Introduction
In September 2024, 9,097 people had slept rough in England over the previous month, an 8% increase on a year earlier. One third of these were new to rough-sleeping (street homelessness) but 34% were long-term rough sleepers. ONS data shows these numbers have been rising since 2010. The numbers sleeping rough have now reached similar levels to the late 1990s, when it was estimated 2,000 nightly slept rough in England, and 10,000 over the course of year. Sleeping rough for more than six months in the UK today cuts your life expectancy in half. This makes it one of the most punishing health inequalities in modern Britain. It comes with additional financial costs too.
Rough Sleeping has always existed, but became a highly visible feature of most cities in the UK during the 1990s. In 1990, Major’s Conservative administration launched a Rough Sleepers Initiative in London, which was expanded to selected cities outside the capital in 1996, but when Labour came to power in 1997 rough sleeping was still at extraordinarily high levels and widely considered a ‘national disgrace’.
New Labour afforded rough sleeping a high priority on election, setting a target of reducing the numbers of street homeless by two-thirds by April 2002 which it achieved early in November 2001. Labour also developed policies that succeeded in sustaining this reduction, and by 2010 rough sleeping numbers were at their lowest ever recorded level, and homelessness had ceased to be a visible issue. This paper seeks to understand how this was achieved and whether there are applicable lessons for today’s policy-makers
The extraordinary success of New Labour’s street homelessness initiatives has largely been forgotten, and this amnesia extends to the institutional memory of Whitehall, where lessons learned from the period have not been applied to contemporary homelessness policy making. This policy paper is underpinned by research that gathered the oral testimonies of ninety individuals involved in the formation and delivery of New Labour’s homelessness policy 1997-2010, from government ministers to frontline workers. It was followed up by a ‘witness event’ staged in December 2024, where key individuals from the era reflected on their experiences of the New Labour period in conversation with academics and current and former practitioners in the field.
Cross-cutting bodies and Prime Ministerial patronage
The mechanisms of government employed by New Labour were vital in its success. Homelessness is an archetypical ‘wicked’ issue, peripheral to many government departments but requiring a co-ordinated approach that cuts across departmental silos. To address such issues, Labour established a Social Exclusion Unit (SEU) in 1997, with Blair’s mandate. Homelessness was designated one of its first four priorities. The location of the SEU in the Cabinet Office and Prime Ministerial patronage were key to Labour’s success. Moira Wallace (SEU director) was clear that it was ‘only thanks to Blair’s advocacy [that] individual departments were brought into line’. The SEU co-opted staff from the homelessness sector, consulted widely, and produced a well-received report (July 1999). This seeded a second cross-cutting body, the Rough Sleepers Unit (RSU) (formed March 1999) that sought to address the most visible and extreme aspect of the homelessness spectrum.
The RSU – composition and working practice
The creation, composition and working practices of the RSU were important elements in its success. Located within Whitehall, the RSU co-opted expertise from the homelessness sector including its leader, Louise Casey, formerly deputy director of Shelter, a leading homelessness campaigning organisation. Unusually, it was given its own budget (combining four previously independent funding streams), media and legal teams. Set the specific target of reducing rough sleeping by two-thirds, it was afforded considerable autonomy in how to achieve this aim. The unit’s working practice was also very different to conventional civil service methods, described by its members as ‘non-hierarchical’, ‘bottom up’ rather than ‘top down’ and ‘hands on’, with its members taking part in street counts and outreach work on a regular basis.
Combined, these elements became vital in its success. Employing a charismatic and controversial leader and her designation as ‘homelessness tsar’ helped keep the issue firmly in the public eye and ensured continued government attention. Staffing was a blend of civil servants who knew their way around Whitehall and individuals co-opted from the homelessness sector. Both Casey and her deputy Ian Brady (from Centrepoint) described themselves as ‘poachers turned gamekeepers’, their understanding of the strengths and weaknesses of the homeless sector enabled them to drive through the changes necessary to improve delivery. The working practices adopted by the RSU helped generate a strong esprit de corps among its members and enabled insights gained from the frontline to be fed quickly back into the programme. The target galvanized the team, and focused voluntary sector providers on achieving resettlement outcomes.
Adapting services to address the needs of ‘entrenched’ rough sleepers
The focus on ‘entrenched’ rough sleepers – those who had been on the streets a long time and had become accommodated to street life – was critically important. This cohort, it was felt, had been overlooked in previous responses, and because of their complex needs and/or challenging behaviour much of the sector had unconsciously adapted its practices toward those with fewer issues who were easier to work with and resettle. Outreach services were re-targeted to focus on long-term rough sleepers and specialist mental health and substance misuse workers were recruited into outreach teams. The Homeless Mentally Ill Initiative (HMII) – originally established by Major for London – was expanded nationally, and co-ordination between the homeless and HMII outreach teams was enhanced. Rolling shelters, offering minimal barriers to entry were set up, and the number of hostel beds was increased including an additional 250 ‘high support’ beds. ‘Wet’ hostels and day centres were opened for alcohol dependent homeless people and significant investment was made in improving their access to detox and rehabilitation services.
Disruption
More controversially, but equally important, was the RSU’s adoption of a robust approach to getting people off the streets. Employing forms of ‘disruption’, were, according to Brady, vital to break established patterns of street living that trapped people in homelessness. Services, like soup runs, that the RSU considered to be not only failing to assist in resettlement, but to be actually sustaining street-living, were targeted for closure. Direct investment was instead channelled into day centres where food and clothing could be equally well be provided and homeless people could also gain access to resettlement and other services. Forms of ‘assertive outreach’ were introduced, co-working with the police was made mandatory, and agencies unwilling to adapt to these changes were deprived of funding and replaced by agencies more willing to embrace the new directives. These aspects of the RSU’s approach were resisted by much of the sector and viewed as inhumane by the public and press. However, in retrospect, the vast majority of the sector came to reject this negative interpretation, arguing that as rough sleeping exacts a terrible toll on health and life expectancy, it was vital to get people off the streets fast, and that effective rehabilitation and resettlement work could only be achieved if people were in some form of settled accommodation.
Long-term resettlement, meaningful activity and prevention
Importantly, despite the primary focus of the RSU on reducing rough sleeping numbers, steps were also undertaken to ensure that homeless people did not return to the streets or remain stuck permanently in inadequate hostel accommodation.
Key to this was the promotion of a culture change in the voluntary sector. Against an ethos of ‘unconditional acceptance’ that was widespread in the sector, the RSU strongly promoted approaches based on ‘rehabilitation and change’. Funding was only granted to hostel and day centre providers willing to work in a pro-active way, resettlement targets were set, and forms of conditionality were introduced into day centres where clients were required to ‘actively engage with services’ in order to continue attending. Measurable outcomes had to be fulfilled in order to maintain funding. This approach has been highly criticised by many in faith-based organisations and some academics, but is integral to achieving positive resettlement outcomes.
The RSU believed resettlement would only succeed long-term if resettled homeless people engaged in some form of ‘meaningful activity’ and were reconnected with wider society. The importance of work for successful resettlement had long been recognised by homeless agencies such as St Mungo’s, but such initiatives were rare prior to 1999. The RSU encouraged hostels to introduce programmes providing education, training and opportunities for volunteering or paid employment. New Tenancy Support Teams for people moving into independent accommodation were tasked with being much more proactive, and were set a target of 75% of tenants becoming involved in ‘meaningful activity’ within six months of taking up their tenancy. Not only did engagement in meaningful activity aid resettlement, but it was noted that when activity programmes were introduced in hostels, evictions, abandonments and violent incidents went down sharply.
The RSU also began to engage with homelessness prevention. Prevention measures remained ‘in their infancy’ according to a 2002 review, but steps were taken to prevent the flow of care leavers, ex-offenders and service veterans onto the streets. These included a ‘safe-stop’ scheme for young people, funding for prisoners at risk of homelessness on discharge, and links with the service veterans’ charity SSAFA. Louise Casey noted that small actions can be decisive in prevention, citing the success of prison discharges switching to Mondays rather than Fridays when accommodation services were more likely to be available.
Second phase – moving upstream
In the second phase, from 2001, the programme moved ‘upstream’, with a coherent series of initiatives designed both to sustain resettled homeless people in their accommodation and to prevent homelessness. This second phase of Labour’s programme has been largely forgotten, but the lessons learnt here are even more important for contemporary policymaking than the work of the RSU. Significantly, in 2001, the RSU was merged with the Bed and Breakfast Unit. The Homelessness Directorate, as it was now called, was tasked with ending the use of Bed and Breakfast for families with children and, in 2005, was set the target of reducing by 50% the number of homeless households in temporary accommodation. This was achieved in 2010. Since 2010 the numbers have risen from 48,010 households to 126,040 (September 2024).
Labour’s interventions included primary and secondary legislation, a new role for local authorities, a significant increase in funding for support services and a new emphasis on service-user empowerment. Key to Labour’s achievements was a profound transformation of the voluntary homelessness sector. Subject to new forms of monitoring and control, the sector was professionalised, creating a vastly improved skill-base that, in turn, enabled it to adopt of more sophisticated ways of working with vulnerable people. There were three key interventions in New Labour’s second and third terms; the passing of the Homelessness Act 2002; the creation of the Supporting People programme from 2003, and the Hostels Capital Improvement Programme, ‘Places of Change’ from 2006.
The Homelessness Act 2002: Homelessness prevention and a new role for Local Authorities
The impact of the Homelessness Act 2002 completely transformed the way in which local authorities responded to single homeless people. Mandated to develop ‘local homeless strategies’, councils were obliged, for the first time, to make an audit of local need and provision, provide advice and assistance for single homeless people, develop homelessness prevention strategies, and ensure joint working across local government departments and with the voluntary sector. At the Witness seminar of December 2024, local homeless strategies were the aspect that delegates would most like to see reinstated. Rebecca Sycamore (Director of Regional Development Homeless Link 2000-2009) argued that their imporatnce was they obliged ‘local authorities to take a real hard look at their approach to homelessness…to use data, to talk to people who are experiencing homelessness…to talk to agencies’ and to develop ‘really local [solutions] ‘focused on the specific conditions and the specific needs of a local area’.
The new focus on homelessness prevention in the act was also vital. Steve McKinley (YMCA/De Paul Trust 2003-2014) notes that, ‘if you fund homelessness you get homelessness, you fund prevention you get prevention’. Prior to the passing of the act, local authority housing departments had not seen prevention as a core part of their remit, but under the aegis of the ‘Housing Options’ approach, it became a central part of their work. Previously, local authorities had engaged with private landlords principally to secure temporary accommodation, but DCLG guidance now required them to focus ‘on negotiation and conflict resolution’ to prevent evictions. From a very low base, by 2005, 87% of local authorities had rent deposit schemes, and 77.8% had tenancy support schemes aimed at prevention. For young people, services such as mediation, crash-beds and supported lodgings were rapidly developed. The Housing Options approach was controversial, with critics arguing that it could potentially undermine applicants’ rights to social housing and given and could be employed as a ‘new form of gatekeeping’.
The re-application of the Housing Options approach has limitations in the contemporary context, particularly in areas of high demand. At the December 2025 Witness Seminar, Neil Morland (Stoke City Council 2003-7) pointed out that with Local Housing Allowance (LHA) levels set at the 30th percentile since 2011, access to private rented accommodation was unviable. LHAs should be reset at the 50th percentile (the median cost of housing in a local area). It would also be inapplicable for the much-increased number of those with ‘no recourse to public funds’.
A vital component of the successful implementation of local homeless strategies was Labour’s determination that they would not be merely paper documents. This was achieved by close monitoring, detailed advice, the introduction of a Best Value Performance Indicator for homelessness prevention, and the allocation of additional funds for high performing local authorities. Reflecting on his own experiences with Stoke City Council, Neil Morland thought the introduction of a ‘tackling homelessness’ theme for the ‘Beacon Council award’ strongly incentivised investment in local homeless strategies and prevention.
The Supporting People (SP) programme 2003
The Supporting People programme from 2003 was perhaps Labour’s most important homelessness intervention. It injected £1.6 billion annually into the voluntary homeless sector. SP funding was ‘needs-led’ and local authorities worked in partnership with voluntary sector agencies to submit bids that met the full recovery costs of their services. Equally importantly, the funds were ‘ring-fenced’ and could only be spent on homelessness prevention, resettlement and support and could not be leached away into the wider council budget. By making service-user consultation compulsory for the award of SP contracts, it also enabled the development of services that empowered homeless people.
There were flaws in the SP funding model. It did not follow a national audit of need but was built upon existing services, meaning that areas with limited prior provision missed out. There were problems in two-tier authorities where funding went to the top tier but homelessness was the responsibility of the lower (district) tier, and in the capital, agencies providing London-wide provision now had to deal with thirty-two separate boroughs. The total cost of SP had come as a shock to the treasury, and the total SP pot was cut in each successive year. In 2009 the ring-fence was removed (in England – in Wales it was retained), and once local authorities had discretion over the allocation of grants the money was often diverted away into other council services. This was hugely impactful on the sector, resulting in the closure or scaling back of many projects after 2010. The removal of the ring-fence was described by Mike Barrett (CEO Porchlight 1999- 2024) as ‘the biggest mistake in homelessness prevention and homeless support this country has ever seen’.
An equivalent of SP funding should be resurrected, with a new funding stream following a national audit of need, it should be awarded at the district level in two-tier authorities, and adjustments for London-wide provision made. Most importantly, any future funds must be ring-fenced.
NPM and the delivery of Supporting People
The mechanism for dispersing SP funds led to profound changes in the composition, capacity, and size of homeless voluntary sector organisations. Typical of New Labour’s approach to governance, New Public Management (NPM) techniques were employed, with SP funds distributed by local authority commissioning teams on the basis of competitive tendering for contracts and with a requirement to achieve specific measurable outcomes.
Historically homelessness service provision has been located almost entirely within the voluntary sector, and that this sector was highly diverse in 1997, containing pockets of high-quality services, but where an amateur ethos remained predominant, and much of the provision was of a very poor quality. Pre-1997 the sector had been subject to only minimal degrees of monitoring, control and accountability.
As a consequence of the historic legacy of poor services, the more developed parts of the sector welcomed the new mechanisms of monitoring and accountability, seeing them as both long overdue, and vitally necessary to improve the quality of services. The imposition of competitive tendering for contracts, the negative effects of which have been well documented, was also probably beneficial for the homelessness sector. Dom Williamson (Homeless Link/DCLG 2004-2009) stresses the importance of the power to ‘take the contract of people who are doing it badly’, as it ‘drove the bad providers out of the sector’. It resulted in sector rationalisation, with many smaller agencies forced to merge or go out of business. The net result was the development of larger, national agencies and regional centres of excellence although at the expense of diversity. SP funding enabled these larger agencies to professionalise, and the need to meet the new quality standards obliged them to invest staff training, and make significant improvements in the quality of management and supervision which had hitherto been woefully neglected.
The homelessness sector experienced the well-documented problems of the application NPM techniques: onerous volumes of paperwork, management time diverted to tendering and re-tendering for contracts, excessive and inappropriate targets, a negative effect on cooperation between agencies, and some loss of autonomy. Practitioners, whilst acknowledging the overall benefits of improved accountability, were unconvinced that NPM techniques were the best way of delivering the programme.
The new role for local authorities as commissioners of homeless services proved challenging. Responsible since 2002 for formulating local homelessness strategies it was logical to give local authorities oversight and control of the disbursement of funds. Problems arose because many commissioners had little understanding of homelessness, often commissioned on cost only not quality, and by seeking to reduce the number of funding recipients in the interests of administrative ease, often failed to protect the ‘eco-system’ of local providers.
Local authorities should retain their central role in the distribution of a future equivalent to SP, but extensive training in the skills of commissioning, as distinct from procurement, must be given. In addition, homeless services should be commissioned from a broad range of service providers, perhaps bidding jointly as consortia, as has become the practice in some local authority areas such as Plymouth.
User-Empowerment
User-empowerment grew due to the obligation to consult and engage with service-users in SP contracts, which in turn improved aided long-term resettlement away from the streets. User-engagement adoption by the sector had been very slow and working practice remined highly paternalistic before 2003. Athol Halle (Director of lived-experience-led agency Groundswell 1999-2017) believes that it was only as a consequence Labour’s SP requirements that homeless agencies engaged with client involvement, ‘not because they believed in it…we were banging on the door and then they opened it when they had to, not before’. Over the remainder of the decade, service-user engagement gradually became mainstream practice. . Schemes to train homeless people in homeless work were established in many agencies, and by 2005, 30% of Porchlight in Canterbury’s staff were ex-homeless people. Changing Lives in Newcastle followed a similar trajectory, with its Director, Stephen Bell, describing it as ‘the best thing we ever did’. It is abundantly clear that Groundswell’s adage that ‘homeless people are not part of the problem, but part of the solution’, must be central to all future policy-making.
‘Places of Change’ – Trauma-informed care and Psychologically Informed Environments
New Labour’s homelessness programme had a cumulative effect. By the latter part of the decade better run organisations, enacting user-engagement and with more highly skilled staff, were able to take on board new understandings of the relationship between trauma and homelessness to develop much more effective forms of support and rehabilitation.
A weak point had been the grim institutional settings in which homeless work was performed, and this was addressed by the ‘Places of Change’ hostel capital improvement programme from 2006. Robin Johnson, who developed the concept of Psychologically Informed Environments (PIEs) at Labour’s National Social Inclusion Programme, describes them as ‘enabling environments’ where ‘people feel emotionally safe’. Whilst Johnson is clear that PIES are ‘not a place’ but an ethos, they required the kind of spaces that Places of Change funding made possible. Good examples are the rebuilds of Snow Hill hostel in Birmingham, and St Mungo’s Endell Street. Places of Change funding also required applicants to demonstrate how they would involve service users, provide training and employment and favoured the development of social enterprises. Snow Hill hostel created a basket weaving coffin making business, a cupcake bakery and a coffee shop, all run by people with lived experience of homelessness.
New trauma-informed approaches were pioneered by psychologist Nick Maguire. This acknowledged childhood trauma as both a cause, and aggravating factor, in homelessness and argued resettlement would likely fail without addressing underlying trauma. Implementing such an approach required highly-skilled staff. The most developed sections of the homeless sector, such as St Basil’s in Birmingham, have embedded a trauma-informed approach into their practice, but successive funding cuts since 2010 have hampered their ability to maintain and expand their work. Investment in the capacity of these organisations at a level that enables them to pay competitive salaries to conduct this skilled work is the essential.
Conclusion
It is clear that there is a great deal of New Labour’s homelessness programme that was highly effective and could easily be reintroduced in the contemporary context with adjustments made in the light of the lessons learned between 1999 and 2010. To rapidly reduce the number of rough sleepers, Keir Starmer must firstly make it clear that the issue is a prime ministerial priority – only this will afford sufficient weight to ensure a sustained commitment by government and ensure the necessary cross-departmental co-operation. A cross-cutting body, akin to the RSU, must be established, it should be led by experts from the homeless sector and be awarded with the necessary authority and resources to achieve its aims. To achieve permanent resettlement away from the streets, a funding stream equivalent to Supporting People must be re-established, ring-fenced for homeless support and prevention, and administered by local authorities working in partnership with voluntary homeless sector providers. However, it is important to note that not all of New Labour’s programme will be applicable today. Hostels are now regarded with some suspicion, and programmes of ‘Housing First’ are now considered the best approach. This may be so, although in the immediate task of getting people off the streets hostels still have a place, and most certainly should be designed and run like those developed under Places of Change. Local government finance is now much more highly stressed, and although a ring-fenced grant would circumvent this, central government funding is also more tightly constrained. But funding is always tight, and the moral imperative to act remains as strong as ever. Most significantly, perhaps, is the changing demographics of rough sleeping with a much higher percentage of people sleeping rough today being non-British nationals, with many having no recourse to public funds. Different solutions will be required, but much expertise still exists in the sector which, at its best, has a remarkable capacity for adaptation and innovation and will undoubtedly be able to deliver if given the necessary resources.
