Reconviction patterns of offenders managed in the community: A 48-months follow-up analysis

Summary

This report summarises patterns of reconviction and imprisonment, over a 48 months period, amongst almost 35,000 offenders who started community sentences (Supervision, Community Work) and orders (front-end Home Detention) during the 12 months period 1 July 2002 to 30 June 2003.

The analysis reveals important differences of reconviction behaviour of different sub-groups of offenders who started community sentences in 2002/03.  Offender characteristics such as gender, age at start of sentence, ethnicity, offence type and offenders' previous criminal history are each examined with reference to reconviction and imprisonment.

The report is part of a sequence of reports, available on the Department of Corrections' website; earlier reports have presented reconviction and re-imprisonment amongst released prisoners with 36 and 48 months follow-up periods 1.

Introduction

Reducing re-offending is an important outcome objective for most correctional services.  As such, measures of recidivism, particularly reconviction and re-imprisonment rates, are key indicators of performance.

The data presented here are based on the "recidivism index" (RI) methodology used in the Department of Corrections' ("the Department's") annual reporting of reconviction and re-imprisonment.  This method quantifies the rate of reconviction and re-imprisonment for specified sub-groups of offenders, over follow-up periods of defined length, after release from a custodial sentence or from the start date of a community sentence. Conviction and sentencing data is obtained from the Ministry of Justice's Case Management System (CMS) database.

This report summarises patterns of reconviction and imprisonment amongst almost 35,000 offenders who started community sentences (Supervision, Community Work) and orders (front-end Home Detention) during the 12 months period 1 July 2002 to 30 June 2003 2.

Recidivism figures are produced for two potential outcomes: reconvictions leading to any sentence administered by the Department (community-based or prison), and reconviction leading solely to a term of imprisonment.  Imprisonment figures are generally considered the more critical of the two measures, as this outcome is associated with more serious offences and higher costs of sentence administration.  Recidivism figures exclude reconvictions which do not result in sentences administered by the Department (fines, convicted and discharged, etc) - all references in this report to reconvictions should be interpreted as restricted to convictions resulting in imprisonment or a community-based sentence. Reconvictions include breaches of previous sentences.

It is also important to note that most of the figures here represent any reconviction or imprisonment during the 48-month follow-up period: some of the individuals reconvicted will have had multiple, successive convictions during that period.

The figures below reveal important differences of reconviction behaviour of different sub-groups of offenders who started community sentences in 2002/03.  Offender characteristics such as gender, age at start of sentence, ethnicity, offence type 3 and offenders' previous criminal history are each examined with reference to reconviction and imprisonment.

Recidivist offenders are a challenge both to criminal justice sector agencies and to society at large.  Therefore, this report examined the entire reconviction data set for the current sample by dividing the population into two groups, "first-timers" and "recidivists".  A number of important findings emerge from this perspective on recidivist offenders.

Offence types by individual sentence types

The graphs below provide some background information on the distribution of offenders who started Supervision and Home Detention, by their most serious offence category.  Additional detail of this nature is summarised in Appendix 1.

Figure 1: Proportion of offenders by offence category, Supervision

Figure 2: Proportion of offenders by offence category, Home Detention

Recidivism rates; all community sentences

Across the entire sample of offenders who commenced Supervision, Community Work and Front-end Home Detention in 2002/03, 18 percent were convicted of a new offence and started a prison sentence at least once during the 48-months follow-up period.  For the same period, 36 percent were convicted and started a community sentence.  This means overall reconviction rate of offenders who started Supervision, Community Work and Front-end Home Detention was 54% over four years.  To provide more background to the reconviction rate, the relationship between time and first re-offence is shown below.

Figure 3:  Reconviction rate by time to first re-offence

To clarify the above graph, the shape of the two curves is illustrative of the relationship between the volume of offenders reconvicted and elapsed time over four year follow-up period.  The lower (blue) line is the cumulative proportion reconvicted , and indicates that numbers rise steadily early in the follow-up phase, such that by the one-year mark, 32 percent of the sample had already been reconvicted.  By the two-year mark 43 percent had been reconvicted, by the three-year mark 50 percent of the sample, with the 54 percent figure attained by 48 months.  The shape of this curve predicts that, in a further twelve months time (after five years follow-up), around 57 percent of the entire sample will have been reconvicted.

The upper (red) curve in the above graph shows the proportion of all reconvicted (in percentages) who had been reconvicted by successive three-monthly intervals.  This shows, for example, that of those who were reconvicted, over half (59 percent) were reconvicted within the first twelve months.  From that point the rate of "relapse" slows, with a further 20 percent reconvicted by the 24-month point, 12 percent reconvicted by 36 months, and the remaining nine percent convicted over the final twelve months.

Comparisons of reconviction or reimprisonment rates with other countries is usually a fraught exercise, given that countries differ markedly in how criminal justice data are handled – for example,  whether new offences are counted from the commencement of the initial sentence (as in New Zealand) or only after completion of the community sentence (Australia).  However, a United Kingdom Home Office report 4 provided data which appear to have been derived from a similar methodological approach to ours.  Their figures (using a 24 months follow-up period only) indicated an overall reconviction rate of 51% for all community sentences within two years, which is somewhat higher than the rate in New Zealand (43%) within that same period.

Reconviction and imprisonment rates by age; Supervision sentences

Supervision is a community-based sentence with a rehabilitative focus. Sentences of Supervision range from six to 24 months; each year about 6000 Supervision sentences are handed down.  Offenders sentenced to Supervision tend to have longer criminal histories compared to those receiving Community Work or Home Detention.  Across the entire sample of offenders who started Supervision in 2002/03, 16% were under the age of 20 years, and 15% of offenders were aged 40 or over.

Of the Supervision group of offenders, 26 percent were convicted of a new offence and received a prison sentence during the 48-months follow-up period.  In the same period, 35 percent were convicted and started a community sentence.  The overall reconviction rate of offenders who started Supervision therefore was 61% over four years.

The graph below gives rates of reconviction and imprisonment for offenders of different age bands (note that offenders' ages here are as at the start of Supervision).  Eighty one percent of those aged under 20 were reconvicted within 48 months.  Only 39 percent of those aged over 40 were reconvicted.  In other words, offenders aged under 20 are twice as likely to be reconvicted as those aged over 40.  Thirty eight percent of those aged under 20 were imprisoned compared to only 14 percent of those aged over 40 were imprisoned within 48 months.

Figure 4:  Reconviction and imprisonment rate by age, Supervision

Reconviction and imprisonment rates by gender; Supervision

Across the entire sample of offenders who started Supervision in 2002/03, 17% were female.  The rate of reconviction over 48 months for male offenders (64%) is higher than the rate of reconviction for female offenders (52%).  The rate of imprisonment for male offenders (28%) is considerably higher than the rate of imprisonment for female offenders (16%).

Reconviction and imprisonment rates by ethnicity; Supervision

Across the entire sample of offenders who started Supervision in 2002/03, 50% were Maori, 38% were European and 11% were Pacific offenders.

The reconviction rate over 48 months for Maori offenders (67%) is considerably higher than the rate for both NZ Europeans (55%) and Pacific offenders (58%).  This difference is likely to be a reflection of a number of variables such as age and offence type: there are higher numbers of young Maori offenders, and Maori are more likely to be convicted for offences which have high base-rates, especially dishonesty offences (burglary, car conversion, theft, etc – see below) 5.

The imprisonment rate over 48 months for Maori offenders (30%) is also significantly higher than the rate for both NZ Europeans (23%) and Pacific offenders (20%).

Figure 5:  Reconviction and imprisonment rates by ethnicity, Supervision

Reconviction and imprisonment rates by original offence type; Supervision

Across the entire sample of offenders who started Supervision in 2002/03, 36% of offenders were convicted of either violent or sexual offences.  About 26% offenders were sentenced for dishonesty offences and 23% of offenders were sentenced for traffic offences.

The following graph shows reconviction and imprisonment rates by the most serious offence for which the offender had previously been sentenced.  These clearly reveal that reconviction rates vary considerably between different types of offences: rates are highest among those who committed property and dishonesty offences (burglary, car conversion, theft, etc) and lowest among those who were convicted of sexual offences.

Figure 6:  Reconviction and imprisonment rates by offence type, Supervision

Reconviction and imprisonment rates by age; Community Work

In 2002 the sentence of Community Work replaced sentences known as Periodic Detention and Community Service.  Community Work is the most widely used community sentence: each year more than 30,000 offenders start Community Work sentences, more than all other community sentences combined.

Of offenders who started Community Work in 2002/03, 16 percent were convicted of a new offence and started a prison sentence during the 48-months follow-up period.  For the same period, 38 percent were convicted and started a community sentence.  This means overall reconviction rate over four years for offenders who started Community Work was 54%.

Across the entire sample of offenders who started Community Work in 2002/03, 19% of offenders were under the age of 20 and 14% of offenders were aged 40 or more.

The graph below gives rates of imprisonment and reconviction for offenders of different age bands (again, offenders' ages are as at the start of the Community Work sentence).  Seventy percent of those aged under 20 were reconvicted within 48 months, and only 33 percent of those aged over 40 were reconvicted.  In other words, offenders aged under 20 are twice more likely to be reconvicted than those aged over 40.

Figure 7:  Reconviction and Imprisonment rate by age at start, Community Work

Reconviction and imprisonment rates by gender, Community Work

Across the entire sample of offenders who started Community Work in 2002/03, 20% were female.  The reconviction rate over 48 months for male offenders (58%) is much higher than the rate for female offenders (41%).  The imprisonment for male offenders (18%) is considerably higher than the rate for female offenders (8%).

Reconviction and imprisonment rates by ethnicity, Community Work

Across the entire sample of offenders who started Community Work in 2002/03, 46% were Maori, 42% were NZ European and 9% were Pacific offenders.

The reconviction rate over 48 months for Maori offenders (60%) is considerably higher than the rate for both NZ Europeans (50%) and Pacific offenders (49%).  This difference is likely to reflect of a number of variables such as age and offence type.

Figure 8:  Reconviction and imprisonment rates by ethnicity; Community Work

Reconviction and imprisonment rates by original offence type; Community Work

Across the entire sample of offenders who started Community Work in 2002/03, 26% of offenders were sentenced for traffic offences for their original sentence.  About 25% offenders were sentenced for dishonesty offences and only 14% of offenders were sentenced for either violent or sexual offences.

Figure 9:  Reconviction and imprisonment rates by offence type, Community Work

The graph above shows reconviction and imprisonment rates by the most serious offence for which the offender had previously been sentenced.  These clearly reveal that reconviction rates vary considerably between different types of offences: rates are highest among those who committed property and dishonesty offences (burglary, car conversion, theft, etc) and lowest among those who were convicted of sexual offences.

Reconviction and imprisonment rates by age; Home Detention

Home Detention in 2002/03 was a community-based order that allowed offenders to serve their prison sentence at an approved place of residence, under electronic monitoring and close supervision by a Probation Officer 6. For each of the last few years over 1000 offenders commenced on Home Detention.  The graph below gives rates of imprisonment and reconviction for offenders of different age bands (offenders' ages are as at the start of the Home Detention period).

Across the entire sample of offenders who started Home Detention in 2002/03, 23 percent were convicted of a new offence and started a prison sentence during the 48-months follow-up period.  For the same period, 21 percent were convicted and started a community sentence.  This means the overall reconviction rate of offenders who started Home Detention was 44%.

Figure 10:  Reconviction and Imprisonment rate by age; Home Detention

Reconviction and imprisonment rates by gender; Home Detention

Across the entire sample of offenders who started Home Detention in 2002/03, 19% were female.  The reconviction rate over 48 months for male offenders (47%) is much higher than the rate for female offenders (31%).  The imprisonment for male offenders (25%) is considerably higher than the rate for female offenders (12%).

Reconviction and imprisonment rates by ethnicity; Home Detention

Across the entire sample of offenders who started Home Detention in 2002/03, 51% were European, 39% were Maori and 6% were Pacific.

The reconviction rate over 48 months for Maori offenders (55%) is considerably higher than the rate for both NZ Europeans (40%) and Pacific offenders (37%).  Interestingly however, in contrast to the figures reported above for both Supervision and Community Work, the imprisonment rate for Pacific offenders (27%) is almost same as the rate for Maori offenders (28%), and the imprisonment rate for NZ Europeans (19%) is considerably lower than both.  No obvious explanation presents itself for this anomaly.

Figure 11:  Reconviction and imprisonment rates by ethnicity, Home Detention

Reconviction and imprisonment rates by original offence type; Home Detention

Across the entire sample of offenders who started Home Detention in 2002/03, 30% were sentenced for traffic offences; 23% were sentenced for dishonesty offences and 22% for drug offences.  The remaining 25% of offenders were sentenced for other offences such as violence, sex, property and administrative offences.

The graph below shows reconviction and imprisonment rates by the most serious offence for which the offender had previously been sentenced.  Rates are lower among those who were convicted of sexual, drug and property offences, compared to the higher rates observed for dishonesty, violence and traffic.

Figure 12:  Reconviction and imprisonment rates by offence type, Home Detention

Reconviction rates by number of previous sentences

The following section examines more closely the reconviction dataset by disaggregating offenders according to the number of previous sentences.  Note that the records used here do not include previous conviction resulting in fines or other minor penalties (e.g., convicted and discharged).  Those for whom the community sentence started in 2002/03 was their first Corrections-administered sentence are designated in the following as "first-timers".  The remainder are designated "recidivists".

Of interest here is a finding that 15 percent of those who started Supervision had never previously served either a community sentence or a prison sentence.  In contrast, 36 percent of those who started Community Work and 34 percent of those who started Home Detention never had previously served either a community sentence or a prison sentence.

Reconviction rates: "first-timers" and "recidivists"

The graph below shows clearly that the reconviction rates of first-timers are much lower than the reconviction rates of recidivists: further, first-timers' reconviction rates differ also between those who started Supervision, Community Work and Home Detention.   The reconviction rates of recidivists for both Supervision and Community Work is 65%, and for Home Detention is 55%, over four years.  Other analyses reported by the Department 7 confirm a picture of many recidivists cycling rapidly through a sequence of offending, conviction, community sentence (or imprisonment), and rapid return to further offending.

Figure 13:  Reconviction rate by sentence type: First-timers vs Recidivists

Conclusion

For the 2002/03 cohort of community-sentenced offenders, a number of important patterns emerge in the reconviction rates over the 48 months period for which data has been analysed.  As is commonly found in most studies of this type, the highest rates of reconviction tend to occur amongst those who are younger, male, and repeat offenders.  Those convicted of dishonesty offences also tend to recidivate more frequently and rapidly.

In terms of the sentences and orders themselves, offenders who started Home Detention orders tended to record fewer convictions than did offenders who started Supervision or Community Work. However, it is likely that Home Detainees were at lower risk of reoffending than those on Supervision at least, based on their shorter histories of offending.

For the same cohort (2002/03), five year follow-up full reconviction analysis will be published by end of June 2009, which will include further analysis of community sentences as well as the community orders of parole and release under conditions.

Appendix 1: Counts of offenders in each sub-group

Reconviction Rate by Age Group, Supervision

Reconviction Rate by Gender, Supervision

Reconviction Rate by Major Ethncity, Supervision

Reconviction Rate by Most Serious Offence Group, Supervision

Reconviction Rate by Age Group, Community Work

Reconviction Rate by Gender, Community Work

Reconviction Rate by Major Ethnicity, Community Work

Reconviction Rate by Most Serious Offence Group, Community Work

Reconviction Rate by Age Group, Home Detention

Reconviction Rate by Gender, Home Detention

Reconviction Rate by Major Ethnicity, Home Detention

Reconviction Rate by Most Serious Offence Group, Home Detention

1 See for example http://www.corrections.govt.nz/research/reconviction-patterns-of-released-prisoners-a-48-months-follow-up-analysis

2 The data set includes reconvictions where the offence date was prior to 30 June 2007, but includes offences where the reconviction date occurred up to 31 March 2008.

3 As a great many offenders are convicted for multiple offences, offenders in this analysis are grouped according to the "most serious offence" (MSO) for which they were convicted on the original sentence.  MSO rankings are determined from the Ministry of Justice Seriousness of Offence Scale, which orders offences in accordance with the average number of days imposed by judges, for that specific offence type, over the past five years.

4 http://rds.homeoffice.gov.uk/rds/pdfs07/hosb0607.pdf

5 The high base rates of dishonesty offences can be appreciated from the fact that in the year to June 2007, over 230,000 dishonesty offences were recorded by Police; on the other hand, just on 3600 sexual offences were recorded.

6 In this paper, only "front-end" Home detention is considered.  Since October 2007, Home Detention became a sentence in its own right.

7 E.g., see http://www.corrections.govt.nz/research/offender-volumes-report-2007

Back to top >