Critical Minerals in 2026: The Strategic Constraint Is Moving Downstream

1. Executive assessment

Critical minerals are often discussed as a map of reserves. That is necessary but insufficient.

The IEA reports that concentration in refining increased for most key energy minerals in 2025. Excluding rare earths, the average share of the leading refining country increased from 70% in 2023 to 72% in 2025. Indonesia is the dominant refiner for nickel and China for most other key energy minerals. Over the same two-year period, those leading countries accounted for more than three quarters of growth in refined supply. IEA, Global Critical Minerals Outlook 2026

At the same time, the long-run supply balance remains tight in important markets. The IEA's projected copper supply deficit for 2035 narrowed from roughly 30% to 25%, but did not disappear. Lithium's outlook improved, while cobalt risks increased after policy changes in the Democratic Republic of the Congo.

The central distinction is therefore between resource availability and industrial availability.

A mineral can exist in the ground, a mine can be permitted, and a concentrate can be produced while the economically decisive conversion step remains elsewhere.

2. The material chain

Marginal Thinking will analyse materials through the full industrial chain:

Each stage has its own capital, energy, technology, environmental, logistics and policy requirements.

For battery materials, the path can extend into cathode and anode materials, cells and packs. For rare earths, extraction and separation are distinct from metal and magnet production. For copper, mine supply is distinct from concentrate treatment, smelting, refining, wire and cable fabrication.

The analytical error is to treat ownership of geological reserves as equivalent to control of the usable material.

3. Copper illustrates the transition from geology to processing

Copper is essential to grids, motors, transformers, buildings, electronics and many forms of electrification.

The IEA still projects a 25% supply deficit in 2035 against anticipated demand based on the current project pipeline. New projects in the DRC and Zambia have narrowed the expected gap, which shows that mining investment matters.

But another constraint appears downstream.

The IEA reports that China accounted for more than 90% of global copper-smelting capacity growth since 2005, increasing its share of global smelting capacity from around 15% to about 50% by 2025. Benchmark copper treatment charges reached USD 0 per tonne in 2026, while spot charges had been negative since 2024, signalling pressure created by rapid smelting expansion relative to concentrate availability. IEA, Global Critical Minerals Outlook 2026

This is a useful example of a system in which the constraint can move.

If mine supply expands faster than refining, processing can become scarce. If smelting expands faster than concentrate supply, smelters compete for feedstock. If refined metal is available but cable, transformer or component manufacturing is slow, the constraint moves again.

The strategic asset is the coordinated chain, not one stage in isolation.

4. Small material markets can support very large downstream value

The IEA identifies strategic minor minerals such as gallium, germanium, tungsten, tellurium and others as particularly vulnerable because their markets are small, production is concentrated and downstream applications can be economically important.

This creates an asymmetric risk.

A relatively small volume of material can constrain high-value output in semiconductors, aerospace, defence, energy equipment or advanced manufacturing. The economic value at risk can therefore be many times larger than the value of the mineral market itself.

The IEA estimates that full implementation of certain rare-earth export restrictions could put trillions of dollars of annual downstream production outside China at risk. It also estimates that a complete disruption of battery-grade graphite trade could place more than USD 300 billion of downstream production outside China at risk. These are scenario estimates, not forecasts that such disruption will occur.

The analytical lesson is that market size is not a good proxy for strategic importance.

5. Diversification is capital-intensive and uneven

Governments and development institutions are increasing financial support. The IEA estimates that public-finance commitments for critical-mineral projects in advanced economies reached roughly USD 65 billion in 2025, more than four times the level in 2023.

Yet announced diversification remains uneven across the chain.

Mining projects outside dominant suppliers can advance while refining, precursor production, magnet manufacturing or other downstream capacity lags. That matters because a diversified mine feeding a concentrated processing network does not fully diversify the industrial system.

Cost differences are also material. The IEA finds that capital costs for refining projects outside the dominant supplier can be 20% to more than 150% higher, depending on the mineral and jurisdiction, while operating costs can be around 50% higher in some cases.

Diversification therefore requires an economic mechanism, not only a strategic objective. Long-term contracts, anchor demand, price floors, tax incentives, public procurement, concessional finance or strategic stockpiles can change project economics, but each mechanism transfers cost or risk to another actor.

6. Energy and materials are mutually dependent

Materials are inputs into the electricity transition, but energy is also an input into materials.

Mining, smelting, refining and advanced materials manufacturing can be electricity- and heat-intensive. A country with mineral deposits but unreliable or expensive energy may struggle to move downstream. Conversely, low-cost reliable electricity can become a comparative advantage for processing.

The link is circular:

That circularity is why this series will not treat critical minerals as a commodity-only topic.

Copper availability affects grid expansion. Grid availability affects copper processing. Industrial policy affects where refining is built. Trade controls affect where manufacturers source inputs. Capital costs determine whether alternative supply becomes commercially viable.

7. New and advanced materials require a stricter test

The same framework applies to newer materials and material systems.

A laboratory result becomes relevant to Marginal Thinking only when the analysis can move beyond a property claim into questions such as:

• can it be manufactured repeatably?
• what feedstocks and purity levels are required?
• what is the yield?
• what is the cost at relevant scale?
• is production dependent on specialised equipment or intellectual property?
• can incumbent materials substitute for it?
• does it change energy density, thermal performance, conductivity, durability or other system economics enough to alter adoption?

The series will distinguish scientific capability, engineering demonstration, commercial deployment and industrial scale.

8. What would change the assessment

The view that strategic constraints are moving downstream would weaken if:

• refining concentration falls materially across several major minerals;
• new downstream plants outside dominant suppliers achieve competitive utilization and cost;
• processing technology becomes sufficiently standardized and transferable to reduce entry barriers;
• substitution materially reduces demand for the most concentrated inputs;
• recycling becomes a large, reliable source of industrial feedstock;
• strategic stockpiles and diversified contracting materially reduce disruption exposure.

It would strengthen if mining diversification proceeds while processing concentration rises, if export controls expand to equipment and processing technology, or if downstream plants outside dominant suppliers persistently fail to reach competitive scale.

9. Tracking framework

Sources

• International Energy Agency — Global Critical Minerals Outlook 2026
• International Energy Agency — Global Critical Minerals Outlook 2026: Executive Summary

Method note. Reserves, mine output, refined output and downstream manufacturing capacity are not interchangeable measures. Concentration is assessed by industrial stage wherever data allow, and disruption scenarios are identified as conditional estimates rather than forecasts.