Update: GHO Coverage Jumped +16pp to 40.4% — Displacement Still Carried at 117.8M

What changed since the May update

Our May Global Trends / Mid-Year Review update closed on a scissors that finally bent on the people side: forced displacement down to 117.8 million at end-2025, returns up 50%, resettlement halved, LMIC host share at 68%, and GHO appeal coverage stuck near 24.4% even after the ask was stripped back to $33.66B. That post answered what moved once the first decade-long stock decline landed. This 2026Q3 refresh asks the next vintage question markets and desks actually trade on a summer cadence: what changed once late-summer FTS and UNHCR’s Global Appeal 2026 prints replaced the 31 May Mid-Year Review cash ledger — and did the funding side finally heal while people meters stayed frozen?

The cash answer is loud. OCHA’s GHO 2026 monitoring snapshot now shows requirements at $34.87B and funding at $14.08B, for coverage of 40.4% — a +16.0 percentage-point jump from the May Mid-Year Review’s 24.4%. Funded inflows rose +$5.87B; the unfunded gap narrowed from $25.45B to $20.79B. The people answer is quieter by design: displacement stocks, host shares, and GHO people-in-need remain carried from Global Trends 2025 / the May MYR until UNHCR’s Mid-Year Trends (typically October/November). Meanwhile UNHCR’s own 2026 needs budget was deliberately cut to $8.505B (−19.8% versus the 2025 Global Report’s $10.604B), with early pledges covering only about 18% of that smaller ask, and mid-year operational messaging still flagging roughly 11.6 million people at risk of losing assistance from prior funding cuts.

The dashboard above is built as a May→August vintage delta: dumbbell meters, a coverage waterfall, a dual stock×cash path, diverging “who bears it” bars, an UNHCR budget scissors panel, a host-income scatter, and an editorial crisis-plan coverage scatter. Use the View, Burden lane, and Sort controls to isolate cash, hosts, donors, agency cuts, or plan-level geometry.

The headline table: May Mid-Year Review → August FTS

The story is not “displacement reversed again.” The stock is carried. The lever that moved is the cash ledger on a stripped-back appeal — and even that healing leaves a ~$21B hole while agency budgets were already rewritten downward.

Coverage healed — the ask still crept up

Toggle View → Cash ledger. The coverage waterfall decomposes the +16pp into a large funded-inflow contribution and a small negative from ask creep. Requirements rose from $33.66B to $34.87B even as funded dollars nearly doubled from the May print. That matters for narrative discipline: coverage can rise because donors write cheques or because the ask shrinks. Here donors wrote cheques and the ask edged higher — so the coverage gain is the harder kind.

The GHO cash-path panel puts requirements, funded, and gap on one chart. Funded climbs from $8.21B to $14.08B; the gap falls from $25.45B to $20.79B but does not close. Pair this with our prior update: the May vintage already showed that shrinking the ask (versus 2025’s $45B-class print) does not by itself heal coverage. August shows the complementary truth — inflows can heal coverage without healing the people stock, and without erasing a twenty-billion-dollar shortfall on a deliberately prioritized appeal.

People meters are carried — and that is the point

The dual stock×cash panel is the visual punchline. Forced displacement sits flat at 117.8M across the May and August vintage points while coverage jumps from 24.4% to 40.4%. That is not a data failure; it is calendar hygiene. UNHCR’s Global Trends 2025 remains the last official year-end stock. Mid-Year Trends for H1 2026 are not yet the street print as of this Q3 note. GHO people-in-need (252.1M) and people targeted (143.2M) are likewise carried from the May Mid-Year Review rather than invented from an incomplete FTS people tab.

Carried meters are still analytical content. They tell you which side of the scissors moved this quarter: cash, not census. Desks that equate “coverage up” with “crisis resolved” are reading the wrong axis. The people axis is waiting for the autumn Mid-Year Trends print; until then, the honest headline is a funding vintage delta sitting on top of a frozen stock.

Who still bears the costs

Toggle View → Who bears it and filter Burden lane. Host shares are carried from Global Trends 2025: LMIC hosts still at 68%, LDCs at 26%, high-income hosts at 29%. Those percentages did not improve because coverage improved. Hosting is a people geography problem; FTS coverage is a donor cash problem. The diverging Δ bars make that asymmetry explicit: donors show the large positive move; host lanes sit near zero because they are carried; the agency lane shows a −20% budget rewrite and a new 11.6M at-risk flag.

That distribution answers the theme’s core cost question more cleanly than arrival headlines. Rich-country political narratives still orbit border encounters and asylum backlogs. The Q3 official vintage says the cash side of the multilateral ledger moved late in summer, while the host side of the ledger — who shelters refugees day to day — has not been restated and was already majority-LMIC when last measured. Pair the burden panel with our global refugee hosting burden and OECD DAC ODA drop posts: private remittances and bilateral ODA fights do not substitute for the hosting stock that low- and middle-income countries still carry.

UNHCR’s scissors cut the ask before coverage healed

The UNHCR budget panel is the agency-level version of the same story. The 2025 Global Report closed at a $10.604B needs budget with only 37% funded. The 2026 Global Appeal resets the needs budget to $8.505B — about 20% lower — while planning around 136 million forcibly displaced and stateless people. Early pledging covered roughly 18% of that smaller envelope. The comparison is not like-for-like year-end funded percentages, and the dashboard labels the early-pledge share as estimated relative to the 2025 year-end meter.

Read the cut as strategy under constraint, not as evidence that needs fell 20%. UNHCR’s own mid-year messaging still ties funding cuts to roughly 11.6 million people at risk of losing urgently needed assistance. The solutions-channel bars keep the asymmetry from the May update: 14.7 million returns (carried) dwarf 81,800 resettlement arrivals, while the new at-risk bar sits near the scale of the entire refugee-return corridor. Durable solutions are still mostly returns; third-country pathways remain a rounding error; and operational cut risk is now a first-class meter beside them.

Crisis plans: average coverage hides the thin corridors

Toggle View → Crisis plans. The scatter is an editorial mix scaled to the August GHO total — not a cite-ready FTS extract for any single plan. X is requirements, Y is coverage, bubble size tracks people in need. A horizontal reference marks the GHO average at 40.4%. The geometry matters more than any one bubble: large African and multi-country corridors often sit below the average line even when the global coverage meter looks “healed.” Ukraine-class plans can clear 50% while Sudan / Sahel-class plans stay stuck in the mid-20s in this mix.

That is the vintage lesson for desks that only quote the global percentage. 40.4% is a portfolio average. It does not mean every crisis corridor is 40% funded. Use the scatter for relative burden geometry; use official FTS plan pages when you need a single-plan citation. The dashboard’s source footer says the same thing in plain language.

How to read the dashboard

Start with View → Cash ledger and Sort → Largest Δ to see which meters actually moved. Flat dumbbells are carried people prints; teal bars are cash healing; red marks worsening or cut risk. Open the waterfall for the +16pp decomposition, then the dual stock×cash path to keep the frozen census honest. Switch to Who bears it and filter by hosts / donors / agency when the question is distribution rather than headline coverage. Finish on Crisis plans if the question is whether the average hides thin corridors. Hover any mark for confidence labels — disclosed, estimated, or carried.

Caveats and source boundaries

1. Displacement stocks and host shares are carried from UNHCR Global Trends 2025 until Mid-Year Trends land. Do not treat the flat 117.8M as a new H1 2026 census.
2. GHO coverage is derived from disclosed requirements ($34.87B) and funding ($14.08B) on the humanitarianaction.info monitoring snapshot; FTS figures move daily.
3. People-in-need / targeted remain the May Mid-Year Review prints (252.1M / 143.2M) unless a newer GHO people restatement is cited.
4. UNHCR 2025 year-end funded % (37%) is not like-for-like with 2026 early-pledge share (~18%). The scissors panel is a budget-path story, not a matched funding-rate YoY.
5. Crisis-plan scatter points are estimated editorial mixes, not official plan extractslabeled as such in the viz.
6. At-risk caseload (~11.6M) is an operational UNHCR flag, not a Global Trends stock category.
7. Returns and resettlement remain Global Trends 2025 carried solutions meters in this Q3 note.

Methodology

Prior vintage mirrors meters from migration-humanitarian-update-2026: end-2025 Global Trends stocks, GHO 2026 Mid-Year Review FTS as of 31 May 2026 (requirements $33.66B, funding $8.21B, coverage 24.4%, PiN 252.1M), and UNHCR Global Report 2025 budget/funded ratios. Newest cash ledger uses the OCHA GHO 2026 monitoring snapshot on humanitarianaction.info ($34.87B requirements, $14.08B funding; coverage 40.4% derived). UNHCR 2026 budget and planning figures follow Global Appeal 2026 ($8.505B, 136M planning figure); early-pledge share and at-risk caseload follow UNHCR pledging / mid-year operational messaging. Dashboard deltas are arithmetic comparisons of those prints; carried meters are labeled carried; plan-level bars are estimated mixes for geometry only.

Unlike a border-apprehension dashboard, this Q3 update measures the late-summer humanitarian cash vintage against a carried global forced-displacement stock — who paid into the stripped appeal, who still hosts, and what the agency cut before coverage healed.

Related reading: May Global Trends update, migration–humanitarian research, global refugee hosting burden, and OECD DAC ODA first drop.