Raising the minimum age of criminal responsibility Lessons from the Scottish experience

This paper explores the history of the reforms in Scotland which led up to the raising of the minimum age of criminal responsibility (MACR) in 2019, from age 8 to 12. It also showcases research evidence, from the Edinburgh Study of Youth Transitions and Crime, which indicates that there is both an ethical and empirical case for raising the MACR further, to at least 15. As we will argue, the Scottish experience demonstrates the dangers for governments of deploying punitive narratives around youth crime as a means of building political capacity. While progressive policy transformations, aimed at avoiding the criminalisation of children, have been hard won, fears that such practices are ‘ soft options ’ that will drive up youth o ﬀ ending are unfounded. Edinburgh Study ﬁ ndings demonstrate that the lives of children involved in serious o ﬀ ending are blighted by poverty and early trauma: factors beyond their capacity to control. They also demonstrate that diversionary and holistic approaches to dealing with children ’ s needs are associated with reductions in o ﬀ ending. Consequently, whilst the Scottish experience should give con ﬁ dence to policy-makers about the positive impact that raising the MACR brings, we conclude that this is only a minimum necessary step in delivering justice for children.


Introduction
In 2019, the Scottish Parliament legislated to raise the minimum age of criminal responsibility (MACR) from age 8 (one of the lowest ages globally) to age 12.To many this change may appear somewhat modest, not least because the UN Committee on the Rights of the Child (UNCRC) has recommended a minimum acceptable standard of age 14.However, change in Scotland was many years in the making, the culmination of a series of policy reforms for dealing with children who come into conflict with the law, which began around 2006/07 at the end of a short-lived punitive phase in Scottish youth justice.In this paper, we trace the history and impact of juvenile justice reform in Scotland from the inception of its unique Children's Hearing System in the 1960s to the passing of the Age of Criminal Responsibility (Scotland) Act in 2019.Focusing particularly on the last two decades, we show how competing policy narratives around childhood and youth offending served to shape and direct very different responses to the problem of youth crime in Scotland with markedly divergent results.
To contextualise this discussion, we highlight the influential role of evidence from the Edinburgh Study of Youth Transitions and Crime (the Edinburgh Study), a prospective longitudinal programme of research tracking the lives of 4300 young people. 1 Cited extensively in policy documentation, Edinburgh Study findings gave policy makers confidence that reform of the MACR could be effective in tackling youth offending.The research evidence highlighted an ethical imperative for further reform, as it showed that young people involved in the most serious and persistent offending are amongst the most victimised, vulnerable, and traumatised groups in society and are not responsible for the many structural deficits (including poverty) which suffuse their lives.The evidence also highlighted an empirical imperative for reform, as it showed that diversionary practices and more holistic approaches to dealing with children's needs are associated with reductions in offending and shrinking youth justice populations.
Drawing on both the history of reform and our research findings, we point to three principal lessons from the Scottish experience that should influence policy-makers and those campaigning to raise the MACR in other jurisdictions: 1.The politics of policy transformation mean that governments exhibit an abundance of caution when attempting to implement more progressive modes of policy, which can hinder the potential for robust research to influence the policy making process.In other words, reform is likely to be slow-paced and campaigners need to be persistent in making the case for change.2. Policy phases that focus on punishing or 'responsibilising' persistent and serious young offenders serve to draw more young people into the justice system, which results in detrimental outcomes; whereas, phases of policy that actively seek to avoid the criminalisation of children are associated with better individual outcomes.
In other words, claims that diversionary practices are 'soft options' that will drive up crime are unfounded.3.There is compelling evidence that the MACR should be raised to at least age 15 and this should be regarded as a minimum necessary step to delivering justice for children in conflict with the law.Indeed, once the ethical and empirical bases for reform of the MACR are acknowledged, it follows that a 'whole system approach' should be adopted in which cross-cutting policy portfolios (including, but not restricted to, youth justice) focus their attention on the underlying needs of children.Further, efforts should be made to address the structural factors which shape in negative ways the contexts in which they grow up.In sum, the Scottish case highlights the need for a praxis: 1 The Edinburgh Study of Youth Transitions has been funded by grants from the Nuffield Foundation, The Scottish Government and the Economic and Social Research Council (ESRC grant numbers R000237157 and R000239150 fusing the ethical and empirical imperatives for policy transformation to enable all children and young people to flourish. The paper begins with an overview of the Scottish system of juvenile justice, highlighting three very different phases of policy reform and the impact these had on young people (lessons one and two).We then set out key research evidence from the Edinburgh Study that influenced the most recent phase of Scottish policy in which the MACR was raised from age 8 to 12 (lesson three), but which shows strongly that the MACR should be raised even further.The paper concludes with a review of the implications of the Scottish case for other jurisdictions.
Part 1: the Scottish system of juvenile justice-timeline, contexts and policy transformation In this section, we trace the history of juvenile justice in Scotland and examine how transformations in dominant conceptions of 'youth offending' became caught up in governmental efforts to build a sense of political identity and assert a right to rule.As we will show, the rehabilitative ethos which had characterised the Scottish system of juvenile justice since 1971 suffered major disruption during a short-lived punitive phase of policy from around 2002 to 2005.During this period, the incumbent government in Scotland placed emphasis on a populist narrative around the 'problem' of persistent offenders, ignored the research evidence on the aetiology of offending and the impacts of youth justice interventions, and introduced policies which led to a dramatic rise in the rate of youth justice referrals.In contrast, the more compassionate phase of policy which followed this period (from 2006 onwards) was overtly evidence-led and has resulted in historically low youth justice referrals and a dramatic reduction in young people being dealt with by court and prison services.Despite such significant success, however, the journey to reforming the MACR has been incremental and, arguably, over-cautious.

Origins of the Scottish system of juvenile justice
The modern system of juvenile justice in Scotland was established following a major inquiry set up in the 1960s to review legal responses to 'juvenile delinquents and juveniles in need of care and protection or beyond parental control' (Kilbrandon, 1964, para 5).Named after the chair of the inquiry, the Kilbrandon Committee concluded that offending amongst young people was caused by a failure in their upbringing, typically due to shortcomings in their home, family, or school environments.The Committee's recommendations emphasised the importance of avoiding criminalisation, of dealing with troubled and troublesome children in the same way (with deeds being symptomatic of deeper seated needs), and focusing on an educational model of care.Almost all of its recommendations were enacted through the Social Work (Scotland) Act 1968, chief amongst them the creation of a new system of juvenile justice known as the Children's Hearing System.Predicated on the so-called 'Kilbrandon philosophy', this system represented a major divergence from juvenile justice systems in the rest of the UK, which continued to operate a youth court model, despite the fact that legislation in England and Wales (the Children and Young Person's Act 1969) contained similar welfarist principles to those in Scotland (McAra, 2017). 2mplemented in 1971, the Children's Hearing System accepted referrals for any child or young person in need of care and protection from birth to age 16, and for those involved in offending behaviour from age 8 to age 15 (although the lower age threshold has now increased to 12 following the change in the MACR).Referrals for those aged 16 or 17 were only accepted where the young person was already subject to a statutory order under the Children's Hearings System.Referrals were reviewed by an official known as the Children's Reporter, who would determine whether or not there was a prima facie case that at least one of a number of statutory grounds for referral had been established and that the young person was in need of compulsory measures of care.If both of these criteria were met, the case was referred to a lay tribunal of specially trained members of the public who would determine the appropriate disposal (which mostly involved a home supervision order, but could extend to a period in secure care) (McAra & McVie, 2019).
While the so-called 'Kilbrandon philosophy' underpinning the Children's Hearing System was arguably the epitome of a rehabilitative model of justice, some aspects of the Scottish system remained anomolous.This included the retention of a very low age of criminal responsibility (age 8 was originally set in 1932).The Kilbrandon Report had recommended the repeal of any rule or statutory provision establishing a minimum age of criminal responsibility, stating that the MACR bore no resemblance to any observable phenomena concerning child development (Kilbrandon, 1964).However, this was one of the very few Committee recommendations not subsequently taken up by government.A further anomaly was the routine use of adult criminal courts to deal with 16 and 17 year olds charged with committing an offence, as well as those between age 8 and 15 who were accused of serious offences (such as murder or sexual crimes).
A shift in policy narrative: from rehabilitation to punishment The system of juvenile justice in Scotland remained largely unchanged, with some modernisation but no major transformation, for 50 years (Donnelly, 2020).However, the Children's Hearing System came under scrutiny following a referendum in which the public voted for greater powers to be transferred from the UK government to Scotland.A new Scottish Parliament was established in 1999, led by Scotland's first devolved government (a coalition between the Labour and Liberal Democrat parties).Very quickly, the 'problem' of youth crime became heavily politicised in Scotland-a means of building political capacity for a nascent government (McAra, 2017).In doing so, the coalition government drew inspiration from the 'New Labour' Government at Westminster, which had taken office in 1998 with a manifesto pledge to be 'tough on crime, tough on the causes of crime' (Labour Party, 1997).
Aligning with the New Labour ideology, Scotland's political narrative began to focus on responsibilising 'persistent young offenders'.An Advisory Group on Youth Crime was established to examine 'the scope for improving the range and availability of options aimed at addressing the actions' of such offenders (Scottish Executive, 2000).The Group acknowledged that the approach to addressing youth offending should balance the care and protection of young people against the concerns of the public; nevertheless, it emphasised the importance of a strategic multi-agency approach and the need for a range of accredited programmes and interventions aimed at supporting young people involved in persistent offending, which were broadly in line with the Kilbrandon philosphy.The appointment of a new First Minister in 2001 heralded a tougher stance on youth offending, however, and a distinctly punitive agenda emerged which espoused a 'fairly narrow emphasis on individual responsibility and personal (and parental) accountability for the behaviour of young people, with a practice focus on correcting personal (and parental) deficits' (McNeill & Batchelor, 2004, pp. 13-14).A whole series of policy reports were dedicated to the topic, including a 'Ten point action plan on youth crime' (Scottish Executive, 2002), which set out a range of new measures to tackle the problem of persistent offending, including the re-introduction of youth courts.Moreover, the Antisocial Behaviour etc (Scotland) Act 2004 introduced a raft of legislative orders, replicating those available in England/Wales, such as Dispersal Orders3 , Parenting Orders and Anti-Social Behaviour Orders for 12-15 year olds.The then Cabinet Secretary for Justice in Scotland proclaimed that 'punishment was an essential part of the youth justice process' (Scottish Executive, 2003)-a statement that would have been unthinkable in the previous decade, and in complete contradiction to the Kilbrandon philosophy (McAra, 2017).
Core to the Scottish Executive's new measures was a 'Fast Track Hearings' pilot, which would specifically target persistent offenders up to age 16 who committed 5 offences within a 6 month period.Remaining under the jurisdiction and overall ethos of the Children's Hearing System, the Fast Track Hearings had four broad aims: to speed up decision-making processes; conduct more comprehensive assessments, including around offending risk; ensure referral to specialist services; and reduce rates of reoffending (see Hill et al., 2005).Such was the confidence in the success of the Fast Track Hearings that performance targets were set to reduce the number of persistent young offenders by 10 per cent over a three year period, from a precise starting point of 1201 (Scottish Executive, 2002).
A characteristic feature of this punitive policy phase was a reluctance on the part of Scotland's coalition government to engage with academic research which demonstrated that the new arrangements were likely to fail.Early findings from the Edinburgh Study (published later in McAra & McVie, 2007) showed that the Children's Hearing System already had a propensity to draw in and label the most deprived and vulnerable children and young people, which increased their risk of subsequent criminalisation.However, when confronted with this evidence and faced with the proposition that the Fast Track approach was likely to amplify, rather than reduce, youth crime, the Cabinet Secretary for Justice reportedly said: 'I don't care' (Personal Communication with Lesley McAra).
Established in six local authority areas in early 2003, the Fast Track pilot scheme halted abruptly in 2006 following publication of a damning evaluation which revealed that reductions in offending observed in the pilot sites were lower than those in the comparison (non-Fast Track) sites (Hill et al., 2005).In addition, many young people had only met the criterion for fast tracking because of very minor offences or offences which had been committed in residential care (86% of those living in residential establishment were found to have such an offence, Bradshaw, 2005).Rather than reducing the number of persistent offenders by 10 per cent, the new policies actually increased it by 15% (Audit Scotland, 2007).And, as shown in Figure 1, the policy changes were accompanied by overall rises in the rates of offence referrals to the Children's Reporter (by 50%), criminal convictions for 16 and 17 year olds (by 19%), and direct receptions to youth custody (by 10%).
In addition to the demise of the Fast Track pilots, the Scottish Executive had to face the fact that most of its other new measures had also failed.Very few areas in Scotland were designated dispersal zones (see above) (Crawford & Lister, 2007), and there was limited evidence that they reduced crime or anti-social behaviour in communities (McMillan & Robertson, 2012).Furthermore, few Anti-Social Behaviour Orders (ASBOs), and no Parenting Orders, were ever issued (McAra, 2010).Indeed, despite the First Minister of Scotland taking a delegation of police officers and local authority representatives to Manchester to encourage the use of ASBOs (Flint & Nixon, 2006), there was never an appetite for such punitive measures amongst Scottish practitioners.Ironically, the new legislation had helped to promote better multi-agency working at local levels when considering the needs of young people proposed for an ASBO, and 'the prevailing ethos was that ASBOs should really only be used when they have greater positive impact than other support that can be offered' (Manders, 2010)  From the ashes of the Scottish Executive's failed punitive policies rose a new policy framework-Getting it Right for Every Child-which re-emphasised the need for early and effective intervention, using more inclusive and less punitive language than other aspects of policy.Whilst Getting it Right For Every Child had its origins in the early years of devolution, in the context of developing more effective joint working practices between agencies (Lightowler et al., 2014), anecdotal evidence suggests it was propelled to prominence by senior civil servants (rather than politicians) who took advantage of the emerging policy hiatus (Personal Communication with Susan McVie).This watershed moment, with a shift back to a child-focused, rather than offender-focused, narrative was drawn on by the incoming Scottish National Party (SNP) government (from 2007 onwards) and became the first in a series of staging posts towards raising the MACR.

The compassionate years: 2007 onwards
Drawing on a more progressive ideology, the SNP government quickly strived to put clear water between it and the previous Scottish Labour/Liberal Democratic coalition administration (as well as the prevailing Labour Government at Westminster).Taking the heat out of political debates about offending, the language around persistent offenders was immediately toned down, with a more inclusive discourse linked to a new National Performance Framework (Scottish Government, 2007).This Framework set out 11 core outcomes, of which one highlighted the revitalised commitment to the needs of children and young people: 'we grow up safe, loved and protected so that we realise our full potential'. 4The incoming Cabinet Secretary for Justice placed particular emphasis on a more 'compassionate' approach to those who come into conflict with the Law (both adults and children) and tied this to a sense of Scottish cultural identity.As he noted in his autobiography: 'Compassion and mercy are about upholding the beliefs that we seek to live by, remaining true to our values as a people' (McAskill, 2016).
Importantly, the new administration proved itself to be a listening government, and utilised evidence from research, including findings from the Edinburgh Study, to develop and implement a series of policy reforms over the following decade.Thus, in addition to embracing the Getting it Right for Every Child principles, the early years of the new SNP government saw a greater commitment to diverting young people involved in offending away from formal measures of intervention.In 2008, funding was made available to support the piloting of a major diversionary initiative in one Scottish Local Authority and, following a successful evaluation, the 'Whole System Approach' (as it became known) was rolled out to every Local Authority in Scotland from 2011 and reaffirmed in subsequent Youth Justice Strategies published in 2015 and 2021.Drawing on findings from the Edinburgh Study (McAra & McVie, 2007), the aim of the Whole System Approach was to invest in services and support (including youth work, educational outreach and restorative practices) to enable Early and Effective Intervention and diversion away from formal systems of justice.A key ambition was to avoid the unnecessary criminalisation of children whilst at the same addressing their needs: a re-embracing of the Kilbrandon philosophy.
Widespread commitment to multi-agency working and diversionary practice across all local authorities created the political conditions to support raising the age of prosecution in Scotland from 8 to 12 via the Management of Offenders (Scotland) Act, 2010.Whilst this fell short of raising the age of criminal responsibility (which remained at age 8), symbolically it was de-facto decriminalisation of those under the age of 12 who could no longer be prosecuted for offences in the criminal courts.Although those between the ages of 8 and 12 could still be referred to the Children's Hearing System on offence grounds, changes introduced by the subsequent Children's Hearing (Scotland) Act 2011 meant that any admitted offences would count as alternatives to prosecution rather than as a criminal conviction (as previously5 ).In practice, so few children under the age of 12 were ever prosecuted in the criminal courts, that these changes seemed uncontroversial. 6Nevertheless, it still took a further nine years-during which time a Scottish Government Advisory Group was established to examine the evidence for raising the MACR and there was significant campaigning from researchers, the Children's Commissioner and key third sector organisations amongst others-for the Government to bring the age of criminal responsibility into line with the age of criminal prosecution-via the Age of Criminal Responsibility (Scotland) Act 2019.
A key staging post in the campaign for change in the MACR was the publication of the Kilbrandon Again report in 2018. 7This report was commissioned by the Chief Executive Officer of Action for Children and the Children's Commissioner as a result of discussions between them and Lesley McAra.The Edinburgh Study formed the baseline evidence for the Report and, amongst its wide ranging conclusions, it recommended that the MACR be raised to a minimum of 16.Such momentum for change was also supported by official crime trend statistics (see Figure 2) which showed a strong association between the abandoning of punitive and fast-tracking practices (at the end of the Labour/Liberal Democratic coalition years) and the widespread subsequent adoption of Early and Effective Intervention, diversionary practices, and raising the age of prosecution (over the SNP government years).The data show dramatic reductions in offence referrals (by 89% from their peak in 2005/06), criminal convictions for older children (by 93% since their peak in 2006/2007) and receptions to custody (reducing by 98% over the same time frame and reaching their lowest level for at least 50 years). 8mportantly, the development of the Age of Criminal Responsibility (Scotland) Bill was directly influenced by evidence from the Edinburgh Study.It was the only piece of academic research cited in the policy memorandum accompanying the passage of the Bill, which stated: 'We know that responding to childhood behaviour in a criminalising, stigmatising manner serves only to promote escalation and further harm.Scotland has proven approaches to confronting and correcting this childhood behaviour that do not need a criminal justice response' (Paragraph 6).However, many involved in the campaigning for an increase in the MACR were deeply disappointed that the Scottish Government did not raise it further than age 12.In particular, Scotland's Children and Young People's Commissioner registered his concern that the Act was in direct contradiction to the government's progressive commitment to international human rights (Adamson, 2019).Correspondingly, the Edinburgh Study findings suggest that there is a strong ethical and empirical imperative to raise it still further, an argument that we discuss further in part 2 of the paper.

Final reflections: two lessons from policy transformation in Scotland
To conclude this section, what then are the lessons that can be learnt from this short history of Scottish policy reform for other jurisdictions that may be considering reform of the MACR?The first lesson is that reforming the MACR may be a prolonged process that involves many incremental stages, so patience and resilience are requireduntil it is fully accomplished, the case for change has to be continually made.Even in a country with a strong history of welfarist values, like Scotland, the issue of youth crime can be politicised when parties deem it to be to their electoral advantage and where governments feel the need to assert their right to rule and a sense of political identity.The performative role that youth crime comes to play in such contexts renders progressive policy transformation somewhat slow and unpredictable as governments strive to balance their moral sensibilities about children and young people who come into conflict with the law with an over-weaning drive to build political capacity.A strong sense of caution, driven by populist punitive rhetoric, can negate the impact of even the most robust research that goes against the prevailing narrative.The second lesson is that when governments adopt more punitive approaches that seek to responsibilise persistent and serious young offenders, this widens the net of young people who will be drawn into justice systems, whereas more compassionate approaches are associated with far lower youth justice populations.The Edinburgh Study has demonstrated the preventative effect of keeping young people out of formal systems, which flies in the face of populist notions that diversion is a 'soft option' that drives up youth crime.
Having set out the history, context and impacts of reform to the MACR in Scotland so far, we turn to the evidence base supporting a higher age.
Part 2-making the ethical and empirical case for further reform on the MACR in Scotland While raising the MACR in Scotland from 8 to 12 was a significant achievement in both political and practical terms, within months of the Age of Criminal Responsibility (Scotland) Act 2019 receiving Royal Assent in June 2019, and well before it was enacted in December 2021, it no longer met the minimum international standard.In September 2019, General Comment 249 from the UN Committee on the Rights of the Child raised the accepted minimum age of criminal responsibility from 12 to 14, but recommended that States set an even higher age of 15 or 16.Fortunately, the Scottish legislation stipulated that a review be carried out within three years of enactment 'with a view to considering the future age of criminal responsibility' (2019 Act, Part 6: Section 78), and an Advisory Group10 has been established to consider the case for raising the MACR further in Scotland.
In this section, we consider research findings from the Edinburgh Study that support a further increase in the MACR to at least age 15.11There are three key sets of findings from over 25 years of fieldwork that support such a change.First, the relationship between 'needs' and 'deeds' does not stop at age 12 but continues into, and intensifies during, the early teenage years, so raising the MACR would retain a critical focus on needs for longer with, arguably, better outcomes.Second, most of those who come into contact with juvenile justice for the first time are older than 12, so the criminogenic effects of system contact which our research has demonstrated, disproportionately impact on those who will not be protected by the MACR.And, third, minimal intervention during the early teenage years is likely to support the natural process of desistance from offending, whereas formal system contact acts to precipitate, rather than diminish, the development of criminal careers.Each of these arguments is developed further below.
The relationship between needs and deeds intensifies in the teenage years with longer term effects in adulthood The Policy Memorandum accompanying the ACR (Scotland) Bill as it passed through the Scottish Parliament, setting out the rationale for increasing the MACR to age 12, stated: 'We know that a disproportionate number of the under-12s currently dealt with for offending concerns have faced significant prior disadvantage and multiple adversity in early childhood, and that better outcomes will flow from attending to those, rather than focussing on children's deeds in isolation' (ACR (Scotland) Bill 2018 Policy Memorandum: Para 5).This statement drew directly on findings from the Edinburgh Study which demonstrated that young people's deeds were symptomatic of deeper-seated needs, driven by factors that were rarely under the control of young people themselves.
At age 13, which is only just above the current MACR in Scotland, we found that engagement in violent behaviour was associated with a wide range of co-existing vulnerabilities and adversities (McAra & McVie, 2016).Modelling the independent effects of these 12 , Table 1 shows that living in a low socio-economic status household and growing up in a neighbourhood characterised by high levels of deprivation were both significant predictors of violent behaviour.This was true even when a range of other potential risk factors were controlled for, including violence in earlier childhood, early experience of violent victimisation, family conflict, substance misuse, school detachment and truancy, and an impulsive personality.As shown in many other studies (see Moffitt & Caspi, 2001;D'Unger et al., 2002), we found that being a boy was a particularly strong predictor of violence at age 13; although, when examining interaction effects, the relationship between poverty and violence was disproportionately elevated amongst girls.
The findings in Table 1 highlight that, to be successful, strategies aimed at reducing engagement in violence by age 13 need to address a multitude of underlying problems and vulnerabilities experienced during childhood.The increase in the MACR to age 12 in Scotland recognised that this would be facilitated by keeping them out of the juvenile justice system.However, the Edinburgh Study also demonstrates that this complex relationship between needs and deeds endures and intensifies into early adolescence.Indeed, at age 15, we found that violent offenders were significantly more likely than non-violent youths to be: victims of crime and harassment; involved in self-harm and para-suicidal behaviour; engaged in risky health behaviours; exhibiting mental health conditions, including depression and disordered eating; participating in under-age sexual encounters; and living in a dysfunctional family (McAra & McVie, 2007).Living in poverty was a distinguishing feature in the lives of teenagers who engaged in violence, with girls continuing to be at particular risk (McAra & McVie, 2015).And there was a reciprocal relationship between violence to others and violence against the self (through self-harming injuries) for both sexes at age 15 (McAra & McVie, 2010).
The level of trauma and adversity experienced by those who engage in violence in their early to mid teenage years is painfully evident years later.Table 2 presents new findings 12 These findings were first published in a special edition (on violence) of the Journal of Criminal Justice (McAra & McVie, 2016) and are the product of multivariate analysis.We regressed a binary measure of violence at age 13 on a variety of risk and protection factors, while controlling for gender and two measures of poverty.'Enter' was used as a method for insertion of variables into the modelling (placing all covariates into the regression in one block), as we consider this to be the most robust in terms of theory building and testing.Further details of the analytical approach can be found in the article.
from survey data collected in our most recent phase of fieldwork (at age 35), which show that those involved in the highest levels of violence at age 15 were significantly more likely than those with no such history to be heavily victimised in adulthood, to have a moderate to severe clinical level of depression and to have suffered 'adverse experiences' as an adult (as measured by relationship breakdown, losing a job, having a serious accident or illness, being interfered with sexually or having someone close to them die through homicide or suicide).
In-depth interviews held with a sub-sample of cohort members at age 35 reinforce the findings of the survey, highlighting how family interpersonal violence in childhoodbeyond the control of the young person-often formed the context of expressive violence at that age, followed in the adult years with ongoing mental health issues-including selfharm, anxiety and depression (see McAra & McVie, 2022).As two interviewees commented: I was like in first year, second year, third year … I was involved in a sexual relationship with a much older person and I didn't understand what was going on … There was a person at home who was very, uhm, creepy … you just had to let him letch over you … I grew up in [an area of Edinburgh] and it was a bit mixed … quite a lot of drugs and theft and people  (1) Poorest, Highest: binary measures created by specifying more than one standard deviation from the mean in the direction of risk as 1 and the remainder specified as zero.
being on benefits and dealing drugs and things … … at 17 I had a manic episode and a psychotic episode … At the moment [age 35] it's more anxiety and depression and alcoholism.
(Interviewee 371, involved in highest level of violence at age 12).
The first time I ever sort of properly retaliated was in the high school … I think I gave a young lad some serious facial injuries.I think stuff must have been leading up to that, cause I remember going into school and saying nobody better mess with me today … I'm learning to deal with anxiety and depression a bit better.I used to just pick stuff up and throw stuff or be quite loud.And then in 2019 I got a diagnosis of borderline personality disorder.(Interviewee 307, with history of family violence) In demonstrating this strong and consistent link between deeds and needs, we reiterate the argument of the 1964 Kilbrandon Committee that offending is often symptomatic of a broad spectrum of emotional and situational vulnerabilities experienced by some young people.The fact that such a strong relationship between violence and vulnerability is apparent in childhood but also endures into adolescence with longer term impacts into adulthood, raises serious questions about setting the MACR at such a young age as 12.There is strong evidence that brain development does not reach full maturity until into the mid-20s (Steinberg, 2007); and that adversities stemming from developmental immaturity are exacerbated by other social and environmental risk factors which constrain young people's ability to act freely and maturely (Farmer, 2011).Moreover, adolescence is a period of particular risk for traumatic brain injury, especially amongst boys, which is also associated with offending behaviour (Williams et al., 2010), and raises further questions about the culpability of those in pre-and earlyadolescence.Such tensions between the notion of 'competency' and 'capacity' have been highlighted in debates about the age of criminal responsibility, with arguments that age 14 or 15 are the minimum at which competence may be proved by a prosecutor (Delmage, 2013).Our Edinburgh Study findings strongly suggest that offending is the mechanism by which many young people cope with the negative consequences of both their neurological state and the wider social and environmental conditions within which they live.Therefore, considerations about the minimum age at which a young person should be held criminally responsible must take appropriate account of such complex interactions.

The criminogenic effects of juvenile justice system contact are most acute in early adolescence
As highlighted in part 1 of the paper, the risk of causing further damage to children who offend was a particularly salient reason for increasing the MACR from 8 to 12 in Scotland.Again, this decision was explicitly based on findings from the Edinburgh Study which confirmed that early contact with formal systems of social control had the potential to both increase the risk of future offending behaviour and stimulate further justice system contact (McAra & McVie, 2007).However, by focusing on the negative impacts of system contact up to age 12, the Bill failed to take account of the much greater risk of labelling effects during adolescence.
A key issue that should be taken into consideration in setting the MACR is the extent to which justice system contact impacts on the subsequent behaviour of young people, especially with respect to those who are pulled furthest into its reaches and are subject to the most intensive forms of intervention.Thanks to the multiple sources of data collected by the Edinburgh Study, we were able to examine the characteristics of those young people who experienced differential levels of system contact.We looked at three stages of youth justice contact to establish which young people progressed furthest into the system at age 15, what factors appeared to influence decision making around selection, and what impact this had on their behaviour a year later (McAra & McVie, 2007).At stage one, we found that, while police officers were more likely to charge the most serious and risky offenders, those who had received police charges at an earlier age were over seven times more likely to be charged again at age 15.At stage two, we established that young people who had been charged by the police and then referred to the Children's Reporter on offence grounds were four times more likely to receive another such a referral at age 15.And at stage three, we demonstrated that young offenders who were referred to the Reporter were three times more likely to be brought to a Children's Hearing at age 15 if they had already been referred to a Hearing at an earlier age.In other words, a history of system contact was a significant pre-cursor to system contact at age 15.More importantly, we demonstrated that for each additional stage that young people progressed into the youth justice system at age 15, the less likely it was that they showed signs of desisting from offending at age 16.
These findings highlight that contact with youth justice organisations involves a complex filtering process which serves to propel certain young people deeper into formal systems of intervention.Unfortunately, those who were most likely to be subject to such filtering-commonly referred to as 'the usual suspects'-tended to have been exposed to the greatest levels of physical, social and economic adversity (McAra & McVie, 2007, 2010).And the consequences of filtering such vulnerable young people into the justice system were being felt years later, since the Edinburgh Study found that those who exhibited the most adversity during childhood and adolescence were most likely to continue offending beyond age 18, which was also associated with further adversity in adulthood (McAra & McVie, 2022).In addition, justice system transitions from adolescence to adulthood were related to a wide range of behavioural, familial, contextual and experiential factors, but those who persistently came into contact with the justice system over the life course tended to be the poorest and most vulnerable children in the study cohort.
Looking at cohort members at age 35, it is clear that adult criminal careers were undergirded by processes of early labelling and criminalisation, with early and intensive formal system contact (especially care experience) being strongly associated with later justice system contact (McAra & McVie, 2022).Testimonies from cohort members in adulthood support the argument that being subject to formal interventions and labelled a 'troublemaker' in adolescence results in long term damage across a range of outcomes (including mental and physical health, educational achievement, labour market success and, of course, criminal careers).Outcomes for care-experienced young people were especially negative, their lives blighted by poverty, mental health and drug problems, relationship breakdowns, and unemployment.
While not asked to comment specifically on the issue of the MACR, cohort members offered negative reflections on their experiences of formal system contact during childhood and adolescence.Justice system involvement was typically viewed as a set of hurdles to be negotiated, while memories of the care system often exposed ongoing feelings of vulnerability and trauma.Recalling their time growing up in care, one interviewee noted: Obviously if you're put in foster care [or] sent to a home you know there's like a massive lump of rejection there … I've always felt like … I'm the problem or, or you know everything is going wrong … and then having a cycle of negative thoughts (Interviewee 995).
When agencies did make a positive difference in people's lives, it was generally due to the efforts of one particular worker or carer with whom a strong and loving relationship had been built.For example, another interviewee stated: My foster carers … played quite a big part in our lives.If it wasnae for them, I don't know how I would've turned out to be honest (Interviewee 210).
Similarly, reflections on school experiences highlighted the negative effects of bullying, teacher labelling and undiagnosed learning needs, which often led to bad behaviour and truancy-actions which, as we've described above, increased the risk of youth justice contact.
I just got into trouble at school a lot.There's one teacher and she told me that I was attention seeking … I was like, you know what, I will show you the difference and I would totally play up.I would totally wind her up to the point where she couldn't handle the class.And it was because in my hearing … she said I was attention seeking.(Interviewee 415) So they were bullying me basically, and then I got hold of one of them by themselves.And then yeah, maybe I went a bit overboard and I might have broken his arm … I wasn't the happiest of kids to be honest.School wasn't great for me, like, I was quite dyslexic.So they didn't pick that up at all at school, like sort of they messed that whole up.I started skiving school, back then I was smoking cannabis … I was just not the happiest of kids to be honest.(Interviewee 899).
With particular relevance to the MACR, we have argued that 'youngsters are powerless to alter the majority of the factors that propel them further and further into the system at age 15 (including family structure, social deprivation, gender, and being known to the police and Reporter in earlier years)' (McAra & McVie, 2007, p. 338).Our most recent findings from cohort members in adulthood support this assertion, and demonstrate the damage and misery that can result from criminalising people during their teenage years.We contend that raising the MACR and providing a welfare-based and traumainformed response to those who offend during early adolescence would considerably reduce the long term damage caused by justice system contact.It is ethically incumbent on governments to ensure that legal protections are in place to safeguard young people from the type of adverse system effects we have described here.Raising the age of MACR in Scotland to 15 would be a major step in the right direction.

Minimal intervention in the teenage years supports desistance from offending
Concluding their enquiry into the provisions for dealing with juvenile offenders in Scotland, the 1964 Kilbrandon Committee noted that: 'Society is, we believe, seriously concerned to secure a more effective and discriminating machinery for intervention for the avoidance and reduction of juvenile delinquency'.In setting out their recommendations for such machinery, they were clear that young people should be dealt with, as far as possible, away from formal systems of justice.When raising the MACR in Scotland, the Policy Memorandum accompanying the Bill acknowledged that: 'It is clear … that eight to 11 year olds who offend are mainly vulnerable for other reasons-the 'criminal' label does not help the system to deal with them effectively and it is neither the test for intervention nor the focus of any work with these children'.We agree wholeheartedly with this sentiment, but argue that it could be equally be applied to those beyond the age of 11.
To support our argument we draw on criminal convictions data for the Edinburgh Study cohort, which span from age 8 (the MACR in Scotland at that time of the research) to age 21 (for details of the analysis conducted, see McAra & McVie, 2010).As would be expected based on population data, the majority of cohort members (85%) had received no convictions over this period.Focusing only on those who had received at least one conviction 13 , we modelled the data to identify typical group-based 'trajectories' or pathways, based on the probability of being convicted at each year of age.We identified three trajectory groups which were distinct from each other in terms of age of onset and probability of conviction.The shape of each trajectory is illustrated in Figure 3.
The vast majority (89.8%) of Edinburgh Study cohort members with a conviction were in a group that we labelled 'later onset decliners'.For members of this group, likelihood of having a conviction started at around age 15, peaked at age 19 (at which point they had around a 35% probability of being convicted), and declined steadily thereafter.The trajectory followed by this group broadly reflects the age distribution of convictions observed in official statistics at that time (Scottish Government, 2011), which suggests they were typical of the wider population.This finding in itself suggests that a MACR of age 15 in Scotland would be sufficient to deal with the bulk of those who seriously contravene the law.However, we also identified two small groups with a much earlier onset of conviction, starting in childhood.Less than one in twenty of the convicted cohort members clustered into a group that we labelled 'early onset desisters'.This group had a remarkably similar trajectory and probability of conviction to that of the later onset decliners; however, onset of conviction was far earlier, starting at around age 10, and the peak age of conviction was around age 15.The remaining six per cent of convicted cohort members were in a group that we labelled 'early onset chronic' as it showed a steep rise in the probability of conviction from around age 11 to 17 before levelling off and peaking at around age 19 (at which point individuals in this group had around an 80% chance of being convicted).These two very small groups are not evident in official population statistics, but they represent an extremely important minority of offenders that have a direct bearing on the setting of the MACR.
Using the Edinburgh Study's extensive longitudinal dataset, we examined the background characteristics of these two small groups of individuals to answer two questions: why did they have a much earlier age of onset of criminal conviction compared to the majority; and why did their trajectories and probability of conviction differ from one another?
With respect to the first question we found that the two early onset groups had shared certain characteristics with the later onset decliners in childhood.For example, they had similarly difficult family backgrounds, in that they were just as unlikely to be living with at least one birth parent, and equally likely to report poor parental monitoring and high levels of conflict with parents.However, the two early onset groups were more problematic in terms of their behaviour by age 12 than the later onset decliners.They were significantly more likely to have engaged in serious offending and their average volume of offences was almost twice as high as the later onset decliners.They were more likely to report hanging around in public places, taking drugs and associating with other offenders, and to exhibit a lower level of moral concern about engaging in offending.In addition, they were more likely to be exhibiting problems at school, including bad behaviour in class and truancy, resulting in a higher likelihood of exclusion.However, the early onset groups were also more economically and socially vulnerable than the later onset decliners, as measured by low socio-economic status of the household, entitlement to free school meals and growing up in a more highly deprived area.And they were more likely to have experienced agency intervention at an early age, in the form of adversarial police contact, police warnings and charges, offence referrals to youth justice, and statutory supervision.Overall, therefore, the Edinburgh Study provides clear evidence that those who started to be convicted at an earlier age were as much a cause for concern in terms of their needs as their deeds and, as a result, the addition of a 'criminal label' was hardly in their best interests.
To answer the second question, we examined what changed over the period from age 12 to age 15 and found this to be a significant turning point in terms of explaining the divergence between these two early onset conviction trajectories.Those in the early onset chronic group were significantly more likely to have experienced a more intensive period of agency scrutiny, characterised by an increase in school truancy and exclusion, adversarial police contact, police warnings or charges, offence referrals to the Children's Reporter, and periods of statutory supervision.Importantly, however, there was no difference between the two groups in terms of change in the prevalence or frequency of their self-reported involvement in serious offending over this period.Overall, therefore, these findings from the Edinburgh Study cohort suggest that formal agency contact, rather than offending behaviour, during early adolescence played the defining role in increasing the probability of conviction amongst those who were convicted at an early age (McAra & McVie, 2010).
Final reflections: the lesson from the Edinburgh study on the need for further reform Taken together, the findings from the Edinburgh Study provide strong ethical and empirical imperatives for raising the MACR in Scotland further, to at least age 15.However, raising the MACR on its own will not deliver justice for children in conflict with the law unless wider policy transformations are there to undergird and support it.Young people who become involved in offending, including violence, are highly vulnerable and unable to control many of the factors (especially structural problems, such as poverty) which could push them towards a longer-term criminal career.Contacts with systems appears to exacerbate, rather than diminish, the propensity to offend, and diversionary practices are associated with desisting patterns of offending, which provides strong empirical support for raising the MACR, thus avoiding unnecessary criminalisation of many young people.The impacts of early trauma and system contact continue well beyond the teenage years, with outcomes at age 35 being significantly poorer than for those without such histories.In acknowledging the ethical and empirical bases for increasing the MACR, the rationale for a 'whole system approach', which cuts across a range of policy portfolios (e.g., justice, education, health, economic, etc) and focuses on the underlying needs of children, becomes clear.However, this will require adequate and continued investment in services and supports for children and young people away from formal systems of justice, and a continued political commitment to a child-centric, welfare-based philosophy.

Conclusion
In this paper we have set out the history of juvenile justice in Scotland, described the key policy changes which led to an increase in the MACR to age 12, and considered the compelling evidence for a further increase to at least age 15.In doing so, we have highlighted three inter-connected lessons from the Scottish case which may be of value to other jurisdictions considering changing the MACR.The success of raising the MACR after decades of being the lowest in Europe has been momentous in many ways, but Scotland still lags behind many other countries and contravenes UNCRC recommendations on the minimum accepted standard.Nevertheless, we have argued in this paper that raising the MACR is only the minimum necessary step to achieve justice for children and young people who come into conflict with the law, and there is cause for optimism that Scotland's wide policy frameworks are sufficiently child-centred to support a further increase.
Our first lesson from the Scottish experience is that policy making around the MACR is somewhat fragile and subject to the vagaries of electoral cycles and efforts to build political capacity on the part of new governments.When governments consider it to their electoral advantage to politicise youth crime, there is a tendency to embrace a sharper, more punitive, policy narrative and ignore even the most robust research evidence.Even in a country with a long history of welfare-based principles and child-centred practices, politicians can succumb to a populist punitive rhetoric around the problem of persistent youth offending to assert their political dominance.Governments are far more cautious about shifting towards progressive reforms that some may view as 'soft on crime'.Those who wish to persuade policy makers that such approaches are more likely to be both effective and economically prudent must, however, be patient and persistent.
Our second lesson is that punitive policies that aim to punish or responsibilise young people who engage in offending are misplaced and counter-productive.As the Scottish experience highlights, punitive policies tend to increase the number, and widen the net, of those who are drawn into the juvenile justice system-a problem that is even more severe when the MACR is so low.Whereas, policy reforms that seek to divert young people away from the justice system, and instead focus on their underlying wellbeing and developmental needs, avoid criminalisation and increase the likelihood of more positive individual outcomes.Since introducing the Getting it Right for Every Child framework and adopting a 'whole system approach' to deal with young people who get involved in offending, Scotland has seen a dramatic drop in young people entering the Children's Hearing System, adult courts, and custodial establishments.And, as predicted by the Edinburgh Study, this has been accompanied by no increase in youth crime (in fact, quite the opposite).This alone, we believe, should give confidence to policy-makers in other jurisdictions contemplating MACR reform.
Our third lesson is that raising the MACR is an important, but minimum necessary, step on the way to delivering justice for children and young people who offend.Drawing on a body of robust evidence from the Edinburgh Study, we have made both the ethical and the empirical case for raising the MACR further, to at least 15.The data clearly show that the young people who get involved in the most persistent and serious forms of offending are the most vulnerable, victimised and traumatised in society.Punitive regimes ensure that these are the critical factors that determine which young people get caught up and recycled by justice organisations, and such contact serves to perpetuate further vulnerabilities and adversities well into adulthood.It is irrational, therefore, to increase the MACR without taking into account the wider context of youth crime, the cultural practices of institutions which make up the juvenile justice system, and the ethos of all policies that impact on children and young people (including economic policies focused on eradicating poverty).It is only by taking a holistic approach to the needs of such children (such as the Getting it Right for Every Child framework) and providing a wider set of services beyond youth justice which avoid stigmatising and labelling (as provided by the whole systems approach) that reductions in youth offending and justice system contact can be sustained over time.
As the Edinburgh Study evidence has shown, only by embracing both the ethical and empirical imperatives for change, can justice for children be delivered.All of us-as academics, civil servants or youth justice practitioners-have a responsibility to bring the research evidence to bear on policy debates and continue to make the case for further reform.

Disclosure statement
No potential conflict of interest was reported by the author(s).

Figure 1 .
Figure 1.Trends in offence referrals, criminal convictions and receptions to custody over the punitive phase.Sources of data: Scottish Children's Reporter Administration; Criminal Proceedings in the Scottish Courts (Scottish Government); Scottish Prison Statistics and Office for National Statistics Mid-Year Population Estimates.

Figure 2 .
Figure 2. Trends in offence referrals, criminal convictions and receptions to custody over the compassionate phase.Sources of data: Scottish Children's Reporter Administration; Criminal Proceedings in the Scottish Courts (Scottish Government); Scottish Prison Statistics and Office for National Statistics Mid-Year Population Estimates.

Table 1 .
(McAra & McVie, 2016)ted involvement in violence at age 13(McAra & McVie, 2016).Victimisation includes: threats to hurt you; actually hurt you by hitting, kicking or punching; actually hurt you with a weapon; stole something of yours; used threat or force to steal or try to steal something from you.(3) Wave 1: self-report at age 12-'ever involved in violence', 'ever a victim'.(4) Most, Highest, Lowest: binary measures created by specifying more than one standard deviation from the mean in the direction of risk as 1 and the remainder specified as zero.

Table 2 .
Outcomes at age 35 from survey data for those involved in highest levels of violence at age 15 compared with the rest of the cohort