A Dependency Investigation (#03) – May, 2026
About This Series
The Magician’s Hands is a series of dependency investigations. Each report examines a single case in which a structural dependency, between a state and an infrastructure owner, a farmer and a seed company, a continent and an energy supplier, was created, normalised, leveraged, and converted into power. The cases span domains and decades. The grammar beneath them does not change.
The series takes its name from a simple observation: the most consequential things happening in the world are rarely the things that take centre stage. While we watch the visible hands, something else is being built in the structural layer underneath. These reports are an attempt to make that layer legible.
The founding article, “”, sets out the full grammar. Each investigation that follows applies it to a specific case.
Scene-setter
There is a story told about rare earths that is technically accurate and structurally misleading. The story goes like this: China has a lot of them. Geography gave China an advantage. Western economies, recognising the risk, are now working to correct it. Diversification is underway. The problem is being addressed.
What this story omits is that geography was never the point. Rare earth deposits exist on every inhabited continent. The United States, Australia, Brazil, India, and several African nations sit on substantial reserves. What China built was not a monopoly on the ground; it built a monopoly in the middle, in the crushing mills, the chemical separation plants, the smelters, the processing infrastructure that transforms raw ore into the refined materials that wind turbines, electric vehicle motors, missile guidance systems, and consumer electronics actually require. The mining is visible and mappable. The processing is the part that cannot be easily seen or quickly replicated, and it is the part that matters.
This investigation follows the creation of that processing monopoly: how it was built deliberately over four decades while the nations that would come to depend on it were looking elsewhere, how it became so embedded in global supply chains that it ceased to register as a dependency at all, and how it was converted, at a moment of its choosing, into geopolitical leverage. The five moves will show that China’s rare earth dominance is not a fortunate accident of geology. It is a patient, coordinated act of structural construction, and the Western world is only now beginning to understand what it consented to while it was not paying attention.
Move 1: Creating the Dependency
The dependency was not captured from an existing holder. It was built from near-nothing, through a decades-long industrial policy whose strategic ambition was stated openly and largely ignored.
The starting point matters. In the early 1980s, China was not a dominant force in rare earth production. The Mountain Pass mine in California was the world’s primary source, and the United States held a commanding position in both mining and processing. What changed was a deliberate policy decision by the Chinese government, under Deng Xiaoping, to develop rare earth processing as a strategic industrial priority. During his Southern Tour in January 1992, Deng stated it directly: “The Middle East has its oil, China has rare earth.” The comparison was not incidental. It was a statement of intent, and it was carved, later, in stone at an industrial park in Inner Mongolia.
What followed was the patient construction of an entire industrial ecosystem. China invested in processing infrastructure, subsidised domestic producers, and, critically, priced its processed rare earths at levels that made competition from other nations economically irrational. Western producers could not match Chinese prices without government support they did not receive. The Mountain Pass mine, once the centre of global supply, ceased active mining operations in 2002, unable to compete on cost and facing mounting environmental regulatory pressure. Processing expertise, once distributed across the United States, Japan, and Europe, gradually concentrated in China because that was where the economics pointed. In 1995, a consortium of Chinese state-owned companies acquired a majority stake in Magnequench, a General Motors subsidiary that had pioneered rare earth magnet manufacture; by 2001, its US production lines had shut down and operations had shifted to China.
The hook for adopting nations was genuine. Chinese rare earths were cheaper, reliably supplied, and available at scale precisely when the global economy needed them most. The 1990s and 2000s saw explosive growth in consumer electronics and the early expansion of clean energy technology. Both sectors required rare earth elements in increasing quantities. Sourcing from China was not naivety; it was rational procurement. The structural consequence, invisible at the point of each individual purchasing decision, was that an entire global industry was outsourcing not just its supply but its processing knowledge, its workforce expertise, and its ability to function independently.
By the period spanning roughly 2005 to 2011, China controlled approximately 97% of global rare earth production. No single policy decision created this position. It was the aggregate result of thousands of rational individual choices made by companies and governments that were optimising for cost and not asking what the map of the whole looked like.
Move 2: Normalising the Dependency
The normalisation of China’s rare earth dominance operated through both mechanisms the framework identifies, and they reinforced each other in ways that made the dependency exceptionally durable.
The narrative that carried the dependency was the narrative of comparative advantage. Free trade theory provided the intellectual infrastructure: countries should produce what they produce most efficiently, and global supply chains should be allowed to find their natural equilibrium. Under this reading, China’s rare earth dominance was not a strategic vulnerability but an example of the global trading system working as designed. Raising alarms about it invited the accusation of protectionism, of wanting to subsidise uncompetitive domestic industries for nationalist reasons. The economists who warned that processing monopolies in strategic materials created asymmetric risks were marginalised not because they were wrong but because their argument was structurally inconvenient for a policy consensus committed to liberalisation.
This narrative was carried by international trade institutions, by the academic economics mainstream, and by the procurement culture of multinational corporations whose performance metrics rewarded cost reduction and penalised supply chain risk premiums that had not yet materialised. The counter-narrative, that strategic dependencies in critical materials required policy intervention regardless of short-term cost, did exist. It was present in security community analysis and in the thinking of some industrial policy advocates. It did not land because it ran against the dominant framework, and because the consequences it predicted were not yet visible.
The operational normalisation was perhaps more consequential than the narrative one. Rare earths stopped being a supply chain input and became infrastructure, simply the way things are made. Design engineers at electronics companies built products around the assumption of available processed rare earths. Wind turbine manufacturers specified permanent magnets whose performance characteristics depended on Chinese-processed neodymium. Defence contractors integrated rare earth components into weapons systems whose specifications were set by the materials available, not the materials that could be sourced independently. Each design decision deepened the dependency without anyone making a decision about dependency at all.
By the late 2000s, the switching cost had become enormous before any leverage had been applied. This is the defining signature of successful normalisation: exit becomes expensive in the absence of any threatening event, purely through the accumulation of reliance.
Move 3: Leveraging the Dependency
The leverage moment arrived in 2010, and it arrived through all three forms simultaneously.
The direct leverage was the export quota reduction. In 2010, China cut its permitted rare earth exports to approximately 30,259 metric tons, a reduction of around 37% from prior permitted levels. The effect on global prices was immediate and severe. Rare earth prices rose from approximately $9,461 per metric ton in 2009 to nearly $66,957 per metric ton in 2011, a sevenfold increase in two years. This was not a market correction. It was a demonstration.
Simultaneously, in the context of a territorial dispute with Japan over the Senkaku/Diaoyu Islands, Japan experienced restricted access to Chinese rare earths for approximately two months in late 2010. The precise mechanism is analytically contested: academic analysis, including work published in peer-reviewed economics journals, has questioned whether China formally singled out Japan in trade data as distinct from the broader global quota cuts being applied simultaneously, noting that the trend of declining export quotas had been underway since 2006 and that prices took a full year after the diplomatic incident to peak. What is not contested is the effect. Japan, which was at that time importing over 90% of its rare earths from China, felt the restriction acutely, and the leverage relationship was made visible in a way it had not been before. Whether the targeting was formal or operational, the message was legible.
The structural leverage operated more quietly. Dependent nations discovered that their policy options in areas adjacent to rare earth supply were constrained by the dependency relationship in ways they had not previously had to calculate. Industrial policy decisions, trade negotiation positions, and technology partnership choices all now carried an implicit variable: how would this affect rare earth access? The dependency had inserted itself into strategic decision-making before anyone had formally decided that it should.
The option-space leverage was the most durable form. Western governments and corporations, observing the 2010 episode, understood that certain postures toward China were now more expensive than they had been before. Some adjusted their behaviour not because pressure was applied but because the possibility of pressure was now part of their planning. No directive was needed. The dependency had already reshaped what felt available.
Move 4: Obscuring the Mechanism
The rare earth dependency was not hidden in any simple sense. The facts were available. China’s growing market share in processing was documented in trade statistics. The Mountain Pass closure was reported. The consolidation of Chinese producers under state coordination was observable. What was missing was not the data but the institutional architecture to interpret it collectively as a strategic problem.
The temporal displacement is central here. The dependency was built over roughly two decades, from the early 1980s to the early 2000s. The leverage was first applied in 2010. That gap spans multiple electoral cycles in every affected democracy, multiple corporate leadership tenures, and a complete generational turnover in the procurement and policy communities that made the original decisions. The politicians and procurement officers who approved the outsourcing of processing capacity were not the politicians and procurement officers who faced the consequences of the 2010 quota cuts. The institutional memory of the choice had dissipated; what remained was a supply chain whose origins were no longer visible to the people operating it.
The narrative suppression was structural rather than active. No coordinated campaign was needed to keep rare earth processing dependency off the policy agenda. The comparative advantage framework did the work on its own. Questions about strategic vulnerability in supply chains were categorised as trade policy questions, which placed them in a framework that presumed the answer: liberalise and trust the market. The question of what happens if a single state controls the processing of a material essential to defence and clean energy was structurally unaskable within that framework, because it presupposed the legitimacy of the concern the framework was designed to pre-empt.
The structural opacity was compounded by the distribution of responsibility. Mining decisions were made by mining companies. Procurement decisions were made by electronics manufacturers. Defence acquisition decisions were made by procurement agencies. Clean energy policy was made by environment ministries. No single institution held the whole picture, and the architecture for assembling it did not exist. The dependency was legible in aggregate and invisible in each of its component parts.
Move 5: Converting the Value
What China converted through its rare earth dominance was not primarily financial, though the financial conversion was real enough. The price spike of 2010 to 2011 transferred substantial value from dependent industries to Chinese producers and to the state through export taxes and quota management. But the financial extraction was the least consequential form of conversion.
The strategic conversion was the more significant one. By demonstrating in 2010 that it could impose severe costs on global supply chains through a unilateral administrative decision, China established itself as a power that held a structural veto over the industrial capacity of its principal geopolitical competitors. The United States, the European Union, Japan, and South Korea all manufacture defence systems, clean energy infrastructure, and consumer technology whose supply chains run through Chinese processing facilities. That is not a trade relationship in any ordinary sense. It is a structural condition that reshapes the balance of options in every negotiation those states conduct with China across every domain.
The normative conversion is subtler. China’s dominance in rare earth processing gave it disproportionate influence over the technical standards that govern how processed materials are characterised, traded, and specified. Standard-setting authority follows processing expertise, and processing expertise had concentrated in China. The dependent party does not just rely on the dominant actor for supply; it relies on it for the technical language in which supply is discussed.
The long-term consequences are still accumulating and accelerating. On 4 April 2025, in direct response to US tariffs announced on what the Trump administration called Liberation Day, China introduced export licensing requirements on seven heavy rare earth elements, as well as all related compounds, metals, and magnets. Carmakers across the United States, Europe, and elsewhere struggled to obtain permanent magnets; some were forced to cut production rates or temporarily halt assembly lines. In October 2025, China escalated further, adding five additional elements to the controlled list and, for the first time, applying extraterritorial reach: products manufactured outside China using Chinese-origin materials or Chinese processing technology now require Beijing’s approval for export. The dependency had been extended into a new domain, not just what China exports but what knowledge and technology others are permitted to use anywhere in the world.
The cost is not borne evenly. The procurement officers who optimised for cost in the 1990s are not the engineers who cannot source components today. The politicians who declined to subsidise domestic processing capacity are not the defence ministers who now face supply chain vulnerabilities in weapons systems. The clean energy transition, urgently needed and politically committed to across the Western world, runs through a processing infrastructure controlled by a state whose geopolitical interests do not align with those of the nations that need it most. The dependency was created in one era and is being paid for in another, by people who made none of the choices that produced it.
Analytical Notes
The WTO ruling of 2014 against China’s export restrictions deserves specific attention as a case of institutional resistance that left the dependency entirely intact. The WTO found China’s export quotas to be in violation of its trade obligations. China removed the quotas in 2015. The dependency did not diminish. Processing capacity remained concentrated. Prices adjusted, but the structural condition that made the leverage possible in 2010 was unchanged. This is a textbook illustration of the grammar’s resilience: legal challenge can remove the visible mechanism of leverage without touching the underlying structural condition that makes leverage possible. The dependency creator does not need the lever to remain in place; it needs the dependency to remain in place. These are different things, and institutions designed to address the former are structurally unable to address the latter.
Closing
What China’s rare earth case reveals about the grammar is this: the most durable dependencies are not the ones that require active maintenance. They are the ones that become self-sustaining through adoption, through the embedding of the dependent relationship into design specifications, procurement cultures, and industrial knowledge bases that outlast any political decision to reverse them. China did not need to threaten anyone. It needed the world to need it, and to need it in a way that felt like simply how things work.
The founding article argued that the most consequential things happening in the world are rarely the things that take centre stage. The rare earth case is a four-decade demonstration of that claim. The transfer of processing dominance occurred in plain sight, documented in trade statistics, visible in the closure of Western facilities, legible in the price data. It was not hidden. It was simply not seen, because the institutional architecture for seeing it collectively as a strategic problem did not exist, and the intellectual framework that dominated policy thinking was specifically designed to redirect that kind of concern.
The question to carry forward is not whether Western diversification efforts will succeed, though that question matters practically. Japan has reduced its overall dependence on Chinese rare earths from over 90% to below 60%, through sustained investment in Australia’s Lynas Corporation and other measures, but its dependence on Chinese processing for heavy rare earths remains acute, and the path others must travel is longer still. The deeper question is what other processing dependencies, in materials, in data infrastructure, in biological supply chains, in the standards that govern digital systems, are currently in the normalisation phase of this same grammar, accumulating switching costs quietly, becoming load-bearing before anyone has decided they should be. The hands you are watching are not the ones doing the consequential work. built. The hands are visible. The question is whether we are watching the right ones.
Comments are closed