Anhydrous ammonia, US retail
$945/ton
+22% y/y
Potash, US retail
$495/ton
+2% y/y
Vancouver spot sulphur
~$1,040/ton
peak $1,090
Corn needed to buy a ton of NH₃
197bushels
at $4.80/bu
Illinois corn operator return
$36/acre, 2027
from $59 in 2026
  • This is not a broad fertiliser crisis. Anhydrous ammonia is +22% year on year at $945 per ton while potash is +2% at $495 S. The shock is concentrated in nitrogen and in the sulphur that phosphate production depends on. Potash is absent from it, and the reason is a policy change rather than a market one.
  • A ton of anhydrous now costs 197 bushels of corn at the USDA season-average price of $4.80 KR. The nitrogen-to-corn price ratio is 0.121, in the upper third of the range agronomic rate tables are built across, which mechanically pulls economically optimal nitrogen rates down by roughly 15 to 25 pounds an acre against a normal year KR.
  • The rational grower response and the profitable one point in opposite directions. Nitrogen is the expensive nutrient and the one that cannot be skipped without an immediate yield penalty. Phosphate and potash are bankable in the soil. The correct deferral is P and K, not N, and we expect widespread single-season skipping of P and K maintenance across the coming application window.
  • The most telling number of the quarter is a producer's income statement. The world's largest phosphate exporter reported first-half 2026 revenue down 7.3% year on year through the sharpest phosphate price spike in four years, with second-quarter capital expenditure down 36% S. Volume loss and the sulphur cost squeeze more than offset price, and the capex signal is bearish for any 2027 and 2028 supply response.
  • The Strait of Hormuz is widely described as closed continuously since 28 February 2026. It has not been. There was a documented partial reopening in April, a reopening under a June memorandum, and a re-closure on 22 June S. The distinction matters because the sulphur move usually attributed to the closure is measured from a base that predates it.
01

What the sulphur shock actually is, and what we will not print

Sulphur is the cost that moved, and most published accounts of it are imprecise in the same two ways. This section fixes the timeline and the base before the rest of the note builds on them, because the affordability arithmetic that follows is only as good as these two numbers are.

The strait has not been continuously closed. The 28 February 2026 start date holds. What does not hold is continuous closure. The sequence on the public record is closure on 28 February, restricted escorted passage in early March, a US reopening operation from 19 March, partial operation under a ceasefire in April, a memorandum in mid-June under which the waterway was to reopen, a renewed shutdown on 22 June, collapse of the interim ceasefire on 8 July, and reinstatement of a naval blockade in mid-July, with mine clearance from the traffic separation scheme confirmed in late August S.

We will not print a transit count. Two commercial trackers disagree materially on current throughput, and one of them is internally inconsistent within a single page. Neither is a primary maritime authority. We will report the direction, the war-risk insurance multiple and the crude price as a market-verifiable proxy, and we will not print a vessel count we cannot source.

The sulphur base is widely misstated, and the precise version is a better claim. The move is commonly given as roughly $400 to $1,000 or $1,200 a ton. The $400 level is end-2025, before the closure: sulphur had already doubled from below $100 in January 2024 before the strait shut S. The Hormuz-attributable move is more precisely $525 to $1,150 per tonne CFR Brazil between late February and early May 2026, a 119% rise, alongside sulphuric acid CFR US Gulf moving from $155 to $400, up 158% S. That is the tighter and more defensible statement, and it is the one the rest of this note uses.

Exhibit 1: Sulphur, not phosphate rock, is what broke the phosphate market
Vancouver spot sulphur, US$ per ton
US$/ton100Jan 24200Spring 25400End 251,040Aug 26
Note: January 2024 and spring 2025 points are stated as below $100 and above $200 respectively and are plotted at those thresholds. August 2026 peaked at $1,090. The end-2025 level of about $400 predates the Strait of Hormuz closure: sulphur had already doubled before the strait shut.
Source: farmdoc daily, University of Illinois, 25 August 2026
02

The affordability arithmetic, and the asymmetry inside it

At DTN's retail prices for the survey week of 14 to 18 September 2026 and the USDA season-average corn price of $4.80 a bushel from the 11 September WASDE, a ton of anhydrous costs 197 bushels, DAP 193, urea 137 and potash 103 KR. On the $5.00 corn assumption used in the 2027 Illinois budgets those fall to 189, 185, 132 and 99.

Affordability improved slightly, and it is worth being precise about why.

Anhydrous rose 2.4% over the three weeks to 18 September while the WASDE lifted the season-average corn price 6.7%, from $4.50 to $4.80, on a yield cut to 178.5 bushels an acre and ending stocks down to 1.567 billion bushels P. Corn outran ammonia. The bushel cost of a ton of anhydrous therefore fell from 205 to 197 even though the dollar price rose KR. This does not change the deferral argument below, which is about which nutrients can be skipped rather than about absolute affordability, but anyone modelling a worsening squeeze through the autumn should note that the squeeze eased marginally.

Nitrogen unit cost is $0.58 per pound of N in anhydrous, against $0.72 in urea and $0.75 in UAN28 KR. Divided by $4.80 corn, the nitrogen-to-corn price ratio is 0.121, and at $5.00 corn it is 0.116. Agronomic rate tables typically span 0.05 to 0.15, so 0.12 sits in the upper third. That pulls economically optimal nitrogen rates down by roughly 15 to 25 pounds an acre against a 0.07 to 0.08 ratio year KR. Across 96.7 million planted corn acres that is a demand signal of the order of 0.7 to 1.2 million tons of nitrogen, and we would state it as an order of magnitude rather than a point estimate.

The recommendation we would actually give a co-op

A grower cutting the fertiliser bill will cut nitrogen first, because it is the largest line and it is bought fresh every year. Nitrogen is also the one nutrient that cannot be deferred without an immediate yield penalty. Phosphate and potash sit in the soil, and Corn Belt soil test levels are generally adequate to high after a decade of build.

So the profitable advice inverts the intuitive one: defer P and K, hold N. Maintenance-only P and K on a corn acre runs to roughly $88 at DAP $925 and potash $495 KR, against a projected 2027 Illinois corn operator return of $36 an acre S. Maintenance alone is 2.4 times the projected return. That is not a demand collapse; it is a deferral, and the tonnage returns in 2028 and 2029 as a restocking cycle. For a retailer this is an inventory timing call.

Exhibit 2: A ton of anhydrous now costs 197 bushels of corn
Bushels of corn required to buy one ton of product
Product@ $4.80/bu@ $5.00/bu
Anhydrous ammonia197189
MAP201193
DAP193185
Urea137132
Potash10399
Note: Kastra Research calculation from DTN retail prices for the survey week of 14 to 18 September 2026 against the USDA season-average corn price of $4.80 per bushel in the September 2026 WASDE, and against the $5.00 assumption used in the 2027 Illinois crop budgets.
Source: DTN, 23 September 2026; USDA WASDE, 11 September 2026; farmdoc daily, 1 September 2026; Kastra Research

One further detail belongs in the fall application conversation, and it has turned since our last survey. Nutrient prices stopped falling: six of eight products rose month on month in the 14 to 18 September DTN survey, none of them by the 5% DTN treats as a sizeable move S. The case for waiting on a further decline has weakened. Meanwhile the fuel line has moved violently. US on-highway diesel averaged $6.529 a gallon in the week ending 21 September 2026, up $2.780 year on year, a rise of 74% P. The nutrient has flattened and the tractor pass has repriced. Application timing advice that ignores the fuel line is incomplete this autumn, and on these numbers the fuel is now the faster-moving half of the applied cost.

03

Sulphur is the transmission mechanism, and OCP has told us what it thinks

DAP is roughly 46% P₂O₅ and requires both ammonia and sulphuric acid. Producing a tonne of P₂O₅ consumes on the order of 2.5 to 3 tonnes of sulphur as acid. At $1,040 to $1,150 a tonne against a sub-$150 historical norm, the sulphur input alone can approach or exceed the historical total production cost of DAP.

Three shocks arrived at once, each independently documented. Saudi Arabia and the UAE produced 13,500 thousand tons of world sulphur in 2025, essentially all byproduct from sour gas processing and refining, and essentially all shipped through the Hormuz corridor P. China announced a de facto export ban on smelter byproduct sulphuric acid on 10 April 2026, effective from May, removing 4.65 million tonnes of 2025 exports from the market S. And Kazakhstan and Iraq, at 5% and 4% of US elemental sulphur imports respectively, are exposed to the same corridor P.

The most informative response came from the producer, not the price.

OCP cut output by approximately 30% in the second quarter of 2026 and pivoted its product mix toward TSP, which requires far less sulphur and no ammonia. TSP is now 65% of its crop nutrition volumes S. Then, on 2 September 2026, it reported first-half revenue of MAD 48.37 billion against MAD 52.17 billion a year earlier, down 7.3%, with second-quarter capex down 36% S.

Read those two together. The world's dominant phosphate producer posted falling revenue through the sharpest phosphate price spike in four years, and is conserving cash rather than expanding into the shortage. The product pivot is a producer telling us through capital allocation, rather than through a press release, that it does not expect sulphur to normalise soon. We would expect the DAP to TSP spread to compress structurally and TSP availability to improve relative to DAP, which is a concrete procurement recommendation for any phosphate buyer who can agronomically substitute.

Exhibit 3: Two Gulf producers alone account for a fifth of world sulphur, and it ships through Hormuz
World sulphur production by country, thousand metric tons, 2025
China
19,000
Russia
7,500
Saudi Arabia
7,200
United Arab Emirates
6,300
Canada
5,000
Kazakhstan
4,800
Note: Saudi Arabia and the UAE together produced 13,500 thousand tons, essentially all of it byproduct from sour gas processing and refining, and essentially all of it exported through or immediately adjacent to the Strait of Hormuz. Kazakhstan is highlighted separately because it supplies 5% of US elemental sulphur imports and is exposed to the same corridor.
Source: USGS, Mineral Commodity Summaries 2026, sulfur chapter, published February 2026
04

The nitrogen cost curve has never been this wide

Henry Hub is forecast at $2.87 per MMBtu for the third quarter of 2026, cut fifty cents from the prior month's outlook on reduced LNG feedgas demand and strong domestic production P. European TTF has moved above €80 per MWh in late September 2026, against €71.96 on 4 September, on low storage into winter S, which converts to roughly $27 to $28 per MMBtu.

At the industry convention of approximately 34 MMBtu of gas per tonne of ammonia, that is a gas cost of about $98 a tonne in the United States against $930 to $950 in Europe, a spread of roughly $830 to $850 KR. Against a US anhydrous retail price of $945 a ton, the entire European gas bill for making a tonne of ammonia now roughly equals what a US farmer pays at retail for the finished product. The curve was already the widest on record at our last cutoff; three weeks of European gas took it wider.

Four consequences follow, and they compound rather than offset. US nitrogen producers are running the widest feedstock advantage on record, so US anhydrous at +21% is a margin story as much as a cost story. European curtailment removes supply from the Atlantic basin, supporting US and Middle East netbacks, so the Hormuz shock and the European gas shock reinforce each other. The EU's escalating nitrogen tariff sits on top of that, leaving European growers with the world's highest gas-based production cost and a tariff wall against the imports that would relieve it. And the LNG channel runs the other way: EIA explicitly cut its US gas forecast on reduced LNG feedgas demand, so the disruption is subsidising US nitrogen producers by trapping gas domestically.

Exhibit 4: European ammonia is structurally underwater on gas alone
Gas cost per tonne of ammonia at approximately 34 MMBtu per tonne, US$
US Henry Hub
~$98
Europe TTF
~$940
Note: Henry Hub at the EIA's Q3 2026 forecast of $2.87/MMBtu; TTF above €80/MWh in late September 2026, converted at 3.412 MMBtu per MWh and an assumed $1.15 to $1.18 per euro. The 34 MMBtu per tonne conversion is an industry engineering convention, not a sourced figure, and the FX assumption is ours.
Source: EIA Short-Term Energy Outlook, 11 August 2026; TTF quotation via market data aggregator, 24 September 2026; Kastra Research
05

Trade policy, and the exposure nobody is writing about

The Federal Register carries a phosphate duty story that has not reached the trade press. On 24 July 2026 Commerce preliminarily found that revoking the countervailing duty order on Moroccan phosphate fertiliser would likely lead to continued subsidy, with a rate likely to prevail of 20.04% ad valorem P. On 17 April 2026 the final results of the 2023 administrative review set the Russian producer's cash deposit rate at 12.71% P. Then, on 19 August 2026, following a Court of International Trade judgment, the amended final results for the 2020 to 2021 period raised that producer's rate to 22.86% P.

The operative forward rate is 12.71%, while a court decision has just pushed the historical rate up to 22.86%. Importers with unliquidated entries from the 2020 to 2021 period face retroactive exposure, and we have seen this discussed nowhere. Separately, the ITC elected on 17 June 2026 to conduct full rather than expedited five-year reviews for both Morocco and Russia P, which pushes the injury determination into 2027 and creates a second catalyst.

The largest open question in this market

China's phosphate export restriction was never a published regulation. It was an industry consensus reached in December 2025 under National Development and Reform Commission direction that no new export plans would be scheduled before August 2026 S. There is therefore no legal instrument to expire and no announcement to wait for.

That window lapsed on 31 August 2026. The urea half has since resolved, and it resolved exactly as an unpublished regime must: through volumes rather than through a notice. Chinese urea shipments ran approximately 403,000 tonnes in July against 7,000 tonnes in June, and the 2026 export allowance is now reported at roughly 5.0 to 5.5 million tonnes S. The global urea benchmark stood at $443 per tonne on 4 September 2026, with an Indian tender clearing below $400 S, far beneath the $660 to $670 FOB floor that governed the June to August window. A floor that low a benchmark has passed through is no longer an operative floor.

The phosphate half has not resolved. We could find no source, primary or secondary, establishing what replaced the NDRC arrangement after 31 August U, and US retail DAP and MAP at +2% and +5% year on year are not yet telling us. That is now the single highest-value unknown in this market. The urea precedent says to expect it in customs volumes before any document, and we will lead with it the moment the volumes turn.

Potash, finally, explains itself. The United States eased sanctions on Belaruskali and the Belarusian Potash Company on 19 March 2026 S. EU sanctions remain fully in force and the Klaipėda rail route stays shut, so tonnes must route through Russian rail. The market largely shrugged. But it is why potash is +2% while everything around it moved, and it is why potash is the nutrient a buyer can afford to be relaxed about this season.

Position summary

Defer P and K, hold N. The intuitive cut and the profitable cut are opposites this year.

Sulphur is the binding constraint on phosphate, not rock and not ammonia.

Substitute into TSP where agronomy allows. The largest producer already has.

US nitrogen margin is the widest on record. Gas cost per tonne of ammonia is roughly $98 here against $940 in Europe.

Potash is not in this crisis. Do not price it as though it is.

Exhibit A: This is not a broad fertiliser crisis. It is a nitrogen and sulphur crisis with potash absent
US retail fertiliser prices, survey week 14 to 18 September 2026
Product$/tonY/y
Anhydrous ammonia945+22%
MAP967+5%
DAP925+2%
10-34-0718+8%
Urea659+6%
UAN32458−3%
UAN28421+1%
Potash495+2%
Note: Six of eight products rose month on month, none by the 5% DTN treats as a sizeable move: prices have flattened at a high level rather than continuing to fall. Anhydrous is the outlier on the year. Potash at +2% year on year has not participated in the crisis at all.
Source: DTN retail fertiliser survey, published 23 September 2026
Exhibit B: Barge freight rises 27% into harvest, and barges carry fertiliser upbound
Mississippi system barge tariff rates, spot against October forward
tariff index02505007501,000931Twin Cities909Mid-Miss886Lower Ill.14 July spotOctober forward increment
Note: Barges move grain downbound and fertiliser upbound. A 27% rise in the forward is a direct delivered-cost addition to fall-applied phosphate and potash into the upper Midwest.
Source: USDA AMS Grain Transportation Report bundle, 14 July 2026

How to read our figures

Every material number in this note carries a provenance tag. We do not present sourced data and our own estimates as the same thing.

PConfirmed at a primary source we name
SSecondary reporting of primary data; reporter named
KRKastra Research calculation; method stated
UCirculating but unverified; do not rely on it

What we could not verify

Chinese phosphate export policy after 31 August 2026. The arrangement lapsed and no successor was located. The urea half of the same arrangement has resolved and is reported in section 05.

Current Strait of Hormuz transit counts. Commercial trackers conflict and neither is a maritime authority.

Russian fertiliser export quota volumes and expiry dates. Asserted in trade press with no volumes, dates or ministry named.

Chinese glyphosate and glufosinate pricing. We found no verifiable 2026 figure and have dropped crop protection from this edition rather than run soft numbers.

September 2026 World Bank Pink Sheet levels. Our latest sourced commentary is from May 2026.

06

Politics and policy

Fertiliser is the most politically administered of the markets we cover, and this season it is administered by four separate authorities pulling in different directions. Two of the four do not publish. The section below names the instrument in each case, because the instrument is what tells a buyer whether a change is announced in advance or discovered from volumes.

  1. United States

    Two dated trade-remedy catalysts, and a retroactive exposure

    On 24 July 2026 Commerce preliminarily found that revoking the countervailing duty order on Moroccan phosphate would likely lead to continued subsidy, at a rate likely to prevail of 20.04% ad valorem P. The 2023 administrative review set the Russian producer's cash deposit rate at 12.71% on 17 April 2026 P; then, following a Court of International Trade judgment, the amended final results for 2020 to 2021 raised that producer's rate to 22.86% on 19 August 2026 P.

    Separately the ITC elected on 17 June 2026 to conduct full rather than expedited five-year reviews for both Morocco and Russia P, pushing the injury determination into 2027.

    Read. The forward rate is 12.71% while a court has just pushed the historical rate to 22.86%, so importers with unliquidated entries from 2020 to 2021 face retroactive exposure that we have seen discussed nowhere. The full reviews create a second, dated catalyst in 2027. Neither is priced.

  2. China

    Two administered regimes lapsed together, and neither was ever published

    China's phosphate export restriction was never a published regulation. It was an industry consensus reached in December 2025 under National Development and Reform Commission direction that no new export plans would be scheduled before August 2026 S. The parallel urea arrangement reopened exports for June to August 2026 under a price floor of $660 to $670 per tonne FOB and a quota of 1.5 to 1.6 million tonnes S. Both lapsed on 31 August 2026. The urea regime has since reopened at scale - roughly 403,000 tonnes shipped in July against 7,000 in June, on a 2026 allowance now put at 5.0 to 5.5 million tonnes S - while the phosphate side remains unlocated U.

    Because there is no legal instrument, there is nothing to expire and no announcement to wait for. The policy will be observable only in customs volumes, after the fact.

    Read. The urea reopening is the proof of the method: it was legible in customs volumes weeks before it was legible in any document, because there was no document. Phosphate is now the open half. Buyers should monitor monthly export volumes rather than watching for a notice, and should assume any change is already several weeks old by the time it is visible.

  3. Belarus, Russia and the EU

    Two sanctioning blocs diverged, and potash is the tell

    The United States eased sanctions on Belaruskali and the Belarusian Potash Company on 19 March 2026 S. EU sanctions remain fully in force and the Klaipeda rail route stays shut, so tonnes must route through Russian rail.

    The market largely shrugged, which is itself informative: the easing removed a tail risk rather than adding supply, and the routing constraint that actually governs cost is a European instrument that did not change.

    Read. This divergence is why potash is +2% while anhydrous is +21% S, and why potash is the nutrient a buyer can afford to be relaxed about this season. The risk to that view is European rather than American: an EU move on Klaipeda, in either direction, moves potash more than anything Washington does.

  4. Domestic farm politics

    An affordability squeeze inside an application window

    A ton of anhydrous now costs 197 bushels of corn at the USDA season-average price of $4.80, with a nitrogen-to-corn price ratio of 0.121 KR. An input cost shock that lands in the autumn application window, in front of a farm constituency, reliably generates a political response.

    The two available responses point in opposite directions. Direct support payments leave the duty structure intact and are broadly neutral for import pricing. Trade relief, whether a duty suspension or an exclusion, cuts directly against the Morocco and Russia cases above.

    Read. Watch which lever is pulled, not whether one is. Support payments are the higher-probability outcome and change nothing in this note. Any serious move toward phosphate duty relief would invalidate the trade-remedy catalysts above and is the political development most likely to force us to rewrite this edition.

07

Key debates

  1. Is this demand destruction or deferral?

    Fertiliser affordability is at multi-year lows, Illinois operator returns fall to $36 an acre on corn in 2027, and Brazilian deliveries are already down 2.2% through May.

    Our answer. Deferral, and the distinction is commercially important. Nitrogen demand is genuinely rate-sensitive and will fall with the price ratio. Phosphate and potash demand is being postponed against generally adequate soil test levels, which means the tonnage returns rather than disappearing. Plan inventory for a 2028 and 2029 restocking cycle, not for a structurally smaller market.

  2. Does Brazil cushion or amplify this?

    Brazilian deliveries fell 2.2% in the first five months of 2026, with imports down 2.7%, which reads as demand softening in the largest import market.

    Our answer. Amplifies. The number that matters is domestic Brazilian production down 17.1% S, against imports down only 2.7%. Brazil is becoming more import-dependent precisely as the seaborne market tightens. That is additional call on the same constrained tonnes, not relief from them.

  3. Will OCP expand into the shortage?

    The classic supply response to a price spike is capacity, and OCP has both the reserve base and the balance sheet.

    Our answer. Not on the evidence of its own capital allocation. First-half capex rose 6% but second-quarter capex fell 36% year on year, and revenue declined through the spike S. A producer conserving cash while its product prices rise is telling you it does not believe the margin is durable. We would not underwrite a 2028 phosphate supply response on announced capacity from this producer.

  4. Is European ammonia curtailment a one-winter problem?

    TTF is elevated on Middle East tension and low storage, both of which are arguably transient.

    Our answer. The gas spread is cyclical; the policy is not. Even if TTF halves, European producers face an escalating nitrogen import tariff through 2028 that removes the relief cheap imports would provide. The structural read is that a slice of European merchant ammonia capacity does not come back, and that the Atlantic basin is permanently shorter as a result.

08

What would change our mind

09

Watch list

DateEventWhy it matters
OngoingChinese phosphate export volumes, monthly customs data. The no-export consensus lapsed 31 August with no located successor; the parallel urea window has reopened at scale The highest-value unknown left in the market. There is no published instrument, so the change will be legible in volumes before any document
~9 Oct 2026USDA WASDE and Crop Production Sets the corn price that drives every affordability ratio in this note. The September report moved it from $4.50 to $4.80
Sep–Oct 2026India's decision on nutrient-based subsidy rates for the Rabi season India is the marginal global DAP buyer; the rate determines whether Indian importers can clear cargoes above $900
~Oct 2026Commerce final results, Moroccan phosphate sunset review, 240 days from the 2 March initiation Determines whether the 20.04% preliminary rate stands, and sets US delivered phosphate cost
Oct–Nov 2026Fall anhydrous application window The demand destruction test, against $0.58 per pound of N and $6.53 diesel
Winter 2026–27European gas storage stress TTF up 125% year on year into low storage is the trigger for a second wave of ammonia curtailment
2027ITC full five-year review determinations, Morocco and Russia Full rather than expedited reviews push the injury determination into 2027
~Mar 2027USDA Prospective Plantings First hard read on 2027 acreage and implied nutrient demand
Q3 27 – Q1 28Kazatomprom's TQZ sulphuric acid plant commissioning, delayed from Q1 2027 A reminder that uranium in-situ recovery and phosphate production compete for the same molecule
Independent research · Not investment advice · See important disclosures belowData cutoff 24 September 2026

Method and data sourcing

Sulphur production, trade and import-source data are from the USGS Mineral Commodity Summaries 2026 sulfur chapter, published February 2026. Countervailing duty rates and review schedules are from Federal Register notices as dated in the text. US grain balance sheets and the season-average corn price are from the USDA World Agricultural Supply and Demand Estimates of 11 September 2026. Henry Hub forecasts are from the EIA Short-Term Energy Outlook of 11 August 2026.

US retail fertiliser prices are the DTN weekly survey for the week of 14 to 18 September 2026. The diesel price is the EIA weekly on-highway series for the week ending 21 September 2026, which replaces the secondary figure we used at the prior cutoff. Illinois production cost figures and 2027 budget projections are from farmdoc daily at the University of Illinois, which restates the USDA AMS Illinois Production Cost Report. We were unable to retrieve the live AMS report or the September World Bank Pink Sheet at this cutoff and have said so rather than substituting stale figures.

Affordability ratios, the nitrogen-to-corn price ratio, the implied rate response, maintenance P and K cost per acre, and the gas cost per tonne of ammonia are Kastra Research calculations from the sources named alongside them. The 34 MMBtu per tonne ammonia conversion is an industry engineering convention rather than a sourced figure, and the euro to dollar assumption is ours.

Nutrient removal coefficients underlying the drawdown discussion follow the conventional Corn Belt agronomic values. We have not re-verified them at source in this cycle and have therefore expressed the maintenance cost as an approximation rather than a precise figure.

Strait of Hormuz status is drawn from contemporaneous reporting and is described as a sequence of events rather than a continuous state. We have deliberately not published a vessel transit count, a Russian export quota volume, or any crop protection price, because no reliable source for them was available at this cutoff.

Exhibits numbered 1, 2, 3 run in the body of the note; exhibits lettered A, B, C sit in the margin column. Each series is in the order it is meant to be read.

Notes

  1. 1Much trade coverage describes the Strait of Hormuz as closed continuously since 28 February 2026. Section 01 sets out the documented sequence: the start date holds, continuous closure does not.
  2. 2The sulphur move is commonly quoted as approximately $400 to $1,000–1,200 per ton. The $400 level is an end-2025 figure and predates the closure. The Hormuz-attributable move we use is stated in section 01.
  3. 3The TTF quotation is taken from a market data aggregator rather than an exchange feed. It is adequate for a spot reference and should be replaced with an ICE Endex settlement for any contractual purpose.
  4. 4OCP first-half 2026 results are sourced from a single news outlet. The underlying investor release has not been retrieved. This is our strongest finding on the weakest single source in the note and is tagged accordingly.
  5. 5China's phosphate and urea export arrangements were administered consensus and quota mechanisms rather than published regulations, so no expiry notice exists and none should be expected.

Principal sources

USGS, Mineral Commodity Summaries 2026, sulfur chapter. Federal Register notices on phosphate fertilizers from Morocco and the Russian Federation, 17 April, 17 June, 30 June, 24 July and 19 August 2026. US Energy Information Administration, Short-Term Energy Outlook, 11 August 2026. USDA, World Agricultural Supply and Demand Estimates, 12 August 2026, and AMS Grain Transportation Report bundle, 14 July 2026. farmdoc daily, University of Illinois, 11 August, 25 August and 1 September 2026. DTN retail fertiliser survey, 2 September 2026. World Bank commodity markets commentary, 14 May 2026. S&P Global Commodity Insights, 7 May 2026. ANDA Brazil delivery statistics via trade reporting, 10 August 2026. Reporting on OCP first-half 2026 results and product mix, June and September 2026, and on Chinese urea and sulphuric acid export policy, April to July 2026, attributed in text.

Abbreviations

Important disclosures

About this publication. Kastra Research is an independent research publisher. We are not a broker-dealer, investment adviser, or registered investment research firm. We do not underwrite, trade, or make markets in any security or commodity, we publish no ratings or price targets on securities, and we make no recommendation to buy, hold, or sell any security.

Analyst certification. The author certifies that the views expressed in this report accurately reflect their personal views about the subject commodities, markets, and entities, and that no part of their compensation was, is, or will be directly or indirectly related to the specific views or conclusions expressed.

Intended recipients. This report is prepared for institutional and professional recipients with the resources and expertise to make their own procurement and investment decisions. It is not directed at retail investors, is not a personal recommendation, and takes no account of the objectives, financial situation, or particular needs of any recipient.

No reliance. Information in this report is drawn from sources we believe to be reliable, but we make no representation or warranty, express or implied, as to its accuracy, completeness, or timeliness, and none should be inferred. The provenance tags exist precisely because reliability is not uniform across the figures we print. Nothing here is a substitute for a recipient's own verification, and no recipient should act on any figure, estimate, or conclusion in this report without independently confirming it against the underlying source, which we name so that they can.

Forward-looking statements. This report contains estimates and scenario analysis. Forecasts depend on the assumptions stated alongside them and actual outcomes will differ, potentially materially. Historical relationships are not a reliable indicator of future results.

Political and policy commentary. This report discusses legislation, regulation, litigation, sanctions and trade measures because they are material to the markets covered. That discussion is analysis of market consequence, not political advocacy, and it is neither an endorsement of nor opposition to any government, party, candidate, official, or policy position. Where we describe what a political actor is likely to do, we are forecasting behaviour in order to price risk. The forward-looking caveat above applies to those passages with additional force: political outcomes are less tractable than industrial ones, and our conviction is graded lower in that section for that reason.

Use of artificial intelligence. This report was produced with the assistance of artificial intelligence. AI tools were used in gathering and summarising source material, in drafting and editing the text, and in preparing the exhibits and the calculations behind them. AI systems misattribute sources, transpose figures, and state false things fluently, so the provenance discipline in this note is the control on that failure mode rather than a presentational device: every figure carrying a P or S tag was checked by the author against the named source, and figures the author could not confirm are tagged U or listed under "What we could not verify" rather than presented as established. The judgements, the conviction levels, and any errors are the author's. The use of AI does not qualify the analyst certification above and does not reduce the author's responsibility for anything in this document.

Conflicts. Kastra Research receives no compensation from any entity discussed in this report. No subject entity reviewed this report prior to publication, and we do not accept payment for coverage.

Jurisdiction and distribution. This report is published from the United States and is not directed at any person in any jurisdiction where its publication, distribution, or use would be contrary to law or regulation, or would subject Kastra Research to any registration or licensing requirement. It is not an offer to sell, or a solicitation of an offer to buy, any security, commodity, or instrument in any jurisdiction. Recipients are responsible for compliance with the laws applicable to them, and for observing the redistribution terms below.

Limitation of liability. To the fullest extent permitted by law, Kastra Research, its owners, and its contributors accept no liability for any loss or damage, whether direct, indirect, incidental, consequential, or otherwise, arising from any use of or reliance on this report or any part of it. Recipients use this report at their own risk.

© 2026 Kastra Research. Licensed to the named subscriber. Redistribution in whole or in part requires prior written consent. Quotation is permitted with attribution.

Kastra Research Document AG-2026-01 · v1.0 · Data cutoff 24 September 2026