The Bottleneck Trade, Part II: Own the Metals Layer

Everyone wants the next hardware winner.

I’m more interested in the metals that hardware cannot ship without.

The market still talks about next-gen hardware as if this is mainly a compute story. I think that misses the deeper choke points. Rare earths are critical for high-performance magnets in power generation and electric motors. Copper is still the backbone of electrical and electronic systems. Tin still matters because solder still matters. And tungsten is becoming more strategically important as both semiconductor and industrial supply chains run into tighter export controls and tighter non-China supply.

That is the lens I used here.

I do not want random miners with a good deck. I want smaller companies tied to the metals layer of the next hardware buildout that already have one of two things: operating proof, or unusually strong evidence that the market is being forced to finance them anyway.

These are the five names I would watch now.

1) Arafura Rare Earths (ASX: ARU)

This is the highest-torque name on the list.

NdPr is the magnet metal. If robotics, power electronics, industrial motors, drones, and electrified infrastructure keep scaling, the West will need more non-China magnet feedstock. Arafura is still only about a A$1.24 billion company, but Nolans has moved well beyond a concept. The company says it has secured the remaining conditional approvals for US$775 million of senior debt plus an US$80 million cost-overrun facility, while Reuters reported this week that Nolans is designed to produce 4,440 tonnes per year of NdPr oxide from late 2029, or roughly 4% of global supply, with supply agreements already in place with Hyundai, Kia, Siemens Gamesa, and Traxys.

The bet here is simple: if non-China magnet supply becomes strategic rather than optional, Arafura can rerate hard. This is still a development story, so the risk is obvious. But that is also why the upside is still large.

2) Metals X (ASX: MLX)

Tin is one of the least appreciated bottlenecks in the whole hardware stack.

You can talk about AI servers, advanced packaging, industrial controls, and electrification all day, but electronics still need solder. USGS still lists solder as a major tin use, and the United States still does not mine or smelt tin domestically. Metals X gives you direct exposure to that layer through its 50% interest in Renison, one of the key Western tin assets. The company is still roughly a A$1.14 billion market cap, and Renison just posted 3,319 tonnes of tin-in-concentrate in the December 2025 quarter, its second-highest quarterly output on record.

This is not the sexiest stock on the list. That is exactly the appeal. It already has operating proof, it sits in a real bottleneck metal, and the market still does not talk about tin the way it talks about flashier parts of the hardware chain.

3) EQ Resources (ASX: EQR)

Tungsten sits one step behind the headline, which is usually where I want to be.

It matters in industrial tooling, high-spec manufacturing, defense systems, and increasingly in the semiconductor conversation as buyers worry about concentrated supply. USGS says China remained the leading tungsten producer in 2025, while industry groups have warned that tungsten is a critical semiconductor input and that prices surged after China’s expanded export controls. EQ gives you smaller-cap exposure to that squeeze, but with more operating credibility than most people realize. In FY2025, the company signed five new offtake contracts worth an estimated US$124 million over 24 months, and it also locked in a separate five-year tungsten concentrate agreement with Elmet valued at about A$30 million plus a A$2 million advance payment. Its Barruecopardo operation in Spain also delivered record quarterly output in FY2025.

This is the kind of stock that can move a lot if tungsten shifts from “niche industrial metal” to “strategic hardware input” in investors’ heads.

4) Perpetua Resources (NASDAQ: PPTA)

Perpetua is the antimony choke point.

Antimony is still underrated as a hardware metal. The company’s own materials explicitly tie it to semiconductors, printed circuit boards, and phone glass, on top of the better-known defense angle. What makes Perpetua different is that this is not just a narrative anymore. Its February 2026 investor presentation says Stibnite contains the only U.S. reserve of antimony at 148 million pounds, the company has received more than $80 million in government awards, and it is targeting EXIM debt financing and a final investment decision in 2026 after starting early works construction in October 2025. The stock is now around a $3.3 billion market cap, so it is less hidden than it was, but it still controls one of the clearest domestic bottlenecks in the market.

If the market keeps moving toward “strategic inputs first,” Perpetua stops looking like a normal development miner and starts looking like critical infrastructure.

5) Ero Copper (NYSE: ERO)

Copper is the least exotic name here.

It may also be the hardest to avoid.

The next hardware cycle is also a power-and-interconnect cycle. Copper’s electrical uses account for about three quarters of total use, and the IEA expects copper demand to grow 30% by 2040 in its stated-policies scenario. Ero is not the smallest name on this list, but it is one of the cleanest. The company reported record 2025 copper production of 64,307 tonnes and $395.1 million of operating cash flow, and its February 2026 PEA for Furnas outlined a 24-year initial mine life with roughly 70,000 tonnes of annual copper production over the first 15 years, plus meaningful gold and silver by-product output.

This is the quality anchor in the group. Maybe not the cleanest literal ten-bagger from here. But it is one of the better ways to own a metal that every hardware cycle eventually runs back into.

Final thought

The easiest mistake in this market is stopping at the machine.

I would rather go one layer deeper.

That is where the real bottlenecks usually are. Magnet metals. Solder metals. Tooling metals. Power metals. The names above are not identical setups, and they do not carry identical upside. Arafura, EQ, and Perpetua have more rerating torque. Metals X and Ero have more operating proof.

But all five sit where the next wave of hardware demand eventually gets physical.

And when that happens, the bottleneck usually gets paid first.

Disclaimer: This is not financial advice. This article is for informational purposes only and reflects personal opinion, not a recommendation to buy, sell, or hold any security.