In a recent article in the British Journal of Criminology (2015, 55/5: 883-900), William Wood sets out the reasons “Why Restorative Justice Will Not Reduce Incarceration”. His analysis is perceptive and, to my mind, largely persuasive, even if it has the potential to dampen the growing political confidence in the potential of restorative justice to improve crime control.

For those of us who believe in the power of restorative justice to promote healing and change in participants and to strengthen social bonds, Wood’s treatment is an important reminder of the need never to lose sight of how structurally embedded the modern prison system is, of the inherent limits of any intervention aimed primarily at delinquent individuals to bring about systemic change, and of the dangers of rhetorical excess in promoting the benefits of restorative justice.

At the same time, there is still good reason to invest in expanding restorative justice practices, both within the criminal justice system and beyond it, as an important part of the solution to mass incarceration, and in ways that are not always immediately obvious.

Three Levels of Impact
Wood begins by distinguishing three levels at which RJ is said to be effective. At the micro-level, it strives to repair the harm caused to victims and to hold their offenders accountable. At the meso-level, it facilitates the involvement of the wider community in responding to crime and reducing its occurrence. At the macro-level, it seeks to transform criminal justice practices and policies by advocating a new way of dealing with the problem of crime.

Most empirical research has focused on assessing the effectiveness of micro-level practices, such as family group conferencing and victim-offender mediation. These practices are said to deliver a strong sense of procedural justice, to enable the affected parties to take greater ownership of their conflicts and needs, to bring about a deeper sense of satisfaction to victims, to reduce recidivism in offenders, and to enhance the reintegration of wrongdoers into the community.

But in addition to these micro-level claims, it is not uncommon for RJ proponents to claim that restorative practices also have the potential to transform to criminal justice system as a whole, at the macro-level, and to substantially reduce the use of incarceration. But is this the case?

The Modern Prison Boom
​Wood focuses on four countries that have each experienced galloping growth in prison populations over the past 40 years. Between 1980 and 2010, the rate of incarceration in Australia increased by 100%, in the United Kingdom by 83%, in New Zealand by 183%, and in the USA by a staggering 500%. To imagine that a phenomenon of such massive proportions can be substantially changed through the use of RJ conferencing is not only lacking in empirical evidence, Wood claims, it is patently naïve.

It is lacking in empirical evidence because there is a dearth of empirical research on the impact of RJ on incarceration. Numerous studies and meta-analyses have demonstrated the impact on recidivism in general, including two by our own Ministry of Justice. But reduced rates of reoffending do not translate automatically into reduced rates of imprisonment, for several reasons.

One reason is that the majority of cases referred to RJ, especially in the youth jurisdiction, are for offences that would not normally carry a prison sentence anyway, such as vandalism, theft, burglary, minor assaults and traffic violations. Only a handful of studies have focused specifically on the impact of RJ on those who would likely face imprisonment for their crimes, and these studies have come up with mixed results.

There are also very few examples in the world where RJ is available as an alternative to prison. Certainly in its earliest days, victim-offender mediation was envisaged as an alternative track to the mainstream punitive system. “Since this time however”, Wood explains, “RJ has developed not as an alternative to state practices, but as an extension of them, so that today few restorative programmes exist as alternatives to the criminal justice system” (8).

A possible exception is the NZ Youth Justice system, which uses Family Group Conferencing in place of formal prosecution and punishment for young offenders. Following the introduction of new system in 1989, there was a marked reduction in juvenile detention. But this laudable outcome cannot be attributed simply to the inherently restorative power of FGCs. It was the result of a conscious legislative intent to use mandatory diversionary mechanisms to severely limit custodial placements.

As well as being lacking in empirical evidence, the claim that RJ has the potential to solve the prison problem is also sociologically and politically naïve. It rests on the assumption that our currently high rate of imprisonment is due to increased levels of criminal offending, and that by reducing reoffending rates through restorative interventions, the prison population will eventually fall.

But reality is not that simple. The recent growth in incarceration has not come as a result of real increases in crime, but because of changes in sentencing practice that have criminalized less serous offences, extended sentence lengths, increased the use of remand, tightened bail and parole requirements and increased the practice of recalling released persons to prison for technical violations. These harsher policies are in turn part of a larger pattern of more punitive and exclusionary practices of social control that extends through multiple policy areas, including social welfare, mental health, employment law and education. No amount of RJ conferencing, in and of itself, will be able to undo the deleterious impact of such policy settings.

It is probably no accident that the growth of incarceration has followed hard on the heels of neo-liberal economic reforms involving the deregulation of labour markets, reductions in social welfare provisions, reduced worker protection, regressive taxation, growing inequalities of wealth and income, and relatively punitive policies towards the poor and minority populations. These structural inequalities and the discursive practices that have justified them, such as the so-called “war on drugs”, are the major contributing factors to the burgeoning prison population of recent decades.

If recent trends towards more punitive sentencing are not the result of increased criminal offending, they cannot be reversed simply by working more restoratively with individual criminals. Micro-level practices, like RJ, cannot readily fix the macro-level or structural problems that have given rise to the prison boom, and it is naïve to think otherwise. Only if it were able to reduce poverty, social inequality and marginalization would RJ be able to significantly affect rates of incarceration.

So Where To For RJ?
Wood’s analysis is a helpful warning against sloppy talk and sloppy thinking on the part of RJ advocates. In promoting the benefits of restorative justice to politicians and policymakers, and to the general public, it is all too easy to commend it on the grounds that it reduces reoffending, and will therefore reduce the prison population, and in so doing save the government money. The evidence that it can help to reduce reoffending is encouraging. But there is no automatic correlation between offending rates and rates of imprisonment, between the drivers of crime and drivers of incarceration. There are deep systemic and ideological reasons for what has been called “the expanding prison”, and these cannot be undone solely by increased access to restorative conferencing.

This is not to say that RJ can make no difference whatsoever to the problem. There is still good reason for RJ advocates to work hard at extending the reach and deepening the quality of restorative practices, as one part of a much larger campaign to transform the wider system and, indeed, society itself.

The bulk of RJ work around the world is still for relatively minor offending, especially by young people. There is nothing wrong with this, but if RJ is to play even a small role in reducing incarceration it needs to be used more regularly for imprisonable offences. As Wood concludes, RJ will continue to have a negligible impact on rates of imprisonment “unless RJ practices can be more systematically implemented into cases of serious offending and developed more broadly as a sentencing alternative for offenders who otherwise may face incarceration” (p.12, cf. 8).

At the same time, much more effort should be made to expand the use of restorative practices within prisons and especially as part of the process of reintegration following release. New Zealand is still doing very little in this space, certainly compared to the pre-sentence arena. More work is needed on the potential of restorative programmes and practices to reduce the number of people who return to prison a second or third time because they fail to reintegrate successfully into law-abiding society.

Another way in which RJ can play some role in reducing incarceration is by interrupting the so-called “school-to-prison pipeline”. There is a direct correlation between failure at school and entanglement in the criminal justice system and the extent to which restorative practices in schools can help to keep kids engaged in education is the extent to which they will help to reduce the prison population. It is also important to recognize that quantitative research has its limits. Not everything that matters can be measured statistically. One of the most important roles that RJ can play is in cultivating the values and social commitments needed to move society toward greater civility, equality, inclusiveness and compassion.

The highly retributive and exclusionary Corrections system we have created over the past generation is, amongst other things, the symptom of a deep shift in our social and economic values, marked by a growing individualism and diminishing sense of our interconnectedness and interdependence as citizens and as human beings. RJ promotes on a different value system, one focused on whanaungatanga, and those who participate in its practice often discover or rediscover in themselves and in others something that is profoundly humanizing and unifying.

It is a mistake, then, to limit the role of RJ in redressing the prison problem solely in terms of its ability to keep more offenders out of jail. For every offender who engages in a RJ conference there will often be half a dozen or more other participants involved in the process, whether as victims, supporters or community representatives. They witness what takes place. The impact their attitudes, values, actions and relationships subsequently helps to remove or reduce the fears, hostilities and rivalries on which prison monster has grown fat.

Researchers have often found that victims who participate in RJ typically experience a reduction in feelings of vengefulness and fearfulness, the very emotions that have been manipulated by the advocates of tougher sentencing. Who can say what long term shifts in social climate will flow from increasing numbers of people engaging in restorative processes?

Wood is correct. Micro-level practices cannot, by themselves, effect macro-level change. But they can still play an important role in creating the conditions necessary for macro-level policy changes to be contemplated and implemented.

When David Seymour rose to speak at ACT’s 2017 Conference, no one (including myself) expected him to propose a prison policy, aimed at the early release of prisoners.
From what I can garner, ACT’s policy was directed at a significant expansion of the literacy and numeracy programme in prison, on the basis of the following principles:

• Offenders who complete numeracy and literacy courses in prison, or passed their driving licence test, would have the length of their sentences reduced by up to six weeks a year, with a cap of 18 week’s reduction on a three year sentence – an 11.5% reduction;

• Well-educated prisoners who mentored other prisoners would also be eligible for a similar reduction in their sentence;

• ACT would make it easier for volunteers to gain approval to carry out this kind of work in prisons;

• Decisions on eligibility for the course, would be made by the Parole Board;

• The policy would be part of any coalition arrangement with National.

Just as exciting was the thinking behind the proposal.  Prisoners needed positive incentives to become productive, law abiding citizens, David Seymour said.  Many lacked the skills required to lead normal, productive lives.  They needed to take responsibility for their lives, and this policy would provide  that incentive.  What is more, suitable volunteers and well-educated prisoners would run these programmes at a reasonable cost.
It took me back to 1992-93, when the Prisons Division of the Department of Justice increased literacy programmes four-fold, and trusted and suitably qualified prisoners tutored other prisoners in computer skills, te reo Māori, literacy and numeracy. Some of the external tutors were former gang members.  A time when the emphasis was not on managing risk, but encouraging prisoners to take charge of their own lives, with the support of whānau and family.  I was also reminded of the efforts of Prison Fellowship efforts in 2003/4, when formal approaches to the Department of Corrections to introduce qualified volunteers into the prisons to teach literacy and numeracy was roundly rejected, as was a proposal to expand the role of volunteers in the prison as mentors, teachers, across a range of skills based programmes, and recreational activities.
In one fell swoop, David Seymour has put those issues firmly back on the agenda; and this time with the support of the Prime Minister.  How did that happen?  Easy – David Seymour and Bill English witnessed a transformational event, an occasion when all the empirical evidence, the ‘what works’ literature, the scepticism, is swept aside as they listen to the emotive, spirit-filled testimony of an offender whose life has been transformed by an intervention, a programme or a person – the korero of someone who genuinely knows that his or her life will never be the same again.
The Magical Moment of Transformation 
Every Corrections service provider worth their salt, longs for the moment when an offender or prisoner, stands before fellow prisoners, staff, whānau or family, speaks from the heart, and in a moment of transcendence, talks about  how their lives have been changed, re-defined, re-directed, through the efforts of their transformer.  It can take many shapes and forms.  The prisoner or offender who has taken part in a restorative justice conference, and whose expression of remorse has been fully accepted by a victim, allowing both of them to move forward on their healing journey.  The drug dependent prisoner who successfully completes a drug treatment programme; knowing that while they may relapse in the future, they are on a journey toward recovery.  The violent gang member, who decides that he doesn’t want his mokopuna to live the life he lived; and starts doing something about it.  The culturally-deprived Māori prisoner, whose experience of learning te reo and tikanga in a Māori Focus Unit, gives them a sense of genuine identity and self-worth.  Those whose adoption of a belief system provides a set of values and principles for living a pro-social and meaningful existence in the future.   And of course, those who learn to read and write, and feel confident  about finding their way into employment.
When that moment is witnessed by outsiders (a policy analyst, a prison officer, a philanthropist, or best of all, a member of parliament), it becomes even more special.  The magic of the moment challenges their own perceptions, causes them to rethink policy, and understand how transformation occurs – although Garth McVicar continues to contend that no prisoner is capable of remorse.    I recall that in about 2005 Prison Fellowship ran a ‘Sycamore Tree’ programme at Waikeria Prison (Parker, 2016), and the Mayor of Hamilton attended the graduation.  He was so impacted by the witness of prisoners to the effectiveness of a programme funded through charity, that the Hamilton City Council funded the next two.  A number of MP’s have similarly been impacted by restorative justice in prisons, (terminated by Corrections in 2009) (Workman, 2016).  Prison Unit Managers who witness that ‘moment of magic’, will often work to ensure the longevity of a programme, sometimes in the face of Head Office scepticism.  Like it or not, they become partners in the offender’s journey of desistance.
The ultimate protection against an early demise of a promising intervention, is political support.  When Mike Williams, CEO of the Howard League of Penal Reform, Auckland, invited ACT member David Seymour, and the Prime Minister, the stage was set for a policy coup.  Both Bill English and David Seymour report being ‘moved’, after listening to the testimony of prisoners. When David Seymour learnt from Mike Williams that there was not much demand for the literacy programmes, he set about to design a policy to create that demand.  The outcome may well change the shape of the way we think about prisoners.
What is the Evidence for the Effectiveness of Literacy and Numeracy Programmes, in Reducing Reoffending?
Prior to 2000, prisoner education programmes were regarded as ‘time fillers’, as there was no evidence that they reduced reoffending.  Since then, the evidence for their effectiveness has grown.  One of the best summaries is found in a 2009 Department of Corrections own publication, ‘What Works Now – a review and update of research evidence relevant to offender rehabilitation practices within the Department of Corrections which concluded that, ‘Good outcomes are being widely achieved through educational and employment training; provision of education and employment in conjunction with other forms of correctional rehabilitation is likely to bring about the best results.’ (p.4).
The logic is inescapable:

• offenders who attend corrections-based education have higher rates of employment;
• stable employment is generally associated with lower rates of reoffending;
• offenders who participate in corrections-based education programmes have higher (post-release) earnings than non-participants (Ibid: 28-9).

Despite the evidence for their success, the department decided in 2009 that that ‘while prisoners would continue to have access to a range of educational and employment activities and opportunities, relative to the major cognitive-behavioural treatment (CBT) interventions, these would be treated as a lower priority in sentence planning, especially for higher-risk offenders’ (Ibid: 26).
Seymour’s policy changes that – there will now be an increased focus on interventions for lower risk offenders.  In time, it may force the Department to think more broadly about the range of positive interventions that are available outside of CBT.
Addressing Inequality and Justice
What is there not to like about David Seymour’s policy?  The most obvious shortcoming is that it is too narrowly drawn.  If one of the goals is to promote prisoner self-improvement , then limiting the incentive of an early release to literacy and numeracy programmes, promotes inequality of opportunity – something at odds with Act’s own stated political principles.  It needs to be extended to the successful completion of other interventions and programmes; such as drug treatment and violence reduction programmes, active engagement in a restorative justice conference, becoming a fluent speak of te reo and demonstrated understanding of tikanga Māori, acquisition of other employment-based  skills , and so on.  Some prisoners will be limited in their ability to respond to such opportunities.  The US states take a wider approach, and provide ‘good time’ early release  for prisoners who respond positively to the prison environment, and ‘work time’ early release for prisoners who demonstrate a strong work ethic (Lawrence, 2009).
There is no rational basis for excluding all sex offenders from an earned incentive scheme, upon successful completion of a sex offender’s treatment programme.  This structural stigmatisation of sex offenders does not align with the evidence; the reoffending rate of treated sex offenders is extremely low. The reconviction data estimate is that the overall rate of sexual reoffending is 14% over a 5 year period (Harris and Hanson, 2004).
Criminal justice providers shudder at the idea of the Parole Board being involved in determining eligibility for earned privilege.  First, it would create a bureaucratic nightmare involving a lot of red tape – something that David Seymour wants to avoid.  Second, the Parole Board is extremely conservative, cautious and risk averse.   There are progressive and enlightened members within the Board, but the prevailing environment is one of distrust, and negativity toward prisoners.  The Parole Board’s approval of parole for prisoners declined from 31% of all parole applications in 2011 to 23% in 2016 (Johnson, 2017:25). Enough said.
Will it Reduce Reoffending?
David Seymour puts a convincing case for the effectiveness of this initiative, citing US earned incentive programs, such as the New York Corrections Department, which saw a 20% lower recidivism rate among prisoners who earned early-release.  A proportionate saving for New Zealand’s population would be $113 million for Corrections, (adjusted for inflation, NZD and population).
The key issue is whether released prisoners reoffend less because of their newly acquired skills, or because they have spent less time in prison.  The critical question then is ‘Can prison populations be reduced without endangering the public?
Perhaps the best recent example is in the state of California, where the California Public Safety Realignment Act 2011 forced the California State to reduce the size of the total prison population.  Within 15 months, the prison population was reduced by  by 27,527 inmates, prison crowding declined from 181%to 150% of design capacity, approximately $453 million was saved, and there was no adverse effect on the overall safety of Californians (Sundt et al, 2016). The authors concluded:
“We make a mistake when we assume that prisons are the only meaningful or viable response to crime. Instead, we should ask whether the relative benefits of imprisonment are greater than the broad array of policies available. The answer to that question is becoming increasingly clear: Imprisonment may affect crime, but it does so at a high social, human, and economic cost and is far less cost-effective than alternatives”.
A growing number of states in the US, are actively downsizing prisons, and finding that the recidivism rate is dropping as a result.   On the basis of that evidence, it would be safe to reduce the sentences of  non-violent offenders by say 20%, i.e. 10 weeks for every year sentenced.  The evidence is that there would be a significant saving,  a reduction in reoffending, and all without any risk to public safety. A far better outcome than the 12% maximum reduction proposed by the ACT policy and at less cost.
Is the Prison Blowout due to a Blowout in Reoffending?
David Seymour contends that our prison population blowout is actually a reoffending blowout i.e. Corrections figures show a massive 69% of people on new sentences have been sentenced previously; 48% of released prisoners are back inside within four years. For prisoners aged under 20, that figure is 70%.
Despite the government’s BPS ‘Reducing Reoffending’  hype, recidivism rates have worsened.  Over the last five years, there has been some success in reducing re-conviction rates for released prisoners— in the order of 1% to 3% reduction.  Re-imprisonment rates have, however, deteriorated—increasing as a percentage by 1% to 3% over the past five years (Johnson, 2017:25).  There are a whole lot of reasons why our recidivism rate is so bad, in comparison to countries like Norway, where the reoffending rate after five years is in the order of 20% (Sterbenz, 2014).
But the recidivism rate is not the reason for the recent blowout.  Between 2014/15 and 2015/16, there was a 3% rise in recorded offences.  However, an increasing proportion of those convicted of criminal offences are being sent to prison. We have become more punitive.  During the year to 30 June 2016, almost one-in-eight people convicted of a crime received a prison sentence, although because of falling rates of recorded crime, the total numbers of people sent to prison is 10% fewer than in 2011 (Ibid: 24)
The major drivers for the blowout are two fold; a huge increase in the number of people remanded in custody, (due to Judith Collins Bail Amendment Act 2013) and the increased reluctance of the Parole Board to release prisoners.  At the end of December 2016, our prisons detained a record 9,914 people, of whom a record 28% or 2,774 were on remand.  Ten years earlier, just 21% or 1,765 of the prison population of 7,700 was on remand. During 2016, 60% of the increase of the prisoner population was due to this growth in the number of people held on remand (Ibid: 23).
The Parole Board’s approval of parole for prisoners declined from 31% of all parole applications in 2011 to 23% in 2016.  While the decline in actual numbers of this period is relatively small—from 1,542 approvals in 2011 to 1376 in 2016—this decline is against a background of a rising prisoner population. As a share of the sentenced prisoner population, parole approvals declined from a high 22.8% in 2012 to 20.2% in 2016 (Ibid: 25).
Policy Implications
ACT must be commended for not engaging in ‘law and order’ politics so close to a general election, and instead coming up with a ‘smart on crime’ proposal.  If successful, it has the potential to challenge and change current thinking, and the way the community relates to prisoners.
It also demonstrates what can happen, when we engage meaningfully with prisoners and offenders and their whānau.  When perspectives change, policy is challenged.
However, a greater challenge awaits.  The real struggle occurs when the ex-prisoner starts door knocking for a job.  The mere mention of a criminal conviction, however distant or minor, is usually sufficient for the door to shut.  My submission on a petition to the Justice and Electoral Select Committee by Eric Knight and 156 others, addresses some of those issues in detail.
We can only hope that this initiative may encourage the other political parties to disengage from the ‘tough on crime’ mantra, and start to explore policies that provide smart solutions to the issue of prison overcrowding.
Kim Workman is an Adjunct Research Associate at the Institute of Criminology, Victoria University of Wellington. This article was originally published by the Criminology Collective, and is republished with permission of the author.
References
Department of Corrections (2009) ‘‘What Works Now – a review and update of research evidence relevant to offender rehabilitation practices within the Department of Corrections, DoC: Strategy, Policy and Planning Section.
Harris, A.J.R, and Hanson, R.K. (2004) Sex offender recidivism:  A simple question (No. 2004-03) Ottawa.  Public Safety and Emergency Preparedness Canada.
Johnson, A (2017) ‘Off the Track’ Salvation Army State of the Nation Report.
Lawrence A (2009) Cutting Corrections Costs: Earned Time Policies for State Legislatures, National Conference of State Legislatures, Denver, 2009.
Parker P (2016) Restorative justice in prison: A contradiction in terms or a challenge and a reality? UK Prison Service Journal, November 2016, No. 228, pp15-20.
Sterbenz C (2014) Why Norway’s Prison System is So Successful, Business Insider Australia, 12 December 2014.
Sundt, J. et al (2016) Is Downsizing Prisons Dangerous?  The Effect of California’s Realignment Act on Public Safety, Criminology & Public Policy , Volume 15 , Issue 2, pp 315-341.
Workman K (2016) Restorative Justice in New Zealand Prisons: Lessons from the Past, UK Prison Service Journal, November 2016, No.228, pp21-29.