The impact of employment upon young offenders’ identities

Northumbria University has developed Northumbria Research Link (NRL) to enable users to access the University’s research output. Copyright © and moral rights for items on NRL are retained by the individual author(s) and/or other copyright owners. Single copies of full items can be reproduced, displayed or performed, and given to third parties in any format or medium for personal research or study, educational, or not-for-profit purposes without prior permission or charge, provided the authors, title and full bibliographic details are given, as well as a hyperlink and/or URL to the original metadata page. The content must not be changed in any way. Full items must not be sold commercially in any format or medium without formal permission of the copyright holder. The full policy is available online: http://nrl.northumbria.ac.uk/policies.html

in the youth justice system today are the most persistent and troubled offenders (Taylor, 2016).It is very important therefore to investigate which factors can support youth desistance, in order to improve the lives of young people and reduce the persistence of offending into adult criminal careers.
Both scholars and offenders explain that desistance constitutes a fluid process of abstaining from crime over time, often with numerous setbacks, rather than an instantaneous transformation to the status of 'non-offender' (Halsey, Armstrong & Wright, 2017;Maruna & Farrall, 2004;Paternoster & Bushway, 2009).A concept that is central among criminologists' theorising surrounding desistance is that of 'identity reconstruction'.Many report that the process of desistance involves the offender reconfiguring their self-view and having a new conceptualisation of the person they wish to be (Copp et al., 2020;Farrall & Calverley, 2006;Giordano, Cernkovich & Rudolph, 2002;Maruna, 2001;Vaughan, 2007).However, these studies have primarily been conducted with adults and there has been less exploration of the form identity change takes among young offenders.Furthermore, research by desistance theorists into the role of identity often focuses upon the extent to which the development of a non-offending identity can predict desistance (Bachman et al., 2016;Maruna, 2001;Paternoster et al., 2016;Rocque, Posick & Paternoster, 2014;Shapland & Bottoms, 2011).Less attention has been paid to what can trigger or promote a change in identity.While there is limited evidence that engaging in employment can prompt identity reconfiguration among offenders (Fontin-Dufour & Brassard, 2014;Weaver & McNeill, 2015), again this research has not been conducted with youths.
Therefore, to address this lacuna, this article investigates the influence of engaging in employment at a social enterprise upon the identities of offenders aged 16-18 years.Through detailed analysis of young people's self-narratives at various stages throughout their participation in this programme, this article seeks to ascertain whether young people's identities altered throughout this period and -if a change was observed -the role that engaging in employment had in this.Consequently, this study aims to enhance criminological understanding of the process of identity development among youths, as well as the mechanisms by which employment might promote the formation of a more 'prosocial' identity.

IDENTITY RECONSTRUCTION
Identity is a primary factor in attitudes and behaviour; our self-view provides a direction for, and will be consistent with, our actions (Burke & Reitzes, 1991;Matsueda, 1992;Reynolds & Ceranic, 2007).Therefore, understanding an offender's identity should aid an understanding of their criminal behaviours.Based upon this, Maruna's (2001) Liverpool Desistance Study investigated the life narratives of desisters and persisters.As Dingfelder (2011) states: 'we create ourselves out of the stories we tell about our lives ' (p.42).Thus, the narrative can be understood as more than a retrospective record of life events; it aids the formation of an individual's identity.Maruna discovered that persisters and desisters formed different types of self-stories.Desisters form 'redemption narratives'.They describe themselves as always being a good person; however, they were a victim of bleak chances in life that made them get involved in crime, but they have now taken control and turned their lives around.They desired 'generative' goals, now wishing to give something back to society.Thus, by reconfiguring their past, desisters can conceptualise a 'prosocial' identity for themselves (Maruna, 2001, p.7).This supports desistance because continued involvement in criminal activity would be incongruent with their new identity.Conversely, persisters were found to employ 'condemnation scripts' where they saw themselves as victims of forces outside their control; they had no real hope for change.With such an understanding of their lives, it is unsurprising that these individuals would continue to engage in criminal offending.Thus, Maruna's research led to the understanding that certain narratives may be criminogenic, while others might promote desistance.
A wealth of research has built upon these findings.Prosocial identity formation as a predictor of desistance has received confirmation by several quantitative studies (Bachman et al., 2016;Na, Paternoster & Bachman, 2015;Rocque, Posick & Paternoster, 2014).Moreover, scholars explain that identity formation involves more than the development of a self-narrative that unifies the past events of one's life.Identity also consists of future goals -the 'ideal self' that one is working towards becoming at the moment (Hunter & Farrall, 2018;Presser, 2010;Vaughan, 2007).Paternoster & Bushway (2009) added to this the notion of a 'feared self': the person they might become if they fail to change.The imaginings of the two possible selves provide the individual with guidance on how to achieve the positive future self and avoid the negative possible self.

THE DRIVERS OF IDENTITY CHANGE
The formation of a non-offending or 'prosocial' identity can promote desistance because offending is no longer compatible with who the individual sees themselves as.However, it is less clear what drives such a change in identity.Many studies emphasise the importance of building 'agency' (Healy, 2014;King, 2013;Munford & Sanders, 2015;Rocque, Posick & Paternoster, 2014).However, feelings of agency are conditioned by each individual's social context.There needs to be an opportunity available within their immediate social setting that might support an alternative identity; Giordano, Cernkovich & Rudolph (2002) term these 'hooks for change'.If such opportunities are limited then this necessarily restricts feelings of agency and the ability to envision a prosocial identity (Healy & O'Donnell, 2009;King, 2012;Rumgay, 2004).
A small number of studies acknowledge that employment can be a 'hook' for identity change.Fontin-Dufour & Brassard (2014, p.324) found that once offenders take on the social identities of workers they begin to 'assimilate into a culture of good citizenship', and gradually leave the criminal identity behind them.Furthermore, Weaver & McNeill (2015) reported that the development of new social relationships through work 'afforded a concrete way of enhancing one's own identity as a respectable person' (p.101).Their participants, offenders who commenced employment in the steel-fixing industry, replaced their criminal identities with more constructive reputations as 'workers'.However, overall, the impact of engaging in employment upon identity remains unclear.In particular, there is little understanding of the mechanisms by which employment might aid prosocial identity formation.

YOUTH AND IDENTITY CHANGE
Most studies of identity and desistance have explored this topic with adults.The research that has considered youths suggests that -as with adults -desistance is associated with developing a non-criminal identity (Hazel et al., 2017;McMahon & Jump, 2018).However, identity change may take a different form for these individuals.Nugent & Schinkel (2016) examined narratives of five desisters in the UK aged 16-21 years and found that their identity was based on seeking to avoid old places/associates to ensure they no longer offended.However, it did not go beyond this; due to a lack of opportunities, they had not conceived of a future 'self' as part of conventional society.
Although not a study of desistance, Munford & Sanders (2015) explored identity development with 13-to 17-year-old youth justice service users in New Zealand.They found that due to severe social exclusion, their identity horizons were limited -they had little that could act as a foundation in the search for a new identity.This research suggests that young people's identity reconstruction may be constrained by a lack of resources/opportunities.Because children have less ability than adults to influence their lives, they may feel less able to draw successfully upon opportunities that could support the conception of a 'prosocial' identity.Consequently, they may be more likely to require external sources to aid their change in identity (Haigh, 2009).Thus, scholarship demonstrates that a change in identity is an important part of the desistance process.However, whether engagement in employment can promote the formation of a prosocial identity is unclear.Moreover, very few studies have considered the identity development of offenders under the age of 18 years.This article aims to expand criminological understanding in these areas by investigating the impact of engaging in an employment programme upon the identities of offenders aged 16-18 years.

METHODOLOGY
This article is based on data collected during 2017-2019 for a PhD project involving young people (n = 23) employed by the Green Light 1 (GL) social enterprise.The GL provides young offenders with six months' paid employment, as an opportunity to turn their lives around.Any youths between the ages of 16 and 18 years, who are, or have recently been, under the supervision of the local youth offending team (YOT) and are assessed as either low or medium risk of harm, are eligible to be involved in the scheme, though participation is voluntary.All the work that the GL youths undertake is outdoors, such as fencing, allotment clearance, paving, grass cutting, flood mitigation and litter picking.For risk-management and mentoring purposes, young people work in small cohort groups -a maximum of five employees and the supervisor.Young people work approximately 30 hours per week and receive considerable support to gradually become attuned to working and develop a more routine and structured day.Access to participants was gained through the social enterprise, which acted as a gatekeeper.The GL currently operates in ten locations in the UK.No selection process was made for participants; they were the entire cohort groups for the three GL sites (all in northern England) surveyed for the period of this research.Twenty-two of the young people were male and one was female.One young person was Black, two young people were of mixed race, and all the others were white.All the employees had multiple convictions prior to engaging in the GL, the average -referring to official records -was twelve.The most common offences on their records were criminal damage, burglary, common assault and shoplifting -the Youth Justice Board grades these as mid-serious offences.
A relatively small number of participants was chosen to allow repeated contact and to build relationships with each of the youths.Understanding 'identity', requires more than a surfacelevel investigation into the social realties and life-worlds of participants.Furthermore, while the number of participants may appear small, the number of youths entering the youth justice system in England and Wales is decreasing (Ministry of Justice, 2021).Indeed, the participants of this study still represented approximately 7% of all medium-risk, 16-to 18-year-old service users of the three localities surveyed during the period of the research.
Interviews were the primary source of data for this article.In previous studies, to explore identity, researchers conducted narrative interviews (Bachman et al., 2016;Carlsson, 2013;Gadd & Farrall, 2004;King, 2012;Nugent & Schinkel, 2016;Presser, 2016).As this article is based on data collected for a PhD project, that had other aims beyond exploring the identity of young offenders, interviews were conducted that included questions designed to elucidate self-narratives from young people, but the interview was not exclusively focused upon producing an autobiographical account of their lives.I was particularly interested in how young people described their offending, now that they are working at the GL.Scholars explain that the formation of a prosocial identity involves redefining past negative behaviours to be consistent with their new law-abiding self (Harris, 2011;Maruna, 2001;Maruna & Roy, 2007).In interviews with young people, I also discussed their plans and aspirations.The formation of a desistance-supporting identity requires a shift in goals: to achieve an imagined prosocial self (Healy & O'Donnell, 2009) and to avoid an imagined feared self (Paternoster & Bushway, 2009).To aid my investigation into whether there had been a change in young people's identities during their participation in employment, narrative questions were included in semi-structured interviews with young people on their first week, three months in, on their last week of their employment with the GL (six months), and six months after their employment ended (one year).While the first-week, three-month and six-month interviews were conducted face-to-face, one-year follow-up interviews were conducted either by telephone or via social media.Data were evaluated using a narrative analysis approach, encompassing both what was narrated and how it was narrated.Guidance was taken from O'Connor (2000) and Presser & Sandburg (2015) who outline the importance of genre, symbolic boundary drawing, justifying devices, deflecting agency, repetitions and inconsistencies, when conducting narrative analysis.
Interviews with GL supervisors and participant observations at GL worksites also provided information on the identity development of young people.I inquired with supervisors and observed myself if the behaviour and attitudes of each young person changed throughout their placement at the GL.Such a change might indicate an alteration in self-view.The records held for each young person at the YOT were a source of data on the offending activities of participants.These included the date and description of every offence as recorded by the police.When collecting data from these records the primary focus was on offences for which young people had been convicted; those that had been 'withdrawn', 'dismissed' or where they were found 'not guilty' were discounted.
The project was approved by the University Ethics Committee.Conducting observations and interviews with young people, most of whom had long been disengaged from school, presented a potential ethical issue.They may have been less able to comprehend the nature of the research and the risks of participating (Caulfield & Hill, 2014).I mitigated this by designing user-friendly information sheets for young people and I spent time explaining the nature of the project to participants.The consent of a parent/guardian for those aged under 18 years to participate was also obtained.

FINDINGS
In presenting the findings, I first discuss the development of young people's narratives during their participation at the GL.Second, I explore how engaging in employment influenced young people's identities.Third, I detail young people's identity development and offending behaviours once they left the GL.

GL participants' self-narratives
During this research, a number of transitions in both the content and structure of young people's self-stories were observed, which may therefore indicate a transformation in self-view.Broadly, the findings revealed that during their time at the GL, young people began to construct a story of change/reform.At the same time, they sought to distance themselves from past offending and present a more prosocial self in their stories.They also detailed a more concrete 'future self' engaged in legitimate employment.Each of these key transitions was explored in further detail.
Most young people began the GL scheme being open to change.This can be interpreted not only from their voluntary participation in the employment programme, but from the assertions made by several participants at both first-week and three-month interviews that they needed to change their behaviours.However, as they advanced through their employment, there was an observable shift in young people's narratives, and many described themselves as already having changed.Self-stories depicted the young person reaching a critical 'turning point' in the past, where they knew their lives would get much worse if they kept offending.Thus, for example, Joseph described the realisation that he was limiting his future work opportunities by offending: I was just realising that it stops you from doing a lot of shit, like being able to get a good job . . .like you've only got specific jobs you can do now, now that you've got a criminal record and shit.It just gets worse if you keep doing it.
Furthermore, William depicted his relationship with his family reaching a critical point: It were mainly like my family, when I were getting in shit and that they were like basically like pushing me out of the family and obviously I like started to realise and just thinking like if I don't stop I'm going to like be on my own, so . . .Paternoster & Bushway's (2009) 'identity theory of desistance', which purports that what triggers a change in identity is the realisation on the part of the individual that their criminal offending is more costly than beneficial -there is a 'crystallisation of discontent' with offending.That young people described their 'turning point' arising prior to their engagement in the GL, could suggest that they began to reconstruct their identity before participation in employment.However, it was only in the six-month interviews that young people described having reached a turning point in the past.There was no mention of this in any of the prior interviews.Thus, after experiencing employment, participants appeared to be rewriting their pasts, to describe themselves as being on a path to change they did not see themselves on before.

These findings complement
Another notable shift was in how young people described their experiences with the GL.In the first-week interviews, young people depicted their YOT caseworker 'putting' them on the GL.Participants did not express any agency in choosing to partake in this scheme and were very passive regarding their efforts in attaining a place on this employment programme.However, by the six-month interviews, most young people rewrote their stories; their involvement in the GL was an agentic move to support their resolve to change.While, as specified above, several young people described in the six-month interviews having reached a 'turning point' and made a decision to stop offending before engaging with the GL, they also stated that the GL helped support this choice.My interview with John illustrated this: Interviewer: So why did you stop offending?John: Because I didn't want to go back in the jail.I didn't want to be that person anymore.I want to be this person.I want to work.
Interviewer: So would you say you were already changed before you had started the [GL]?
John: Aye, I was on the right tracks.But then that's [the GL] just led us.It's like, I were on the track and then getting on this, has put us more on that track, know what a mean?This is exactly what I needed, this here.
Indeed, most participants recalled in their narratives both having chosen to change themselves and the GL having assisted their change.Thus, as participants progressed through this employment programme, self-narratives altered, and the GL became integrated into their story of change.
Coinciding with this emerging story of change, was a significant transition in the way in which young people accounted for their past offending.During their employment, GL attendees increasingly sought to deny their responsibility for past offences and separate themselves from them.At the first-week interviews, at the beginning of their GL placement, young people minimised their involvement in offending.It was not just what they told in their narratives, but also how they told it, which demonstrated this intention to minimise.Some would not mention their criminal past, while others would use words other than 'committing crime' to depict their past involvement in offending, such as: 'got in some bother' (George); 'I used to go to the YOT' (Julie); 'I got an, emm, YOT order' (John).In the three-month interview, participants continued to use minimising techniques.However, young people also used techniques to deflect the focus from their involvement in the offence.Again, how they told their stories was very revealing.For example, Scott's deliberate avoidance of 'I' in his account of past offending -'At YOT for fighting and affray.Then stole a car', and Glenn's awkward phrasing -'I had a . . . got a . . .stand-off'.O'Connor (2000) highlights how passivising structures can be used to remove focus from the storyteller's agency.The way in which young people told the story was to detach themselves from the offence.
By the six-month interview, in addition to minimising past crimes and detaching themselves from the offence, participants were also actively attaching the responsibility for past offences to external circumstances and other individuals.Table 1

displays this.
The categories in this table can be classified as 'techniques of neutralisation' (Sykes & Matza, 1957).According to Maruna & Copes (2005), neutralisations allow offenders to avoid a negative self-image, as they still see themselves committed to the dominant normative system.Likewise, Harris (2011) reported that offenders use excuses for past behaviours to distance themselves from their former selves and 'to re-create a possible self still worthy to be redeemed in the future' (p.2).That GL employees used such methods in telling the story of their criminal pasts, suggests that they wished to portray themselves as inherently prosocial; past offending had been for reasons outwith their control or had been minimal anyway.
This desire to separate themselves from their past offences also corresponded with a more overt presentation of a prosocial self in the GL youths' narratives.As highlighted earlier, in the first-week interviews, some young people minimised their past involvement in crime, while others acknowledged that they needed to change their behaviours.However, it was not until the three-month and six-month interviews, that young people made efforts to present a particularly 'altruistic' self-image in their stories.For example, Scott stated: 'I always try and be polite to people 'cos I always act like how I would want people to be to me'.Furthermore, in several of the TA B L E 1 Avoidance techniques of youths when accounting for past offences

Minimising the crime
It was seven year ago . . .if it had been that much of a great deal they wouldn't have allowed us to work in the nursery (Julie) Just a bit of daftness (John) We just went daft (Glenn) It was just common assault and affray, that's all it was really (Stephen) Just fighting with the bissies * (Gary) It was just having a fight with someone outside me door.That's all it was (Jay)

Extenuating circumstances
On the streets you get bored, so that's when you start doing stuff (Scott) There wasn't really nowt to do was there on the streets (George) I was trying to get the money for the drugs . . .you need to get the money from somewhere (John) I was just going out to get money really (Scott) Literally it was just for money (Dean) I needed to get a name for me-self, if you have a reputation no-one will mess with you (Stephen)

Blame attached to another
In the care home like the kids would make us lose our temper (Julie) Me mates were egging us on (Sam) I grew up with these lads, I had to just join in (Alexander) It's the people I hang about with that made us do stuff (Max) Note: *Slang for police.
participants' narratives, they distinguished between themselves and those who they perceived to be more anti-social.For example, Jay was keen to portray to me that he was not like the others in his work group: Interviewer: Do you think you've got stuff in common with the other lads?Jay: Hmm . . .nah.I'm different from them . . .I haven't got anything in common with them.They just sit outside the bus station and get pissed . . .I would only drink in a bar . . .or in the house, I'd never think let's sit in this field and get pissed.
Presser & Sandburg (2015) describe the importance of symbolic boundary drawing as a method for the storyteller to present a positive self-image.
A final distinct transformation that could be observed in young people's self-stories involved their depiction of future goals.Almost all young people expressed desiring a 'full-time' 'reliable' job in the future even in the first-week interviews, before they had much experience of working at the GL.However, most were vague about which occupations they aspired to at this early stage.Some participants 'feared' future self focused on unemployment and financial hardship.For example, William described: I would want to be stable at a young age -and have a nice bit of money in the bank, house, car.'Cos a lot of people don't, you know what I mean?I see people at 30 and they've got no life . . .I want to be secure.These desires/fears for the future were also present in the three-month narratives.At this point some young participants named specific occupations they desired, such as roofer, personal trainer and electrician.Furthermore, some young people described beginning to take proactive steps to improve their chances of gaining future employment.
The six-month interviews revealed that many young people had envisioned, and in some instances had taken action towards, the attainment of specific employment opportunities post-GL.In their narratives, young people described being hopeful about securing future employment, however, many also gave expressions of a lack of confidence in this coming to fruition.Thus, during their involvement in the GL, most young people were increasingly able to envision a prosocial 'ideal self' engaged in full-time employment and some had taken steps towards achieving this imagined self.However, even by six months into the scheme, young people's confidence in their ability to realise this future self was limited.
Most participants' self-stories revealed that their ideal 'future self' was also a non-offender.Indeed, as they progressed through the programme, increasingly young people mentioned the possibility of being imprisoned as their most 'feared' future self.As John articulated: You could be sitting in a cell on a day like this and looking out your window.You cannot open your window you cannot do nothing.I don't want to go back in the jail, if that's me path then I swear to God just shoot me now, there's no point going back there.
Similarly, Glenn communicated in his narrative: If I don't change myself I'm gonnay end up in the doghouse.That's not a place I want to be . . .I want to be outside working not inside in a cell.As long as I work hard I should be free from getting locked up all the time.However, even by the six-month interviews, when I asked young people whether they thought they would reoffend in the future, their replies avoided expressing their full agency.For example, responses included: 'not purposely' (Joseph); 'hopefully not' (Sam, John, Kyle); 'can't say' (Darrell); 'you never know what the future holds.I don't want to, but there could be a reason in the future why I have to' (Glenn); 'touch wood no' (Scott); and 'who knows?' (Jay).Presser & Sandburg (2015) highlight the importance in narrative analysis of the linguistics that can present someone as passive to other forces.It appears that, despite young people determining over their time at the GL that incarceration was their 'worst case' future self, many could not guarantee that their future self would be a non-offender.
Overall, the findings presented in this section suggest that participants (on the whole and not without exceptions) formed a more coherent prosocial self-view during GL participation.Most young people began the GL scheme being open to change.However, as they advanced through their employment, they described themselves as firmly on a changed path, a decision they had made previously, but which had been aided by their choice to participate in the GL.Concurrently, young people sought to greater separate themselves from past offences and negative behaviours.This is further illustrated in how they recounted their lives at different time intervals; these also became more coherent.At the first-week interviews, young people's self-story was disjointed; they admitted needing to change their behaviours (although they also minimised these), they felt they had been 'put' on the GL by the YOT and they wanted to gain full-time employment in the future, though they were unsure in what.However, by six months, young people's self-story was that they had -for reasons primarily outside their control -been involved in offending but decided to make a change because they knew things would get much worse.They chose to partake in the GL to help with this change and (for most) it aided their vision of a positive future self, which they were now working towards.Thus, by six months, the participants had a more coherent story, with a moral message contained within it, particularly those who included examples of their new altruistic self; they were the hero in the story who saved themselves from the abyss.

The influence of employment
Building upon the findings presented thus far, the following sections consider how engagement in employment at the GL supported a stronger prosocial identity for youths.

New purposes
Working at the GL strengthened young people's vision of a prosocial 'future self' employed in the legitimate sphere.Learning new skills boosted young people's confidence that they could engage in legitimate work in the future, reinforcing their imagined prosocial self.As Scott stated: It covers quite a lot . . .a wide variety of stuff . . .We've done from picking weeds to painting benches to bricklaying to like all sorts, cleaning cars and we've done absolutely all sorts.So loadsa those jobs now I reckon I'd be ok at . . .I won't look like a complete idiot.
Similarly, Max affirmed: It's learning aspects of jobs I wouldn't normally do.We've done fencing, paving, made those river steps, strimming and gardening and making signs.If I come across it in another job I'll be ready to do it . . .I'm more confident now.That's the best thing about the [GL], it's getting us ready.
Furthermore, experiencing personal achievement at work aided some young people's ability to envision a prosocial future self.As John's words illustrated: Doing all this stuff, it's a challenge and I've seen me better self.Now that I've stuck at these six-months, done me job, I know I can do it.I could go to work out there now . . .I don't need to make me money from crime.
Thus, the GL introduced young people to some of the skills necessary for future employment and allowed them to experience a degree of success in the conventional sphere.This gave participants confidence that a future as a legitimate worker was a realistic possibility and aided their prosocial identity development.
Completing physically demanding work helped some young males visualise a prosocial future self.Scholars report that traditional ideals of masculinity, such as 'toughness' and 'strength' and 'machismo', can be achieved through manual labour (Halsey, Armstrong & Wright, 2017;Uggen, 2000;Weaver & McNeill, 2015).From observations, I witnessed some young people adopting a 'grafter' role at the GL and taking pride in this newfound purpose.As Darrell asserted: this is the sort of work males my age should be doing, its physical work.They need tough lads like us to get this done; we can get the job done on time.
Likewise, Alexander described how participating in manual work was more appropriate for males of his age group than offending: I'm doing good now.Instead of knocking the shit out of some arsehole, I'm taking it out on that tree, getting it cut down so people and their dogs and that can get past.And that's what I should be doing now, I'm not a fucking kid . . .I've seen 20 year olds and that put windows through and that and run off 'cos they want a police chase.That's something a 14, 15 year old should be doing.Not a 20 year old.People just don't know how to grow up properly.
Achieving masculine gender expectations was particularly valued by participants.Carlsson (2013) reports that participating in law-abiding employment can be a way to meet masculine gender expectations after adolescence when offending is no longer status-building.Being able to adopt the role of tough manual worker at the GL allowed young males to still achieve valued gender norms in the legitimate sphere.This realisation strengthened some young people's emerging prosocial identity.

Relatedness
At the GL, all employees were of a similar age and had comparable histories of YOT involvement.
Frequently work tasks required the youths to work together in small teams.Consequently, most GL employees formed bonds with their co-workers.This supported their prosocial identity development.As stated by McNeill & Maruna (2008) if an offender is going to take the risk to transform their identity, they need to feel that they have social support to undertake this precarious venture.For example, Scott distinguished his friends on the GL from those outside: My friends outside, although they are my friends, they don't have my best interests at heart.These guys like want to see you do well, they want to see you go further.We all want to get a job at the end of it so we are all teaching each other new skills, we are all helping each other.
Likewise, Joseph described how friendships with other employees supported his process of change: It's like when I weren't on it [the GL] and I were hanging around with my mates I were starting to like realise and obviously I started to want to change myself and change my ways, but they weren't seeing it from that point of view.They were like -'oh you won't get caught, do this, do that'.It were just like being around people who were realising like obviously it's stupid and they want to change themselves and they actually want a decent life.
It may seem counter-intuitive that bonds with other offenders would support prosocial identity development.However, the findings indicated that even those with a history of criminal involvement can act as prosocial role models.Promoting close bonds among young offenders who -crucially -were open to change, provided social support and created a safe environment to try out new identities.
The GL supervisor also functioned as a role model for identity change.In some instances, supervisors admitted to participants that they, too, had been deviant in their youth.Consequently, supervisors were 'credible' role models; through their efforts to be relatable and their work at the GL, they demonstrated a different lifestyle or 'future self' that young people could choose.Indeed, some youths described wishing to do the supervisor's job in the future.

5.2.3
Changing the perceptions of others Much of the work at the GL visibly benefitted local communities.The community frequently responded positively to the GL employees; often passers-by thanked and complimented them on their work.For youths who were used to the negative reactions of others because of their antisocial behaviour, this was a big change.Indeed, Joseph expressed: everyone in that neighbourhood thinks I'm a little bastard, because -to be fair -I was, but now they are glad I'm working.They all recognise me, you nah, and they say you're doing good.I was like -fucking hell.Nice to hear though innit?
Scholars explain that how one conceives of their identity may be influenced, or even determined, by how they feel others around them perceive their identity (Castree, Kitchen & Rogers, 2013;Cooley, 1902;Goffman, 1956;Jenkins, 2004;Mead, 1934).Therefore, when others label us as prosocial this will inevitably make it more likely that we will internalise a prosocial identity.As Scott stated: It's nice that these people don't know your past and they don't know what you've done they just know who you are now.They just see us fixing up the neighbourhood, helping people out.
The GL youths saw their developing prosocial self-view reflected back to them in the eyes of the general public.Resonating with Maruna's (2001) notion of generativity, the types of work young people engaged in at the GL (and the resultant response from others) allowed young people to distance themselves from past labels as the 'local troublemakers' and perceive themselves as giving something back to the community, strengthening their configuration of a prosocial identity.Some young people also described their family viewing them more positively since they had been working at the GL.This was illustrated in my conversation with John: Interviewer: Do you think working at the [GL] has affected your relationship with your family?
John: It's made it better really.Better now I'm deying something good with meself.I wasn't in this place a couple of years ago, I wasn't in this headspace, so I've made them proud ya naw, I've changed me life around.I've been to jail and that, ya naw, so now they can see that I'm not that person and I'm willing to work, I can work, I can help people.They know I'm not that person.They see me as a different person, 100%.
Several other participants expressed that their families were proud of them for working at the GL.Due to their offending, young people's relationship with their families had often become strained prior to their engagement with the GL.John's comments demonstrated how this changed when working at the GL; family members began to see them as a 'new person' and were proud of their commitment to prosocial activities.This necessarily impacted upon their identity.
Supervisors also made clear to young people that because they were working, they would regard them differently.As Greg stated: I drill it in to them, they're ex-offenders; you've done that, you've moved on.Everybody has a past and this is the start of your new life.You're a worker now, and what you are doing is good for people and good for the community.Thus, GL supervisors made an effort to treat young people as conventional workers, rather than young offenders while they were engaging in the programme.As the GL employees progressed through their placement, supervisors allowed them to complete work tasks with less supervision.Supervisors described how, because they were trusted at the GL, this became a self-fulfilling prophecy; young people believed that they must indeed be trustworthy.This strengthened their prosocial identity.

Identity and offending post-GL employment
Although the GL provided only six months' work, supervisors and YOT workers assisted youths to progress into further employment.Indeed, many young people were employed upon leaving the GL, examples included: casual construction worker; call-centre worker; bartender; warehouse operative; cleaner; and factory worker.Some young people also progressed into various vocational training programmes.There were limitations to the data collected from one-year interviews; the reliance upon phone calls and social media to regain contact with participants meant that I could not obtain the detailed self-narratives that the face-to-face interviews had provided.Nevertheless, follow-up interviews indicated that most youths had maintained the stronger prosocial identity they developed during their time at the GL.In the one-year interviews, they still described participation in the GL as a good decision they had made and several acknowledged that their involvement in the programme had prevented them from reoffending.Furthermore, despite participants primarily being employed in insecure work, with little opportunity for progression, most were still working towards an 'ideal future self' engaged in stable, full-time employment.For example, Jay expressed being 'here there and everywhere trying to look for something more permanent, a career rather than a job'.Did this stronger prosocial identity influence a reduction in criminal activities among youths?Young people's narratives by the end of their employment with the GL undoubtedly told a story of a changed self, and most reported having ceased offending at both six-month and one-year interviews.This was also confirmed by supervisors.Furthermore, police records demonstrated that -compared with pre-GL involvement -youths' volume of criminal convictions decreased by 78.1% during GL participation and 46.5% in the six months after GL participation.All those who persisted in offending after GL participation were Not in Education, Employment, or Training (NEET).Several supervisors described how these individuals returned to spending their time with pro-criminal families and friends who refused to recognise their changed self.For example, the Landington supervisor described Gary having changed significantly since he had engaged in the GL.However, his family were constraining this transformation: Gary wants to work.Gary really, really wants to . . .Like if he got a job that could take him away from his family . . .that's what he wants.Sometimes it's frustrating when you see it and all, and they are trying to draw him back.
Unfortunately, post-GL, Gary returned to this pro-criminal environment.Not surrounded by anyone who would accept and reaffirm his changed identity, he slipped back into offending.This, again, demonstrated the importance of positive reaffirmation by others for identity change.

DISCUSSION AND CONCLUSIONS
The change in identity observed in this study was not a dramatic move from a 'condemnation script' to 'redemption narrative' as exhibited in Maruna's (2001) research.Many participants, even by the six-month interviews, expressed real uncertainties as to whether they would gain legitimate employment and stop offending in the future.However, I do not believe that this discounts my assertion that participants formed a stronger prosocial identity while at the GL.While young people could not say with confidence that they would not reoffend, this was not comparable with the 'doomed to deviance' narratives of the persistent offenders in Maruna's study.A lack of agency when discussing the future has been found in other research with vulnerable youths, whose lives are frequently characterised by powerlessness and instability (Fitzpatrick, McGuire & Dickson, 2015;Haigh, 2009;Munford & Sanders, 2015).Indeed, the main reason why the identity change among participants was subtle was that none of the young people displayed a strong criminal identity when they commenced the scheme.Many participants had been actively offending in the few months prior to engaging in the GL and indeed (according to official records) during their first few months' employment.Yet, even in the firstweek interviews, most participants tried to minimise their involvement in offending and nearly all expressed desiring legitimate employment in the future.This was a reasonably surprising finding.Giordano, Cernkovich & Rudolph (2002) purport that the fashioning of a new 'conventional self' occurs only after, and as a result of, involvement in conventional roles or hooks, such as employment.Yet most GL employees had envisioned a future self engaged in full-time work before they had had much involvement with the scheme.Furthermore, existing research would suggest that active offenders would possess a deviant self-view, as scholars purport that our actions are consistent with our 'identity' (Maruna, 2001;Matsueda, 1992;Reynolds & Ceranic, 2007).
It is a possibility that young people portrayed a largely prosocial self in their interviews because of the difference in positionality between us.The context of data gathering can shape the telling of self-narratives (Pasupathi & Rich, 2005;Rajah, Kramer & Sung, 2014).Appearing as both a female and a student put me firmly in the non-offending category to young people.This might have made them less willing to represent themselves as an offender.Nonetheless, by including participant observations in the research design, this gave me an opportunity to build rapport with young people.Moreover, by displaying empathy and acknowledging what participants were saying without judgment I believe participants felt that they could represent their 'true selves' to me in interviews.Consequently, it is more likely these findings affirm the assertions of Walters (2019) and Wigzell (2021) -that youths are less likely to possess a robust criminal identity because they have not been involved in offending long enough.Furthermore, the participants in my study were not the extreme persisters of Maruna's sample; despite some continuing with criminal activities, they had voluntarily engaged in an employment programme, suggesting that they were at least open to change.
This article, therefore, has implications for criminologists' theorising surrounding identity change.In considering the identity change of a younger population than those typically surveyed, this study has identified the complexity of the connection between self-view and criminal behaviours.Youths may not have a strong criminal identity, even when actively offending.This also has policy implications.A number of recent advocates for change within the youth justice system in England and Wales purport that more needs to be done to support youth transitions from a deviant to a prosocial identity (Case & Browning, 2021;Goodfellow et al., 2015;Hampson, 2018;Hazel et al., 2017).Such reports assume that research based on adult offender identities can be directly translated into youth justice practice.Yet the findings of this research, confirming that of Wigzell (2021), urge caution in assuming that all young offenders possess a pro-criminal identity, which requires youth justice intervention to change.
Although youths did not display the deviant identity that might have been expected of persistent offenders, the findings demonstrate that employment at the GL had a positive impact upon the development of a more coherent prosocial identity.Many participants had experienced poverty, workless households, homelessness, mental/physical ill health and educational underachievement and thus emanate from a population frequently termed the 'socially excluded' (Percy-Smith, 2000).When scrutinising the findings further, it becomes apparent that what was key to the youths' development of a stronger prosocial identity is how employment at the GL promoted their inclusion within conventional society.For instance, by successfully completing tasks, learning new skills and adopting a 'grafter' role, GL employees perceived that they could function effectively in law-abiding society, making them feel part of the conventional sphere.In addition, the findings illustrate that employment affected participants' relationships to create a sense of inclusion within law-abiding society.For example, doing work that benefitted local communities engendered the compliments of the public, which gave young people a sense of being accepted by wider society.Moreover, employment at the GL created bonds between co-workers, which led to social capital built with individuals within conventional society (or -in the case of the GL -offenders who were open to engagement with conventional society).Thus, these aspects of employment at the GL made participants feel more integrated within conventional society -reducing their social exclusion according to Burchardt, Le Grand & Piachaud's (1999) definition -and aiding prosocial identity development.
Would any employment promote social inclusion?It should be acknowledged that the GL was a social enterprise and not analogous to regular employment.Only at an 'employment programme' could employees form bonds with other offenders, which would provide them with credible role models and support for change.Similarly, the extra support and mentoring offered by the supervisor might not be expected otherwise than at a social enterprise.Furthermore, Crutchfield (2014) and Standing (2011) explain that due to deindustrialisation and neoliberalism, an increasingly significant portion of employment is now 'precariat work'.Such occupations are insecure, low-skilled, provide few opportunities for advancement and can be socially isolating.Thus, precariat work prevents the formation of an occupational identity.This may be particularly the case for males who find themselves unable to find a traditional masculine 'role' in a labour market that is orientated to the service sector and customer care (Simpson, Hughes & Slutskaya, 2016).
Engagement in precariat employment might therefore do little to reduce the social exclusion of young offenders and further their development of a prosocial identity.As Rajah, Kramer & Sung (2014) detail, youths' prosocial identities can diminish under conditions of social exclusion.Yet, due to their criminal records and (typical) lack of qualifications, criminalised youths are often precariat workers (Standing, 2011).Indeed, most participants in this study progressed into such employment upon leaving the GL.Nevertheless, the one-year interviews indicated that young people's prosocial identity had held, despite potentially being little there to support it.However, it is possible that the 'good effects' of participating in the GL had not worn off yet for these participants.It is important that future research explores this further; can a non-offending identity be built upon, or sustained by, precariat employment?This is particularly important in the wake of the Covid-19 pandemic, where precarious employment is predicted to increase and become even more insecure, both in the UK and worldwide (Gibson et al., 2021;Matilla-Santander et al., 2021;O'Keeffe, Johnson & Daley, 2021).
Overall, this article demonstrates that social inclusion through employment can aid the formation of a stronger prosocial identity among youths.This highlights the importance of young people being able to gain meaningful, rather than precarious, employment and the value of social enterprise in providing such employment for those who would otherwise struggle to obtain this in contemporary labour markets (for a more detailed discussion see Oswald (2021) and Soppitt, Oswald & Walker (2021)).While acknowledging the limitations of a small dataset, self-reports, supervisor's reports and official records at least suggest that this more coherent prosocial identity influenced a reduction in offending activities for young people.Nonetheless, time constraints on this project did not allow for exploration of longer-term processes of prosocial identity development and desistance.It is important, therefore, that future research investigates this further with other instances where young offenders engage with employment, potentially incorporating a longitudinal research design.

A C K N O W L E D G E M E N T S
I would like to thank the 23 young people who participated in this research.I would also like to express my gratitude to all the social enterprise staff at the three locations I researched in this project.This work was supported by the award of a Northumbria University PhD Scholarship.

E N D N O T E
1 The Green Light (GL) is a pseudonym, as are all names presented in this article.