Diverse Diversions: Youth Justice Reform, Localized Practices, and a ‘New Interventionist Diversion’?

The recent resurgence of practices aimed at ‘diverting’ young people from prosecution appears to suggest a sea change from the interventionism which characterized New Labour’s approach to young law-breakers. Drawing on interviews with youth justice practitioners at two sites in England, we argue this is overly simplistic, since the ‘interventionist diversion’ they describe reflects the continued influence of New Labour reforms, as well as older approaches. We conclude that more empirical research is needed to establish where such interventions sit within the broader – and increasingly localized – landscape of support provision, as well as the consequences of providing ‘welfare’ in this way.

Various explanations have been posited, which include: a probable decline in the volume of youth crime; 1 the contribution of early intervention programmes in preventing and reducing crime; 'creative maths' in the selection of disposals within counting rules (Morgan, 2009); and the 'main' factor identified by witnesses to a recent inquiry by the Justice Select Committee: '…changes to the way in which offending is dealt with by the authorities' (House of Commons, 2013: 7). Encouraged by shrinking public sector budgets and the introduction of so-called 'austerity measures' following the 2008 global financial crisis, as well as the revision of official performance measures (discussed in more detail below), there is evidence of practice innovations that have outstripped the pace of policy change (see, for example, Bateman, 2014). Significant among these, and the focus of this article, are the various pre-statutory schemes set up by police forces and youth justice services to 'divert' young people accused of first or minor offences from prosecution, and from FTE statistics (Haines et al., 2012;Home Office 2012;Youth Justice Board, 2011).
The examination of these diversion schemes is a growing area of interest, politically and among scholars (see also Bateman, 2012Bateman, , 2014Carlile, 2014;Cushing, 2014;Haines et al., 2013;Richards, 2014;Smith, 2014). Drawing on qualitative interviews with youth justice practitioners and documentary research conducted as part of a broader research project exploring policy implementation in youth justice, we consider such initiatives in two locations in England. We explore managerial claims that the new diversionary focus represents a 'sea change' in practice, but argue for a more nuanced assessment. We suggest that current practices reflect the continued influence of New Labour reforms, particularly in relation to standardization of practice, actuarial risk assessment and a commitment to welfare-oriented early intervention. This leads us to argue that a 'new interventionist diversion' is observable, which differs in important aspects from the non-interventionist understandings of 'diversion' prominent during earlier historical periods -notably in England and Wales during the 1980s and early 1990s (Allen, 1991;Bateman, 2002).
In noting ambiguity in the meanings of and rationales for diversion, and the ways in which this label has been applied to highly varied diversionary practices, our argument reflects claims recently made by Richards (2014) about diversion in the context of Australian youth justice. The empirical findings presented here indicate, however, that the conceptual project she recommends of detangling the domains of 'primary crime prevention' (which we refer to as universal support services), 'early intervention' and 'diversion' is beset with difficulties -at least within the studied jurisdiction. Our research suggests the notion that 'diversion from crime' depends on intervention through which criminogenic risk factors can be mitigated (for example, Farrington, 2000) is now broadly accepted by practitioners as well as policy makers. Thus, intervention without criminalization in the form of a criminal record was judged to be both necessary and helpful by many of our participants, especially in the context of reduced investment in universal child and family support services in England as in other neo-liberal societies (Connolly and Morris, 2012), the removal of 'ring-fenced' funding for youth justice prevention work in the context of reduced Youth Offending Team (YOT) budgets (House of Commons, 2013) 2 and raised thresholds and an increased early years focus within specialist child protection (Allen, 2011;Carlile, 2014). We would therefore argue, reflecting comments made about 'risk' and 'need' in early intervention services (see for example, Smith, 2005), that rethinking the role of 'diversion' in youth justice has ramifications well beyond the youth justice 'system' in so far as this is equated with the work carried out by youth offending teams. Varied local practices and the diversity of service structures within which YOTs now sit (Fielder et al., 2008), heralds considerable challenges for researchers attempting to assess what forms of 'system contact' help and hinder, or indeed how forms of 'prevention', 'early intervention' and 'diversion' are developing within any given local authority area in England, let alone the jurisdiction as a whole.

Diverse Meanings of Diversion
The drive to divert young people from aspects of formal criminal justice processes has a long history. The Children's Act 1908 established separate juvenile courts to divert young people from the adult court process in recognition of the different needs of child and adult offenders and the responsibility of the state to consider the welfare of young people coming before the courts (Arthur, 2010). A more radical conception of 'diversion' emerged in the United States in the mid-20th century in the context of new sociological theoriesnamely 'labelling theory' -which emphasized the stigmatizing consequences of applying deviant or criminal labels that cast deviants as 'outsiders' (Becker, 1963;Lemert, 1967). This was believed to promote self-identification with the applied label and forms of discrimination, which together serve to 'amplify' deviance. When applied to criminal justice practice, this way of understanding crime and criminalization logically promotes strategies that minimize labelling -in other words, non-intervention (Schur, 1973) or minimum possible intervention and the reduction of contact with formal criminal justice systems.
Critical criminologists and campaigning organizations in England and Wales have also emphasized that most youth crime is relatively minor and that the majority of children will 'grow out' of crime (for example, National Association for Youth Justice, 2013; Rutherford, 1986Rutherford, , 1992. This observation, bolstered by empirical research suggesting contact with formal criminal justice systems inhibits this process (Kemp et al., 2002;McAra and McVie, 2007), also lends support to noninterventionist approaches. The recent Independent Parliamentarians' Inquiry into the Operation and Effectiveness of the Youth Court headed by Lord Carlile found that there was 'wide support for high levels of diversion among inquiry respondents' (Carlile, 2014: ix), in part due to this body of evidence.
What this means in practice can be complex. Cressey and McDermott (1974: 3) discussed the difficulties in securing a single and clear definition of 'diversion' in their research on diversionary programmes in the United States. They identified several central problematic areas including: contested and contradictory definitions of diversion, definitions of 'system contact' and definitions of formal/informal provision. Similar concerns have recently been explored by Richards (2014: 122) who asks four interrelated questions to tease out these issues: … what young people are to be diverted from and to; whether young people are to be 'diverted' from the criminal justice system or offending; whether young people are to be 'diverted' from criminal justice processes or outcomes; and whether diversion should be considered distinct from crime prevention and early intervention.
We would suggest that at least five understandings of diversion can now be identified within academic and policy discourse: i) 'diversion from prosecution/court'; ii) 'diversion from custody'; iii) 'diversion from the youth justice system'; iv) 'diversion into alternative services'; v) 'diversion from crime'. Each of these meanings addresses Cressey and McDermott's (1974) and Richard's (2014) questions in different ways and has its own history in relation to youth justice in England and Wales. Some are also ambiguous: as we discuss below, the notion of diverting from the 'youth justice system', in particular, depends on where the boundaries of 'youth justice work' are drawn. Of particular relevance to current practice is the shift in political rhetoric and policy from 'diversion-fromprosecution', associated in England and Wales with the promotion of youth cautioning during the 1980s and early 1990s (Home Office, 1985, 1990, to 'diversion from crime', associated the more interventionist youth justice strategies identifiable from the mid-1990s (Audit Commission, 1996) and central to the policies introduced by the New Labour Government from 1997 (Home Office, 1997). We then go on to ask to what extent, and with what implications, this has since been reversed by practitioners.

From 'Diversion-from-Prosecution' to 'Diversion-from-Crime'
The principles of minimal or non-intervention were most visibly enacted in juvenile justice practice in England during the 1980s and early-1990s. This period saw a significant reduction in the numbers of young people convicted and imprisoned under an ostensibly punitive Conservative government publicly committed to the pursuit of 'law and order' (Allen, 1991). While acknowledging the role of demographic changes (the overall juvenile population fell by 9 per cent over this period), critical criminologists and practitioners (for example, Bateman, 2002;Goldson, 2000;Pitts, 2001a;Rutherford, 1986;1992) have tended to attribute these reductions to the dual practices of diversionfrom-prosecution, facilitated by official endorsements of informal police cautioning (Home Office, 1985;1990), and bifurcation (i.e. the tendency to impose the most serious sanctions only for the most serious offences, which in practice amounts to 'diversion-from-custody').
Although by 1990 a Home Office circular was able to claim 'widespread agreement that the courts should only be used as a last resort' (Home Office, 1990: 3), this consensus was relatively short-lived. The policy became subject to critique on the grounds of 'ineffectiveness', particularly in relation to practices of repeat cautioning and the use of less 'tough' disposals such as warnings, and 'inconsistencies' in terms of application (Audit Commission, 1996). Although disputed, such critiques became central to the reformulated (New) Labour Party's attempts to claim greater political authority on the issue of criminal justice during this period (Bateman, 2002;Jones, 2001). By the mid-1990s, the dominant direction of policy in England and Wales was therefore tending towards ever earlier intervention into young people's lives under the guise of 'diversion from crime'.
Following their election in 1997, the New Labour government radically and rapidly overhauled juvenile justice with the Crime and Disorder Act (CDA) 1998. Juvenile Justice Departments were replaced by Youth Offending Teams under the management of a new Youth Justice Board. Practice was closely regulated and structured by a new system of national standards, targets and key performance indicators, data recording, management and reporting requirements, standardized risk assessment documentation and procedures and nationally disseminated guidance and training on 'effective practice' (Muncie, 1999;Pitts, 2001b). So significant were these changes, and such was the depth of the new focus on standardization, that policy makers and youth justice scholars began to treat the CDA 1998 as the start of an entirely new era -the start of the 'new youth justice system' (Goldson, 2000).
Significant too was the emphasis on 'preventing offending', which the CDA 1998 introduced as a new 'principal aim' of the youth justice system. Based upon the correlation of 'risk factors' and (future) criminal behaviour, derived in turn from longitudinal studies such as the Cambridge Study of Delinquent Development (Farrington, 2000;West and Farrington, 1973), what became known as the 'Risk Factor Prevention Paradigm' (RFPP) provided theoretical support for the increasingly interventionist youth justice developed by New Labour. Here, prevention was understood to depend upon drawing young people into services in which risk factors could be addressed -diverting them from crime by addressing welfare needs, albeit within or close to the youth justice system.
Identifying young people 'at risk' and drawing them into voluntary services such as Youth Inclusion Projects (YIPs) 3 and Youth Inclusion and Support Panels (YISPs) 4 was presented as a way to prevent crime, but also, in so doing, prevent the criminalization of young people. Thus, 'diversion' became equated with 'diversion-from-crime', which in fact implied 'early intervention', whereas 'diversion from the system' meant 'system contact' of another form (system contact following offending). As Jack Straw, then Justice Secretary, suggested when introducing the Youth Crime Action Plan 2008 (HM Government, 2008): One aim of the new Action Plan is to divert young people away from crime, so that they are not unnecessarily drawn into the criminal justice system (cited in Home Office, 2008).
Clearly, this brief overview of policy change is schematic and reductive. We do not intend to imply that the history of diversionary practices in youth justice form a simple linear progression, nor that practice innovation is driven only by policy change. Indeed, as we go on to argue, such a position would offer little chance of accurately capturing current developments.

The Current Context: Diverse Practices of Diversion
There seems to be an emerging consensus that the dramatic reduction in FTE numbers with which we opened the paper has not been driven by an explicit redirection of national policy (see, for example, Bateman, 2014;Carlile, 2014;House of Commons, 2013;Smith 2014). Commentators point to concurrent local responses to revised performance indicators -particularly the Youth Crime Action Plan 2008 pledge to reduce the numbers of young people entering the youth justice system and the removal of the controversial police 'offences brought to justice' target -but also the strict budgetary constraints occasioned by the 2008 financial crisis and subsequent 'austerity measures'. As in the 1980s, the prospect of reducing costs at a time when local authority budgets have been significantly, although unevenly, reduced (Sheffield Political Economy Research Institute, 2014), and the removal of some performance management and reporting requirements promoting standardization (see Ministry of Justice and Youth Justice Board, 2013a; Youth Justice Board, 2011;and Bateman, 2014 for discussion) mean that there are clear financial incentives for local authorities to innovate to prevent young people being drawn into criminal justice services.
The Legal Aid, Sentencing and Punishment of Offenders (LASPO) Act 2012 made legislative provision for the increasingly flexible system proposed by the Green Paper 'Breaking the Cycle' (Ministry of Justice 2010) by replacing the 'two strike' Reprimand and Final Warning system with the more flexible Youth Caution and Youth Conditional Caution. Both statutory out-of-court disposals, Youth Cautions do not automatically involve intervention programmes (although the police must inform the YOT if given) whereas Youth Conditional Cautions do carry compliance conditions although these should only be punitive when rehabilitative and reparative conditions 'are not suitable or sufficient to address the offending' (Ministry of Justice and Youth Justice Board, 2013b: 8). Most significantly, in specific circumstances the new process allows for repeat cautioning, as was the case pre-1997, thereby aiming to prevent the unnecessary escalation of young people into the court system which characterized the Reprimand and Final Warning system. LASPO also clarified available police responses to young people's lawbreaking, including the kinds of informal non-statutory 'community resolutions' that many areas had already implemented (Bateman, 2014).
Ministry of Justice pilots have explored three overlapping models of these non-statutory responses: Triage (Home Office, 2012), Youth Justice Liaison and Diversion Schemes (Haines et al., 2012) and the Youth Restorative Disposal (Youth Justice Board, 2011). A recent survey of diversion schemes in 71 YOTs found that more than half had dedicated preventions-type teams or dedicated workers dealing with diversion (New Economics Foundation, 2014), most commonly Youth Justice Liaison and Diversion (YJLD) and Triage schemes. The YJLD schemes developed from the recommendations of the Bradley Report (Department of Health, 2009) and aim to respond, at the point of contact (usually a police custody suite), to mental health issues which are disproportionately experienced by young people in trouble with the law. They aim to 'divert' affected young people: either from the criminal justice system to health and social care services or to youth justice services 'better equipped to meet their health, emotional well-being and welfare needs' (Haines et al., 2012: 8). The government has proposed a national roll-out of YJLD schemes for children and adults from 2014 (Public Health England, 2014).
Triage services also place YOT staff in custody suites to work with the police and Crown Prosecution Service in determining the disposal most suited to the alleged offence, their history of service contact and assessed risk (Home Office, 2012). This can involve diverting minor cases (level one), offering supportive interventions for young people kept within the system, sometimes with the involvement of parents and carers (level two) and, in a few pilot areas, supporting but also fast-tracking young people identified as 'serious' and 'prolific' offenders (level three). The Youth Restorative Disposal (YRD), meanwhile, is aimed at providing first-time entrants with a 'supportive' intervention based on restorative justice principles. This entails a young person admitting guilt, taking responsibility for their actions and apologizing to the offended party. YRDs are a police-led intervention, and while the majority of YOTs receive YRD referrals from the police, an evaluation of YRD schemes found that YOT involvement varies across areas. The evaluation found that YOT responses were commonly preceded by a risk assessment, with levels and types of work depending on identified risks (Youth Justice Board, 2011). This 'work' might involve implementing a plan with the young person to address their behaviour and prevent future offending (Ministry of Justice, 2011), a process decided at a local level but which can include workshops with other young people, victims or families, reparative work, or 'offending behaviour programmes'. We explore these models in more detail in relation to our research findings below.

Methodology
We draw on findings from a study funded by the Economic and Social Research Council (ES/J009857/1 and ES/J009857/2) which explored the implementation of youth justice policy in England through an examination of practitioner sense-making. The research comprised a study of youth offending services at two case study sites ('Site A' and 'Site B'), with the primary data collection undertaken between December 2012 and September 2013. Briefly, Site A had a traditional Youth Offending Team made up of seconded probation officers and social workers. Site B had subsumed the YOT into a broader 'young people's service' (YPS) including other agencies such as youth work and Connexions. Practices of 'diversion' were not a focus of the original research but emerged as a key theme in interviews across the sites. There was no sampling in either location: where possible all youth justice professionals and their managers were interviewed. The data collection involved initial focus groups with practitioners at each site (n = 44), followed by the main body of data collection which comprised in-depth qualitative interviews with 71 practitioners (Site A = 31, Site B = 40) ranging from the Heads of Service at each site through to officers, workers and performance managers.

The New Disposal and New Intervention
In the context of increased local decision-making and in response to new national FTE targets each site had developed their own pre-statutory diversion scheme. The Site A scheme, referred to henceforth as the 'New Disposal' (ND), was developed by the YOT in conjunction with the local police. It was similar to a first step in Triage, a 'one chance' before the invocation of formal Caution or Youth Conditional Caution for young people who had come to the attention of the police for the first time and had made an admission of guilt. The ND involved an ASSET-style assessment focused on identifying the welfare needs of the young person (this also triggered a number of other assessments and risk management plans when deemed necessary). Following the initial assessment, the actual 'work' for the young person was likely to be a programme of activities to address 'offending behaviour' (in areas such as victim awareness, consequences of offending and work with families) and any other 'need' the assessment had highlighted. Whilst most NDs were completed, non-compliance with the ND would ultimately revert back to the police and the young person could instead be given a formal caution.
The disposal developed by Site B − the 'New Intervention' (NI) − was similarly a nonstatutory pre-caution measure and had also originally been developed in partnership with the local police. However, in practice the two schemes looked very different. The NI was more akin to the Youth Justice Board's 'Youth Restorative Disposal' model than Triage, and was considered in all cases where the young person admitted guilt for a non-indictable offence (not just for a first offence). All referrals came from the police but, following a decision-making panel made up of police and YPS representatives, NIs could be administered by the police or by YPS practitioners. While the Site A disposal was based largely on welfarist principles (i.e. the assumption that addressing welfare needs would reduce offending), the NI was based also on restorative justice values (i.e. the recognition of harm to the victim of the offence was considered as central to the success of an individual intervention). Most of the official discourse about the NI focused on the victim rather than the offender (Site B considered themselves highly restorative in general practice) and in theory the involvement of the victim had to be agreed for the intervention to go ahead, although at least one practitioner discussed a case where the victim did not want the NI but it was carried out nonetheless. Each NI was made up of an agreed contract with the young person regarding appropriate reparation for their offence, either to the victim or community. Where there was no identifiable victim (such as drug possession) the young person would be expected to undertake other activities to 'make amends' for their offending, for example speaking with parents about their offence. Many young people were also expected to undertake activities such as victim awareness or offending behaviour programmes in conjunction with their reparation or as part of the restorative process.
Whereas Site A's decision-making approach to diversion was offence-focused, Site B practice was flexible in deciding whether an NI was appropriate, taking a more offenderfocused approach in considering broader factors including the young person's offending history, their social circumstances and other factors. This meant that in practice the cohorts of young people holding NDs and NIs at the respective sites was very different -young people could hold more than one NI at a time (and/or at the same time as a formal order) if deemed appropriate, and could be not only first time offenders but also more serious offenders who may have already been in court or custody. Both sites undertook welfare and risk assessments at the outset of the diversion and where appropriate included a welfare component to the measure. Site A in particular, across both management and practitioners, widely believed that pre-court work including the New Disposal was a way in which to get young people and their families 'the help that they needed' (Site A Report), highlighting the blurring of the conceptual boundaries between 'diversion' and 'intervention'.

A 'Sea Change' in Practice?
Our findings illustrate that national policy encouraging diversion via the new out of court framework (Ministry of Justice and Youth Justice Board, 2013b) can be 'understood as formalizing changes that had already occurred organically' (Bateman, 2014: 419). Both New Interventions and New Disposals were presented, by managers and practitioners, as a necessary response to a challenging new economic climate and as an 'opportunity' to undertake more diversionary work which saved money compared to statutory measures. Managers at both sites viewed the diversion schemes as a 'new' way to deal with young people's offending: We've been phenomenally successful in diverting young people from the criminal justice system, and they're not coming back − they're not reoffending. (Site A) There has been a sea-change in the way young people are dealt with by the criminal justice system in [Site B] … diversion and restorative responses and not criminalising kids unnecessarily, and getting them out of the system rather than bringing them into the system. (Site B) The local practices developed at both sites were presented by managers as significant 'innovations', which had pre-empted the publication of the Out-of-Court Disposals National Framework (Ministry of Justice and Youth Justice Board, 2013b). A small number of practitioners were aware of similar innovations occurring elsewhere but most considered their local approach to be unique or knew little of other YOTs' working practices. Managers at Site B stated that the NI was developed for a number of reasons including: as a quick and effective response for the victims, to reduce costs and to lower FTEs (Site B Factsheet), but also within a broader context of preventing the criminalization of young people. Alternatively, Site A's rationale in setting up the ND was primarily described as FTE reduction and better outcomes for young people through early intervention (Site A Effective Practice Report). Although there were similarities and differences in these two diversion schemes occurring at various stages in the process (the officially stated rationale, the decision-makers and decision-making processes, thresholds for decisions, the number of interventions, the nature of intervention work, and the standardization of work within service), nonetheless both sites shared a 'new' commitment (made possible by budget restrictions and loosening of central government control) to 'diversion from prosecution' and 'diversion into alternative services' (the 1980s model). And yet they were also committed to 'diversion from crime' involving forms of work similar to the riskfocused 'early intervention' discussed above (the New Labour model). Whether and to what extent these services amounted to 'diversion from the youth justice system' was also open to question at both sites, as we discuss below. Together, we argue, this indicates that we were witnessing a distinct form of 'interventionist diversion' that differed from (at least some of) the models previously observable in England and Wales and from the forms of early intervention which had preceded it.

Diversion, but Not as We Know It?
As the previous summaries make clear, we observed diverse practices at both our research sites and consequently we became interested in understanding the extent to which these were novel practices. The New Intervention and New Disposal both clearly differed from the forms of early intervention targeting young people 'at risk of offending' which Youth Offending Services were encouraged to develop under the New Labour Government (Youth Justice Board, 2006), crucially because they require a young person to have broken the law and admit guilt. And yet, similar forms of pre-court 'intervention' were also in operation during the period in which informal forms of 'diversion-from-prosecution' were at their height in the 1980s. In 1987, approximately 50 per cent of police forces in England and Wales reported some form of 'caution-plus' scheme to Home Office researchers (Evans and Wilkinson, 1990), involving a formal caution accompanied by a type of intervention. This can be seen as a clear precursor of current interventionist diversionary approaches: it involved a supportive intervention, the involvement of multiagency panels in determining the use and nature of the disposal, and some complexity around compulsion in relation to the 'offer' (Evans and Ellis, 1997). Davis et al. (1989: 232) in their evaluation of the Northamptonshire Bureau also differentiated between the varieties 'diversion-from-prosecution' schemes on offer in the 1980s: … the Bureau offers 'traditional' or 'true' diversion (no further treatment, no service, no follow-up) rather than 'new' diversion, which requires the young person to undergo some kind of rehabilitative programme.
These studies suggest there are continuities in practice from juvenile justice practice pre-1998, both in models of 'diversion' that also involved an 'intervention' and in terms of a highly diverse and locally innovative practice context. Nonetheless, we would argue that what we observed is still something 'new' (see also Smith, 2014) due to the continued influence of New Labour reforms. We discuss three specific aspects of this influence below: 1) embedded managerialism and 'function creep'; 2) the continued influence of 'risk-thinking' and 'risk assessment'; and 3) practitioner commitment to (early) intervention as a preventative measure, i.e. the continued elision between 'diversion-from crime' and 'diversion from the system'.

Embedded managerialism and 'function creep'
The positions that practitioners in our research took about the appropriate balance between intervention and diversion were neither unanimous nor straightforward, but it is clear that many made a distinction between formal involvement in the youth justice system (and the associated criminal record) and informal alternative interventions. We observed, however, that 'informal', i.e. out-of-court, diversion schemes shared some features of 'formal' system contact. This included at one site recording the intervention on the Police National Computer, which could potentially result in Disclosure and Barring Service disclosure (see also Carlile, 2014), and also similarities in the actual intervention work carried out.
At one of our research sites, informal disposals could be granted after a formal disposal (which is permitted by section 138 of LASPO 2012). This meant in practice that young people could be required to complete work simultaneously on a statutory and nonstatutory basis, usually with the same case manager responsible for ensuring completion in both cases. Some youth justice practitioners, although generally not those with a background in voluntary participation services, told us they found it challenging to engage some young people without the threat of breach. Whilst they aimed to prevent the case going back to court, some tried to reproduce the more familiar power dynamic by emphasizing the contingency of the 'out-of-court' disposal: With the New Interventions actually it's not just a carrot, there is a stick as well. 'Do what you said you were going to do or you still go to court'. (Site B) We also found evidence that there appeared to be some 'sharing' of materials with statutorily ordered programmes and a continuity of practice across the different interventions, both formal and voluntary: I just treat it the same because in my world they would have been on a referral order probably years ago… it's the same intervention it's just called a different thing. I'm probably doing exactly what I would do if they were on a referral order but just without a panel meeting so I'd go and meet [them]…. I do my assessment and I go back to the office and write it all up and I discuss with him what I think would be a good way forward, we agree, he's going to do 4 sessions on X, Y or Z, he does those 4 sessions. (Site B) This transfer of practice seems unremarkable in professional contexts where the performance management techniques which pre-existed but were increasingly standardized by the CDA 1998 (Pitts, 2001b) shape the current working lives -and in some cases, the entire professional experience -of practitioners to such a significant degree. We found most practitioners spent a substantial proportion of their time completing standardized documentation or in front of a computer screen (reportedly averaging between 50 and 80 per cent of their working week at both sites) and most were unable to envisage a world in which practice could take place without this paperwork (see Phoenix, forthcoming for fuller discussion): People write too much which is why they spend too long in front of computers.
[…] They feel safer, if we've written it all down then if you've written down every bit of information you feel like you haven't missed anything and that no one will come and criticise you for not doing that but in the process you haven't necessarily weighed what is important information. (Site B) We found that, despite the removal of some national requirements, our studied services were still recording and reporting on many aspects of practice for local management teams and other political actors within local authorities. Some practitioners felt that the pressure to adhere to managerialist targets and 'tick boxes' had increased rather than decreased in recent years: The world is now is that … from a managing point of view, there is a great pressure to ensure that there's [a] tick box element to it. I'm continuously as a manager sending emails saying 'I need this information, have you got this information?', there's a lot of duplication. (Site B) I think the paperwork has to be done and that's an element of the job that I don't always agree with. I think sometimes we do things for the sake of doing it rather than for it to be of purpose, you know for it to have ticked a box. (Site A) We suggest that this embedding of managerialist techniques of performance management have caused elements of 'function creep', which challenge claims of an absolute sea change in practice. We loosely borrow this term from surveillance studies concerned with the spread of technologies designed for use in one domain, to other not originally intended functions (Winner, 1977). Some practitioners felt it was beneficial to adapt existing, already 'approved' ways of working with statutory cases to the new non statutory context. We suggest that this blurs distinctions between 'formal' and 'informal' system contact, while retaining the distinction between criminalization (i.e. a criminal record) and its avoidance which was at the centre of practitioner support for the new 'diversionary' interventions.
The continued influence of 'risk-thinking' and 'risk assessment' Similarities in 'formal and 'informal' work were similarly evident in the use of risk-based assessments which draw upon the formal ASSET framework for these new forms of prestatutory work.
For any other case apart from a statutory court case we have a one sheet assessment. Basically it's a shortened form of an ASSET…but instead of the 12 ASSET sections you've got two or three of the sections put into different categories so it's shorter. (Site B) This is perhaps unsurprising given the extent to which 'risk thinking' has become embedded in youth justice work (for example, Baker, 2012;Case and Haines, 2009;Kemshall, 2008). We found some evidence too of defensive practice cultures, especially within our integrated service, as youth justice practitioners responded to a shrinking workload and fears of de-professionalization (see also Coyles, 2014).
We've got practitioners… many of whom are greatly experienced who are now doing things far less often than they used to do… writing pre-sentence reports for example. If you're writing one 3 or 4 times a month it's probably keeping your skill level up. If you're only doing it once every 2 or 3 months I think things are going to be forgotten and the kind of sharpness of them is lost and quality deteriorates. People are less content than they were and I think we will, over time, lose the expertise that we have because either they're not practising the skills that they've got or because they have gone off to other Local Authorities who can offer them more of what they want to do… that's the downside of it. (Site B) It seemed that the categories and scoring of 'risk' underpinning ASSET had not only become important aspects of the reasoning process, but proficient risk-based assessment was now part of the professional expertise which some youth justice workers were keen to defend.

Practitioner commitment to (early) intervention as a preventative measure
Assessments of the new practices were mixed, but at both sites, we found that many practitioners were concerned that the New Disposal and New Intervention were being used inconsistently or disproportionally. Some were in favour of heavier and earlier intervention: I think, if I compare how we work now to how we worked in the YISP when it was …'at risk of offending', reprimand, final warning. If a young person came through on a reprimand they didn't even get referred to the Youth Offending Service, until they received a final warning …[with a] reprimand they could go to the police station, sign this, don't get into trouble no more. No intervention, walk away and nine times out of ten we're going to see that young person back. Whereas now every young person that goes to the police station and is arrested and charged for an offence is referred to the Youth Offending Service. So that very start, that early stage, they've been interviewed, they've had the photos took, they're referred to the youth, it's the whole impact of it all. (Site A) I know they're meant to be soft touch which is the point of them but there's not a lot of change that you can implement within that short period of time unless you are very very good or very very lucky I think. (Site B) Echoing the welfarist justifications of 'early intervention' (e.g. France and Utting, 2005), practitioners were concerned about the erosion of broader universal provision and felt they were supporting, not punishing those with whom they worked, particularly in light of the removal of the dedicated funding for YOT prevention work. I mean some of the families we're getting on the New Disposal have been crying out for services and they just haven't been delivered and this is the first time they're getting them from ourselves. Now I would love a world where they did get them delivered and they did have all that and they didn't need to go into criminal justice services and kids were kept out of it, but it doesn't exist, and whilst we've got the situation we've got I'm happy for us to do it in the right way. (Site A) The new approaches to 'diversion' that we observed were therefore both 'interventionist', since they draw young people into structured programmes believed to bring behavioural change, the nature of which is shaped by the assessment of risk, and 'new', since they share features with both pre-1998 'diversion' and post-1998 'early intervention'. We have shown how practitioners identified benefits associated with both 'diversion' (especially the avoidance of criminalization for minor or first time offences) and early intervention (especially the chance to offer welfare-focused support). Although we have insufficient space to analyse them here, the concerns raised by practitioners were similarly hybrid. Despite these similarities, the 'New Disposal' and 'New Intervention' were very different to one another, particularly in respect of repeated use and use with young people previously sentenced at court (see also Smith, 2014 for an extended comparison of current services). The teams we studied also sat within very different surrounding service structures, which meant 'pre-court' work was delivered by a dedicated team at one site, but by workers also dealing with young people formally involved with the service at the second. Both these observations -the emergence of a 'new interventionist diversion' and the very different forms this is taking in different local authority areas -raise important but very challenging questions about what helps or hinders in terms of 'system contact', which we reflect on further below.

Conclusion and Future Research Directions
Newly developed diversionary approaches were represented as a 'sea change' in practice by some participants in our study. This article makes a case for a more complex reading, which reflects multiple understandings of diversion and its relation to other youth justice priorities, particularly the prevention of crime. We have suggested that the models of pre-court work we observed do not represent an absolute break from the 'new' youth justice established by the CDA 1998, since the elision between 'diversion from the youth justice system' and 'diversion from crime' that characterized New Labour's youth justice continues to shape some aspects of this work. Indeed, our participants described an 'interventionist diversion', where young people were 'diverted from prosecution' and 'into alternative services' which, though voluntary, closely resembled other formal interventions. Similarly, while some aspects of this work clearly echo older forms of 'diversion' − the embedding of the 'risk factor prevention paradigm' and the exportation of existing practice challenge easy comparisons with the past -it is not possible to understand current trends as simply the 'rebirth' of the 1980s 'orthodoxy' in youth justice practice (Bateman, 2012).
Our research suggests that, while criminal justice agencies are responding to the same national drivers of revised performance indicators and budgetary constraints following the financial crisis of 2008 and the introduction of so-called 'austerity' measures, the removal of many of the mechanisms for standardized youth justice practice by the Coalition Government and broader diversification in local authority service structures (e.g. Fielder et al., 2008) have brought considerable local variation. Recent policy developments in this area seem unlikely to reverse these practice trends. Sections 133-8 of LASPO specify the conditions for out-of-court disposals, but there is little detail about intervention requirements. Guidance recently issued on the new pre-court sentencing framework (Ministry of Justice and Youth Justice Board, 2013b) is non-statutory, and again offers very little that is prescriptive in relation to the substance of intervention, particularly in terms of the pre-caution or 'community resolution' work discussed above. This diversity of practice has far-reaching consequences for youth justice scholarship. In this area at least, the assumption that policy represents practice, while always reductive (Fergusson, 2007), is now unsound, while the assumption that findings can be easily transferred even between local authority areas is questionable.
As noted at the outset, we support earlier research (for example, Cressey and McDermott, 1974) and, more recently, Richards (2014) in highlighting the need to distinguish between 'diverse diversions'. We are uncertain, however, that the conceptual disambiguation Richards (2014) recommends will significantly advance older debates about the appropriate balance between 'intervention' and 'diversion' in youth justice without clear understanding of the specific relevant services in any given context. Unpicking the conditions under which the lessons of the past can help understand the present is, we would argue, first an empirical task. Outcomes -both in terms of FTE statistics and long-term outcomes for young people -will depend not only on the specific 'diversionary' practices adopted in any given area but also, at least for vulnerable young people, on the broader network of support services that sit outside the youth justice system and the connections between them. If researchers in England are able to exploit the current diversity of service provision to arrive at more nuanced definitions of 'system contact', as well as how this is experienced by young people in trouble with the law, we may gain clearer understandings of when it helps or hinders. A focus, therefore, on youth justice processes rather than outcomes may provide transferable if not representative research findings, help unpick the extent to which existing assumptions about 'diversion' held by practitioners, criminologists and policy makers apply to the current youth justice landscape, and forge a new terrain of debate.