Housing First vs the staircase model: what works for ending homelessness?

The debate in brief

For decades, the dominant approach to homelessness in the UK has been the "staircase" model: people move through stages of temporary accommodation, demonstrate they are "housing ready" — sober, stable, engaged with treatment — and eventually earn a permanent tenancy. Housing First inverts this entirely. It gives people an unconditional home first, then provides intensive, open-ended support to address other needs once they have a stable base.

The evidence strongly favours Housing First for people with the most complex needs. The UK government's own evaluation of three regional pilots found 84% tenancy sustainment after three years. Finland has reduced long-term homelessness by 68% since adopting the approach nationally. Yet despite this evidence, England's National Plan to End Homelessness, published in December 2025, did not include a national rollout of Housing First — a decision that drew sharp criticism from Crisis and Homeless Link.

The arguments

The case for Housing First

Housing First starts from a simple premise: you cannot address addiction, mental health, or unemployment while living on the streets or cycling through temporary hostels. A stable home is not the reward for recovery — it is the foundation for it.

The model was developed by psychologist Sam Tsemberis in New York in 1992, and has since been adopted across North America, Europe, and Australasia. Its core principles are unconditional access to permanent housing, choice and control for the individual, separation of housing from treatment, and personalised, open-ended support.

The UK's three regional Housing First pilots — in Greater Manchester, Liverpool City Region, and the West Midlands — supported 1,061 people into independent tenancies between 2018 and 2023. The government-commissioned evaluation, published by MHCLG in October 2024, found tenancy sustainment of 84% after three years, with average costs of £7,700 per person per year and estimated long-term savings of £15,880 per person per year through reduced use of emergency services, hospitals, and the criminal justice system.

Jon Sparkes, then Chief Executive of Crisis, described Housing First as offering "the stability and security of their own front door" as "a life changing first step towards recovery," arguing that it "makes sense for the public purse" and "has the potential to deliver savings for local councils right across the country." (Note: this quote is from a 2017 Crisis press release, not from the 2024 pilots evaluation.)

The strongest objection to Housing First is that it requires a supply of affordable one-bedroom homes that does not currently exist. Advocates acknowledge this, but argue that the housing supply problem is a reason to build more homes, not a reason to keep people in hostels.

The case for the staircase approach

The staircase model is not without its defenders. Its logic is straightforward: some people need structured support — help with substance use, mental health treatment, life skills — before they can sustain a tenancy. Putting someone with active addiction into an unsupported flat, the argument goes, sets them up to fail.

Supporters point out that Housing First is designed for a specific group: people with the highest and most complex needs who have experienced chronic homelessness. It was never intended to replace the entire homelessness system. The House of Commons Library briefing on Housing First (CBP-8368) is clear that Housing First "is not targeted at all homeless people" but focuses on those with "the highest and most complex needs." For people experiencing short-term homelessness due to relationship breakdown, job loss, or benefit delays, supported accommodation and structured pathways may be more appropriate and more cost-effective.

There is also a practical question about fidelity. Housing First requires intensive, open-ended, personalised support — typically a caseload of no more than five to seven people per worker. When scaled without adequate funding, programmes that call themselves Housing First may deliver something closer to "housing only," without the wraparound support that makes the model work. The UK pilots themselves identified staff recruitment and property acquisition as significant barriers.

The strongest objection to the staircase model is that it demonstrably fails the people it is meant to serve. Research published by Crisis found that people frequently become "stuck" in staircase services, unable to meet the escalating conditions required to progress, and are often evicted back to earlier stages — or out of the system entirely — when they relapse.

The emerging consensus: housing-led systems

In practice, the debate has moved beyond a binary choice. Most homelessness organisations now advocate for a "housing-led" system in which rapid rehousing into settled accommodation is the default, with Housing First reserved for those with the most complex needs and lighter-touch support available for others. Crisis has recommended this approach as part of its Plan to End Homelessness, and the Centre for Homelessness Impact rates Housing First among the strongest evidence-based interventions for ending chronic homelessness.

This consensus position accepts that the staircase model's fundamental premise — that people must be "ready" for housing — gets the causation backwards. But it also recognises that Housing First alone cannot solve homelessness without addressing housing supply, welfare adequacy, and prevention.

The evidence

The evidence base for Housing First is unusually strong by social policy standards. A systematic review commissioned by the Centre for Homelessness Impact through the Campbell Collaboration, covering 28 studies reported across 51 articles and involving thousands of participants, found that programmes like Housing First result in significantly better housing stability than models with less or no tailored support.

The UK government's final synthesis report on the three regional pilots, published in October 2024, found 84% tenancy sustainment at three years, with improvements in physical and mental health, reduced drug and alcohol use, and significant reductions in antisocial behaviour and criminal justice contact. The cost-benefit analysis found average delivery costs of £7,700 per person per year against estimated savings of £15,880 per person per year.

Scotland's Housing First Pathfinder, evaluated by Heriot-Watt University, supported 579 people across five areas from 2019 and achieved tenancy sustainment of 88% at 12 months and 80% at 24 months.

Internationally, Finland's national adoption of Housing First from 2008 reduced long-term homelessness by 68% between 2008 and 2022, with pilot testing demonstrating annual savings of approximately EUR 21,000 per person through reduced hospital visits and rehabilitation services.

The evidence gaps are real, however. Cost-effectiveness findings are less consistent than housing outcomes — a systematic review found that studies using experimental designs sometimes reported net cost increases, while pre-post studies reported savings. The impact on substance use and mental health, while positive in the UK pilots, is harder to attribute solely to housing. And almost all the robust evidence comes from programmes targeting chronic homelessness with complex needs, leaving questions about effectiveness for broader homeless populations.

Current context

The UK government published its National Plan to End Homelessness in December 2025, backed by £3.5 billion of investment. The plan pledges to halve long-term rough sleeping, end the unlawful use of B&Bs for families, and prevent more households from becoming homeless. It includes a new £124 million supported housing scheme for over 2,500 people.

However, the plan does not include a national expansion of Housing First — a notable omission given the evidence. Crisis responded that ministers were "falling short of what's desperately needed," and the Institute for Government concluded that the strategy's "actions fall short of ambition." Homeless Link had estimated that delivery of Housing First at scale would require approximately £150.3 million per year.

The scale of the problem continues to grow. Rough sleeping in England reached a record 4,793 people on a single night in autumn 2025, a 3% increase on the previous year and a 171% increase since 2010. The number of households in temporary accommodation hit 134,760 by September 2025 — another record — with over 172,000 children living in temporary accommodation.

Last updated: April 2026

What this means for charities

For homelessness charities, this debate has immediate operational implications. Organisations delivering traditional staircase services face growing evidence that their model produces worse outcomes for people with the most complex needs. But transitioning to Housing First requires different staffing models (lower caseloads, higher-skilled workers), different commissioning relationships, and — critically — access to affordable housing stock.

Charities should be honest about what the evidence says, even when it challenges their existing service models. They should also resist the temptation to rebrand existing services as "Housing First" without adopting the model's core principles, particularly unconditional tenancy access and open-ended support.

For funders and commissioners, the evidence points clearly toward investing in Housing First for people with complex needs and housing-led approaches for the broader population. The biggest risk is not the choice between models — it is the chronic underfunding of both, combined with a housing supply crisis that no support model can solve on its own.

Common questions

What is Housing First?

Housing First is an approach to ending homelessness that provides people with an unconditional permanent tenancy before addressing other needs such as addiction or mental health. Unlike the staircase model, it does not require people to demonstrate "housing readiness" first. Once housed, intensive and personalised support is provided for as long as it is needed. The model was developed by psychologist Sam Tsemberis in New York in 1992 and has since been adopted across North America, Europe, and Australasia.

How many people are in temporary accommodation in England?

As of September 2025, 134,760 households were living in temporary accommodation in England — a record high. More than 172,000 children were living in temporary accommodation at the same point. The number of people sleeping rough on a single night reached 4,793 in autumn 2025, a 3% increase on the previous year and a 171% increase since 2010.

Does the UK use Housing First?

The UK has run limited pilots rather than adopting Housing First nationally. Three regional pilots — in Greater Manchester, Liverpool City Region, and the West Midlands — supported 1,061 people between 2018 and 2023. Scotland ran a separate Housing First Pathfinder across five areas from 2019. Neither England nor Scotland has committed to a full national rollout. England's National Plan to End Homelessness, published in December 2025, did not include a funded national Housing First programme.

What is the staircase model?

The staircase model — also called "treatment first" or "housing readiness" — is the traditional approach to homelessness in the UK. People move through a series of stages, from street outreach to night shelters, then hostels, then supported accommodation, and eventually to a permanent tenancy. Progress through the stages depends on demonstrating stability, engagement with treatment, and compliance with conditions. Critics argue that this model gets the causation backwards — it is very hard to address addiction or mental health problems without the security of a stable home — and that people frequently become stuck or are evicted back to earlier stages when they relapse.

What did the UK Housing First pilots find?

The government-commissioned evaluation of the three English regional pilots, published by MHCLG in October 2024, found that 84% of people sustained their tenancy after three years. Average delivery costs were £7,700 per person per year, against estimated long-term savings of £15,880 per person per year through reduced use of emergency services, hospitals, and the criminal justice system. The evaluation also found improvements in physical and mental health, reduced drug and alcohol use, and significant reductions in antisocial behaviour and criminal justice contact.

Key sources and further reading

• Evaluation of the Housing First Pilots: Final Synthesis Report — MHCLG/ICF Consulting, October 2024. The definitive evaluation of the three English regional pilots, covering outcomes, costs, and implementation lessons over five years.

• Housing First: Tackling Homelessness for Those with Complex Needs (CBP-8368) — House of Commons Library. A regularly updated briefing covering the evidence base, UK policy, and international comparisons.

• Staircases, Elevators and Cycles of Change — Crisis, 2010. The foundational UK analysis comparing Housing First with staircase models for people with complex needs.

• A National Plan to End Homelessness — UK Government, December 2025. The cross-government strategy backed by £3.5 billion of investment.

• Housing First Intervention Tool — Centre for Homelessness Impact. Evidence rating and systematic review of Housing First effectiveness.

• Scotland's Housing First Pathfinder Evaluation — Heriot-Watt University/I-SPHERE, 2022. Independent evaluation of the Scottish programme across five local authority areas.

• Housing First Europe Hub: Evaluation Framework — Housing First Europe. European evidence base and evaluation methodology for Housing First programmes.

• Statutory Homelessness in England: Financial Year 2024-25 — MHCLG, 2025. Official statistics on homelessness assessments, temporary accommodation, and outcomes.