Global LFP Cathode Active Material Market Size, Share & Forecast 2026–2033

Forecast Snapshot 2026–2033

Market overview

The global LFP cathode active material market is valued at USD 8.30 billion in 2025, rising to USD 26.20 billion by 2033 — a 15.45% baseline CAGR. That figure is the conversion basis: the fees the cathode industry actually bills for turning iron phosphate and lithium carbonate into a specification-grade powder. On an invoiced basis — the full price on the purchase order, including the lithium that passes straight through — the 2025 number is USD 15.99 billion. Both are true. Only one is a market this industry controls, and the difference between them is the finding.

Volume is 3.77 million tonnes in 2025, of which China produced 3.75 million — 99.47% — up roughly 60% in a single year. The forecast reaches 9.90 million tonnes by 2033 at a 12.83% CAGR, against 2.33% for the fee per tonne. Value therefore compounds faster than tonnage by design, not by assumption: the fee rises a fifth over eight years while tonnage rises 2.6 times.

The growth shape is unusual for this series. Every other report in this run has a peak annual growth year somewhere in 2028–2030; here the peak is 2026, the first year of the window, when Chinese and global output step from 3.77 to 5.60–5.80 million tonnes in a single move. Everything after 2026 is deceleration: a market that is being built out faster than the demand curve that justifies it, and that then has to live with the capacity.

Why the perimeter is the finding

A tonne of LFP cathode contains 0.2342 tonnes of lithium carbonate on stoichiometry — 73.89 over twice 157.76, the molar masses of Li₂CO₃ and LiFePO₄ — and 0.956 tonnes of iron phosphate. Both are quoted, traded commodities with their own markets and their own price charts. When the cathode maker invoices a cell maker, it passes the lithium through at a linked price and adds a conversion fee. Nothing in that invoice is a cathode-maker price except the fee.

The industry says so itself. At the April 2026 price of RMB 55,800 per tonne of lithium carbonate, a tonne of LFP cathode carries RMB 13,068 of lithium — 0.2342 tonnes of it — and a converted processing fee of RMB 18,000 that includes the iron phosphate. The invoice is therefore RMB 31,068, and the fee is 57.94% of it. Of that fee, iron phosphate at 0.956 tonnes × RMB 11,650 is RMB 11,137 — 61.87% of the fee and 35.85% of the invoice. What is left for energy, labour, depreciation, yield loss and profit is RMB 6,863 per tonne, 22.09% of what the customer pays. That is the LFP cathode industry: roughly a fifth of its own invoice, and more than half of what the cell maker pays is the fee the cathode maker negotiates. The 51.91% in the snapshot table is the same ratio struck on 2025's weighted fee of RMB 15,806 rather than on the April 2026 quote — the fee share rises with the fee, which is the whole argument.

So the conversion basis is not a modesty. Sizing this market at the invoiced value would make it a lithium derivative with a fixed 0.2342 multiplier, and would produce a market that doubled in 2026 for reasons originating in a mine in Jianxiawo rather than in a kiln in Guizhou.

The three pools

By application. LFP cell demand in 2025 was 1,477.86 GWh: 937.86 GWh of EV and power cells (1,421 GWh of global power battery output at 66% LFP) and 540 GWh of stationary storage, which grew about 73% in a year and is now 36.54% of the chemistry's demand. LFP consumes 2,551 tonnes of cathode per GWh, which is how 1.48 TWh becomes 3.77 million tonnes. The storage pool is the one that changed the industry's year: it is why roughly twenty Chinese cathode plants ran at or above nameplate in the second half of 2025, and why the monthly output peak reached more than 410,000 tonnes.

By generation. The 2025 mix was 44.8% third-generation (2.50 g/cm³ compaction), 33.5% second-generation (2.40), 11.5% third-and-a-half (2.55) and 10.3% fourth-generation (2.60). Fourth-generation material is made by perhaps four to six companies in the world. The fee the market pays for it, RMB 18,800, is 21.29% above the RMB 15,500 it pays for second-generation powder — for a route that requires a second calcination. A technical frontier that is rewarded at a fifth of a premium is a frontier being asked to fund itself out of somebody else's cost base.

By route. 83% of 2025 output came from the iron-phosphate solid-state route, 6% from ferrous oxalate, 5% from iron oxide red and 6% from liquid phase. The route matters commercially because only the first consumes merchant iron phosphate at scale — and iron phosphate is the input that the Chinese export-control catalogue now lists by specification.

What the leader tells you, and it is uncomfortable

Hunan Yuneng sold 1,137,100 tonnes in 2025, up 60.03%, at a realised RMB 30,450.06 per tonne. In 2024 it sold 710,554 tonnes at RMB 31,804.10. Volume rose 60%; price per tonne fell 4.26%. Revenue rose 53.22% to RMB 34.62 billion and net profit rose 115.18% to RMB 1.277 billion — a 3.69% net margin, or RMB 1,123 per tonne. Gross margin was 9.15%, meaning RMB 2,786 of gross profit on a tonne that sold for RMB 30,450.

This is the most integrated producer in the world: it self-supplies iron phosphate, it has a phosphate mine ramping in Guizhou, and it has 70,000 tonnes a year of battery-grade lithium carbonate conversion. It ran at 113.76% capacity utilisation while the Chinese industry ran at 54.5%. And it kept 3.69 cents of every yuan. That is the honest counterweight to a 15.45% value CAGR: the growth is in the tonnage, and the tonnage is where the competition is.

Method and limits

The volume base is reported output, not shipments, and comes from Chinese industry survey data; it carries an estimated ±10% band. The fee per tonne is a weighted construction from the published generation-by-generation processing-fee schedule applied to the reported generation mix, so it inherits both. The fee path assumes 2.33% a year — below the 44.0% the fee rose between early 2025 and April 2026 — on the view that capacity additions of 2.5 million tonnes a year hold the fee down while lithium, iron phosphate and energy float underneath it. All values are converted at a constant RMB 7.18 per USD so that the growth rates measure the market and not the exchange rate.

Excluded: lithium carbonate and lithium refining (a separate market, and 0.2342 tonnes of every tonne here); iron phosphate as a standalone merchant market where it is sold to third parties rather than consumed in the fee; cell manufacturing, module and pack assembly; and sodium-ion and lithium manganese iron phosphate chemistries, which appear as substitution risk rather than as volume.

Related research

This report is the material layer beneath Global EV Battery Market (USD 91.7 billion in 2025 to USD 369.5 billion in 2033, 19.03%) and Global Battery Electrolyte Production Market (USD 11.49 billion to USD 41.08 billion, 17.27%), both of which are sized on cell-level value and therefore sit above this one by construction rather than by contradiction. It is the demand side of Global Phosphorous Acid & Phosphonate Derivatives Market, which covers the thermal branch of the phosphorus chain; LFP consumes the purified wet-process branch instead, running at 64.75% utilisation while the thermal furnaces ran at 35.46%.

Competitive Landscape

There are two markets inside this one product, and the concentration data only makes sense once you separate them. The top five held 54.8% of Chinese output in 2025 and the top ten 74.0% — both down on 2024, in a year when the leader grew 60%. Concentration fell while the leader grew because tier-three and tier-four plants won orders for conventional second- and third-generation powder on price, while the leaders sold out of high-compaction material. One market competes on qualification and clears; the other competes on cost and sets the floor.

Hunan Yuneng sold 1,137,100 tonnes, up 60.03%, for 30.16% of world tonnage and its sixth consecutive year as the largest producer. Hubei Wanrun and Dynanonic held second and third. Fulin Shengda entered the top five on the oxalate route, and Youshan Technology held its place. Behind them sits a long tail of named producers each above 10,000 tonnes — Guoxuan High-Tech, Changzhou Liyuan, Andi Technology, Taifeng Xianxing, GCL Lithium, Rongtong High-Tech — which is the structural reason a fee increase cannot be held: there is always a plant with spare kiln time and a conventional product.

The most informative pair of numbers in the entire report is a utilisation gap. The Chinese industry ran at 54.5% of its 6.88 million tonnes of nameplate capacity — or 57.5% against the 6.52 million tonnes a second compiler reports — while Yuneng ran at 113.76%. Capacity is not the constraint. Qualified capacity is. The difference between those two figures is the moat, and it is not made of capital: it is made of approved-vendor status on a powder that determines a cell’s cycle life and its safety file.

That is the economics of qualification, and it is unusually strong here. A cell maker weighing a second source is not comparing prices on a commodity; it is comparing a supplier against a qualification programme that consumes months and a production line that runs at gigawatt-hour scale. Saving RMB 1,000 a tonne on a material that is 51.91% of the invoice sounds material until it is set against the cost of a failed qualification lot. So price competition is confined to the tiers where switching is cheapest — conventional product, short qualification, buyer already multi-sourced — and the frontier clears at whatever the ladder says.

Which produces the paradox at the top. Thirty per cent of world tonnage and a 3.69% net margin, on a 9.15% gross margin, in the best demand year the industry has ever had. Yuneng is the most integrated, lowest-cost producer in the world and it kept 3.69 cents of every yuan. Share leadership in this market is not a licence to price; it is a licence to run at 113.76% utilisation while somebody else sets the fee.

Outside China the competitive position is close to nil on volume and real on intent. Aleees in Taiwan has made LFP since 2005, holds 160 patents, and synthesises its own iron phosphate specifically so that it does not depend on a Chinese precursor — a strategy that now reads as foresight. ICL has committed USD 285 million of initial investment to a joint venture with Dynanonic at Sallent in Spain, on a former potash site. Ford has spent USD 3 billion on the Marshall, Michigan plant, 60% complete in mid-2025, running CATL’s process under licence. None of these is a supply threat to a Chinese producer before 2028. All of them are the reason Chinese producers are opening plants in Spain and Malaysia rather than shipping powder into those markets.

What follows for the next three years is a structure that consolidates at the top and attrites at the bottom. The top consolidates on qualification, on precursor ownership and on fourth-generation yield; the bottom competes on a fee that a hundred plants can quote. And the fee itself is set by the marginal conventional tonne, not by the frontier — which is why the most profitable position in this industry belongs to whoever can sell the 2.60 rung while the price of everything else is decided by someone running at 40% utilisation.

Value Chain

A tonne of LFP cathode is a bill of materials with a service attached, and the bill of materials is the larger half. On stoichiometry it holds 0.2342 tonnes of lithium carbonate — 73.89 over twice 157.76, the molar masses — and 0.956 tonnes of iron phosphate, 150.82 over 157.76. Neither material is made by the cathode industry. Both are quoted, traded commodities with their own producers, their own price charts and, in iron phosphate’s case, their own export-control specification. That single fact governs the whole structure of this market: the industry does not sell a material, it sells a conversion.

So there are only three lines on a cathode invoice and the industry owns one of them. Lithium is passed through at a linked price, because no cathode maker is going to take a directional position in a commodity that moves on a Chinese mine auction. Iron phosphate sits inside the processing fee, which is why the fee moves when the precursor moves. What remains is conversion: energy, labour, depreciation, yield loss and profit. The fee pool is 51.91% of the invoiced value the customer pays — USD 8.30 billion of USD 15.99 billion in 2025 — and the other 48.09% is lithium that arrives and leaves at somebody else’s price.

The fee is not one price. It is a ladder, published generation by generation: RMB 15,500 for second-generation powder at 2.40 g/cm³ compaction, RMB 16,500 for third at 2.50, RMB 17,800 for third-and-a-half at 2.55 and RMB 18,800 for fourth at 2.60. The spread from bottom to top is 21.29%, and it is earned by a second calcination — a second firing cycle, a second carbon-coating pass and roughly double the kiln time per tonne. In no other part of this chain is a specification published alongside its own price. That is both the industry’s discipline and its problem: the ladder tells every buyer exactly what the frontier is worth, and it is worth a fifth.

Iron phosphate is where the fee actually goes. At RMB 11,650 a tonne and 0.956 tonnes of it per tonne of cathode, the precursor inside one tonne of LFP is worth RMB 11,137 — 61.87% of the RMB 18,000 fee. What is left for the cathode maker to run a plant on is RMB 6,863, or 38.13% of the fee. That ratio is not stable, and 2025 showed why: iron phosphate went from RMB 10,000–10,300 a tonne in September to RMB 11,500–11,800 by late December in three rounds, and the cathode producers passed the increase through as a fee rise within weeks rather than absorbing it. The linkage is contractual and index-referenced, not negotiated tonne by tonne. Which means the fee is a float on two commodities the cathode industry does not control, with a conversion margin of about 38% of it.

The only durable lever is ownership of the precursor. Hunan Yuneng, the largest producer, now self-supplies iron phosphate entirely, took first ore from the Huangjiapo phosphate mine in Guizhou in the fourth quarter of 2025, and has built 70,000 tonnes a year of battery-grade lithium carbonate conversion from spodumene and from recycled black mass. A producer buying iron phosphate at RMB 11,650 against an RMB 18,000 fee has RMB 6,350 for conversion and margin. A producer making its own captures whatever the gap is between the market price and its own cost — and that gap, not the fee ladder, is what every tier-two Chinese cathode plant is now chasing.

The export price is the cleanest evidence of where value is captured. China shipped 28,300 tonnes of LFP cathode in January to November 2025 for USD 151 million, or USD 5,328.49 a tonne — RMB 37,156.6 at the period rate, 22.02% above the RMB 30,450.06 the domestic leader realised on 1.14 million tonnes. The premium is not scarcity. It is mix: export volumes are weighted to high-generation material and the buyers are qualification-constrained. And 28,300 tonnes is 0.8% of national output — an industry that industrialised for a home market and exports a rounding error.

The perimeter, stated in the body rather than in a footnote. This market is merchant LFP cathode powder: the fee pool plus the precursor consumed inside it. Excluded are lithium carbonate and lithium refining, which is a separate quoted market and 0.2342 tonnes of every tonne here; iron phosphate sold to third parties rather than consumed in a cathode fee; cell, module and pack manufacturing; and sodium-ion and lithium manganese iron phosphate chemistries, both of which are named in the risk chapter as substitution rather than counted as demand. The invoiced total of USD 15.99 billion is disclosed alongside the USD 8.30 billion headline so that a reader can reconcile to any figure built on the full purchase-order price.

Investment Activity

The capital going into this market is being spent to buy a fee, and the fee’s margin is thin. That is not a reason to avoid the investment case; it is the whole shape of it.

Yuneng’s own numbers are the benchmark every other project in the world is implicitly measured against. Revenue of RMB 34.62 billion on 1.14 million tonnes at a 9.15% gross margin and a 3.69% net margin, in the best demand year the industry has ever had. Total assets of RMB 39.09 billion against net assets of RMB 12.88 billion. It raised equity in the first quarter of 2026 and finished the period with RMB 6.73 billion of cash, and it is now spending that outside China: environmental approval granted and site works under way in Spain, an entity established in Malaysia with preparatory work begun. Its own 2026 guidance is earnings of RMB 3.86 billion — a five-fold jump on 2025 — which reads less as a price story than as operating leverage applied to a thin base.

That guidance is the investment case in one line, and it cuts both ways. On a 9.15% gross margin, a 44% fee rise and 113.76% utilisation take RMB 1.28 billion of net profit to RMB 3.86 billion on roughly the same asset base. The same arithmetic runs backwards with equal force: a fee that gives back a third of its rise takes most of that earnings growth with it, and nothing on the balance sheet changes. There is no capacity decision that protects a business structured like this; there is only a price decision somebody else makes.

ICL’s USD 285 million at Sallent is the clearest read on what a Western entrant thinks a tonne of European LFP capacity is worth. It is not a phosphate play, although ICL has the phosphate and the former potash site. It is a qualification play: buy Dynanonic’s process and a European address in one transaction, and inherit a specification ladder rather than rebuilding one. That is the correct thing to buy, and it is priced accordingly.

The capacity arithmetic is where the investment case gets uncomfortable. Chinese nameplate reached 6.88 million tonnes in 2025 with 46% added in a single year. 2026 is forecast at 9.50 million tonnes nominal and 9.00 million effective, with 2.50 million tonnes a year of new build landing against demand growth of 1.90 million tonnes. Newbuild is running ahead of demand at exactly the moment the fee is recovering, which is the classic shape of a margin that gets given back rather than kept.

Two things are worth underwriting anyway. First, precursor capacity: iron phosphate and lithium carbonate conversion. Whoever owns the precursor owns the only part of the fee that is defensible, and Yuneng’s 70,000 tonnes a year of lithium carbonate conversion is the template every tier-two plant is now copying. Second, fourth-generation and LMFP qualification capacity, because the premium of RMB 3,300 a tonne over second-generation exists, is published, and is defended by a yield curve that only four to six producers can climb. What is not worth underwriting is conventional second- and third-generation capacity: it is 78.3% of the market’s tonnage, it is sold at a fee any of a hundred plants can quote, and it is the product the tier-three entrants are buying share with.

One accounting caution that changes how announced capacity should be read. A double calcination roughly doubles kiln time per tonne, so capital per tonne of qualified fourth-generation output is materially above capital per tonne of nameplate. The industry reports nameplate. The return is earned on qualified output. The gap between the two numbers is the entire difference between a project that pays and a project that fills a warehouse with third-generation powder sold at a third-generation fee.

And the geography is itself a purchase. Spain and Malaysia are not cheaper than Guizhou on any operating line that matters; they are capital spent to buy market access in a decade when market access is being rationed by rule rather than by price. Investors in this chain are now underwriting two things at once — a fee and a jurisdiction — and the second one is not yet priced in anybody’s model.

Technology & Innovation

LFP is a mature chemistry being improved at the powder level, and the industry scores itself on one number: compaction density, the grams per cubic centimetre of electrode that can be pressed from a given powder. Higher compaction means more active material in the same can, which means more range from the same cell without changing the cell design. Everything else — carbon coating uniformity, particle-size distribution, phase purity — is in service of a number that a press can measure in minutes and a customer can specify in a purchase contract.

The 2025 mix was 44.8% third-generation at 2.50 g/cm³, 33.5% second-generation at 2.40, 11.5% third-and-a-half at 2.55 and 10.3% fourth-generation at 2.60. The step from 2.40 to 2.50 is process discipline: tighter control of firing temperature, atmosphere and carbon content on equipment the plant already has. The step to 2.60 is a different plant. It requires a double calcination — a second firing cycle and a second carbon-coating step — with tighter particle-size control and roughly double the kiln time for every tonne produced, in exchange for a product the market pays 21.29% more for.

Only four to six companies in the world can hold stable production at 2.60, and the barrier is yield, not chemistry. The specification is published. The equipment is available from the same Chinese furnace builders who supply everyone else. What fails is consistency: the failure mode is a batch that compacts to 2.57, which cannot be sold as fourth-generation and does not become scrap — it becomes a third-generation tonne at a third-generation fee. The frontier’s risk is therefore margin, not loss: the kiln time is spent either way and the price difference is the whole return.

The route mix explains why the barrier sits where it does. 83% of 2025 output came from the iron-phosphate solid-state route, 6% from ferrous oxalate, 5% from iron oxide red and 6% from liquid phase. The oxalate route reaches high compaction more easily, which is why Fulin Shengda and Hunan Pengbo held sold-out positions in 2025. The liquid-phase route — Dynanonic’s speciality — moved from laboratory to stable high-compaction shipment during the year. Iron oxide red is the cost route, and GCL’s version of it began taking share on price. Four routes, one specification ladder, and the ladder is what the fee schedule prices: a producer’s route determines which rung it can compete on before it ever quotes a customer.

There is a process constant underneath all of this that gets very little attention and should get more: LFP consumes 2,551 tonnes of cathode per GWh. That number is what turns 1,477.86 GWh of cell demand into 3.77 million tonnes of powder. It is also the quiet deflation inside the volume line. Every improvement in compaction or in cell design reduces tonnes consumed per unit of energy delivered, so cathode demand grows more slowly than gigawatt-hours by exactly the rate at which the technology improves. The volume forecast in this report is a bet that GWh growth outruns that efficiency, and so far it has — comfortably.

The next step is manganese. Lithium manganese iron phosphate substitutes manganese for part of the iron and lifts average voltage from about 3.2 volts to 3.8, which raises energy density without a new supply chain for the anode or the electrolyte. Yuneng has closed the research phase on its second-generation high-compaction LMFP and moved a low-interface-impedance grade into trial production. It is the same kiln, the same precursor logic and the same customers, which is precisely why LMFP is a risk to LFP volumes and not a threat to the incumbents’ position: the firms that would lose the tonnes are the firms that would gain the upgrade.

What none of this is, is discovery. There is no new molecule in this industry’s pipeline, no step-change chemistry that would obsolete a plant. The innovation is metallurgical housekeeping conducted at industrial scale, rewarded by a published ladder, protected by a yield curve that most producers cannot climb. That is why the competitive question in this market is not who invents — it is who qualifies, and at which rung of a ladder whose top price is only 21.29% above its bottom.

Market Risk

The largest risk to this forecast is not competition. It is that the denominator changes chemistry.

Sodium-ion is the substitution the lithium price creates. The IEA noted in its 2026 outlook that lithium prices at the start of 2026 were more than twice their level a year earlier, and that if that persists it reinforces momentum behind sodium-ion, which offers lower range and no lithium. Sodium-ion uses no iron phosphate, no lithium carbonate and none of this fee structure. It is currently a small share of storage, which is precisely where energy density matters least — and storage is 36.54% of LFP demand. Every ten percentage points of storage that migrates to sodium removes roughly 1.4 million tonnes of LFP demand at 2026 volumes, which is 37% of the 2025 base. That is the single largest discrete threat in the model, and it is a threat with a two-year lag and a five-year probability.

The second risk is LMFP cannibalisation, and it is the more likely of the two. It is the same plants, the same precursor logic and the same customers, which is good for the incumbents’ position and bad for this report’s volume line: the tonne that moves to LMFP carries manganese, commands a different fee, and does not appear here at the same price. Falling on the right side of the competitive question and the wrong side of the perimeter question is a real outcome, and no amount of qualification advantage protects the number.

The third risk is the fee itself, and it is the most likely reason a conservative case beats a baseline case. Between early 2025 and April 2026 the processing fee rose 44.0%, from roughly RMB 12,500 to RMB 18,000 for a second-and-a-half generation product. It rose because iron phosphate went up 15% to 18% in three rounds and because a handful of leaders announced maintenance shutdowns at the moment cell makers were building seasonal stock. That is a supply event, not a structural repricing. Against 2.50 million tonnes a year of new capacity arriving into 1.90 million tonnes of demand growth, a fee that has just risen 44% is a fee with room to fall. A 10% cut costs the industry USD 0.83 billion of 2025 value on its own.

The fourth risk runs the other way and could break the model rather than dent it. If the material assistance cost ratio makes licensed Chinese process uneconomic inside the United States, the Western capacity being built to serve that market is stranded before it ever qualifies — and the Chinese response of building in Spain and Malaysia does not solve an American problem. That is a case where the volume materialises, the fee materialises, and the asset that was supposed to earn it is in the wrong jurisdiction.

Geopolitical exposure is extreme by any normal measure: 99.47% of tonnage from one country, with that country’s government controlling the export of the process technology above 2.58 g/cm³. That cuts two ways, which is rarely stated. It is a supply risk to every customer outside China and a competitive risk to every producer outside China, simultaneously and for the same reason.

There is also a technical deflation that belongs in the risk chapter rather than the technology chapter, because it acts on the same line. LFP consumes 2,551 tonnes of cathode per GWh, and that constant is not constant: higher compaction and the higher voltage of LMFP both reduce tonnes consumed per unit of energy delivered. The volume forecast therefore depends on gigawatt-hour growth outrunning chemistry efficiency. It has, decisively, for three years. It is still an assumption, and it is the one nobody in this industry argues about because arguing about it is bad for orders.

The growth shape compounds all of it. The peak annual growth year is 2026 — +51.19% in volume and +54.71% in value, behind the base year rather than ahead of it. Everything after that is deceleration into capacity that has already been built. Any demand disappointment in this market lands on a plant that is already standing, not on a project that can be deferred.

Overall market risk is medium. The volume is close to certain for three years and the price is not. Substitution risk is high on a five-year view and low on a two-year view. Anyone underwriting this chain should treat the tonnage as a planning input and the fee as a position.

Regulatory Landscape

Two export-control regimes now point at this market from opposite directions, and they are not symmetrical. China controls the technology. The United States controls market access by cost accounting. Europe regulates the artefact. None of the three raises supply, and only one of them is aimed at the grade that matters.

China’s control is written as a specification, not a company list. On 15 July 2025 the Ministry of Commerce and the Ministry of Science and Technology published Announcement No. 28 of 2025, amending the Catalogue of Technologies Prohibited or Restricted from Export and adding, under chemical raw materials and products, item 252604X: battery cathode material preparation technology. The thresholds are precise. LFP preparation technology is restricted where the material reaches powder compaction density of at least 2.58 g/cm³ at 220 MPa, 0.1C discharge capacity of at least 160 mAh/g and first-cycle coulombic efficiency of at least 97%. LMFP is restricted at 2.38 g/cm³, 155 mAh/g and 3.85 volts average discharge.

A third limb reaches upstream, and it is the part that gets missed. The catalogue also covers precursor processes — iron phosphate, iron manganese phosphate, battery-grade ferrous oxalate, lithium dihydrogen phosphate and battery-grade lithium phosphate — with iron phosphate restricted where tap density exceeds 1.2 g/cc and magnetic foreign matter is below 10 parts per billion. That second threshold is a purity specification, which means the control does not merely fence off the powder: it fences off the highest-quality merchant precursor as well, and the precursor is the thing that is 61.87% of the fee.

Read the numbers against the commercial ladder and the policy is doing something very specific. Fourth-generation material is specified at 2.60 g/cm³ against a control threshold of 2.58. The state drew the line exactly where the technology premium begins. Commodity grades remain licensable — a Western joint venture can, in principle, buy the right to make second- and third-generation powder — and the fourth-generation route, the one worth 21.29% more, must be built independently or licensed from a counterparty that may not be permitted to grant it. Beijing has effectively ring-fenced a two-hundredths-of-a-unit gap.

The American instrument works the other way, through the invoice. The One Big Beautiful Bill Act of 2025 replaced the Inflation Reduction Act’s sourcing rules with a material assistance cost ratio, defined in Section 70512 and applied to the Section 45X advanced manufacturing credit. The ratio is cost-weighted rather than geographic: it counts technology licensing, equipment and processing relationships as assistance, and the permitted thresholds tighten progressively through 2030. A 2026 assessment by Resources for the Future concluded that the LFP chain is unlikely to be sufficiently friendshored before the restrictions bite, and identified midstream processing as the binding constraint. Ford’s Marshall plant is the test case, and it is an uncomfortable one: the plant licenses CATL’s process, and licensing is exactly what the ratio counts.

Europe regulates the product rather than the process. Regulation (EU) 2023/1542 imposes carbon-footprint declarations, a battery passport from February 2027 and recycled-content thresholds from 2031. None of that constrains where the powder comes from. All of it adds fixed cost to a cell, and fixed cost on a cell compresses the fee a cathode maker can negotiate. It is a compliance instrument that lands on the same margin the market already treats as thin.

The asymmetry is the finding. A Chinese restriction raises the cost of building a competitor outside China by making the only valuable process unlicensable. An American cost ratio raises the cost of using Chinese process inside the United States by making the licence count against the credit. A European product regulation raises the fixed cost of every cell regardless of origin. Each of the three bites on a different line of a different document, and the net effect is that the world’s second-largest battery chemistry is being industrialised everywhere except the places writing rules about it.

One scope note, because it changes what these rules can do in the near term: China exported 28,300 tonnes of LFP cathode in January to November 2025, 0.8% of national output. These instruments govern licensing and future plants. They do not govern current trade, because there is barely any current trade to govern.