Systems-wide evaluation of homelessness and rough sleeping: preliminary findings

Question 1

Foreword

Accepted Answer

This is the first report of the Systems-Wide Evaluation of Homelessness and Rough Sleeping Services. The aim of this innovative evaluation is to identify opportunities to improve the way that the homelessness and rough sleeping system works to ensure it is delivering the best possible outcomes for those who need to make use of services and value for money for the taxpayer.

This first phase of research is focussed on the role of MHCLG’s core homelessness and rough sleeping programmes and how they are affected by the wider the system. The evidence presented in this report combines findings from systems thinking analysis and fieldwork in five local authority areas. The report is clear that the system is not working as intended and Local authorities are facing rising demand from people who have reached crisis point. This report highlights the need for a much greater focus on prevention and how disparate funding schemes are contributing to inefficiencies in the system.

We would like to thank the Centre for Homelessness Impact and their partners for their expertise and continuing hard work to deliver this ambitious programme of research, local authority staff and other stakeholders who participated in the research, and the analysts at MHCLG who provided input to the research materials and reviewed the outputs.

Most importantly, we are hugely grateful to the service users who participated for giving us their time and sharing their experiences with the research team.

MHCLG is committed to continuing to develop its evidence base on the causes of and solutions to homelessness and rough sleeping. This report adds to a range of recent evaluations that MHCLG has published on this topic including a process evaluation of the Homelessness Reduction Act and a range of evidence on our Housing First pilots and Changing Futures programme.

Stephen Aldridge

Director for Analysis and Data & Chief Economist

Ministry of Housing, Communities and Local Government

Question 2

Executive summary

Accepted Answer

This report provides an overview of the initial findings from the first phase of a systems-wide evaluation into homelessness and rough sleeping. It includes evidence collected during the first six months of the evaluation and offers early insights into some of the research questions. In addition to outlining the preliminary findings, the report contains reflections from the evaluation team on the current approach and suggests ways to refine it to enhance learning in the following phases of the evaluation.

The report has been written by the Centre for Homelessness Impact and its partners for this evaluation, RSM and Cordis Bright.

Scope and purpose

Homelessness research in the UK has tended to be concerned with qualitative and conceptual analyses, often focusing on the experiences, perceptions and priorities of people who experience homelessness, with a relative lack of robust evaluative research on the effectiveness of interventions. While recent years have seen a growth of rigorous studies of effectiveness of specific programmes and interventions designed to tackle homelessness, it still lags behind other policy domains in terms of evidence of the effectiveness of funded programmes and interventions. Importantly, there is a shortage of robust evidence on the effective structural and systemic actions that may be needed to tackle the drivers of homelessness and rough sleeping in the UK.

To contribute to filling the evidence gap around the system-wide drivers of homelessness and rough sleeping, this evaluation was commissioned to the Centre for Homelessness Impact and consortium partners by the Ministry of Housing, Communities and Local Government (MHCLG) in October 2023 and marks the first time a government, either in the UK or internationally, has set out to understand the systemic impact of its policies and interventions in relation to homelessness and rough sleeping. In addition, the research brief included a requirement to undertake an evaluation of the operation and impact of three of the system’s main interventions: the Rough Sleeping Initiative, Homelessness Prevention Grant and the Rough Sleeping Accommodation Programme.

This evaluation is as challenging as it is groundbreaking, in particular because of the tight timescales and the complexity of the requirements. But the research is timely: as the report shows, homelessness and rough sleeping numbers have risen dramatically in recent years and increased levels of financial insecurity and the shortage of affordable housing has made it more important than ever to improve our collective understanding of what is driving homelessness and rough sleeping, what works within the system currently, and what doesn’t, and how impact and value for money might be maximised.

Despite an increasing focus on systems thinking in recent years, holistically evaluating complex policy areas like homelessness and rough sleeping remains relatively new. Within this research, therefore, frameworks and methods more closely linked with the public health system have been utilised to support the work.^{[footnote 1]}

This first phase of the research concluded in June 2024, prior to the July 2024 general election. It is retrospective and therefore does not reflect either the approach of the new government or any policy developments since the election. The evaluation is ongoing, and further reports will be published as the work progresses.

Background

Homelessness and rough sleeping policy in England has evolved significantly over the past twenty years. Efforts to both prevent and tackle homelessness and rough sleeping have led to various legislative changes and the adoption of several specific policies and programmes. This has included the Rough Sleeping Initiative, which began in 2018 with the aim of reducing rough sleeping through targeted support and funding given to the majority of local authorities in England, the “Everyone In” initiative, launched to provide emergency housing to people homeless during the COVID-19 pandemic, and the introduction of the Homelessness Reduction Act 2017 which aims to improve support for those at risk of homelessness, requiring local authorities to provide proactive and preventative assistance.

While there is no universal or consistent definition of homelessness, it is generally understood to encompass individuals and families living in unsuitable or inadequate housing, as well as those who are sleeping rough. In England, homelessness is defined legally as when “a person or a household does not have accommodation that is available for them to occupy, that they have a legal right to occupy and that is reasonable for them to continue to occupy”.^{[footnote 2]} This includes people experiencing rough sleeping, people living in hostels and refuges, and hidden homelessness such as sofa surfing. Homelessness in England has risen in all of the forms for which there are reliable statistics in recent years.

Most research divides the causes of homelessness and rough sleeping into individual and structural factors. Common individual factors include relationship breakdown, domestic abuse, trauma, mental ill health, and substance use, while structural factors which influence homelessness include experiences of poverty, particularly in childhood, financial insecurity linked to unemployment and the operation of the benefits system, housing supply and affordability, asylum and migration, and the impact of time spent in public institutions such as prison or the care system. Addressing these system-wide challenges requires a multi-faceted approach, involving collaboration between central, regional and local government, homelessness charities, landlords, and a wide range of other stakeholders.

Research methods

A mixed-method approach was employed to meet the aims of the research, combining the use of qualitative and quantitative data from primary and secondary sources within five local authority areas. Fieldwork in local authorities was undertaken to understand how the homelessness and rough sleeping system functions on the ground in different contexts. The selected local authority areas were Bournemouth, Christchurch and Poole, Herefordshire, Manchester, Southend-on-Sea, and Westminster. The areas were chosen from a group that expressed an interest in participating in the research and were selected to represent various different types and sizes of local authority: covering both urban and rural, unitary, district and metropolitan boroughs.

Engagement in five local authority areas, with council representatives in a range of housing-related roles and with their delivery partners, was primarily aimed at supporting the evaluation of the process and impact of the three specific programmes and interventions set out in the research brief: the Rough Sleeping Initiative, Homelessness Prevention Grant, and the Rough Sleeping Accommodation Programme. However, the research also sought to determine how the delivery of the programmes was influenced by system-wide factors.

To support understanding of system-wide factors, a detailed systems map of the homelessness and rough sleeping system was developed by reviewing existing maps, conducting workshops with experts, and refining the map based on feedback. The map shows the number and strength of the connections between the different parts of the system as a means of demonstrating where to focus and where interventions may have the most impact.

The evaluation also included a review of 30 major homelessness and rough sleeping policies and programmes funded by various government departments and other funders using a methodology developed from studies by Donella Meadows called the Intervention Level Framework.^{[footnote 3]} This framework assesses policies and interventions on how deeply and at what point they act to influence the system.

Interviews were undertaken with strategic stakeholders from six government departments, including MHCLG, and further interviews and focus groups were conducted with local-level representatives from within the five selected local authorities and their delivery partners as well as with individuals with lived experience of homelessness and rough sleeping.

Findings were triangulated using multiple research methods and were largely consistent across the different local areas, which increases our confidence in their robustness. However, the sample of local authorities that participated in the research is small and findings should not therefore be taken to be representative of all local areas.

Summary of findings

The findings of this phase of the research have been derived through combining the systems mapping insights, the analysis undertaken using the Interventions Level Framework, the analysis of primary and secondary data, and the interviews and focus groups conducted with people with lived experience, strategic governmental representatives, and those working within the selected local authority areas either for the councils or for their delivery partners.

National and local stakeholders highlighted how economic conditions in the UK have contributed to severe strains on the homelessness and rough sleeping system, with a number of factors highlighted as combining to create significant pressures and barriers to tackling the problem more effectively, including rising rents and the reducing value of welfare benefits, an overall shortage of affordable housing, rapidly rising housing demand, particularly from families facing cost of living issues and from new refugees experiencing homelessness, and the limited availability of wider specialist support services, most notably, mental health support.

The systems mapping work highlighted the extent to which financial insecurity was both a driver and result of homelessness and rough sleeping and determined the importance of people at risk of homelessness and rough sleeping being able to access effective services to meet their wide-ranging needs. In particular, support to address mental health needs.

Stakeholders were clear that these wider structural pressures have affected the extent to which programmes like the Rough Sleeping Initiative and funding streams like Homelessness Prevention Grant have been able to reduce the number of people experiencing rough sleeping or homelessness for the first time, as local authorities have had to respond to the rise in demand.

The majority of government activity and expenditure is on crisis relief, rather than prevention and delivered at a level which does not create system-level improvements

The policy review conducted using the Intervention Level Framework illustrated that there is an emphasis within existing government-funded policy/programmes on crisis relief and the provision of services rather than on longer-term prevention, or on tackling the causes of homelessness at a system-level. This was underlined by the local authority survey and interviews which illustrated how pressures on services means that funding allocated to homelessness and rough sleeping is typically short term and focused on responding to immediate need. Under the Rough Sleeping Initiative, for example, it was determined that across the five areas participating, less than 1% of the funds provided were spent on prevention. This restricts both local authorities’ ability to plan into the longer term and to implement longer-term or system-level prevention activities.

There are significant competing policy priorities within national government

It was acknowledged consistently by research participants, including those in central government, that while homelessness and rough sleeping is a complex policy area requiring strong cross-departmental collaboration, conflicting departmental priorities, and a lack of cross-governmental understanding, had hindered efforts to tackle homelessness and rough sleeping. Interviewees stated that departments were often focused on the specifics of their own departmental agendas and limited in the resources they are willing to allocate, or the policy focus they are willing to give, to developing opportunities for cross-departmental collaboration to prevent homelessness and rough sleeping. Some areas of positive cross-government working were identified by those interviewed at a national and local level, providing something upon which to build.

There is some good evidence of strong local partnership working on homelessness and rough sleeping but there is room for improvement

It was reported by research participants that effective partnership working resulted in more efficient use of resources, reduced duplication of services, and enhanced communication, decision-making, and coordination among stakeholders. While most local authorities and delivery partners in the five areas sampled reported positive examples of partnership working, there was a consensus that further integration of services and funding could enhance the effectiveness of their work to tackle and prevent homelessness and rough sleeping. In particular, it was suggested that bringing more of the relevant services together, including health, housing, social care, police and crime, and pooling funding, would support more integrated partnership working, fostering a collaborative approach to tackling homelessness where everyone has a clear stake in achieving the desired outcomes. The importance of accessible, multi-faceted services for those in housing need was supported by both the system mapping and the testimony of those with lived experience.

Despite rising demand and numbers, the main homelessness and rough sleeping programmes were viewed as making a positive impact

The preliminary findings of the programme evaluations suggest that the programmes were successful in increasing the overall number of people who were able to access accommodation and support to prevent or resolve their homelessness or rough sleeping.

The Rough Sleeping Accommodation Programme (RSAP) provides move-on homes and accompanying support to help people who are sleeping rough find long term accommodation solutions. The programme was considered to have delivered well against its outcomes in the local authority areas interviewed, with the provision of good quality, wraparound support for those accommodated a key success factor.

The Rough Sleeping Initiative (RSI) provides councils with funding to support those who are rough sleeping or who are at risk of rough sleeping. Local councils and their delivery partners reported that the programme had been effective in contributing to reducing the number of people experiencing rough sleeping, largely due to the availability and success of off-the-street accommodation and outreach services. However, the programme has not led to an overall reduction in the number of people sleeping rough due to increases in the flow of people into the rough sleeping system. Reducing rough sleeping overall was felt to be beyond the reach of the programme, given its lack of focus on prevention, and the much wider structural factors driving higher levels of new rough sleeping.

The Homelessness Prevention Grant (HPG) is a flexible source of funding to support local authorities to prevent and tackle homelessness and rough sleeping. This phase of the evaluation demonstrated that while local authorities are making good use of Homelessness Prevention Grant (HPG) to intervene critically to meet their statutory duty to offer temporary accommodation to those who are homeless and in priority need, a very limited share of the funding is spent on prevention activity occurring before the 56 days provided for within the Homelessness Reduction Act or on improving the operation of the local system, as perhaps was envisaged. This was linked clearly to the challenges associated with rising levels of demand from those who are already homeless.

The partnerships developed as a result of collaborating with other partners on delivery of the programmes, and the funding included for non-housing related wraparound support, enabled beneficiaries to also access support they may not have otherwise been able to access.

Measuring success of MHCLG funded programmes

The evaluation has also highlighted that local authorities, at times, used funding from the three homelessness and rough sleeping programmes interchangeably – and in combination with other sources of funding. This was considered positive by research participants as it allowed funding to be tailored to the needs of local people, supported a degree of cross-cutting working, and enabled local authorities to respond to surges of demand where necessary. However, it also contributed to the challenge of measuring the success of the programmes, given that programme beneficiaries were supported by services funded across all three programmes and with other sources of funding mixed in.

Unintended consequences

During the course of this phase of the evaluation, some issues came up that could best be described as unintended consequences of the activity of the programmes or the way in which the homelessness and rough sleeping system currently operates. These included a concern on the part of one local authority that the absence of a preventative offer for single people in housing need (but not yet roofless) might be creating a perverse incentive for some people to sleep rough as a means of accessing support with their housing needs. Two local authorities also told us that the high quality of Rough Sleeping Accommodation Programme properties made some beneficiaries reluctant to move to other housing options, typically in the private rented sector, which were of lower quality, thus creating move-on difficulties. Further, local authorities and their delivery partners highlighted that the increased funding for homelessness and rough sleeping given to local authorities at the same time (ie: at the start of a new programme cycle) has led to recruitment difficulties due to multiple councils or providers seeking to recruit to the same type of job roles at the same time.

Conclusions

The conclusions of this phase of the research are set out below and will be built on during further phases of the evaluation. Findings were triangulated using multiple research methods and were largely consistent across the different local areas, which increases our confidence in their robustness. However, the sample of local authorities that participated in the research is small and findings should not therefore be taken to be representative of all local areas.

The importance of a whole-government approach

There was clear agreement that there would be value in the collaborative development and implementation of a cross-government strategy focused on ending homelessness in its widest sense, including but not limited to rough sleeping. The findings outlined in this report suggest it would be important to both take a more ‘upstream’ public health-type preventative approach to tackling this problem, which would involve looking at drivers of homelessness and rough sleeping and seeking to intervene before households reach a crisis point, and to have collective ownership of that strategy across government. There is a need for a clear leadership role that can operate across departments to deliver on this. This would include agreeing a clear set of shared outcomes, navigating and coordinating competing priorities, looking for policy gaps and conflicts, and driving a cohesive cross-departmental approach.

Preventative programmes, activities or funding streams sitting within other government departments which are used to alleviate or prevent homelessness and/or are acting as drivers of homelessness could play a bigger system role in supporting homelessness prevention.

Building confidence in taking a preventative approach

To effectively move interventions and policies upstream, all partners within the system need to be more confident in understanding where to target prevention efforts as evidence and practice in this space is limited. In addition, there is a need for more structural support in terms of funding, focus, policy and partnerships to enable a better focus on prevention in particular, prevention that occurs outside of the 56-day prevention duty provided for in the Homelessness Reduction Act. The lack of resources makes prevention challenging and pressure on services leads to a gatekeeping approach which can be ineffective if many of those most in need of help are kept out of services and those most able to navigate the system get help. Using data and assessments to identify those most at risk and understanding within the system the places where preventative support can be most effectively targeted at those most at risk of homelessness could be a key starting point for any new approach to developing a whole-government approach to tackling homelessness and rough sleeping.

Programme and funding consolidation

Consolidating funding and programmes is likely to improve effectiveness and flexibility, avoiding both unhelpful silos and fragmentation of efforts. This approach has the potential to help local areas to enjoy some flexibility to meet urgent needs and support improved monitoring and understanding of impact, by avoiding duplication and mapping activities to desired outcomes. Within any combined funding stream, it would be important to safeguard funding for specific activities, such as homelessness prevention, and expectations regarding this should be made clear when allocating the funding.

There was limited evidence of local authorities pooling funding across local authority boundaries and given the economies of scale that may be possible, the mobility of the client groups being served and the issues around how geographical footprints across health, probation, housing are sometimes misaligned, consideration could be given to ways of incentivising joint commissioning within the allocation of funding. Combined Authorities offer one possibility for this, as do existing local authority sub-regions and other groupings.

Better evidence of what works

If the standalone homelessness and rough sleeping programmes remain in place, it could be beneficial to review the current processes for designing objectives per programme and in particular how these are being used to inform reporting on progress. There could be benefit in reviewing the current metrics captured and mapping these to programme objectives. It was noted from the evaluation that the focus of the metrics was on outputs rather than outcomes.

There also appears to be a need to address the lack of impact data against the programme terms and conditions. It is recognised that evidencing outcomes may require multiple partners to be involved and contributing to these. A sensible first step might be to review metrics that could support central government to monitor the performance of the homelessness and rough sleeping system as a whole and hold all stakeholders to account.

Given the system-level interactions and overlaps, care should be taken to ensure that any wider outcomes framework monitors the system-level drivers of homelessness and rough sleeping that have been determined during this first phase of the research so that the issues about housing affordability are considered alongside drivers which may sit outside of MHCLG’s remit, like financial insecurity, the operation of the benefits system, unemployment, limited access to specialist support services (especially mental health), and institutional discharge (e.g. from asylum support accommodation, prison, care).

Better local partnership working

It could be beneficial to explore opportunities for how the government can encourage more effective local partnership working. This could cover setting out examples of how partnership working has been effective in addressing homelessness and what contributed to this happening. An effective approach might be facilitating opportunities (eg: webinars) for partners to share their experiences of what works well in their own context.

These considerations for the future are proposed based on preliminary findings and will require testing with national and local stakeholders in future fieldwork in support of this continuing evaluation.

This Report

The following report provides an overview of the initial findings from the first phase of the systems-wide evaluation. It includes evidence collected during the first six months of the evaluation and offers early insights into some of the research questions.

The research undertaken was completed prior to the general election in July 2024. It is important to note that the findings and conclusions reflected within this report do not represent any policy developments within central government since that date.

Chapter 1: Introduction and Methodology

Chapter 1 defines homelessness and rough sleeping and provides some contextual data on the current pressures within the homelessness and rough sleeping system. It goes on to describe the range of factors, both structural and individual, that are understood to be drivers of homelessness and rough sleeping and explains the need for a systems-wide approach to the evaluation. It also outlines the objectives of the evaluation and the research methodology.

Chapter 2: The system affecting homelessness

Chapter 2 describes the work that has been done to map the system and understand the causal connections between different elements. It highlights the feedback of people with lived experience on the drivers, barriers and enablers that they experienced on their journey through the homelessness and rough sleeping system. The need for a systems-wide approach to prevent and tackle homelessness and rough sleeping is further explored using feedback from fieldwork with local authorities, their delivery partners and government departments. It includes the outcome of a review of current interventions and programmes using the Intervention Level Framework, describing where in the system interventions currently operate and what the optimum distribution might be.

Chapter 3: The role of three programmes in the homelessness and rough sleeping system

Chapter 3 sets out findings from an evaluation of three main programmes and initiatives funded by government to prevent and tackle homelessness and rough sleeping: the Rough Sleeping Initiative, Homelessness Prevention Grant and the Rough Sleeping Accommodation Programme. It provides a description and rationale for each of the programmes and sets out the operating context in each of the five local authority areas where the evaluation took place. This chapter also highlights the system factors that might inhibit or support the achievement of intended outcomes and sets out what worked well and less well about each of the programmes to support future learning.

Chapter 4: Discussion, conclusion and next steps

Chapter 4 presents the conclusions that can be drawn from the systems-wide and programme evaluation activity undertaken to date and sets out some next steps in terms of considerations for the future.

Appendix 1 Detailed methodology

Appendix 1 provides a more detailed description of the activities and methodology including: a description of the system mapping process; an explanation of how the Intervention Level Framework was used to classify homelessness and rough sleeping programmes and initiatives; details of the qualitative research undertaken with government departments, local authorities and their delivery partners and people with lived experience of homelessness and rough sleeping; and, information about the quantitative data collection and analysis that has been undertaken to date.

Appendix 2 Intervention Level Framework

Appendix 2 lists the 30 programmes that were reviewed using the Intervention Level Framework and shows them categorised according to the type and level of intervention.

Question 3

Chapter 1. Introduction and methodology

Accepted Answer

1.1 Overview

This chapter provides a definition of homelessness and rough sleeping and offers some contextual data around the current pressures within the homelessness and rough sleeping system. It also highlights the range of factors, both structural and individual, that are understood to be drivers of homelessness and rough sleeping and describes why the complexity and changing nature of homelessness and rough sleeping requires a systems-wide approach to the evaluation. It outlines the objectives of the evaluation and the research methodology and activities being used to achieve these evaluation objectives. This includes system mapping, the categorisation of existing government programmes and interventions using the Intervention level Framework and the gathering of data and qualitative insights from a wide range of stakeholders, including government departments, selected local authorities and their delivery partners, and people with lived experience. Limitations of the methodology are also included in this chapter.

1.2 Homelessness and rough sleeping in England

There is no universal or consistent definition of homelessness. It is generally understood to encompass individuals and families living in unsuitable or inadequate housing, as well as those who are sleeping rough. In England, homelessness is defined legally as when “a person or a household does not have accommodation that is available for them to occupy, that they have a legal right to occupy and that is reasonable for them to continue to occupy”.^{[footnote 4]} This includes people experiencing street homelessness (referred to as sleeping rough), people living in hostels, refuges, and hidden homelessness such as sofa surfing.

Official homelessness statistics show that homelessness in England has risen in all of the forms for which there are reliable statistics in recent years.

The number of households in temporary accommodation has increased, rising from fewer than 49,000 in March 2011 to around 117,000 households by 31 March 2024. This figure is higher than at any other point since the data series began in 1998.

The number of people estimated to be sleeping rough on a single night rose from 1,768 in 2010 to 4,751 in 2017. Thereafter, the figures began to decrease slowly (by 2% in 2018 and 9% in 2019) until a more rapid decline in 2020 and 2021, when more than 33,000 people were helped to find accommodation under the “Everyone In” initiative during the COVID-19 pandemic.^{[footnote 5]} Since 2022, rough sleeping levels have been climbing with the latest single night figure at the time of writing (2023) standing at 3,898 – similar to the level of rough sleeping recorded in 2015.

Though research often divides the causes of homelessness into individual and structural factors, it is also acknowledged that there can be a complex inter-relationship between these factors. Common individual factors include relationship breakdown, domestic abuse, trauma, mental ill health, and substance use, while structural factors which influence homelessness include experiences of poverty, particularly in childhood, housing supply and affordability, unemployment, and leaving public institutions such as prison or having been in care.

Homelessness research in the UK has tended to be concerned with qualitative and conceptual analyses, often focusing on the experiences, perceptions and priorities of people who are experiencing homelessness, with a relative lack of robust evaluative research on the effectiveness of interventions. While recent years have seen a growth of rigorous studies of effectiveness of specific programmes and interventions designed to tackle homelessness, there is a shortage of robust evidence on effective structural actions to reshape the underpinning factors driving the homelessness and rough sleeping system in the UK.

This has important implications for policy and practice questions, which typically require mixed methods approaches that attain both breadth and depth of understanding.

1.3 Systems approach to the evaluation

Many different factors may lead to people experiencing homelessness or sleeping rough. The particular ways in which a person experiences homelessness; actions to reduce the likelihood of it happening in the first place; responses to it at individual, neighbourhood, local authority and national levels; and the multiple interactions between all these factors, combine to represent a complex system.

The systems-wide evaluation has been commissioned to the Centre for Homelessness Impact and consortium partners to look across the homelessness and rough sleeping system as a whole. This report summarises findings from the first phase of research conducted within the evaluation which examined the key drivers of homelessness and rough sleeping and the role of the core homelessness and rough sleeping programmes funded by the Ministry for Housing, Communities and Local Government (MHCLG). Chapter 2 sets out our working definition of the system affecting homelessness and rough sleeping and the results of our initial system mapping, informed by the first phase of research. Our understanding of the system and how it operates will change and improve as evidence is gathered through the evaluation. In subsequent phases of the evaluation the focus will turn to the influence of policy areas outside of MHCLG’s remit including the criminal justice, immigration, welfare, health and social care systems.

The homelessness and rough sleeping system is adapting and changing all the time, reacting to legislative and policy changes, to the presence or absence of interventions, and to what is happening in related systems or policy areas. This evaluation is attempting to identify the ways in which the homelessness and rough sleeping system might have been affected or impacted not just by changes made in the immediate policy sphere of housing and homelessness (ie: new programmes, changing funding levels, housing market conditions), but by wider systemic and policy issues. The intention is to consider how other systems might interact with the homelessness and rough sleeping system to support, undermine or inhibit intended outcomes and generate both desirable and undesirable consequences. A systems-wide evaluation aims to achieve this by stepping back and looking at the big picture. This involves considering outcomes and interventions operating at different levels of the homelessness and rough sleeping system and looking for impacts generated from outside that system.^{[footnote 6]}

To that end, it should be noted that many of the wider factors that affect homelessness and rough sleeping sit outside the direct remit of the Department, and will require action by other government departments, arms-length bodies, and a range of agencies. There are, for example, multiple interactions between the criminal justice system, asylum and migration system, benefits system, provision of drug and alcohol treatment, and wider physical and mental health care services.

Effective action to prevent homelessness and rough sleeping requires concerted and integrated action across all these systems and services, and many more, as per a ‘public health approach.’ In this context, the need is for prevention activity to take place at a whole population level (‘universal prevention’) or to be focused on those individuals and families likeliest to be at risk of homelessness or rough sleeping before they reach a crisis point (‘targeted prevention’) rather than being something that occurs once there is a statutory duty or a household is in imminent risk (‘crisis prevention’).

One of the core functions of the systems-wide evaluation is to go beyond describing these factors, and the interactions between them, to identify a range of potential levers that could be activated across the various systems to drive improvements in homelessness and rough sleeping outcomes over the short, medium and long term.

1.4 The objectives of the evaluation

The objectives for the systems-wide evaluation were established by MHCLG in October 2023 as part of the evaluation commission. The main objectives are to:

Objective 1: Gain insight into the most effective organisation, distribution and balance of central government funding and the effect on commissioning, strategic decisions, and service delivery at local level.

Objective 2: To provide a better understanding of how users enter, interact, and move through homelessness and rough sleeping services, barriers and enablers and establish what components deliver an effective and efficient system at local level.

Objective 3: To measure causal chains in the Theory of Change and the homelessness and rough sleeping service map using a common set of metrics to inform this programme of evaluation.

Objective 4: To better understand and make recommendations on ‘what works’ and for who, why and in what context to deliver outcomes and Value for Money in the homelessness and rough sleeping system.

Objective 5: To better understand the interactions between central government policies and their overall contribution to impact and to identify the most effective areas of intervention through a complexity-sensitive systems approach.

These objectives will, in combination, meet the overall objective of the systems-wide evaluation to support the UK government to design and deliver interventions that will impact positively to reduce homelessness and rough sleeping in England.

The objectives have been translated into research questions to meet the needs of the evaluation.

This report focuses on supporting objectives 1, 2 and 5 with a particular focus on the three standalone programmes/initiatives that were reviewed: the Rough Sleeping Initiative (RSI), Homelessness Prevention Grant (HPG) and the Rough Sleeping Accommodation Programme (RSAP).

1.5 Overview of the methodology

The research used a variety of different methods to produce a range of data types that complement each other, as is best practice in a system-wide evaluation. To inform the preliminary results described in this evaluation report, qualitative and quantitative data was collected from primary and secondary sources. This enabled the evaluation to combine existing information with new data to fill the gaps, elicit additional themes and ultimately develop a coherent narrative. Such a mixed method approach was particularly valuable to triangulate insights from fragmented or divergent sources of data from different parts of the system.

Primary data collection methods comprised:

workshops to inform system mapping (including representatives from MHCLG, other government departments, local authorities, expert advisors and representatives from the homelessness sector);
interviews and focus groups with strategic stakeholders from MHCLG and other government departments;
interviews and focus groups with local area representatives (both local authority staff and their delivery partners);
interviews and focus groups with people with lived experience;
a survey in five chosen local authority areas.

In addition, secondary data analysis of existing publicly available data was undertaken to support insights emerging from qualitative data.

Data collection approaches that focused on local areas and their stakeholders covered five local authorities across England. These were Bournemouth, Christchurch and Poole; Herefordshire; Manchester; Southend-on-Sea and Westminster. The areas were chosen from a group that expressed an interest in participating in the research and were selected to represent various different types and sizes of local authority: covering both urban and rural, unitary, district and metropolitan boroughs.

At the core of the methodology used to give a systems perspective is the development of a detailed systems map of the homelessness and rough sleeping system. To understand how the system currently works, the research team created an interactive systems map.^{[footnote 7]} The map will evolve throughout the evaluation as more evidence is gathered and our understanding of the system deepens. The systems map is being used as a tool to:

identify the root causes and interconnected factors contributing to homelessness;
visualise the connections between factors and develop an understanding of how different parts of the system influence each other;
identify ways that the system either reinforces or mitigates the factors contributing to homelessness, known as ‘feedback loops’;
inform the development of more effective policy interventions that are sensitive to how the system operates;
facilitate collaboration by providing a shared language and tool that will enable stakeholders to work together more effectively toward common goals.

System mapping is a method for examining the multiple ways in which programmes and interventions exert influence across the system. The map is therefore not an end in itself, but rather a tool for stimulating discussion, generating data, and encouraging systems thinking with regards to homelessness and rough sleeping.^{[footnote 8]}

A high-level review of all the major homelessness and rough sleeping policies, programmes and interventions being funded by the government was undertaken. There are clearly many potential approaches when deciding how to intervene in complex systems. Some needs require crisis prevention, such as providing temporary accommodation to those in priority need. At the other end of the spectrum, policies could focus on universal prevention, such as fundamental changes to macroeconomic policies (like reducing child poverty). Balancing these contrasting types of interventions - some immediate/crisis, some long term/preventative - is a common challenge in any complex system.^{[footnote 9]}

A true system approach to a complex societal challenge takes time, and requires coordinated, phased delivery across all levels of a system. A good example is the approach that has been taken by the government and others to reduce smoking prevalence over the last 50 - 70 years. This has combined actions at very different ‘levels’ of the system over an extended period. For example, early efforts were focused on interventions aimed at individual smokers such as pack warnings and health education. Over time, the focus shifted to more structural approaches, such as harm reduction strategies (for example, regulation on tar/nicotine content); legislation (for example, the indoor smoking ban to protect employees); and campaigns to influence social norms (encouraging positive feedback in the system). Finally, recent proposals to phase out smoking in society through successive age-related bans would represent a true ‘paradigm shift’ in the role of the state. The net effect of these combinations of actions has been to make a significant measurable impact on the prevalence of smoking.^{[footnote 10]}

Applying a method called the Intervention Level Framework, this research included a review of a total of 30 programmes and interventions funded either wholly or partially by the government (with one exception) to determine where in the system they are being used to intervene. One intervention, the Voluntary and Community Sector capacity grant, is split into four lots. Each of these have their own focus, operating at different levels of the system. Each lot was allocated its own Intervention Level Framework classification.

Interviews conducted virtually with strategic stakeholders across the system. These captured the perspectives of 12 individuals from across six government departments, including the Ministry of Justice, the Office of Health Improvement and Disparities, His Majesty’s Treasury, the Department for Work and Pensions, the Home Office and the Ministry of Housing, Communities and Local Government. In the case of the latter, interviews were conducted with seven representatives from across the homelessness and rough sleeping systems in five virtual interviews. Overall, the ambition was to gather their views and experiences of what is working well and less well for homelessness outcomes, and the interactions across the system.

Interviews with people with lived experience of the homelessness and rough sleeping system. Fieldwork was undertaken with 13 people who had lived experience of the homelessness and rough sleeping system. The people interviewed were from a variety of different backgrounds and were currently living in Bournemouth, Herefordshire, Southend-on-Sea or Westminster. It was not possible to undertake fieldwork with users of services in Manchester due to restrictions associated with local elections taking place at the time the research was undertaken. The interviews captured people’s journeys through the system, from becoming homeless to accessing housing, their experiences of the services they accessed, the barriers they faced and their touchpoints with other systems. In addition a focus group was undertaken with five members of the Centre for Homelessness Impact’s Lived Experience Network.

Engagement was undertaken across the five selected local authorities to evaluate the process and impact of Homelessness Prevention Grant, the Rough Sleeping Accommodation Programme and the Rough Sleeping Initiative. As well as evaluating the three programmes, focus was given to exploring the influence and interaction of the wider system in the delivery and effectiveness of the programmes.

Table 1.1 below provides an outline of the sample sizes for each of the primary sources of data, showing that 34 interviews and workshops were delivered, with a total of 85 people participating overall. The 15 programme-level interviews and the 3 programme-level focus groups also included discussion of system-level issues. The interviews and focus groups were supplemented with a survey completed by the five local authority areas involved in this phase.

Table 1.1: The sample sizes for primary data collection

Method/Programme	RSAP	RSI	HPG	System
Number of interviews with strategic stakeholders from MHCLG (no. of participants)	0	0	0	5 (7)
Number of interviews with strategic stakeholders from other government departments (no. of participants)	0	0	0	6 (12)
Number of interviews with local authorities and local delivery partners (no. of participants)	5 (9)	5 (18)	5 (13)	15 (40)
Number of focus groups with local area representatives (no. of participants)	3 (13)	3 (13)	3 (13)	3 (13)
Number of interviews with people with lived experience (no. of participants)	0	0	0	4 (8)
Panel interview with members of the Centre for Homelessness Impact’s Lived Experience Panel	0	0	0	1 (5)
Local authority survey (please note that while survey respondents were asked to provide evidence in support of responses this was not always provided)	5	5	5	5

In addition to the primary collection of both qualitative and quantitative data, secondary data was also collected in the form of strategy documents and programme data. This was complemented by secondary data from publicly available sources that was collated as part of the monitoring framework, such as the Homelessness Case Level Information Collection (H-CLIC) data on local authorities’ actions under the statutory homelessness legislation in England and the Ending Rough Sleeping Data Framework.^{[footnote 11]}

1.6 Limitations of the methodology

Local authorities participating in the research were selected to be diverse and representative in terms of geography, type and size of local authority to minimise any bias in the research. However, the fieldwork was based on a small sample of five local authorities and is not therefore representative of all areas. To ensure depth of coverage across the key issues within the homelessness and rough sleeping system, multiple interviews per local authority were conducted with different stakeholder groups, and multiple stakeholders were included in each interview (details as in Table 1.1).

The fieldwork was partially impacted by the timing of the research as it coincided with the pre-election period for the local and Mayoral elections in parts of country. This meant it was not possible to conduct an intended site visit and interviews with people with lived experience in Manchester. Overall, just eight people with lived experience were interviewed across the four local authority areas participating in the research and to strengthen this area of work, a focus group with five members of the Centre for Homelessness Impact’s Lived Experience Network was used to supplement insights.

Additionally, when scheduling group interviews, it was necessary to negotiate limited stakeholder availability to meet the study requirements. In some cases, this meant conducting joint interviews with providers and commissioners together, which could have impacted on the ability of both groups to discuss sensitive issues candidly in the presence of others. The sequencing of data collection, via the surveys being carried out, and interviews with local authorities did not always allow for data to be discussed in interviews.

There were inconsistencies in the availability and quality of quantitative data across local authorities, which limited our ability to report on their spending on specific activities and beneficiaries and precluded an assessment of value for money.

The Intervention Level Framework analysis considered just 30 of the programmes and interventions focused on homelessness and rough sleeping.

Similarly, there was limited information on specific targets for each programme (delivery plans or offer letters were not available) and the monitoring data (where available) was focused on programme outputs and not outcomes. Altogether, this did not allow us to assess the relative success of each programme at either the local or national level.

Question 4

Chapter 2. The system affecting homelessness

Accepted Answer

2.1 Overview

This chapter defines the homelessness and rough sleeping system and its boundaries and describes the process of mapping that system. An initial analysis of the systems map demonstrates the system’s complexity and the wide range of interconnected factors that cause, and are caused by, homelessness and rough sleeping.

An analysis of existing government initiatives and programmes has been undertaken by researchers and is described through use of the Intervention Level Framework. This methodology determines where in the system current interventions operate, what they are intended to achieve and whether they are impacting at a structural or lower level.

The chapter further explores the homelessness and rough sleeping system through the expert lens of people with lived experience and summarises feedback on the drivers of homelessness and rough sleeping, the barriers and enablers to exiting homelessness and rough sleeping and the experiences of navigating and receiving services.

The need for a systems-wide approach to tackling homelessness and rough sleeping was clearly demonstrated through the interviews with government departments, local authorities and delivery partners. The chapter sets out feedback from interviews and surveys, alongside the results of the system mapping, to highlight many of the system-wide factors that can enable or make more difficult the intended outcome of tackling homelessness and rough sleeping. The factors are explored in detail but include the importance of a sufficient supply of affordable housing (and some of the barriers to affordability that exist within the system currently), the challenge of meeting wider support needs beyond housing need, the importance of cross-government collaboration and long-term funding given the complexity of the system and some of the variations at a local level that may impact success.

An analysis of existing government initiatives and programmes has been undertaken by researchers and is described through use of the Intervention Level Framework. This methodology determines where in the system current interventions operate, what they are intended to achieve and whether they are impacting at a structural or lower level.

2.2 Defining the homelessness and rough sleeping system and its boundaries

Homelessness is an example of a ‘complex adaptive system’: an issue affected by and affecting multiple, connected factors that interact with each other and change or adapt over time. This contrasts with a ‘system’ that produces more predictable or linear consequences. Given this, there is a value in conceptualising homelessness and rough sleeping as products of a complex system. The homelessness system is influenced by national and international macro-economic policy, national and local housing and benefit policy and implementation, income and housing security, physical and mental health, relationship stability, public attitudes, investment in prevention and investments and quality of support services.

A key first step to any systems-focused evaluation is to decide what parts of the system will be evaluated.^{[footnote 12]}

The aim of this evaluation is to identify areas where the UK government could take action that would have a direct impact on homelessness and rough sleeping outcomes. The focus of the evaluation will therefore be restricted to elements of the system that are within the UK government’s sphere of influence and have a direct connection to homelessness and rough sleeping. This excludes external economic and political factors such as climate change that are not solely within the control of the UK government and where the impact of policy changes on homelessness and rough sleeping outcomes is harder to predict. The scope of this evaluation includes:

statutory homelessness and rough sleeping policies, programmes or interventions, funded by MHCLG and delivered by local authorities or commissioned providers;
wider housing-related policies, programmes and interventions that influence or are influenced by homelessness and rough sleeping, funded and/or led by MHCLG, local authorities, relevant statutory bodies and commissioned providers;
policies, programmes or interventions owned by other government departments which influence or are influenced by homelessness and rough sleeping, such as health, asylum and migration, welfare, education and justice, and delivered by local authorities or commissioned providers;
policies, programmes or interventions undertaken by or relating to non-statutory homelessness and rough sleeping bodies, such as voluntary, community and faith sector organisations.

2.3 Initial analysis of the system map

The systems map has been used to explore wider systems issues and unintended consequences of programme activities with local authority staff. It will act as a continuously developing reference point for the systems-wide evaluation, both informing and being informed by other elements of the evaluation, allowing us to situate insights from these elements of the evaluation within the wider homelessness and rough sleeping system. This will support the evaluation team to use these insights to identify the most effective places to intervene in the system to achieve the highest impact.

The core system map for this evaluation was developed through three key stages:

1. Review of existing Centre for Homelessness Impact (CHI) homelessness system map and identification of adaptations required to create a new core system map aligning with the scope of the evaluation;

2. Initial system mapping workshop gathering insight from homelessness and rough sleeping experts to inform adaptations to the CHI map, and subsequent development of a draft core system map;

3. Sense-testing of the draft core system map, through a workshop with homelessness and rough sleeping experts and written feedback, and subsequent revisions and finalisation of the core system map.

In our current systems map, the factors with the highest number of connections across the system are mental health and wellbeing, financial resilience and the importance of effective services. While the map is not primarily a quantitative tool, these highly linked factors have the potential to impact more areas of the system than less connected factors. The three most connected factors on the system map are shown in the online version and described below:

Mental health and wellbeing

Poor mental health can increase the risk of someone experiencing homelessness or sleeping rough and be worsened by these experiences. The system map shows links to issues such as social isolation; access to healthcare; physical health; anti-social behaviour and lack of financial security. The importance of this issue is reinforced by interview data, for instance in Manchester, Herefordshire and Southend, local mental health teams are reporting increasing proportions of people experiencing homelessness or rough sleeping presenting with severe mental health issues and multiple needs.

Financial resilience

Financial resilience is connected to issues such as employment; ability to access and maintain a tenancy; ability to meet housing/living costs; mental health and wellbeing; and stable employment. Interviews with local authority officers in Manchester, Herefordshire, Southend and Bournemouth, Christchurch and Poole also highlighted the importance of financial security, with stakeholders stating that the cost-of-living crisis has rapidly increased housing costs and detrimentally affected affordability. For those people relying on Local Housing Allowance (LHA), the fact that LHA levels have not kept pace with rising rents has led to incredibly tight housing markets across the country, with local authorities seeing growing numbers of rent arrears and evictions as people are priced out of their accommodation. Higher numbers of evictions from the private rented sector are driving increasing levels of homelessness and placements into Temporary Accommodation; therefore the combination of both individuals’ and local authorities’ lack of financial resilience are exacerbating one another.

The importance of effective services

Effective services have the potential to positively impact many of the causes of homelessness and rough sleeping, such as by improving mental health and wellbeing; physical health; employability; and educational engagement. It is also important that services are accessible to ensure that people engage with the support they need. A key finding emerging from the interviews with national and regional stakeholders and from the systems mapping workshops is that the lack of engagement that results from inaccessible services is a key factor contributing to someone’s likelihood of becoming or remaining homeless. Access to effective services can help prevent people from experiencing homelessness, and support people who are at the point of crisis to access and maintain accommodation.

Interview and workshop participants suggested that the accessibility and effectiveness of services is determined by factors such as their use of evidence, the extent to which they collaborate, the leadership they offer, and the inclusion of people with lived experience in service planning. For example, Southend-on-Sea City Council reported good working relationships with local homelessness charities. They explained that this increased engagement as service users tend to have more trust in community organisations. The City Council felt they also benefited from the additional services provided by community organisations.

The lack of long-term funding, and the overall inability of commissioners to uprate funding in line with inflation, were both reported by local authorities and their delivery partners as having a destabilising effect on service provision.

Central government stakeholders also felt that better collaboration, particularly between central government departments, would improve the effectiveness of services.

This initial analysis shows domains that could benefit from further exploration in the next phases of the evaluation, for example on specific drivers of homelessness and rough sleeping such as poor mental health, or on specific segments of the population such as refugees and asylum seekers.

2.4 An analysis of government homelessness and rough sleeping interventions

2.4.1 Introduction

To understand what types of interventions currently exist within the homelessness and rough sleeping system, researchers conducted a review of 30 programmes that are fully or partially funded by Government (with one exception) and intended to tackle homelessness or rough sleeping. This included 29 standalone programmes and the Voluntary and Community Sector Capacity Grant which involved four programmes or ‘lots’, some of which were operating at different levels of the system. We therefore reviewed each ‘lot’ individually.

The Intervention Level Framework was used to categorise programmes according to what level of the system they operate at. The results were used to identify gaps in the system where increased focus is needed. The Intervention Level Framework is an approach grounded in systems thinking, which aims to provide a deeper and more cohesive understanding of the most effective ways to address complex problems. It has been used in other policy settings like climate change and obesity.^{[footnote 13]} ^{[footnote 14]}

Table 2.1 provides a description of each of the intervention levels within the Interventions Level Framework along with examples of interventions from the homelessness and rough sleeping system that have been given to illustrate each category.

Table 2.1 An explanation of the type of Intervention Level Framework category with example

Intervention Level Framework category	Homelessness/rough sleeping example
Paradigm (or deep-seated mindset): this describes the system’s deepest held beliefs giving rise to the system’s goals, rules and structures. Intervening at this level can produce the most significant impact.	None currently. An example might be setting out fundamental changes to economic and social policies to ensure that no-one ever became homeless by reason of poverty (eg: introducing universal basic income).
Goals (high level, able to influence the system at highest level): interventions at this level can change the aim or mindset of the system.	Housing First, as it prioritises the provision of stable and permanent housing without requiring individuals to first address support needs.
System structure (interventions, flows, connectivity, trust): these elements make up the system as a whole. Interventions at this level can change the system structure and create effect at a population level.	Changing Futures, as it sets out to change linkages across a range of supportive services and subsystems with the intention of shifting the behaviour of the system as a whole.
Feedback and delays (self-regulation, adaptability, reinforcement); this allows the system to regulate itself by acting on feedback.	The Rough Sleeping Indicator Framework provides information about outcomes with different cohorts of people, allows benchmarking and identification of weaknesses in performance.
Structural elements (sub-systems, actors, operating parameters): interventions affect specific parts of the system. Many actions at this level are needed to create system change.	Accommodation for Ex-Offenders programme, as it only addresses the sub-system of prison discharge and only targets a specific population group (prison leavers).

There is no agreed optimum combination of types of system actions, but it is widely argued in the available literature that system change is more likely when there are actions taken across all levels of the system, with some effort being taken to tackle the wider determinants of the outcome in question, and the way that the system operates. There are several examples from different policy domains which support this.

One relates to the introduction of the soft drinks industry levy in the UK in 2018; this more structural, population-level attempt at reducing sugar intake contrasts with campaigns exhorting individuals to consistently reduce their sugar consumption. Less efficient systems tend to have a greater proportion of lower-level interventions affecting specific subsystems, actors or elements of the system, which may create competition between groups/subsystems or demand high levels of individual agency, leading to unintended detrimental consequences for some groups/subsystems.^{[footnote 15]} In this context, while a wide range of policy interventions had been implemented, there was a much greater emphasis on diet than on physical activity, with governments also mostly tending to prioritize information and capacity building in their obesity strategies rather than directly shaping the choices available to individuals through population-level fiscal and regulatory measures.

When homelessness and rough sleeping programmes and interventions are looked at, there is a greater proportion of lower-level interventions affecting specific subsystems, actors or elements of the system, such as prison discharge in the case of the Accommodation for Ex-Offenders programme which is targeted specifically at people leaving prison. Policies at this level are less likely to succeed than deeper, structural ones, because they may create competition between groups/subsystems, or they demand high levels of individual agency, leading to unintended detrimental consequences for some groups/subsystems, particularly the most disadvantaged. These findings analysed alongside other data; help point towards ways that systems change could happen as laid out below in section 2.7.

It should be noted that this pilot analysis only included a limited number of policies/programmes. In addition, a limitation of the Intervention Level Framework is that its five categories are unlikely to be sufficiently nuanced to precisely divide interventions. The tool is a distillation of Donella Meadows’ ‘Places to Intervene in a system’ which described 12 levels of a system. In either case though, the categorisation can be subjective. It remains, however, a useful tool for showing how programmes and interventions are distributed and how the homelessness and rough sleeping system has more individual-level, responsive interventions, and few that would lead to structural, preventative changes.

The understanding from the Intervention Level Framework is supported by the homelessness and rough sleeping systems-wide Theory of Change. The Theory of Change and how it informs programme objectives is explained in Chapter 3.

More detail on the operation of the Intervention Level Framework and the full list of interventions reviewed, and their classification, are available in Appendices 1 and 2.

2.4.2 Results of interventions review

The interventions review determined (Figure 2.1) that the majority of programmes (18/30; 60%) were classified as ‘structural elements’ in that they (as described above) are directed at specific people or groups within the system. They typically involve the direct delivery of services to people who are already homeless. Examples (full list appears in Appendix 2) include the Local Authority Housing Fund, which aims to increase the stock of housing within the homelessness and rough sleeping system, the Rough Sleeping Drug and Alcohol Treatment Grant which focuses on the drug and alcohol treatment subsystem within the wider homelessness and rough sleeping system and the Accommodation for Ex-Offenders programme and Community Accommodation Service Tier 3, both of which are directed specifically at prison leavers.

While useful in isolation, these programmes operate at a level within the system that means they are responding to homelessness and rough sleeping rather than seeking to prevent it in a systemic way. There are no examples in the list of policies and programmes analysed that demonstrate attempts to shift the paradigm around homelessness and rough sleeping, meaning that the capacity for existing policies, programmes and interventions to achieve meaningful system level change is limited.

There were four programmes that were classified as providing ‘Feedback and Delays’ as per the definitions set out in the framework. These included the Housing Advisors Programme, a team engaged to share learning across the homelessness and rough sleeping system, and the Home Office Homelessness Escalations Service, which supports case escalation for non-UK nationals who are homeless. The programmes classified in this section are mainly concerned with information provision. Identifying and working on feedback loops can be an important part of a systems approach as they enable interventions to have a more lasting and far-reaching impact.

Six programmes were thought to represent changes to ‘System Structures’. These included Capital Letters, which is seeking to make system-level changes by connecting London boroughs, investors and landlords to make the housing market work better for people experiencing homelessness, and Changing Futures, which aims to change linkages across a range of support services and subsystems with the intention of shifting the behaviour of the homelessness and rough sleeping system as a whole.

Two of the programmes, while not quite operating at the level which could be said to represent a paradigm shift within the homelessness and rough sleeping system, were certainly aimed at changing the goals of the system. The Housing First pilots aimed to change the aim of the system by re-defining who is deemed eligible for housing and the Employment Covenant funding (Department for Work and Pensions) sets a new target across the system to create a more accessible employment landscape for people experiencing homelessness.

There are few programmes that build on possible feedback loops within the homelessness and rough sleeping system. Those that have been classified as such are mainly concerned with information provision. Figure 2.1 shows the distribution of homelessness and rough sleeping programmes and interventions according to the categories within the Intervention Level Framework.

Figure 2.1 The major homelessness and rough sleeping programmes and interventions categorised according to the Interventions Level Framework

In addition to classifying the 33 programmes using the Intervention Level Framework, they were also categorised using the focus of the funding programmes (Figure 2.2). This analysis demonstrated that there are only five programmes offering ‘preventative’ support of any kind, none of which are aimed at changing the paradigm or goals – which are the intervention levels most likely to have large scale, system-wide impact - and only three of those operating at the middle ‘system structure’ level, can be said to be preventative. Overall, the majority of programmes (28/33; 85%) involve offering accommodation or support.

Figure 2.2: The major homelessness and rough sleeping programmes and interventions categorised by Intervention Level Framework and the focus of the funding

This Intervention Level Framework analysis suggests that, whilst there are a wide variety of programmes addressing a diverse set of needs, programmes and funding for homelessness and rough sleeping is piecemeal, fragmented, and dominated by interventions that target a specific cohort or subsystem. Additionally, the majority of interventions tend to focus on the symptoms, rather than the causes of homelessness and/or rough sleeping.

There are few interventions that seek to address or improve the connections between subsystems. The Homelessness Prevention Grant, which is a flexible source of funding to support local authorities to prevent and tackle homelessness, is intended to operate at a system-level. However, our evaluation of the core MHCLG programmes found that in practice this grant is increasingly being spent on interventions for those who face an immediate risk of homelessness or are already at the point of crisis due to rising demand. Analysis of Homelessness Prevention Grant and MHCLG’s other core programmes is presented in Chapter 3.

2.5 The experiences of people with lived experience

Interviews with people who had lived experience of the homelessness and rough sleeping system were conducted to understand drivers of homelessness and rough sleeping, their engagement with and journeys through the system (and other related systems) and the barriers/enablers for them in exiting homelessness and rough sleeping.

The purpose of the interviews was not to develop an exhaustive list of drivers but to identify common themes in personal experiences of interacting with the homelessness and rough sleeping system. The 8 interviews were supplemented with a group discussion with 5 members of the Centre for Homelessness Impact’s Lived Experience network. The headlines can be summarised as follows:

2.5.1 Lack of preventative support

People with lived experience participating in the panel discussion and in the focus groups reported that it was common for people who ended up rough sleeping to have had a period of insecure housing/hidden homelessness beforehand (for example, living in a van, sofa surfing) and could not access the support they needed during this time to prevent homelessness before being asked/forced to leave. The absence of preventative support, while they were in housing need but not yet roofless, meant rough sleeping was not avoided.

Many of those participating in the discussions also reported that their first experience of homelessness often led them to find a temporary living arrangement with friends or family. However, this was usually a short-term arrangement with significant levels of uncertainty and reliance on the goodwill of others. After a length of time, interviewees found themselves at risk of rough sleeping as “[their] connections ran out”. It was common for friends and family to eventually ask them to leave because of limited space, lack of privacy, or financial concerns. In another instance, an interviewee described themselves as hidden homeless, as they were living out of their van until difficulties with drug and alcohol use necessitated the sale of their van, resulting in the participant having to sleep rough:

I’ve been living in the back of my van and then I had to sell that. I had a drink driving [offence]. I couldn’t afford insurance.

2.5.2 Challenges with mental health

One major theme within the interviews revolved around the impact of mental health, echoing findings from the systems mapping. People with experience of homelessness identified existing mental health conditions as contributing factors in becoming homeless, noting they had often been exacerbated by current life stressors or discrete adverse life events. In fact, some service users identified traumatic life experiences as the primary reason for homelessness. One user, for example, described a bereavement precipitating a deterioration in their mental health.

Basically, [my experience of homelessness] started with the loss of my twins… [and led to a series of events and circumstances as a result of which I struggled with] drug addiction, broke down a relationship, lost a job, moved from a different town to [current location]. Things didn’t work out [the way] I was hoping for it to work out.

At the same time, another reported that they had been the victim of modern slavery and exposed to multiple traumatic events. The psychological consequences of these events were reported as impacting their social and occupational functioning.

2.5.3 Challenges with substance use

The majority of those interviewed also disclosed alcohol and substance use difficulties, which they reported resulted in loss of employment and financial instability.

Then, it just became too much. I was doing it every day…I lost my job because I couldn’t…properly [work] because of my drug dependency and became homeless.

There were also reports of engaging in drug-related crime, with one interview reporting a period of incarceration following a deterioration in their personal circumstances. They reflected:

I committed a crime [at a] time of my life [when I had] lost my job…lost my home.

2.5.4 Impact of financial insecurity

Another common theme among interviewees revolved around the cost-of-living crisis in the UK. When faced with rising rents and limited alternative accommodation options, one service user reported that they left London to move to their current local area to “start a new chapter in their life.” However, upon their arrival, they were unable to afford the accommodation and subsequently slept rough on the street.

I used up all my money staying at a guest house and stuff like that, and I had to sleep rough.

Another interviewee reported facing a significant monthly rental increase in their private accommodation over two consecutive years, first by 17% and then again by 50%, rendering the non-working single parent at risk of rough sleeping. Council intervention mitigated the risk and prevented homelessness in the first year but could not prevent the second round of increased rental costs for the service user. The interviewee stated:

The first time I experienced homelessness was when my landlord put the rent up. Because I’m a single parent and I’m not working, and you know I was struggling because obviously I had to look after the home, groceries, child expenses. I had to ring the council up, and they asked me to fill in a homelessness application form, then the caseworker got back to me, and agreed with my landlord, and went ahead with a top up DHP payment, which was OK, and then that was for just a year. And then my landlord put the rent up again. [it was] then I experienced homelessness - that second time - and I couldn’t afford to pay that much rent.

2.5.5 Routes into stable housing

When considering the factors that supported their access and entry into stable housing, those interviewed highlighted the importance of the local authority and community organisations, wider systems and outreach efforts. Though it should be noted that there is likely to be a bias towards people who have accessed housing services due to the method of selecting interviewees.

The most common route into stable housing was reported as being facilitated by the local authority or community providers of homelessness and rough sleeping services. For example, an interviewee who had recently completed a prison sentence described how their parole officer provided details of a local charitable housing foundation and arranged for a meeting.

Maybe I was lucky because it’s my probation officer. Because of the crime I committed, I was on probation for two years. And he knew about [accommodation] and he invited a representative to tell me about it. I knew nothing if it hadn’t been for that.

Interviewees described reaching out to their local authority and receiving support towards their rent or getting help with mediation with their landlord. Others reported accessing support with substance use and general health problems from their GP, noting that this interaction with the medical professional resulted in a referral to a social enterprise supporting individuals with alcohol and substance-related difficulties and mental health concerns. That charity, in turn, facilitated their access to night shelters and referred the participant to the local authority housing solutions team. This culminated in the participant being placed into stable accommodation where they currently reside.

Beyond the homelessness service provided by the local authority, the most common services that service users interacted with were local food banks. Several interviewees acknowledged that they currently, or in the past, relied on food banks to ensure access to a minimal level of food and drink. However, this reliance was accompanied by an expressed ambivalence in some cases, as the interviewees did not enjoy the idea of depending on them.

But I wouldn’t want to [depend on foodbank services] because there are some people out there needing all that much more, [and] I wouldn’t take anything from foodbank if I don’t need it.

Two individuals commented that they were now claiming benefits which they wouldn’t have done before they received support from the local authority. One person specifically mentioned support from the Citizens Advice Bureau to access welfare benefits.

2.5.6 Barriers to accessing stable housing

Interviewees indicated significant barriers to reaching out for help as people experience homelessness. With serious mental health difficulties and substance use playing critical roles in facilitating and maintaining homelessness, many of the interviewees noted from their personal experience a reluctance or inability for people to reach out for help straight away, often reaching a breaking point before feeling able to access support:

Oh, [I was] just caught up in my own thing, you know, drinking on the drugs and all that…but you know the depression got to me so much and …paranoia…[I was] schizophrenic…I was hallucinating from the drugs and [things were] just bad.

Where interviewees did report asking for help, a common barrier identified consisted of a lack of knowledge or understanding of where to go and who to contact. Even with the knowledge of what organisations to contact, the interviewees reported struggling to achieve consistent and fruitful communication with the right person.

Interviewees also suggested that the actual process of gaining stable housing was often difficult and protracted, as they struggled with completing the forms required, and received little assistance:

Yes, I did have a bit of a problem filling out some bits and uploading it and all that, and then I had to ring back again, speak to them and you know, they explained it to me again and then I had to go back. So, I did have a bit of a problem.

2.5.7 Engagement with local authority homelessness services

Many interviewees had a positive experience of local authority homelessness and rough sleeping services in their respective local areas. For instance, one interviewee felt that the council was very supportive and their engagement “always worked really well…no barriers, no challenges.” They reported that the council had provided all the support that was needed in securing stable housing, commenting on the positive experience of having everything explained and being given assistance with the administration needed.

Another interviewee shared a similar positive experience and was satisfied with their interactions with the council. They appreciated being able to work with the same caseworker, which was an efficient and consistent way of working, noting:

it was very quick. It was amazing…I just called the council. Then they said they [going to] call me. I spoke to [case worker]. She sent me an e-mail, and everything was very quick. Yeah. The process was very, very quick. Amazing.

However, not all interviewees had positive experiences and reported that local authorities could do more for people experiencing homelessness. They highlighted a lack of knowledge of who to contact to gain access to vital services and raise safeguarding issues for themselves or others:

The other day I saw a young girl sitting with a blanket over basically begging. To me, she [feels] so vulnerable. But who do I phone? Even though I’ve got some real high reasonable knowledge, that is something not [commonly known].

One particular interviewee expressed a deep-seated pessimism about how services are commissioned. The interviewee believed the policies were in place to win over local voters in short-term elections rather than fix the structural issues in the homelessness system.

The politicians can…say, well, we’re commissioning something. It’s to be seen to be doing something so they can get re-elected at the next election …. the governments haven’t been interested in thinking long term, particularly with housing.

2.5.8 Factors that might support improved homelessness and rough sleeping outcomes

Finally, the people with lived experience that were interviewed shared their reflections about factors that might improve their experience of the homelessness and rough sleeping system, suggesting:

that co-production would be valuable, envisioning a system designed with the input of those who have experienced homelessness, which would ensure that services are accessible and appropriate;
the need for effective communication so that the right communication channels are used, and clear and consistent messages are shared with people experiencing homelessness. Relatedly, they also discussed the importance of inter-organisation communication so that services are coordinated, information is shared and work conducted in tandem, with appropriate data sharing safeguards;
that education and awareness may help to address the stigma associated with homelessness and rough sleeping. It was particularly perceived in the case of political leaders, who, according to the interviewees, were not placing enough importance on the issue to tackle it.

2.6 The need for a systems approach

Interviews with local authorities, delivery partners and relevant officials within MHCLG and a range of other government departments (Home Office, Ministry of Justice, Department for Work and Pensions, Office for Health Improvement and Disparities and His Majesty’s Treasury) have been combined to derive insights into the importance of a systems approach to tackling homelessness and rough sleeping.

The insights have been considered alongside those derived from the system mapping to strengthen understanding of where, and in what way, a new systems approach may need to impact. It should be noted that, while there was consistency while there was consistency in views of the drivers, interviewees were not asked to evidence their assertions with supporting data. The insights have been grouped thematically as follows:

2.6.1 Housing Supply

Limited housing stock, issues with affordability and suitability of existing stock, local authority funding, and housing demand were frequently perceived to be key drivers of homelessness raised by central and local government interviewees and their delivery partners.

Housing stock and affordability. Alongside inflation and interest rates rises, the view was that high land and house building costs were also contributing to the shortage of affordable accommodation. The shortage of affordable rented accommodation was perceived to be particularly acute in town and city centres, something backed up by rental market reports.^{[footnote 16]} The impact being that local authorities are often forced to accommodate people out of their local area. Additionally, the lack of council owned stock increases reliance on expensive temporary accommodation.

Anecdotally, interviewees also reported that landlords were nervous about the legislative changes that were being proposed under the Renters (Reform) Bill, banning ‘no fault’ evictions and requiring them to increase property checks. Some interviewees expressed a concern that landlords would sell their rental properties as a result, with landlords remaining in the market able to charge higher rents due to the reducing supply.

Unsuitability of available housing. It was noted that the accommodation that is available can be unsuitable for people’s needs. For example, households experiencing homelessness are often housed in B&Bs and places with shared facilities, which are not always appropriate. In particular, it was expressed that there is inadequate accommodation for specific groups of people who may have more specialist needs when it comes to accommodation, such as those experiencing domestic abuse, people who identify as LGBTQ+, and disabled people. Living in unsuitable accommodation was often felt to contribute to increasing people’s support needs.

Lack of supported accommodation. Cuts to local authority budgets were said to mean there was a lack of funding to commission sufficient supported housing, meaning there has been a reduction in the availability of this type of accommodation. Interviewees also stated that some of the supported housing being offered was not of sufficiently good quality with the effect that some people reportedly preferred to sleep rough or stay in other forms of insecure housing such as sofa surfing, rather than live in the supported housing they are offered. Conversely, it was also noted that where high quality supported housing is provided, its quality could sometimes disincentivise people moving on when support is no longer needed.

Funding shortage. Funding for local authorities was considered insufficient to respond to housing demand and support needs. Local Housing Allowance rates have not kept pace with rental costs, further limiting local authorities’ ability to secure suitable housing and leading to people staying longer in unsuitable and expensive Temporary Accommodation.

Housing demand. Whilst housing is becoming less affordable, the number of people needing assistance to access affordable housing is increasing. Again, this is in part a result of the cost-of-living crisis, with high levels of inflation and increasing rental costs, resulting in increasing numbers of people, including families, being evicted from their homes due to rent arrears. An increase in refugees, particularly from Afghanistan, Syria and Ukraine, has also heightened housing demand, further straining local authority resources. Given the level of demand, local authorities are unable to house everyone and are forced to prioritise the people they house.

Negative public perceptions. Interviewees noted that they believed the housing of people experiencing homelessness and rough sleeping is further complicated by negative public perceptions of people who experience homelessness and/or receive benefits. The examples interviewees gave included that many private landlords appear to be reluctant to accept tenants paying with Universal Credit or Housing Benefit, or to house people who had previously experienced rough sleeping. People in neighbouring properties are also thought to be reluctant to live next to people who have experienced homelessness, though no evidence of this was provided. There is also significant stigma against people involved with the justice system, and housing for people leaving prison sometimes requires approval by the Police, restricting options for their placement on release from prison.

2.6.2 Wider support needs

In addition to a lack of suitable housing, interviewees also felt that the wider system is currently unable to respond sufficiently to the wider support needs of people who are either at risk of or already sleeping rough. The key issues identified by central and local government stakeholders and their delivery partners were:

that many people experiencing homelessness are unable to access primary care, including GP services, as many erroneously require patients to provide a fixed address to show they are within catchment areas;
that funding pressures on local authorities, leading to cuts in staffing and support services, such as substance use and advice services, is another reason that the system fails to meet the demand for support, noting that this could lead to increased demand for homelessness services;
that there was a lack of mental health support which was noted to limit a person’s ability to secure and maintain a tenancy. Interviewees felt that the limited availability of mental health support and lack of integrated care pathways for dual diagnoses of mental health and substance use needs further exacerbate these issues.

There were also concerns expressed by interviewees in all five areas that support services are regularly overwhelmed by the range of needs that people who are experiencing homelessness may present with. They stressed the importance of wraparound support that responds to wider support needs alongside the provision of accommodation, but noted that services are often unable to provide it due to a) a lack of understanding of people’s needs and the support required to respond to them; b) limited availability of services due to high demand and c) lack of coordination between services, limiting ability to provide a cohesive, holistic support package that is easy for people with complex needs to engage with.

In particular, it was noted by local authorities that services’ ability to work with and maintain engagement with people was affected by a lack of understanding of trauma and the impact this has on people’s behaviour.

There is a poor understanding in public policy of the impact of trauma on the way people think, feel and behave.

2.6.3 Focus on short term funding limiting sustainable outcomes

Local authority officers and their delivery partners also highlighted the fact that funding allocated to responding to homelessness and rough sleeping has typically been short term (on two-to-three-year cycles) and focused on responding to immediate needs. This was viewed as restricting local authorities’ ability to plan for the longer term and to implement sustainable initiatives, such as the development of new supported accommodation, or to take preventative measures that reduce the risk of people becoming homeless in the future. The focus on short term interventions and outcomes and their immediate costs and benefits means there has typically been little support for those more sustainable interventions where impacts may be larger but take longer to be achieved. Interviewees felt that, without addressing the underlying causes of homelessness and implementing preventative measures, interventions will remain reactive and limited in their capacity to produce long term change.

Delivery partners suggested more broadly that increasing the availability of housing for vulnerable groups (for example, for refugees, for those with a dependency, or those who have been in prison) has not garnered much public support and therefore has not previously captured much political attention. They felt that the resulting outcome had been a lack of political impetus around reducing homelessness that was present in other policy areas.

2.6.4 Limitations to central government cross-departmental collaboration

The causes and consequences of homelessness and rough sleeping are interrelated and as such have relevance across government, requiring strong cross-departmental collaboration to tackle the issue.

Civil servants felt that, whilst significant progress has been made, structural barriers to cross-departmental collaboration remain especially where departmental priorities are in conflict. They also identified a lack of cross-governmental understanding of homelessness and rough sleeping policy and available data. Specifically, interviewees reported:

Conflicting departmental priorities. Interviewees reported that departmental priorities had often conflicted with one another, which prevented positive collaboration and caused negative implications for programmes seeking to address homelessness and rough sleeping. An example often cited related to the Home Office prioritising clearing the backlog of asylum applications at a pace and in a way that had contributed to both homelessness and rough sleeping as there was simply not enough time (or available housing) for local authorities to act to prevent homelessness for people newly granted refugee status who have been asked to leave their asylum support accommodation.

Another example commonly offered related to welfare policies and processes that created financial insecurity, contributed to debt and arrears that drive tenancy failure and made both move on and homelessness prevention more difficult. As well as some policies being directly in conflict with the aim to prevent homelessness, it was noted that sometimes tackling homelessness was simply less of a priority for other departments than it was for MHCLG. Interviewees reported that this was particularly true for the minority of people sleeping rough because this group accounts for such a small proportion of demand on services that other government departments are responsible for.

Resource pressures. Amid resource and funding pressures, civil servants told us that departments have been focused on remaining within the scope of their own agendas and have been limited in the resources they have been able to allocate to developing opportunities for cross-departmental collaboration.

Departments get tunnel vision about what is within and outside of the scope of massively busy agendas, particularly if there isn’t more funding for something.

A lack of understanding of other departmental remits and responsibilities. Interviewees reported a general lack of understanding across departments of one another’s remits, and subsequent misconceptions about one another’s responsibilities. For example, staff were somewhat unsure about aspects of the system, structures, and responsibilities of departments outside of their own. Improved shared understanding between departments would support more meaningful collaboration on homelessness and health. Stakeholders also suggested that government colleagues may be apprehensive about working on issues outside of their usual remit, due to risk of feeling exposed engaging with issues they feel they do not fully understand.

Data sharing across government. The data on homelessness and rough sleeping outside of MHCLG was described as fairly limited, hindering the collective ability of government to build a comprehensive picture of the population, their health, housing and support needs, and their interactions with non-homelessness services. This in turn limits government’s ability to understand how departments should collaborate to plan and deliver appropriate and comprehensive support. There is also a lack of clarity from within other departments on how homelessness and rough sleeping funding is spent and how outcomes from this spending are measured. Interviewees reported a need for more openness on these issues from MHCLG to increase efficiency in resource allocation.

The lack of information and clarity around outcomes and measures means it feels like a bit of a grey area. It feels like there isn’t the information we need to be able to jump on to something or attach our own work to it.

Need for clearer leadership and collaboration across government. There was a view expressed among interviewees that MHCLG should assume leadership for promoting wider work to tackle homelessness across all relevant government departments, navigating and coordinating competing priorities, and driving a cohesive evidence-led cross-departmental strategy for ending both homelessness and rough sleeping. It was stated by local authorities and their delivery partners that any such strategy should also involve significant collaboration with local authorities in its development and implementation. As the department responsible for local government, interviewees suggested that MHCLG is in a unique position to initiate this collaboration and support other government departments to collaborate with local authorities, as they take forward a strategy for ending homelessness.

2.6.5 Variation at local authority level

Local authorities have discretion to interpret and apply central government guidance. This means that their approach to tackling homelessness and rough sleeping varies. Central government stakeholders highlighted the tension between the benefits of allowing local authorities the flexibility to tailor service to local need, and the challenges this presents for consistent monitoring and reporting, and therefore central oversight of homelessness and rough sleeping.

We know the greater the flexibility, the greater the ability to tailor to local needs. But this also reduces our level of oversight. At the end of a spending period, we need to know, ‘did this deliver what it was supposed to deliver’. If we’re allowed complete flexibility, it’s very hard to know what it was supposed to deliver.

Local authority autonomy in applying homelessness and rough sleeping guidance can also complicate collaborative working across agencies. Interviewees noted this as a particular issue within the Duty to Refer process.^{[footnote 17]} For example, MHCLG developed a template referral form for the Duty to Refer process, but some local authorities have developed their own mechanisms for processing referrals, which other public bodies cannot always access. In some cases, to process a referral, local authorities request certain information, which referring bodies are not always able to provide due to data protection. Furthermore, local authorities are not obliged to update referring bodies on the progress of the referral.

2.6.6 Misalignment between local authority and health services

Central and local government stakeholders agreed that local authority housing services and health systems are misaligned. For example, in some places, Integrated Care Boards operate over different geographical areas than local authorities and are typically larger. This reduces their capacity to engage with all local authorities within their boundaries and complicates partnership working between the two.

Interviewees also told us that differing priorities can be a barrier to collaboration as health providers, and their commissioners, tend to focus on things like the immediate need for housing for those ready for discharge from hospital, rather than thinking about long-term health and housing needs.

2.6.7 Local Authority funding and support

Civil servants agreed with local authorities that funding should be awarded over a longer term to support long term planning and more sustainable responses to preventing and tackling homelessness and rough sleeping. They felt that this would improve the deliverability of services and maximise value from providers. Local authority officers also requested they be given more notice of funding allocations to aid effective service planning. Further upstream, they suggested that this could be supported by considering addressing multiple policy areas and needs concurrently during spending reviews.

Despite the flexibility of programmes such as the Homelessness Prevention Grant and the Rough Sleeping Initiative, local authority officers still felt that some of the funding for homelessness and rough sleeping is fragmented and overly restricted to a defined set of interventions and services, which may not be the most effective in meeting local needs or preventing homelessness. Local authorities would like to be able to use funding even more flexibly and across programmes to adapt interventions, services and approaches to reducing homelessness and rough sleeping based on what is or is not working.

However, local authority officers still saw a need for central government to support via frameworks and tools to help them to use funding effectively and measure what is working. This could include support with developing tools to better plan cross-sector interventions, such as theories of change for programmes involving multiple services.

2.6.8 Positive examples of cross-departmental collaboration

Whilst there is significant work to be done, civil servants, local authorities and delivery partners all noted several areas of positive collaboration in tackling homelessness and rough sleeping that may be built upon to increase the effectiveness of government work to reduce homelessness and rough sleeping.

Several interviewees cited the Ending Rough Sleeping for Good strategy as having contributed somewhat to improved collaboration across government departments and in fostering a sense of shared responsibility among ministers.

The Ending Rough Sleeping for Good strategy shows that clear vision drives action. Ministers feel accountable to central government and can hold others across government to account.

The Ending Rough Sleeping strategy board, which brought together departments to focus on homelessness and rough sleeping was also noted as important, providing the opportunity to focus on different cohorts of people experiencing homelessness and rough sleeping, including care leavers, ex-offenders, non-UK nationals, veterans, and families, and keeping outcomes for these groups on the agenda.

The strong working relationships reported between MHCLG and the Ministry of Justice with regard to accommodating people leaving prison was also noted, with the example of a joint accommodation board and the development of a national partnership agreement to sit alongside this being mentioned.

Other examples included cross-departmental programmes such as the Disabled Facilities Grants, which helps both to provide appropriate housing offers and to relieve pressure on health and social care, the collaboration between MHCLG and Department for Education to train Ofsted inspectors on accommodation suitability and a joint review of subsidy systems for temporary accommodation by MHCLG and the DWP to ensure access to work and benefits are reviewed as part of a holistic housing options approach.

Finally, the inter-ministerial group on domestic abuse, taking forward the Violence Against Women and Girls strategy, was also noted as a good example of cross-government working.

2.6.9 Positive governmental interventions

Central and local government stakeholders also highlighted the seeming success of a number of governmental programmes seeking to tackle homelessness and rough sleeping. These included early intervention and targeted prevention programmes, such as Staying Put and Staying Close, both of which provide ongoing support to care leavers to reduce their risk of homelessness, and the Tier 3 Community Accommodation Service funded by the Ministry of Justice which provides interim accommodation to people leaving prison who are homeless on release.

The Homelessness Escalations Service, run by the UK Visas and Immigration Service within the Home Office, was also mentioned positively. This service helps local authorities and their providers to clarify the status of non-UK nationals who are homeless or rough sleeping and to escalate cases of those who are waiting for an immigration decision.

Finally, Changing Futures was highlighted as a programme that is supporting positive cross-government collaboration to improve outcomes for people experiencing multiple disadvantages, including those who are experiencing homelessness. The programme is led by MHCLG on behalf of the Ministry of Justice, the Department for Work and Pensions and the Department of Health and Social Care and the co-design of interventions and the systems-level changes aimed for were viewed positively.

2.6.9 Local Authority funding and support

Interviewees agreed that funding for local authorities should be awarded over a longer term to support local authorities to plan more sustainable responses to preventing and tackling homelessness and rough sleeping. Stakeholders also highlighted that local authorities should be given more notice of funding allocations to aid effective service planning. Further upstream, they suggested that this could be supported by considering addressing multiple policy areas and needs concurrently during spending reviews.

Interviewees raised concerns that some of the funding for homelessness and rough sleeping is fragmented and restricted to a defined set of interventions and services, which may not always be what is needed to meet local needs. Local authorities would like to be able to use funding flexibly and across programmes to adapt interventions, services and approaches based on what is or is not working. Making that change is not always possible or can take time to be agreed.

However, stakeholders still saw a need for central government to support local authorities with frameworks and tools to help them to use funding effectively and measure what is working. This could include support with developing tools to better plan cross-sector interventions, such as theories of change for programmes involving multiple services.

2.7 System-wide conclusions

The system mapping, the review of interventions and the fieldwork interviews suggest there is a case for government to take a more system-wide approach to preventing and tackling homelessness and rough sleeping, and to have a more even distribution of policies, programmes and interventions that are targeted at tackling the causes of homelessness and rough sleeping, rather than simply responding to them. Interviewees at a national and local level gave several examples of policies that could reorient the system towards tackling the root causes of homelessness, including policies to reduce the likelihood of people losing their homes, such as making adjustments to welfare policy so that benefit payments rise in line with rising housing costs, and reforming Section 21 of the Housing Act to remove ‘no fault’ evictions. They also suggested policies to improve the availability of housing for those at risk of homelessness, such as tax break incentives to landlords accepting tenants on lower incomes.

Stakeholders suggested homelessness and rough sleeping should be treated as a public health issue, and programmes should have an upstream focus, aiming to adopt measures that offer universal prevention (eg: reducing financial insecurity) and to do more to identify those at risk of homelessness and rough sleeping and undertake targeted prevention activity well before the 56-day crisis period at which the prevention duty set out in the Homelessness Reduction Act applies. To do this successfully, they suggested that a cross-government strategy on homelessness prevention which has collective ownership across government was needed.

Stakeholders also suggested there is a need for more early intervention, addressing structural and individual factors that contribute to homelessness and rough sleeping such as debt, rent arrears, well-being, resilience and supporting better management of tenancies.

The analysis also suggests that there is also a lack of connectedness between parts of the homelessness and rough sleeping system, and between the homelessness and rough sleeping system and other related systems like justice, health and immigration. Given the high level of crossover between homelessness and rough sleeping and the remits of other government departments, stakeholders felt more could be done to support collaboration at a central government level and more effectively enable integration at regional and local levels.

Services should be commissioned, led, and promoted in a way that responds to the needs of the person, and are integrated like their needs are.

Civil servants in other government departments had a range of suggestions for how to achieve better integration. These included taking a more holistic view at spending reviews, and for these to address multiple policy areas and multiple needs concurrently. Some favoured the creation of joint units across government departments to establish collective priorities and integrate efforts, and the promotion of joint needs assessments and pooled funding to facilitate joint commissioning. Stakeholders argued that local authorities need more support from central government to integrate services, and that central government could lead by example by providing infrastructure and allowing more freedom within the funding provided to innovate. Interviewees also felt that central government should incentivise and support more upstream interventions and build joint working into government contracts.

The need for improved infrastructure for joint working at local level was seen as vital for housing, health and care in particular. Improved integration may support local housing, health and care services to put in place joint strategic plans to better meet the needs of their local populations. Stakeholders noted that housing and health services in some local areas are already working well together, and best practice from these areas could be shared to help inform the integration of services in other areas.

Interviewees suggested several changes in the use, collection and sharing of data as important to reducing and preventing homelessness and rough sleeping, including:

enhancing data collection and sharing to inform commissioning and support, ensuring that robust evidence leads to funding the most effective interventions;
increasing the use of data from frontline workers and people with lived experience to inform service design and delivery;
facilitating the sharing of best practice between local authorities, particularly in relation to the use of programme funding and interventions that appear to be working.

The need to make homelessness and rough sleeping prevention a cross departmental priority was highlighted by interviewees as a key issue. They felt that this requires MHCLG to take a leading role to articulate to other government departments why homelessness is a critical issue, demonstrating the causes, consequences, and financial repercussions of homelessness and rough sleeping, and emphasising the opportunities that could be taken to prevent homelessness rather than simply respond to it. Stakeholders in other government departments would like to be supported to understand homelessness and rough sleeping from the perspective of their policy areas. For example, how reducing and preventing homelessness and rough sleeping positively impacts on drug use and offending behaviour as well as how the work they undertake impacts on homelessness and rough sleeping.

According to interviewees, existing and future programmes supporting people at risk of or experiencing homelessness should take a more person-centred approach. By this they meant that barriers to accessing services should be reduced and support should be tailored to individual needs.

We need to be starting with where people are at and working through the range of issues they are experiencing.

Question 5

Chapter 3. Assessing the role of three main homelessness and rough sleeping programmes

Accepted Answer

3.1 Introduction

This chapter sets out findings from an evaluation of the three main programmes and initiatives funded by government to prevent and tackle homelessness and rough sleeping: the Rough Sleeping Initiative, Homelessness Prevention Grant and the Rough Sleeping Accommodation Programme. It provides a description and rationale for each of the programmes and sets out the operating context in each of the five local authority areas where the evaluation took place. This chapter also highlights the system factors that might inhibit or support the achievement of intended outcomes and sets out what worked well and less well about each of the programmes to support future learning.

3.2 The Theory of Change and evaluation methodology

The homelessness and rough sleeping system includes multiple programmes, funding streams and interventions intended both to end rough sleeping and reduce homelessness. The homelessness and rough sleeping systems-wide Theory of Change was designed to establish a comprehensive framework that allows for a holistic view of the system.^{[footnote 18]} It illustrates how policies and programmes delivered across government and by other public bodies contribute to outcomes. This framework clarifies how funding streams transform into services which address homelessness and rough sleeping effectively.

Funding streams and policies are categorised into four pillars that align to the previous government’s strategy on ‘Ending Rough Sleeping for Good’^{[footnote 19]}:

1. Prevention: Funding schemes aimed towards preventing homelessness and rough sleeping from every occurring and ensuring experiences are rare;

2. Intervention: Funding schemes which have an appropriate and timely offer of support to ensure the experience of homelessness and rough sleeping is brief and people are quickly moved into settled accommodation;

3. Recovery: Funding schemes which tackle long-term rough sleeping and homelessness and ensure homelessness and rough sleeping is non-recurring;

4. Transparent and joined-up system: Resources are allocated to enhance collaboration across departments and third sector organisations and develop evidence-based strategies.

This chapter focuses on our evaluation of three key national funding schemes: the Homelessness Prevention Grant (HPG), the Rough Sleeping Accommodation Programme (RSAP) and the Rough Sleeping Initiative (RSI). A brief summary of the evaluation of each of the schemes is set out, bringing out the key common themes observed across them, and addressing the three programme-level research questions as listed below (contextualised for each programme):

1. To what extent are the main homelessness and rough sleeping programmes and interventions (HPG, RSI and RSAP) delivered as intended and what learning can be gleaned from how they are being delivered?

2. To what extent have the HPG, RSI and RSAP programmes and interventions achieved the anticipated outcomes? What worked well, less well and what could be improved, and for who?

3. What, if any, unintended consequences were realised through the delivery of HPG, RSI and RSAP and to what extent were indirect impacts realised in other areas of the system?

The data and evidence analysed and reported here has been gathered from five local authority areas (Bournemouth, Christchurch and Poole; Herefordshire; Manchester; Southend-on-Sea and Westminster) over April and May of 2024. The areas were chosen from a group that expressed an interest in participating in the research and were selected to represent various types and sizes of local authority: both urban and rural, unitary, district and metropolitan boroughs.

3.3 Brief summary of each programme

Rough Sleeping Initiative (RSI). This most recent phase of the Rough Sleeping Initiative was launched in March 2018 as part of the Rough Sleeping Strategy (2018) and was targeted at the 83 local authorities identified to have high numbers of people sleeping rough in the 2017 annual rough sleeping snapshot statistics. This initial allocation provided the 83 local authorities with a share of £30 million over a one-year period (2018-2019), to provide services and programmes specifically targeted towards rough sleeping.

In September 2022, the Rough Sleeping Strategy was updated and replaced with a new cross-government strategy Ending Rough Sleeping for Good which set out how the government intended to reduce and eventually end rough sleeping. Under the strategy, the Rough Sleeping Initiative was extended until March 2025.^{[footnote 20]} The extension included an increase in the number of local authorities receiving funding to 303 and an increase in the level of funding to £499 million across three years (2022-2025).

Rough Sleeping Accommodation Programme (RSAP). The Rough Sleeping Accommodation Programme was announced in 2020 and included in the ‘Ending Rough Sleeping for Good’ strategy published in September 2022. The programme forms part of the government’s approach to ‘recovery’ by enabling a sustainable route out of rough sleeping through the provision of accommodation and support for individuals and a move on to longer-term/permanent accommodation and independent living. It is a capital and revenue programme and therefore, differs from both the Rough Sleeping Initiative and the Homelessness Prevention Grant which are revenue only.

To fund the programme, the government committed £433 million to support up to 6,000 people rough sleeping across England to access and remain in long term accommodation. The programme provides the funding for local authorities to secure long-term supported housing with an intended move on into independent housing. The housing units are intended to be retained by the local authorities as long-term assets designed to support future people at risk of rough sleeping.

Homelessness Prevention Grant (HPG). This grant was introduced in 2021 and brings together the funding allocations of both the Homelessness Reduction Grant (HRG) and the Flexible Homelessness Support Grant (FHSG). It has the overall purpose of bringing together the three aims of prevention, early intervention, and flexibility under one umbrella to give local authorities control in how they manage homelessness pressures and support those who are at risk of homelessness locally. The components of the funding (£1.6bn in total over the period 2021-2025) are allocated through different formulas which consider the local need for prevention, temporary accommodation and the cost of providing housing services. Set out below are the allocations for the Homelessness Prevention Grant from 2021 to 2025 and what they comprise:

2021/22: £310 million allocated (including £263 million through existing HRG and FHSG funding streams and £47 million uplift through the new Homelessness Reduction Act uplift formula) with an additional £65 million allocated through a winter top-up;
2022/23: £315.8 million allocated (including the same allocations as above as well as an additional £5.8 million to meet domestic abuse burdens) with an additional £50 million allocated through a winter top up;
2023/24: £322.8 million allocated (including funding to address Temporary accommodation pressures, homelessness prevention and relief pressures, single homelessness pressures and domestic abuse new burden, reflecting an increase of £7 million on the 2022/23 allocation) as well as £150 million top-up for the Homes for Ukraine scheme support;
2024/25: £331.3 million allocated (same as 2023/24 but reflecting an increase of £15.5 million on the 2022/23 allocation) as well as £120 million top-up for Homes for Ukraine scheme support.^{[footnote 21]}

Unique rationale for the three programmes

The Rough Sleeping Accommodation Programme was developed to address the need for long-term, affordable and stable housing with the understanding that the foremost focus of other programmes is to provide a short-term intervention of temporary, emergency or supported housing. The key intention with the introduction of the programme was to increase the availability of ‘move-on’ accommodation for local authorities to use as part of their housing pathway, so that people who are ready can ‘move on’ from the streets and into independent accommodation.

The Homelessness Prevention Grant was born out of an understanding that prevention is cheaper for the system and better for people. In other words, it is anticipated to be more cost effective in the long-term to prevent a person from experiencing homelessness in the first place than to intervene in a crisis. With the introduction of the Homelessness Prevention Grant, the government sought to empower local authorities to flexibly spend (ringfenced) funding on prevention-related activity wherever it was needed within the system. One of the main ways in which this was expected to impact was in reducing the local authority’s reliance on temporary accommodation.

The Rough Sleeping Initiative in its current form was introduced in 2018 and has seen many iterations to build on its reported successes.^{[footnote 22]} It aims to achieve the following four outcomes:

Prevention – People at risk of rough sleeping are successfully diverted from spending a night on the streets;

Intervention – People currently sleeping rough are able to move from the streets into accommodation;

Recovery – Those who have slept rough previously do not return to the streets;

Systems support – Local authority systems and structures help to keep strategic focus on ending rough sleeping and can monitor and drive forward progress.

It is important to note that these outcomes align with the Theory of Change and pillars of intervention mentioned earlier in this chapter. It is also important to note that it was not the case that equal priority was to be given across the four aims listed above. Local authorities allocating the Roush Sleeping Initiative funding had the discretion to choose their focus areas within the four outcomes based on local need.

Although the Rough Sleeping Initiative focuses predominantly on street-based interventions, the funding stream has extensive overlap with the two other programmes, and there are significant dependencies between Rough Sleeping Initiative outcomes and those of other homelessness and rough sleeping funding streams. In particular, the outcome of ‘Prevention’ is shared with the Homelessness Prevention Grant and the outcome of ‘Recovery’ is shared with the Rough Sleeping Accommodation Programme.

Whilst each programme has a clear rationale, the need for three separate programmes is less clear given overlap between their objectives. It is also challenging to conduct independent evaluations of programmes that are so interlinked in delivery and outcomes, thus preventing the long-term lines of attribution from programme to impact to be clearly identified.

3.4 Brief context of homelessness in each local area

This section details the current rough sleeping and homelessness situation in each of the five local areas sampled for the programme evaluations. This enables an understanding of the current landscape and context relating to homelessness and rough sleeping over the evaluation period. Figures 3.1 to 3.5 show homelessness and rough sleeping statistics for each area.

Figure 3.1 shows that the number of people sleeping rough per month increased in each of the five local authorities over the period 2021/22 to 2022/23, with the increase ranging from 22% in Westminster to 179% in Herefordshire, though where actual numbers are small this can exaggerate the degree of change when stated in percentage terms. The national average increase over the same period was 27%, Furthermore, each of the sampled local authorities had levels of rough sleeping above the national average in 2022/23.

Figure 3.1 The average of number of people sleeping rough per month in 2022/23 and percentage change since 2021/22

Source: Rough Sleeping Management Information, June 2020 to December 2023, Ending_Rough_Sleeping_Data_Led_Framework__December_2023.ods

Figure 3.2 shows that the number of new people sleeping rough per month increased in four of the five sampled local authorities between 2021/22 and 2022/23, with the increase ranging from 16% in Westminster to 277% in Herefordshire, though where actual numbers are small this can exaggerate the degree of change when stated in percentage terms. Southend-on-Sea had a reduction in new people sleeping rough of 33%. All of the sampled local authorities were above the national average in terms of the number of people new to sleeping rough in 2022/23.

Figure 3.2 The average of number of new people sleeping rough per month in 2022/23 and percentage change since 2021/22

Source: Rough Sleeping Management Information, June 2020 to December 2023, Ending_Rough_Sleeping_Data_Led_Framework__December_2023.ods

Figure 3.3 shows that the number of homeless households owed a relief duty increased in four of the five sampled local authorities over the period 2021/22 to 2022/23, with the rate of increase ranging from 3% in Bournemouth, Christchurch and Poole to 21% in Westminster Southend-on-Sea experienced a 31% reduction. Furthermore, in all four of the five sampled local authorities that had an increase in relief duty acceptances, all had increases that were above the national average in 2022/23.

Figure 3.3 The number of households owed a relief duty in 2022/23 and percentage change since 2021/22

Source: Statutory Homelessness Live Tables, years 2021/22 and 2022/23, Tables on homelessness - GOV.UK

Figure 3.4 shows that for the number of households owed a prevention duty, the data is more mixed. For three of the five sampled local authorities, the number of households owed a prevention duty increased from 2021/22 to and 2022/23, with the increases ranging from 2% in Bournemouth, Christchurch and Poole to 31% in Westminster. For the two local authorities where the number decreased, this ranged from 16% in Manchester to 21% in Southend-on-Sea.

Figure 3.4 The number of households owed a prevention duty in 2022/23 and percentage change since 2021/22

Source: Statutory Homelessness Live Tables, years 2021/22 and 2022/23, Tables on homelessness - GOV.UK

Figure 3.5 shows that, for the number of households in temporary accommodation, there was an increase between 2021/22 and 2022/23 for all five local authorities, with the increase ranging from 2% in Manchester to 41% in Herefordshire, though where actual numbers are small this can exaggerate the degree of change when stated in percentage terms.. Furthermore, all of the sampled local authorities had increases above the national average in 2022/23.

Figure 3.5 The number of households in temporary accommodation at the end of Q4 in 2022/23 and percentage change since 2021/22

Source: Statutory Homelessness Live Tables, years 2021/22 and 2022/23, Tables on homelessness - GOV.UK

3.5 Process evaluation

This section describes how the main homelessness and rough sleeping programmes and interventions (the Homelessness Prevention Grant, the Rough Sleeping Initiative and the Rough Sleeping Accommodation Programme) are being delivered and what learning can be gleaned from programme delivery. This includes how funding had been used within each of the five local authority areas participating in the evaluation and what structures and resources were used to deliver the programmes and the core enablers and barriers that were highlighted as impacting programme delivery.

However, although national objectives were available against which to assess activity, local delivery plans were not available to the evaluation team. This made it difficult to be clear on how each funding stream was planned to be delivered at the local level. Additionally, the scale of delivery, local objectives, and timeframes were unclear. Therefore, a full assessment of delivery against the intended objectives was not methodologically possible.

Whilst primary data collection was undertaken robustly, interviews and focus groups were sometimes undertaken before the survey containing the primary data was returned. This resulted in less opportunity to ask follow up questions to gain a deeper understanding of survey responses.

3.5.1 Delivery of activities

The Rough Sleeping Initiative (RSI): Local authorities suggested that funded activities have:

“been entirely delivered as intended because [the funding was] intended to be spent on reducing rough sleeping and supporting rough sleepers off the streets. That’s what we used it for.”

Below, a brief summary of the activities delivered using the funding is provided. These activities are aligned with national programme objectives, but in the absence of being able to access local delivery plans, it is not possible to assess if the activities were delivered as local authorities intended.

The RSI programme was intended to fund activities that were aimed at the following:

Prevention - to identify those at risk of sleeping rough early and intervene before a crisis. Examples include early alert systems, liaison with hospitals, prisons and care, and influencing landlords to reduce short-notice evictions;
Intervention - to support people experiencing rough sleeping into a housing pathway off-the-street. Examples include outreach, off-the-street accommodation and supported accommodation;
Recovery - to identify options to supply longer-term accommodation away from the streets, along with appropriate floating support. Examples include social rented sector and private rented sector access schemes, Housing First and wraparound services addressing health, work and education;
Systems support - to increase capability through specialist roles to address the needs of the most vulnerable people, such as women, victims of domestic abuse, young people, LGBTQ+, those with drug and/or alcohol needs or mental ill-health.

Local authorities reported an unequal spread of resources for the four activity categories, as seen in Figure 3.6 below. Activities were broken down into 12 categories and then grouped into how they related to the four pillars of intervention, recovery, systems support, and prevention. An ‘other’ category was used for activities that did not seem to relate to the four pillars directly.

Activities related to intervention were by far the most prevalent – two thirds of funding directed towards off the street accommodation, supported accommodation and outreach services. Activities related to recovery utilised around a fifth of spending, and this included services intended to support people to access private rented sector accommodation, health services and, to a smaller extent, employment support services. Systems support activity – mostly legal support and service navigation – accounted for around 10% of the spend on activities There was relatively little spend on prevention activities – less than 1% overall.

Figure 3.6 The activities delivered by the 5 sampled local authorities and total funding allocated 2022/23 – 2024/25

The Rough Sleeping Accommodation Programme: The main activity associated with the programme was to provide ‘move-on’ homes to people currently sleeping rough and those with a history of rough sleeping. There were a variety of ways in which homes could be provided under the programme. Listed below are the types of activities undertaken and the number of successful homes delivered across the five areas (aggregated) compared with the targets reported by the local authorities at the outset.

purchase and repair of properties – 172 out of 187 delivered;
maintenance of existing, satisfactory properties – 10 out of 10 delivered;
refurbishment and repurposing of existing stock – 17 out of 17 delivered;
development of new build properties – 14 out of 32 delivered;
private sector leasing and contribution towards social investment programmes – 43 out of 43 delivered.

The gaps in delivery, particularly in the development of new build properties, were largely reported by one local authority (Manchester City Council) but it is important to note that the programme was still ongoing at the time of data collection, and this aim may now have been met.

The second activity associated with the programme was to design and deliver services that support those people housed in the move-on accommodation and work with them in moving on into permanent accommodation.

Examples of these types of services include mental health support, skills development, employment support and provision of funding for occupational health workers. Each local authority was expected to outline for MHCLG the specific activities they would deliver or commission with the funding awarded, but in the absence of clear performance reporting associated with delivery of these activities, it is not possible to say if these support services were delivered as planned.

The Homelessness Prevention Grant. There were four overall objectives expected to be fulfilled through HPG funded activity:

Preventing homelessness by early intervention, providing funding and support through early intervention to address housing instability before it arises;
Supporting vulnerable populations, including those experiencing domestic abuse, refugees, and those facing financial hardship;
Promoting stable and suitable accommodation, including preventing evictions, mediating landlord-tenant disputes, and offering financial assistance to stabilise housing as required;
Allocating resources strategically based on local needs and market dynamics.

To support these objectives, local authorities were expected to use the funding for staffing, temporary accommodation or prevention or relief duty activities. The remit for the Homelessness Prevention Grant is broad and flexible, and intended to support homelessness prevention systemically at the local level. As there were no delivery plans or local targets agreed with each local authority, it is not possible to evaluate whether or how the objectives have been met. However, it was reflected by local authorities that the Homelessness Prevention Grant programme was not achieving the overall objective of focusing on prevention because:

it’s subsidising the provision of temporary accommodation, and…funding [the assessment of the needs of] people when they become homeless.

Figure 3.7 shows a breakdown of activities delivered through Homelessness Prevention Grant funding in each of the five participating local authorities. It was not possible based on locally available data to separate spend on prevention from other activities, such as administering relief and support duties under the Homelessness Reduction Act.

As can be seen from the chart below, the bulk of the spending across all five local authority areas made under the Homelessness Prevention Grant was focused on the provision of emergency housing (temporary accommodation, including B&Bs and hostels). The second highest spend category in four of the five local authorities was related to the delivery of duties under the Homelessness Reduction Act – administration, prevention, relief and support. Westminster City Council was something of an outlier in reporting no spend on delivery of the Homelessness Reduction Act and more significant spend on local authority and housing association stock.

Figure 3.7 The activities delivered by five sampled local authorities 2020/21 – 2022/23 and spend on each

Local authorities felt pressured to make this shift away from using the Homelessness Prevention Grant for prevention and using it instead to meet the costs of temporary accommodation because of the number of households accepted as homeless and in priority need. This increased demand was linked to rising pressures, such as cost of living, loss of private rented sector tenancies, and lack of affordable housing stock.

3.5.2 Structures and resources involved in delivery

Four out of five of the local authorities participating in the evaluation mentioned using both internal staff and external service providers to deliver the three programmes. Capacity was often added to both delivery and management functions to existing teams, as well as through funding external specialist services. However, information on the exact numbers of internal staff employed by local authorities to deliver either Rough Sleeping Initiative or Homelessness Prevention Grant activities or programmes was unavailable.

For the Rough Sleeping Accommodation Programme, the five local authorities provided internal staff numbers involved in programme delivery that ranged from 0.30 to 6.00 FTE. Local authorities also worked in partnership with voluntary and community sector organisations, private providers, and other services (for example, probation, substance use, employment advice and housing) to deliver wraparound support services to ensure that tenancies are more likely to be sustained. Some of the local authorities raised concerns about the challenges in recruiting staff to deliver the programme and also highlighted capacity issues in services that tenants might rely on:

We’ve got issues [around capacity] locally, I think as most of the local authorities have, with mental health services, mental health provision, mental health outreach - too many people [in need of support] and not enough staff to be able deliver the service.

It was not clear whether this capacity issue related to these services having limited funding or a shortage in staff with the appropriate skills.

Local authorities also mentioned experiencing difficulty in recruiting and retaining skilled staff, especially for more specialised support roles. This was particularly challenging as local authorities were all given the programme funding at the same time, and thus were on similar recruitment cycles, creating inter-council competition for the same pool of specialist staff.

3.5.3 Partnership working

All local authorities agreed that partnership working was key to successfully delivering programme activities. Partnerships included those with third-sector organisations and wider statutory services. For example, all local authorities referenced the importance of working with the voluntary and community sector, as well as mental health, probation, and employment services. Given the varied and multiple needs of some people experiencing homelessness, support from these specialist services was fundamental to delivering effective interventions.

Interviewees felt partnership working enhanced communication, decision-making, and coordination among strategic and operational stakeholders involved in programme delivery. Herefordshire Council has implemented a three-tiered governance structure (Bronze, Silver, Gold) for the delivery of homelessness and rough sleeping services that facilitated the successful management of internal and external partnerships. The Bronze group is largely operational working directly with individuals experiencing rough sleeping; the Silver group consisted of programme managers; and the Gold group included the strategic officials who were responsible for the setting and delivery of the overall strategy. Statutory and voluntary sector partners involved in the delivery of services are represented at all levels within the structure, depending on their role. Herefordshire Council reflected that having this ‘chain of command’ in place is a key enabler for the successful delivery of homelessness-funded services as it established a clear structure for communication and decision-making between partners.

Across all three programmes, partnership working within the system was emphasised as a key success factor in programme delivery in the majority of local authority interviews:

partnership working is huge, and without it, [the programme delivery] wouldn’t happen.

Strong partnerships with diverse stakeholders facilitated a comprehensive support network and the flexibility to address local needs. Some local authorities discussed how partners were embedded into the programme delivery team right from the beginning. For example, Westminster City Council described how they brought together supported accommodation providers and specialist mental health services to develop the initial bid for the Rough Sleeping Initiative funding. Effective partnership working enabled local authorities to build a scaffold of support around individuals at risk of or experiencing rough sleeping and ensure the system within the respective local areas could more readily adapt their approach to reducing rough sleeping if and when required. The authority acknowledged that:

[The programme] needs everyone’s input, and it needs…open information sharing around these people if we’re truly going to solve the problem.

Overall, local authorities and their delivery partners stated that current partnership working was effective. Local authority officers working at Manchester City Council cited effective partnership working as the reason for their successful delivery of the Rough Sleeping Accommodation Programme, despite the challenging local market conditions experienced, such as increasing house prices, growing demand from the increase in refugees and asylum seekers and extremely limited access to mental health services. However, some local authorities felt that partnership working could be made even stronger. It was suggested that bringing all relevant services together, including health, housing, social care, police and crime, and pooling funding would support more integrated partnership working, fostering a collaborative, whole-system approach to preventing and tackling homelessness.

3.5.4 Enablers and barriers

A key enabler raised was flexibility of funding, which allowed local authorities to spend funding in line with local needs and respond to changing demands. For example, a representative from Manchester City Council shared that if the Homelessness Prevention Grant funding had been strictly ring fenced for homelessness prevention only, many local authorities would find that restrictive, as they would not have been able to use it to cover urgently needed temporary accommodation. On the other side of this, it was acknowledged that this very funding flexibility meant that crisis spending often won out over spending on services that may prevent homelessness in the medium to longer term. Funding flexibility also enabled local authorities to meet unexpected needs like those associated with increased decision-making activity for those seeking asylum housed in Home Office hotels in the autumn of 2023.

Other success factors highlighted across the programmes included enhancing capacity for delivering support services by establishing specialist roles, regular communication with working partners, fostering a positive relationship with MHCLG and use of local data to target services in areas with the greatest need.

A common challenge raised for both the Rough Sleeping Initiative and activities delivered under the Homelessness Prevention Grant was the timing of additional ‘surge’ funding. While obviously welcome and very much needed, local authorities said funding was often given to them with short notice and with a requirement for it to be spent within a short timeframe. Local authorities felt this required short-term thinking and rapid responses, which was seen to impede long-term planning in favour of delivering “what is quickest and easiest” rather than things that may be more strategically valuable.

The majority of the local authorities and delivery partners agreed that the pressure on resources from a rising demand for services has also made it difficult for them to prioritise prevention over immediate solutions and crisis interventions. This rise in demand was often mentioned in connection with the rising cost of temporary accommodation:

temporary accommodation expenditure is our largest pressure at the moment, and that is increasingly taking up a larger percentage of our Homelessness Prevention Grant allocation to offset the costs that we are occurring in the temporary accommodation space.

3.5.5 How does local context impact on delivery

In the local authorities involved in the evaluation, the delivery of all three programmes was affected by similar contextual factors. The main external contextual factor raised was the lack of affordable housing. This contributed towards the increased demand for homelessness and rough sleeping services in all five areas and made it more challenging for local authorities to provide cost-effective options for supported and long-term housing.

All local authorities reported increases in the level and complexity of needs among individuals sleeping rough. Insufficient capacity in support services, especially in mental health support services, often made it difficult to address this increase in demand. Finally, due to variation in the provision of rough sleeping support across different local authorities, some had experienced an increase of people who had moved from local authorities where their desired support was not available. This was seen to increase demand in areas that are viewed as having more services and therefore as more likely to provide this support.

3.5.6 How do system factors impact on delivery

Interviewees mentioned several wider system factors that they felt were driving homelessness and rough sleeping and/or impacting on programme delivery. These included the ongoing impact of the cost-of-living crisis on peoples’ ability to afford housing, limited housing stock suitable for addressing homelessness and rough sleeping needs and challenges in coordination across government departments (as also mentioned in Chapter 2 in interviews with key stakeholders from MHCLG and other government departments). For example, one local authority referenced the current definition of a person being at risk of homelessness by MHCLG is that they will lose their accommodation within 56 days. However, the notice period used by the Home Office when people leave asylum support accommodation is 28 days. This discrepancy in definitions leads to confusion and inconsistent support. For example, a person may still qualify for support under the 56-day rule, but they might already face eviction or removal actions based on the 28-day notice, meaning individuals may not receive support on time.

Local authorities found it difficult to distinguish between funding streams and to accurately identify which programme was funding which intervention. This was because funding for the three programmes was often used in combination with funding from other sources (like from health or probation) and while this flexibility was perceived to have positive impacts on the delivery of homelessness services and improving outcomes for people, it made attributing costs and outcomes difficult.

Overall, these findings suggested that reducing homelessness cannot be achieved by programmes operating in isolation as wider services and factors influence the ability of homelessness services to have meaningful impact. A whole system approach that tackles not only the root causes of homelessness but also promotes collaboration and cross sector working may deliver better outcomes for people at risk or experiencing homelessness than standalone programmes or interventions.

3.6 Outcomes evaluation

The outcomes evaluation sought to answer the question about the extent to which the Homelessness Prevention Grant, the Rough Sleeping Initiative and the Rough Sleeping Accommodation Programme have achieved their anticipated outcomes? With a focus on what worked well, what worked less well, what could be improved, and for whom?

3.6.1 Emerging outcomes

Overall, findings from the interviews and survey suggested that the three programmes were successful in supporting the number of people experiencing, or at risk of experiencing, rough sleeping or homelessness. Below the emerging evidence is summarised for each outcome from local authorities, aligned with the national aims of each respective programme.

This research assessed how funding has been used within each of the five local authority areas participating in the first phase of the evaluation. However, although national objectives were available against which to assess activity, local delivery plans were not available to the evaluation team. This, combined with the lack of a counterfactual, made it difficult to be clear on how each funding stream was planned to be delivered at the local level. Additionally, the scale of delivery, local objectives, and timeframes were unclear. Therefore, a full assessment of delivery against the intended objectives was not methodologically possible.

Rough Sleeping Accommodation Programme

Local authorities felt that the Rough Sleeping Accommodation Programme (RSAP) provided a means to address “a specific housing need” of its target beneficiaries, i.e. those who had “a long term [history] of rough sleeping” and therefore had not been appropriately supported through previous programmes. The aims of the programme and how the five evaluated areas have performed against those aims are set as below.

Increase the availability of move-on accommodation for individuals sleeping rough or at risk of rough sleeping or homelessness by building more units: The five local authorities surveyed aimed to deliver a total of 289 move-on units by the end of 2025. Survey returns suggested that 256 move-on homes or 88% of the overall target had been provided by May 2024 when research was undertaken.

Aim 2: Provide support funding for RSAP accommodated individuals (appropriate to needs). The five local authority areas evaluated reported that RSAP helped them to increase capacity to provide support for those individuals accommodated in move-on homes. This included: Mental health support, skills development, employment support and funding for occupational health workers. Local authorities also reported that the support was successful in ensuring tenancies were sustained long-term, with one council reflecting that “it has gone quite well. [They] have had no failure…no evictions and…no complaints”.

Aim 3: Achieve a sustainable reduction in rough sleeping. Whilst there has been an overall rise in the numbers of people experiencing rough sleeping across all the evaluated local authority areas, four out of five reported that RSAP had contributed to curbing larger increases in the number of people experiencing rough sleeping, the number of long-term rough sleeping and the number of people returning to rough sleeping. RSAP was reported to be particularly effective in reducing rough sleeping for those individuals who had long-term, complex needs and who could not be placed in shared housing or in independent housing without support.

Rough Sleeping Initiative

The Rough Sleeping Initiative (RSI) was described by one local authority stakeholder as the “jewel in the crown” in terms of supporting individuals experiencing rough sleeping to move from the streets into accommodation and prevent them from returning to experiences of rough sleeping.

Below is a summary of the aims of the programme along with evidence about how the five local authorities involved in the evaluation are meeting these national aims.

Aim 1: Prevention – People at risk of rough sleeping are successfully diverted from spending a night on the streets. This is an area for improvement. Local authorities acknowledged that they had not focused on prevention activities, as funding was often used for crisis relief instead. One council highlighted that:

“RSI has been less effective at meeting the expectation of…developing true rough sleeping prevention activity. Services commissioned remain predominantly about supporting people in crisis, rather than preventing them from hitting the streets in the first place.”

Aim 2: Intervention – People currently sleeping rough are able to move from the streets into accommodation. Four out of five local authorities reported that the initiative had somewhat contributed to reducing the number of people experiencing rough sleeping; while one local authority reported that RSI has significantly contributed. This was largely due to the funding of off-the-street accommodation and outreach services.

Aim 3: Recovery – Those who have slept rough do not return to the streets. All five local authorities reported that the initiative had somewhat contributed to reducing the number of people experiencing rough sleep long-term and those returning to rough sleeping.

Aim 4: Systems support – Local authority systems and structures help keep strategic focus on ending rough sleeping and can monitor and drive forward progress. Developing systems and structures to provide wraparound support was described as game-changing in targeting entrenched rough sleeping. For instance, one local authority claimed that:

any one person will have that multidisciplinary wraparound approach, and that’s kind of enabled us to really deal with those very entrenched rough sleepers and get Care Act assessments for them, get mental health assessments for them without passing them from team to team. It’s all kind of contained within one team and one service, and I think…that’s been kind of game-changing for that particular group.

Homelessness Prevention Grant

This funding stream has a clear aim to drive increased homelessness prevention activities across the country.

In the survey, four out of five local authorities reported that prevention or relief activities funded by the Homelessness Prevention Grant had had a positive impact on the number of people experiencing homelessness for the first time. Nationally, based on the H-CLIC data, 52.7% of prevention duties in 2022/23 ended with the household securing accommodation for 6+ months.

Table 3.1 below shows that for the five sampled local authorities, the percentage of prevention duties in 2022/23 that ended with the household securing accommodation for more than six months ranged from 25.5% in Herefordshire to 80.5% in Westminster, with Manchester not providing the data.

Table 3.1 showing the % of successful prevention duty outcomes

Area	% successful prevention duty outcomes
Bournemouth, Christchurch and Poole	55.6%
Herefordshire	25.5.%
Manchester	Not reported
Southend-on-Sea	55.1%
Westminster	80.5%

Monitoring data also showed that there is strong compliance with providing data on HPG-funded activities through H-CLIC, with 96% of all local authorities submitted H-CLIC data in 2023/24, including the five local authorities included in the evaluation. It also suggested that the sampled local authorities all spent 100% of their HPG allocation. Evidence from the interviews with the five local authorities suggested that their view was that overall, the Homelessness Prevention Grant had positive impacts on preventing people from experiencing homelessness. However, gaps remain in the analysis regarding detail of precisely what impact activities had in each local authority and whether intended outcomes were met. This missing data made it difficult to assess how local authorities have performed.

3.6.2 System factors impacting on outcomes

Despite the perceived success of the programmes and the objective progress made towards achieving their aims, monitoring data for the sampled local authorities showed that the number of people experiencing rough sleeping across England had increased year-on-year, as shown in Figure 3.2. Local authority officers explained this increase by reference to wider system pressures, which they felt had counteracted the positive impact of the programme funding:

rough sleeping numbers . . . continued to increase across this period owing in part to the rising cost of living, inflation and market pressures in the private rented sector, both in relation to affordability and availability. The complexity of rough sleepers with multiple support needs, increasing flow and reduced move-on options are all contributing factors to rising rates of rough sleeping.

The local authority staff and delivery partners we spoke to cite a wide range of external pressures acting to increase pressure on the homelessness and rough sleeping system in recent years. A full list of these factors is provided below for completeness:

cost-of-living crisis;
shortages of available mental health services;
ongoing impact of austerity;
limited availability of affordable accommodation;
shortages of social housing;
welfare provision not meeting needs;
staff retention challenges due to burnout;
reduced funding for other public services;
Home Office policy on asylum seekers/refugees.

Taken together, local government stakeholders working on the frontline in the five local authorities sampled felt that these wider system factors had combined to outweigh gains made through the introduction of Rough Sleeping Accommodation Programme, Rough Sleeping Initiative and Homelessness Prevention Grant. The number of people being supported had increased in all five areas, but not enough to keep up with rising demand. Because of this, local authorities were also of the opinion that rates of homelessness and rough sleeping would have been even higher without these funds.

In particular, local authority staff/housing officers believed that the Rough Sleeping Accommodation Programme enabled them to address a crucial housing need:

to develop proper next step accommodation for people who have been on the streets or have been in shared accommodation…where shared accommodation wouldn’t be appropriate for them.

Similarly, the Rough Sleeping Initiative and the Homelessness Prevention Grant supported local authorities to target their services to specific underserved groups. In Manchester, council staff reported that these programmes were used to fund workers focused specifically on women, improving outreach efforts, facilitating access to emergency accommodation, and increasing connections with other essential services such as healthcare. Council representatives anticipated that in the absence of these funded services, women who were experiencing rough sleeping, working on the streets in the evening and at risk of being homeless would have poorer physical and mental health outcomes and be at risk of increased levels of domestic abuse and sexual violence. Without the wraparound support provided through the Rough Sleeping Initiative, councils predicted higher levels of antisocial behaviour, crime, unmet mental health needs and the potential for increased suicides.

Local authority staff in one area felt the Rough Sleeping Accommodation Programme enabled them to serve groups such as single adults who are not in priority need and may therefore receive less attention. One local authority suggested that there would be more deaths amongst the people experiencing rough sleeping in the absence of supported move-on accommodation, as the council have “used RSAP funding to provide support to those who [they] have no statutory duty to accommodate.”

Local authorities and delivery partners reported that being able to move people off the streets and into supported accommodation through the Rough Sleeping Accommodation Programme provided the stability for them to benefit from wraparound support as well as access wider opportunities. Stable accommodation can have positive downstream impacts on other domains of life, such as health, education and employment, which in turn can reduce risks of homelessness in the future.^{[footnote 23]}

One council reflected that due to legal duties placed on local authorities, they would have to find alternative sources of funding to address and prevent homelessness for those with priority need. In the absence of other funding, the pressures of bearing costs such as temporary accommodation may result in more councils having to serve S114 notifications (equivalent to declaring bankruptcy) if they are unable to meet their statutory obligations.

3.6.3 Outcomes relating to changes in the wider system

Two local authorities reported a decrease in hospital admissions for mental ill health following the delivery of RSI and RSAP funded activities, which they attribute to partnership working and provision of mental health wraparound support. For instance, Centrepoint in Manchester found that the rate of hospital admissions for young people they supported decreased from 90% to 15%. Centrepoint reported that having support staff funded through the programmes meant they were able to identify needs, conduct assessments, and refer people to the relevant mental health, drug and alcohol-related services more rapidly. The positive changes and learnings from this bolstered the case to recruit specialist staff, which was further critical to addressing unmet needs associated with homelessness and rough sleeping.

A similar trend was also found for employment, wherein service delivery partners found that wraparound support provided through Rough Sleeping Initiative funding increased downstream signposting to employment services.

Together this highlighted the importance of conceptualising homelessness and rough sleeping as a systemic issue and developing solutions that are able to address several dimensions through a coordinated, multidisciplinary approach. In fact, one local authority suggested that:

unless [they] have the partnerships that wraparound that support for those clients when they are in accommodation, then those tenancies will not succeed as effectively.

This indicates a reciprocal relationship between homelessness and rough sleeping services and the wider support services, where the likelihood of seeing positive change in terms of rough sleeping is maximised when the intervention also covers other domains of the beneficiary’s life, and vice versa.

Downstream positive changes were also reported as a result of moving families from bed-and-breakfast to private rented accommodation, supported by Homelessness Prevention Grant funding. According to local authorities, this was less expensive to provide and unlocked further cost savings by preventing issues that often arise from bed-and-breakfast stays, such as children missing school or families experiencing health problems.

3.6.4 Outcomes by beneficiary groups

The groups of people that benefited from the three programmes varied according to the target priority groups of the respective programmes. The Rough Sleeping Accommodation Programme and the Rough Sleeping Initiative are targeted towards people sleeping rough, whereas the Homelessness Prevention Grant aims to support a wider group of people at risk of experiencing homelessness.

The Rough Sleeping Initiative had a particularly positive impact on reducing the number of individuals experiencing rough sleeping long-term. In addition, four out of five local authorities said they achieved the best outcomes for individuals under age 25 and UK nationals. For other priority groups, there was a more mixed picture. A few local authorities had also used Rough Sleeping Initiative funding to somewhat reduce the rates of rough sleeping among non-UK nationals, refugees and asylum seekers.

The Rough Sleeping Accommodation Programme also had the most positive impact on under-25s, UK nationals and those with a long-term history of rough sleeping. The programme was perceived to be particularly valuable in supporting those individuals who traditionally struggled in shared accommodation but were unable to independently support their own tenancies. By housing people in supported accommodation offered as ‘move-on’ from the streets for a period of time, the Rough Sleeping Accommodation Programme enabled them to “gain employment, return to college and develop community connections”.

The Rough Sleeping Accommodation Programme also provides tailored support services alongside accommodation. This was felt to be effective at supporting individuals who had been sleeping rough long-term and had multiple vulnerabilities including substance use or disabilities. However, staff working in to support those accessing the programme emphasised the need for this support to be sustained over the long term: “as soon as the funding drops away, you still have people who are [sleeping rough].”

Interestingly, the evidence suggested that Homelessness Prevention Grant funded staff and access to temporary accommodation had a greater impact on those leaving institutions. For instance, Manchester City Council reported observing around 1% of people sleeping rough who had left prison in the previous 3 months, compared to 4% nationally. This could be because more reactive solutions are required for this beneficiary group to alleviate the immediate risk of homelessness, rather than long-term prevention activities. One local authority also mentioned that they had used the grant to set up a vulnerable renters’ fund to support families and individuals who were at risk of losing their privately rented accommodation due to increasing market rents.

3.7 Learnings

What has worked well for the programmes?

Strong partnership working. Across all three programmes, emerging evidence from the fieldwork suggested that building strong partnerships and working together with community, social and third sector organisations, as well as with other councils and wider public services was critical to the delivery of activities. This facilitated positive outcomes for beneficiaries experiencing, or at risk of experiencing, homelessness or rough sleeping. This learning chimed with other reports considering what works best in tackling homelessness and rough sleeping such as the Kerslake Report from 2021.^{[footnote 24]}

It was noted that partnership working was at the heart of the delivery across all programmes, and local authorities were positively surprised by the extent to which it was beneficial. For instance, a Southend-on-Sea Council member reflected that “a lot has been achieved because people pulled together” to not only place people into accommodation but also provide follow-on multidisciplinary support in other areas of the person’s life.

Better identifying and meeting wider needs. The funding streams allowed the local authorities to undertake research to fully understand the experiences of people rough sleeping and to better identify, contact and assess the needs of those people who are homeless in their area. This prompted a better understanding and visibility of a population that is usually “forgotten about”. Uncovering these needs, such as those related to physical or mental health or criminal justice, led to pressure on associated wider public services to be able to address these unmet needs.

Relatedly, providing wraparound support was highlighted as instrumental in achieving positive outcomes for people supported by the programmes. For instance, in the case of RSAP, support staff were critical to ensuring that people accessing move-on homes were well settled and any issues could be managed so that the tenancy continued to be sustained.

Similarly for the Rough Sleeping Initiative, linking the beneficiary with other specialist staff dedicated to resolving challenges with mental health or substance use was seen to be a fundamental part of the delivery model.

What has worked less well for the programmes?

Overlap and lack of clarity around outcomes and targets. Most importantly from the perspective of this evaluation, there was a lack of clear, defined objectives and targets associated with each programme. While rough sleeping and homelessness metrics are regularly monitored and reported by local authorities, there isn’t clear alignment between these and the programme activities. This makes it challenging to assess whether the programmes were successful in delivering activities and outcomes as intended.

Relatedly, because the three programmes overlap and because beneficiaries can be supported within all three, there are clear challenges understanding the contribution of each programme on the observed outcomes. While the Rough Sleeping Initiative and the Rough Sleeping Accommodation Programme come with clear plans and some deliverables that can be monitored and reported on, the Homelessness Prevention Grant is functioning more like a system-level contribution to existing activity around homelessness prevention and the provision of temporary accommodation and because of this attribution of activity to outcomes is very difficult.

Lack of spending on prevention. While local authorities reported benefiting from the flexibility associated with the funding streams, this flexibility came with a clear risk that funding was much more likely to be spent reactively instead of on long-term prevention activity. Four out of five areas reported spending the bulk of their Homelessness Prevention Grant funding on temporary accommodation and less than 1% of Rough Sleeping Initiative funding in the five areas was spent on prevention. Where prevention activity was taking place, it was largely crisis prevention with households already at risk of homelessness and within the 56-day period specified within the Homelessness Reduction Act.

Need for more flexibility within Rough Sleeping Accommodation Programme (RSAP). Local authority interviewees highlighted that the impact of RSAP could be improved by increasing their ability to build purpose-built properties, allowing them to explore alternative commissioning models and allowing for greater flexibility in the support given to specific target groups. They also noted that it was not always easy to find the right property for the Rough Sleeping Accommodation Programme target population, for example to meet needs in terms of convenient locations or the physical state of the property.

Behavioural concerns from neighbours of move-on homes. Anecdotally, there were instances reported of challenging and stigmatising behaviour displayed towards Rough Sleeping Accommodation Programme tenants by those who lived close to Rough Sleeping Accommodation Programme accommodation, which were believed to have reduced the chances of people successfully maintaining their tenancy and putting them at risk of becoming homeless again.

Perverse incentives to stay/become homeless. There were also anecdotal reports that individuals without a priority need started sleeping rough because they believed they would be in a better position to access support with their unmet housing needs through the local rough sleeping pathway. Local authority staff felt that this was partially due to gaps in support for inadequately or insecurely housed single people experiencing ‘hidden homelessness.’

It was also reported by a local authority that the quality of Rough Sleeping Accommodation Programme properties was sometimes higher than the private rented sector or shared housing options that people were expected to move on to, making it hard to persuade people to move on from the Rough Sleeping Accommodation Programme property.

Demand rose beyond expected levels. Although local authorities had expected an increase in the overall number of people experiencing rough sleeping and homelessness due to various socioeconomic factors, the actual demand placed on their services was higher than anticipated. The rising number of people new to rough sleeping was described as “shocking” by one local authority. The duties placed on local authorities by the Homelessness Reduction Act and the increasing numbers of families becoming homeless and needing temporary accommodation have also placed considerable pressure on local authority capacity.

Competition between local authorities for recruitment. Local authorities were all granted Rough Sleeping Initiative funding at the same time, which meant they had similar timelines for recruiting staff. This created a high demand for relevant individuals with appropriate skills and experience in the same areas and led to the same individuals being employed by multiple councils. This prevented local authorities from developing sustainable staffing models and delivery structures in their local area.

Funding uncertainty. Specific aspects about the funding structures could be improved. For instance, local authorities suggested that the timing of additional Rough Sleeping Initiative funding could be more effective - while additional funds are welcome given pressures in the system, the one-off, very short-term nature of the awards made to local authorities mean that money could not be spent strategically or on prevention activity. Additionally, clarity on the continuation of funding beyond March 2025 was required to allow local authorities to continue to build on the positive impacts and working relationships established to-date.

Limitations on capacity. There was a need expressed by local authorities to increase their capacity to deliver activities focused on the universal and targeted prevention of rough sleeping and homelessness, which proved to be a challenge for both Rough Sleeping Initiative and Homelessness Prevention Grant funded programmes and activities. It was found that the focus on these types of prevention interventions reduced over time as local authorities were under pressure to provide crisis prevention and relief and cover costs of temporary accommodation, particularly from Homelessness Prevention Grant funding.

Programme success was sometimes further inhibited by capacity issues in wider support services (such as mental health support, substance use support) which were beyond the control of local authorities. In some cases, Rough Sleeping Initiative funds were used to try and mitigate these gaps with specialist mental health services or dedicated substance use workers funded from the programme grant.

In summary, the programme evaluations illustrated some of the challenges present within the wider homelessness and rough sleeping. Overall, the programmes are focused on the relief of homelessness and rough sleeping rather than its prevention (or on the goals or paradigm shift as described in Intervention Level Framework terms), and not cross cutting enough. The operation of the programmes would benefit from fundamental reform, with a greater focus on enabling and encouraging more upstream universal or targeted prevention of homelessness (ie: much earlier than the 56-day prevention duty period provided for in the Homelessness Reduction Act) and clearer, more ambitious goals.

Question 6

Chapter 4 Discussion, conclusion and next steps

Accepted Answer

4.1 Conclusions

This chapter presents the conclusions that can be drawn to date from the systems-wide and programme evaluation activity set out within this report and sets out some next steps in terms of considerations for the future. As this report contains preliminary findings, full recommendations were not felt to be appropriate based on the strength of evidence at this stage of the evaluation.

Both national and local stakeholders, including local authority representatives and their delivery partners, highlighted how economic conditions in the UK have contributed to severe strains on the homelessness and rough sleeping system. Research participants highlighted a number of critical issues that they believe have combined to create significant pressures and barriers to tackling the problem more effectively, including:

rising rents and reductions in the value of welfare benefits (for example, Local Housing Allowance rates, Housing Benefit and Universal Credit) leading to increased evictions and/or people being unable to find affordable housing;
a shortage of affordable housing, which is particularly acute in the private rented sector, due to private landlords exiting the market, a failure to build enough affordable social rented homes and competition for the homes that are available acting to drive up rents;
lack of affordable housing leading to an over-reliance on expensive temporary accommodation, and/or use of lower quality/unsuitable accommodation such as B&Bs or placements of housing in cheaper areas which may be far away from where people have connections;
rapidly rising housing demand, particularly from families facing cost of living issues and from new refugees experiencing homelessness or requiring resettlement;
the demand for specialist support services and the poor availability of, most notably, mental health support, was stated to have meant that people are more at risk of losing tenancies and more likely to experience prolonged homelessness.

People with lived experience stated very clearly that rent levels, debt and arrears, coupled with an inflexible and inadequate system of benefits, made tenancy sustainment difficult, especially in the private rented sector.

It is likely that these wider structural pressures have affected the extent to which programmes like the Rough Sleeping Initiative and funding streams like the Homelessness Prevention Grant have been able to reduce the number of people experiencing rough sleeping or homelessness for the first time, as local authorities have had to respond to the rise in presentations from people who qualify (as statutorily homeless and in priority need) for crisis intervention in the form of temporary accommodation provision.

4.1.2 The majority of government activity and expenditure is on crisis relief, rather than prevention and delivered at a level which does not create system-level improvements

The policy review conducted using the Intervention Level Framework (ILF) analysis illustrated that there is an emphasis within existing government-funded policy/programmes on crisis relief and the provision of services rather than on longer-term prevention, or on tackling the causes of homelessness at a system-level. This was underlined by the local authority survey and interviews which illustrated how pressures on services means that funding allocated to homelessness and rough sleeping is typically short term and focused on responding to immediate need. This restricts both local authorities’ ability to plan into the longer term and to implement longer-term or system-level prevention activities.

At a local authority level, budgets allocated for prevention (notably the Homelessness Prevention Grant) were increasingly being spent on crisis relief and those prevention activities that are being funded are limited to the 56-day period set out in the Homelessness Reduction Act rather than activities related to more upstream prevention, whether that be universal or more targeted prevention. Similarly rough sleeping services were commissioned predominantly to support people after they are already sleeping rough, rather than preventing them from reaching the streets in the first place.

The data collected from the five local areas participating in the research showed that less than 1% of Rough Sleeping Initiative funding was spent on prevention. Government interviewees outlined how a political focus on short term economic benefits and immediate outcomes means there is less support for longer term preventative interventions but recognised a need to focus more on the underlying causes of homelessness and to enable the implementation of more preventative measures to produce longer term change.

People with lived experience echoed this finding, stating that access to preventative support prior to homelessness was hard to find, with local authorities not having much to offer.

4.1.3 There are significant competing policy priorities within national government

It was acknowledged repeatedly by research participants that homelessness and rough sleeping is a complex policy area requiring strong cross-departmental collaboration to tackle the issue.

Some areas of strong cross-government working were identified by those interviewed at a national and local level, notably the work stemming from the Ending Rough Sleeping For Good strategy, which was highlighted as contributing to improved collaboration across government departments, and the development of specific work focused on preventing homelessness for those leaving prison and hospital. Programmes with promise in delivering cross-cutting outcomes included the Tier 3 Community Accommodation Service for people leaving prison homeless, the Homelessness Escalations Service for those needing help to resolve their immigration application, the Rough Sleeping Drug and Alcohol programme, and the Changing Futures programme.

While there are examples of good cross-government working, overall conflicting departmental priorities, and a lack of cross-governmental understanding, appear to be hindering efforts to tackle homelessness and rough sleeping. Interviewees stated that departments seemed focused on the specifics of their own departmental agendas and are limited in the resources they are willing to allocate, or the policy focus they are willing to give, to developing opportunities for cross-departmental collaboration to prevent homelessness and rough sleeping. Interviewees particularly noted that connections between homelessness and health (identified as a key area in the systems mapping) are currently under-developed within government policy.

People with lived experience echoed this from their own perspective, suggesting that more attention needs to be paid to meeting the overlapping and interconnected support needs (mental health, addictions, prison stays, experiences of care, trauma, lack of trust) of people experiencing homelessness when policies are made, or services are commissioned. There was a clear statement that too often homelessness is caused or prolonged by factors other than a lack of housing.

4.1.4 There is some good evidence of strong local partnership working on homelessness and rough sleeping but there is room for improvement

It was reported by research participants that effective partnership working resulted in more efficient use of resources, reduced duplication of services, and enhanced communication, decision-making, and coordination among stakeholders.

Different local authorities have adopted varied approaches to partnership working, utilising a mix of internal staff and external providers. It was suggested by local authorities and their delivery partners that bringing together community, social and third sector organisations, as well as other councils and wider public services, was critical to the successful delivery of activities. As part of this, oversight mechanisms such as internal coordinators (i.e. staff within the local authorities who liaised between the local authority staff and external stakeholders or delivery partners), cross-cutting programme boards, and regular progress meetings were described as being key to the effectiveness of partnership working.

While local authorities and their delivery partners reported positive examples of partnership working, there was a consensus that further integration of services and funding could enhance the effectiveness of their work to tackle and prevent homelessness and rough sleeping. In particular, it was suggested that bringing all relevant services together, including health, housing, social care, police and crime, and pooling funding, would support more integrated partnership working, fostering a collaborative approach to tackling homelessness where everyone has a clear stake in achieving the desired outcomes.

At a local level, interviewees reported on some instances of councils working together, and with other agencies, to share approaches and services and sometimes pool funding where appropriate. However, it was noted that this is frequently complicated by the multiple support needs of those at risk of or experiencing homelessness, requiring a more cohesive, and holistic support intervention than is often available, especially given the capacity and resourcing issues in many of those wider support services.

Competing policy priorities for partners and misaligned funding programmes and priorities may also mean that despite their willingness, they find it hard to argue to do more for people experiencing homelessness in the area. Partnership working with health can be confounded by the misalignment of local authority housing services and health systems, reducing their capacity to work together effectively. Also, local authority interviewees noted that health partners tend to have to focus on immediate pressures, like hospital discharge, rather than on meeting longer-term health improvement needs or improving the capacity of much needed community-based mental health services.

Key enablers to improve partnership working were stated to include more flexible and pooled funding, increased capacity for delivering shared support services by establishing specialist roles, and regular communication with working partners.

4.1.5 Despite rising demand and numbers, the main homelessness and rough sleeping programmes were viewed as effective

The preliminary findings of the programme evaluations suggest that the programmes were successful in increasing the overall number of people who were able to access accommodation and support to prevent or resolve their homelessness or rough sleeping. However, a full impact evaluation making use of suitable quantitative data and a robust counterfactual would be needed to verify this claim.

The Rough Sleeping Accommodation Programme was developed to address the need for long-term, affordable and stable housing for people with complex needs who may have had a long-term experience of sleeping rough. It was considered to have delivered well against its outcomes in the local authority areas interviewed, with the provision of good quality, wraparound support for those accommodated a key success factor.

The Homelessness Prevention Grant was born out of a shared understanding that it is better for the individual and the public purse to prevent a person from experiencing homelessness in the first place than to intervene in a crisis. This evaluation has shown that while local authorities are making good use of the Homelessness Prevention Grant to intervene, critically to meet their statutory duty to offer emergency accommodation to those who are homeless and in priority need, a very limited share of the funding is spent on upstream universal or targeted prevention activity outside of the 56-day prevention duty or on improving the operation of the local system, as perhaps envisaged. This was linked clearly to the challenges associated with rising levels of demand from those who are already homeless. It was also acknowledged by research participants that the relief of homelessness is also meant to be a key function of the funding and to that extent it is being spent as intended.

The Rough Sleeping Initiative exists to prevent rough sleeping, to ensure people already rough sleeping can be helped quickly and do not return to the streets and to support local authorities to maintain the system and structures required to achieve outcomes and make progress. Four out of five local authorities reported that the Rough Sleeping Initiative had somewhat contributed to reducing the number of people experiencing rough sleeping; while one local authority reported that the Rough Sleeping Initiative has significantly contributed to this metric. This was largely due to the availability and success of off-the-street accommodation and outreach services.

In some cases, the accommodation and wraparound support provided under the Rough Sleeping Initiative, the Rough Sleeping Accommodation Programme and the Homelessness Prevention Grant enabled beneficiaries to also access support from wider services, such as healthcare, employment, and education. It was noted that outcomes were often bolstered by effective partnership working at the local level between local authorities, registered providers, charities, and community organisations. Without the Homelessness Prevention Grant, the Rough Sleeping Accommodation Programme, and Rough Sleeping Initiative funding, local authorities reflected that there would be:

significantly higher rates of rough sleeping, homelessness presentations, and temporary accommodation placements.

However, these positive changes were reported in the context of an increasing flow of people experiencing, or at risk of experiencing homelessness and rough sleeping due to wider structural drivers that were beyond the ability of local authorities to tackle.

4.1.6 Measuring success of Departmental funded programmes

The evaluation has also highlighted that local authorities, at times, used funding from the three homelessness and rough sleeping programmes interchangeably – and in combination with other sources of funding. This was considered positive by research participants as it allowed funding to be tailored to the needs of local people, supported a degree of cross-cutting working, and enabled local authorities to respond to surges of demand where necessary.

However, it also contributed to the challenge of measuring the success of the programmes, given that programme beneficiaries were supported by services funded across all three programmes and with other sources of funding mixed in.

Local authorities regularly monitor and report rough sleeping and homelessness metrics to MHCLG via tools such as the H-CLIC performance dashboard.^{[footnote 25]} However, there is a lack of clear alignment between these metrics and the programme activities. It was noted from the evaluation that the focus of the metrics was on outputs. There appears to be a need to address the lack of outcome and impact data against the programme terms and conditions. The absence of this is another factor which makes it difficult to assess whether the programmes are successfully delivering the intended activities and whether they are having the intended impact. This is more true for the Homelessness Prevention Grant than the Rough Sleeping Initiative and the Rough Sleeping Accommodation Programme, where there are delivery plans and funding agreements with some negotiated targets.

From this evaluation, it is evident that the flexible design of programmes, and the welcome ability that local authorities have to flex interventions according to changing needs, makes monitoring of success difficult. The lack of an obvious counterfactual also makes it difficult to measure impact and value. This relates to the fact that the Homelessness Prevention Grant, the Rough Sleeping Initiative and the Rough Sleeping Accommodation Programme are available across England and most local authorities have received grants to support activities to reduce or prevent homelessness and rough sleeping in their local areas. It is therefore challenging to identify or select specific local authorities that were not allocated funding for one or more of these programmes but are matched on other criteria such as type, size, geography, and complexity to the funded local authorities. The absence of such a counterfactual group of local authorities makes it challenging to measure impact and value for money, when all other influencing factors are considered.

4.1.7 Unintended consequences

During the course of this phase of the evaluation, some issues came up that could best be described as unintended consequences of the activity of the programmes or the way in which the homelessness and rough sleeping system currently operates.

Multiple interviewees suggested that the increased availability of Rough Sleeping Initiative funded services at a time when access to housing for single people was made difficult by market conditions, might have created a perverse incentive for some people to sleep rough as a means of accessing support with their housing needs, despite not actually being roofless. It was felt that making space to offer preventative support for single people in housing need who approach services before they were roofless was needed moving forward.

It was also noted on multiple occasions that the high quality of the Rough Sleeping Accommodation Programme properties made some beneficiaries reluctant to move to other housing options, typically offered in the private rented sector and which were of lower quality, when their time in the Rough Sleeping Accommodation Programme units ended, thus creating move on difficulties.

The increased funding for homelessness and rough sleeping given to local authorities at the same time (ie: at the start of a new programme cycle) has led to recruitment difficulties due to multiple councils or providers seeking to recruit to the same type of job roles at the same time. This was felt to be particularly challenging when roles were more specialist (eg: in drug and alcohol services or in housing options).

4.2 Considerations for the future

This report relates to the first phase of the systems-wide evaluation of homelessness and rough sleeping. This phase of the work was completed before the end of June and before the General Election in July 2024. Work on examining some of the themes and issues is continuing, however it is possible to highlight at this preliminary stage some considerations for the future for government, for local authorities, and for others involved in the work to prevent and tackle homelessness and rough sleeping. These are set out below:

4.2.1 The importance of a collaborative cross-government strategy

There was clear agreement that there would be value in the collaborative development and implementation of a cross-government strategy focused on ending homelessness in its widest sense, including but not limited to rough sleeping. Given the findings outlined in this report, it would be important to both take a more ‘upstream’ public health preventative approach to tackling this problem, and to have collective ownership of that strategy across government. There is a need for a clear leadership role that can operate across departments to deliver on this. This would include agreeing a clear set of shared outcomes, navigating and coordinating competing priorities, looking for policy gaps and conflicts, and driving a cohesive cross-departmental approach.

Preventative programmes, activities or funding streams sitting within other government departments which are used to alleviate or prevent homelessness and/or are acting as drivers of homelessness could play a bigger system role in supporting homelessness prevention. Examples include Discretionary Housing Payments (Department for Work and Pensions), advice funded through Legal Aid (Ministry of Justice), community accommodation for prison leavers (Ministry of Justice), community mental health support (Department of Health and Social Care), resettlement support for those leaving the asylum support system (Home Office).

4.2.2 Programme and funding consolidation

MHCLG could consider consolidating funding for Homelessness Prevention Grant, the Rough Sleeping Accommodation Programme, the Rough Sleeping Initiative and some of the other programmes analysed as part of this research into a single fund which could be used flexibly to effectively prevent and address homelessness and rough sleeping. Such a fund, allocated to local authorities for them to design interventions relevant to local needs, would need a clear outcomes framework.

Within any combined funding stream, it would be important to safeguard funding for specific activities, such as homelessness prevention, and expectations regarding this should be made clear by MHCLG when allocating the funding. Local authorities would need to clearly articulate the expected outcomes (within a local homelessness strategy) and provide robust monitoring processes to check these are being achieved.

This approach has the potential to help local areas to enjoy some flexibility to meet urgent needs and avoid silos for people who are experiencing homelessness who have to try and fit their needs under a specific programme to get help. At the same time, it would encourage local authorities to maintain a sustained focus on prevention while also supporting accommodation and crisis relief efforts. A key requirement for this approach to be successful would be to ensure that activities clearly map to the desired outcomes, and it should not be overlooked that combining funding streams could also present new risks that would need careful consideration in the set-up.

There was limited evidence of local authorities pooling funding across local authority boundaries and given the economies of scale that may be possible, the mobility of the client groups being served and the issues around how geographical footprints across health, probation, housing are sometimes misaligned, consideration could be given to ways of incentivising joint commissioning within the allocation of funding. Combined authorities offer one possibility for this, as do existing local authority sub-regions and other groupings.

4.2.3 Better evidence of what works

If the standalone homelessness and rough sleeping programmes remain in place, it could be beneficial to review the current processes for designing objectives per programme and in particular how these are being used to inform reporting on progress. Ensuring that there are clear and measurable objectives could improve accountability and support a clear vision for what each programme funding stream is looking to achieve. There could be benefit in reviewing the current metrics and mapping these to programme objectives. It was noted from the evaluation that the focus of the metrics was on outputs rather than outcomes.

There also appears to be a need to address the lack of impact data against the programme terms and conditions. It is recognised that evidencing outcomes may require multiple partners to be involved and contributing to these. Better evidence would require stronger partnership working agreements that reflect the focus on delivering local outcomes. If the suggestion were taken up to create a more integrated approach to ending homelessness with a combined funding programme, the review of metrics and the suggestion of putting in place a monitoring framework that would support the collection of impact data would be even more valid.

Given the system-level interactions and overlaps, care should be taken to ensure that any wider outcomes framework monitors the system-level drivers of homelessness and rough sleeping that have been determined during this first phase of the research so that the issues about housing affordability are considered alongside drivers which may sit outside of MHCLG’s remit, like financial insecurity, the operation of the benefits system, unemployment, limited access to specialist support services (especially mental health), and institutional discharge (eg: from asylum support accommodation, prison, care).

For reasons outlined in the main body of the report, local authorities have limited experience in designing and delivering homelessness prevention activities. A priority should be to showcase examples of what is working across the country and ensuring that the system can collectively learn from these. The 2019 report on the Homelessness Prevention Trailblazers work found that doing more ‘upstream’ prevention activities had no detectable impact on the level of statutory homelessness.^{[footnote 26]} This finding was not surprising given that evidence of impact would only be expected further in the future, and this was not part of the study. It does further illustrate the importance of using evaluation and longitudinal analysis to understand where and how, in the journey into homelessness, prevention interventions can be effective.

4.2.4 Better local partnership working

It could be beneficial to explore opportunities for how MHCLG can encourage more effective local partnership working. This could cover setting out examples of how partnership working has been effective in addressing homelessness and what contributed to this happening. An effective approach might be facilitating opportunities (eg: webinars) for partners to share their experiences of what works well in their own context. By following these recommendations, local authorities can build on existing successes and further strengthen their partnership working, leading to more effective and efficient programme delivery.

These considerations for the future are proposed based on preliminary findings and will require testing with national and local stakeholders in future fieldwork in support of this continuing evaluation.

Question 7

Appendix 1 – Detailed Methodology

Accepted Answer

A1. Core system map development

The core system map for the evaluation was developed through three key stages:

review of existing Centre for Homelessness Impact (CHI) homelessness system map and identification of adaptations required to create a new core system map aligning with the scope of the evaluation;
initial system mapping workshop gathering insight from homelessness and rough sleeping experts to inform adaptations to the CHI map, and subsequent development of a draft core system map;
sense-testing of the draft core system map, through a workshop with homelessness and rough sleeping experts and written feedback, and subsequent revisions and finalisation of the core system map.

A1.1 Review of Centre for Homelessness Impact map

The consortium lead (Centre for Homelessness Impact) has previously developed a system map of homelessness, based on extensive consultation with stakeholders in the homelessness and rough sleeping system across England. It provides a detailed and nuanced picture of the interconnected causes and consequences of homelessness. This map has enabled us to make a quick start on the system mapping for the evaluation, by using the existing map as a basis for developing one that is applicable to the system-wide evaluation.

Through a review of the Centre for Homelessness Impact map, the consortium identified two key adaptations that would be required to align with the scope of the evaluation.

The first was to refine the map to focus on factors within the sphere of influence of MHCLG and other government departments. The Centre for Homelessness Impact map identifies a broad range of factors relating to the causes and consequences of homelessness, some of which are outside of the sphere of influence of MHCLG and other government departments, and therefore outside of the scope of the evaluation. A priority for the development of the core map was therefore to determine which factors on the Centre for Homelessness Impact map are within and outside of the scope of the evaluation and needed to be removed or refined.

The second was to expand and densify the broad factors that encompass multiple causes and consequences of homelessness. Consortium partners identified three factors in the Centre for Homelessness Impact map that require a greater level of detail to allow for their thorough interrogation through the course of the evaluation. These were: prevention services; crisis relief services; and health/ capabilities/ wellbeing

The evaluation must continually bear in mind the wider system depicted in the Centre for Homelessness Impact map. Using it as the basis for the development of the core system map for the evaluation will allow us to understand how the homelessness and rough sleeping system, as defined by the evaluation, fits within the wider system of causes and consequences of homelessness, some of which may be outside of the sphere of influence of MHCLG and other government departments.

A1.2 Initial system mapping workshop and map development

In December 2023, the consortium convened an initial system mapping workshop, drawing on Centre for Homelessness Impact’s strong stakeholder relationships to ensure a high level of engagement among homelessness and rough sleeping stakeholders with a strong insight into the homelessness and rough sleeping system. In total, 25 key stakeholders from the following groups attended the workshop:

homelessness officials from local authorities;
officials from MHCLG homelessness and rough sleeping team;
officials from other government departments, including Health and Justice;
homelessness and rough sleeping expert advisors;
systems expert advisor;
representatives from the homelessness sector.

The objectives of the workshop were to introduce the system-wide evaluation, explore with stakeholders why a systems approach is needed for homelessness and rough sleeping and frame and refine the Centre for Homelessness Impact map by determining which factors on the Centre for Homelessness Impact map are within and outside of the evaluation scope and expanding three broad areas of the homelessness system identified in the Centre for Homelessness Impact map: prevention; crisis relief; and health, wellbeing and capabilities.

A1.3 Determining factors within the scope of the evaluation

Workshop participants were organised into four groups, aligning with the four elements of the homelessness and rough sleeping system (statutory homelessness and rough sleeping policies funded by MHCLG; wider housing related policies funded by MHCLG; policies owned by other government departments that influence or are influenced by homelessness and rough sleeping; and policies undertaken by or relating to voluntary and community sector organisations). Participants identified factors on the Centre for Homelessness Impact map over which their group has influence and provided explanation and evidence to support this.

Evidence gathered through this exercise helped us to identify factors in the Centre for Homelessness Impact map that are within the sphere of influence of the four elements, and therefore within the scope of the evaluation. It also provided further detail on these factors and the ways in which they are influenced by stakeholders working within the homelessness and rough sleeping system, thereby increasing the utility of the core system map, and providing a database for the development of subsequent focused maps.

A1.4 Expansion of key nodes / areas of activity

Workshop participants were then organised into three groups, aligning with three factors within the Centre for Homelessness Impact map which encompass multiple factors and therefore require further interrogation:

crisis relief services;
prevention services;
health/capabilities/wellbeing.

Workshop participants identified types of services provided in each of these domains, and detailed the nature, intended beneficiaries, and delivery and funding partners for each of these.

Evidence gathered through this exercise has allowed us to expand the broad, high-level factors listed above and incorporate more nuanced factors into the draft core system map, which sit within these critical areas of homelessness and rough sleeping activity. Again, it has also provided an initial database of services operating within these domains, which can be utilised in the development of subsequent focused maps.

A1.5 Initial draft core system map development

Data from the initial system mapping workshop, combined with homelessness and rough sleeping and systems expertise of the consortium, informed the removal, adaptation and addition of nodes and connections to the existing Centre for Homelessness Impact map.

Table A1 below shows the criteria used during the workshop in determining whether nodes should be retained, amended, deleted or added during the development of the systems map.

Table A1: Criteria for determining whether to retain, amend, delete or add nodes in the systems map

Retained	Identified as relevant, and aligned with the scope of the evaluation and the consortium’s homelessness and rough sleeping and systems knowledge and expertise
Amended	Identified as relevant, and either data from the workshop or the consortium’s homelessness and rough sleeping and systems knowledge and expertise deemed an amendment to the existing wording to be necessary.
Deleted	Identified as not relevant, and this aligned with the scope of the evaluation and the consortium’s homelessness and rough sleeping and systems knowledge and expertise
Added	identified as necessary to fill key gaps, aligned with the consortium’s homelessness and rough sleeping and systems knowledge and expertise and within scope of the evaluation

Once nodes were finalised, the consortium’s homelessness and rough sleeping and systems knowledge and expertise was used to amend existing connections and identify new connections on the draft core system map.

A1.6 Core system map sense-testing and revision

A virtual workshop was convened in February 2024 with a subset of participants from the initial workshop to sense-test the draft core system map and generate feedback to inform further refinements. To avoid overburdening stakeholders, the remaining participants from the initial workshop were asked for feedback via email.

The draft map was shared with participants prior to the workshop, along with information regarding the map’s purpose and scope, instructions for navigating the map, and guidance on the feedback the workshop will be gathering. Feedback from this workshop was used to identify further refinements which were then incorporated to produce the final version of the core system map.

A2. Intervention Level Framework classification

The researchers collected details of all the major homelessness and rough sleeping (homelessness and rough sleeping) policies, programmes and interventions being funded by the government and others. This totalled 30 policies, programmes and interventions. One intervention, the voluntary and community sector capacity grant, is split into four lots, each of which operates at different levels of the system, giving a total of 33 interventions to classify.

These were split into categories based on whether they were directly funded by MHCLG, funded by other government departments, or funded by a mix of government and others.

For each of these policies, programmes and interventions, policy documents were sourced and data extracted on key elements:

1. name;

2. funder and funding amount;

3. timescale;

4. delivery organisation(s);

5. detail of policy, programme or intervention.

These details were used to assign each policy or programme to an Intervention Level Framework category. This was done on the basis of published documents only, focusing on how the policies, programmes and interventions were intended to be implemented, rather than how they have been implemented in practice. The categorisation was conducted by one researcher, with conflicts or doubts discussed with two senior evaluation team members. Finally, the categorisations were validated by a senior researcher. The details of this are contained in Appendix 2.

A3. Qualitative interviews and local areas

Data collection: Semi-structured interviews were conducted with local authority representatives and delivery partners who had been involved in the delivery of the Homelessness Prevention Grant, the Rough Sleeping Accommodation Programme and the Rough Sleeping Initiative. In total, the evaluation comprised five programme-level interviews and included a representation of 40 individuals across the five local areas. A trained member of the evaluation team conducted virtual interviews. With consent, the content of the interviews was transcribed.

A topic guide was used to direct conversation and ensure all themes were covered comprehensively. The interview structure covered the following topics:

1. programme activities and inputs;

2. outcomes and impacts of the programme;

3. the homelessness and rough sleeping system and broader context;

4. cost and value for money;

5. learnings, challenges and successes.

Analysis: Anonymised interview data was analysed thematically and aligned to questions in the topic guide. Deductive thematic analysis was applied through a coding framework. The analysis was conducted in Microsoft Excel (linking of verbatim data extracts to relevant themes and sub-topics in the framework) and aligned to principles described by Braun and Clarke. A gap analysis was undertaken to identify questions and responses that may not have been fully addressed or required further clarification. Email follow-up communication was undertaken with the relevant interviewee to address specific queries.

A4. Qualitative interviews with policy leads from across central government

Data collection: Semi-structured interviews were conducted with identified policy leads within MHCLG and other government departments. A specific interview topic guide was prepared. These interviews captured the perspectives of 12 individuals from across six government departments, including MHCLG, the Ministry of Justice, the Office of Health Improvement and Disparities, His Majesty’s Treasury, the Department for Work and Pensions, and the Home Office. A trained member of the evaluation team conducted virtual interviews. With consent, the content of the interviews was transcribed.

A topic guide was used to direct conversation and ensure all relevant themes relating to the programme were covered comprehensively. The interview structure covered the homelessness and rough sleeping system and broader context.

Analysis: Anonymised interview data was analysed thematically and aligned to questions in the topic guide.

A5. Focus groups

Data collection: Semi-structured focus groups were conducted with local authority representatives and delivery partners who had been involved in the delivery of the Homelessness Prevention Grant, the Rough Sleeping Accommodation Programme and the Rough Sleeping Initiative. These were undertaken in person and across five local areas. In total, the evaluation comprised three focus groups and included a representation of 13 individuals. A trained member of the evaluation team conducted focus groups, and with consent, the content was transcribed. A focus group topic guide was used to direct conversation and maintain relevance to the research questions. Focus groups covered the following topics:

1. programme activities and inputs;

2. outcomes and impacts of each programme;

3. learnings, challenges and successes.

Analysis: The data gathered was synthesised thematically and analysed using qualitative analysis, which is in line with our approach for interview analysis.

A6. Interviews with people with lived experience

Data collection: Semi-structured interviews were conducted with people with lived experience of homelessness. These were undertaken in a range of forums, including in person, via the telephone or MS Teams. In total, the evaluation included a representation of eight people with lived experience. A trained member of the evaluation team conducted the interview, and with consent, the content was transcribed. A topic guide was used to direct conversation and maintain relevance to the research questions.

Analysis: The data gathered was synthesised thematically and analysed using qualitative analysis, which is in line with our approach for interview analysis.

A7. Local authority survey

Data Collection: A survey was conducted online via the SmartSurvey platform. A webpage link to the survey was disseminated to key leads in each of the five sampled local authorities. These leads then coordinated the completion in their local area, including a number of relevant individuals in the process to ensure that a depth of information was collected to answer the research questions and to ensure a range of perspectives were represented. The survey included questions on the following areas:

activities delivered and beneficiaries of the programme;
funding of the programme, including whether funding covered programme costs;
outcomes and impact of the programme on specific homelessness and rough sleeping measures.

Analysis: Survey questions were mapped to process, impact, and economic categories to help answer the research questions related to each of these areas. Descriptive analysis was then conducted in Microsoft Excel for the quantitative data to assess the programme’s outcomes and impacts and whether it has been delivered as intended.

A8. Secondary quantitative data collation and analysis

Data collection: Secondary data from publicly available sources was collated for the monitoring framework to support the evaluation of Homelessness Prevention Grant, Rough Sleeping Accommodation Programme and the Rough Sleeping Initiative. Table A1 outlines the datasets included, alongside the data source and the years of data collected. The data was collated for each of the five local authorities, where possible, on a longitudinal basis.

Analysis: The longitudinal data was reviewed, where appropriate, to see whether there was strong convergence or divergence between the secondary data and the findings from other research methods. A descriptive analysis was conducted using Microsoft Excel.

Table A1: Years of data collected and the source of secondary quantitative datasets

Dataset	Source	Years of data collected
H-CLIC	Gov.uk website	2020/21 - 2022/23
Rough Sleeping Snapshot in England	Gov.uk website	2020/21 - 2023/24
Ending Rough Sleeping Data Framework	Gov.uk website	2021/22 - 2022/23
Continuous Recording of Social Housing Lettings and Sales (CORE)	Gov.uk website	2020/21 - 2022/23
Local authority revenue expenditure and financing: Homelessness services and expenditure	Gov.uk website	2020/21 - 2022/23
Children looked after in England	Gov.uk website	2020/21 – 2023/24
Dwelling Stock (including vacant)	Gov.uk website	2020/21 - 2022/23
Local authority Housing Statistics	Gov.uk website	2020/21 - 2022/23
Housing Supply: net additional dwellings	Gov.uk website	2020/21 - 2022/23

Question 8

Appendix 2 – Categorisation of programmes and interventions according to the Intervention Level Framework

Accepted Answer

This appendix sets out the funding programmes or initiatives operating in the homelessness and rough sleeping system and then categorised using the Interventions Level Framework.

Group 1: funded by MHCLG – homelessness and rough sleeping directorate

1. Homelessness Prevention Grant

Funded by MHCLG. See main body of report for full details of past and current funding. Introduced in 2021. Delivered by local authorities. The components of the funding are allocated through different formulas which consider the local need for prevention, temporary accommodation and the cost of providing housing services. The grant aims to change the structure of the system by empowering local authorities to flexibly spend (ringfenced) funding on prevention-related activity, though in reality it is operating more at the level of structural elements.

ILF Categorisation - System structure

2. The Rough Sleeping Initiative

Funded by MHCLG. See main body of report for full details of past and current funding. This iteration of the programme was introduced in 2018. Delivered by local authorities and their commissioned delivery partners. Funding is allocated according to data on rough sleeping levels. The Rough Sleeping Initiative focuses predominantly on street-based interventions for people experiencing homelessness, though it can also be spent on prevention.

ILF Categorisation - Structural elements

3. The Rough Sleeping Accommodation Programme

Funded by MHCLG. See main body of the report for full details of past and current funding. Launched in 2021. Delivered by local authorities and their commissioned delivery partners. The programme provides the funding for local authorities to secure long-term supported housing and support with an intended move on into independent housing.

ILF Categorisation - Structural elements

4. The Housing Advisors Programme

Funded by MHCLG, £20,000 provided to each successful project. Launched in 2017 and held six rounds of funding with the final round awarded in the 2022/23 financial year. Delivered by local authorities and partner organisations, supported by the Local Government Association. This programme facilitates organising shared learning across the wider sector which in turn changes the approaches taken up by local authorities and other delivery organisations across the homelessness and rough sleeping system. Outcomes of these approaches are fed back and can adapt the learning shared.

ILF Categorisation - Feedback & delays

5. Capital Letters

Funded by MHCLG and London local authorities. Capital Letters has been set up as a non-profit organisation with the goal of becoming self-financed in the long-term. The organisation launched in 2019 and is ongoing. Their most recent corporate strategy dates to 2028. Capital Letters intends to change the entire homelessness and rough sleeping system by creating new connections between London boroughs and altering the housing landscape.

ILF Categorisation - System structure

6. Accommodation for Ex-Offenders programme

Funded by MHCLG. Phase 1 funding ranged from £4,800 to £470,842. Phase 2 funding ranged from £21k to £1.5m. Phase 2 launched in April 2023 and will run until March 2025. Delivered by local authorities across 87 final schemes. These include the 33 local authorities found to have the highest need. The programme has the greatest impact on the criminal justice subsystem within the wider homelessness and rough sleeping system as well as on prison leavers specifically.

ILF Categorisation – Structural elements

7. Night Shelter Transformation Fund

Funded by grant funding agreement, with £13.3 million set aside for funding between 2022 to 2025, £3.3 million of which was added during round 2 of the fund. The fund is opened bi-annually. Delivery organisations can apply for short-term, one year or multi-year funding awards. Funding is provided to Night Shelter providers from the faith and community sector. The bidding process was organised by Homeless Link and Housing Justice. The Night Shelter Transformation Fund has the greatest impact on the faith and community sector subsystem within the homelessness and rough sleeping system and more specifically, organisations providing short-term support via temporary accommodation.

ILF Categorisation - Structural elements

8. StreetLink

Principally funded by MHCLG with additional funding from the Greater London Authority and the Welsh government. First created in 2012 and relaunched in September 2023. Individuals alert StreetLink of people sleeping rough, with alerts passed along to outreach teams with the intention better connecting people experiencing rough sleeping to appropriate services. StreetLink updates individuals using the platform with the outcomes of their alert, potentially incentivising continued use of the platform and people sleeping rough accessing services.

ILF Categorisation - Feedback & delays

9. Housing First

Funded by MHCLG, £28 million total funding originally committed in 2017. Delivery launched in 2019, and funding has extended to March 2025. Delivered in three regions: Greater Manchester, Liverpool and West Midlands. Housing First looks to change the aim of the homelessness and rough sleeping system by re-defining who is deemed eligible for housing.

ILF Categorisation – Goals

10. Single Homelessness Accommodation Programme

Funded by MHCLG, £200 million total funding to allocate across councils. Bidding opened January 2023 and ran until March 2025. Outside London, Homes England is the delivery partner, and within London, the Greater London Authority. Councils can deliver the programme themselves or work with housing associations, charities and other organisations to do so. The central aim of this programme is to increase the supply of long-term affordable housing, in essence increasing the physical elements or capital within the homelessness and rough sleeping system.

ILF Categorisation - Structural elements

11. Supported Housing Improvement Programme (previously Supported Housing Oversight Pilots)

Funded by MHCLG, with allocations less than £1 million in total across the 3 years but in some instances, it may be more (capped at £2 million). Operational over financial years 2022 – 2025. Delivered by local authorities including of the original pilot local authorities (Birmingham, Blackpool, Blackburn with Darwen and Hull). This programme has the greatest impact on the supported housing subsystem within the homelessness and rough sleeping system.

ILF Categorisation - Structural elements

12. Move On Fund

Funded by MHCLG, £44 million grant funding and £6 million revenue funding to cover ongoing tenancy costs. The Greater London Authority added £50 million to fund schemes in Greater London using capital grant funding and revenue funding to also cover ongoing tenancy costs. The fund prospectus was published August 2017 and schemes were due to have been completed by March 2021. Delivery was led by organisations qualified as Homes England investment partners. This includes housing associations, local authorities, private sector developers and community groups, among others. The Move On Fund looks to impact the short-term housing subsystem within the homelessness and rough sleeping system.

ILF Categorisation – Structural elements

13. Next Steps Accommodation Programme

Funded by MHCLG, total shorter-term/interim accommodation and immediate support funding was £105 million. Total longer-term move-on accommodation funding was £161 million (£130 million capital funding and £31 million revenue funding). Launched in 2020 with funding to be used by March 2021. Long-term accommodation and support was administered by the Greater London Authority in Greater London. Immediate and interim support in Greater London and across England was administered by MHCLG, as was long-term accommodation and support outside of Greater London. This programme looks to increase short- and long-term housing stock, the physical elements or capital within the homelessness and rough sleeping system.

ILF Categorisation – Structural elements

14. Changing Futures (pilots)

Funded by MHCLG, £55.4 million in funding from the government’s Shared Outcomes Fund and £21.6 million from the National Lottery Community Fund. Announced in 2020 with delivery beginning July 2021 until March 2025. Delivered by 15 local authorities working with partner organisations. Changing Futures changes linkages across a range of supportive services and subsystems with the intention of shifting the behaviour of the homelessness and rough sleeping system as a whole.

ILF Categorisation - System structure

15. Voluntary and Community Sector Capacity Grant

Funded by MHCLG, grant funding agreement. £7.3 million total funding. The grant is managed by Shelter, Homeless Link and Housing Justice. Grantees are voluntary and community sector organisations who provide specialist service delivery within the homelessness and rough sleeping sector. Lot 1 Shelter seeks to increase the capacity and capability of the voluntary and community sector subsystem. Lot 2 Shelter is largely providing specialist advice to actors within the voluntary and community sector subsystem. Lot 3 Homeless Link is placed under system structure. Lot 4 Housing Justice is placed under structural elements due to largely affecting the voluntary and community sector subsystem, developing the capabilities of actors within the voluntary and community sector subsystem.

ILF Categorisation – a mix of structural elements and system structure

16. National Homelessness Advice Service

Funded by MHCLG, grant funding agreement, ongoing. Delivered by National Homelessness Advice Service. The National Homelessness Advice Services provides information to organisations giving housing or homelessness advice. This impacts the type of advice provided and in turn the outcomes of the service users seeking advice.

ILF Categorisation - Feedback & delays

Group 2: funded by MHCLG – other directorates

17. Respite Rooms programme

Funded by MHCLG – Domestic Abuse, total funding of £5.4 million which ranged from £118k to £491k. Operated from October 2021 to March 2023. Delivered by local authorities and partner organisations. This programme specifically impacts people sleeping rough affected by domestic violence and sexual violence and looks to increase short-term housing stock, a physical element or capital of the homelessness and rough sleeping system.

ILF Categorisation - Structural elements

18. Local Authority Housing Fund

Funded by MHCLG, funding of £500 million for round 1 and £250 million for round 2. Round 1 was announced in December 2022 and ran until March 2023. Round 2 was announced in March 2023 and will run until March 2024. Delivered by local authorities, but involvement of housing association partners suggested. This fund looks to increase the stock of short- and long-term housing, a physical element or capital within the homelessness and rough sleeping system.

ILF Categorisation - Structural elements

19. Affordable Homes Programme

Funded by MHCLG, £11.5 billion total funding with £7.39 billion to Homes England and £4 billion to the Greater London Authority to deliver affordable homes. Runs from 2021 – 2026 with full completions expected in 2028/29. Managed in Greater London, Greater Manchester and In the West Midlands by the Combined Authorities.

Everywhere else in England is managed by Homes England, the UK government’s housing accelerator. Eligible delivery organisations include housing associations, local authorities, developers, for-profit providers, community-led organisations and others interested in developing affordable homes. This programme looks to increase the stock of affordable housing, a physical element or capital within the homelessness and rough sleeping system.

ILF Categorisation - Structural elements

Group 3: funded by other government departments

20. Community Accommodation Service Tier 3

Funded by Ministry of Justice, the level of funding available for the service is unclear. Launched in July 2021 within 5 ‘vanguard’ regions. Since rolled nationally across 12 regions. Delivered by a variety of different housing suppliers. This programme has the greatest impact on the criminal justice subsystem within the homelessness and rough sleeping system and prison leavers specifically.

ILF Categorisation - Structural elements

21. Staying Close

Funded by Department for Education, £36 million of total funding ranging from £22,000 to £32,000 per care leaver. Piloted in 5 local authorities beginning in 2017/18. 15 additional LAs were then funded in 2022/23. The Department for Education is now looking to increase the number of LAs in 2024/25. Delivered by local authorities and partner organisations. Private children’s homes, charities and other voluntary organisations are not eligible to apply but can partner with local authorities. Staying Close has the greatest impact on care leavers and the care subsystem within the homelessness and rough sleeping system.

ILF Categorisation - Structural elements

22. Staying Put

Funded by Department for Education, funding ranges from £9,294 for Rutland local authority to £3,141,597 for Croydon local authority. Began in 2017/18. Grant funding has been set up for local authorities up until 2025. Delivered by local authorities and partner organisations. Similar to Staying Close, Staying Put has the greatest impact on the care subsystem within the homelessness and rough sleeping system but specifically for former foster carers.

ILF Categorisation - Structural elements

23. Out of hospital care funding (also called Out-of-Hospital Care Model)

Funded by Department of Health and Social Care; MHCLG and Ministry of Justice, £16 million of funding from the Shared Outcomes Fund. Launched in 2020. There were 17 out of hospital care funding pilots delivered by local authorities. Delivered by Integrated Care Systems and partner organisations across England. The fund for out-of-hospital care affects the healthcare subsystem within the homelessness and rough sleeping system and more specifically individuals experiencing homelessness entering into hospital.

ILF Categorisation - Structural elements

24. Homelessness Escalations Service

Funded by Home Office. The Home Office merged the Rough Sleeping Support Service, and an interim ‘escalation’ process to form the Homelessness Escalations Service in June 2023. It remains in operation. Delivered by UK Visas and Immigration as of June 2023. Previously under Immigration Enforcement as the Rough Sleeping Support Team. This service provides information on non-UK nationals which then dictates their entry into or out of various homelessness and rough sleeping subsystems.

ILF Categorisation - Feedback & delays

25. Rough Sleeping Drug and Alcohol Treatment Grant

Funded by Office for Health Improvement and Disparities, funding for 2023/24 ranged from £244,757 for Richmond upon Thames to £1,903,739 for Westminster. Launched in 2020 and will run until 2025. Delivered by local authorities and partner organisations. This grant has the greatest effect on the drug and alcohol treatment subsystem within the wider homelessness and rough sleeping system.

ILF Categorisation - Structural elements

26. Housing Support Grant

Funded by Department of Health and Social Care, £53 million in total funding. Funding ranges from £1,187,650 for Lincolnshire County Council to £2,206,250 for Essex County Council. Funding allocated for financial years 2022/23, 2023/24 and 2024/25. Delivered by Local authorities and partner organisations. Based on the intended actions of the Housing Support Grant, the Grant has the greatest effect on the drug and alcohol treatment subsystem within the wider homelessness and rough sleeping system.

ILF Categorisation - Structural elements

27. Employment covenant funding

Funded by Department for Work and Pensions. Launched in 2023 and is ongoing. Delivered by Crisis. Employment covenant funding sets a new target across the system to create a fairer and more accessible employment landscape for people experiencing homelessness.

ILF Categorisation - Goals

28. Mental health support

Funded by Department of Health and Social Care. Cumulative £30 million investment until 2023/24 financial year. The NHS Long-Term Plan (2019) included the aim of establishing new specialist mental health provision for people sleeping rough by 2023/24. Delivered by NHS and local authorities and partner organisations within the following areas: Brent, Westminster and Camden, Somerset, Devon, Slough, Windsor and Maidenhead, Surrey Heartlands, Peterborough, Great Yarmouth, Hereford and Worcestershire, Shropshire and Telford and Wrekin, East Riding, Sheffield and Doncaster, Greater Manchester. Additional funding for mental health support birthed from the NHS Long-Term Plan has brought about a novel type of structural element in the form of multi-service outreach teams made up of local authority and NHS services who bring together services and staff across a number of subsystems.

ILF Categorisation - System structure

Group 4: government funding mixed with non-governmental funding

29. Homeless Link

Funded by Grant funding agreement, funded by a number of different sources in the form of grants and donations from central and local government; other public sector bodies; trusts; foundations; the National Lottery Community Fund; companies; and individuals. Founded in 2001 and is ongoing. Homeless Link aims to join people and organisations across the system together in order to influence policy, so can be seen as a change to system structures

ILF Categorisation - System structure

30. Big Society Capital

Funded by the four main UK high street banks and £15m of investment from the UK government in social investment pilots. Launched in 2012 and ongoing. Delivered by Big Society Capital Board and involving a novel type of structural element in the form of introducing a new way of funding homelessness and rough sleeping services.

ILF Categorisation - System structure

Cookies on GOV.UK

Applies to England

Foreword

Executive summary

Scope and purpose

Background

Research methods

Summary of findings

There are significant social, economic and external policy pressures impacting the homelessness and rough sleeping system

The majority of government activity and expenditure is on crisis relief, rather than prevention and delivered at a level which does not create system-level improvements

There are significant competing policy priorities within national government

There is some good evidence of strong local partnership working on homelessness and rough sleeping but there is room for improvement

Despite rising demand and numbers, the main homelessness and rough sleeping programmes were viewed as making a positive impact

Measuring success of MHCLG funded programmes

Unintended consequences

Conclusions

The importance of a whole-government approach

Building confidence in taking a preventative approach

Programme and funding consolidation

Better evidence of what works

Better local partnership working

This Report

Chapter 1: Introduction and Methodology

Chapter 2: The system affecting homelessness

Chapter 3: The role of three programmes in the homelessness and rough sleeping system

Chapter 4: Discussion, conclusion and next steps

Appendix 1 Detailed methodology

Appendix 2 Intervention Level Framework

Chapter 1. Introduction and methodology

1.1 Overview

1.2 Homelessness and rough sleeping in England

1.3 Systems approach to the evaluation

1.4 The objectives of the evaluation

1.5 Overview of the methodology

1.6 Limitations of the methodology

Chapter 2. The system affecting homelessness

2.1 Overview

2.2 Defining the homelessness and rough sleeping system and its boundaries

2.3 Initial analysis of the system map

Mental health and wellbeing

Financial resilience

The importance of effective services

2.4 An analysis of government homelessness and rough sleeping interventions

2.4.1 Introduction

2.4.2 Results of interventions review

2.5 The experiences of people with lived experience

2.5.1 Lack of preventative support

2.5.2 Challenges with mental health

2.5.3 Challenges with substance use

2.5.4 Impact of financial insecurity

2.5.5 Routes into stable housing

2.5.6 Barriers to accessing stable housing

2.5.7 Engagement with local authority homelessness services

2.5.8 Factors that might support improved homelessness and rough sleeping outcomes

2.6 The need for a systems approach

2.6.1 Housing Supply

2.6.2 Wider support needs

2.6.3 Focus on short term funding limiting sustainable outcomes

2.6.4 Limitations to central government cross-departmental collaboration

2.6.5 Variation at local authority level

2.6.6 Misalignment between local authority and health services

2.6.7 Local Authority funding and support

2.6.8 Positive examples of cross-departmental collaboration

2.6.9 Positive governmental interventions

2.6.9 Local Authority funding and support

2.7 System-wide conclusions

Chapter 3. Assessing the role of three main homelessness and rough sleeping programmes

3.1 Introduction

3.2 The Theory of Change and evaluation methodology

3.3 Brief summary of each programme

Unique rationale for the three programmes

3.4 Brief context of homelessness in each local area

3.5 Process evaluation

3.5.1 Delivery of activities

3.5.2 Structures and resources involved in delivery

3.5.3 Partnership working

3.5.4 Enablers and barriers

3.5.5 How does local context impact on delivery

3.5.6 How do system factors impact on delivery

3.6 Outcomes evaluation