The Bell Tolls for Bangladesh

On the first of April this year, WFP food vouchers for more than a million Rohingya refugees in Cox's Bazar were quietly re-priced. 

The flat $12-a-person monthly entitlement that had defined survival in the camps was replaced by a tiered system: $12 a month for the "extremely food-insecure," $10 a month for the "highly food-insecure," $7 for everyone else, a dollar more on Bhasan Char. No press conference marked the change.

There didn't need to be one; camp residents had lived through this before, in 2023, when two ration cuts inside three months pushed acute malnutrition among children to the highest rate ever recorded there.

This time, aid officials told journalists plainly what the number meant: A diet below 2,100 calories a day, the floor line for emergency food assistance, breached again.

That single decision in Cox's Bazar is a local echo of something much larger and, until recently, almost unknown outside humanitarian and diplomatic circles: The "Humanitarian Reset."

It is worth understanding properly, because it will shape how much help Bangladesh gets -- for the Rohingya and for its own climate-battered communities -- for years to come.

What the Humanitarian Reset actually is

In March 2025, the UN's Emergency Relief Coordinator, Tom Fletcher, wrote to the heads of every major UN agency and international NGO grouped under the Inter-Agency Standing Committee (IASC) -- the body that runs the global humanitarian system -- announcing a process he called the Humanitarian Reset.

It was not a new funding mechanism or a pledging conference. It was, in effect, an admission: The system built after the 2016 World Humanitarian Summit could no longer sustain itself at its existing size, and rather than wait for it to collapse under funding shortfalls, its own leadership decided to shrink it deliberately.

The Reset's instructions to country teams worldwide were blunt. Humanitarian Coordinators were told to identify only the most severe, life-threatening needs -- what the system internally grades as "severity 4/5" -- and build response plans around those alone, discarding broader relief and recovery work.

Coordination structures, the sprawling architecture of "clusters" covering everything from shelter to child protection to gender-based violence, were to be merged and simplified; in several countries, the Protection Cluster and its three sub-groups on child protection, gender-based violence and mine action were folded into a single unit.

Fletcher ordered an "accelerated transition" out of eight country operations altogether. And humanitarian agencies were told, unusually candidly, to justify every dollar against a single test: Is this a more efficient use of the money than simply handing it to the affected family directly?

By June 2025, at the IASC's biannual meeting, Fletcher described this as a shift "from a system driven by the money we can raise to one based on greatest need" -- a system, he said, "significantly" smaller than before.

The language of efficiency and localization was genuine, and reformers inside the sector had pushed for exactly this kind of streamlining for years.

But the Reset did not emerge from a strategy session. It emerged from a cash crisis, principally the sudden, sweeping cuts to US foreign assistance in early 2025 that removed the system's single largest donor almost overnight, compounded by parallel retrenchment from several European governments.

The Reset is, in that sense, less a reform than a controlled demolition -- an attempt to manage a collapse in donor confidence before the collapse manages itself.

The Numbers Behind the Retreat

The scale of the pullback is visible in the UN's own annual planning documents. The 2023 Global Humanitarian Overview appealed for just over $56 billion.

By 2025, that ask had already fallen to roughly $47Bn for 190 million people across 72 countries -- and even that reduced target was met only partially: By the close of 2025, the appeal was just 27.85% funded, with total humanitarian funding for the year landing around $12B, the lowest in a decade.

The 2026 Global Humanitarian Overview, launched in December 2025, formalizes the retreat rather than reversing it. It estimates 239m people in need worldwide -- itself a number revised downward by methodology as much as by improving conditions, since needs on the ground have arguably worsened.

Of those, planners are targeting only 135m people, at a cost of $33Bn. And within that, a "hyper-prioritized" core tier -- the truly minimum, life-or-death response -- asks for just $23Bn to reach 87m people.

That is the lowest global humanitarian ask since 2017, even as the drivers of crisis -- conflict, climate shocks, political fragmentation -- are by the UN's own account intensifying.

Fletcher's own framing of the shortfall was pointed: the world spent an estimated $2.7 trillion on defense last year; the entire 2026 humanitarian appeal amounts to roughly one percent of that.

The consequence of "hyper-prioritization," aid officials now openly concede, is not efficiency but exclusion. Every year since 2023, more than 100m people categorized as being "in need" have simply fallen outside the boundaries of any funded response plan -- not because their situation improved, but because the plan's border was redrawn around them.

Yemen has become the sector's own cautionary example: The gap between people who need help and people the system will admit to counting keeps widening, prompting even senior humanitarian officials to warn that hyper-prioritization is "a symptom of underfunding, not a solution to it."

Bangladesh's Twin Exposure

Bangladesh sits at the intersection of two of the Reset's sharpest edges, and few countries are exposed on both fronts simultaneously the way it is. The Rohingya response.

The Cox's Bazar and Bhasan Char operations are entering their ninth year almost entirely dependent on international largesse that is now visibly draining away.

The 2025-26 Joint Response Plan, jointly issued by the UN and the Bangladesh government, was itself revised down by roughly a quarter from the previous year's ask before it even launched, reflecting donors' own signals about what they were prepared to fund. 

Programs for the Rohingya were only around half-funded through 2025; barely a fifth funded so far in 2026.

The April ration re-pricing was the visible tip of that funding line -- but the deeper damage has been happening beneath the surface for two years: School closures tied to funding gaps have been linked to rises in child marriage, child labour and kidnapping in the camps; malnutrition treatment admissions spiked sharply after the 2023 cuts; and by one WFP survey that year, 90% of camp households could not afford an adequate diet.

Layered onto this is a fresh complication the Reset's planners did not anticipate when they drew up their 2025 numbers: Renewed fighting in Rakhine State has pushed an estimated 100,000 additional Rohingya across the border since, stretching an already-shrinking pool of resources over a growing population.

The domestic disaster-risk picture. Bangladesh's own climate exposure is not shrinking to match the funding available to address it.

Germanwatch's Climate Risk Index, released at COP30 in November 2025, placed Bangladesh 13th globally on its long-term risk ranking spanning three decades of extreme weather -- a period in which more than 9,700 recorded events killed over 830,000 people worldwide and caused an estimated $4.5T in economic losses.

Bangladesh's own domestic toll, by Germanwatch's separate annual accounting, runs to roughly $3B in economic losses and 6.3M people affected in an average year, driven by floods, cyclones, drought and -- increasingly -- heatwaves; the 2024 heatwave alone, which pushed temperatures to 43.8°C, affected more than 33M people, placing Bangladesh among the countries with the highest number of heat-affected citizens anywhere in the world that year.

The genuinely good news in that data is real and worth stating plainly, because Bangladesh rarely gets credited for it internationally: Cyclone-related mortality has fallen more than a hundred-fold since 1970, from roughly half a million deaths in a single storm to just over 4,000 in the far larger Cyclone Sidr of 2007, thanks to early-warning systems and cyclone shelters that are now studied as a global model.

But disaster-risk reduction of that calibre is expensive to sustain and expand, and it was built substantially with international humanitarian and development finance that is now the very category being squeezed hardest by the Reset.

The Localization Promise, and Its Non-delivery

If there is one part of the Humanitarian Reset that ought to work directly in Bangladesh's favour, it is the renewed emphasis on "localization" -- shifting money and decision-making power toward local and national organizations, who are cheaper to run, faster to mobilize, and permanently present after international agencies move on. 

The trouble is that this promise is a decade old and consistently broken. Donors and agencies committed at the 2016 World Humanitarian Summit's "Grand Bargain" to channel at least 25% of humanitarian funding directly to local and national actors.

Development Initiatives found that just 0.6% of funding from Grand Bargain signatories went directly to local and national actors in 2023, while independently verifiable direct and indirect funding amounted to 4.4%.

The substantial gap between these figures and the 25% commitment highlights the continuing challenges of meaningful localization and transparent tracking of funding flows. Bangladesh's own experience mirrors the global pattern almost exactly: At a Dhaka seminar in August 2025 assessing progress on "Grand Bargain 3.0," speakers noted that only about 6% of international humanitarian funds channeled through Bangladesh actually reach Bangladeshi organizations, with several participants accusing international NGOs of registering local branch offices specifically to satisfy localization reporting requirements without shifting any real decision-making power.

There are honourable exceptions -- the Start Fund Bangladesh model has been repeatedly cited internationally as a rare example of locally-led disbursement working as intended, and OCHA's own country-based pooled funds report raising the local-NGO share of their disbursements from under a quarter to over a third between 2016 and 2021 -- but these remain islands in an otherwise unmoved system.

This matters enormously for what the Reset claims to want to achieve. A humanitarian system that says it is retreating to "greatest need" and "efficiency" while continuing to route the overwhelming majority of a shrinking pot through expensive multilateral and international-NGO layers is not actually optimizing for either goal.

In 2023, the last year for which detailed tracking exists, 58 percent of all Grand Bargain donor funding went directly to multilateral institutions and another 23% to international NGOs -- leaving a sliver for the organizations that are, by the Reset's own stated logic, best positioned to deliver cheaply and close to the ground.

What This Means for the Years Ahead

Three things follow from this data, and Bangladesh's policymakers, NGOs and diplomats would do well to plan around all three rather than treat the Reset as a temporary funding dip that will self-correct.

First, the era of a predictable, roughly flat Rohingya response budget is over. The 26% downward revision built into the current Joint Response Plan is not a one-off correction; it reflects a structural repricing of how much the international system is willing to commit to a protracted, unresolved refugee crisis with no repatriation pathway in sight.

Bangladesh needs contingency planning -- on nutrition surveillance, on camp security, on education continuity -- built around the assumption that funding will remain at or below current depressed levels rather than the assumption that a crisis moment will attract a rescue package, as happened intermittently in 2023 and 2024.

Second, domestic disaster financing cannot continue to lean as heavily on international humanitarian appeals as it has historically.

Bangladesh's own record on anticipatory action and early-warning investment is genuinely strong, and it is precisely that kind of pre-positioned, domestically-owned system -- rather than post-disaster international appeals -- that the shrinking global pot will reward going forward.

Expanding domestic disaster-risk financing instruments, sovereign risk pools, and the national Disaster Management Fund, alongside continued advocacy for the climate finance (as opposed to humanitarian finance) that Bangladesh is owed as a frontline state, is now less a matter of good practice and more a matter of necessity.

Third, and most immediately actionable: Bangladeshi civil society and the government's own coordination bodies (the NGO Affairs Bureau, RRRC, and the Ministry of Disaster Management) have real leverage to press the localization argument at the exact moment the international system is publicly committed to it in principle.

The Reset's own architects have staked their credibility on efficiency and closeness to communities as the justification for shrinking the system.

Holding donors and UN agencies to the 25 percent commitment they signed nearly a decade ago -- with hard, published, disaggregated data on how much actually reaches Bangladeshi organizations, not intermediary claims -- is a fight winnable on the Reset's own stated terms, not against them.

A System Reorganizing Itself Around Less

None of this is cause for fatalism. Bangladesh has, in genuine and internationally recognized ways, out-performed its resource base before -- the collapse in cyclone mortality since 1970 is proof that domestically-driven, well-designed systems can save lives even when the money behind them is thin. 

But the Humanitarian Reset is not a passing storm to be weathered until the old funding levels return. It is the international system's own leadership telling the world, in writing, that the old levels are not coming back, and that what remains will go to whoever can prove it delivers the most for the least, closest to the people who need it.

For a country carrying both the largest protracted refugee population Asia has produced this century and one of the world's most climate-exposed populations, that is either an existential funding threat or the strongest argument Bangladesh has ever had for demanding the resources -- and the decision-making power -- be routed to where they were always supposed to go.

Which of those two futures materializes will depend less on what happens in Geneva than on what Bangladesh does with the argument between now and the next Joint Response Plan.

Inmanuel Chayan Biswas, a humanitarian and development practitioner, is working for a prominent NGO in Bangladesh.