The US Has the Geology and the Capital. What It Doesn’t Have Is the Industrial Muscle Between the Mine and the Finished Product.

July 24, 2026

Between 1965 and the mid-1980s, the United States produced approximately 15,000 metric tonnes of rare earth elements per year. That was roughly three times the rest of the world combined. Mountain Pass in California was the centre of global rare earth supply, and American metallurgists and chemical engineers were the best in the world at separating and refining these materials. That expertise does not exist anymore.

Washington is spending billions to secure critical mineral supply chains, but US rare earth processing capacity remains the binding constraint, not mining. The policy response runs through executive orders, expanded critical minerals lists, permitting reform, and defence production authorities. The US currently has two domestic rare earth mining locations. They produced approximately 51,000 metric tonnes of rare earth concentrates in 2025. The country still imported roughly 21,000 metric tonnes of rare earth compounds, most of them from China. The gap is not in the rock. It is in the refinery.

What Mountain Pass Was and What It Became

Mountain Pass in the Mojave Desert of southern California was the world’s dominant rare earth mine from the 1960s through to the late 1990s. At its peak, the operation produced more rare earth oxide than the rest of the world combined. The metallurgical knowledge required to extract, separate, and process the 17 rare earth elements, which share very similar chemical properties that make separation technically challenging, was developed in American university and government research programmes and commercialised at Mountain Pass.

China’s emergence as the dominant rare earth supplier in the 1990s and 2000s was driven by a combination of lower production costs, looser environmental standards, and deliberate government policy to subsidise domestic production and undercut international competitors. Mountain Pass became unprofitable, closed in 2002, and reopened briefly under different ownership before closing again in 2015. The closure was not simply an economic event. It was a knowledge event. Two decades on, the case for rare earth refining capacity outside China is still being made against the same accumulated Chinese advantage.

When Mountain Pass closed, the workforce dispersed. The metallurgists and process engineers who had developed the separation chemistry moved on or retired. The university programmes that fed talent into rare earth processing were not maintained at the scale that a functioning industry requires. Two decades of no commercial-scale rare earth processing in the United States produced a gap not just in physical infrastructure but in the embodied knowledge required to run a competitive separation operation.

Why US Rare Earth Processing Cannot Be Solved With Capital Alone

Building a rare earth processing facility takes years of permitting, highly specialised equipment, and a workforce trained in metallurgy, chemical engineering, and hydrometallurgical plant operation. The people who can separate neodymium from praseodymium or run a solvent extraction circuit at commercial scale are not sitting idle waiting for a policy signal. Those skills take years to develop and require active industry to train them in.

Washington’s policy response treats the problem primarily as a capital allocation and permitting issue. Executive orders direct investment into rare earth supply chains. The Department of Defense has provided funding to MP Materials for Mountain Pass expansion and processing capability development. The IRA creates domestic content incentives that make US-processed rare earths more valuable for EV manufacturers seeking tax credits.

These instruments are necessary. They are not sufficient. Capital can fund equipment. Permitting reform can accelerate facility approvals. Neither solves the workforce development problem that sits at the centre of the processing gap. A rare earth separation facility with capital, permits, and equipment but without the process knowledge required to operate it efficiently is not a functional processing operation. The learning curve at an industrial hydrometallurgical plant is measured in years of operating experience, not months of training.

This is the kind of analysis we publish daily in The Drill Down.

Subscribe free →

What Canada and Australia Are Doing Differently

Canada has started connecting mining projects directly to battery manufacturing through funded processing hubs and workforce training tied to specific industries. The model makes processing capacity and the workforce to run it conditions of the same funding agreement, rather than treating them as separate policy domains. When a processing facility is funded, the training programme for the workforce that will operate it is funded alongside it, with commitments from manufacturers who will consume the output.

Australia is combining critical minerals incentives with expanded university and vocational programmes in metallurgy and mineral processing. Curtin University’s minerals separation research, the University of Adelaide’s hydrometallurgy programmes, and TAFE-level vocational training in mineral processing are being coordinated with government-funded processing infrastructure to build the human capital pipeline alongside the physical capital. State-level frameworks are doing the coordinating work, as Queensland’s State Strategic Projects legislation for critical minerals does for approvals and enabling infrastructure.

These models reflect an understanding that the processing gap is a knowledge and workforce problem as much as a capital and permitting problem. They are not yet operating at sufficient scale to create the same depth of processing expertise that China has accumulated over three decades. But they acknowledge the nature of the problem in a way that US policy, which continues to treat capital and permitting as the primary levers, has not fully incorporated.

A Generational Knowledge Problem With a Generational Solution

The United States let rare earth processing expertise atrophy over three decades while China was quietly building it. That atrophy cannot be reversed by a funding announcement. It requires rebuilding an industry that trains people in the specific skills required, maintains operating facilities where those skills can be practised and transmitted, and sustains that industry through market cycles that will inevitably create periods where government support is required to keep commercial operations open. The scale of what has to be matched is set out in the rare earth refining bottleneck, where Chinese control of heavy rare earth separation runs at 98 to 99%.

The US has the geology. Mountain Pass and the Rare Element Resources deposits represent genuine rare earth resources at commercial scale. The US has the capital markets. MP Materials, Lynas’ North American processing ambitions, and a range of smaller developers have access to public and private capital. The US has the research institutions. NREL, AMES Laboratory, and multiple university programmes are advancing rare earth separation chemistry.

What the US does not have is the industrial muscle between the mine and the finished product. That is not a funding problem, and it cannot be solved on a funding timeline. It is a generational knowledge problem. The solution is generational: it requires building an industry, training a workforce, and sustaining both through cycles for long enough that the knowledge becomes embedded in institutions rather than dependent on specific individuals. The parallel with China’s own development of rare earth processing, which took three decades, is not accidental.

This analysis is from The Drill Down, a daily briefing on critical minerals, junior mining, and capital markets. Join 3,200+ investors and operators who read it before the market opens.

Subscribe free →

Kamoa Capital is not licensed and does not know your circumstances. Nothing here is financial, legal or tax advice – seek professional advice and read any PDS before acting. We aim for accuracy but make no guarantees and accept no liability. Views are opinions only and may include forward-looking statements that may not occur.