Seven Propositions for the Next Eighteen Months

A field guide for operating inside a contracting system

Demand is surging. Supply is collapsing. The space between those two curves is the crisis, and it is structurally different from anything the sector has absorbed before, including 2011, 2014, and 2022.

Two shocks are now landing at once. The Iran conflict's economic transmission is moving through fertiliser, diesel, shipping, and insurance channels into the food systems of the most exposed economies, locking in caseload increases through late 2026 and into 2027. The humanitarian funding base is collapsing into that same window. Neither shock alone would produce the curve the sector is about to absorb. The two together do.

I have spent most of the last year writing about the funding collapse in instalments. Grandi's final briefing. Adapt Shrink or Die. The Fine Print. What If the Strait Stays Closed. Each piece tracked one edge of the picture. This one steps back and tries to name the whole of it, because the distance between what the sector says in public and what it is actually doing in private has now become wide enough to cause harm.

What follows is a set of propositions I think hold up. They are written for people who have to make operational decisions in the next six months and have no appetite for sector happy-talk.

Executive summary

• The Iran shock is already locked into food systems. Six-month lag for the main transmission channels, twelve to eighteen months for the deeper ones. Caseload effects land across food-import-dependent economies, including parts of the Sahel, the Horn, South and Southeast Asia, the Caribbean, and Latin America, through Q3 and Q4 2026, and into 2027.
• The same price shock hits depleted households harder. Coping capacity is the lowest it has been in fifteen years, which means the caseload curve will be steeper than transmission analysis alone suggests.
• The humanitarian architecture is contracting unevenly. What survives will be decided by where money concentrates, not by which reforms are formally adopted.
• Localisation is now being implemented in a form its original proponents would not recognise. Proximity to specific bilateral and regional funders is doing the selecting.
• 2026 and 2027 sector plans are underestimating need and overestimating response capacity. The gap shows up in numbers by Q2 2026 and in mortality and acute malnutrition data by Q4 2026. This is arithmetic, not forecast.

People will die

A significant number of people will die in the next eighteen months who would not have died if the donor system had held. This is locked in, regardless of what the sector does in the next six months. The remaining question is whether the death toll is the one current trajectories produce, or a smaller one produced by serious analytical and operational triage.

The rest of this piece is an attempt to describe why, and what can still be done.

The shock already moving through the system

The Iran conflict's secondary effects are doing what secondary effects do. They are working through fertiliser prices, diesel costs, shipping and insurance premiums on Gulf and Red Sea routes, and jet fuel constraints on air freight. They are translating into farm-gate and consumer prices in the most exposed economies. I set out the transmission channels in What If the Strait Stays Closed. Mercy Corps's Crisis Analysis team has now published The Long Reach of War, which documents these channels across five country cases and names the operational consequences with the specificity the sector needs. Read it. It is the clearest institutional statement of the transmission problem currently available.

What matters now is the lags. Six months for the main channels. Twelve to eighteen for the deeper ones. Q3 and Q4 2026 is where the caseload effects cluster. One finding from the Mercy Corps paper is worth pulling out here because it sharpens the argument. The FAO Chief Economist has set forty days of conflict disruption as the threshold at which farmer behavioural responses (reduced fertiliser application, crop switching, smaller planted areas) carry through to subsequent harvests. That threshold was crossed on 9 April, during the ceasefire window. The food security consequences of this war are already written into harvests that have not yet been planted. Ceasefire or no ceasefire, the caseload curve through late 2026 and into 2027 is locked in by agricultural decisions being made now.

What makes this shock different is the population it is hitting. Households across the Sahel, the Horn, South and Southeast Asia, the Caribbean, and parts of Latin America have already absorbed the 2022 shock, the 2023 donor contraction, the elevated price plateau of 2024 and 2025, and in several contexts sustained conflict-driven asset depletion. The buffer is gone. The same absolute price rise hits a resilient household and a depleted household very differently, and the populations we are talking about are now depleted. The caseload curve from this shock will be steeper than transmission channel analysis alone predicts, because it is landing on people with less resilience than at any point in the last fifteen years.

Food insecurity at this scale is not only a humanitarian outcome. It is a conflict driver, a displacement driver, a governance stressor that feeds back into further need. The loop is already visible in early indicators. It will be undeniable by mid-2026.

Seven propositions

Each of the seven below is a lens for decisions that need making now. They are propositions rather than predictions. I think they hold. Test them against what you are seeing.

One. Emergency systems outlast their emergencies, and contracting systems contract unevenly.

The cluster architecture. The pooled funds. The IASC structure. The Grand Bargain commitments. Each of these started as an emergency response and became permanent. Each accumulated staff, budgets, and institutional interests that resisted reform for a decade. The current fiscal collapse is now forcing contraction at a pace the sector chose not to manage voluntarily, and the contraction will not be uniform. Some structures will survive in diminished form. Others will disappear. The deciding variable will be where the money concentrates, not which reforms are formally adopted at the next high-level event.

Two. Fiscal collapse delegitimises faster than any intellectual critique.

Thirty years of critique on localisation, colonial structures, efficiency, and accountability produced no meaningful reform. What is producing reform now is not the critique. It is the collapse of the funding base. Which means the reform agenda is being shaped by the urgency of survival rather than by a coherent vision of what the sector should become. Settlements built under those conditions reflect short-term political necessity rather than long-term institutional logic, and they rarely last in their initial form. Watch what gets built in the next eighteen months. Most of it will be rebuilt within five years.

Three. Removing central constraints produces discretion, not liberty.

The collapse of the centralised donor architecture is being described in some quarters as an opening for localisation and direct funding to national actors. What it is actually producing is a system in which bilateral political relationships and proximity to specific donors determine who gets funded. National organisations with strong ties to particular donors or Gulf funders may do well. National organisations without those ties are being destroyed. Localisation is being implemented in a form its original proponents would not recognise, because the fiscal conditions have changed what the word means in practice. This is worth saying plainly, because the official line has not caught up with it yet.

Four. The gap will be filled, but not by the system that is contracting.

When centralised institutions fail to respond to acute need, other actors move in. National governments pursuing their own agendas. Regional powers pursuing their own interests. Private sector actors making their own calculations. Diaspora networks. Gulf state humanitarian machinery. Chinese and Turkish state-linked actors. Religious networks. In some contexts armed groups using aid as a tool of territorial control. Some of this is useful. Some is actively harmful to the populations it reaches. Most of it operates on different principles from the post-Cold War humanitarian system, with different accountability structures and different purposes.

The pattern is already visible in Sudan, where Sudanese mutual aid networks, diaspora funding, and selective Gulf engagement are carrying a disproportionate share of the response while the formal international system operates at a fraction of assessed need. Sudan is not the exception. It is the early template.

Five. Response capacity concentrates where political interest aligns, and that geography is not the geography of need.

Contexts in which Western donors retain specific political interests will keep disproportionate response capacity. Contexts in which political interest has faded will be triaged out, regardless of the scale of need. The official line will not say this. The funding flows already do. The implication is unavoidable. The geography of humanitarian need and the geography of humanitarian response will diverge further, and the gap will be widest in exactly the contexts now showing the steepest caseload increases.

Six. Evidence of outcomes moves resources that evidence of need no longer moves.

Traditional needs-based advocacy is losing traction. Donors do not believe the aggregate numbers translate into impact at the promised scale, and they are making political choices to disengage from the contexts the numbers describe. What still works is evidence of specific interventions achieving specific results at specific cost, and evidence of what happens when those interventions stop. Cost-per-outcome data on measles vaccination campaigns. Effectiveness evidence on cash transfers in acute food crises. Mortality displacement analysis when therapeutic feeding programmes are cut. This is the category of argument that still moves donor decisions in rooms where moral claims no longer do. It is empirical humanitarian economics, and the marginal value of investing in the evidence infrastructure that produces it is higher now than at any point in the sector's recent history. If you take one thing from this piece, take that.

Seven. The settlement that emerges will be smaller, harder, more bilateral, and more politically conditioned than the one being dismantled.

The propositions above describe pressures and patterns inside the current system. This one describes what comes out the other side. The system that replaces the current one will also, in some respects, be more honest, because the fictions the current system sustains (universal coverage, needs-based allocation, neutrality in contexts where no donor is neutral) will be harder to maintain. Whether that is a better system or a worse one depends on what you value. It will certainly be different. And the transition to it will be worse than either the system that preceded it or the one that eventually stabilises.

What this means operationally

Sector planning documents for 2026 and 2027 underestimate need by a significant margin and overestimate response capacity. The gap between those two will be visible in numbers by Q2 2026 and in measurable outcomes, meaning excess mortality and acute malnutrition data, by Q4 2026. This is not a forecast of something that might happen. It is the arithmetic of commitments already made and capacities already withdrawn.

Three tasks follow from this.

The analytical task. Make the gap visible early enough that donors, governments, and sector leadership can make informed triage decisions rather than sliding into triage by underfunding. Triage by design is morally harder. It produces better outcomes than triage by default, because it concentrates remaining resources where they have the most effect. The sector has historically resisted explicit triage, partly on principle and partly because explicit triage requires comparative effectiveness evidence the sector has not built. The current moment may force that reckoning.

The advocacy task. Contest the framing that the funding collapse is a rational response to reduced effectiveness. It is not. It is a political shift in donor countries that has been retroactively justified by effectiveness arguments. Pushing back on that framing with evidence of what humanitarian intervention actually achieves in specific contexts, and what happens when it stops, is the most useful advocacy work available now. Generic needs-based advocacy is losing. Specific outcomes-based advocacy is still winning some battles.

The operational task. For organisations still active in the worst-affected contexts, identify the sub-contexts where intervention is still feasible and effective, and concentrate effort there while producing the evidence that makes the case for doing so. This is not defeatism. It is acknowledging that the old assumption, that resources will approximately match assessed need even if imperfectly, has broken. Operating inside the new reality requires planning on two time horizons at once. The immediate caseload. And the institutional settlement that will handle caseloads in the decade after this one.

The durable question

The counterfactual that matters is not the pre-2022 system performing well. That system is no longer available. The counterfactual is a worse version of the current system performing even less well, which is what emerges by default. Marginal effort applied to triage, evidence, and concentrated operational presence has disproportionate effect against that counterfactual, because the default is bad enough that any deviation from it is meaningful.

The settlement that eventually stabilises the sector will not be designed by the people currently leading it. It will be shaped by a combination of donor governments setting harder terms, national governments asserting more control over humanitarian response within their territories, private sector and Gulf actors filling gaps on their own terms, and a new generation of humanitarian leaders who did not build the current system and have no loyalty to it. The work of the current generation, increasingly, is to preserve what is worth preserving through the transition, and to build the evidence base that allows the next settlement to rest on something firmer than political necessity.

The durable question, for anyone working in the sector now, is not which reform agenda wins the current arguments. It is which capacities, institutions, norms, and relationships survive the next five years. Those are what the next system will be built from. Directing effort toward the durable elements rather than the fashionable ones is the practical implication of everything above. What is durable is mostly not what is currently being discussed at high-level events.

That is where the work is. It is grim work. It is also the work where marginal effort has the largest marginal effect.

Over to you

I am putting this out as a framework, not a conclusion. I want pushback. Specifically:

• Which of the seven propositions holds least well in your context, and what's the evidence?
• Where are you already seeing the discretion-not-liberty pattern in practice, and which national organisations are being destroyed by it?
• If evidence of outcomes is the advocacy tool that still works, what is the single piece of evidence from your programming that you wish donors would actually look at?
• And the hardest one. What conversations are you having internally about triage priorities in the next six months, and what is shaping them?

Reply on the thread, or message me directly if it needs to stay off the record. I will work through the responses in the next instalment.

---

Aid and Dev Dispatches is a newsletter on the collapse and reconstitution of the humanitarian funding architecture, written for practitioners. Previous posts: Grandi's Final Briefing, Adapt Shrink or Die, The Fine Print, What If the Strait Stays Closed.