The scale of youth homelessness in the UK is significant. Research has found that around 78,000–80,000 people under the age of 25 experienced homelessness in 2008/09. In addition, however, because many homeless young people are not counted, the total number is likely to be considerably higher; these people are known as the ‘hidden homeless’.
Homelessness no doubt has significant negative effects on the quality of life of the homeless individuals themselves. In addition, it leads to substantial costs at a societal level. These include direct expenditure in providing support to the individual, as well as the costs of the adverse outcomes of homelessness such as higher crime rates, mental health problems and drug abuse, and lower employment rates and wages. In the UK, around £1bn of annual government spending is targeted at either the causes or the effects of homelessness.
Homelessness is therefore generally seen as worthy of intervention by the state and charitable organisations. In the UK a number of charities are involved in helping homeless people, including Centrepoint Soho (Centrepoint), St Mungo’s and Thames Reach, for example. Of these, some, such as Centrepoint, focus on tackling the issue early on by targeting efforts to help young homeless people between the ages of 16 and 25. Centrepoint’s services involve providing housing services, skills development and health services to these young people before life on the streets has taken its toll (see the box).
Given the significant societal costs of homelessness, early intervention might be expected to be critical to avoid escalation of the condition and thereby its negative effects on society. To test this expectation, Oxera undertook a cost–benefit analysis of the early intervention by Centrepoint relative to interventions later on, from the point of view of the impact on public spending. The analysis focused on determining whether each £1 that Centrepoint spends on interventions at an early stage of homelessness is higher or lower than likely future public expenditure, should the intervention not take place at that time. The assumption underlying the analysis was that if Centrepoint does not intervene early, another organisation would undertake the same intervention five years later when the homeless person is older and their problems may have escalated. The five-year period was based on the average duration of homelessness among young people—i.e. that there is ‘delayed intervention’. The benefit of Centrepoint’s work was therefore measured by the costs that society avoids and/or the benefits it receives through the avoidance of escalated problems that are more expensive to address after five years.
Benefits to society of intervening early
Social benefits may arise in various ways:
improved education and lower barriers to securing employment and wages, and thereby higher tax revenue to the state and a lower drain on welfare benefits;
less involvement in crime, leading to lower costs to the criminal justice system;
fewer health and substance abuse problems, and improved mental health conditions, leading to a lower burden on the public healthcare system.
Improved education and employment. A large proportion of young homeless people are NEET (not in education, employment or training). The provision of education and employment skills to these young people can have two separate impacts:
by increasing employment and wages, it reduces the level of welfare benefits drawn;
the higher wages in turn lead to higher tax collection.
The magnitudes of these impacts are determined by the employment rate after early intervention and the likely employment rate of young people after delayed intervention.
Oxera’s analysis (based on data from Centrepoint and from public sources) showed that, with Centrepoint’s intervention, employment levels stabilised at around 46% over the long term, while the impact of delayed intervention after five years was lower (28% employment rate) because the problems were more entrenched and difficult to address. This higher employment rate due to early intervention resulted, in turn, in a greater reduction in welfare benefits and higher taxes raised over the lifetime of the young person relative to intervention five years later.
Decreased crime rates. Owing to their financial situation, homeless people may, and often do, turn to crime to fund survival. This can involve robbery, shoplifting, stealing to order for criminal gangs, and prostitution. Such crime imposes costs on the criminal justice system. These include costs associated with police work in investigating crime incidents, court proceedings (which vary depending on the type of crime), prison costs, the enforcement of community sentences, and/or the provision of probation services.
The impact of early intervention on reducing the costs incurred by the criminal justice system has been assessed by combining two sources of information:
data on crime rates among Centrepoint’s clients and among older homeless people;
estimates of monetary costs to the criminal justice system, available from the National Audit Office.
This information was used to determine the expected costs per homeless young offender with and without early intervention by Centrepoint.
Decreased drug and alcohol abuse. Homeless young people also impose costs on society through substance abuse. Without early intervention, the incidence rate is likely to increase, as is the intensity of an addiction. This, in turn, means greater need for drug intervention programmes and counselling services, and, at the extreme stages, significant costs through residential treatments. By providing direct support (through accommodation and counselling) to tackle substance misuse by its clients, and by referring them to NHS substance abuse treatment early on, Centrepoint helps to avoid the higher treatment costs that would have been incurred in the absence of early intervention.
Improved mental health conditions. Early intervention can also lead to improved mental health conditions for young homeless people, thereby saving public health costs by reducing visits to hospital accident and emergency (A&E) departments, in the use of crisis mental health teams, and owing to the decreased need to be hospitalised. To estimate Centrepoint’s impact in reducing mental health problems, and thereby the associated costs, Oxera used information from Centrepoint on mental health problems among its clients before and after intervention, as well as other sources of data on such problems suffered by older homeless people before and after being resettled, and information on the costs of different treatments.
Aggregate impact of early intervention
The net benefit of the early intervention is estimated by netting off the costs incurred by Centrepoint in providing these services from the benefits (or public sector costs avoided) estimated above. In addition, it is relevant to account for the cost of intervention at a later stage by other charitable organisations, as Centrepoint saves these costs through its early intervention.
The analysis suggests that £1 spent by Centrepoint in intervening during the early stages of homelessness, compared with similar intervention at a later stage, results in potential costs avoided by the public purse of £2.40. This equates to a net benefit of at least £19,900 per young homeless person. Table 1 gives a breakdown of the values of the different types of benefit. Sensitivity analysis around some of the assumptions in the main scenario indicates that the benefits to the public purse could range from £2.21 to £2.48.
Table 1 Net benefit of Centrepoint intervention to the public purse, averaged per Centrepoint client (£, real 2010/11 prices)
Note: Numbers may not add up due to rounding. 1 A discount rate of 3.5% has been used; see HM Treasury (2011), ‘The Green Book. Appraisal and Evaluation in Central Government’.
Source: Oxera analysis.
Key challenges and lessons
This cost–benefit analysis is typical of the type of work done by economists, but applying it to Centrepoint’s activities brought about new challenges. Although Centrepoint provided good information on its own clients before and shortly after its intervention, a key challenge was to create the counterfactual path that the young people would have taken without such intervention. While there was indeed a significant amount of data on the relevant elements of the calculation, in most cases the data had been collected for a different purpose, and often did not concentrate on the same categories of homeless people who come to Centrepoint. Significant work was therefore required to pull together this large body of independent, and not ideal, sources of information to create a well-founded assessment of the counterfactual.
It was also challenging to capture the other, less tangible benefits from Centrepoint’s intervention, such as those from the improvement in the quality of life of the young people and those from reduced costs to victims of crime. Evidence indicates that these benefits are likely to be substantial. The results set out in this article are therefore conservative and the actual social benefit delivered by Centrepoint is likely to exceed the above estimates.
Overall, the analysis highlights that sooner is indeed better, and indicates the importance of intervening in the lives of homeless people at an early stage, and thereby preventing escalation of conditions that may prove costly for both society and the individuals if left to be tackled at a later stage.
Contact: Fod Barnes
 For general information on homelessness in the UK, see the website and publications of Crisis, an organisation that focuses on homelessness issues. See, for example, Crisis (2011), ‘The hidden truth about homelessness: Experiences of single homelessness in England’.
 National Audit Office (2005), ‘More than a roof: Progress in tackling homelessness’, 23 February.
 Names have been changed throughout this article to protect identities.
 Some homeless people may not receive the necessary help at a later date, and some reverse their homelessness situation without intervention from homelessness services organisations. The analysis includes sensitivity analysis to test for these possibilities.
 In addition, there are the costs incurred through the crime itself (e.g. damage to the victim and property). Oxera did not estimate these owing to a lack of sufficiently robust data.
 Crane, M., Warnes, T. and Coward, S. (2011), ‘The FOR-HOME Study: Moves to Independent Living: Single Homeless People’s Experiences and Outcomes of Resettlement’, Shefﬁeld Institute for Studies on Ageing, University of Shefﬁeld. The treatment costs for different mental health conditions were obtained from a study on the costs of mental healthcare in England: King’s Fund (2008), ‘Paying the price: the cost of mental health care in England to 2026’.
 Centrepoint’s costs of intervention are available from its annual accounts. The implementation costs of delayed intervention could be proxied by those costs incurred by a provider of all-age homelessness services that would work with older and more entrenched people. However, with no relevant cost data being available, Oxera assumed that the costs of services to homeless people of all ages are similar to those of Centrepoint, although the costs of early intervention may be lower than those of delayed intervention by a comparable service provider owing to the difference in the entrenchment of clients.