Australia's China Problem

In 1935, off the coast of Western Australia, a Japanese mining company chaired by a former Imperial Japanese Navy admiral signed a lease over Koolan Island. The deal had to be run through a British front company, because Australian law forbade direct Japanese investment in iron ore. It would have been the largest iron-ore venture then attempted east of Suez.

Three years later, the Lyons government banned all iron-ore exports to Japan, on strategic grounds. Seven years after that, Japanese aircraft bombed Darwin. The iron that was supposed to bind Japan to peace had become its foundry.

The argument made in Canberra for shipping ore to Nippon Mining in 1937 is the argument made in Canberra for shipping ore to Chinese state-owned mills in 2026. Commerce will restrain them. Dependency is mutual. Severance would cost us more than them. It is the same case, made from the same assumptions, with the track record of the last time it was tested on file.

I. What coercion actually looked like

From November 2020, Beijing ran a calibrated campaign against seven Australian commodities. No written directive was ever issued. Instructions moved through state media and in verbal guidance to customs officials. An 80.5 per cent barley tariff went on in May 2020. Wine tariffs of between 116 and 218 per cent followed in March 2021. By 6 November 2020, some sixty-six bulk carriers loaded with Australian coal sat at anchor off Chinese ports, their cargoes undischarged. Rock lobsters worth about $700 million a year, almost all of it to China, were left to die on tarmacs. Timber, copper concentrate, cotton, and selected beef abattoirs followed. The Lowy Institute estimates the campaign cost Australian exporters around $20 billion at its peak.

The two commodities Beijing did not sanction were the two it could not replace: iron ore and LNG. The campaign was not chaotic. It was a demonstration of calibrated economic warfare, conducted by a state that had identified exactly which of Australia's exports it could afford to punish.

Most Australian commentary drew the wrong lesson. Because exports rebuilt – barley reconcentrated to China within eighteen months of the tariffs lifting, iron-ore revenue rose through the campaign – engagement-era voices argued the system had held. The lesson, on their reading, was that Beijing could not, or would not, really hurt us. That is the reading of someone who has just been hit and is telling themselves the next hit will be weaker. It is also the reading pressed publicly by James Laurenceson at the Australia-China Relations Institute at the University of Technology Sydney, and by Geoff Raby, the former ambassador to Beijing. Raby now sits on the board of Yancoal, an Australian subsidiary of a Chinese state-owned coal company, and has registered himself as an agent of foreign influence under Australian law. That disclosure is his own.

ACRI itself has a disclosure problem. The institute was founded in 2014 with an A$1.8 million donation from Huang Xiangmo, a Chinese property developer whom the Australian government banned from re-entering the country in 2019 on foreign-influence grounds. Huang was ACRI's founding chairman, and the person who chose Bob Carr as its first director. He also chaired the Australian Council for the Promotion of Peaceful Reunification of China, an overseas arm of the Chinese Communist Party's United Front Work Department. State-owned Bank of China and China Construction Bank are among ACRI's more recent donors. None of this makes the institute's work wrong. It does mean it should be read with an awareness of who paid for the building it is housed in, and why.

II. Why engagement was misread from the start

The liberal-interdependence theory was tested on China between 2001 and 2020. It came back with a negative result. WTO accession was meant to embed Beijing as what Robert Zoellick, as US Deputy Secretary of State, called a "responsible stakeholder". Instead the Chinese Communist Party learned that international rules applied to it about as much as they applied to its 218 per cent wine finding.

Commerce did not democratise China. Commerce financed the People's Liberation Army Navy's blue-water fleet, the Belt and Road, the surveillance architecture in Xinjiang, and the island-building campaign in the South China Sea.

John Mearsheimer's position on this has been steady since the early 2000s. In The Tragedy of Great Power Politics (2001), he argued that a rising power in a region without a peer-competitor will seek hegemony. In "Can China Rise Peacefully?" (The National Interest, October 2014) he wrote that "China's rise is unlikely to be tranquil". By 2021, reviewing two decades of Western policy in Foreign Affairs, he concluded that "engagement may have been the worst strategic blunder any country has made in recent history". Mearsheimer was right. The engagement consensus that built Australian and American policy against his warning was wrong.

III. A material-capability seeker, not an evil empire

It is important to say what this is, and what it is not. This is not the case that the CCP is uniquely wicked. It is the case that the CCP is uniquely well-placed, and perfectly rational. All great powers, democratic or not, compete for power in what is, at root, a zero-sum game. China is doing what a continental state with 1.4 billion people, a one-party system, and fourteen sensitive land borders does. It builds capability. It secures supply. It presses the weak points it can find.

Understanding this requires recognising that the society running China is not reading the liberal international order the way we read it. In September 2001, in Shanghai, I watched the first plane hit the North Tower with two Chinese friends – both in their mid-twenties, both educated, one a doctor, one a Tsinghua engineer. When the second tower fell, they cheered. Reflexively. There is a saying among people who have lived there: the longer you are in China, the less you understand about it. That was true that night, and it has not been untrue since.

None of which makes China a rogue state. It makes China a country. The error in Western policy for twenty years was to confuse polite interaction in English with agreement about what world we were both in.

IV. The material map Canberra has not drawn

Canberra's reports on China exposure run to thousands of pages. None of them is a map. An integrated map would show, for each Australian export, the share going to one buyer, the alternative buyers available, and the time it would take to redirect supply in a shock. It would also show, for each critical import, the share sourced from one supplier and the time to stand up an alternative.

The numbers are uncomfortable. Of Australia's iron ore, about 74 per cent goes to China; of Western Australia's iron ore in the 2023-24 financial year, 84 per cent ($116 billion worth) went there. Of our lithium spodumene, about 97 per cent is refined in China. Of our wool, 84 per cent. Of our rock lobster, 90 per cent. Running in the other direction: China refines nineteen of the twenty critical minerals the International Energy Agency tracks. It controls 92 per cent of magnet rare-earth refining. It controls 95 to 99 per cent of battery-grade graphite. The Albemarle lithium-hydroxide refinery at Kemerton in Western Australia, in which more than US$4 billion has been invested, was idled in February 2026. Kwinana is running at about a quarter of its design output.

The Pilbara advantage itself is narrowing. Rio Tinto's benchmark blend now grades around 60.8 per cent iron. Simandou in Guinea – ramping to 120 million tonnes a year by 2028-29, all contracted to China – grades 65 to 66 per cent. Steelmaking is moving toward direct-reduced-iron and green-hydrogen routes that require at least 67 per cent iron content. The Pilbara's moat is not quality, it is price and proximity. That is a real moat. It is not a permanent one.

V. What resilience actually looks like

Resilience is not severance. The world does not run on bravado. But it also does not run on complacency. The right target is not self-sufficiency; it is plural supply on both sides. Plural buyers for our exports. Plural suppliers for our critical inputs. Plural processors for the minerals we mine. The test of Australian policy is whether, in twelve months, the country could redirect a quarter of its exposure without crippling any of the industries involved. For most commodities, at the moment, it could not.

The machinery is being built, slowly. The Critical Minerals Production Tax Incentive was legislated in February 2025, worth about $7 billion over the decade. The Critical Minerals Strategic Reserve, $1.2 billion, becomes operational in the second half of 2026. Iluka Resources' Eneabba refinery is due to produce separated rare earths from 2026. The United States-Japan-Australia trilateral gallium venture at Alcoa's Kwinana site is in construction. These are directional. They are not sufficient, and they are not coherent.

What they lack is conviction, and conviction is cheaper than capability. The United States, under both parties, has a standing strategic interest in critical-minerals supply outside Chinese control, and is willing to put price-support agreements and long-term offtake contracts against it. Japan and Korea are building analogous programs. The question an Australian government should be asking is not whether critical-minerals sovereignty is achievable. It is what the cost would be if Canberra acted as though it were, and what the bill would have to look like to deliver it. That answer is knowable. It has not been seriously asked.

VI. The test Canberra has chosen not to run

The test of a country's strategic literacy is whether it has modelled the shocks that would define its next decade. Bloomberg Economics in January 2024 modelled a Taiwan Strait conflict: in a full-war scenario, global GDP falls 10.2 per cent in the first year; Japan falls 13.5 per cent; Korea falls 23.3 per cent. In a blockade scenario, global GDP falls 5 per cent.

Australia's equivalent numbers are not in the Bloomberg report. The Australian Strategic Policy Institute has done useful work on the shape of the crisis in its "What if…?" report; the Lowy Institute has written on the policy envelope; the Australian Institute of International Affairs has considered preparedness. What has not been published, in any document the Reserve Bank of Australia or the Commonwealth Treasury has been willing to release, is the government's own estimate of what a Taiwan Strait contingency would cost the Australian economy.

The world's most China-exposed advanced economy has not told its public what its primary strategic risk would cost it.

It is not that the work is difficult. The Japanese and Korean figures are on the public record, and the methodology is the same. The gap is not technical. It is political. The answer, once published, would produce policy consequences that a decade of Australian commentary and successive Australian governments have preferred not to confront. You cannot be held to account for a number you have refused to calculate.

VII. What to do

Canberra should publish a Commodity Exposure Register. One document, updated annually, tabled by the Treasurer, covering every major export and every critical import. For each entry: the share of trade going to one buyer or coming from one supplier; the available alternatives; the time it would take to redirect; and a dissent footnote where DFAT's reading and Defence's reading differ. Public, legislated, uncomfortable.

Beijing does not need to read it. The register is not for Beijing. It is for Canberra, and it is for the Australian public, who are entitled to know which of the commodities paying for their schools and hospitals also serve as pressure points they did not know existed. The point of the register is not to cut trade. The point is to know what we are holding.

In 1938, the country that knew what it was holding banned iron-ore exports to Japan. It did so late, but it did so. The country that had built its prosperity on the trade understood, in the end, that commerce was not the same as peace. That understanding took a war to ripen. It should not have to ripen that way twice.

---

Sources

• John Mearsheimer, The Tragedy of Great Power Politics, W. W. Norton, 2001.
• John Mearsheimer, "Can China Rise Peacefully?", The National Interest, October 2014.
• John Mearsheimer, "The Inevitable Rivalry: America, China, and the Tragedy of Great-Power Politics", Foreign Affairs, November/December 2021.
• Richard McGregor et al., China's Coercion Playbook, Lowy Institute, 2022.
• Australian Bureau of Statistics, International Trade Supplementary Information, Calendar Year 2024.
• Department of Foreign Affairs and Trade, China Country Brief, updated 2025.
• Department of Industry, Science and Resources, Resources and Energy Quarterly, December 2024.
• International Energy Agency, Global Critical Minerals Outlook 2024.
• Bloomberg Economics, "Taiwan Strait Conflict Would Cost Global Economy $10 Trillion", January 2024.
• S&P Global Commodity Insights, "Iron ore industry challenges Simandou's 'Pilbara killer' status", 5 January 2026.
• Australian Strategic Policy Institute, What if…? Economic consequences for Australia of a US-China conflict over Taiwan.
• Lowy Institute, Taiwan Flashpoint: What Australia Can Do to Stop the Coming Taiwan Crisis.
• Clive Hamilton, Silent Invasion: China's Influence in Australia, Hardie Grant, 2018.
• Clive Hamilton and Mareike Ohlberg, Hidden Hand: Exposing How the Chinese Communist Party Is Reshaping the World, Hardie Grant, 2020.
• University of Technology Sydney, ACRI Funding disclosures.
• Commonwealth of Australia, Foreign Influence Transparency Scheme Public Register.
• R. G. Menzies, The Forgotten People broadcast address, 3 April 1942.
• Wayne Lee, "The Yampi Sound Iron Ore Embargo, 1938", Australian Mining History, vol. 18.