Editorial Article
The Most Qualified Tonne Sells for the Least
Two reports, one tonne, two prices
Global Phosphate Rock & Yellow Phosphorus sizes the whole phosphorus complex at USD 32,310.0 million in 2025, and inside that total it allocates USD 1,320 million — 4.09 per cent — to battery-grade purified phosphoric acid. The purified acid report sizes its own market at USD 6,160.0 million, and inside that total it puts battery-bound acid at 2.75 million tonnes of 5.60 million, 49.1 per cent of volume, at USD 1,040 a tonne — which is USD 2,858.2 million and 46.4 per cent of value.
So the same tonne appears as USD 1,320 million in one report and USD 2,858 million in the other. The ratio is 2.165. Neither number is wrong, and the reason they differ is the reason this bridge is worth writing down: one counts what purification adds, the other counts what the customer pays, and the rock in the middle is counted once, in the rock pool.
| Measure | Rock & yellow phosphorus report | Purified acid report |
|---|---|---|
| Battery pool, value | USD 1,320m | USD 2,858.2m |
| Share of its own report | 4.09% | 46.4% |
| Battery volume | — | 2.75 Mt of 5.60 Mt (49.1%) |
| Basis | Value added above rock | Realised invoice price |
What four point zero nine per cent actually is
Reconcile it on a single tonne. Battery-bound acid realises USD 1,040. A tonne of 85 per cent purified product carries 3.68 tonnes of phosphate rock, and at USD 526 of rock per tonne of product that bill is 50.58 per cent of the finished price. Subtract it and you have USD 514 of value added per tonne.
Now take the rock report’s number and put it on the same footing. USD 1,320 million divided by 2.75 million tonnes is USD 480 of value added per tonne. The two constructions land 6.61 per cent apart — the gap being the conversion margin the acid report counts inside its 5.49 per cent value line and the rock report deliberately leaves outside its perimeter.
Which means 4.09 per cent is not a small market. It is a purification margin pool of USD 1.3 billion sitting on a rock bill of USD 1,446.5 million, and the only reason it reads small is that the tonnage underneath it is already counted, as rock, in the fertiliser line. Anyone quoting 4.09 per cent as evidence that battery phosphates are immaterial has read the denominator and not the construction.
The discount
Here is the number that should stop a reader. Battery-bound acid realises USD 1,040 a tonne. Food-grade acid realises USD 1,250. The most tightly specified product in the market sells 16.80 per cent below the least tightly specified one.
And the premium that is supposed to compensate for that specification is 2 per cent. China’s exported refined acid is quoted at USD 1,335 to 1,380 a tonne FOB, within 2 per cent of food grade. A cathode maker is not paying for purity. It is buying a product priced as a commodity, by a customer base of perhaps a dozen accounts, on contract, into a market that commissioned 2.48 million tonnes of new purified capacity in 2026 against 4.64 million tonnes of output — a surplus of roughly 570,000 tonnes.
That is a cost-plus business with the plus negotiated. Which is what a 16.8 per cent discount to food grade looks like from the seller’s side of the table.
The moat is not the price
If the premium is 2 per cent, the barrier cannot be economic, and it is not. It is a specification: 10 parts per billion of magnetic foreign matter at the iron phosphate stage, which is where the export-control threshold is written. That is not a certificate attached to a shipment. It is a property of a plant’s equipment, its solvent recovery, its water and its housekeeping — and it is demonstrated by the cathode maker’s own qualification run, not by an assay on a consignment.
This is why the pool that grows is the pool that cannot raise its price. Volume compounds at 8.17 per cent; value at 5.49 per cent; realised revenue per tonne falls 2.48 per cent a year, from USD 1,100 to USD 900 by 2033. Growth in this market subtracts price by arithmetic rather than by competition. A supplier planning a purification plant off the 8.17 per cent is planning a tonne count, not a revenue line.
The strangest fact in the chain
Run the rock backwards. 2.75 million tonnes of battery-bound acid at 3.68 tonnes of rock per tonne is 10.12 million tonnes of rock — 4.05 per cent of the world’s 250 million tonnes.
And the rock report says about 9.2 million tonnes of that rock, 3.70 per cent, goes into an electric furnace to become elemental phosphorus.
Which means the battery chain already draws more phosphate rock than the entire global thermal phosphorus industry. The branch built for detergents, flame retardants and glyphosate is now the smaller of the two non-fertiliser consumers of rock, and it took the battery chain a decade to pass it — while the rock itself grew at only 2.52 per cent a year.
Both sit inside the same four per cent, and that is the real bridge: not a growth rate, a queue. Ten million tonnes of rock is a mine, and what comes out the far end is priced by the two bills that stand between the mine and the cathode — USD 526 of rock and, since May 2026, USD 429 of sulfur.
Two routes, one input bill each
The rock report’s thermal branch and the acid report’s wet branch are the same argument in two currencies. Yellow phosphorus at RMB 26,000 a tonne is 87.4 per cent of the RMB 8,000 thermal acid price — a gross spread of 12.6 per cent before power, electrodes, labour and depreciation. On the wet side, sulfur went from 5.3 per cent of the product price to 43.7 per cent between April 2025 and May 2026, while rock held at 3.68 tonnes and moved by single digits.
So the thermal route is now a bet on a grid tariff and the wet route is a bet on a shipping lane. Neither is the cheap route, and the market price sits where the two input bills meet. That is the honest answer to what 4.09 per cent raises: the pool is small, it is growing fastest, it consumes more rock than the furnace branch, and it is the one pool in the chain where the producer has no pricing power at all.
What falsifies it
Falsifier one, dated to the next published assessment. The discount closes. Battery acid currently realises USD 1,040 against USD 1,250 for food grade, and exported refined acid is within 2 per cent of food grade. If a battery-grade assessment appears at a genuine premium to food grade — not parity, a premium — then qualification has become pricing power and this argument is wrong.
Falsifier two, dated to the first qualified non-Chinese tonne. The purified acid report names North America as the fastest-growing region with no merchant battery-grade capacity. The first qualified merchant battery-grade tonne outside China breaks the specification-as-moat claim, because a moat made of a plant’s housekeeping is not a moat; it is a head start.
The one thing not to do
Do not add the reports. USD 32,310.0 million plus USD 6,160.0 million is not a market. The purified acid pool is 19.07 per cent of the rock pool and the acid’s rock is already inside it, so the sum double-counts a fifth of itself before a single tonne is examined.
The correct reading is a stack: rock, then the furnace branch, then purification, then the cathode. Three conversions, each now dominated by a single input bill, and the smallest one carrying the hardest specification while paying the smallest premium for it.
Frequently Asked Questions
Why is battery-grade purified phosphoric acid only 4.09 per cent of the phosphate rock and yellow phosphorus market?
Because that figure counts only the value purification adds, not the whole invoice. The rock report allocates USD 1,320 million to battery-grade acid out of a USD 32,310.0 million total, while the purified acid report sizes the same battery-bound tonnage at 2.75 million tonnes and USD 2,858.2 million. The difference is the rock bill, which is already counted once in the rock pool. Put per tonne, the two constructions are USD 480 and USD 514 of value added — 6.61 per cent apart.
Does battery-grade phosphoric acid sell at a premium to food grade?
No. Battery-bound acid realises USD 1,040 a tonne against USD 1,250 for food grade, a 16.80 per cent discount, and China's exported refined acid is quoted at USD 1,335 to 1,380 a tonne FOB, within 2 per cent of food grade. The barrier to entry is a specification — 10 parts per billion of magnetic foreign matter at the iron phosphate stage — not a price premium. The buyer is concentrated, contracted, and writes the specification.
How much phosphate rock does the battery chain consume?
2.75 million tonnes of battery-bound acid at 3.68 tonnes of rock per tonne of 85 per cent product is 10.12 million tonnes of rock, or 4.05 per cent of the world's 250 million tonnes. That is more than the roughly 9.2 million tonnes — 3.70 per cent — that goes into electric furnaces to become elemental phosphorus, so the battery chain already draws more rock than the entire global thermal phosphorus industry.
Can the two phosphate reports be added together?
No. The purified acid market of USD 6,160.0 million is 19.07 per cent of the USD 32,310.0 million rock pool, and the acid's rock is already inside that pool, so adding the headline figures double-counts about a fifth of the combined number. Read them as a stack — rock, furnace branch, purification, cathode — with each conversion's own value added counted once.