The Sentenced Prisoner Population

Wayne Goodall
Principal Strategic Analyst, Department of Corrections

Author biography:
Wayne has been employed by the Department of Corrections for two years. For almost all of the last 22 years he has worked in the Justice sector. Wayne has a PhD in criminology, his thesis addressed regional disparity in sentencing in the District Courts.


Since 1980, New Zealand’s sentenced prisoner population has grown by 170%. A full explanation of the change would encompass, among other factors, changes in the volume and mix of offenders and offences, societal change, policy change and changes in sentencing approach. A paper within the confines of this journal cannot disentangle the relative contribution of each factor. This paper primarily focuses on the generalised link between policy change and movements in the sentenced prisoner population with limited reference to other factors. It concludes with a description and discussion of the changing composition of the population and the link back to policy changes.

The sentenced prisoner population 1980-2016

The growth in the sentenced prisoner population is depicted in Figure 1 along with a series of key policy changes. At a very general level the population was stable from 1980 through to 1985 before rising fairly constantly for more than 20 years through to 2007. The detail reveals three noticeable short term drops in 1985, 1993 and 2007 and a plateau around the turn of century. All of these can be linked to policy changes or the absence of change.

Figure 1: The sentenced prisoner population 1980-2016 and key policy changes

Impacts of legislative change

The following narrative works through the series of major policy changes included in Figure 1. It is followed by some examples of the influence of changes to maximum penalties for individual offences and an example of what might be classed as non-legislative responses to a sentinel event (a single or small number of horrific offences in close proximity). The discussion concludes with a brief reference to impact of the custodial population on the sentenced prisoner population.

The Criminal Justice Act 1985

The Criminal Justice Act 1985 was the first policy change post-1980. The immediate effect of the Act was a reduction in the prison population. The reduction was primarily due to the introduction of a presumption against imprisonment for property and other non-violent offending and a heightened expectation of imprisonment for more serious violent offending. The Act introduced universal parole eligibility at half sentence and reduced mandatory release from three-quarters of time served to two-thirds.

The immediate effect of the 1985 Act was not sustained; the population returned to close to 1985 levels within two years. This was a consequence of the more severe sentences for violent offending coupled with a growth in convictions for violent offending.

The 1985 Act marks the start of a period of change in the composition of the prison population. This is discussed in more detail in a subsequent section of this paper.

The 1987 Criminal Justice Amendment Act

In 1987 the first of a series of amendment acts responded to rising concern about law and order issues. The concern was spurred by a series of sentinel events – the abduction and murder of two primary school girls (Louisa Damodran and Teresa Cormack), the abduction and rape of a woman by a recently released prisoner and a gang rape. Parole eligibility was abolished for offenders serving sentences for specified serious violent offences. Parole eligibility for those serving life sentences or on preventive detention was increased from seven to ten years, and the scope of preventive detention was broadened by reducing the minimum age from 25 to 21 and adding specified serious violent offences to the list of qualifying offences.

1993 Criminal Justice Amendment Act

Like the 1985 Act, the immediate effect of the 1993 Criminal Justice Amendment Act was a reduction in the prison population. This occurred despite a further strengthening of provisions governing the sentencing and release of sex offenders and serious violent offenders. The changes were also in response to disquiet about high profile crimes; the rape and murder of 15 year-old Kylie Smith and the rape and murder of primary school girl Sarah Curry. The maximum penalty for rape and unlawful sexual connection was increased from 14 to 20 years. Despite resulting in a marked increase in sentence lengths, the impact of the change on the prison population was delayed. This is common for increases in maximum penalties because the impact only becomes noticeable when the first offenders sentenced under the new law reach the point at which they would otherwise have been released but for the longer sentence.

The immediate decrease after the passage of the amendment and stable population through to early 1997 is linked to three measures: The introduction of suspended sentences, provision for the release of non-violent offenders on home detention, and reduction of parole eligibility from half to one-third for prisoners other than serious violent or sexual offenders. Judges were able to suspend any prison sentence of six months to two years, which was expected to divert offenders from prison and result in a long term reduction in the prison population.

The expected longer term reduction in the population due to suspended sentences did not eventuate. There were “unintended consequences” as suspended sentences began to add more prisoners than were diverted. This has been attributed to a high proportion of suspensions in cases where the offender would not otherwise have been imprisoned. The subsequent activation of some of these suspensions outweighed the number diverted from prison.

1999 Criminal Justice Amendment Act

The severity of sentences for serious violent and sexual offences was further increased in the 1999 Criminal Justice Amendment Act if the offending took place during a home invasion. In effect a new set of offences was created since the maximum sentence for certain offences became five years greater than for the same offence absent from the element of home invasion. A 13 year minimum non-parole period was introduced for murder. Any impact the change had on the prison population was either negated by other shifts in offending or sentencing or was too small to be noticeable; the population was relatively stable from 1999 to 2002. The latter explanation is the most likely for two reasons. First, the number of offences occurring in these circumstances would likely have been small, but more importantly, as described above, any effect would have been delayed until the prisoner went beyond the point at which they would otherwise have been released. The changes were short-lived, being repealed as part of the 2002 reforms.

Sentencing Act 2002 & Parole Act 2002

In 2002 the sentencing and release provisions were comprehensively reformed with the Sentencing Act 2002and theParole Act 2002 replacing the relevant parts of the Criminal Justice Act 1985. The Sentencing Act is very different to the Criminal Justice Act, but many of the changes had little effect on sentencing patterns because they codified the sentencing practices already followed by the Courts. There were two changes targeted at the most serious violent offenders. First, the qualifying age for preventive detention was further reduced from 21 to 18. Second, a minimum non-parole period of at least 17 years was required for offenders convicted of murder in any of nine aggravating circumstances unless it would be manifestly unjust to do so. In addition to abolishing the “home invasion” provisions, the Act also abolished suspended sentences.

The Parole Act was revolutionary. All offenders sentenced to more than two years on a determinate sentence became eligible for parole at a minimum of one-third or up to two-thirds if the sentencing judge ordered a minimum non-parole period. The almost automatic release at two-thirds was removed meaning that offenders could serve the full sentence.

The change was expected to result in an increase to the prison population; it was designed to make dangerous offenders, who posed a risk to the community, serve a greater proportion of their prison sentence. Although it was estimated that the prison population would increase by about 400, the actual increase has been much higher. The effect has been estimated to be an increase of about 1,500, mostly occurring between 2002 and 2008.

Sentencing Amendment Act 2007

The Sentencing Amendment Act 2007 introduced home detention as a sentence in its own right, and the new sentences of community detention and intensive supervision slowed the growth in the prison population. The slowing of the growth was primarily due to judges making much greater use of home detention as a sentence in its own right, compared to previously where it was a way for an offender to serve all or part of a short sentence of imprisonment (two years or less). Under the old regime, judges granted leave to offenders to apply to the Parole Board to serve the sentence on home detention. The Parole Board was possibly more risk averse than judges have proven to be, but also appeared to consider whether the prisoner deserved to be released on home detention rather than just considering whether the risk they posed warranted keeping them in prison. It has been estimated that the change has resulted in the prison population being about 1,000 lower than would otherwise be the case.

Sentencing and Parole Reform Act 2010

The Sentencing and Parole Reform Act 2010 introduced the three strikes regime. It is another example of a change which will have delayed effects. It takes time for offenders to accrue strikes, and it takes time for those offenders to reach the point in their sentence where they would otherwise have been released but, due to the three strikes regime, remain in prison. At present there are around 2,400 “first strike” offenders in prison, and almost 100 “second strike” offenders, with the possibility that the first “third strike” case is likely to be received in the near future. The prison forecast allows for an additional 250 prisoners over the next 10 years due to the three strikes regime.

Increased maximum penalties and a sentinel event

There have been periodic increases to maximum penalties over the years, the following three examples provide a sense of the effects such changes can have:

  • In 1998 the maximum penalty for a third or subsequent drink driving offence increased from three months to two years, resulting in an increase of about 50 in the prison population.
  • In 2003 methamphetamine was re-classified from Class B to Class A, lifting the maximum penalty for dealing offences from 14 years to Life Imprisonment, resulting in an increase of about 150 in the prison population.
  • In 2013 the maximum penalty for breaching a protection order increased from two to three years resulting in a four percentage-point increase in the imprisonment rate and a seven-week increase in average sentence length resulting in an additional 50-70 prisoners.

Sentinel events can also play a significant role outside the contribution to policy changes discussed above. For instance, the serious re-offending while on parole by Graeme Burton in 2006 resulted in a rapid increase in the prison population. Not only did the Parole Board adjust its decision-making, courts became more inclined to remand defendants in custody.

The custodial remand muster

A discussion of the sentenced prisoner population would be incomplete without a brief mention of the relationship with the custodial remand population. Increases and decreases in the custodial remand population may be partially matched by a reverse effect on the sentenced prisoner population. An increase in the rate of custodial remand for an offence type is, all other things being equal, likely to suppress the sentenced prisoner population because more of those sentenced to prison will have served time on remand. Likewise an increase in the average time on remand is likely, all other things being equal, to suppress the sentenced prisoner population because offenders have less time left to serve post-sentence. It is likely that the increase in remand population over the last 18 months (more than 700 additional prisoners) has suppressed the sentenced prisoner population.

The changing composition of the sentenced prisoner population

The composition of the prison population by high level offence groupings has changed markedly since 1980. The change is depicted in Figure 2 (the change in volume) and Figure 3 (the proportionate composition of the population).

Figure 2 shows very sharp changes beginning in 1985. Prior to 1985, burglary and dishonesty offenders were the largest group in prison. The Criminal Justice Act 1985 created a swift change. By the end of 1987 there were more violent offenders in prison than burglary and dishonesty offenders and in July 1993 the number of sexual offenders overtook the number of burglary and dishonesty offenders. To some extent this was a consequence of growing numbers of offenders but it was also substantially influenced by the legislative changes. More serious violent and sex offenders could expect to be imprisoned whereas there was a presumption against the imprisonment of others. The subsequent changes in 1987 and 1993 increased the severity of sentencing for violent and sex offenders. It took until 1997 for the number of burglary and dishonesty offenders to return to pre-1985 levels. All three groups rose in the five years immediately after the Parole Act 2002 came into force. Many offenders were not released until later in their sentence and some were not released until sentence expiry; notably more than one-third of burglars served their entire sentence in prison. In 2007 the number of burglary and dishonesty offenders fell whereas the other two groups continued to increase.

While these changes were taking place, the number of drug offenders in prison slowly increased through the 1990s and subsequently grew more quickly. The number of offenders imprisoned for offences outside the four named groups has been relatively constant.

Figure 2: The composition of the sentenced prisoner population by offence type

All these shifts mean that the composition of the sentenced prisoner population is very different now compared to 1980; this is depicted in Figure 3.

As expected there has been substantial growth in the proportions of violent and sexual offenders and a decrease for burglary and dishonesty offenders. Consistent with the changing numbers, the population has an appreciably higher proportion of prisoners serving sentence for drug offences, whereas the relatively unchanging number of prisoners for other offences means they now account for a much smaller proportion than the 20% up to 1985.

There is a surprising dip in the violent offender share of the population in the mid-2000s; there is a drop and then a “correction”. There is a much shallower dip for sex offenders. Both dips can be explained by a combination of factors. The first is that there was an upturn in the number of prisoners serving sentence for drug offences which reduced the proportionate shares for both groups. The second and third factors apply to policy changes that had disproportionate impacts on burglary and dishonesty offenders. The beginning of the dip coincides with the application of the new parole regime. The regime impacted more quickly on burglary and dishonesty than other groups because these offences attract a greater proportion of parole governed sentences. As the Parole Board began to decline parole for these prisoners because they imposed an undue risk of re-offending, the number of prisoners began to climb. The effect was relatively short-lived because of the 2007 changes, in particular home detention becoming a sentence directly available to judges. The proportion of short sentences converted to home detention was much greater for burglary and dishonesty offenders and the change was reversed.

Hidden within the violent and sex offender numbers and proportions is the growing significance of indeterminate (life and preventive detention) sentences. In June 1980 offenders serving these sentences (80) accounted for 3.2% of the sentenced prisoner population. By June 2016 the number had grown to 829 and accounted for 12% of the sentenced prisoner population. The increase is partially attributable to the periodic changes expanding the eligibility criteria for preventive detention. The increases in sentence length result from changes to sentencing maximums for sexual violation and unlawful sexual connection and longer minimum non-parole periods.

Figure 3:
The proportionate composition of the sentenced prisoner population by offence type

Closing comment

It is readily apparent that a large part of the growth in the prison population can be linked to changes in policy. In general the growth was expected, but on some occasions, for example the parole reform of 2002, the impact was much greater than anticipated. Those measures intended to reduce the prison population have had mixed results. The introduction of home detention as a sentence was successful, whereas suspended sentences did not have the desired result.