Neodymium

Market Overview: China Mines 69% and Refines Roughly 90% of the World's Neodymium

Reserves by country: China leads, but Brazil and Australia hold large undeveloped bases

Per USGS Mineral Commodity Summaries 2026, global rare-earth reserves totalled more than 75 million tonnes of REO equivalent at end-2025, with China holding 44 million tonnes (roughly 52% of the world total). Brazil follows with an estimated 21 million tonnes, then Australia at 6.3 million tonnes, Russia at 3.8 million tonnes, Vietnam at 3.5 million tonnes, the United States at 1.9 million tonnes, and Greenland at 1.5 million tonnes (Supply Chain Informs summary of USGS MCS 2026 data). Neodymium and praseodymium are not separately reported by USGS — both are light rare earths (LREE) grouped inside the bastnaesite- and monazite-hosted ore bodies that dominate these national totals, typically representing 15–20% of contained REO in a bastnaesite deposit such as Bayan Obo or Mountain Pass.

Mine production by country, 2023–2025

Source: USGS MCS 2026. World mine production grew roughly 2.6–3% year-on-year in 2025, driven almost entirely by the 12% increase in U.S. output (Mountain Pass) rather than any change in China's flat production quota (CTIA Group, compiling USGS data). China's mine-production share has held remarkably stable near 69–70% for three straight years even as reserves data show China's share of the reserve base is only about 52% — a gap that reflects China's decision to keep extracting at a high, quota-managed rate rather than any physical resource constraint.

Refined separation capacity: the real chokepoint, not mine output

Mine production numbers understate China's true leverage. The critical bottleneck for neodymium supply is not ore extraction but chemical separation — isolating individual rare-earth oxides from a mixed concentrate via solvent extraction. China holds an estimated 85–90% share of global separation and refining capacity, meaning that even rare-earth concentrate mined in the United States, Australia, or elsewhere has historically been shipped to China for separation into usable NdPr oxide (Supply Chain Informs, June 2026). The Congressional Research Service states plainly that China “mines about 60% and processes and separates about 90% of global REEs,” and that Europe depends on China for 98–99% of its rare-earth supply, a dependency the CRS calls structurally unlikely to change quickly (Supply Chain Informs).

U.S. import reliance and the 2025 partial reversal

USGS reports U.S. net import reliance on rare-earth compounds and metals fell from >90% in 2023 to an estimated 67% in 2025, down from 53% in 2024 but still far above the near-total reliance of the early 2020s (USGS MCS 2026). The swing reflects Mountain Pass's ramp-up of domestic separation and metal-making capacity rather than a reduction in end-use demand. U.S. domestic mine output reached an estimated 51,000 tonnes REO in 2025, up 12% from 45,500 tonnes in 2024, valued at $240 million, with bastnaesite mined as a primary product at Mountain Pass, California, and monazite recovered as a byproduct of heavy-mineral-sand mining in the southeastern United States (USGS MCS 2026).

Supply Chain: From Bayan Obo Concentrate to Sintered NdFeB Magnet Block

Bayan Obo, Sichuan, and Shandong: China's three production heartlands

China's rare-earth mine output is concentrated in three geographies with different ore chemistry and different corporate control. Bayan Obo, near Baotou in Inner Mongolia, is the world's largest bastnaesite-hosted light-rare-earth deposit and is mined as an iron-ore byproduct by Northern Rare Earth (Baotou Steel's rare-earth subsidiary), making it the anchor supply point for light rare earths including lanthanum, cerium, praseodymium, and neodymium. Ion-adsorption clay deposits in Jiangxi, Fujian, Guangdong, and southern China supply most of China's heavy rare earths (dysprosium, terbium), while hard-rock deposits in Sichuan (Liangshan/Mianning) provide a second major light-rare-earth source, and processing/separation hubs in Shandong (home to Shandong-based separators and magnet fabricators) convert concentrate from both northern and southern China into oxides, metals, and alloy. China's own government-set 2023 quota allocated 235,850 tonnes for light rare earths and 19,150 tonnes for heavy rare earths across mining, with a near-matching smelting/separation quota, administered by just four state-designated corporate groups (USGS 2023 Minerals Yearbook, China chapter).

MP Materials: Mountain Pass to Fort Worth, the only vertically integrated Western producer

MP Materials operates the Mountain Pass mine and processing facility in San Bernardino County, California — the only active rare-earth mining operation in the United States — with current capacity to produce roughly 40,000 metric tonnes of total rare-earth oxide (TREO) annually, about 11.5% of the global market, and plans to expand upstream capacity toward 60,000 tonnes TREO (Payne Institute, Colorado School of Mines). Downstream, MP's Independence facility in Fort Worth, Texas, converts NdPr oxide into metal, alloy, and finished sintered NdFeB magnets, beginning commercial-scale magnet production in January 2025 (Dallas Business Journal, 22 Jan 2025). On 10 July 2025, MP Materials and the U.S. Department of Defense announced a multibillion-dollar public-private partnership: a $400 million DoD equity investment (with a commitment for up to $350 million more) making DoD MP's largest shareholder at roughly a 15% as-converted stake, a $150 million DoD loan to expand heavy-rare-earth separation (dysprosium, terbium, samarium) at Mountain Pass, a $1 billion private construction loan from JPMorgan Chase and Goldman Sachs, and construction of a second magnet plant — the “10X Facility” — expected to begin commissioning in 2028 (North American Mining, 11 Jul 2025; MP Materials 8-K, 10 Jul 2025). Once complete, MP's total U.S. magnet manufacturing capacity will reach an estimated 10,000 metric tonnes per year, up from 1,000 tonnes previously, alongside a 10-year DoD price floor of $110/kg for NdPr products — roughly double the prevailing Chinese domestic price — structured as a contract-for-difference under which DoD pays the shortfall when market prices sit below the floor and receives 30% of any upside above it (MP Materials 8-K, 10 Jul 2025).

Lynas Rare Earths: Mt Weld ore to Kalgoorlie cracking-and-leaching to Kuantan separation

Lynas Rare Earths is the largest rare-earth producer outside China, mining concentrate from its Mt Weld deposit in Western Australia, performing cracking and leaching at its Kalgoorlie processing facility, and completing final separation into individual rare-earth oxides at its Kuantan, Malaysia plant — historically the world's largest rare-earth separation facility outside China. In 2025, Lynas achieved a landmark expansion of its downstream capability: on 19 May 2025 the company announced it had successfully produced its first dysprosium oxide in Malaysia, and on 18 June 2025 confirmed first terbium production at Kuantan — the first heavy-rare-earth separation achieved anywhere outside China at commercial scale (Yahoo Finance/Lynas, 19 May 2025; Mining.com.au, 18 Jun 2025). Lynas' quarterly report for the period ended 30 June 2025 confirms continued ramp-up of the Mt Weld concentration plant and the Kalgoorlie rare earths processing facility feeding both light rare-earth and the new heavy-rare-earth separation lines (Lynas Rare Earths, Quarterly Report, 30 Jun 2025).

Intermediate products: from mixed concentrate to NdPr oxide, metal, and alloy strip

The supply chain runs through four discrete intermediate products, each historically a separate chokepoint. First, mixed rare-earth concentrate (roughly 60–70% total REO) comes off the flotation circuit at the mine. Second, solvent extraction (cracking and leaching, then hundreds of mixer-settler stages) separates the concentrate into individual oxides — lanthanum oxide, cerium oxide, and crucially neodymium-praseodymium (NdPr) oxide, typically specified at a 75:25 Nd:Pr ratio matching natural ore abundance. Third, metallization (molten salt electrolysis or calciothermic reduction) converts NdPr oxide into NdPr metal or alloy ingot. Fourth, strip casting and powder metallurgy (jet milling, pressing, sintering, and magnetizing) converts NdPr-iron-boron alloy into finished sintered magnet blocks, often with dysprosium or terbium grain-boundary diffusion added at this final stage for high-temperature grades. Fastmarkets publishes daily assessments for both stages: Neodymium-Praseodymium oxide 99% (75:25 ratio), fob China and Neodymium-Praseodymium metal (Nd 75%/Pr 25%), fob China.

End Uses: NdFeB Magnets Now Dominate Demand, Led by EV Traction Motors and Wind Turbines

EV traction motors: 1–3 kg of NdFeB per vehicle, ~99% market share for PMSM

Within battery and hybrid electric vehicles, permanent-magnet synchronous motors using NdFeB rotors have captured near-total market share. Academic literature estimates that NdPr magnet motor technology is used in approximately 99% of all passenger EVs, with each vehicle motor typically containing 1–2.5 kg of sintered NdFeB, rising to as much as 8 kg in performance-oriented dual-motor vehicles such as the Tesla Model S Plaid (Semantic Scholar, NdFeB permanent magnet uses and projected growth; AIM Magnetic, Jul 2025). The U.S. Department of Energy's own supply-chain analysis projected that 90–100% of battery and hybrid EVs would use synchronous traction motors with NdFeB magnets by 2025, each requiring 1–2 kg of permanent-magnet material (U.S. Department of Energy, Rare Earth Permanent Magnets supply chain report). Market research from Mordor Intelligence puts battery-electric-vehicle consumption of NdFeB at 2,800 tonnes in 2024, projected to exceed 8,000 tonnes by 2030 as annual global EV sales approach 30 million units, citing IEA's Global EV Outlook 2025 (Mordor Intelligence, citing IEA Global EV Outlook 2025). The dedicated EV NdFeB magnet market is estimated at $5.3 billion in 2025, growing to $9.5 billion by 2030 at a 12.4% CAGR (GlobeNewswire, EV Magnet Market Insights, 2 Apr 2026).

Wind turbines: direct-drive PMSG designs use several hundred kilograms of NdFeB per megawatt

Direct-drive permanent-magnet synchronous generators (PMSG) eliminate the gearbox in a wind turbine's nacelle, trading mechanical complexity and maintenance cost for a heavy reliance on NdFeB magnet material — typically several hundred kilograms of sintered magnet per megawatt of rated capacity, concentrated in the generator rotor. This design has become the dominant architecture for offshore wind, where minimizing maintenance visits matters more than magnet cost, and is a major driver behind the projected sixfold increase in rare-earth demand for high-performance magnets by 2030 that the European Commission cites in justifying the Critical Raw Materials Act's permanent-magnet circularity provisions (European Commission, CRMA Q&A, IP_24_2749). Wind and EV traction together are treated by the U.S. Department of Energy as the two demand segments that dominate future growth in the “energy sector critical materials” category of NdFeB consumption, projected to grow far faster than legacy consumer-electronics and industrial motor demand (U.S. Department of Energy supply chain report).

Industrial motors, HDDs, and MRI: the legacy base that keeps volume steady

Before EVs and wind turbines became the dominant growth drivers, NdFeB demand was anchored by three large, mature applications: industrial motors and actuators, hard-disk-drive voice-coil motors, and MRI magnet assemblies. The DOE's 2024 supply-chain report notes that “consumer electronics and industrial motors make up the largest share of current NdFeB magnet demand,” even as wind and vehicle demand represent the fastest-growing segments (U.S. Department of Energy, Rare Earth Permanent Magnets report). Consumer electronics alone — speakers, headphones, smartphone haptics and cameras — represents almost 30% of NdFeB demand due to a large and continually expanding device base, even though the magnet content per device is small (Semantic Scholar, NdFeB uses and growth rates). HDD voice-coil motors and MRI systems remain structurally important because they are simultaneously (a) large single-unit consumers of high-grade magnet material and (b) the leading feedstock streams for magnet-to-magnet recycling programs (see Section 6).

Magnet-type market share: NdFeB versus ferrite and samarium-cobalt

Across the broader magnetic-materials market, NdFeB commanded roughly 44.2% of global volume share in 2025 according to Mordor Intelligence, with ferrite (a rare-earth-free, lower-performance alternative) projected to grow at a slower 6.28% CAGR through 2031 as NdFeB continues to take share in performance-critical applications (Mordor Intelligence, Magnetic Materials Market Report, Feb 2026). Within the narrower rare-earth magnet category specifically (excluding ferrite), NdFeB's dominance is far starker: one industry estimate puts NdFeB at 85% of rare-earth magnet type share and 86% of application share in 2025, versus samarium-cobalt magnets, which retain a smaller niche in extreme-temperature aerospace and defense applications where their lower energy product is offset by superior thermal stability above 300°C (IMARC Group, Rare Earth Magnet Market). China holds an estimated 55% regional revenue share of the rare-earth magnet market, reflecting its vertically integrated mine-to-magnet supply chain, with Japan a distant second at 18.2% (IMARC Group).

Why EV and Wind Magnets Need Dysprosium and Terbium, Not Just Neodymium

Pure NdFeB alloy loses coercivity — its resistance to demagnetization — rapidly as temperature rises, which is a serious problem inside an EV traction motor or a turbine nacelle where operating temperatures regularly exceed 150°C. The commercial fix, used across the industry, is to substitute a fraction of the neodymium in the magnet's grain-boundary phase with dysprosium or terbium, both heavy rare earths that dramatically raise the magnet's intrinsic coercivity and thermal stability at the cost of a lower remanence and a much higher raw-material bill.

Grain-boundary diffusion versus bulk alloying: two doping routes, different Dy/Tb intensity

Bulk alloying — mixing dysprosium directly into the melt before strip casting — is the older, simpler approach but consumes the most heavy rare earth per unit of coercivity gained, typically requiring 6–10% dysprosium by weight for high-temperature automotive grades. Grain-boundary diffusion (GBD), the more advanced route pioneered in Japan and now used across leading magnet makers, coats the sintered magnet block's surface with a dysprosium- or terbium-bearing compound and diffuses it selectively into the grain boundaries via heat treatment, concentrating the heavy rare earth exactly where it is needed to block reverse magnetic domains while leaving the grain interior richer in neodymium. This can cut total dysprosium content by roughly half for an equivalent coercivity gain, which is why GBD process licensing and know-how has become a genuine competitive moat for magnet producers trying to manage heavy-rare-earth cost and supply exposure.

DOE's own material-intensity data: dysprosium content by magnet grade

The U.S. Department of Energy's neodymium magnet supply-chain report tabulates magnet grades by maximum operating temperature and corresponding heavy-rare-earth content, showing that magnet grades rated for 200–220°C service — the range required for “hybrid and electric traction drives, high-temperature motors and generators” — require roughly 8.5–11% dysprosium and 19–21.5% combined heavy-rare-earth content by some measures, confirming that the highest-temperature-rated magnets used in traction motors are precisely the grades most exposed to dysprosium and terbium supply risk (U.S. Department of Energy, Rare Earth Permanent Magnets report, Table 3).

Why this matters for the 2025 export-control story

This doping dependency is the direct link between neodymium-magnet manufacturing and China's April and October 2025 export-control actions (detailed in Section 5): the controls targeted dysprosium, terbium, samarium, gadolinium, lutetium, and other heavy/mid rare earths specifically because they are the doping elements that make high-temperature NdFeB grades viable for EV and defense applications — not because of neodymium itself, which remained comparatively uncontrolled throughout 2025. MOFCOM's April 2025 notice explicitly listed “permanent magnetic materials (such as high-performance NdFeB...magnets) containing any of these...elements” among the controlled items, meaning any finished NdFeB magnet containing so much as trace dysprosium or terbium content became subject to Chinese export licensing (MOFCOM/GAC Announcement No. 18 of 2025 summary).

The Western response: separating heavy rare earths outside China for the first time at scale

This doping chemistry is precisely why Lynas' 2025 achievement of commercial dysprosium and terbium separation at Kuantan, Malaysia, and MP Materials' DoD-funded expansion of heavy-rare-earth separation capacity at Mountain Pass, matter disproportionately relative to their tonnage. A single percentage point of dysprosium content in a magnet can determine whether that magnet meets an automotive OEM's thermal-derating specification, so even modest non-Chinese heavy-rare-earth supply materially changes negotiating leverage for Western magnet buyers. One industry funding announcement projects a combined non-Chinese platform reaching approximately 30 tonnes of dysprosium oxide and 10 tonnes of terbium oxide per year by early 2027, with a Phase 2 target of roughly 200 tonnes per year of dysprosium metal and 45 tonnes per year of terbium metal later in the decade — still a small fraction of Chinese output, but described as potentially the largest heavy-rare-earth source outside China once achieved (PR Newswire, 5 May 2026).

China's 2025 Export-Control Escalation: April Licensing, October Expansion, November Suspension

The April 2025 licensing regime: end-user certificates and no appeal route

Announcement No. 18 was issued jointly by MOFCOM and the General Administration of Customs on 4 April 2025, formally extending export-control coverage to samarium, gadolinium, terbium, dysprosium, lutetium, scandium, and yttrium and all related alloys, oxides, compounds, targets, and magnets (Nautical Log, summarizing Announcement No. 18). Exporters must apply for a licence on a per-shipment basis, submitting an end-user certificate signed by the foreign importer, an attestation of non-military end-use, a statement of final application naming the downstream product at HS-8 level and the physical consumption site, and a consent clause permitting MOFCOM or GACC to conduct a post-shipment audit within 12 months. MOFCOM retains discretion to deny applications on national-security, foreign-policy, or non-proliferation grounds with no formal appeal route (Sourzi, one-year retrospective, Mar 2026). The immediate downstream impact was severe: Ford Motor Company reportedly experienced “hand-to-mouth” magnet shortages and briefly idled its Chicago assembly plant for a week in May 2025 due to a lack of rare-earth magnets (Rare Earth Exchanges, 10 Aug 2025).

October 2025: MOFCOM adds five more elements and extraterritorial reach

On 9 October 2025, MOFCOM issued a coordinated package of six new notices (Announcements 55–58, 61, and 62 of 2025) that dramatically escalated the scope of rare-earth controls. Notice 57 added europium, holmium, erbium, thulium, and ytterbium to the restricted list, bringing the controlled-element total to twelve, effective 8 November 2025 (China-Britain Business Council, policy update, Oct 2025). Notice 56 added 60 additional rare-earth-related processing and equipment items to China's Dual-Use Items Control List. Most consequentially, Notice 61 imposed an extraterritorial “0.1% rule”: foreign entities outside China would be required to obtain a MOFCOM export licence to ship any goods containing at least 0.1% Chinese-origin rare earths by value, or goods produced using Chinese rare-earth technology — even for shipments between two non-Chinese countries with no Chinese company involved (White & Case, 13 Oct 2025). Notice 62 extended controls to rare-earth technology itself — mining, smelting, separation, metal production, magnetic-material manufacturing, and even recycling process technology — a direct attempt to prevent the kind of technology transfer that let Lynas and others build non-Chinese separation capability.

November 2025: a one-year suspension, but the April controls stay in force

Following the Trump–Xi summit, MOFCOM issued Announcement No. 70 on 7 November 2025, suspending implementation of the six October directives (Notices 55–58, 61, and 62) for rare earths, magnet materials, lithium-battery inputs, and super-hard materials. Two days later, Announcement No. 72 (9 November 2025) separately suspended Article 2 of the December 2024 Announcement No. 46 — the gallium/germanium/antimony dual-use licensing requirement — through 27 November 2026 (Clark Hill, 24 Nov 2025). Crucially, the April 2025 Announcement No. 18 controls on the original seven elements were not suspended and remained fully operational; China also began issuing “general export licences” to selected pre-vetted exporters under the April regime, easing but not eliminating the licensing burden (USGS MCS 2026). A detailed one-year retrospective published in March 2026 confirms that “Announcement No. 18, issued on 4 April 2025, remains fully operational twelve months on” and that the Busan-summit de-escalation “did not touch Announcement No. 18” (Sourzi, Mar 2026).

U.S., EU, and other jurisdictional responses

In the United States, Senators Mark Kelly and Tom Cotton introduced legislation in February 2025 aimed at reducing dependency on China for rare-earth elements, part of a wider wave of congressional critical-minerals bills tracked through 2025–26 (Senator Mark Kelly, press release; Northeast-Midwest Institute, Critical Minerals Legislative Tracker 2025–26). Separately, the Congressional Research Service notes that in January 2026 the Trump Administration announced a critical-minerals trade zone, an equity investment in USA Rare Earth, and convened a critical-minerals ministerial, followed on 2 February 2026 by White House announcement of Project Vault, a $12 billion initiative to establish a U.S. Strategic Critical Minerals Reserve (Congressional Research Service, IF13171, 4 Mar 2026). On the defense-procurement disclosure side, an earlier House bill (H.R. 8272) would require any DoD contractor supplying a system containing a rare-earth permanent magnet to disclose the country of origin at each supply-chain stage — mining, oxide refining, metal/alloy-making, and sintering/magnetizing — reflecting how granular U.S. policymakers want visibility into the neodymium supply chain to become (GovInfo, H.R. 8272). In Europe, the Critical Raw Materials Act (Regulation (EU) 2024/1252), which entered into force in 2024, designates rare earths as both Critical and Strategic Raw Materials and sets 2030 benchmarks of 10% domestic extraction, 40% domestic processing, and 25% recycled content for strategic raw materials as a category (European Parliamentary Research Service, CRMA briefing, 2024). In August 2025, the European Commission selected 13 rare-earth-and-critical-mineral Strategic Projects in third countries and, together with earlier domestic project selections, backed a total of roughly 60 Strategic Projects under the CRMA framework (Ecomondo, 19 Aug 2025; European Commission, IP_25_1419).

Prices: From the 2011 Bubble Through the 2022–2024 Collapse to the 2025–2026 Rebound

USGS annual average price history, 2021–2025

Source: USGS MCS 2026. The pattern is unambiguous: prices peaked in 2022 on post-pandemic EV and wind-turbine demand optimism, collapsed by more than half through 2023–2024 as Chinese mining quotas expanded faster than demand, then partially recovered in 2025 as China's April and October export controls reintroduced scarcity risk even without a change in the underlying mining quota.

The 2011 rare-earth bubble: the historical benchmark for volatility

Neodymium and praseodymium oxide prices spiked roughly 120% in a matter of weeks in early 2011, with praseodymium oxide and neodymium oxide reaching approximately 459,000 yuan and 575,000 yuan per tonne respectively (roughly $70,000 and $88,000/tonne, or $70–88/kg at the time), driven by China's 2010 export-quota cuts and a subsequent speculative bubble across the entire rare-earth complex (China.org.cn, 30 Mar 2011). A composite index of Chinese rare-earth export prices reportedly reached roughly $109,000/tonne at the peak of the 2011 episode before collapsing over the following three years, falling by roughly a third each year through 2013–2015 to a floor near $180/kg for some rare earths by 2016 (Business Insider, 7 Apr 2011; MINING.COM, 25 Apr 2025). This boom-bust cycle is the historical reference point cited by both Chinese and Western policymakers when discussing the demand-destruction risk of excessive rare-earth price spikes: high prices in 2011 accelerated substitution efforts (rare-earth-free motor designs, ferrite substitution) that persistently dampened demand growth for years afterward.

2022–2024 correction: oversupply, not demand collapse

Unlike 2011, the 2022–2024 price decline was driven primarily by supply growth rather than demand weakness: China's mining and separation quotas expanded materially (from 210,000/202,000 tonnes in 2022 to 255,000/244,000 tonnes in 2023 for mining/smelting respectively, with an unprecedented third-batch quota added mid-year), while MP Materials and Lynas both ramped production, together outpacing EV and wind demand growth and pushing NdPr oxide from $124/kg (2022 average) to $55/kg (2024 average) (USGS 2023 Minerals Yearbook, China chapter; USGS MCS 2026). MP Materials executives have publicly noted that pre-2025 NdPr oxide prices in the $50–60/kg range sat below the roughly $40/kg production cost floor needed to sustain new non-Chinese investment, explaining why the DoD's subsequent $110/kg price-floor guarantee was structured at nearly double the prevailing Chinese benchmark (Columbia University Center on Global Energy Policy, 11 Jul 2025).

2025–2026 rebound: export-control scarcity meets a still-loose Chinese quota

China's domestic NdPr oxide benchmark, tracked by Shanghai Metals Market (SMM), climbed from approximately $49/kg in December 2024 to a multi-year high of $111.5/kg on 25 February 2026, before easing slightly to $103.76/kg by 10 March 2026 — roughly double the December 2024 level (Rare Earth Mining News, 10 Mar 2026). A separate SMM-based tracker shows NdPr oxide at $90,317.58/tonne ($90.32/kg) as of 1 June 2026, with neodymium metal at $124.87/kg and praseodymium metal at $126.16/kg as of 1 April 2026 (Critical Minerals News, citing SMM, 1 Jun 2026). By 1 July 2026, domestic Chinese praseodymium metal had risen further to $149.19/kg (up 19.5% from June) and the NdPr alloy benchmark had climbed to $133.02/kg (up 21.4% month-on-month), suggesting the rebound accelerated through mid-2026 rather than fading (Rare Earth Mining News, citing SMM, 1 Jul 2026). For historical comparison, dysprosium — the heavy rare earth most directly targeted by the April 2025 controls — had already risen from roughly $238/kg in 2018 to about $528/kg by the time of a 2023 Oxford Institute for Energy Studies analysis, illustrating how much more volatile the heavy-rare-earth doping elements are relative to neodymium and praseodymium themselves (Oxford Institute for Energy Studies, China's rare earths dominance, 2023).

Benchmark providers and the absence of an exchange-traded contract

Unlike copper or aluminium, there is no LME- or COMEX-listed neodymium or NdPr futures contract. Price discovery instead runs through assessment agencies: Fastmarkets publishes daily fob-China assessments for NdPr oxide (99%, 75:25 ratio) and NdPr metal (Fastmarkets NdPr oxide assessment; Fastmarkets NdPr metal assessment), Argus Media runs a parallel metals-platform price-reporting service, and Shanghai Metals Market (SMM) is the dominant domestic-China reference used by most Chinese producers, traders, and increasingly by Western analysts tracking the China/ex-China price gap. Adamas Intelligence and the former Industrial Minerals Company of Australia (IMCOA) framework are the leading independent research houses for volume and demand-forecasting analysis rather than daily price assessment, and were cited directly by market participants reacting to the MP Materials-DoD deal (Reuters via Yahoo Finance, quoting Adamas Intelligence managing director Ryan Castilloux, 10 Jul 2025).

Recycling: Magnet-to-Magnet Loops Are Commercializing, But Still a Rounding Error on Global Supply

HyProMag's HPMS process: hydrogen decrepitation at Tyseley Energy Park

HyProMag, majority-owned by Mkango Resources through its Maginito subsidiary, commercializes the Hydrogen Processing of Magnet Scrap (HPMS) technology developed over roughly two decades and $100 million of R&D at the University of Birmingham's Magnetic Materials Group. HPMS exposes whole NdFeB-bearing components — hard-disk drives, e-bike motors, wind-turbine generators, MRI magnet assemblies — to hydrogen gas, causing the sintered magnet to absorb hydrogen and disintegrate into a demagnetized alloy powder that separates cleanly from coatings, adhesives, and housing metal without any need for manual disassembly or shredding (HyProMag, company overview). The resulting powder can be purified and directly re-sintered into new magnets with more than 95% recycled content — a “short loop” process that HyProMag states requires approximately 88% less energy than primary mine-to-magnet production (HyProMag, technical progress bulletin, Jun 2025). A third-party life-cycle assessment by Minviro found HyProMag USA's process carries a product carbon footprint of just 2.35 kg CO₂-equivalent per unit (HyProMag technical bulletin, citing Minviro LCA). The Tyseley Energy Park facility in Birmingham, UK — the first commercial-scale sintered NdFeB magnet manufacturing site in the UK since Philips closed its Southport plant in December 2003 — achieved first commercial recycled-alloy production in June 2025 and was officially opened in early 2026, with capacity of over 400 kg of recycled alloy per HPMS batch and a target of 100–350 tonnes per annum scaling toward a longer-term ambition of roughly 1,000 tonnes per annum (HyProMag, About page; Critical Minerals Association, HyProMag UK update, Apr 2026). As of the April 2026 update, HyProMag UK reported 9.2 tonnes of recycled alloy powder produced cumulatively and a collaboration with Siemens AG on a recycled-magnet SIMOTICS servomotor rotor showcased at Hannover Messe 2026, alongside parallel HyProMag GmbH (Germany) and HyProMag USA rollouts targeting first U.S. production in 2027 (Critical Minerals Association, Apr 2026).

Noveon Magnetics: magnet-to-magnet (M2M) recycling in San Marcos, Texas

Noveon Magnetics, based in San Marcos, Texas, developed its own patented Magnet-to-Magnet (M2M®) process, created by chemist Miha Zakotnik, which the company states is currently the only method able to manufacture full-performance NdFeB magnets from recycled sources without compromising magnetic performance (Noveon Magnetics, History and Future). Noveon's process disassembles end-of-life products — motors, MRI machines, hard drives, hybrid-vehicle motors — via a “focused extraction methodology” that explicitly avoids shredding technology, then sorts recovered magnets, steel, aluminium, copper, and plastics before recycling 100% of recovered magnet material back into new production (Noveon Magnetics, Our Process). The company can blend virgin and fully recycled feedstock flexibly, stating it can “create a superior quality magnet using 100% recycled materials” (Noveon Magnetics, Our Products). In February 2025, Noveon signed an off-take agreement with Nidec, one of the world's largest motor manufacturers, explicitly framed as strengthening U.S. rare-earth magnet supply-chain resilience (PR Newswire, 3 Feb 2025).

MP Materials and Apple: recycled feedstock enters the largest U.S. magnet deal to date

On 15 July 2025, MP Materials and Apple announced a $500 million partnership under which MP will supply Apple with magnets made from 100% recycled rare-earth material, manufactured at MP's Fort Worth Independence facility using recycled feedstock processed through a new, dedicated recycling line at Mountain Pass, California (MP Materials, press release, 15 Jul 2025). Apple paid $200 million upfront against magnet shipments expected to begin in 2027, ramping to support hundreds of millions of Apple devices, with the deal separately triggering an expansion of Independence's magnet capacity from 1,000 to 3,000 tonnes per year in combination with the DoD transaction (Yahoo Finance, 25 Jul 2025; MP Materials, Automotive Symposium transcript, 4 Nov 2025). Industry commentary projected the deal could enable the U.S. to produce approximately 20% of global recycled rare-earth output by 2030, though this figure comes from third-party analysis rather than MP or Apple disclosure and should be treated as an estimate (Discovery Alert Australia, LinkedIn analysis, 15 Jul 2025).

Recycling's structural ceiling: feedstock scarcity, not technology, is the constraint

USGS notes only that “limited quantities of rare earths were recovered from batteries, permanent magnets, and fluorescent lamps” in 2025, without providing a specific tonnage or percentage estimate for neodymium specifically, reflecting how immature formal reporting of rare-earth recycling volumes remains even as the underlying technology matures (USGS MCS 2026). The binding constraint on scaling magnet recycling is not process technology — HPMS and M2M are both proven at increasing commercial scale — but feedstock collection logistics: NdFeB magnets are dispersed across billions of small consumer devices and thousands of larger industrial/EV/wind assets with no dedicated collection infrastructure comparable to lead-acid battery or aluminium can recycling. The EU CRMA's push for minimum recycled-content thresholds in permanent magnets is explicitly designed to force collection-infrastructure investment by creating guaranteed demand for recycled material, rather than assuming collection will scale organically (European Commission, CRMA Q&A).

Forward Look 2026–2030: A Race Between Western Capacity Build-Out and China's Suspended Controls

Capacity pipeline: 10,000 tonnes of U.S. magnet capacity by 2028, but China still sets the marginal price

The most concrete forward-looking commitment on the Western side is MP Materials' path to 10,000 metric tonnes of annual U.S. magnet manufacturing capacity once its 10X Facility begins commissioning in 2028, up from 1,000 tonnes previously — a tenfold increase underwritten by a 10-year DoD offtake agreement and $110/kg NdPr price floor (Columbia University Center on Global Energy Policy). Even at full build-out, however, 10,000 tonnes of magnet capacity remains a fraction of global NdFeB demand, which industry estimates place in the hundreds of thousands of tonnes annually and growing at double-digit CAGR through 2030–2034 (AIM Magnetic, NdFeB market outlook to 2033). Lynas' heavy-rare-earth separation ramp in Malaysia and a separate industry platform targeting roughly 20,000 tonnes per year of heavy-rare-earth permanent magnet capacity later this decade add incremental non-Chinese supply, but China's flat mining/separation quota policy means Beijing retains the ability to flood or tighten the market at will, continuing to set the marginal price for the foreseeable future (PR Newswire, 5 May 2026).

Substitution R&D: heavy-rare-earth-free and rare-earth-free motor designs

A distinct market segment — heavy-rare-earth-free NdFeB for EV motors — was valued at $2.8 billion in 2025 and is projected to reach $12.4 billion by 2034 at an 18.5% CAGR, reflecting automaker efforts to specify magnets that avoid dysprosium and terbium entirely through grain boundary diffusion optimization and alternative alloy chemistry, even while remaining neodymium-based (Market Intelo, Heavy Rare Earth Free Magnet for EV Motors Market). Separately, automakers including Valeo and MAHLE are expanding magnet-free (induction or wound-rotor synchronous) electric motor product lines into premium vehicle segments specifically to reduce rare-earth exposure altogether, though these designs generally carry an efficiency or power-density penalty versus PMSM architecture that limits their applicability to cost- or supply-risk-sensitive segments rather than the full EV market (Strategic Market Research, EV Motor Market report).

Key risks 2026–2030: policy reversal, financing execution, and demand-destruction feedback

Three risks dominate the outlook. First, geopolitical reversal risk: the October 2025 controls are only suspended, not repealed, and MOFCOM retains the standing legal authority to reinstate the twelve-element licensing regime, the 0.1% extraterritorial rule, and technology controls after November 2026 if U.S.-China relations deteriorate again (Dhit sp. z o.o., anatomy of the 2025 crisis, Nov 2025). Second, financing and execution risk on the Western build-out: MP Materials' 10X Facility, Fortune Minerals-style byproduct projects, and EU CRMA Strategic Projects all depend on multi-year construction timelines, continued government funding certainty (MP's own 8-K flagged that “ongoing funding from the U.S. Congress cannot be assured indefinitely”), and successful technology scale-up for heavy-rare-earth separation outside China (Reuters via Yahoo Finance, 10 Jul 2025). Third, demand-destruction feedback: the 2011 bubble demonstrated that sustained high rare-earth prices accelerate substitution and efficiency efforts that can suppress demand growth for years afterward — a dynamic Western policymakers must now balance against price floors like MP's $110/kg NdPr guarantee, which intentionally holds prices above the free-market clearing level to sustain investment, but at the risk of incentivizing the same substitution response that followed 2011 (MINING.COM, rare earth export restrictions and demand-destruction risk, 25 Apr 2025).

Demand scenarios: energy transition and defense stockpiling both point the same direction

The European Commission's own justification for CRMA permanent-magnet provisions cites a projected sixfold increase in rare-earth demand for high-performance magnets by 2030 and sevenfold by 2050, driven jointly by EV adoption and wind-turbine deployment (Ecomondo, citing European Commission analysis, 19 Aug 2025). Layered on top of this civilian demand growth, the FY2025 U.S. National Defense Stockpile listed potential acquisitions of 300 tons of neodymium-praseodymium oxide, 450 tons of NdFeB magnet block, and 60 tons of samarium-cobalt alloy — explicit government recognition that defense stockpiling now competes directly with civilian EV and wind demand for the same constrained NdPr and heavy-rare-earth supply (USGS MCS 2026, Government Stockpile table). Whether Western capacity additions from MP Materials, Lynas, HyProMag, Noveon, and the EU's Strategic Projects can keep pace with this compounding civilian-plus-defense demand growth — without triggering another China-driven price shock or supply cutoff — is the central question for the neodymium market through 2030.