Editorial Article
The Write-Down That Looked Like a Margin Collapse
A 960 million yuan loss, and where it actually sits
The Chinese trade press has 2025 in the phosphorus flame retardant industry as an involution story: capacity floods in, prices collapse, the leader loses money. The direction of that story is right. The mechanism is not, and the difference is worth RMB 1.1 billion.
Zhejiang Wansheng’s 2025 filing, read line by line, says something considerably more specific than a price war.
| Wansheng, 2025 | RMB m | What it says |
|---|---|---|
| Revenue | 3,377.79 | +13.98 per cent — the company sold more, not less |
| Gross profit / margin | 618.0 / 18.2% | −0.30pp — the operating margin barely moved |
| Net profit, Q1–Q3 | +50.1 | Nine months profitable |
| Net profit, Q4 | −1,010.4 | The entire annual loss, in one quarter |
| Asset impairment, full year | −1,097.45 | 2024: −84.27 — a thirteenfold increase |
| Net profit, full year | −960.35 | +50.1 − 1,010.4 = −960.3 |
The quarterly halves reconcile to the annual figure to within half a million yuan. This is not a reconstruction — it is the reported arithmetic, and it locates the loss in a single quarter and, within that quarter, a single line.
A flat gross margin in the year the price fell ten per cent
The TCPP assessment went from about RMB 15,000 a tonne in 2024 to about 13,500 in 2025 — down 10.0 per cent, the only down year in five and the steepest move in either direction across the period. A price war of that size should appear in the gross margin as a wound. It appears as thirty basis points.
The reason is on the other side of the ledger, and it is the part of this story that gets left out. Propylene oxide — 0.532 tonnes of every tonne of TCPP — fell 12.87 per cent year on year to about RMB 7,500 a tonne by September 2025. Bisphenol A, the feedstock for BDP, fell 21.67 per cent to about 7,755. The price collapse was not absorbed by the producer. It was transmitted, from a feedstock that fell first and further than the product did.
Which produces the correction this article exists to make: the price war of 2025 did not destroy Chinese flame retardant gross margin. It compressed the entire chain, and the producers travelled with it. What it destroyed was the carrying value of assets built on the assumption that the old price level would persist. Those are not the same event and they do not have the same remedy.
Four hundred and fifty thousand tonnes of nameplate against five hundred and forty thousand tonnes of world demand
The impairment has an address, and the capacity table explains it.
| Wansheng base, 2025 | Nameplate | Utilisation | Status |
|---|---|---|---|
| Jiangsu Taixing | 65.3 kt | 46.93% | Operating |
| Shandong Hanfeng | 121.5 kt | 39.21% | Operating |
| Shandong Weifang | 189.6 kt | Being cut | Capacity reduced and relocated to Thailand; impairment of RMB 1,097.45m taken in Q4 |
| Guangdong Zhuhai | 44.0 kt | 38.99% | 38.99 per cent including 25 kt of impact modifier capacity still under construction; 90.29 per cent excluding it |
| Thailand | 32.0 kt | Commissioning | Phosphate ester project; main works complete, equipment being installed |
| Total disclosed | 452.4 kt |
Four hundred and fifty-two thousand tonnes of nameplate against 540.0 thousand tonnes of global consumption. One company, with the disclosed capacity to serve 83.8 per cent of the world’s annual volume. That is the structural fact standing behind every utilisation figure in the table — and behind the observation that the two running sites with the best numbers still ran below half.
Now price the write-down against the thing it was written against. RMB 1,097.45 million over 189.6 kilotonnes is RMB 5,788 per tonne of capacity — 42.9 per cent of a full year’s TCPP price, written off before a single tonne was made at that site again. And RMB 1,097.45 million is 1.78 times the year’s entire gross profit. The company took one year of operating earnings, and most of another, and put all of it against one plant.
The impairment is not “too much capacity”. It is “the wrong jurisdiction”
That distinction is commercial, not academic, and the filing makes it unmissable by what the company does next. In the same year it impaired Weifang, Wansheng had two more projects in build at Weifang — 23.1 kilotonnes of specialty functional materials and 14.2 kilotonnes of high-end new materials phase one, both due for completion during 2026 — and a 32,000-tonne phosphate ester plant commissioning in Thailand, disclosed by the company as a route around trade barriers.
A company with too much capacity closes capacity. This company cut a site, kept building at the same site, and moved production to another country. That is not a capacity judgement. It is a location judgement, and the variable being priced is tariff and trade access, not phosphorus.
The customs data agrees with that reading to an uncomfortable degree. China’s exports under HS 291990 — phosphoric esters and their salts, the heading that carries these molecules — went from USD 952.1 million in 2021 to USD 452.2 million in 2025, a fall of 52.5 per cent. Into 2026 the decline steepened: USD 279.8 million in January to August, down 9.2 per cent, with August alone at USD 28.1 million, down 31.5 per cent.
Volume did not halve across those four years. Capacity grew. The difference is price — and a price that halves while capacity grows is what a market looks like when its largest customers begin to price the origin of the goods into the quotation. Relocating 32 kilotonnes to Thailand is what a producer does once he has concluded that the answer is yes.
What a hundred thousand tonnes says about the industry’s own expectation
Nantong Jiangshan discloses the other half of the Chinese supply side: 80,000 tonnes a year of TCPP, 15,000 of BDP, and a 100,000-tonne flame retardant and supporting unit complex planned for Guizhou. In August 2025 an investor asked the company directly why its flame retardant gross margin was low. The company’s answer was the industry’s answer: because it is low.
That combination is the most informative thing in this file. A producer who states publicly that the margin is low is at the same time committing 100,000 tonnes of new capacity, and impairment or no impairment, the industry is building. Which means the industry does not expect the margin to be the return. It expects the position to be. Scale in a market that is fragmented by count and oligopolistic by capacity is an option on a consolidation that has not happened yet, and Chinese chemical producers have repeatedly paid to hold that option through a downcycle.
What this does to the forecast
The market grows from 540.0 thousand tonnes to 664.1 at 2.62 per cent a year, and from USD 1,041.7 million to 1,447.7 million at 4.20 per cent. Those are not demand forecasts in any ordinary sense. Global volume growing at 2.62 per cent while one producer holds nameplate for 83.8 per cent of it is a forecast of capacity discipline, not of consumption — and it is the assumption most likely to be wrong in either direction.
What the impairment adds is that the discipline is being enforced by balance sheets rather than by agreement. That is a slower and more expensive mechanism, and it has a signature worth watching: further impairments concentrated at export-oriented Chinese sites, more announcements in jurisdictions with trade access, and a persistent gap between a recovering gross margin and a flat net one.
What would falsify this reading
Honesty requires the other interpretation to be stated, because it is not weak. An impairment is a valuation of the future. A write-down of this size is partly the board agreeing, on the record and under audit, that the margin level is not coming back. On that reading Q4 is not an accounting event at all — it is the moment the accounts caught up with what the price had been saying since 2023.
The two readings are not exclusive, and the numbers available do not yet separate them. What separates them is what happens next, and the test is simple enough to state. If this was a margin problem, the response is price, and a producer with nameplate for 83.8 per cent of the world has every incentive to hold it. If it was a footprint problem, the response is more capacity where the customer will buy from it and further closures where he will not. Watch the Weifang projects still under construction, and watch whether the next impairment lands at a site that sells into Europe or one that sells into China.
Nine months of profit followed by one quarter of write-down is not a company in retreat. It is a company that has decided where it intends to stand — and paid 1.78 times a year of gross profit for the privilege of standing there.