During Chinese President Xi Jinping’s state visit to Washington in late September, the two sides extended their trade truce to January 10, 2027, keeping tariff relief and rare earth shipments in place for two more months. However, they did not settle the issue many manufacturers were watching most closely. Bonny Lin, director of the China Power Project at the Center for Strategic and International Studies, noted that Xi “did not publicly address U.S. concerns about rare earths and other critical minerals.”

Critical minerals are the foundation of almost every industry that governments deem strategic: artificial intelligence and semiconductors, electric vehicles, defense, renewable energy, and advanced manufacturing. Take rare earths, a group of 17 metals, several of which go into the permanent magnets inside EV motors, wind turbines, and fighter jets. For these magnet metals, China accounts for around 60% of mined production, more than 90% of refining, and almost 95% of permanent magnet production, according to the International Energy Agency. The pattern goes beyond rare earths. China is the top refiner of every other key energy mineral except nickel, where Indonesia leads. On average, the top refining country’s share rose to 72% in 2025, up from 70% in 2023.

That kind of concentration means a single policy decision in China can affect distribution and factories worldwide. Since 2023, China has placed export controls on gallium and germanium for chips, graphite for EV batteries, and antimony and tungsten for defense. In April 2025, it added seven rare earths, including dysprosium and terbium, which are used in high-performance magnets. The number of mineral product categories covered by Chinese export controls has tripled since 2023, the IEA says. U.S. magnet maker Arnold Magnetics said in January that the export approval process had become “slow, opaque and highly discretionary.”

The central question in the minerals race, then, is who will finance the supply chains that reduce this concentration risk. So far, the market alone has not. Investment in critical minerals fell 9% in 2025, ending several years of growth. Exploration spending dropped by more than 10%. Volatile prices are a big part of the reason. Because a mine or refinery takes years to build, lenders are wary of funding one that could be unprofitable by the time it opens. Lithium companies cut their investment by roughly 40% last year. Regarding rare earths, Columbia University’s Center on Global Energy Policy notes that excess Chinese capacity has depressed prices in recent years, which “hurt non-Chinese projects.”

Exploration and mining

At the start of the supply chain, new pools of capital are forming. The Orion Critical Mineral Consortium, set up last year by investment firm Orion Resource Partners with the U.S. International Development Finance Corporation and Abu Dhabi’s ADQ, started with $1.8 billion and is aiming for $5 billion. It invests in mines within emerging markets that have already commenced production or are close to it, rather than in long-term greenfield exploration. Orion founder Oskar Lewnowski called it “a significant step forward in bridging the substantial funding gap.”

This kind of money favors mines that already produce. That leaves the riskiest stage, finding new deposits, with the least capital support, since most exploration projects never become mines. Exploration for lithium and nickel fell about 45% last year.

Security policy is also shaping who gets to invest in new deposits. Greenland has vast and largely untapped reserves of rare earths, a key consideration behind the security agreement that the United States, Denmark, and Greenland signed on September 22. Under the deal, investors from countries outside NATO, the EU, and NATO’s partner countries will need approval from all three governments to put money into sensitive sectors such as mining, according to the Atlantic Council.

Refining and processing

Refining, the step that turns mined ore into the pure metals and oxides manufacturers can use, is the real chokepoint. It is also one of the hardest stages to finance. For rare earths, it means separating elements that are chemically almost identical. Their ions are almost the same size and react in nearly identical ways, so refineries have to repeat the same separation step over and over, sometimes several hundred times. Plants are expensive, slow to build, and exposed to price swings set largely in China. Even including announced projects, rare earth refining capacity is expected to cover only a quarter of the projected demand outside China by 2035, according to the IEA.

That is why governments are responding by becoming investors themselves. In July 2025, the Pentagon agreed to take a $400 million equity stake in MP Materials, which owns the Mountain Pass rare earth mine in California. It also guaranteed a minimum price of $110 per kilogram, for 10 years, for MP’s neodymium-praseodymium oxide, a key magnet ingredient. At the time, that was nearly twice the price inside China, where the same material sold for under $60 per kilogram, according to Columbia’s analysis. If the market price falls below the floor, the Pentagon pays MP the difference every quarter. That protection now matters. In August, the oxide was selling for about $95 a kilogram in Europe and North America, below the floor, while the price inside China had risen to about $107, according to Benchmark Mineral Intelligence.

The partnership has since gone international. Last November, MP and the Pentagon agreed with the Saudi Arabian Mining Company (Maaden) to build a rare earth refinery in the kingdom, with Maaden holding at least 51% and the Pentagon financing the U.S. share.

Australia is taking a similar approach. In June, Iluka Resources secured a A$1.65 billion ($1.15 billion) government loan for its Eneabba refinery in Western Australia, which the company says will be the country’s first fully integrated rare earths refinery. Lynas Rare Earths, another Australian company, already separates heavy rare earths such as dysprosium at its plant in Malaysia.

Europe is moving too. The EU’s RESourceEU plan, unveiled last December, aims to mobilize up to €3 billion within a year for projects that can bring new supply online quickly. Early support has gone to Vulcan Energy’s lithium-extraction project in Germany and Greenland Resources’ Malmbjerg molybdenum project.

Magnets and components

In magnets, where China’s lead is widest, buyers have started putting up money directly. The Pentagon’s deal with MP includes a commitment to buy all the magnets from a new MP factory for 10 years after it opens. Apple has committed $500 million to buy magnets made at MP’s plant in Fort Worth, Texas. Agreements like these, known as offtake deals, give a new plant a guaranteed buyer, which makes it far easier to raise the rest of its funding. Even so, the IEA expects existing and announced capacity to meet less than a fifth of magnet demand outside China by 2035.

Recycling

Recycling offers a partial answer. Apple’s agreement includes a recycling line at MP that will take material from used electronics and factory scrap. The EU has announced plans to keep that kind of material at home by restricting exports of magnet scrap. The IEA expects average recycling rates for key energy minerals to rise from about 10% today to nearly 20% by 2040 under current policies. That helps, but most minerals will still have to come out of the ground.

Strategic stockpiles

Until that new supply arrives, stockpiles are the fallback. In February, Washington announced Project Vault, a strategic minerals reserve that manufacturers can draw on if supplies are cut off. It is backed by up to $10 billion in financing from the Export-Import Bank of the United States, plus nearly $2 billion in private money. In September, commodity traders Glencore and Mercuria became its first partners, committing $1 billion between them. Manufacturers, not federal officials, decide when to draw on the reserve. Gracelin Baskaran, director of the Critical Minerals Security Program at CSIS, describes it as “effectively an insurance mechanism for U.S. industry.”

The IEA suggests this kind of insurance is affordable. For 11 high-risk materials, it estimates the net annual cost of stockpiling for countries outside the dominant supplier at less than $900 million, which it calls “modest relative to the potentially major economic impacts of disruptions.” That is roughly twice what the Pentagon put into MP Materials alone.

A new financing model

Across the supply chain, the same approach keeps appearing. Governments take on the price risk that private investors will not, by buying equity, guaranteeing prices, committing to buying output, and holding inventory. Sovereign investors, commodity traders, and manufacturers such as Apple then invest alongside them.

The model does not come without cost. Price floors can leave taxpayers covering the difference if markets fall. Moreover, many of these projects will not start production for years.

“Recent disruptions have underlined how quickly these vulnerabilities can translate into real economic risks,” IEA Executive Director Fatih Birol said in April. “Addressing them will require sustained investment, stronger resilience measures and deeper international cooperation.”

Sustaining that cooperation will not come easily either. Trade truces are measured in months; the current U.S.-China truce expires on January 10, 2027. Meanwhile, mines run on a different clock: it takes more than 16 years on average to bring one from discovery to first production, according to the IEA. The supply chains being financed today will reduce concentration risk only if the money outlasts the truces.