Why this session matters to us
Read as a crisis briefing, this FAO side event was about the Strait of Hormuz, an incoming El Niño and a fertilizer squeeze. Read the way SDGCounting reads a room, it was a tour of the instruments the UN uses to see food risk: the hunger headline in the SOFI report, the hidden-cost accounting in SOFA, the FAO Risk Monitor, the Agricultural Market Information System, and WMO’s seasonal forecasts. What made the hour unusual was how candid the panel was that several of those instruments are either missing for the inputs that now matter most, or running on funding that may not last.
The policy playbook was broadly agreed and repeated by everyone: anticipatory finance before a crisis, social protection for the most vulnerable, no export bans, no blanket subsidies on energy or fertilizer. The harder admission was upstream of policy. The world can model the shock and forecast the weather. It still cannot see fertilizer markets the way it sees oil, and the observing systems behind the forecasts are fraying. The gap this session revealed is a measurement gap.
The just-in-time world is over
FAO’s chief economist set the scene with numbers he flagged as provisional, ahead of the State of Food Security and Nutrition in the World report due to launch on 21 July in Rome. On those pre-release figures, 673 million people were hungry in 2024, about 8.7 percent of the world, and more than 266 million people faced acute food insecurity. The largest driver was conflict, at 147 million people, followed by weather and climate at 87 million, then economic shocks. WFP put the same headline at 266 million people across 47 countries. The point both agencies pressed is that the world produces enough food to feed everyone; the failure is one of access and affordability, not supply.
He also cited FAO’s attempt to price what the sector costs as well as what it delivers: the hidden costs of agrifood systems, around 11.6 trillion US dollars, roughly 10 percent of global GDP, with the largest share tied to health and the cost of healthy diets. That number is itself a measurement product, an effort to put on the books externalities that GDP and food-price statistics never capture.
His framing for the moment was that comparative advantage is giving way to geopolitics, and supply chains with it.
FAO’s modelling team, led by David Laborde, is running that logic across three classes of threat: biological (his example was soybean Asian rust reaching a genetically modified crop), nuclear (land taken out of production by contamination), and logistical (Hormuz now, the Red Sea and Suez, the drought that throttled the Panama Canal). The resilience answers he named are as much about data and genetic infrastructure as about trade: the CGIAR gene banks, including CIMMYT’s maize and wheat collections, as insurance against a disease wiping out a variety.
Oil has a dashboard and a stockpile. Fertilizer has neither.
The sharpest measurement point of the session was an analogy. The disruption at the Strait of Hormuz, on FAO’s reading, removes roughly a quarter of the key inputs to the agrifood system at once, a supply shock that cascades into fertilizer, energy and seed prices and into irrigation, since around 70 percent of pumping runs on diesel. When oil does this, the world has a response: the International Energy Agency can release strategic stocks into a price spike, because oil is measured, monitored and stockpiled. For fertilizer, none of that exists.
Two of the panel’s concrete initiatives are, in effect, attempts to build that missing infrastructure. FAO is extending the Agricultural Market Information System, the G20-era transparency platform for grain markets, to cover fertilizers, a move Italy’s development-cooperation official singled out for praise. And Italy, with Croatia, has launched the Rome Coalition on fertilizers, a roughly 40-country grouping convened in May 2026 to keep supply flowing. Behind both sits FAO’s Risk Monitor, the early-warning layer that tries to score where the next input or logistics shock is most probable. The through line is simple and it is ours: you can only stabilise what you can first see, and for the input driving this particular crisis the world is only now learning to look.
The panel from Malawi grounded the stakes. The country runs a fertilizer-heavy farm subsidy programme; soaring input prices land directly on its budget and on the vulnerable households the programme reaches, with an El Niño forecast on top. Its response, described as a transformation agenda rather than a relief one, includes a national commodity exchange to improve price transparency and cut post-harvest losses, an example of a country building its own market data where none existed.
The forecasts are sharper. The instruments behind them are fraying.
WMO’s New York head brought the one piece of genuinely improving measurement, and then its vulnerability. El Niño, she said, has been confirmed and is forecast to strengthen into a strong event peaking around the turn of the year, with unusually high agreement between the world’s forecasting centres. But the baseline has shifted: the last major El Niño was more than a decade ago, and every one of the eleven years since has been among the hottest on record, so past events are no longer a reliable guide to this one. The value of the forecast is the lead time it buys governments, farmers and markets to act before impacts land.
The catch is that the forecast depends on an observing system that is quietly eroding. The ocean network, where El Niño originates, is largely funded through individual research projects rather than any coordinated commitment, and the single largest funder has signalled it is pulling back. The land network is unevenly thin.
She was equally pointed about the technology everyone reaches for. AI can democratise forecasting, putting tailored predictions in the hands of countries that could never afford a supercomputer, “but AI is also only as good as the observations that it learns from, and so better and trusted data remains the foundation.” WMO’s fixes are all about paying for the data layer: the Systematic Observations Financing Facility, the Global Ocean Observing System co-sponsored with UNESCO, and an emerging ocean alliance to broaden who funds the network. For a site about how the UN counts, this was the clearest case of the session: the quality of every downstream decision is capped by an upstream data system that no one is reliably paying to keep alive.
What even counts as “rural”
The most useful challenge came from the floor. A UN-Habitat representative pointed out that the entire conversation was happening at a level that the SDG framework barely reaches: there is nothing subnational in SDG 2, no urban or peri-urban dimension to how zero hunger is measured, even though local governments sit on the front line of every shock the panel described. Torero’s reply went further, and it cut to the heart of comparability.
Rurality, he noted, is defined country by country: what counts as rural in India is not what counts as rural in China, so the category does not travel across borders. FAO’s own work has moved to treating it as an urban-rural continuum rather than a binary, precisely because value chains, energy and labour cross the line constantly. It is an unglamorous point with real consequences: an indicator you cannot compare between countries is a weak basis for a global goal.
Preparedness is a measurement problem too
WFP’s contribution translated the same logic into money. Its case for anticipatory action, acting on a forecast before a crisis is declared, rests on a single number it wants everyone to remember.
That ratio only exists because someone measured it, and it only triggers if the early-warning and risk-finance systems are wired to release money on a threshold rather than on a headline. WFP reported scaling pre-arranged and disaster-risk finance to millions of people across dozens of countries, and, like every speaker, closed on the theme that no single institution can hold the whole picture. IFAD, in the closing, made the structural version of the argument: today’s emergencies are “symptoms of deeper structural vulnerabilities,” and the fix is to finance resilience earlier and more predictably, at what it calls the first mile, rather than pay more for relief after the fact.
Our read
A room holding much of the UN’s food-risk machinery spent 75 minutes in near-total agreement on what to do, and was unusually honest that the binding constraint sits somewhere else. The world can now model a biological or logistical shock, price the hidden costs of the food system, and forecast an El Niño with high confidence. What it cannot yet do is see the fertilizer market the way it sees oil, keep the observing systems behind its forecasts funded, or compare “rural” across two countries. Each of those is a counting problem, and each is what turns the next foreseeable shock into a surprise. The encouraging part is that the response is institutional and already under way: AMIS is being extended to fertilizers, WMO is chasing sustainable finance for the observing system, FAO is treating rural and urban as a continuum. Watch whether any of it is funded before the crop calendar, which no one can renegotiate, forces the issue.
Why it matters for the SDGs
The session sits on SDG 2 (zero hunger), but its lesson is about how SDG 2 is measured: against inputs it does not count, such as fertilizer and energy, and through a rural-urban category that does not compare between countries. It reaches SDG 13 (climate) through the El Niño forecast and the observing systems that make it possible, and SDG 6 and SDG 7 because water stress and diesel-run irrigation are the channels through which an energy shock becomes a food shock. Above all it lands on SDG 17, the means of implementation: the recurring plea of the afternoon was not for a new metric but for someone to fund the data, the early-warning systems and the partnerships that make any of the metrics actionable.
Watch & read
- FAO at HLPF 2026, the FAO Liaison Office programme page listing this side event (14 July, Conference Room 6).
- The State of Food Security and Nutrition in the World (SOFI), the flagship whose 2026 edition launches 21 July in Rome and updates the hunger figures previewed here.
- The State of Food and Agriculture (SOFA), whose true-cost editions produced the hidden-cost estimate for agrifood systems.
- Agricultural Market Information System (AMIS) and the FAO Risk Monitor, the market-transparency and early-warning tools discussed.
- The Rome Coalition on fertilizers, launched by Italy and Croatia (May 2026), and WMO’s El Niño / La Niña updates.
- Related HLPF 2026 coverage: SDG 7 (energy), SDG 6 (water), the UN-Habitat SDG 11 briefing on the urban-rural question, and Beyond GDP on measuring what GDP misses · Full HLPF 2026 coverage.
Quotations are lightly edited from an automated (Otter.ai) transcript of the UN Web TV recording and should be read as close paraphrase. Speakers whose names could not be confirmed against an official source are cited by role. Figures are as speakers and their agencies reported them and were not independently verified; the hunger numbers are pre-release previews that the SOFI report updates on 21 July.