The Shock That Reports Late

Analysis · 1 September 2026 · predictions registered before research · honesty score −10, reported at a loss

The 2026 fertiliser shock is being written up as absorbed. The indicator carrying that verdict is a price that round-tripped inside six months. The indicator that would carry the damage reports in the 2026/27 marketing year, and it has not reported.

1 September 2026

The all-clear is being sounded, and it is worth quoting precisely.

Writing for the Economist Intelligence Unit on 24 August, Anushree Ganeriwala concluded that "resilience came from a combination of buffers and adaptability. Inventories bought time, while flexible producers and responsive markets prevented a fertiliser shock from becoming a food shock." Attention, she noted, is now shifting to weather.

That verdict rests on a set of facts that are not in dispute. What is in dispute is whether the facts it rests on are the ones that will decide the question.

Nobody missed this

It is tempting to reach for a story about institutional blindness — a technical channel overlooked while everyone watched the oil price. That story is unavailable, and the record is emphatic about it.

The Strait of Hormuz closed effectively at the end of February. By 25 March, CNBC had run the fertiliser transmission for a general audience. On 1 April, Charlotte Hebebrand, Joseph Glauber, Rob Vos and Brendan Rice published it at IFPRI. On 29 April, Shawn Arita and co-authors modelled disruption scenarios for US crop producers in farmdoc daily. The World Bank's Open Data blog, the WTO's Data Blog in July, AMIS, CSIS and the IFDC have each published on it since, and the European Commission's Fertiliser Action Plan is a policy response to it. Rob Vos's July survey for IFPRI and Welthungerhilfe is the most complete single statement of the channel.

Eleven institutions, five months, and the mechanism identified inside four weeks of the closure. Whatever went wrong here, nobody was looking the other way. That is what makes the remaining question sharp: if attention was not the binding constraint, what is?

The fast indicator

Urea is the price everyone watched, and it has been round and come back.

The EIU's series, built on ICE and CBOT data, has urea rising about 56 percent between March and mid-April, from around $462 a tonne to a peak of $720, and falling by more than 45 percent from that peak since. World Bank Pink Sheet figures as reported run higher at the top — from roughly $400 to over $850 in April, back to $453 in June — and record a 77 percent jump from mid-December 2025 to 9 March. The series disagree about amplitude. They agree about shape.

Three things brought it down, all named by the EIU: lower energy prices, China's resumption of urea exports at the end of May after restricting them further on 19 March, and improving supply availability. India's most recent fertiliser import tender was heavily oversubscribed. Vos records that other suppliers partially replaced Gulf exports, at higher prices.

Measured against the shock that produced it, this is a substitution success. Roughly 3.9 million tonnes of Gulf fertiliser exports were suspended after end-February — about thirty percent of the Gulf's annual fertiliser exports, from a region carrying about a third of world urea trade. The price is back where it started.

On the fast indicator, the system passed.

What the adaptations actually were

The EIU's account of why it passed names three producer responses, each offered as evidence of resilience. US farmers shifted acreage from maize to soyabeans, which need less nitrogen. Some Thai rice farmers cut fertiliser application rates. Brazilian producers moved to alternative nutrient blends and cheaper options.

And one system response, in the same piece: inventories cushioned the shock initially, and "those buffers have since been drawn down, reducing the capacity to absorb further supply disruptions."

None of those four is a costless substitution. Each moves a cost out of 2026 and into the following crop year.

Acreage switched away from maize produces less maize. Application rates cut below agronomic optimum produce lower yields on the acres they were cut on. Drawn-down inventories are, by definition, grain that has already been eaten. The price recovered in part because demand for the input was destroyed and stock was consumed, and both of those show up on a different schedule than the price does.

The American figures give the demand destruction a size. At the end of January a US producer needed 127.74 bushels of corn to buy a tonne of urea, about fifteen percent worse than a year earlier. Corn was trading around $4.40 to $4.60 a bushel against $7.50 and above in 2022. The American Farm Bureau Federation's own survey of its own members reports 78 percent of Southern-region farmers unable to afford all needed inputs, with 69 percent in the Northeast and 66 percent in the West.

The 2022 comparison runs backwards

Every account of this shock reaches for 2022, and every one of them draws comfort from the difference. Unlike 2022, there is no grain shortage. Unlike 2022, cereal prices are stable. Unlike 2022, markets entered the crisis with adequate stocks.

All true, all good for consumers, and on the specific question of next year's harvest it runs the other way.

In 2022 fertiliser and grain rose together. The affordability ratio had two moving terms, and a farmer who paid more also received more. In 2026 grain and oilseed prices rose 11.6 percent over the conflict's first 161 days, against 18.4 percent in the equivalent window after Russia's invasion.

The absence of a grain price response removes the only term in that ratio that was ever going to correct on its own. A farmer deciding how much nitrogen to buy in 2026 faces a high input price and no compensating output price, which is a worse decision than the same farmer faced in 2022 at a comparable urea quote.

The slow indicator

The 2026/27 numbers exist as forecasts, and they are not reassuring.

The International Grains Council has 2026/27 world total grains production at 2,416 million tonnes, cut by five million, against consumption of 2,444 million tonnes. That is a twenty-eight-million-tonne gap filled from stocks, with cumulative inventories at 606 million tonnes and the outlook described as continuing to tighten. USDA has global demand outpacing production for 2026/27, with stocks down two percent for wheat, two percent for rice and seven percent for corn.

Corn is the largest decline, and maize is the crop US producers switched away from on nitrogen cost. That is a correlation, and this article does not assert it is the cause.

Sitting in the same window: NOAA declared El Niño conditions in June, and attention has moved to a possible "super El Niño" across the 2026/27 crop cycle, with rice producers in India, Thailand, Vietnam and Indonesia named as most exposed. Vos notes that even had the conflict ended in mid-June, fertiliser markets would not normalise before the end of 2026, and cites a scenario extending the crisis into 2028. Restarting a shut fertiliser plant takes five to eight weeks; with structural damage, years.

The claim

Urea price reports daily and reverted in five months. Yields and stock-to-use report annually and have not reported. The resilience verdicts are being written in August, on the only variable that had time to come back.

What decided whether this shock was recorded as absorbed is the reporting lag of the indicator used to score it, and that is a property of the measurement rather than of the shock.

The test is dated and this article can be checked against it. If 2026/27 production and closing stocks come in at or above the pre-shock trend, the August verdict was right and this piece was wrong.

The weakest joint

The attribution is the soft place, and it is soft in a way no amount of further reading fixes.

The 2026/27 stock tightening, the fertiliser shock and a declared El Niño occupy the same window, and nothing found here separates their contributions. The argument is that the slow indicator was not consulted before the verdict was issued. It does not establish what the slow indicator will say, or that fertiliser will be the reason it says it.

The second soft place is the size of the application cuts. Controlled agronomic work finds that nitrogen reductions of ten to twenty percent do not significantly reduce maize growth, while a fifty percent cut reduces corn yield by about seventeen percent. Nothing found establishes where 2026's actual cuts fell in that range. If they fell at the shallow end, the deferral this article describes is real and small.

What this does not claim

No part of the WFP's projection of up to 45 million additional people in acute food insecurity, nor of the 2.5 million in Somalia, 2.3 million in Afghanistan and 1.3 million in Sri Lanka recorded as at 4 June. Those are the agency's own figures about its own caseload, and the projection is explicitly conditioned on oil near $100 a barrel through end-June. The exposures named for those countries are energy, remittances and trade — Sri Lanka draws 63 percent of its energy from the Middle East, 44 percent of remittances from the Gulf, and sends 45 percent of its tea there. Fertiliser is not a named driver of the 2026 caseload, and nothing here suggests it should be.

Nor is the fertiliser channel resolved everywhere. Regional exposure is uneven and documented: Brazil imports about forty percent of its fertiliser through Hormuz and saw local prices rise over thirty-five percent in the first fortnight; India had enough for kharif and faces an uncertain rabi; East African farmers draw about a third of their supply from the Gulf.

Sources

1. Anushree Ganeriwala, "Beyond the farm gate: lessons from the 2026 fertiliser shock" — Economist Intelligence Unit, 24 Aug 2026.
2. Rob Vos, "Iran war: From fertilizer to food crisis?" — IFPRI / Welthungerhilfe, 2 July 2026.
3. Hebebrand, Glauber, Vos and Rice, "The Iran war's impacts on global fertilizer markets and food production" — IFPRI, 1 April 2026.
4. Arita, Wang, Kim, Chakravorty and Steinbach, "Strait of Hormuz Disruption Scenarios and Fertilizer Purchasing Risks for U.S. Crop Producers" — farmdoc daily 16:75, 29 April 2026.
5. "Fertilizer prices surge as Strait of Hormuz disruptions tighten supplies" — World Bank Open Data blog.
6. "Fertilizer trade impacted by Strait of Hormuz conflict" — WTO Data Blog, 10 July 2026.
7. "Chokepoint: How the War with Iran Threatens Global Food Security" — CSIS.
8. "It's not just oil and gas. The Strait of Hormuz blockage is rattling another vital commodity" — CNBC, 25 March 2026.
9. "Fertilizer Crisis Response Bulletin #26: From Energy Shock to Food Production Risk" — IFDC, 25 Aug 2026.
10. Fertiliser Action Plan, COM(2026) 310 final — European Commission.
11. Hormuz shock: global and regional impacts on fertilizer markets — AGRICULTURAL MARKET INFORMATION SYSTEM.
12. Grain Market Report summary — International Grains Council, 2026/27 production, consumption and stocks.
13. World Agricultural Supply and Demand Estimates, August 2026 — USDA.
14. "Fertilizer markets struggle with supply constraints, farmer affordability" — CoBank.
15. "Farm Bureau Survey Reveals Real Impact of Fertilizer Availability and Price" — American Farm Bureau Federation, reporting on its own members.
16. "Conflict and Food Insecurity, August 2026 Monthly Forecast" — Security Council Report, carrying the WFP projection and country figures.
17. "European ammonia production costs exceed imports as gas prices surge" — S&P Global, 4 March 2026.
18. "China allows fresh urea exports amid Iran war-fueled fertilizer crisis" — Hydrocarbon Processing.

Produced under provenance discipline. Falsifiable predictions were registered to a signed chain before any research began, verdicts were recorded before any prose existed, and the registered thesis was destroyed by prior art on the third research step and replaced. Honesty score −10, reported at a loss: one point for every research step the platform could not carry. The run is sealed on the host that took the writes. The chain establishes the order in which the work was recorded. It does not establish that any source cited here is correct.

sha256 9ab9702e14625eb6a0271514fcdd74db064096d17700caffa744ba8572efb8cd