Homelessness has rapidly become a key challenge facing local authorities. Spending by councils in England on temporary accommodation has reached £2.8 billion a year, up from £1.7 billion a couple of years ago.
This is causing great harm. There were 132,410 households staying in temporary accommodation in June 2025 because they risked homelessness, a record high. Almost two thirds are families with children.
Evidence shows us that children who experience homelessness do less well at school, have poorer health and are more likely to be impacted by homelessness in adulthood.
Costs are soaring but outcomes are worse. The reason? Homelessness spending is heavily skewed towards responses once individuals and families reach crisis. Our system reacts rather than prevents. Our aim must be to drive this system towards prevention upstream.
There are challenges in embracing prevention. Many local authorities are losing their institutional memory of what effective homelessness prevention involves; things like tenancy sustainment, legal support, advice on debt. Resources and capacity have been sucked into crisis interventions.
We also lack evidence of which prevention programmes work and provide value for money. This requires experimentation, coupled with independent evaluation.
Investing in prevention requires an up-front cost. Councils have to meet current demand and fund preventative services whose benefits take longer to emerge.
That’s why it was significant that Rachel Reeves, in the Spending Review in June, allocated £100 million over the next two years specifically for homelessness prevention. Its purpose is to ‘prevent homelessness through investing in early intervention measures’. Most of the money will come from the Transformation Fund to support fundamental reform of public services, overseen by the Cabinet Office.
Ideally, this should endow a national fund for research and development of early prevention initiatives to create change at a system level. Without national scaffolding – shared data, common evaluation standards and support for implementation – we risk repeating the “pilot malaise” that has held back progress for a decade. It appears, however, that the Government’s preference is to pass this additional funding to local areas.
If so, it is critical that this additional funding does not top up councils’ current day-to-day homelessness spending, which will be drawn inexorably towards reactive crisis responses. More of the same will produce the same bad outcomes.
Local authorities should use part of this money to adopt or test promising early interventions.
This will require leadership. Prevention work is often less visible than crisis services. Polling evidence shows public support stronger for tangible things like emergency shelters, hostels and refuges than for prevention activity, although this still has majority support. When asked what would make the greatest difference in reducing homelessness, between 82% and 84% say refuges, safe houses and emergency housing, while 56% say investing in prevention services.
It will also require support. Some councils have shown they can innovate: the London Borough of Barking and Dagenham, for instance, used data it holds on residents’ financial stress, such as council tax arrears, to offer them targeted support such as a debt repayment plan.
Such initiatives are very much the exception, however, not the rule. Councils vary enormously in capability. Expecting each to design, procure and evaluate interventions independently leads to duplication, slow progress and uneven quality.
And, for local authorities that do develop effective prevention approaches a big question remains about how this learning can be shared and scaled across the system.
This is where metro mayors and combined authorities can advance the devolution agenda by stepping up and leaning in to this homelessness prevention challenge. Combined authorities provide an appropriate scale for rigorous testing and shared learning: large enough for unified datasets and specialist expertise, small enough to stay rooted in place. They can take on the capabilities individual councils cannot realistically build alone – intervention design, behavioural science, predictive analytics, flow management, and evaluation methodology – and help ensure prevention becomes the norm, not the exception.
The rigorous evaluation and learning that determines what truly works must be coordinated. Combined authorities can ensure councils are not left to reinvent the wheel, and that learning can be generated once and shared many times. Councils should be held to account for investing in prevention, but not for running their own isolated R&D functions.
And city mayors and combined authorities have the heft to build approaches that sit well beyond the homelessness sector and embed primary prevention within mainstream public services, such as health settings, schools, and the criminal justice system.
It is also why national government must play a role as a catalyst: not delivering services, but providing the enabling infrastructure – data architecture, evaluation capability, shared procurement routes and implementation support – that allows the strongest approaches to spread with fidelity and pace. With national scaffolding behind them, combined authorities can turn individual successes into system-wide progress.
Labour’s manifesto for the 2024 general election promised a new strategy ‘working with mayors and councils across the country, to get Britain back on track to ending homelessness’. The £100 million allocated from the Transformation Fund for homelessness prevention offers them the opportunity to do that. They must seize it.