England now has a plan to end homelessness

Home

Edition

Africa Australia Brasil Canada Canada (français) España Europe France Global Indonesia New Zealand United Kingdom United States

Edition: Global

Africa
Australia
Brasil
Canada
Canada (français)
España
Europe
France
Indonesia
New Zealand
United Kingdom
United States

s Newsletters

Academic rigour, journalistic flair An adult and child sized tent on a street in the UK

Yau Ming Low/Shutterstock England now has a plan to end homelessness – here’s how to test whether it will work Published: December 23, 2025 3.55pm GMT Michael Sanders, Julia Ellingwood, King's College London

Authors

Michael Sanders
Professor of Public Policy, King's College London
Julia Ellingwood
Research Fellow, School for Government, King's College London

Disclosure statement

Michael Sanders receives funding from the Centre for Homelessness Impact.

Julia Ellingwood receives funding from the Centre for Homelessness Impact.

Partners

King's College London provides funding as a member of The Conversation UK.

View all partners

DOI

https://doi.org/10.64628/AB.tjhnvawtr

https://theconversation.com/england-now-has-a-plan-to-end-homelessness-heres-how-to-test-whether-it-will-work-272377 https://theconversation.com/england-now-has-a-plan-to-end-homelessness-heres-how-to-test-whether-it-will-work-272377 Link copied Share article

Copy link Email Bluesky Facebook WhatsApp Messenger LinkedIn X (Twitter)

Print article

The UK has proved before that it can end homelessness. The Everyone In scheme during COVID lockdowns accommodated tens of thousands of people in emergency and supported housing, who would otherwise have continued sleeping rough.

But this was only temporary. Nearly six years later, the scale of the challenge is immense. In June 2025, 132,410 households were living in temporary accommodation, almost two-thirds of which were families with children.

The UK government has published a new homelessness strategy for England. The strategy speaks to different forms of homelessness, from rough sleeping to more hidden forms of homelessness, like sofa surfing.

This is a wide-ranging plan, bringing in approaches from different government departments. The £3.5 billion strategy aims to “address the root causes” of homelessness, firstly through a number of universal approaches.

Some of these have already been announced. The plan emphasises the government’s plan to build 1.5 million new homes during this parliament, reforms to renters’ rights, ending the two-child benefit cap and the youth guarantee to get more young people into work or education. It introduces a new commitment to update social housing allocation guidelines, and a legal “duty to collaborate” for public services to address homelessness.

Several more targeted measures look at specific at-risk populations. This includes care leavers, of whom a worrying proportion still go on to experience homelessness. It also includes people leaving institutions including healthcare settings and prisons. These targeted approaches are essential to move from a crisis-based approach to managing homelessness towards a more proactive approach to preventing it.

How do we know what works?

Our research focuses on methods such as randomised trials to evaluate policies across a range of topics, including homelessness. This is why we were pleased to see an emphasis on the government’s “test and learn” approach to technology and AI being applied to the homelessness strategy:

We want to adopt a test and learn approach to evidence, where local areas trial innovative practice, roll this out where it is effective, and subsequently share learning with others.

Given the UK government’s precarious financial situation, it’s important that policies work for the people who need them, without wasting money on untested approaches.

Randomised controlled trials, most common in medicine, are the gold standard method to show simply what the effect of an intervention or policy is. They work by assigning who gets an intervention (such as a vaccine or targeted homelessness support) at random, and comparing those that do to a control group.

A randomised trial in Canada has shown that modest, unconditional cash transfers for people experiencing homelessness can significantly reduce the number of days the average participant spent homeless in a year.

New studies are evaluating whether this approach can have the same kinds of effects in the UK. We are leading the evaluation of one project – administered by the charity Greater Change – which will test how giving people a personalised budget of around £4,000 can help them change their trajectory in life. We are looking forward to having results next summer.

Another trial already underway in the UK involves local councils working with data science company Xantura to give early warnings of households at risk of homelessness. In this trial, households identified by algorithm as being high-risk are randomly assigned either to receive proactive support, or not to. The outcomes of both groups are followed up later.

The evidence for this kind of approach is mixed. There is a burden of proof that companies selling these tools, and governments purchasing them, bear before making their use widespread. Randomised trials, while not perfect, are arguably the most rigorous and efficient way to test them.

Close up hand holding a stack of 20 pound notes

Randomised trials can test the effects of approaches like giving people grants or housing. Alexey Fedorenko/Shutterstock

Sometimes, randomisation is not possible. But we can still use experimental approaches or evaluations to test the effects of policies.

In recent projects, we’ve seen that two approaches have statistically meaningful effects on reducing homelessness for these young people. These are Staying Put, a policy that allows young people to remain with their foster carers after they turn 18, and Lifelong Links, an approach that supports young people in care to build and maintain relationships with people from their birth families. The government has continued to fund the expansion of these programmes, which the evidence suggests will have substantial effects on reducing homelessness for care leavers.

Of course, research evaluations can and do find when interventions are not successful – or are even actively harmful. For example, we found that a model of “extended families” for young people in foster care actually increased the rate at which they go on to experience homelessness.

It’s a positive sign that the government is embracing testing and learning. But this should mean making use of rigorous methods like randomised trials. Shying away from them risks imperilling the UK’s ability to actually end homelessness.