipments to Japan have fallen to roughly zero since the February 2025 controls.

So the problem isn't entirely WF₆ capacity. It's the powder.

Korea's WF₆ makers, market leader SK Specialty (unlisted) and Foosung (KOSDAQ: 093370), are picking up Japan's redirected orders, but they scale by importing Chinese material, an access Beijing has granted Korea and denied Japan.

WF₆ prices are up 70-90% for 2026, and Peric (SHA: 688146), the Chinese gas maker set to absorb the share Japan is vacating, has run ~5x in two months.

So who outside China actually processes tungsten, from high-purity powder to finished parts?