WFP projects acute food insecurity across 45 vulnerable countries rising to 274 million people by the end of 2027. FAO forecasts world cereal stocks rising, with a stock-to-use ratio of 32 percent. Both are true at once, and understanding why is the whole point. No single stress now moving explains what is coming. The concern is that several of them are weakening different buffers at the same time.
On 24 August, Reuters reported Volodymyr Zelenskiy saying Russia was "not ready" for a ceasefire on attacks against Black Sea ships carrying agricultural produce, because Moscow wants any deal extended to cover Ukrainian strikes on Russian refineries and pipelines. In his words, "for [Russia], it’s not about the grain. It’s exclusively about their energy sector." That is one belligerent describing another’s motives and should be read as such, but the shape of the impasse is clear: grain shipping and energy infrastructure are now the same negotiation.
Behind it sits a fortnight of damage. S&P Global reported two Novorossiysk terminals suspending operations after an overnight attack on 12 August: NKHP, which shipped 6.23 million tonnes in the 2025-26 marketing year, and NZT, with 8.5 million tonnes of annual capacity. A third, KSK, kept working slowly. Strikes on shipping followed around 17 and 18 August. One Russian wheat seller told S&P: "This means that Russian South is almost fully blocked."
That is a supply-chain event with dates attached. It is also the fifth item on a list assembling since March, and the list matters more than any entry on it.
Those three numbers do not contradict each other. Needs are projected to rise, the aggregate cereal balance is adequate and improving, and the funding that converts available supply into food on plates has more than halved. That is the story.
El Niño. NOAA CPC holds an El Niño Advisory and, on 13 August, gave greater than 90 percent odds of a very strong event in Northern Hemisphere fall and winter, with 69 percent odds that October to December exceeds +2.5 C on three-month RONI. WFP expects the peak between September and December. What exists today is a forecast of shifted odds, not a documented crop loss, a distinction this site learned the hard way when it retracted drought percentages derived from an index rather than observations.
Diesel, which is not crude oil. US on-highway diesel averaged $5.454 a gallon in the week of 17 August 2026, $1.741 above a year earlier, up about 47 percent. Gasoline rose about 30 percent. The gap is the point. Hormuz removed heavier Middle Eastern grades that yield proportionally more diesel and jet fuel, and constrained Gulf refining and export routes, so the acute pressure sits in refined products and suitable crude grades rather than in total crude availability. EIA’s cost breakdown shows the same asymmetry: in May 2026 refining was 25 percent of the US retail diesel price against 22 percent for gasoline, while crude was 42 percent of diesel against 52 percent of gasoline. Calling this a global crude oil crisis would misdescribe it.
Fertilizer, which is really natural gas and shipping. Urea rose above $850 per tonne in April 2026, up 80 percent since February and the highest since April 2022, on World Bank data. The Middle East supplies nearly a quarter of global urea exports, and Iranian, Qatari and Indian output all fell. Phosphate followed, with sulfur prices doubling since January. Potash did not, and those markets remain comfortably supplied.
Black Sea export capacity. Disruption was in the price data before the August strikes. FAO’s index for July shows wheat up 5.8 percent on the month and 9.9 percent on the year, attributed to "heightened concerns over continued disruptions to Black Sea export flows and damage to export infrastructure," compounded by heat effects elsewhere. That reading closed before 12 August, so it cannot capture the terminal strikes, and the August effect is not yet in any monthly index. Early signals are mixed: S&P’s Platts Milling Wheat Marker was assessed at $222 a tonne on 11 August, down 2.84 percent on the day, traders citing execution risk. When buyers doubt a cargo will load, that can push a benchmark down as easily as up.
Humanitarian capacity. Funding for food assistance, emergency agricultural assistance and nutrition in food crises fell an estimated 59 percent between 2022 and 2025, returning to levels last seen nearly a decade ago.
They operate through different channels, and conflating them produces bad analysis.
Diesel enters as a cost across the chain: tractors, diesel-powered irrigation, harvesters, trucks and parts of the shipping system. It first appears as a higher cost of producing and moving food. If the shortage persists, reduced irrigation, field operations and distribution can eventually affect physical supply as well. In fuel-importing countries it can make groundwater irrigation uneconomic exactly when a failed monsoon makes it necessary, which is the sharpest El Niño-fuel link and the subject of an earlier piece here.
Fertilizer works differently, entering through application and planting decisions. It can affect a crop already in the ground when farmers reduce later applications, or the next crop through lower planting, reduced application and crop switching. Timing depends on the crop calendar and on when inputs were secured. FAO’s chief economist noted in March that fertilizer has a nonlinear yield response, so modest cuts can produce disproportionate declines where baseline application is already low.
El Niño enters through production in exposed regions, unevenly. FAO offers a concrete example rather than a hypothetical: the global wheat forecast was cut to 806.5 million tonnes, down 4.3 percent year on year, driven mainly by Australia, where "an increased likelihood of El Niño-induced below-average rainfall and elevated input costs are foreseen to curtail both plantings and yields." Two of the five stresses, named together, in one line of a supply forecast.
Black Sea disruption enters first through reduced reliability, delayed or cancelled cargoes, higher insurance and potentially lower export tonnage. Importers with thin foreign-exchange reserves feel those costs first. Conflict and currency weakness enter last, and are decisive. They determine whether a higher world price becomes a household that eats less.
These stresses are often described as arriving together. They do not.
| Window | What moves |
|---|---|
| Now | Diesel, freight rates, war-risk insurance, port and terminal availability |
| Next planting cycle | Fertilizer affordability and application rates, crop switching |
| Harvest | El Niño yield effects, where they materialise |
| Late 2026 into 2027 | Household food access, humanitarian caseloads and funding requirements |
The staggering cuts both ways. No single month delivers a synchronised shock, which is real resilience. But each stress lands while the last is still working through, so buffers may not be replenished between them.
On 5 August WFP published early analysis of 45 already food-insecure countries where El Niño is expected to significantly influence rainfall, temperatures and the frequency of floods and droughts. Across them, acutely food insecure people would rise from approximately 225 million to 274 million by end-2027, an increase of at least 49 million, about 22 percent. Central America shows the largest proportional rise at 83.1 percent, with more than 18 million additional people in East and Southern Africa, 8.2 million in Asia-Pacific and nearly 6 million in West and Central Africa.
Two points, both from WFP's own note to editors. First, it measures IPC Phase 3 or equivalent and above, which is Crisis-level acute food insecurity. It is not a famine projection and does not claim to be.
Second, and more often missed: "This analysis focuses on countries where El Niño is expected to deepen food insecurity, rather than on regions where its effects may be limited or beneficial." The 274 million is deliberately a selected-sample estimate, not a global net figure, and treating it as one overstates it. WFP calls this initial modelling built on historical evidence, baselines, risk factors and expert judgement: a reasonable method, not a measurement.
The FAO-WFP Hunger Hotspots report of 17 June identifies 13 hotspots for June to November 2026, with around 266 million people already facing high levels of acute food insecurity across them. Sudan, South Sudan, Yemen and Palestine remain of highest concern, joined by Nigeria and Somalia.
The most important line in that report, for anyone attributing this to El Niño, is this: armed conflict and violence remain the primary drivers of acute food insecurity, affecting 12 of the 13 hotspots. El Niño appears in the same paragraph as a compounding risk alongside economic shocks and funding shortfalls, not as the lead.
That is the exposure map. It is not where crops fail, but where crop failure cannot be absorbed.
Famine is a technical classification, not an intensifier. Under IPC an area requires extreme critical levels of acute malnutrition and mortality alongside extreme food deprivation. Risk of Famine, a warning that famine could occur, has been formally identified in 14 areas of Sudan through September 2026, four counties of South Sudan through July, and one Somali district. Those risk findings already exist, and they are not the same as an IPC finding that famine is occurring. Conflict is the dominant driver across the hotspot report, although individual crises also reflect drought, economic disruption, low crop production and restricted humanitarian access: Somalia’s deterioration is attributed to multiple years of drought and record-low crop production alongside conflict. They should not be presented as consequences of the developing El Niño.
Nothing in the current evidence supports a forecast that El Niño will produce famine. What it supports is narrower and still serious: a strong El Niño is arriving in a system whose shock absorbers are thinner than usual.
A chain is only as strong as its weakest link, which cuts both ways. There are real buffers here, and several are already working.
The aggregate cereal balance is adequate. FAO forecasts 2026 world cereal production at 2,983 million tonnes, the second largest on record despite being 1.9 percent below 2025. Stocks at the close of 2027 are forecast at 957.8 million tonnes, up 0.9 percent, stock-to-use at 32.0 percent. Rice stocks are the second highest ever. FAO’s aggregate cereal balance does not show a worldwide shortage, although global stocks do not guarantee affordable or geographically accessible supplies, and a large share of world inventories sits in countries that do not export them.
Aggregate food prices are not in crisis. The FAO Food Price Index averaged 131.1 points in July 2026, up 1.0 percent year on year and 18.2 percent below its March 2022 peak. Dairy sits 24.8 percent below a year earlier and meat fell in July, its first decline of 2026. The stress is in wheat, sugar and vegetable oils, not across the basket.
The fertilizer shock has been partly absorbed. The World Bank attributes the subdued response, against 2021 and 2022, to three things: Northern Hemisphere growers had already secured much of their supply, gas prices rose less sharply than after the 2022 invasion of Ukraine, and Middle East flows are increasingly rerouted overland, bypassing Hormuz.
Regional variation is real. FAO raised maize forecasts on stronger harvests in Argentina, Brazil, China and Zambia. Barley prices fell on favourable prospects in Australia and the Black Sea. Rice held steady. An El Niño does not damage every agricultural region, and assuming otherwise is the most common error in this genre.
Anticipatory action has a track record. WFP has triggered plans in six countries since May, reaching half a million people with more than $14 million, and cites evidence from prior events that $1 spent this way saves between $3 and $7 later.
Against those: the funding decline is the buffer with no obvious substitute. Alternative shipping routes, fertilizer suppliers and grain exporters all exist. There is no alternative source of humanitarian financing at the scale that has been withdrawn.
The testable version. The claim is that no single stress explains the projected deterioration, and that the binding constraint is absorptive capacity rather than global food availability.
What would support it: acute food insecurity rises in the WFP-identified countries while the FAO cereal stock-to-use ratio stays near 32 percent, meaning need grows without a physical shortage. Watch the FAO Cereal Supply and Demand Brief on 4 September 2026 and the Food Price Index the same week.
What would weaken the access-first interpretation: a sustained, material decline in global cereal production, exporter inventories and export availability, accompanied by a falling stock-to-use ratio. That would add a genuine aggregate supply problem to the existing access crisis. Note that this would supplement the access argument rather than replace it, since a crisis can involve both.
What would weaken the El Niño contribution: repeated post-harvest assessments showing no unusual production or food-security deterioration in the rain-fed regions WFP identified, after allowing for local crop calendars and separating weather effects from conflict and prices. WFP’s projection runs to the end of 2027, so no single forthcoming IPC update can settle this.
Where it gets settled: FAO CSDB and FFPI monthly, IPC and Cadre Harmonisé country analyses, the next Hunger Hotspots edition, and CPC's ENSO discussion on 10 September 2026.
The useful posture is neither alarm nor dismissal. FAO’s aggregate cereal balance does not show a worldwide shortage, though global stocks do not guarantee affordable or accessible supplies. A strong El Niño is arriving. Several expensive frictions now sit between the grain and the people who need it, and the humanitarian system responsible for food assistance, emergency agriculture and nutrition received 59 percent less funding in 2025 than in 2022. All of those are true at once, and any account that drops one is describing something simpler than what is happening.
Method and data vintage. WFP figures are from its news release of 5 August 2026 and its accompanying note to editors. Cereal supply, stock and production figures are from the FAO Cereal Supply and Demand Brief released 3 July 2026; there was no August edition and the next is due 4 September. Price index figures are from the FAO Food Price Index released 7 August 2026, covering July. Hotspot and funding figures are from the FAO-WFP Hunger Hotspots report of 17 June 2026. Diesel and gasoline prices are US on-highway retail averages from the EIA release of 18 August 2026 for the week ending 17 August; percentage changes are calculated from EIA's published dollar changes and are US retail figures, not global. Fertilizer figures are from the World Bank Data Blog of 14 May 2026, drawing on the April 2026 Commodity Markets Outlook, and describe benchmark prices rather than fertilizer purchased or applied. ENSO figures are from NOAA CPC's discussion of 13 August 2026. Accounts of the August Black Sea strikes rest on trade and news reporting rather than official damage assessments: terminal names, throughput figures and the Platts wheat assessment are from S&P Global Commodity Insights dated 12 August 2026, and the ceasefire account is Reuters dated 24 August 2026. Zelenskiy’s description of Russia’s negotiating position is one belligerent characterising another and is attributed rather than asserted. Figures issued by parties to the conflict are not used. The World Bank blog states that it draws on the April 2026 Commodity Markets Outlook.