In New Zealand a person can be sentenced for one or more of the following purposes under the Sentencing Act 2002: accountability of the offender, responsibility for and acknowledgement of harm by the offender, to provide for the interests of the victim, reparation, denunciation, deterrence, protection of the community and rehabilitation and reintegration. The Court of Appeal has said that deterrence is a “fundamental requirement” and a “primary sentencing objective”.

But what is the evidence that imprisonment, and in particular imposing long prison terms, furthers deterrence, or any of the Sentencing Act purposes? In the absence of good evidence, how can we justify this very significant incursion on one of the most basic human rights, and the associated costs?

New Zealand has one of the highest imprisonment rates in the OECD

A 2016 Ministry of Justice study found that 71% of respondents thought national crime rates were increasing.

Since the 1990s, crime rates in New Zealand have, in fact, declined so that they are now lower than in the 1970s. In the same period, incarceration rates have shot up so that by 2018, New Zealand imprisoned more people per capita than any other OECD country, save the US, Turkey, Israel and Chile. New Zealand imprisons over 200 people per 100,000; the OECD average is below 150. Between 1998 and 2017 the number of people in New Zealand prisons almost doubled. The total cost of operating prisons has correspondingly doubled since 2005, and tripled since 1996.

Many studies confirm that there is low correlation between crime rates and imprisonment, over time and between jurisdictions. So it is not plausible that this increase in use of imprisonment has caused the decrease in crime rates in recent decades.

The costs of New Zealand’s imprisonment of such large numbers of people are enormous and troubling. In addition to the direct financial cost of over $100,000 per prisoner per year, there are very significant costs to society in terms of lost jobs, inability to secure jobs in future, adverse health impacts, harm to families and increased recidivism. Māori have borne the brunt of the human cost, with devastating consequences.

No evidence that lengthy sentences rehabilitate, deter or increase protection for the community

In R v Wellington [2018] NZHC 2196, Palmer J drew attention to the absence of evidence that long prison sentences deter reoffending. The President of the Court of Appeal has spoken publicly about the growing concern that imprisonment, and longer prison terms, are likely to increase recidivism.

An assumption that underlies much of the rhetoric in this area is that imprisonment prevents offending by prisoners, and this protects the public from crime. But 2016/2017 Corrections figures suggest that over 99% of those imprisoned will be returned to the community. Research increasingly suggests that, while imprisonment may reduce an individual’s criminal offending (outside the prison walls) as long as that person is in prison, at best it has no effect on reoffending later and often results in a greater rate of recidivism. The threat of imprisonment generates a small general deterrent effect and increases in the certainty of apprehension and punishment demonstrate a significant deterrent effect. But increases in the severity of penalties, such as increasing the length of terms of imprisonment, do not produce a corresponding increase in deterrence.

Imprisonment also has poor rehabilitative effects. Those involved in the sector have expressed concern that prison criminalises people and erodes their ability to contribute positively to their society. This has implications for the likelihood that they will reoffend, as well as their communities’ resilience and productivity. To the extent that specific rehabilitation programmes can be shown to have an effect (and any effect shown has invariably been small), these programmes can be provided (and are more likely to be effective) in contexts other than prison.

Canadian researchers Paul Gendreau, Claire Goggin and Paula Smith warned in a 2002 study that excessive use of prison may be indefensible and indeed “fiscally irresponsible”, given the significant wider social costs of even modest increases in recidivism.

As it becomes increasingly difficult to avoid reckoning with the evidence on the absence of deterrent and rehabilitative effect from lengthy sentences, a number of New Zealand judges have fallen back on consistency as a justification for imprisoning people for lengthy periods. How long are we going to continue to consistently do something indefensible and fiscally irresponsible?

What about denunciation and accountability?

There may be a more obvious relationship between lengthy sentences and retributive justice. But even retributive justice requires that the punishment is proportionate to the seriousness of the crime. There is little or no analysis to demonstrate this relationship. And it is not easy to see why denunciation requires such significant prison sentences in New Zealand, but not in other comparable OECD countries. This is an outcome at odds with New Zealand’s pride in its kindness, tolerance and inclusiveness.

While it is difficult to assess proportionality, it is not impossible. Some reasonably sophisticated attempts to do so have been made, including by the Law Commission in a September 2013 Study Paper, which suggested that there are substantial anomalies in current maximum penalties.

The Law Commission called for a wholesale revision of maximum penalties to ensure fairness in sentencing practice. At a minimum the approach should be logical and rational so that, for example, punishment should not be obviously disproportionate to the crime, especially when compared with punishment for other crimes generally considered more serious. The prime example of this is the very lengthy sentences imposed for drug offences compared with sentences for murder and other violent offences. One of the anomalies identified by the Law Commission in 2013 was the maximum penalty of life imprisonment for dealing in a Class A drug (that is otherwise reserved only for a handful of the most serious offences including murder, manslaughter, treason and piracy involving murder).

Serious concerns about consistency of sentencing have also been raised by Wayne Goodall’s research. His 2014 PhD thesis found very substantial variability between circuits in use of imprisonment for aggravated drink driving and burglary. The result was that people were being incarcerated in some circuits when they would not have been had they been sentenced in other circuits.

These kinds of anomalies in sentencing are a risk to the legitimacy of the criminal justice system. They can be expected to have a negative impact on reporting of crimes and cooperation with the Police. Significantly for present purposes, they can erode the deterrent function of the criminal law (especially given that research suggests that the prospect of being caught is the most significant factor in deterrence).

There are a number of troubling facts that exacerbate these concerns: the disproportionate imprisonment of minority and disadvantaged groups, the effect on the families of those imprisoned and the disproportionate number of people on remand for lengthy periods.

Disproportionate impact on victims, and disadvantaged and minority groups

The most vulnerable and disadvantaged members of society are disproportionately incarcerated. Most of those in prison are victims themselves. The March 2018 report by the Prime Minister’s Chief Science Advisor records that 77% of those in prison have been victims of violence, and 91% have a lifetime diagnosable mental illness or substance use disorder.

Māori are significantly over-represented in the criminal justice system, making up 51% of the prison population (compared with about 15% of the general population). The position is even more extreme for Māori women: they make up over 60% of the female prison population. Pacific peoples comprise about 11% (compared with approximately 7.5% of the general population). Māori and Pacific peoples also form a disproportionate percentage of victims: they are more likely to be a victim of crime than Pākehā New Zealanders.

These statistics also raise wider questions about the experience of indigenous people, minorities, victims and the poor and powerless of our criminal justice system generally. As a country which prides itself on its fair, tolerant and inclusive approach, we need to confront these issues. But in the meantime they underscore the need to urgently address how and why we are imprisoning so many people, at such enormous cost to society.

Impact on families and communities

Even if New Zealand society’s urge for retribution and vindication justifies the impact of lengthy incarceration on offenders, can it justify the effect on their families and communities?

The Prime Minister’s Chief Science Advisor’s 2018 report records that almost one in five of those in prison (19%) has a direct parenting role at the time they are imprisoned. This means that about 3,800 children currently have one active parent in prison. Another 20,000 have a non-active parent in prison.

The effects of imprisonment on families, and particularly children, are still not well understood. In addition to the direct impact that many suffer of losing a caregiver, family income and in many cases, their homes, there are also dynamic, less visible and direct, consequences which play out over time. Evidence suggests that incarceration of parents leads to greater behavioural problems for children. Adolescents with incarcerated parents have a greater risk of mental health problems. Following a first offence, those whose parents have a criminal conviction have a greater risk of subsequent offending. As noted above, almost all these parents will have a pre-existing mental illness or substance-abuse disorder or will develop one in prison.

Disproportionate effect of remand

A large proportion of those in prison are being held on remand (the figure climbed steadily to 30% as at 31 December 2018). When combined with the high prison figures, this means a very large number of people are held on remand. Average time on remand has increased to 66 days in 2017 from under 40 days in 2000. Many of these people will ultimately be either acquitted or given non-custodial sentences.

There is reason to expect that remand is disproportionately costly to society in terms of its impact on individuals and the functioning of already struggling communities. Poorer people are less likely to meet bail requirements and are therefore more likely to lose jobs, custody of their children and housing while held on remand. They are also more likely to plead guilty to something they didn’t do, and face the life-long consequences of a conviction.

Does the law dictate this outcome?

There is a real question mark over the extent to which imprisonment furthers the Sentencing Act purposes. This is routinely assumed, rather than being carefully examined in any given case. Even the reliance on prevention (while a person is in prison) as a justification does not appear to be based on evidence that the particular person is likely to reoffend unless they are incarcerated.

The position is even starker in relation to length of sentence. There is evidence that lengthy sentences undermine purposes such as deterrence, and rehabilitation and reintegration, and are very costly to society. There are also serious questions about their justification on retributive grounds given the questionable rationality of the relationship between particular sentences and particular offending, a comparison with other OECD jurisdictions and a clear picture of those disproportionately imprisoned.

In these circumstances it seems likely that many sentences are inconsistent with the New Zealand Bill of Rights Act provision for freedom from arbitrary arrest or detention. Excessive, arbitrary or inconsistent use of imprisonment cannot be a reasonable limit on this right, which is demonstrably justified in a free and democratic society, as required by section 5. Irrational and disproportionate sentencing that is more likely to undermine the objective it is being relied on to further than it is to support that objective must fail both the requirement that detention not be “arbitrary” and the section 5 test. New Zealand’s high rates of imprisonment and the disproportionate impact on vulnerable people and Māori require a very careful examination of these questions.

The excessive numbers of people held on remand for lengthy periods is even more problematic, given the presumption that they are innocent.

It is far from clear that the legislation requires the approach that has been taken. If it does, then it is a good candidate for a declaration of inconsistency with the New Zealand Bill of Rights Act.

Liesle Theron is currently on sabbatical. She is the convenor of the Law Society’s Law Reform Committee and a member of the Women’s Advisory Panel. The article reflects her personal views, and not those of any organisation.