There is a phrase that resurfaces whenever homelessness rises up the political agenda: we’ve ended homelessness before, so we can do it again.

It is usually offered with good intent. A signal of optimism. A reminder that progress is possible. And at a moment when record numbers of children are growing up in temporary accommodation in the UK – when the human and financial costs of failure are impossible to ignore – that reassurance can feel not just welcome, but necessary.

But the phrase also carries a quiet danger. Because it suggests that the path ahead is already known. That the solution lies somewhere in our past, waiting to be retrieved. That what worked once can simply be reapplied to a very different set of social, economic, and political conditions. And this is not quite true.

The UK has indeed seen periods of real progress on homelessness, most notably in relation to rough sleeping in the late 1990s and again during the pandemic. The Rough Sleepers Unit, and more recently the Everyone In response, showed what can be achieved when political will, funding and coordination align. At its peak, Everyone In brought thousands of people inside almost overnight.

But it is important to be precise about what that success represented. It was progress on one form of homelessness, the most visible, but also the least common. Rough sleeping fell, but wider housing insecurity did not disappear. Families continued to enter temporary accommodation. Young people continued to leave care into instability. Structural pressures in housing, welfare and local government remained unchanged. When the emergency ended and attention shifted, rough sleeping rose again.

The lesson is not that those efforts failed. It is that their success was narrow and never fully embedded in the system. That is not a critique of the people or programmes involved. It is a recognition of how difficult it is to sustain progress when the wider foundations – housing supply, local government capacity, welfare support and public services – are under growing strain. Even well-designed interventions struggle to endure when the system around them is under pressure.

This matters because homelessness is an issue we understand well. We know it should be rare, brief and non-recurring. We know that prevention is more humane and more cost-effective than crisis response. We know that intervening early dramatically improves outcomes. And yet, progress keeps slipping – not because people do not care, but because prevention is structurally disadvantaged. Its benefits accumulate slowly. Its successes are quiet. Its costs are immediate. And when systems come under strain, prevention is the first thing to be eroded.

This pattern is visible internationally. Finland is often held up as the exception: the only country to have achieved sustained reductions in homelessness. Its Housing First approach, backed by long-term investment in social housing and strong welfare infrastructure, reduced long-term homelessness by more than a third over a decade. But even Finland now shows signs of strain. As fiscal pressures increase and housing supply tightens, recent data suggests rough sleeping is beginning to rise again. Success, in other words, is not self-sustaining.

The same lesson emerges closer to home. The UK’s progress on veterans’ homelessness shows what durable change actually requires. Official estimates suggest rough sleeping among veterans fell by around half between 2018 and the early 2020s, following a concerted national effort to treat veterans as a distinct cohort with clear pathways into housing and support. What made the difference was not a single policy or funding stream, but a delivery model built around shared data, named responsibility, and sustained coordination between central government, local authorities and specialist providers. When that focus weakened and pressures elsewhere in the system grew, progress began to stall.  

This is the pattern that matters. Progress depends not only on ambition or political will, but on whether systems are deliberately designed to hold strong under pressure. When they are not, gains fade. This is where a longer-term way of thinking becomes essential, but not in the passive sense it is sometimes used. What is required now is a more deliberate form of system-building, one that recognises that progress on homelessness is not sustained by the way responsibilities are structured, decisions are taken, and delivery is supported over time. In other words, this is not just about thinking long term, but about actively designing systems that can learn, adapt and hold their course under pressure.

One way of framing this comes from the philosopher William MacAskill, who uses the term Viatopia to describe a society focused not on reaching a perfect end state, but on staying on a good path – one that keeps better futures possible rather than closing them off through short-term decisions. Applied to homelessness, the implication is practical rather than philosophical: progress depends on whether systems are actively built to sustain it, not simply whether they are well intentioned.

That distinction matters now more than ever. A National Plan to End Homelessness, the new strategy for England is being rolled out alongside strengthened statutory duties across the UK, placing greater responsibility on public services to identify and prevent homelessness earlier. At the same time, temporary accommodation numbers remain stubbornly high, local authorities are stretched to their limits, and visible homelessness is inescapable in every city centre. The pressure to demonstrate progress quickly is intense.

But this is also where risk lies. Because when pressure to demonstrate results grows, the temptation is to focus on what can be seen and counted quickly, rather than on what will actually reduce homelessness over time. The danger is not inaction, but action that looks decisive while leaving the underlying system unchanged.

Success will need to be more ambitious, not less, but ambitious in the right way. That means moving beyond isolated initiatives and towards a genuinely delivery-led approach: defining cohorts clearly; aligning national and local responsibilities; investing in data and feedback loops; and building in the capacity to test, learn and adapt what works over time, rather than relying on one-off interventions.

This is where the lessons from veterans’ homelessness, and from international experience, are most relevant. Progress was possible because political support was present, delivery was actively managed, and attention was sustained. Where those conditions weakened, progress faltered.

The same logic should now inform our approach to children growing up in temporary accommodation, to young people leaving care, and to families cycling repeatedly through crisis services. In each case, prevention depends on whether systems are designed to notice risk early, to act decisively, and to stay engaged long enough for stability to take root, even though the contexts and needs may differ.

Which brings us back to the question we started with. Yes, homelessness has been reduced before, but not in a way that was built to last. The real challenge now is not whether we can reduce homelessness again. It is whether we are willing to do the harder, less visible work of building systems capable of sustaining that progress over time. 

That means moving beyond short-term fixes. Beyond episodic initiatives. Beyond assuming that commitment alone is enough. It means treating prevention as infrastructure, delivery as a discipline, and long-term thinking as something that has to be actively built, not simply declared. 

That is the test before us, and it is one we will only pass if we are willing to match ambition with the sustained effort required to make it real.

• Ligia Teixeira is Chief Executive of the Centre for Homelessness Impact

‍

‍