Hidden homelessness: do the numbers miss the real crisis?

The debate in brief

When the public pictures homelessness, they picture rough sleeping -- someone in a doorway or under a bridge. Government policy reflects the same instinct. The annual rough sleeping snapshot, conducted on a single autumn night, is the most-reported homelessness statistic in the UK. In autumn 2024, the snapshot recorded 4,667 people sleeping rough in England. That number is real and serious. It is also a fraction of the problem.

Behind the visible crisis sits a much larger hidden one. Crisis estimates that around 299,100 households in England experienced the worst forms of homelessness in 2024, including those in hostels, temporary accommodation, and sleeping in cars, tents, or on public transport. The National Housing Federation calculates that 3.6 million people in England live in overcrowded, unaffordable, or unsuitable housing. Centrepoint estimates that around 124,000 young people aged 16-24 in the UK faced or were at risk of homelessness in 2024-25, the vast majority of them sofa surfing with no contact with statutory services.

The argument is not that rough sleeping counts are wrong -- it is that they define the problem too narrowly, directing political attention and funding toward the most visible manifestation while the structural crisis grows largely unmonitored.

Quick takeaways

The arguments

The case that rough sleeping counts are inadequate

The annual rough sleeping snapshot is, by design, a count of the most visible form of homelessness on a single night. Local authorities in England can choose between a physical count (walking the streets) and an estimate (based on intelligence from outreach teams and agencies). In 2024, only 63 of 309 local authorities conducted a physical count; the remainder used estimates.

The methodology has well-documented limitations. Counts take place on a single night, typically in autumn, missing seasonal variations. They cannot capture people sleeping in hidden locations -- cars, sheds, derelict buildings, or on night buses. They do not include people in hostels, night shelters, or hospital and prison settings. CHAIN, the more comprehensive London-based database, consistently records far higher numbers than the snapshot because it captures individuals encountered over an entire year.

But the deeper problem is definitional. By making rough sleeping the headline measure of homelessness, the government frames the issue as one of visible destitution rather than structural housing failure. This framing shapes everything downstream: public perception, media coverage, political urgency, and how funding is allocated. A rough sleeping count of 4,667 is a crisis. A hidden homelessness figure of nearly 300,000 households is a catastrophe. The difference in response should be proportionate, but it is not.

The case that rough sleeping counts still matter

Rough sleeping is the most dangerous and life-threatening form of homelessness. The average age of death for people experiencing homelessness in England and Wales was 45 for men and 43 for women in the ONS's most recent analysis. People sleeping rough face extreme weather, violence, and acute health crises. The argument for counting and prioritising rough sleeping is not that it captures the full picture -- it is that it captures the most urgent part of it.

There is also a practical argument for a defined, measurable metric. The rough sleeping snapshot, for all its flaws, provides a consistent annual benchmark that can be compared over time and across areas. Attempts to measure hidden homelessness produce estimates with wide confidence intervals, making it harder to hold government to account against specific targets. When the previous government committed to ending rough sleeping, the target was meaningful because the metric was defined. A commitment to end "hidden homelessness" would be harder to measure and easier to evade.

The concern from some frontline organisations is that broadening the definition risks diluting urgency. If someone sofa surfing and someone sleeping in a doorway in January are counted in the same statistic, policymakers may lose sight of the immediate lethality of rough sleeping. The counter-argument is that this is a false trade-off: it is possible to maintain urgency about rough sleeping while also measuring and responding to hidden homelessness.

The structural view

The most challenging argument is that the focus on counting any form of homelessness distracts from its causes. The UK has a housing supply crisis. England needs around 340,000 new homes per year, according to research commissioned by the National Housing Federation and Crisis from Heriot-Watt University, with a backlog of approximately four million homes. Social housing stock has declined from 5.5 million in 1981 to around 4 million today. Local Housing Allowance rates, frozen for years and only partially uprated, cover the bottom 30th percentile of local rents in theory but increasingly fail to do so in practice.

From this perspective, the distinction between rough sleeping, sofa surfing, and overcrowding is less important than the common cause: an insufficient supply of affordable, secure housing. Counting hidden homelessness more accurately is necessary but not sufficient. Without a response to the housing supply crisis, better data simply documents a problem that policy is failing to address.

The evidence

Crisis's research provides the most comprehensive attempt to quantify hidden homelessness. Their 2024 England Monitor estimated that around 299,100 households experienced the worst forms of homelessness in 2024, a figure that has risen sharply since earlier estimates, driven by increases in those sofa surfing, in unsuitable temporary accommodation, and in hostels and supported housing.

MHCLG statutory homelessness data adds further detail. In the financial year 2023-24, local authorities in England received 358,370 initial assessments for homelessness or threat of homelessness under the Homelessness Reduction Act. At the end of September 2024, 126,040 households were in temporary accommodation, including 164,040 children -- the highest figure on record.

The Heriot-Watt University study commissioned by Crisis projected that without policy change, core homelessness in England would rise by 27% between 2023 and 2036. Their model factored in housing supply, welfare policy, and demographic trends, concluding that the upward trajectory was driven by structural factors rather than individual circumstances.

Centrepoint's research on youth homelessness highlights the invisibility of younger populations. Their 2025 report estimated that around 124,000 young people aged 16-24 in the UK faced or were at risk of homelessness in 2024-25. Among this group, sofa surfing was the most common experience, with only a minority engaging with local authority housing services. The main reasons given were family breakdown, domestic abuse, and mental health crises.

Current context

The Labour government elected in July 2024 established a cross-departmental Interministerial Group on Homelessness and committed to developing a new national homelessness strategy for England. As of April 2026, the strategy is in development, with the government having completed a public consultation in late 2025. Deputy Prime Minister Angela Rayner described the temporary accommodation figures as a "national scandal" and indicated that the strategy would take a broader view of homelessness than previous approaches.

The Kerslake Commission on Homelessness and Rough Sleeping, in its most recent report (September 2023), called on the government to adopt a wider definition of homelessness for strategic purposes, encompassing all forms of housing insecurity rather than just rough sleeping. The Commission called for a statutory duty to publish annual estimates of hidden homelessness alongside the existing rough sleeping snapshot.

Local Housing Allowance rates were uprated in April 2024 to the 30th percentile of local rents for the first time since 2020, providing some relief. However, research by the Institute for Fiscal Studies projected that rates would once again fall behind market rents within 18-24 months without a commitment to ongoing uprating. The Spring Statement 2026 did not announce further uprating.

Meanwhile, the number of households in temporary accommodation continues to rise. The financial cost to local authorities exceeded 1.9 billion in 2024/25, with some London boroughs spending more on temporary accommodation than on any other single service. Councils including Birmingham, Bournemouth, and Hastings have declared housing emergencies.

Last updated: April 2026

What this means for charities

Homelessness charities face a practical dilemma: their funding, their metrics, and their public messaging are built around definitions of homelessness that capture only part of the population they exist to serve. Organisations focused on rough sleeping are well understood by donors and commissioners. Those working on hidden homelessness -- supporting people to sustain tenancies, mediating family breakdown, providing advice on housing rights -- do less visible but often more preventative work that is harder to fund and harder to explain.

Charities should advocate for better data on hidden homelessness, including support for the Kerslake Commission's recommendation of annual hidden homelessness estimates. They should also examine their own data for what it reveals about the people they are not reaching. If a homelessness charity's service users are overwhelmingly people who have already reached rough sleeping, the organisation is operating as emergency response rather than prevention.

For charities outside the homelessness sector, the hidden homelessness data is relevant to service design. Mental health charities, domestic abuse services, children's charities, and advice organisations all work with people who are hidden homeless. Screening for housing insecurity should be routine in assessments, and referral pathways to housing advice should be in place.

Trustees and funders should interrogate the metrics used to define success. If a homelessness service reports on the number of people brought off the streets, that is valuable but incomplete. How many people were prevented from reaching the streets? How many were supported out of sofa surfing or overcrowding? These outcomes are harder to count but no less important.

Common questions

What counts as hidden homelessness?

There is no single legal definition, but the term generally encompasses: sofa surfing (staying with others because you have no home), sleeping in cars, sheds, or other non-residential buildings, living in overcrowded or unsuitable conditions, and staying in temporary accommodation that is not a settled home. What unites these situations is that the person has no secure, appropriate housing of their own but does not appear in rough sleeping statistics.

Why don't hidden homeless people approach their council?

Many do, but the system is difficult to navigate and the outcomes are often inadequate. Under the Homelessness Reduction Act 2017, local authorities must assess anyone who is homeless or threatened with homelessness within 56 days. But in practice, many people -- particularly young adults and those without a "priority need" -- are turned away or offered advice rather than accommodation. Others do not approach because they do not recognise themselves as homeless, fear being judged, or have had previous negative experiences with statutory services.

How accurate are the hidden homelessness estimates?

They are estimates with significant uncertainty. Crisis's figures are modelled from survey data, statutory returns, and research projections rather than direct counts. The true figure could be higher -- some researchers argue it is likely to be substantially higher, because the most hidden populations are by definition the hardest to count. The estimates should be treated as indicative of scale rather than precise figures, but even at the lower end they dwarf the rough sleeping snapshot.

Is sofa surfing really homelessness?

Legally, yes. The Housing Act 1996 defines a person as homeless if they have no accommodation they are entitled to occupy, or if their accommodation is not reasonable to continue to occupy. Someone staying on a friend's sofa with no security and no right to remain meets this definition. The distinction between sofa surfing and "real" homelessness is a public perception issue, not a legal one -- and it is precisely this perception that allows hidden homelessness to be overlooked.

Does focusing on rough sleeping affect funding?

Significantly. Government homelessness funding has historically been weighted toward rough sleeping interventions, including the Rough Sleeping Initiative which has distributed hundreds of millions of pounds since 2018. Charitable funding follows similar patterns: donor appeals featuring rough sleeping generate more income than appeals about temporary accommodation or overcrowding. This is rational from a fundraising perspective but contributes to the underfunding of prevention and hidden homelessness services.

Are temporary accommodation figures part of hidden homelessness?

Temporary accommodation occupies an unusual position. The people in it are counted -- MHCLG publishes quarterly statistics -- but they are not captured in the rough sleeping snapshot because they are technically housed. Whether 126,040 households in B&Bs, hostels, and nightly-paid accommodation count as "homeless" is a matter of definition, but for the 164,040 children living in temporary accommodation at the end of September 2024, the distinction between homelessness and temporary accommodation is largely academic.

Key sources and further reading

• Homelessness Monitor: England 2024 -- Crisis/Heriot-Watt University. The most comprehensive annual assessment of homelessness trends in England, combining statistical analysis with policy commentary and projections.

• MHCLG Statutory Homelessness Statistics -- Ministry of Housing, Communities and Local Government, quarterly. The official data on local authority homelessness assessments, duties owed, and temporary accommodation placements, providing the most authoritative picture of statutory homelessness.

• MHCLG Rough Sleeping Snapshot -- Ministry of Housing, Communities and Local Government, annual. The official single-night estimate of rough sleeping in England, published each spring for the previous autumn's count.

• CHAIN Annual Report -- Greater London Authority. London's multi-agency database of rough sleeping, recording every individual encountered by outreach teams across the year. Consistently the most detailed rough sleeping dataset in the UK.

• Kerslake Commission on Homelessness and Rough Sleeping: Turning the Tide -- September 2023. Cross-sector commission chaired by Lord Kerslake, recommending a broader definition of homelessness for policy purposes and annual hidden homelessness estimates.

• "Swept Under the Carpet" youth homelessness report -- Centrepoint, 2025. Research on hidden homelessness among 16-25-year-olds, documenting the scale of sofa surfing and non-engagement with statutory services among young people.

• National Housing Federation housing need analysis -- NHF/Crisis/Heriot-Watt University. Analysis of housing supply requirements, estimating that England needs around 340,000 new homes per year including at least 90,000 social rent homes annually.

• ONS Deaths of homeless people in England and Wales -- Office for National Statistics, ongoing. The most reliable data on mortality among people experiencing homelessness, providing the evidence base for the lethality of rough sleeping.