Preventing Youth Homelessness through Social Procurement in Construction: A Capability Empowerment Approach

: Homelessness is a serious and growing problem in the UK, exacerbated by the COVID-19 crisis. The latest ﬁgures estimate that 160,000 households are at risk of the worst forms of homelessness. Employment is widely recognised as being critical to reducing homelessness, yet there has been no research into the role that the construction industry, as a major UK employer, can play in reducing this problem. The aim of this paper is to address this gap in knowledge and contribute to the emerging social procurement debate in construction by exploring the role that construction employment can play in reducing the risk of homelessness. Mobilising Sen’s and Nussbaum’s capabilities empowerment approach, an in-depth case study is presented of a construction employment program in Wales, UK, which was aimed at supporting young people who had experienced or who were at risk of homelessness. Contributing to the emerging social value and social procurement debate in construction and drawing on documentary analysis and interviews with young people who were homeless or at risk of homelessness who went through the program, ﬁndings indicate that these young people became empowered in ways which reduced their risk of homelessness. It is concluded that the capabilities empowerment framework is valuable in explaining how employment in the construction industry can reduce the risks of homelessness for disadvantaged youth with a care-experienced background or who were known to the criminal justice system.


Introduction
The European Federation of National Organizations Working with the Homeless [1] (p. 1) defines youth homelessness as "when an individual between the ages of 13 and 26 is experiencing rooflessness or houselessness or is living in insecure or inadequate housing without a parent, family member or other legal guardian". Youth homelessness is a significant and growing social problem in the UK, although reliable national statistics are problematic because homelessness is defined differently in England, Wales and Scotland and many people, such as those who have couch surfed or slept in a car (often termed the 'hidden homeless') do not register in formal statistics. Nevertheless, it is estimated that 57,890 households were assessed as homeless in England in 2019, 34,100 in Scotland and 9210 in Wales [2]. In Wales alone, the latest figures show that 7698 young people (aged [16][17][18][19][20][21][22][23][24][25] asked local authorities for help with homelessness problems in 2018 [3]. This high number of young people seeking help with homelessness is not exclusively a UK phenomenon. The European Observatory on Homelessness reports that increasing youth homelessness is the most striking and worrying trend in homelessness demographics [4]. In Australia, young people make up some 42% of the Australian homeless population and it is estimated that approximately 44,000 young Australians under the age of 25 are homeless (although the actual number is likely to be higher) [5].
Care-experienced young people (young people who have spent a significant part of their childhood under state care) and those known to the criminal justice system (CJS) are especially vulnerable to homelessness, in part because of challenges in finding employment [6]. Young people who have been in state care disproportionally experience homelessness as well as other adverse outcomes, including poor mental health, poor educational achievement and unemployment [7]. A pan-Canadian study found that 85.4% of Canadian homeless young people had declining mental health, 42% had attempted suicide at least once and 35.2% had suffered a drug overdose that required hospitalization [8].
In Wales, Hodgson [9] found that 88% of the 77 homeless young people she interviewed met the criteria for a mental health disorder, compared to 32.3% for the general population.
It is widely recognised that employment can play a role in reducing the risks of homelessness, although this is not automatic and depends on the quality and nature of the work involved [6,10]. As Qian et al. [10] notes, for young people, employment marks a key milestone in their transition into adulthood, helping to build self-confidence, selfesteem and financial independence. However, Qian et al. [10] also recognise that young people suffer higher levels of employment uncertainty and underemployment than the general population and that this can result in feelings of hopelessness and stress and higher incidences of depression and other chronic diseases, which can further reduce housing security. The recent COVID-19 crisis has affected young people particularly badly, especially those from a homeless background. Homelessness, substance use and associated psychiatric disorders put young people at higher risk of COVID-19, which is coupled with the fact that they will be less likely to be able to access COVID-19 treatment and testing [11]. In the UK, the Resolution Foundation [12] warns that the economic fallout from the COVID-19 pandemic has taken the UK into uncharted territory, with fears that an additional 640,000 18-24-year-olds could find themselves unemployed this year alone. In other countries such as Australia, Atkins et al. [13] note that prior to the COVID-19 pandemic, the youth unemployment rate was already 11.5% (more than double that of the general population of 5.1%) but in April 2020 it had risen to 13.8%. Underemployment is a much bigger problem in Australia, with one-in-three young people being either unemployed or underemployed.
Driven by the failure of traditional government policies to address intransigent social issues like homelessness, governments in many countries are turning to social procurement to leverage their construction spending to create employment opportunities for people suffering various forms of disadvantage like homelessness [14][15][16][17][18]. Young people are targets of such policies because, as Raiden et al. [19] note, the construction sector is one of the largest employers of young people (in Australia it is the third largest employer of young people, ABS 2020) and it can provide high quality and well-paid jobs to young people who are otherwise disadvantaged.
Surprisingly, despite an emerging social procurement debate in construction [17][18][19][20][21][22] and despite construction being a major employer of young people, there has been little youth-related research in the field of construction management [23][24][25]. In particular, there has been no empirical research into the contribution which social procurement in the construction industry can make to reducing the growing youth homelessness problem. The only studies into the challenges that young homeless people face accessing construction employment [26][27][28][29][30] show that young homeless people need intensive, comprehensive, client-centred support that draws upon multi-sector collaboration to combine employment training with several other fundamental components, including housing, education and personal support. However, while this research is valuable, the relationship between youth homelessness and social procurement in construction remains under-conceptualised, meaning that the potential contribution that these new policy imperatives can make to reducing the risk of homelessness is poorly understood.
The aim of this paper is to address this important gap in social procurement research through a single in-depth case study of the Symud Ymlean Moving Forward (SYMF) program in Wales, UK. More specifically, the paper addresses the question of how the capabilities of young people at risk of homelessness changes as a result of working in the construction industry and how this affects their risk of ongoing homelessness. The contribution of this research is twofold. First, it addresses a gap in knowledge around how social procurement in construction can reduce the risk of homelessness. Second, in contributing to the general under-theorisation of social procurement within and outside construction [31], the value of the capability approach is explored as a conceptual lens to understand the social value which construction can contribute to the communities in which it builds.
This article proceeds first with an examination of Sen's [1] and Nussbaum's [2,31] capability empowerment approach, highlighting its value in relation to understanding how the construction industry can potentially address youth homelessness. After describing the method adopted to collect and analyse the data presented, we then explore, via an empirical case study of a unique intermediate construction employment program, the potential value of the capabilities empowerment framework in addressing the above research question. Finally, the implications and contributions of this research to policymakers and the field of construction management are discussed and the paper closes with some critical reflections on the contribution that the construction industry can make to the empowerment of young people who have experienced homelessness or who are at risk of homelessness.

Theory
Traditional approaches examining the causes of homelessness have been widely criticised because they can imply that homelessness is the result of personal failure rather than structural disadvantage [32]. Rather, the evidence suggests there are factors which make some people more vulnerable to homelessness than others and that the causes of homelessness consist of a complicated set of interconnected factors, often working in a cumulative way over a person's lifetime [33]. For example, Gaetz and Dej [34] use a socioecological model to divide the contributing factors of homelessness into three main areas: individual and relational factors; structural failures; and system failures. Individual and relational factors involve family breakdown or conflict, which can lead to a young person experiencing homelessness. Bramley and Fitzpatrick [35] show that people who experience childhood poverty have a significantly higher probability of homelessness in later life. This is often linked to other problems which can exacerbate the risk of homelessness further such as harmful substance use, mental health difficulties, disengagement from social systems and youths who are not in education, training, or employment (NEET). Schwan et al. [33] also suggests that although individual and relational factors might look like the most direct cause of homelessness, for many people research consistently demonstrates that these factors are often underpinned by structural and systems problems such as poverty, inadequate education, underemployment, discrimination and a lack of affordable housing. This is supported by a recent study of youth homelessness among young people who had been in state care in Wales, which showed that system failures were consistently a contributory factor for youth homelessness [36]. Schwan et al. [33] and Dobson [37] explain that although safe and affordable housing is key to preventing the homelessness or the repeat homelessness of young people, additional factors, including integrated services which provide employment, income, and other forms of support such as mental health, can contribute to housing stability and a meaningful life.
In conceptualising the risk of homelessness, Sen's [38] and Nussbaum's [39] capabilities empowerment approach and Nussbaum's [39,40] capabilities empowerment framework have proved valuable, although the application of this methodology in construction is yet to be explored. For example, Tanekenov [41] and Tanekenov et al. [16] used the capabilities empowerment framework to examine what capabilities empower homeless people to secure and sustain employment and therefore reduce their risk of homelessness. More recently, Foye [42], Watts and Fitzpatrick [43] and Kimhur [44] employed the capabilities approach to explore the potential for housing-related capabilities to advance housing and homeless policy agendas.
Like every theory, the capability approach has its strengths and weaknesses. For example, it has sometimes been criticised as an overly individualistic approach and as an example of methodological individualism that pays little attention to group capabilities or collective capabilities [41]. There are also alternative theories employed in the youth homelessness field. For example, Barman-Adhikari and Rice [45] use social network, social capital and social influence theories as a framework to understand the barriers or facilitators of the use of employment services by homeless young people. The approach of Barman-Adhikari and Rice [45] provides guidance on the network components of service use and can be used to identify features that can increase the use of employment services by homeless young people. However, recent research such as that of Dobson [37] refers to the complex needs of some of the adult homeless population, with their homelessness being linked with other health and social care needs, including mental and physical health, substance misuse and social problems such as offending. The capabilities approach was chosen as the most appropriate way of understanding these multiple barriers young people at risk of homelessness need to overcome because it acknowledges that different people have different opportunities. As Kimhur [41] explains, because the capability approach moves away from a total amount of welfare to the different ability of humans to convert resources into welfare, it is an approach that can be used to evaluate the welfare of marginalised groups like those at risk of homelessness who are less able to convert resources into achievements (functioning), taking into account human diversity.
The capability empowerment approach is a rights-based evaluative approach, which originated in the field of poverty and development studies [44,46]. Questioning the notion of 'equality of opportunities' which lay at the heart of much social policy practice and literature, the basis of the approach is Sen's [38] criticisms of traditional utilitarian approaches for assessing welfare policy success based on opulence (real income, wealth commodities) and utility (satisfaction or desire-fulfilment). Sen [38] offered an alternative approach based on peoples' empowerment and freedoms (what opportunities are afforded to them), capabilities (what people can feasibly achieve through the resources available to them) and functions (what people value-such as stable housing). In other words, welfare policies should focus on developing people's freedom and capabilities to achieve stable and secure housing, although Kimhur [44] also recognises that stable housing also allows people to enhance other capabilities such as employment, family security, education and health. The value of Sen's [38] capability empowerment approach is that it recognises that different people have different opportunities, resources and freedoms to develop the capabilities needed to achieve their end functions. For example, due to the lack of support structures in place, young people from a disadvantaged background are likely to have had less freedom and opportunity to develop the capabilities to avoid homelessness than someone from a privileged background. They are therefore likely to need more help to do so and policy success in the area should be based on what homeless people succeed in doing relative to the opportunities and resources at their command, rather than what they achieve.
Building on Sen's [38] (1985) work, Nussbaum [39,40] drew on the empowerment literature to propose a list of ten central capabilities (capability components) which she argues are required to have a fulfilling life (see Table 1).
According to Nussbaum [40], each of the capabilities in Table 1 are equally important and a shortfall in one cannot be made up for with a surplus in another. Although Tanekenov et al. [16] later argued that capabilities are context-specific and reject any definitive list of core capabilities, their work showed that preventing homelessness is not just about providing shelter to homeless people. Rather, it requires an integrated approach to build personal capabilities in areas that when deficient can lead to a young person experiencing homelessness. For example, Hodgson [9] indicates that factors in the capability empowerment framework such as a young person's mental health can make it more difficult to find appropriate housing and sustain a tenancy. Gaetz and Dej [34] also highlight mental health conditions, harmful substance use and family and relationship breakdowns as important factors which are further exacerbated by homelessness and experience of the care system or criminal justice system.
Utilising the capabilities empowerment framework in Table 1 as our conceptual framework, the following section describes a methodology to explore the extent to which employment in the construction industry can build capabilities in homeless young people which can lower the risk of continued homelessness in the future. Table 1. The Capabilities empowerment framework. Source: adapted from Nussbaum [40] (pp. 103-105).

Empowerment Domain Capabilities List
Life Being able to live to the end of a human life of normal length; not dying prematurely, or before one's life is so reduced as to be not worth living.
Bodily Health Being able to have good health, including reproductive health; to be adequately nourished; to have adequate shelter.
Bodily Integrity Being able to move freely in safety and security and having one's bodily boundaries treated as sovereign.
Senses, Imagination and Thought.
Being able to use the senses, to imagine, think and reason, and to do these things in a "truly human" way, a way informed and cultivated by an adequate education, literacy and basic mathematical and scientific training.

Emotions
Being able to have attachments to things and people outside ourselves and not having one's emotional development blighted by overwhelming fear and anxiety, or by traumatic events of abuse or neglect.

Practical Reason
Being able to engage in critical reflection, personal development and long-term planning in one's life.

Affiliation
Development of positive self-esteem, self-efficacy and dignity through relationships with family, friends, peers and social networks and concern for other people Other Species Being able to live with concern for and in relation to animals, plants and the world of nature Play Being able to laugh, to play, to enjoy recreational activities Control over One's Environment: Political and Material Having the right to seek employment on an equal basis with others; having the freedom to control one's own life, being able to participate effectively in choices that govern one's life

Method
Ontologically, given the socially constructed nature of the capabilities empowerment framework, our methodology employed a social constructivist ontology and an interpretivist epistemology. This involved using qualitative methods of data collection and analysis in close interaction with young people experiencing or at risk of homelessness, engaging with them to understand how they experience, describe and interpret the capability-based impacts of working in construction [47]. Data was collected within a single exploratory case study of an intermediary program which had been set up to provide disadvantaged young people who were homeless, had experienced homelessness or were at risk of homelessness with access to work in the construction industry. Following Flyvbjerg [48] and Yin [49], we adopted a single case study approach because as far as we are aware, the case study described below is the only example internationally of such an initiative. Work elsewhere to help young homeless people into employment has focused on other industries, but this case study is unique to the construction industry [50].

Case Study Description
The case study is a program which was designed to provide employment opportunities in the construction industry for disadvantaged young people from a care or ex-offending background who have suffered or are at risk of suffering homelessness. Called the Symud Ymlean Moving Forward (SYMF) program, it ran from 2013 until 2016, at a cost of £4.8 million through funding provided from dormant bank accounts in the Big Lottery Fund. To be eligible for the SYMF program, young people could not be in education, employment or training (NEET), needed to be either currently in or had been in state care, or working with a youth offending team (YOT) and homeless or at risk of homelessness. Many young people who were referred to the SYMF program were known to the criminal justice system (52.5%) and one third were care leavers with over one in ten (13.8%) being both. Of the 1116 young people referred to the SYMF program, 212 young people (19%) were homeless or living in temporary accommodation at the time of referral. Work placements were provided in a number of industries, including construction, retail, administration, mechanics, catering and many more. In total, 437 young people completed a 26-week work placement as part of the SYMF program. Fifty-two young people undertook a construction placement. We interviewed 10 young people who had undertaken a construction work placement (19.2% of the sample frame). All were homeless or living in temporary accommodation.
The initiative provided participants with employability support and the potential to progress into a paid 26-week work placement in the UK construction industry. In addition to employment opportunities, the program provided support to participants to help them overcome barriers to work, gain essential skills and confidence, alongside the facilitation of practical experience and paid work placements. This was done through individualised and tailored support and training alongside on-going work-readiness support and mentoring, culminating in a twenty-six-week paid work placement. The disadvantages these young people faced were numerous and included poor academic achievement, absence of work history, low levels of confidence, as well as structural factors such as employer stereotypes and attitudes, which compounded poor labor market opportunities. Numerous individualised support strategies were developed to tackle these barriers as part of the SYMF program. For example, as part of the program, young people completed interview practice with sensitive SYMF staff as part of their pre-employability support. Mock interviews with a supportive employer would also be arranged to enable them to practice interview skills in a more realistic setting. Young people could also be supported to attend interviews with an allowance for suitable clothing and they could be taken to the interview by a member of SYMF team. A lack of access to affordable transport options was an ongoing barrier to employment for many participants of the SYMF program once they were placed with an employer, since most lived in remote towns or rural areas. Therefore, transportation was also arranged for candidates as part of their work placements.
Mobilising Nussbaum's [40] capabilities empowerment framework, data were collected using a mix of methods (semi-structured interviews with a sample of young people who went through the SYMF program) and analysis of documents, which traced their path through the program. This is discussed in more detail below.

Semi-Structured Interviews
Using quota sampling, semi-structured interviews were conducted with ten young people who had gone through the program and undertaken construction work placements and who were willing to give their consent to be interviewed (19% of the total population who had undertaken construction work placements). See Table 2 for sample structure.
Questions were designed around the capabilities empowerment framework as described in Table 1.
• Life (Has your lifestyle changed since you have taken part in the project? How would you say your lifestyle has changed? When your construction employer said they were going to take you on how did that make you feel? Have you received support from the project to look for work in the future?) • Bodily Health (Do you think your physical health has improved because you have been working in construction? Did your construction employer have any policies about drugs or alcohol?) • Bodily integrity (Have you ever had to stay somewhere you did not feel safe?) • Senses, imagination and thought (Did you undertake any pre-employability training before you went on to the work placement? Did you get any construction training to help you get a placement or in your placement? Did you get any qualifications or accreditations? How did you feel gaining them?) • Emotions (Do you think you are better at managing your emotions since taking part in the project? Have you changed your behaviour from being with your friends to being in the work environment on a construction site? Did you do any accreditations or qualifications around controlling feelings? • Practical Reason (How did you find setting goals and targets in your personal development plan (PDP) and PDP reviews? Were there any specific goals and targets for you to achieve a construction placement? How did you feel when you achieved your goals?) • Affiliation (Did any of your relationships with friends or family change because you were taking part in the project. Do you think you have more confidence in your abilities knowing that you can do a job in construction?) • Other species (Did you learn about the environment and/or sustainability on the program?) • Play (Are you taking part in more recreational activities, e.g., sport, since participating in the SYMF program?) • Control over one's environment (You are in employment now with the construction company that you did the work placement with, do you think you would have been able to get that job if you hadn't done the work placement there? Do you think there anything that is going to act as a barrier to you getting further employment?). The interviews were kept deliberately open-ended to enable the researchers to enable respondents to express their views in their own terms and to explore their own paths in discussing each of the capabilities in Table 1 [51]. In this way, our approach respected the vulnerability of the respondents, allowing for the co-production of accounts through a dialogic exchange between interviewer and each respondent, providing the researchers with powerful narratives about the impact of the program on the respondent's lives. As Keene et al. [52] shows, semi-structured interviews that enable respondents the flexibility to move outside narrow question frames and tell their stories are a powerful way for researchers and respondents to collectively make sense of complex multidimensional and interconnected concepts such as those in Table 1, which can be difficult to describe through any other means. Interviews typically lasted 90 min and were undertaken in different locations nominated by the respondents to reflect the different experiences and challenges faced by young people in urban, rural, town and city areas. These were undertaken by a team of experienced researchers and all interviews were recorded and then transcribed.

Documentary Analysis
In addition to the ten interviews, various documents relating to each respondent's background and periodic reviews of their progress through the program were analysed. These documents included: referral and risk assessment forms (these provide baseline information on the respondent's background history and capabilities across the capability empowerment framework); assessment packs (this included more details on eligibility and a skills check); soft outcomes monitoring (this consisted of supporting comments for young people to report on their progress throughout the program); personal development plans (PDPs) (documenting young people's aspirations, preferred learning styles and choice of occupation); PDP reviews (this was a review of progress towards goals and accreditations); participant pathway reviews (this recorded a change in pathway for participants); employment monitoring records (this included the details of the work placements including a risk assessment); and exit paperwork (this recorded why a participant left the project and whether they progressed to employment, education or training).

Data Analysis
The value of the qualitative approach elicited through the semi-structured interviews and documentary analysis is that they produced rich and insightful stories into the respondents' journeys through the program and how their capabilities changed over time [53]. Informed by Nussbaum's [40] capability empowerment framework presented in Table 1 as our analytical lens, we undertook a thematic analysis by searching for patterns in the data collected from both the interviews and documents. Braun and Clarke's [54] and Saldana's [55] approach was adopted, which involved several stages, starting with 'immersion' in the data (repeatedly reading the interview transcripts and documents to obtain a high level of familiarity with the data); categorisation/directed coding (organising and generating an initial list of items/codes from the data-set against each component of Nussbaum's [40] capability empowerment framework, presented in Table 1); searching for themes (examining how codes combine to form over-reaching themes, which are phrases or sentences that identify what the data mean in relation to the analytical framework); and reviewing, defining and naming themes (through repeated iteration between data collection and analysis until theoretical saturation occurred). This process was applied to all the data and Table 3 illustrates the coding process employed to develop the themes, which are discussed in the following Results section, using an example of two quotes from the interview transcripts.
To minimise bias and enhance theme validity, and in accordance with good qualitative research practice, coding and subsequent thematic analysis was undertaken by a team of researchers with a variety of backgrounds (construction, social policy/housing policy) to provide different perspectives on data [56]. Results were verified and checked by key stakeholders of the SYMF program, including young people who had participated in the program. When the lived experience of research participants is so different from the researchers, it is important to ensure they are meaningfully included in the work [57]. Instances of disagreement in coding were resolved through discussion, in which the researchers would table, cross-reference and refine their own interpretations of the data, a process which continued until inter-rater agreement of 100% was achieved. This process provided confidence that the themes which emerged in our results had a high level of 'fit' with the data collected and the theoretical constructs in Table 1 that informed the research. In line with the traditions of thematic research, we present our thematic analysis below in narrative form, supported by selected quotes from the interviews and documents analysed.

Life
When asked about the impact of the program on their lives, respondents described feeling more positive and optimistic about their future. This was largely attributed to increased income from employment in construction, which in-turn provided increased housing stability, greater personal independence and self-confidence and reduced welfare dependency.
"(My life has improved) quite a bit really, I've got more money to manage I manage it better now um I got my own flat, so I moved out (of the supported housing project), I've got a lot more confidence as well." (Participant 5, Completer Interview) Seven of the 10 participants in the sample also felt that working in construction brought about lifestyle changes, new opportunities for career progression into sustainable and meaningful employment and more positive social networks, which helped the respondents overcome common employment barriers associated with homelessness such as interactions with the justice system and previous failures in formal education.
"Yeah, cause when you're working and then you see like people being pricks on the streets it makes you think I used to be one of them. (I have changed my behaviour). I am more sensible, I won't offend anymore, I am more aware of situations." (Participant 1, Completer Interview) On joining the SYMF program, respondents described a number of barriers to gaining construction employment. These included generational unemployment, interactions with the criminal justice system and living in rural areas with poor access to transport. However, at the end of the program, seven of the ten participants secured themselves well-paid construction employment (including three apprenticeships) with the opportunity to progress into a career.
"(My) time on placement went so well that after two months the construction company wanted me to work more . . . They offered me a contract after my placement ended." (Participant 8, Completer Interview)

Bodily Health
When asked about the impact of the program on their health, seven participants reported that their physical fitness had improved. This was linked by the respondents to the physical nature of construction work and the discipline of employment. Participants also reported improvements in emotional wellbeing, sleep patterns and feeling healthier.
"At the start of it I was quite lazy, . . . . now you've got stuff to do now, get up, go to the gym and I go for runs, it's just made me a better person especially my health." (Participant 10, Completer Interview) Analysis of referral and risk assessment forms demonstrated some participants had struggled with mental health difficulties before joining the program. Harmful substance misuse was a common issue among young people referred to the SYMF program. Fortynine percent (n = 537) of the participants referred to the program had a risk due to alcohol or substance misuse identified in their risk assessment. Almost a quarter, 22.1%, of participants referred to the project had a mental health difficulty, with depression being the most common mental health issue among participants. In their interviews, participants noted improvements in mental health, reduced anxiety and substance use and greater resilience in dealing with unexpected challenges. One participant felt his mental health had improved because he knew what was expected of him on a construction site and that his self-confidence and resilience had improved by being able to perform prescribed tasks.
"My employer) was against drugs and alcohol all that he said he didn't mind a pint here and there but drugs and all that and getting too drunk was no good." (Participant 7, Completer Interview) "My anxiety (disorder) has improved . . . because I'm in a good part of my life now . . . " (Participant 9, Completer Interview)

Bodily Integrity
None of the participants reported improved feelings of personal safety and security as a result of participating in the program, which was somewhat surprising and is discussed in more detail later.

Senses, Imagination and Thought
Nine of the participants indicated that they valued the opportunity to improve their literacy and numeracy by learning on a construction project rather than in a classroom environment. Referral and risk assessments revealed participants had faced many barriers which made it difficult to get a basic education, including being permanently excluded from school and learning difficulties such as dyslexia or dyspraxia. Assessment forms also showed that many of the participants had struggled to learn academic knowledge through traditional pedagogical means and that the program had enabled them to develop practical skills and competencies in an applied work-based environment, which better suited their learning styles.
"(I enjoyed) learning new skills, getting better at math and English, getting some qualifications where I didn't have any because I dropped out of school," (Participant 9, Completer Interview)

Emotions
Results relating to the emotions domain indicated that three respondents had improved how they managed their behaviours and emotions while participating in the SYMF program. This was linked to the nature of construction work, which required working calmly with other people in high-risk situations. There was also evidence that the program helped nine participants address their structural disadvantage in the labor market (due to a previous lack of networks in the construction industry) and also facilitated better relationship-building skills. This was often linked in the interviews to the team-based nature and comradery of construction work.
"The project put me, with a lot more people, which meant that, to get along with them you got to start talking to them." (Participant 5, Completer Interview) "I have more confidence in other people as well. It turned out the boys they were alright." (Participant 7, Completer Interview)

Practical Reason
An analysis of the personal development plans and completer interviews indicated that all 10 participants had engaged in forward planning, critical reflection and personal development of skills for working in construction. Participants linked this to the hope of ongoing employment in construction, the opportunities for apprenticeships and the individualised support structures incorporated into the program.

Affiliation
Four participants' personal development plans and completer interviews showed stronger friendships, family relationships and compassion for other people as a result of participating in the SYMF program, highlighting the spill-over impacts of construction employment on family members and friends, as well as the participants themselves. Nine of the participants also reported a growth in confidence and improved selfperception. This was often linked to the sense of responsibility afforded by working in construction.
"The work placement made me feel more confident because I had more responsibilities to deal with. I feel a lot better about myself since getting the job." (Participant 5, Completer Interview) "I (have) better confidence now than what I did before I started. My confidence has been the biggest thing, it has got my confidence up a lot." (Participant 9, Completer Interview)

Other Species
Only one young person demonstrated increased connections to nature as a beneficial outcome of the program. This was attributed to the benefits of working outdoors during his construction placement.
"I enjoyed being outdoors and seeing more parts of Wales." (Participant 2, Completer Interview)

Play
Just two participants reported taking part in more recreation activities due to the case study program, which in turn led to improved social networks and reduced substance abuse and re-offending behaviour. This was linked by the respondents to the need to improve their physical fitness to work on a construction site.

Control over One's Environment
The barriers to employment that young people on the SYMF program faced indicated that they could not seek employment on an equal basis to others. An analysis of risk and referral forms and personal development plans shows that six participants faced additional barriers, including being known to the criminal justice system, lack of transport, living in rural areas, having no up-to-date qualifications, having been excluded from school, having literacy and numeracy problems and having special educational needs such as dyslexia or dyspraxia. Participants were paid minimum wage during their construction placement and sixty percent reported feeling more independent and being better at managing money than before the program.
"It made me feel responsible for my own actions and like because I was getting my own money, I could do my own things with my own money." (Participant 7, Completer Interview)

Discussion
It is notable how the results concerning the 'life domain' contrast with the wider literature on the life impacts of homelessness. For example, in mobilising the empowerment capability approach, McNaughton Nicholls [58] found that under the life domain an intersection of complex factors such as addiction, mental illness, despair and destitution led to homeless people feeling they no longer had a life worth living [58]. These feelings were exacerbated by the lack of control they felt within homelessness systems (Bridgeman and Russell 2020). In contrast, reflecting wider employment research [59], our results show the positive impact that construction employment can have on young peoples' attitudes towards their life. In particular, it shows the importance of individualised support structures within the case study program to overcome barriers such as a lack of formal education and access to government support systems facilitated through employers. Construction employment also appears to offer especially important life impact opportunities by providing opportunities for young homeless people to earn a decent wage and progress a career in an industry where they felt comfortable and valued.
Our findings relating to the bodily health domain reflect the literature, which shows that employment in construction can help reduce the risk of harmful substance use among youth at risk [60]. Supporting previous research using the capabilities empowerment framework on an employment program for homeless people [16], our results also point to improvements in participants' mental and physical health due to the highly regulated nature of construction work and an absolute intolerance to drugs and alcohol on site due to the safety risks involved. This adds a mental health dimension to previous research by Currie et al. [27], whose evaluation of a construction employment program in Canada found that construction employment can lead to better physical health due to the stamina required to work in the industry.
The lack of results relating to bodily integrity were somewhat surprising given previous studies which show the potential benefits that employment can bring to the bodily integrity of homeless people [58]. However, this may be explained by the highly casualised nature of work in construction [61]. Gaetz and O'Grady [29] showed that young homeless people are especially vulnerable to such work and that the jobs they get tend to be part-time, low wage and offer few opportunities for progression. Gaetz and O'Grady [29] also show that homeless people are more likely to rely on illegal ways of making money, which is another risk in construction with the Construction, Forestry, Mining and Energy Union [61] estimating that between 26%-46% of all employment contracts in construction are sham (illegal) contracts-by far the highest of any industry.
Our findings relating to senses, imagination and thought are important and reflect previous research, which shows that the development of basic literacy and numeracy skills is critically important in breaking the cycle of unemployment that can lead to homelessness. For example, Edidin et al. [62], Gaetz and O'Grady [29] and Schwan et al. [33] noted that young people experiencing homelessness often face profound difficulties due to a combination of poor education and undiagnosed learning difficulties, mental health conditions, trauma or harmful substance use issues. As previous research on the empowerment capabilities approach by McNaughton Nicholls [58] suggests, senses, imagination and thought are particularly important for homeless people in reducing barriers to employment and achieving housing stability.
Our results for the emotions domain add a new dimension to previous research on the importance of relational and emotional intelligence in construction [63] by showing how working in construction can help to build emotional skills in homeless people who may have missed this developmental dimension of their life. The findings are important because research shows that homeless young people are structurally disadvantaged in the labor market and find it difficult to form secure and trusting relationships and manage emotions and behaviour [33]. Keats et al. [64] link this to traumatic experiences with adults early in their lives and a lack of supportive and trustworthy adults, teachers and other significant mentors to help them develop their relationship building, decision-making and emotional skills. Addressing these historical problems through working in construction is important because as Schwan et al. [33] found, relationships between young people and employers are key to building their confidence and employability and helping young people stay out of homelessness.
The results relating to the practical reason domain are important because research shows that one of the consequences of being homeless is that long-term thinking and planning is subsumed by the more immediate need for day-to-day survival [29]. Our findings support other research, which suggests that construction employment when supported by appropriate wrap-around support can help homeless young people plan for the future and achieve their goals. For example, Trypuc and Robinson [28] explain that one of the factors which led to the success of the BladeRunner construction program in Canada was that the homeless young people knew they had a support system they could rely on. Likewise, research into the Train for Trades construction program for homeless young people in Canada noted the importance of intensive support and case management in providing the certainty needed for young people to plan ahead and attain personal stability [30].
The affiliation domain findings support research which also shows that vocational construction training can encourage self-confidence, independence, social inclusion and create a sense of connectedness to the community [65,66]. It is widely recognised that homeless people's feelings of stigmatisation and isolation from mainstream society can undermine their pathways to employment [58] and researchers have also recognised the importance of positive social support mechanisms in securing and maintaining employment [67]. As Kidd [68] explains, homeless young people tend to internalise stigmatising beliefs about homelessness, leading to loss of confidence and mental health difficulties which further isolate them from work.
While our results only provided limited evidence that the outdoor nature of construction can strengthen connections with the natural world, it is notable that Nicholls [58] found the 'other species' component of Nussbaum's [40] framework to be of importance to homeless people and that other construction training programs for disadvantaged young people incorporated projects which included building bat boxes, ferret boxes and woodland projects [69]. Batterham [46] explains how homeless people can achieve a heightened sense of connection and belonging through interactions with nature and Roberts et al. [65] found that construction training which used the natural environment as a medium to develop vocational skills reduced the depression and anxiety of participants. In Canada, there is also an innovative construction training program for homeless young people which focusses specifically on green jobs through projects such as energy conservation [30].
The results relating to the play dimension are also interesting, since Rutenfrans-Stupar et al. [70] noted the recreational benefits of employment for homeless people in building a sense of belonging, attachment and connection to the community. Although another study on homelessness and employment noted the difficulties employers had in providing such opportunities [41], it also reported that homeless people reported social activities such as football as an important way to engage in mainstream society and avoid harmful substance use.
Finally, the findings relating to control over one's own environment reflect previous research that noted the tolerance of the construction industry to employing people with a criminal justice record [20], highlighting the potential of construction to help homeless people with such backgrounds escape homelessness through construction employment. It is notable also that although many participants faced transport barriers in getting to work, the project-based nature of construction enabled employment opportunities to be found in areas close to where program participants lived. However, on the downside, Gaetz and O'Grady [29] note that homeless young people often lack income or financial support to pay for clothing and equipment required for being able to work in construction, emphasising that these should be provided as part of any construction employment program for homeless people.

Conclusions
Contributing to the emerging social procurement and social value debate in construction [17][18][19][20][21][22], this paper has mobilised Sen's [38] and Nussbaum's [40] capabilities empowerment framework to explore the contribution which the construction industry can make to reducing the UK's growing youth homelessness problem. Our findings indicate that in responding to social procurement policy requirements, construction employment programs like the one studied here can significantly reduce the risk and impacts of homelessness. They do this by helping young people feel more positive about their lives through increased training and apprenticeship opportunities, increased income, greater personal independence and self-confidence, increased opportunities for recreational activities, improved vision and hope for the future, future employment opportunities and ultimately reduced welfare dependency. Adding to the individual benefits which social procurement van produce for the intended beneficiaries, the results show that this can have significant additional positive spill-over effects on friendships, family relationships and compassion for other people. The physical nature of construction work and the discipline of employment (including zero tolerance to drugs and alcohol) also appears to have the potential to produce important bodily health benefits and improvements in mental health, substance abuse, emotional wellbeing and sleep patterns.
The prospect of real employment opportunities, offered by an industry with high tolerance for employees with criminal justice backgrounds, and opportunities to learn outside a traditional classroom environment in an applied work-based environment also enables the development of basic literacy and numeracy skills through social procurement. These skills can often be missing among homeless young people, with low rates of high school completion a common issue, often due to a combination of undiagnosed learning difficulties, mental health conditions, trauma or harmful substance use issues. The highrisk nature of construction employment and its intolerance of unsafe behaviours and need for emotional intelligence also appear to moderate the emotional vulnerabilities and unpredictability which many homeless people can exhibit. The comradery and teambased nature of construction work can also help to develop missing relationship-building skills, which can, in turn, help to reduce structural disadvantage for homeless people in industries like construction, which largely depend on informal social networks to secure work opportunities. Although construction employment has less of an impact on empathy towards the environment, this depends largely on where construction occurs and the types of projects involved, and there appears to be significant potential to use construction social procurement to also achieve a heightened sense of connection and belonging to nature. This could be especially important in indigenous communities which have strong connections to the land, which are a target for many social procurement policies in countries like Australia, Canada and South Africa [71]. However, the findings also caution that if social procurement is to be effective in reducing the risk of homelessness for young people, then they must be supported by individualised support structures in order to overcome barriers such as a lack of formal education, financial literacy and lack of knowledge about what employers expect.
Although these results begin to address an important gap in social procurement research in construction regarding the potential benefits of construction employment in reducing the risk of homelessness, the limitations of the specific case study initiative reported here are recognised. It is thus acknowledged that the findings reported cannot be generalised without further research in other contexts, although finding young people at risk of homelessness may be challenging given the rarity of programs to tackle this issue in construction. Nevertheless, overall, the results of this case study show that the capability approach is valuable in explaining how intermediate employment programs implemented in response to social procurement requirements can help to address multiple factors, which can contribute to young people experiencing homelessness. In developing the framework further it would be advantageous to involve young people at risk of homelessness in drawing up and interpreting the list of capabilities [42] For example, Bretherton and Pleace [72] examined the effectiveness of an employment program within a theory of change model devised by a homelessness organisation, with the input of homeless people, providing an employment service for homeless people. The model emphasised sustainable employment, financial stability, health and wellbeing, housing stability and good relationships and social networks-all of which are aspects that can be included within the capability empowerment approach.
Finally, in terms of further research, it would be interesting to examine if young people accessing opportunities in different industries leads to the development of different capabilities. For example, three young people reported they were better at managing their emotions because they were in a high-risk construction environment. Not only would it be interesting to examine this further by comparison to youth who have worked in other sectors in this program, it would also be interesting to explore whether there are certain construction trades that may develop some capabilities more than others.

Institutional Review Board Statement:
The study was conducted according to the guidelines of the Declaration of Helsinki, The Code of Ethics for Researchers in the Field of Criminology and approved by a steering group consisting of representatives from Llamau, the Youth Justice Board, academia, the third sector and government.