Editorial Article
Phosphate Is 22.7 Per Cent of the Tonnage and 32.5 Per Cent of the Value. Potash Is the Other Way Round.
Fertiliser is three markets sharing a name, and in 2025 they stopped moving together at all. Nitrogen cleared off a gas price, phosphate off an export licence, and potash off a production schedule that was fixed years ago and cannot respond to any of it.
The consequence is arithmetic, and it is the most useful number in this market: phosphate is 22.7 per cent of the tonnage and 32.5 per cent of the value. Potash is 20.1 per cent of the tonnage and 12.7 per cent of the value. The nutrient that most determines this market’s revenue is not the one that determines its volume, and neither of them is the one whose supply is most predictable.
Nitrogen is priced off gas
In the last week of October the Yara–Mosaic settlement put Tampa at USD 645 per tonne CFR, up USD 95 in a month, while Northwest Europe traded at 710–730 CFR and Chinese export ammonia sat at 510–535 FOB. That is a 40 per cent spread on the same molecule inside a single pricing window, and it is not arbitrage that will close quickly: with about one-third of European ammonia capacity curtailed and TTF forwards above EUR 100 per MWh, the marginal European tonne is no longer a competitor in its own market.
Phosphate is priced by an export licence
The NDRC suspended Chinese phosphate exports in December 2025 until August 2026. As of September only TSP and SSP were inspection-eligible. Indian DAP was assessed at USD 890–900 per tonne CFR, Brazilian buyers switched into TSP — which needs no ammonia at all — and sulphur sat at 7,500 yuan per tonne. None of this responds to demand. It responds to a regulatory calendar in one country.
Potash is priced by a schedule
Jansen Stage 1 is 87 per cent complete and produces first tonnes in mid-2027, with roughly 15 million tonnes of new capacity arriving between 2027 and 2030 into a balance the IFA already scores at a 14 per cent surplus of supply capability. Potash is the one nutrient here where the incremental tonne has a date, a budget and a sinking shaft behind it.
The decade is two-speed, and the slow half is potash
On our path this market compounds at 2.52 per cent to 2029 and 4.15 per cent thereafter. The slow half is potash capacity landing before potash demand does. The fast half is phosphate, on soil depletion and purified phosphoric acid demand.
That two-speed shape matters because the near-term slope is not the trend, and the gap between them is enormous. The World Bank fertiliser index went 138.7 in 2025 to 188.2 in the second quarter of 2026 — a 35.7 per cent move in two quarters, against a value-per-tonne trend of 1.85 per cent a year — before falling back to 147.6 in the third quarter. The cycle is larger than the trend, in both directions, and anyone annualising the 2026 slope over-forecasts this decade badly.
Europe is exiting the top of the nitrogen cost curve
The judgement with the longest tail is not about price. It is about who exits, and it cuts both ways in sequence.
Europe is shutting its highest-cost ammonia capacity, and that is bullish now and bearish later. Removing the most expensive running plant raises the clearing price while trade is disrupted, and lowers it once trade normalises — because a commodity market clears on the last survivor, not the last casualty. The capacity that closes was setting the high end of the cost curve; the capacity that stays sets the low end, and it is still there when the disruption ends.
Against that, the green-ammonia offset is a statement of intent rather than a supply forecast: 3.1 million tonnes of nitrogen commissioned by 2029 against close to 155 million tonnes under consideration. A fifty-to-one ratio of pipeline to programme is what an industry looks like when it has announced its transition and not yet funded it.
The market in one line
USD 172.38 billion in 2025 to USD 224.00 billion in 2033, a 3.33 per cent CAGR, on 209.0 to 234.5 million tonnes of N+P2O5+K2O at 1.45 per cent. The split is 44 per cent tonnage and 56 per cent price and mix. Value per nutrient tonne goes from USD 824.9 to 955.2 — 1.85 per cent a year, which is less than the single 2025-to-2026 price move.
That is the most important sentence in this market. It says the long-run return in fertiliser is set by product mix and by the cost of the marginal molecule, and that a decade of forecasting can be erased or doubled by one gas contract and one export licence.
Full dataset, nutrient-level segmentation and the eight-year value path are in the Global Fertiliser (NPK) Market report.
Frequently Asked Questions
Why do phosphate and potash have such different value shares?
Phosphate is 22.7 per cent of the tonnage and 32.5 per cent of the value; potash is 20.1 per cent of the tonnage and 12.7 per cent of the value. The nutrient that most determines market revenue is therefore not the one that determines volume, and the one with the most predictable supply — potash, with roughly 15 million tonnes of new capacity arriving 2027–2030 — is the one that contributes least value per tonne.
What actually sets fertiliser prices across the three nutrients?
Nitrogen is priced off gas: Tampa cleared at USD 645 per tonne CFR in late October, up USD 95 in a month, against Northwest Europe at 710–730 CFR and Chinese export ammonia at 510–535 FOB, with about one-third of European ammonia capacity curtailed. Phosphate is priced by an export licence, with Chinese exports suspended from December 2025 until August 2026. Potash is priced by a schedule, because new capacity takes years to build.
Why is the fertiliser forecast two-speed?
2.52 per cent to 2029 and 4.15 per cent from 2029 to 2033, for a blended 3.33 per cent. The slow half is potash capacity landing before potash demand does; the fast half is phosphate, driven by soil depletion and purified phosphoric acid demand. The 2026 price slope is not the trend — the World Bank index moved 35.7 per cent in two quarters before falling back.
Why is Europe exiting the top of the cost curve bullish now and bearish later?
Because the market clears on the last survivor rather than the last casualty. Closing the most expensive running ammonia plant raises the clearing price while trade is disrupted, then lowers it once trade normalises and the remaining low-cost capacity sets the price. Green ammonia does not offset it in this window: 3.1 million tonnes of nitrogen is commissioned by 2029 against close to 155 million tonnes under consideration.