Niche metals test West’s resilience to Chinese export curbs

Gallium and germanium prices remain at 9-10 times their 2023 levels after China’s export curbs, while China still controls 98.9% of primary gallium supply and 68.6% of germanium supply. AI, fibre-optic and defense demand is tightening markets further, with gallium demand projected to grow about 12% annually through 2030 and germanium demand 3.3% annually. Western projects and more than $3.5B in Australian and U.S. critical-minerals support are accelerating diversification, but projected ex-China capacity will still leave substantial dependence on Chinese supply by 2030.

Analysis

The investable exposure is not broad AI hardware: gallium and germanium are likely immaterial to NVDA’s near-term gross margin, while the sharper risk sits in specialized optical, RF and defense supply chains where qualification cycles prevent rapid material substitution. COHR, LITE, MTSI, QRVO and TDY face the more plausible 1-3 quarter risk of working-capital absorption, expedited sourcing costs and customer delivery delays; defense programs can pass through some cost, but fixed-price contracts cannot do so immediately. This creates a bifurcation between high-value component makers able to surcharge and lower-margin integrators forced to carry inventory.

The announced Western projects should be valued as strategic-option assets rather than near-term earnings drivers. For TECK.A, AA, RIO and MTLN, potential byproduct recovery adds geopolitical relevance and could attract subsidies, offtake agreements and valuation support over 6-18 months, but project execution, recovery yields and purification economics matter far more than headline capacity. The key second-order issue is that new Western supply may require policy-backed price floors or guaranteed procurement: without them, a later normalization in Chinese availability could strand higher-cost capacity and impair returns on invested capital.

Consensus may overstate both the immediate AI linkage and the speed of supply-chain de-risking. Elevated material prices are more likely to accelerate redesign toward indium-phosphide, silicon photonics and alternative infrared substrates than to produce a sustained windfall for every mining major. A meaningful reversal would come from export-license normalization, a visible decline in lead times, or evidence that customers are qualifying substitutes faster than expected; absent those signals, scarcity premia and inventory accumulation should persist through at least 2027.

Key Decisions for Investors

• Maintain a 6-18 month watch-to-buy bias on TECK.A and MTLN, not a momentum entry: initiate only after disclosed offtake/subsidy terms and project-level capex/return guidance. Thesis requires a contracted pricing floor or strategic customer; abandon if recovery economics imply sub-scale, subsidy-dependent losses.
• Prefer a relative-value expression long TECK.A / short diversified-miner proxy RIO over an outright critical-minerals basket for 6-12 months. TECK.A has a more direct potential refining-expansion catalyst, while RIO’s valuation remains dominated by iron ore; cap risk if iron ore strength drives RIO materially ahead or if Canadian support fails to translate into committed capacity.
• Screen COHR, LITE, MTSI and TDY at next earnings for inventory days, material-cost guidance and backlog conversion. A simultaneous rise in inventory and reduction in gross-margin outlook is a short/watch trigger over the following 1-3 months; do not short solely on scarcity because defense and datacom pricing power can offset costs.
• Avoid using NVDA or SMCI as a direct hedge for this theme. Establish a negative catalyst alert only if either company cites optical-component allocation, networking shipment delays, or incremental supply-chain qualification expense; absent that disclosure, the material constraint is too far removed from their core earnings drivers.