Criticality and geopolitics: What criticality means, and why it is not scarcity

1.1 Criticality is about disruption and consequence, not running out

Hayes and McCullough, two USGS scientists, define a critical element as one at risk of supply disruption that also serves a purpose the evaluator considers important. Both halves depend on who is asking. A solar panel maker and a defence agency will draw up different lists, and both will change as technology and politics move. Criticality is a matter of degree and perspective, not a fixed property of an element.

Nor is it scarcity. Scarcity is a long-run question about depletion and rising real costs. Criticality concerns supply that a cartel, a war, an export ban, a strike or a flood could interrupt. Radetzki and Wårell build their chapter on supply security around sudden cuts that last too briefly for economies to adjust. Graedel's group at Yale looks 5 to 50 years ahead, far short of geological exhaustion.

Gallium shows the gap. The USGS estimates that gallium in world bauxite resources exceeds 1 million tonnes, against primary output of about 900 tonnes in 2025 (USGS estimates). Yet China produced 99% of the low-purity primary metal, so one government decision can stop most of the flow. Rare earths make the same point. Scheyder notes that cerium is about as abundant in the crust as copper; what is rare is a deposit rich enough to mine and a plant able to separate it.

The idea is old. In the United States it emerged before the First World War and was written into the Strategic and Critical Materials Stock Piling Act of 1939. Since the 1980s attention has shifted to small-volume elements used in electronics, renewable energy and weapons.

• Hayes & McCullough (2018), Critical minerals: a review, Resources Policy web ↗ · pdf pp. 1-2
• Radetzki & Wårell (2021), A Handbook of Primary Commodities · pdf p. 196
• Graedel et al. (2015), Criticality of metals and metalloids, PNAS web ↗ · pdf pp. 1-2
• USGS Mineral Commodity Summaries 2026, Gallium web ↗ · pdf pp. 86-87
• Scheyder (2024), The War Below · pdf p. 105

1.2 A supply cut hurts most when imports are high, substitutes are poor and supply is concentrated

Radetzki and Wårell list the conditions that turn a supply problem into a crisis. Four sit with the importer. The first is import dependence: if imports are a small share of use, only marginal uses go short. The second is the import's value relative to the economy, which is why oil can move inflation and growth while strategic metals, with tiny trade values, cannot. The third is substitutability, which has a cost and a time dimension. Silver conducts as well as copper but costs about 70 times as much per kilogram, so copper could rise a long way before silver became an economic replacement, and equipment built around one material may need rebuilding before another can be used. The fourth is how essential the end product is: a nickel disruption is worse than a cocoa disruption because stainless steel is harder to do without than chocolate.

Two conditions sit with the suppliers. Concentrated exports magnify the effect of droughts, strikes or upheavals and make cartels easier to organise. And the harder it is to bring in new sources, because capacity is already busy or new mines take years, the longer a disruption lasts.

A cut first shows up as a price rise. Then come rationing, closures among users who cannot pay, a scramble to substitute, and later relief as substitution and savings work through. The damage is greatest in the short run. A country hit alone by a targeted embargo suffers more than one sharing a global shortage, but only if political controls stop the market from reallocating supply to it.

• Radetzki & Wårell (2021), A Handbook of Primary Commodities · pdf pp. 197-201
• Nassar, Graedel & Harper (2015), By-product metals, Science Advances web ↗ · pdf p. 6

1.3 Damaging disruptions are rare, but fears of them come in waves

Physical supply is much steadier than price. Radetzki and Wårell find no sharp man-made disruptions in global farm output; the largest year-to-year falls, up to 15%, were in coffee and cocoa, whose production is geographically concentrated. In minerals they count only four large man-made disruptions in roughly 40 years.

The first was cobalt. In 1976 upheaval in Zaire, which produced more than half the world's cobalt, cut world output by almost 20%. The producer price rose with a lag from US$5.40 a pound to US$25 later in the decade and stayed high for four years. The costly substitution it forced worked: US demand fell 53% between 1978 and 1982, and by 1983 output and price were back to pre-crisis levels. The metal proved dispensable at a price. The second was oil in 1979-80, when the Iranian revolution and the Iran-Iraq war cut world output from 64.6 to 58.2 million barrels a day and the price more than doubled in five months. The third was China's cut in rare earth exports in 2010. The fourth was the drone attack on Saudi facilities in September 2019, which removed 5.7 million barrels a day, about 5% of world supply, for less than a month.

Concern nevertheless comes in waves after scares: the cartel attempts of the early 1970s, oil and gas in the 2000s, rare earths in the 2010s, and now battery metals and China's export controls. Radetzki and Wårell conclude that the cost of security measures can look exaggerated in hindsight, but that history gives no firm guide to a future crisis of great severity, against which even costly insurance might pay.

• Radetzki & Wårell (2021), A Handbook of Primary Commodities · pdf pp. 203-205, 215

See it in the table

Check your understanding

Gallium contained in world bauxite resources exceeds 1 million tonnes, while annual primary output is about 900 tonnes. Why is gallium still treated as critical?

If the short-run price elasticity of demand for a metal is 0.1, roughly what price change is needed to clear a 10% supply shortfall?

What did the cobalt crisis of 1976-83 reveal, according to Radetzki and Wårell?