Refining can be the scarce step even when the ore is not

themePublished 19 October 20262 min readcommodities, equities

Key points

• Excluding rare earths, the average share of the top refining country rose to 72 percent in 2025, from 70 percent in 2023.
• Over two years, the top refining countries accounted for over three-quarters of growth in refined supply. Indonesia is named for nickel and China for other key energy minerals.
• Refining projects outside the dominant supplier have capital costs 20 percent to more than 150 percent higher, and operating costs about 50 percent higher on average.
• Public-finance commitments in advanced economies were around 65 billion dollars in 2025, more than four times 2023, and the IEA says a gap remains before disbursement.

What is going on

The executive summary says supply concentration in refining edged higher for most minerals in 2025, with rare earths the exception because of projects in the United States and higher production in Malaysia. In several markets, including manganese, nickel and graphite, virtually all supply growth came from the dominant supplier. For rare earths, existing and announced refining capacity in more diversified regions equals around two-thirds of expected mined supply by 2035, while planned magnet production is only one-third. Planned cathode capacity is about one-third of projected lithium mining capacity. The summary is explicit that announced mining tonnes do not automatically become separated metals or magnets.

Why it matters

New projects outside the incumbent cost more to build and more to run. The IEA's ranges are wide: capital costs 20 percent to more than 150 percent higher, operating costs around 50 percent higher on average, before permitting, skills and equipment lead times. A policy that announces a mine, or even writes a cheque, has not closed that gap. Commitments of around 65 billion dollars in 2025 are not spending. The summary says disbursements will determine whether supply actually diversifies. Battery recycling is a related concentration: China accounts for over three-quarters of global pre-treatment capacity and 90 percent of material-recovery capacity.

What to watch

The concentration thesis would weaken if the top-country share fell because new refineries produced, not because a project was listed. The IEA already records a modest decline for rare-earth refining. One exception does not reverse the 72 percent average. Export controls are the stress test: the summary says full implementation of expanded rare-earth measures could put an estimated 6.5 trillion dollars a year of downstream production outside China at risk. Those expanded measures were suspended for one year until November 2026. The 6.5 trillion is an estimate of exposure if the controls were fully applied, not a loss that has been recorded.

Risks

Diversifying into a higher-cost plant can raise resilience and lower the return. Governments can commit public finance and then disburse slowly, or disburse into projects that do not reach nameplate rates. By-product economics tie some 'strategic' metals to the output of copper, zinc and lead smelters, so a stress in smelter fees can tighten a minor metal even when its own mine looks ample. None of these sentences is a recommendation to own a refiner.

Mining, refining, manufacturing

Exploration finds the deposit. Mining extracts it. Refining and separation make a specification a factory can buy. Magnets, cathodes and compound semiconductors are manufacturing steps that can stay concentrated after the metal is refined. The IEA's one-third figures for magnets and cathodes are pipeline comparisons with expected mined supply, not current market shares of finished goods. Read them as a warning about the middle of the chain.