Successive governments have promised to tackle the ‘root causes’ of social and economic disadvantage. Yet public spending on individuals experiencing problems such as addiction, homelessness, offending and poor mental health is still largely reactive, funding expensive crisis care services rather than coordinated and preventative support.1

In its final budget statement in 2015, the Coalition government committed to carrying out an assessment of how to reduce the estimated £4.3 billion spent on ‘troubled individuals’ struggling with homelessness, addiction and mental health problems, ahead of the next spending review (HM Treasury 2015a).2 The new Conservative government is now actively considering the case for extending the Troubled Families programme to these individuals as part of their preparations for that spending review.3

The Troubled Families programme has been successful in pioneering integrated support and in producing savings by reducing demand for crisis-led services, for example in Greater Manchester and Oldham.4 A relatively small pot of central funding has generated additional financial commitments from local areas and galvanised agencies to set up local ‘invest to save’ partnerships.

However, there are several aspects of the model that could be improved. Specifically, the programme relies upon an individual-based payment-by-results (PbR) scheme to reward success. Evidence suggests that the effectiveness of PbR schemes is mixed for those with multiple and complex needs, although this could be mediated by using incremental area-based outcomes. In addition, nationally defined criteria could be made more flexible to allow for local areas to define their target groups in order to ensure that no individuals are left behind.

We recommend that, alongside an expanded Troubled Families programme, the government should create a new ‘Troubled Lives’ programme, based upon similar principles. While the Troubled Families programme is aimed at coordinating support for workless families with problems of crime and antisocial behaviour and truancy, Troubled Lives would be targeted at approximately a quarter of a million individuals who experience two or more of the following problems: homelessness, substance misuse and reoffending. Instead of the PbR element of the programme being linked to achieving individual-level outcomes,5 it should be linked to area-level outcomes: upper-tier local authorities should be required to show reduced demand for expensive crisis care services after one year of the programme. This approach would help to improve the lives of some of the most excluded people in society, support the integration of local services, and reform poorly targeted spending.


Major reductions in public spending need to be made in the next parliament if the government is to achieve its goal of eliminating the deficit by 2018–19. In addition to £12 billion of cuts to social security spending, achieving this will require cuts of almost 17 per cent to be made in spending by unprotected departments (Thompson and Stirling 2015).6 In this report, however, we find that as a result of greater numbers of people becoming socially excluded, demand is likely to rise for out-of-work benefits (including incapacity benefits) and expensive crisis services (A&E, emergency services, police call-outs and time spent in custody). For example:

  • rough sleeping has increased by 55 per cent in the UK, and by 79 per cent in London, over the past five years, partly due to the impact of welfare reforms such as benefit caps and rising numbers made homeless from the private rented sector
  • over 13,000 households in England were accepted as homeless by their local council in the first quarter of 2015 – an increase of almost half (49 per cent) on the same point in 2010
  • among working-age adults without children, the proportion living in poverty is now 20 per cent – the highest it has been in at least 30 years, having risen steadily over that period.

As a result of these trends, more people are at risk of finding themselves on the margins of society and facing problems such as addiction, severe debt, offending and poor mental health. This in turn is likely to impose greater costs on the state, with recent research finding that £19,000 per person per year is spent on individuals facing a combination of these problems, at a total estimated annual cost of £4.3 billion (Bramley and Fitzpatrick 2015).

This situation is perpetuated by a failure to reform services that vulnerable and disadvantaged people rely on. Spending still tends to be focused on expensive crisis care services, rather than on coordinated and preventative support. This creates unnecessary additional costs to the taxpayer. One recent study found that better coordinated interventions from statutory and voluntary agencies can reduce the cost of wider service use for people with multiple needs by up to 26 per cent (Battrick et al 2014).

Because services are set up to deal with single issues such as drug or alcohol use, homelessness or mental health, rather than addressing the various needs of the individual, multiple professionals are often working with the same person. It is not unusual for people to receive help from as many as eleven services or more, resulting in gross waste and inefficiency (Anderson 2010). The Troubled Families programme was developed precisely to address this problem. However, there is no framework for disadvantaged adults who do not meet the programme’s criteria.

In this report we examine what lessons can be learned from previous attempts to reform public services for disadvantaged individuals. Based on these lessons we make the case for the government to take forward the March 2015 budget commitment at the next spending review, and invest in a ‘Troubled Lives’ programme to reduce costly duplication in services and reform poorly targeted and ineffective spending.

Lessons from public service reforms

Successive governments have tried and largely failed to shift spending away from standardised and impersonal public services towards those that intervene earlier and provide more tailored support. Reforms have tended to take the form of large-scale programmes led by Whitehall units or departments, and focused on tackling problems in isolation. Under the last Labour government, for example, initiatives such as the Rough Sleepers Unit often achieved impressive short-term results, but these were not always sustained when the national political focus shifted elsewhere.

The driving force of policy to tackle social disadvantage, both under the current Conservative government and the previous Coalition government, is a belief in commissioning at scale and using market-based mechanisms such as payment by results to create efficiencies and provide innovative responses to difficult social problems. But while flagship reforms such as the Work Programme are helping some back into work, they have been ineffective for those facing more complex challenges. Among prison-leavers, for instance, 44,000 people have been referred to the programme, but claims for any form of job entry have been made for just 5,300 (DWP 2015).7

We have examined a number of public service reforms that had the objective of improving the lives of the most excluded under governments both past and current. The examples were chosen because they each represent a different and distinctive approach, and offer useful lessons.

National programmes

National programmes, such as the Rough Sleepers Unit (RSU) introduced in 1999 and Troubled Families introduced in 2011, can provide a clear political focus from central government, often with an unambiguous target to spark action. However, starting with a defined set of problems linked to centralised targets can miss what people need to lead better lives and overlook deeper structural issues, unless there is local flexibility to determine priorities. For example, the RSUs narrow focus on rough sleeping meant that issues such as the quality and availability of hostels, move-on and permanent accommodation were left relatively unresolved. More recent initiatives, including the Troubled Families programme, appear to be having more success in achieving local ownership by making this an explicit aim.

Outsourcing public services – commissioning at scale and payment by results

Payment-by-results (PbR) schemes can focus efforts on specified outcomes rather than processes or inputs, and can help clarify accountabilities because they hold local authorities or providers to account for the delivery of specified outcomes. With the Troubled Families programme, PbR provided incentives to stimulate cooperation between local agencies and helped attract additional local financial commitments. However, the evidence is less clear on the extent to which outcome-based payments can meaningfully be linked to interventions for individuals with multiple and complex needs. PbR schemes are most effective when outcomes are easy to measure and can clearly be attributed to a provider’s intervention. However, effective interventions for this group rely on coordinating support from a range of different agencies, and progress can be affected by factors beyond the control of services, such as wider government policy and economic conditions.


We examined several examples of central funding streams being devolved to local authorities. This can lead to a more engaged and preventative approach locally, as with, for example, the devolution of homelessness budgets following the Homelessness Act of 2002. However, clear local accountability is needed to reduce the risk of finance being diverted away from where it is most needed. For example, the removal of the Supporting People ringfence in 2009 was supposed to allow local authorities greater freedom to commission tailored support informed by local priorities. Instead, fuelled at least in part by the need to offset significant cuts to their overall budgets, many local authorities have replaced specialist support with more generic services such as floating support, and have made cuts to staff salaries and monitoring practices (Homeless Link 2013). Similarly, the removal of the ringfence around the substance misuse budget in 2013 and its absorption into the public health budgets of upper-tier local authorities has left this funding vulnerable to cuts in order to shore up funding for statutory services.8

Social investment

One form of social investment, the social impact bond, has been trialled in various sectors in the UK. Examples include the Peterborough pilot for reducing reoffending, and the Rough Sleeping social impact bond for working with people with entrenched street lifestyles. The previous Coalition government held up the social impact bond model as an option for funding support to tackle intractable problems, as it does not require upfront investment from either frontline agencies or the state. However, we find little evidence to suggest that they will work successfully in future to attract significant private investment into socially beneficial yet commercially risky projects.


Our analysis suggests that what most previous reforms have in common is a failure to give local areas the necessary powers, responsibility and accountability to improve the lives of the most excluded. The Troubled Families programme appears to be an exception, galvanising local areas to bring existing agencies to work together more effectively, rather than adding another layer of intervention onto an already complex system.9 Aspects of the model are flawed and government claims about the results and savings achieved by the programme are dubious (Portes 2015). In particular, evidence of cashable savings resulting from the programme, as opposed to reduced demand for services, is scarce (DCLG 2015). However, the programme is having some success in improving integration where previous strategies (such as City Pathfinders) have struggled.

In addition, a growing number of initiatives around the country are demonstrating that investing in better local coordination and intensive support for individuals with multiple and complex needs can reduce demand for expensive crisis care services.10

Based on these findings, we argue that the core elements of any successful approach to improve the lives of adults with complex and overlapping problems should be:

  • an area-based, decentralised approach, not large-scale national programmes
  • national priorities set by central government, but local responsibility and accountability for design and delivery
  • integrated funding, commissioning and delivery for person-centred support and clear incentives for wider systems to change.

Towards a breakthrough for ‘Troubled Lives’

We propose that at the next spending review multiple and complex needs is chosen as one of a small number of priority issues for investment in local integration and service transformation. A new Troubled Lives programme should be introduced for this, based on the Troubled Families model of centrally driven but locally led reform for vulnerable groups. The focus should be approximately a quarter of a million individuals who experience two or more of the following problems: homelessness, substance misuse and offending.

As under the expanded Troubled Families programme, however, local areas should have flexibility to determine local priorities and set a number of eligibility criteria. A national, collective outcome should be agreed for this investment, bringing together a range of government departments, agencies and upper-tier local authorities to pool funding and deliver joint solutions.

The Troubled Families model should be adapted for a Troubled Lives programme. For example, evidence is mixed on the performance of outcome-based payment schemes for individuals with complex needs. The capacity to work of those with complex needs can be limited for a long time, possibly permanently, and the performance of the Work Programme for this group has been poor.

Instead of the PbR element of the programme being linked to achieving individual-level outcome-based payments, area-level outcomes should be adopted. This retains the incentives for local cooperation provided by the Troubled Families PbR model, but removes the need for artificial or forced results from interventions with individuals.

To summarise, we propose the following six reforms for the next spending review.

Recommendation 1: Multiple and complex needs is chosen as one of a small number of priority issues for investment in local integration and service transformation. A new Troubled Lives programme is established for this, based on the Troubled Families model. The focus of the programme is approximately a quarter of a million individuals who experience two or more of the following problems: homelessness, substance misuse and offending.

Recommendation 2: A central funding pot of up to £100 million a year for four years is established for a Troubled Lives fund11 to support upper-tier local authorities to integrate local services around troubled individuals and provide intensive support from a keyworker. A national, collective outcome is agreed for this investment, bringing together a range of government departments, agencies and upper-tier local authorities to pool funding and deliver joint solutions. As under the expanded Troubled Families programme, however, local areas have flexibility to determine local priorities and set a number of eligibility criteria.

Recommendation 3: Two-thirds of this fund is devolved to local areas to introduce intensive one-to-one support for troubled individuals and to support local service integration and transformation. This funding should be matched by locally pooled budgets.12 The final third of this fund is awarded to local authorities on a pay-for-performance basis on an incremental scale, with payments being made on the basis of area-level, rather than individual-level outcome indicators. Upper-tier local authorities are required to show reduced demand for expensive crisis care services after one year or more of the scheme. The proportion of the fund awarded on a pay-for-performance basis could rise in subsequent years.

Recommendation 4: While central government sets a small number of success indicators, the design of the programme is developed by those who have the strongest insights into what works – upper-tier local authorities in partnership with specialist and voluntary sector organisations and those living with multiple disadvantage. Given the large regional variations in the extent of people with multiple needs, priority areas for this programme should be northern cities, and some seaside towns and central London boroughs.13 Wider links with economic development and regeneration policy should be developed in each area.

Recommendation 5: As part of their proposals for the Troubled Lives fund, upper-tier local authorities should similarly be able to bid for pots of relevant central funding to be devolved, such as employment support, mental health and community safety funding. Given the high levels of spending on prison for those with the most complex needs, probation funding should come under consideration too. However, as the budget for medium- to low-risk offenders will be tied up in large-scale contracts as part of the Transforming Rehabilitation reforms, it may not be possible to go any further than establishing co-commissioning or partnership arrangements in most areas.

Recommendation 6: A lack of integrated data to assess outcomes for those with multiple and complex needs reinforces a silo-based approach to funding and delivery. To support the introduction of the programme, the government should charge the new Administrative Data Taskforces14 with linking data between key government systems dealing with substance misuse, homelessness and offending, together with data from social security and from key voluntary sector organisations. The objective should be for government to introduce integrated measures of multiple and complex needs in this parliament, and to begin to report against these measures to assess progress.

1 As part of this project, IPPR staff carried out one-to-one conversations with representatives from the following organisations: Making Every Adult Matter (MEAM), St Mungo’s (now St Mungo’s Broadway), Revolving Doors, Drugscope, User Voice, Resolving Chaos, LankellyChase, WCEN, the Troubled Families Unit and the Department for Communities and Local Government. The majority of conversations took place in summer and autumn of 2013, and were carried out over the phone as well as face-to-face. Service users were consulted through a discussion group in April 2013 hosted by Revolving Doors Agency and St Mungo’s.

2 The £4.3 billion figure is drawn from Bramley and Fitzpatrick (2015) and represents annual public spending on individuals experiencing two or more of the following: homelessness, substance misuse and offending.

3 Source: authors interviews with stakeholders.

4 This has also been evidenced in reports such as London Councils (2014).

5 Such as achieving significant and sustained progress, or moving off out-of-work benefits and into continuous employment as under the expanded Troubled Families programme.

6 Unprotected departments are the Ministry of Defence, the Home Office, the Ministry of Justice, the Department for Communities and Local Government, the Department for Business, Innovation and Skills, and the Department for Transport. Spending on overseas aid and the NHS is set to continue increasing in real terms, while schools’ spending per pupil is to be protected in cash terms.

7 Since February 2012 anyone claiming jobseeker’s allowance within 13 weeks of leaving prison is automatically referred to the Work Programme.

8 A survey of local authority commissioning intentions found that, of those where a decision had been made, 34 per cent of areas expected to make cuts to substance misuse services in 2015/16, and only 7 per cent expected to fund more services (Public Health England and the Association of Directors of Public Health 2014).

9 It has been described by the head of the programme Louise Casey as an attempt to bring about ‘systems change’ at the local level (Casey 2012).

10 This includes projects run by the Make Every Adult Matter (MEAM) coalition of charities, the Resolving Chaos charity and the Big Lottery Fund’s Fulfilling Lives: Supporting people with multiple needs projects.

11 We propose that this is funded by top slicing budgets in at least five departments – Department for Communities and Local Government (DCLG), Department for Work and Pensions, Department of Health, Ministry of Justice and the Cabinet Office.

12 The funding ratio for the Troubled Families programme is 2:3 – for example, for the £400 million received in total by local authorities DCLG expects them and their partners to contribute an additional £600 million worth of services, including resources ‘in kind’. We would propose a similar ratio for this fund. Pooled funds could include funding for substance misuse treatment, which currently forms part of local public health budgets, and homelessness funding. The annual drug and alcohol treatment budget is worth between £800 million and £1 billion, and councils also receive a share of the £80 million homelessness prevention grant. Other contributions could come from prison substance misuse funding, the police Community Safety budget, community mental health budgets, health services for prisons, the Supporting People grant and clinical commissioning groups, Work Programme providers, Jobcentre Plus, probation trusts and local police forces.

13 See Bramley and Fitzpatrick 2015

14 This was formed in 2011 by the Economic and Social Research Council, the Medical Research Council and the Wellcome Trust.