Why this session matters to us
The session opened, as these sessions almost always do, with a wall of billions: 3.4 billion people without adequate housing, more than a billion in slums and informal settlements, over 300 million facing homelessness. The moral force of those numbers is real. Their precision is not. Minutes after the moderator gave the 3.4 billion figure, UN-Habitat’s Executive Director, speaking minutes later, put the same headline at “almost 3 billion.” UN-Habitat’s own new report says “around 3 billion.” The agency’s strategic plan says 2.8 billion. The single most cited statistic about the global housing crisis moves by hundreds of millions of people depending on who is saying it.
That is not a gotcha. It is the reason SDGCounting was interested in an hour that was otherwise built from testimony rather than data. The housing crisis is the part of the 2030 Agenda where the gap between what we can measure and what we are trying to fix is widest, and the deprivation is measured least well exactly where it is most severe. The useful thing this session did was show what fills that gap when the official count cannot: people counting themselves.
The number that will not hold still
SDG 11’s housing indicator, 11.1.1, counts the share of the urban population living in slums, informal settlements or inadequate housing. It is a genuine advance, and UN-Habitat has spent a decade getting more countries to report it. But two things sit outside it. The first is homelessness. The “over 300 million” figure repeated on this panel has no agreed global method behind it; older UN work put the number nearer 150 million, and there is no dedicated SDG indicator that obliges any country to count people sleeping rough at all. The most extreme form of housing deprivation is the one the framework sees least.
The second is everything around the four walls. The Executive Director caught this while listening to the opening testimony, in which a mother described never taking her children to a neighbourhood park.
The report she meant is UN-Habitat’s SDG 11 synthesis report, launched at a press briefing the previous morning, which finds public space (target 11.7) among the weakest-performing measures on the goal. Adequate housing, in other words, is not a single number but a bundle of them, most of them thinly reported, and a household can clear the shelter test while failing on space, safety, services and tenure. A count that stops at the roof will call that a success.
What the official count cannot see
The sharpest measurement point of the morning was not framed as one. Shirley Pryce, who organised Jamaica’s domestic workers for decades, explained why an informal-settlement address is a liability its residents learn to hide.
An informal settlement is, almost by definition, a place the state has not mapped: no registered addresses, often no street names, its residents actively misreporting where they live to survive the labour market. That invisibility is not a data footnote, it is the condition. It is why UN-Habitat frames the crisis as one of inequality rather than of shelter, and why Pryce reduced it to a line that doubles as a costing argument.
Rossbach made the same point through gender. Her account of a woman in the Brazilian Amazon, rehoused under the Minha Casa, Minha Vida programme, turned on a single change the housing statistics never captured: a tap. Before the house, the woman spent her days carrying water, kept her children out of school to help, and damaged her health doing it; after it, the children went to school and she could work. None of that is legible in a count of dwellings. It is the kind of thing a household survey registers only if it is built to ask, and the panel’s recurring word for building it, that word being co-creation, was really a statement about who gets to define what counts as an adequate home. A youth panellist from Egypt made the definitional problem explicit.
Counting from below
If the top-down count is thin, the floor showed what the bottom-up version looks like, and it is real statistical work. A community organiser from Jamaica described a group of women in the Newlands settlement using a safety-audit tool to document the hazards around them, taking the evidence to the local authority, signing a memorandum of understanding, and winning a concrete list: street lighting, the removal of the zinc fences that hid attackers, and, first of all, names for their streets. Naming the streets was not cosmetic. Without addresses, they said, the fire truck, the utility and the police could not find them. Addressability is the precondition for being counted at all.
A grassroots leader from the Mathare settlement in Nairobi took it one step further into formal data systems. Working through UN-Habitat’s Global Land Tool Network and its Social Tenure Domain Model, women in the settlement enumerated who lived where, which household held which plot, and got their own names written onto tenancy and land records rather than a husband’s or a landlord’s. The instrument matters because roughly 30% of the world’s land is formally registered and women hold documented rights to a tiny fraction of it, so a pro-poor land record that recognises informal claims is, in effect, a measurement tool for people the cadastre was never designed to see. This is the counting that changed lives in the room, and residents did it themselves.
Our read
Two ways of knowing the housing crisis shared the same hour and never quite met. From above came the billions, delivered with conviction and disagreeing with one another by the population of a large country, alongside an indicator set that can certify a family as adequately housed while it lives without a park, a safe address or a name on the lease. From below came a single mother’s five years in a shelter and grassroots women enumerating their own settlements, street by street and plot by plot. The distance between a contested 300-million estimate and one family’s testimony is the distance between a statistic and a life, and this session’s quiet argument was that the count has to start where people actually live. On adequate housing, the numbers we lead with are the softest in the SDG framework, and the data that moves anything is being gathered by the people the framework struggles to find.
Why it matters for the SDGs
This is SDG 11 (sustainable cities and communities), under in-depth review at HLPF 2026, and specifically target 11.1 on adequate, safe and affordable housing, its indicator 11.1.1, and the weak public-space measure under 11.7. It runs straight into SDG 5 (indicator 5.a.1 on women’s ownership and secure rights over land), SDG 1 and SDG 10 (the crisis as inequality), and SDG 16, since legal identity and a recognised address are what make a person visible to any of the others. The immediate stake is next week: the General Assembly holds its high-level meeting on the mid-term review of the New Urban Agenda on 16 and 17 July, where member states will adopt a political declaration on housing. As the panel kept insisting, whether that declaration improves anything depends on whether the people it is about are counted, and counted where they are.
Watch & read
- The Goals Lounge, UN Office for Partnerships, which convened the session.
- UN-Habitat SDG 11 Synthesis Report, the report launched the day before this session, and our summary of its launch briefing (9 July), where Rossbach and Pryce also spoke.
- Social Tenure Domain Model, the Global Land Tool Network’s pro-poor, gender-responsive land-recording tool referenced from the floor.
- 2026 High-level Meeting on the New Urban Agenda (16 to 17 July), whose political declaration on housing several floor speakers were negotiating.
- Cities that work for people and the SDG 11 in-depth review, the Forum’s other housing and cities coverage · 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. Non-English testimony and passages are summarised rather than quoted, and speakers who could not be confirmed by name are cited by role. Population figures are as speakers and UN-Habitat reported them; the global housing-crisis totals cited on the panel range from about 2.8 to 3.4 billion, and the homelessness figure has no agreed measurement method. The date is established from the session’s own references rather than an official programme listing.