Correctional Service Canada
Symbol of the Government of Canada
Skip to content | Skip to institutional links

Common menu bar links

Home > Research > Research Reports

Institutional links

Research Reports

Warning This Web page has been archived on the Web.

Use of the Residency Condition with Statutory Release:
A descriptive analysis.

by:
Brian A. Grant, Sara L. Johnson and Mike Muirhead

Research Branch
Correctional Service of Canada

June 2000

EXECUTIVE SUMMARY

Recent amendments to the Corrections and Conditional Release Act (Amendments to the CCRA, 1995) have allowed for the imposition of a residency condition to offenders' statutory release (SR) when they are considered to be high risk of committing a serious violent offence without such a condition. Prior to this amendment, residency could only be imposed on offenders' as an alternative to detention, either following the initial referral or following a subsequent detention review. Since the amendment, there has been a dramatic increase in the use of residency from approximately 40 offenders per year to approximately 800 offenders per year. The change is primarily due to the use of the residency condition with offenders released at their statutory release date.

The study divided offenders released on SR with a residency condition into four groups: pre-amendment referred, post-amendment referred, SR-residency, SR suspended-residency. The pre-amendment referred and post-amendment referred groups included offenders who were either referred for detention or had their detention reviewed and were subsequently released on SR with residency. The SR-residency group included offenders released at their SR date with a residency condition while the SR suspended-residency group included offenders who had been released on SR, were suspended on this release and then later re-released on SR with a residency condition.

Approximately 25% of offenders released with a residency condition were Aboriginal offenders, which is somewhat higher than their 15% representation in the offender population. Only 17 females were released on SR with a residency condition.

The use of residency among the pre-amendment and post-amendment referred groups was lower in the Quebec region and higher in the Prairie region relative to their representation in the offender population. The use of residency among the SR suspended-residency group was higher in the Quebec region relative to its representation in the offender population, while the SR-residency group was distributed in accordance with the regional proportions of the offender population.

Most offenders released on statutory release with residency were from medium or maximum security institutions, with the referred groups having approximately equal proportions of offenders released from each level of security institutions. In contrast, the SR-residency group had a higher proportion released from maximum security institutions (59%) while the majority (65%) of SR suspended and re-released on residency offenders were released from medium security institutions.

Overall, offenders released on statutory release with residency had average sentence lengths that ranged from 4.5 to 7.1 years. The pre-amendment referred offenders had slightly longer sentences than the post-amendment referred offenders, while the SR-residency group had the shortest average sentence length of approximately four and a half years.

In terms of offence history, the SR suspended-residency group had the largest number of current offences at approximately eight, while the SR-residency group had an average of approximately five offences. Interestingly, post-amendment referred offenders had more offences (5.6) than the pre-amendment referred offenders (4.2) in the current sentence.

The most serious offence for a large portion of the referred and statutory release groups was a sexual offence, although robbery was the second most common for the SR-residency group. In contrast, the most serious offence of the SR suspended-residency group was most often robbery, with assault following closely. Overall, between 92% and 95% of offenders released on statutory release with residency had a violent offence (Schedule I) in the current sentence; while between 4% and 11% had a serious drug offence (Schedule II).

The time from release on residency to the end of the sentence ranged from a median of 4 months for the suspended group to 12 months for the SR-residency group. The pre-amendment referred group had a longer period of time between the release on residency and the end of their sentence than the post-amendment referred offenders.

Following release, an overall readmission rate of 55% within one year was observed for the study groups, higher than that observed among all SR releases (approximately 40%). However, the majority of these readmission's were due to a revocation, not a new offence. New offence rates ranged from 5% for the pre-amendment referred group to 17% for the SR suspended-residency group. The high readmission rate may indicate that residency had little impact on recidivism following release. However, the use of more appropriate control groups in a subsequent study will evaluate this finding more completely.

TABLE OF CONTENTS

LIST OF TABLES

INTRODUCTION

The humane treatment of offenders requires that they be managed in the least restrictive setting while maintaining the security and safety of the community. Legislation in Canada allows offenders to be released on full parole after serving one-third of their sentence, and to be released on day parole six months before that date. In addition, legislation requires that most offenders not released on parole, be released to community supervision after serving two-thirds of their sentence in custody; this is called statutory release. In exceptional cases, offenders can be detained, or held in custody, beyond two-thirds of their sentence if it is believed they are likely to commit a new violent offence that will result in serious harm to their victims. However, there are some offenders who are not a serious enough threat to detain, but who, at the same time, are not ready for full release into the community.

To address this issue, changes to the Corrections and Conditional Release Act (1995) permitted the National Parole Board to add a condition of residency to the statutory release of offenders. The residency condition requires that the offender reside at a designated halfway house for a portion of the remaining sentence to ensure safe reintegration into the community.

Prior to January 1996, only offenders considered for detention beyond their statutory release date could have a residency condition added to their release. This was used in cases where the National Parole Board felt that the offender could be managed in the community, provided that additional resources were available such as those from a halfway house. Rather than be detained in custody, offenders were released to a halfway house where their behaviour could be monitored more closely, thereby reducing the risk to society. These offenders were required to reside in Community Correctional Centres (CCC) operated by the Correctional Service of Canada or at other special federal institutions.

The use of a residency condition should have an impact on the reintegration of offenders into the community if additional programs and treatment resources are available. Residing at a halfway house provides the offender with a transition period from full custody to partial community living and subsequently full community living. Also, the most likely time for failure is the first six to twelve months after release from prison. A follow-up study of both paroled offenders and those released at their statutory release date (Grant, 1996) indicates a residency condition would normally be in place during this critical period of time.

In addition, day parole, which is a release to a halfway house prior to statutory release, has been shown to be associated with reduced recidivism (Grant, Motiuk, Brunet, Lefebvre & Couturier, 1996; Grant & Gal, 1998; Grant & Gillis, 1998). These studies have shown that offenders who successfully completed a period of day parole at a halfway house were less likely to be readmitted to prison following full release (full parole or statutory release) than those who were unsuccessful during the day parole period. Offenders in these studies were frequently released on day parole within six months of their statutory release date making them comparable to offenders released on statutory release with a residency condition. However, day parole represents an early release to the community whereas statutory release with residency represents a delay in full community release and therefore the reaction of offenders to the two different types of release may be different.

The amended CCRA (1995) specifies strict regulations under which the residency condition should be allowed, namely that "in the absence of such a condition, the offender will present an undue risk to society by committing an offence listed in Schedule I1 [violent offences] before the expiration of the offender's sentence". In order to provide proof of the need for residency, the Case Management Manual (1998) specifies criteria to be analyzed which include variables found to increase or decrease the offender's risk of committing a violent offence such as potential stressors, psychiatric or psychological evidence, and previous violent behaviour.

1 The CCRA contains two schedules, or lists of offences. Offences listed on Schedule I are violent offences while those listed on Schedule II are drug related offences.

The purpose of this report is to provide a description of how the amendment to the CCRA that came into effect in January 1996 affected the use of the residency condition.

Specifically, the study will address the following questions:

  1. Has the use of the residency condition for referred offenders changed following the amendment to the CCRA?
  2. Has the residency condition been used with offenders released at their statutory release date?
  3. What are the demographic characteristics of offenders released with a residency condition?
  4. Is residency used equally across all regions?
  5. What are the sentence characteristics of offenders released with a residency condition?
  6. For offenders released with a residency condition, what types of offences have they committed in the past and what is the extent of their criminal activity? Have these characteristics changed from before the amendment to after?
  7. What percentage of offenders released with a residency condition is readmitted to prison within one year of their release?

Comparison Groups

In order to be able to make comparisons between the pre-amendment and post-amendment period, four groups were created as defined below.

Prior to February 1996, residency conditions were only imposed on offenders who were referred for detention or had a detention review. For offenders referred for detention, or whose detention was subsequently reviewed, the National Parole Board could decide to detain offenders, release them on statutory release, or release them with a residency condition. Of the offenders referred/reviewed for detention only a small percentage (approximately 10%) were granted release with a residency condition, so all referred or reviewed offenders from April 1, 1988 to January 31, 1996 were included in the pre-amendment referred group.

The second group in the study consisted of offenders' referred/reviewed for detention after the change in the amendment (January 1996) and whom had a residency condition imposed at the time of their release. This group is referred to as the post-amendment referred group.

After the January 1996 amendment to the CCRA, a residency condition could be imposed on offenders released at the statutory release date if the National Parole Board believed that the condition would make the risk the offender posed to the community more manageable. This group is referred to as the statutory release-residency group since the condition was imposed at the time of their statutory release, and there was no referral for detention.

If an offender fails to meet the conditions of the statutory release and is suspended while on statutory release, the National Parole Board may decide to impose a residency condition either as part of the decision to cancel the suspension or as a condition after the offender has served some additional time in custody. This fourth group is referred to as statutory release suspended-residency.

In summary, the four groups used in comparisons are:

  1. pre-amendment referred
  2. post-amendment referred
  3. statutory release-residency
  4. statutory release suspended-residency

To determine how the changes to the legislation affected the use of residency for referred cases; comparisons will be made between the pre-amendment and post-amendment groups. Comparisons between the referred groups and the statutory release-residency groups (SR-residency, SR suspended-residency) will be made to determine how the characteristics of the residency population changed following the amendment.

METHODOLOGY

Population

All offenders who were released between April 1, 1988 and January 31, 1998 who had a residency condition imposed at the time of their statutory release were included in the study. Offenders were identified using the National Parole Board's decision and conditions databases. For the purposes of the study an offender was only counted once so multiple instances of residency conditions were ignored. Analyses indicated that most (91%) multiple decisions for residency were probably extensions of the original order (i.e., occurred within 12 months of the original residency order) and probably did not represent new conditions being imposed on an offender after release from a halfway house (for more specific information see Table A-1). For those with multiple instances of residency conditions imposed on their SR, their first release was considered.

Groups

The four comparison groups used in the study were defined as follows:

  1. Pre-amendment referred: All offenders referred for detention or who had their detention reviewed and were granted a release with a residency condition prior to January 31, 1996.
  2. Post-amendment referred: All offenders referred for detention or who had their detention reviewed and were granted a release with a residency condition after January 31, 1996.
  3. Statutory release-residency (SR-residency): All offenders after January 31, 1996 who were released on SR with a residency condition and who did not have a previous SR release which was followed by a readmission to prison, and were not referred for detention.
  4. Statutory release suspended-residency (SR suspended-residency): All offenders who were released on SR after January 31, 1996 with a residency condition, but who had a previous statutory release for which they were suspended during the same sentence.

In order to determine group membership, the most recent decision (detention referral or review of detention, suspension on SR) prior to the statutory release with residency decision was considered.

The number of offenders in each of the comparison groups is presented in Table 1.

Table 1: Number of offenders in each comparison group
Pre-amendment Post-amendment
Referred for detention Referred for detention SR-residency SR suspended-residency
Number 298 101 997 473
Percentage of post-amendment sample 6% 63% 30%

Analysis Variables

Demographic variables in the analyses include age, gender, marital status and race. Race is divided into three groups, Aboriginal, Caucasian and Other. The five Correctional Service of Canada regions are used make geographical comparisons: Atlantic, Quebec, Ontario, Prairie and Pacific regions.

The number and types of offences are compared across the groups, but only offences resulting in a federal sentence are included. Type of offence is compared by classifying offences as violent or drug offences based on the offence Schedules of the Corrections and Conditional Release Act (1992). In addition, more detailed comparisons are performed by classifying offences as attempted murder/ manslaughter, sex offences, assault, robbery, other violent offences, and non-violent offences. Sentence length and time to the end of the offenders' sentences from the time of release to the community are also compared.

The number of offenders returned to custody and the reasons for their return, within one year of release are used as measures of recidivism. However, additional comparison groups are needed to properly evaluate the recidivism rates and this will be included in a future study.

Data Sources

Data for the study came from four main sources, drawn from the Offender Management System (OMS) maintained by the Correctional Service of Canada. OMS is an administrative data system used to manage offender movements and case file information. From OMS it is possible to create independent, fixed databases that are used for analyses. For the current study, two databases provide data on all National Parole Board decisions which track decisions following referral for detention, the imposition of conditions following a detention hearing (pre-amendment), the imposition of conditions at the time of review for statutory release and suspensions, and subsequent suspension cancellations or revocations. Additional databases were used to determine release dates, admissions (or readmission's), offences, and descriptive data.

Data Analyses

Data analyses were conducted using the Statistical Analysis System (SAS) version 6.12 (SAS Inc., 1991; SAS Inc., 1997). Most comparisons are made in contingency tables. Whenever possible, means are used. For all comparisons, statistical tests are not needed as comparisons are made across populations of offenders, not samples.

RESULTS

The use of the residency condition increased dramatically after January 1996 as shown in Table 2. Prior to 1996 there were about 40 releases with a residency condition each year. The number increased to over 800 in the first full year that the condition could be imposed for statutory release, and remained at about 800 in the following year.

Table 2: Frequency of use of residency condition by fiscal year.
Fiscal Year Frequency
1988-1989 35
1989-1990 35
1990-1991 29
1991-1992 58
1992-1993 41
1993-1994 42
1994-1995 28
1995-1996 91
1996-1997 833
1997-19981 800
1 This value is an estimate of the annual number based on the releases for the 10 months of April to January. There were a total of 676 releases in this period.

Prior to the amendment, all residency conditions were for offenders who had been referred for detention. However, in 1996-1997 and 1997-1998 only about 6% to 7% of the residency conditions were imposed on offenders who had been referred for detention or had a subsequent detention review. However, there was no substantial change in the rate of residency per detention referral or review from pre-amendment (prior to January 31, 1996) to post-amendment (after January 31, 1996) where the rate ranged from approximately 10% to 15%.

After the amendment to the CCRA, the most common use (63%) of the residency condition was for offenders who were released at their statutory release (SR) date. Approximately one-third of the residency conditions were imposed on offenders who had been released on SR, but who had had their SR suspended for failing to meet one or more conditions of their statutory release. Residency became a condition of their subsequent release to the community.

Demographic characteristics

Demographic characteristics of the offenders released with a residency condition are presented in Table 3. The racial distribution of offenders referred for detention and released with a residency condition did not change significantly from the pre- to the post-amendment period although there was an increase in the percentage of Aboriginal offenders from 28% to 34%. Given that in 1996-97 Aboriginal offenders accounted for about 15% of the offender population (Correctional Service of Canada, 1997), they appear to be over-represented in the sample. However, this is partly the result of a higher rate of referrals for detention for Aboriginal offenders who accounted for approximately 26% of referrals from 1989-90 to 1993-94 (Grant, 1996).

Approximately 70% of the offenders who had a residency condition imposed as part of their statutory release (SR-residency and SR suspended-residency groups) were Caucasian with Aboriginal offenders accounting for about one-quarter of the cases. Again, relative to the inmate population, Aboriginal offenders are slightly over-represented (24% vs. 15%) in these residency groups as compared to the offender population.

Table 3: Race, gender and marital status for each group
Pre-amendment Post-amendment
Referred for detention Referred for detention SR-residency SR suspended-residency
Race
Caucasian 67% (193) 63% (63) 70% (693) 70% (329)
Aboriginal 28% (81) 34% (34) 24% (235) 25% (120)
Other 5% (14) 3% (3) 7% (66) 4% (21)
Number of cases 287 100 994 470
Gender
Male 99% (284) 99% (99) 99% (990) 98% (465)
Female 1%(3) 1%(1) 0.5%(5) 2%(8)
Number of cases 287 100 995 473
Marital Status
Married/ Common-law 37% (107) 41% (41) 37% (365) 42% (196)
Divorced/ Separated/ Widowed 9% (26) 8% (8) 12% (120) 9% (44)
Single 53% (152) 51% (51) 51% (501) 49% (231)
Number of cases 285 100 986 471

Almost all offenders who were released with a residency condition were male. In the post-amendment period only fourteen female offenders were released on SR with a residency condition.

There were no meaningful differences in marital status among the groups. Overall, approximately 50% of offenders released with a residency condition were single while about 40% were married or living common law and 10% were divorced, separated or widowed.

Offenders with a residency condition imposed after a detention referral are older at the time of the decision, by approximately one year, in the post-amendment period (34 years vs. 35 years) as compared to the pre-amendment period (see Table 4). The change is a result of a slight reduction in percentage of offenders in the youngest age group and an increase in the percentage of offenders in the 35 to 49 age group. Offenders with a residency condition imposed at SR or after a suspension from SR, tend to be about one year younger than post-amendment referred offenders.

Table 4 also shows the age at admission for offenders. Offenders referred for detention prior to the amendment tend to be slightly older at admission (31 vs. 30) than those referred after the amendment. The SR-residency group had an average age of 30 years, but the SR suspended-residency group was slightly younger when admitted, with a mean age of 28 years.

Table 4: Mean age and age distribution at residency decision and at admission for each group.
Pre-amendment Post-amendment
Referred for detention Referred for detention SR-residency SR suspended-residency
Age at residency decision
24 and younger 11% (32) 7% (7) 17% (171) 11% (52)
25 to 34 53% (151) 48% (48) 41% (406) 49% (231)
35 to 49 30% (86) 37% (37) 35% (353) 35% (165)
50 and older 6% (18) 8% (8) 6% (65) 5% (25)
Mean 34.1 35.4 34.4 34.3
Number of cases 287 100 995 473
Age at first admission
24 and younger 31% (88) 41% (41) 37% (366) 44% (209)
25 to 34 41% (118) 36% (36) 39% (390) 39% (186)
35 to 49 23% (67) 19% (19) 20% (198) 15% (69)
50 and older 5% (14) 4% (4) 4% (41) 2% (9)
Mean 30.9 29.5 30.1 28.1
Number of cases 287 100 995 473

Region

As compared with the regional representation in the offender population, the use of a residency condition with referred offenders is the most common in the Prairie and Pacific regions as shown in Table 5. Relative to their proportion of the offender population, the Quebec and Ontario regions appear to have used the residency option less frequently. Use of residency with referred offenders appears to have decreased in the Quebec region, but has increased (almost doubled) in the Pacific region after the amendment.

Table 5: Regional distribution of residency groups.
Region Pre-amendment Post-amendment Distribution of Inmate Population
(March 31, 1997)
Referred for detention Referred for detention SR-residency SR suspended-residency
Atlantic 7% (18) 10% (10) 10% (99) 5% (22) 9.6%
Quebec 19% (50) 14% (14) 28% (280) 33% (155) 27.0%
Ontario 23% (61) 19% (19) 22% (217) 19% (90) 25.0%
Prairie 35% (92) 32% (32) 25% (250) 27% (126) 23.8%
Pacific 15% (40) 25% (25) 15% (148) 17% (79) 13.6%
Number of Cases 261 100 994 472)

The use of residency with the statutory release groups (SR-residency and SR suspended-residency groups) appears to be distributed more consistently with the offender populations in the regions compared to its use for referred offenders. The Quebec region was the most frequent user of residency accounting for 28% of the SR-residency group releases and 33% of the residency conditions applied to offenders who had a previous statutory release suspended. Relative to their regional representation, the Quebec, Prairie and Pacific regions were most likely to impose a residency condition on offenders who were suspended following a SR release.

Security level of releasing institution

Similar rates of imposing a residency condition across various security levels of releasing institutions were observed pre to post-amendment. Almost all referred offenders with statutory release with residency were released from either medium or maximum-security institutions. Offenders in the SR-residency group were most likely to be released from maximum-security institutions (59%). However, only 28% of offenders in the SR suspended-residency group were released from maximum-security institutions.

Table 6: Security level of releasing institution
Security level Pre-amendment Post-amendment
Referred for detention Referred for detention SR-residency SR suspended-residency
Minimum 2% (4) 3% (2) 2% (18) 6% (27)
Medium 48% (117) 49% (36) 38% (327) 65% (289)
Maximum 50% (123) 49% (36) 59% (502) 28% (126)
Number of cases 244 74 849 447

Sentence length

Comparing the sentence length of referred offenders in the pre and post-amendment periods, it appears that pre-amendment referred offenders had longer sentences, on average seven years, as compared to the post-amendment period, where the average sentence length for referred offenders was six years (see Table 7).

Offenders released on SR with a residency condition (SR-residency group) had an average sentence length of four years which was lower than the referred offenders' average sentence length, while the SR suspended group had average sentence length of seven years (see Table 7). The distribution of sentence lengths indicates that 20% of the SR suspended-residency group had sentences of ten years or more, while only 7% of the SR-residency group had sentences of this length.

Table 7: Sentence length of offenders released with a residency condition.
Sentence Length Pre-amendment Post-amendment
Referred for detention Referred for detention SR-residency SR suspended-residency
2 to 3 years 14% (38) 17% (17) 36% (356) 21% (100)
3 to 4 years 14% (39) 19% (19) 23% (225) 16% (75)
4 to 5 years 12% (33) 9% (9) 13% (134) 13% (60)
5 to 10 years 38% (107) 39% (39) 21% (214) 30% (141)
More than 10 years 22% (63) 17% (17) 7% (68) 20% (97)
Mean (years) 7.1 6.3 4.5 6.9
Number of cases 280 101 997 473

Number of offences in current sentence

The number of offences the SR with residency offenders were convicted of during the current sentence is presented in Table 8. The number of offences includes all offences with sentences requiring federal custody during the sentence prior to the residency decision, not just the offences the offender was admitted with. The average number of offences for referred offenders with a residency condition was four in the pre-amendment period and six in the post-amendment period. For SR-residency offenders the average number of offences was five, while the SR suspended-residency group had an average of eight. The larger number of offences for previously suspended offenders may be due to additional offences that resulted in the suspension of the previous SR. Table 8 also presents the distribution of the number of offences for offenders and revealed few differences between groups. However, the SR suspended-residency group had fewer offenders with one offence and more offenders with more than ten offences as compared to the other groups. Similar findings were observed when considering offences from both the current and previous sentences (see Table A-2).

Table 8: Number offences during current sentence.
Number of offences Pre-amendment Post-amendment
Referred for detention Referred for detention SR-residency SR suspended-residency
One 21% (62) 16% (16) 18% (179) 7% (33)
Two to three 37% (110) 33% (33) 29% (293) 19% (89)
Four to five 17% (49) 19% (19) 22% (215) 20% (93)
Six to ten 17% (51) 23% (23) 21% (211) 30% (142)
More than ten 7% (21) 10% (10) 10% (99) 24% (116)
Mean 4.2 5.6 5.1 8.2
Number of cases 293 101 997 473

Type of offence

The types of offences committed by offenders during their current sentences are presented in Table 9. Offenders with multiple offences are counted in each offence type so column percentages do not sum to 100%.

Referred offenders in the pre-amendment period were more likely to have been convicted of a sexual offence than those in the post-amendment period (50% vs. 43%). In addition, a higher percentage of referred offenders had committed a robbery in the post-amendment period (32%) than in the pre-amendment period (20%).

Table 9: Percentage of offenders in each group with convictions for selected types of offences.
Offence type Pre-amendment Post-amendment
Referred for detention Referred for detention SR-residency SR suspended-residency
Attempted Murder/ Manslaughter 8% (24) 10% (10) 4% (39) 6% (27)
Sex offences 50% (146) 43% (43) 29% (291) 12% (58)
Assault 36% (106) 35% (35) 28% (282) 38% (179)
Robbery 20% (59) 32% (32) 37% (368) 58% (276)
Other violent offences 12% (36) 6% (6) 11% (108) 9% (41)
Drug offences 7% (21) 8% (8) 12% (125) 18% (86)
Number of cases 293 101 997 473

Note: An offender may have been convicted of more than one type of offence and all different type convictions are included in the table. Therefore the offence categories are not mutually exclusive. Non-violent offences are not included in the table.

The SR-residency and SR suspended-residency groups have lower rates of attempted murder/manslaughter offences (5% vs. 9%) and sex offences (23% vs. 48%) than the referred groups, but have higher rates of robbery (44% vs. 23%) and drug offences (14% vs. 7%). While between one-fifth and one-third of referred offenders had a robbery conviction, almost two-thirds of the SR suspended-residency group had a robbery conviction during this sentence.

Comparisons of offence types were also made for all current and previous convictions with a federal sentence and the results were similar to those presented above for convictions during the current sentence. Table A-3 presents these results. Results presented in Table 10 show the percentage of offenders with violent offences (Schedule I) and drug offences (Schedule II). Almost all, 92% to 95%, of the offenders released with a residency condition had been convicted of at least one violent offence and the percentage varied little across the groups. Between 4% and 11% of the residency offenders had a drug conviction with the overall percentage being approximately 9%. For more detailed information see Table A-4, Table A-5, Table A-6 and Table A-7 which present the number of scheduled offences in current or current and previous sentences.

Table 10: Percentage of offenders with violent offences (Schedule I) and drug (Schedule II) offences.
Offence type Pre-amendment Post-amendment
Referred for detention Referred for detention SR-residency SR suspended-residency
Violent offences (Schedule I) 95% (283) 94% (95) 92% (918) 92% (436)
Drug offences (Schedule II) 4% (13) 5% (5) 10% (101) 11% (54)
Number of cases 298 101 997 473

Another way of looking at offence type is to determine the most serious offence that offenders committed and compare percentages across groups. Determining the ranking of most serious offence is somewhat arbitrary, but in this study it is based on the perceived level of harm that an offence might cause to a victim. The order of seriousness used is:

  1. Attempted murder/manslaughter2
  2. Sex offence
  3. Assault
  4. Robbery
  5. Other violent
  6. Non-violent
2 Murder (first or second degree) is not included in the offences because murder has a mandatory life sentence whereby a statutory release date is not available.

Once an offender is placed into a more serious offence category they cannot appear in a less serious offence category. Results of these analyses are presented in Table 11, while Table A-8 presents the most serious offence type of offenders in their current and previous federal sentences.

For referred offenders, sex offences are the most serious offence with 48% having "committed" a sex offence. For SR-residency offenders, sex offences are also the most common serious offence (29%) with robbery following very closely at 27%. This result may be contrasted with the SR suspended-residency group for whom robbery was the most serious offence in 37% of the cases and only 12% had sex offences as the most serious offence. In addition, assault was the most serious offence for between one-quarter and one-third of offenders within the four groups.

In order to evaluate whether the changing of the order of the assault and robbery offences, in terms of relative seriousness, affected the findings, another analysis was performed where the assault and robbery offences were reversed in order. As can be seen in Table A-9, this had very little effect on the distribution of most serious offence.

Table 11: Most serious offence type in current sentence.
Offence type Pre-amendment Post-amendment
Referred for detention Referred for detention SR-residency SR suspended-residency
Attempted Murder/ Manslaughter 8% (24) 10% (10) 4% (39) 6% (27)
Sex offences 48% (143) 43% (43) 29% (291) 12% (57)
Assault 25% (76) 27% (27) 23% (232) 33% (154)
Robbery 7% (21) 10% (10) 27% (265) 37% (176)
Other violent offences 6% (18) 2% (2) 6% (56) 2% (10)
Non-violent offences 5% (16) 9% (9) 11% (114) 10% (49)
Number of cases 298 101 997 473

Time to end of sentence

The length of time from release with a residency condition to the end of the sentence is presented in Table 12.

These results demonstrate the amount of time the offender had to serve after the residency condition was imposed. A short period of time could mean that the remainder of the sentence is served with the residency condition while longer periods may indicate that the offender will be released into the community after the residency period.

Table 12: Time from release with residency condition to the end of the sentence
Time from residency to end of sentence Pre-amendment Post-amendment
Referred for detention Referred for detention SR-residency SR suspended-residency
Less than 1 year 53% (114) 62% (61) 46% (427) 85% (364)
1 to 2 years 31% (67) 29% (28) 43% (396) 9% (40)
2 or more years 15% (33) 9% (9) 11% (107) 6% (24)
Mean (months) 14.5 11.9 15.0 7.1
Median (months) 11.4 8.6 12.0 3.9
Number of cases 214 98 930 428

Comparing time to serve after release may be most effectively done using the median rather than the mean, although both are presented in Table 12.

The median is the point in the distribution of sentence lengths at which 50% of cases have shorter sentences while 50% of cases have longer sentence lengths. The median is influenced less by cases with very long sentences.

Offenders referred for detention and released prior to the amendment had a median of 11 months to serve before the end of their sentence while post-amendment referred offenders had only 9 months to serve. For SR-residency offenders, the median of time to serve was 12 months while SR suspended-residency offenders had a median of only 4 months to serve before the end of their sentence.

Examination of the distribution of time to serve indicates that 85% of the SR suspended-residency offenders had less than one year to serve when they were released with the residency condition. The majority of offenders released with a residency condition had less than two years to serve in their sentence when they were released with a residency condition, regardless of group membership.

Recidivism

Recidivism was measured from the date of release on SR with a residency condition. In addition to measuring any readmission to a federal institution, results are presented for three types of readmission: revocation without an offence, revocation with an offence, and readmission with a new offence. In general, revocation with an offence is an underestimate of the rate of new offending since new charges may not be processed when the National Parole Board revokes the conditional release. Readmission with a new offence occurs after offenders have completed their sentence and have committed a new crime and returned to a federal institution. This is contrasted with revocation with a new offence where the new offence occurred during the current sentence (i.e., prior to the expiration of the sentence). The follow-up consisted of a fixed one-year period. Of the post-amendment offenders, only those released in the first year after the amendment was enacted were included in the follow-up.

The recidivism rates for each of the groups are presented in Table 13. Overall, approximately 55% of the offenders released with a residency condition were readmitted within one year of their release. The lowest readmission rate was for referred offenders released prior to the amendment to the legislation with only 47% readmitted within one year. These rates of readmission are slightly higher than the 40% readmission rate found for all offenders released at their statutory release date (Grant, 1996).

Table 13: Readmission rates within one year of release and reasons for readmission
Pre-amendment Post-amendment
Referred for detention Referred for detention SR-residency SR suspended-residency
Not readmitted 53% (142) 47% (27) 42% (214) 43% (114)
Readmitted (overall) 47% (128) 54% (30) 58% (291) 57% (149)
Revocation without offence 40% (107) 42% (24) 48% (241) 37% (97)
Revocation with offence 4% (10) 10% (6) 9% (47) 11% (30)
Readmission with new offence1 1% (4) 0% (0) 0.2% (1) 6% (17)
Other2 3% (7) 0% (0) 0.4% (3) 2% (5)
Number of cases 270 57 505 263
1 Refers to readmission to federal custody due to a new offence following expiration of the current sentence
2 Other includes to conditional release inoperative and termination of conditional release.

Analysis of the reasons for return to custody is also presented in Table 13. These results indicate that for the referred groups, the readmission rates are 47% and 54%, with the higher rate for the post-amendment releases. The majority of these readmission's (40% to 42%) were for revocations of conditional release without a new offence. Between 5% and 10% of the referred for detention releases were returned to custody with a new offence, either as a revocation (in the current sentence) or as a readmission (in a subsequent sentence).

For the SR-residency group, 9% were returned to custody with a new offence while 17% of the SR suspension-residency group were returned to custody with a new offence either as a revocation or a readmission.

SUMMARY AND DISCUSSION

In the introduction, a set of seven questions was proposed. The questions and the answers follow:

  1. Has the use of the residency condition for referred offenders changed following the amendment to the CCRA?
    Prior to the amendment to the CCRA in 1996 which allowed a residency condition to be applied to statutory releases (SR), only about 40 offenders a year were released with a residency condition following a detention review. This number did not change following the amendment. However, after the amendment, only about 6% of residency releases were for referred offenders.

  2. Has the residency condition been used with offenders released at their statutory release date?
    Post amendment, approximately 800 offenders were released each year with a residency condition and most residency conditions (63%) were applied to offenders being released on SR or after their SR has been suspended (30%) and they were subsequently re-released with a residency condition.

  3. What are the demographic characteristics of offenders released with a residency condition?
    Very few female offenders have been released with a residency condition. About half of the offenders released with a residency condition were single. Referred offenders in the post-amendment period were approximately one year older than those in the pre-amendment period. Offenders released on statutory release (SR-residency and SR suspended-residency groups) tended to be about one year younger than referred offenders in the post-amendment period, but similar in age to the pre-amendment referred offenders.
    Of the offenders referred for detention and given a residency condition, 30% were Aboriginal offenders in both the pre- and post-amendment periods. However, this is not an indication of bias in the use of the residency condition since 26% (Grant, 1996) of all referrals for detention are Aboriginal offenders. Aboriginal offenders account for 24% of the SR-residency group and 25% of the SR suspended-residency group and this is somewhat higher than their 15% representation in the offender population. However, Aboriginal offenders are also more likely to be released on SR than on full parole so the higher percentage for Aboriginal offenders may reflect a higher rate of SR release for Aboriginal offenders.
    Pre- to post-amendment, referred offenders released with a residency condition were released from medium and maximum-security institutions at approximately the same rate. While 59% of the SR-residency group were released from a maximum-security institution, offenders suspended after SR and then re-released with a residency condition were more likely to be released from a medium security institution (65%).

  4. Is residency used equally across all regions?
    As compared with their representation in the offender population, residency is used with referred offenders most frequently in the Atlantic, Prairie and Pacific regions and is used less frequently in the other two regions. However, for the SR-residency group, the use of the residency condition is used at a rate proportional to regional representations of the offender population. The use of residency for offenders suspended on SR is highest in the Quebec and Prairie regions, relative to their representation in the offender population.

  5. What are the sentence characteristics of offenders released with a residency condition?
    Analysis of the sentence length indicates that referred offenders with a residency condition released after the CCRA amendment had shorter sentences than those released in the pre-amendment period, by approximately one year. While offenders released with residency at their SR date had an average sentence length of four years, the sentence length of those re-released after a suspension was seven years.

  6. For offenders released with a residency condition, what types of offences have they committed in the past and what is the extent of their criminal activity? Have these characteristics changed from before the amendment to after?
    The number of offences averaged between four and six for three of the four comparison groups, but the number was greater for the SR suspended-residency group with an average of eight convictions during the current sentence. Comparisons between the groups for type of offence indicated that almost all offenders in the residency group (approximately 94%) had been convicted of at least one violent offence (Schedule I) and approximately 10% had a drug offence (Schedule II). More detailed analyses indicated that referred offenders were most likely to have convictions for sexual offences (approximately 48%) while SR offenders (SR-residency or SR suspended-residency groups) were more most likely to have been convicted of a robbery offence (37% to 58%). Interestingly, more than half (58%) of the SR suspended-residency offenders had been convicted of a robbery offence.

  7. What percentage of offenders released with a residency condition is readmitted to prison within one year of their release?
    Offenders released on SR with residency had an overall readmission rate of approximately 55% within one year of release. Pre-amendment referred offenders' readmission rate was 47% but this rose to 54% for the post-amendment referred group. The SR-residency and SR suspended-residency groups had return rates of approximately 57%. However, it is important to note that the majority of these readmission's were due to revocations without new offences. The new offence rate ranged from 5% for the pre-amendment referred offenders to 17% for the suspended then re-released offenders. The new offence rate of referred offenders doubled from pre-amendment (5%) to post-amendment (10%).

Discussion

The amendment to the CCRA allowing the residency condition to be imposed on offenders released at their statutory release date resulted in a dramatic increase in the use of this option. In addition, the characteristics of offenders released with the residency condition changed with an increase in offenders convicted of robbery and a decrease in the percentage of offender released with a sex offence. Offenders given a residency condition following the suspension of SR appear to represent a higher risk group of offenders who were younger at their first federal admission, had longer sentences, more convictions, and were most likely to have a robbery conviction.

The impact of residency on outcome appears to be limited given that 55% of releases with a residency condition were returned to custody within a year, a rate higher than the 40% rate of the population of all SR releases (Grant, 1996). However, more appropriate comparison groups will need to be constructed to further understand the recidivism rate. It may be that those offenders released with a residency condition are the most likely to be returned to custody following conditional release and their base rate of return could be higher than 50%. A future study will provide these additional comparison groups.

If the recidivism result is supported with the inclusion of additional comparison groups, it would be surprising given that day parole has been shown to have an effect on release outcome (Grant & Gillis, 1998), and given the potential for the residency period to act as a transition period from custody to full community living. Since the first six to twelve months of release is the most likely time for readmission (Grant, 1996), the additional support afforded by a halfway house would be expected to reduce the likelihood of return. The preceding reasoning assumes that the halfway houses provide programming opportunities for residency offenders which targets their criminogenic needs and the transition from prison to street life. Research in the area of intensive supervision has found that increased surveillance alone is not sufficient to lower readmission rates, and may, in fact, increase the technical violation and return to custody rates (Petersilia and Turner, 1993).

If support structures are not available then it may be that residency is of little benefit in assisting offenders to reintegrate. Future research will need to look at what kinds of programs and community support services are available for residency offenders and what programs they actually participate in.

The results indicate that a substantial number of offenders with relatively serious offences have been released to halfway houses on SR. Given that previous releases to halfway houses were mostly day parole releases, and many of these day parole offenders would have had relatively short and less violent criminal histories, it seems likely that there has been an impact of the residency releases on halfway house operations. A future research project will need to investigate the impact of the residency releases on halfway house operations, and focus particularly on the impact on the mix of offender types in halfway houses after the SR with residency condition was made more generally available.

While there is evidence to suggest that a residency condition could have an impact on release outcome, there is insufficient evidence available in this study to demonstrate a positive impact. Additional research efforts will be needed to determine if the use of a residency condition is having an impact and to determine if modifications in the way the residency condition is used could be enhanced to increase the likelihood of success for offenders after their release to the community.

REFERENCES

Amendments to the Corrections and Conditional Release Act, Statutes of Canada, c. 42, s. 46, 48, (1995).

Case Management Manual (1998). Ottawa, Canada: Correctional Service of Canada.

Corrections and Conditional Release Act, RCS, C-20, (1992)

Corrections and Conditional Release Act, August 1997 Edition, RSC, C-20, (1997).

Correctional Service of Canada (1997) Basic facts about corrections in Canada. Ottawa, Canada: Correctional Service of Canada.

Grant, B.A. (1996) Inmates referred for detention (1989-90 to 1993-94): A comparative analysis (R-45). Ottawa, Canada: Correctional Service of Canada.

Grant, B.A., & Gal, M. (1998) Case Management Preparation for Release and Day Parole Outcome (R-63). Ottawa, Canada: Correctional Service of Canada.

Grant, B.A. & Gillis, C. (1998). Day parole outcome, criminal history and other predictors of successful sentence completion. Ottawa, Canada: Correctional Service of Canada.

Grant, B.A., Motiuk, L., Brunet, L., Lefèbvre, L. and Couturier, P. (1996) Day parole program review: Case Management Predictors of Outcome (R-52). Ottawa, Canada: Correctional Service of Canada.

Petersilia, J. and Turner, S. (1993). Intensive Probation and Parole. In M. Tonry (Ed.), Crime and Justice, A Review of the Research, Volume 17. Chicago: The University of Chicago Press.

SAS Institute Inc. (1991). SAS Language and Procedure, Usage 2, Version 6, First Edition. Cary, North Carolina: SAS Institute Inc.

SAS Institute Inc. (1997). Statistical Analysis System (SAS), Version 6.12. Cary, North Carolina: SAS Institute Inc.

APPENDIX A

Table A-1: Period between first and second residency decision for offenders with more than one decision
Period of time between decisions Frequency Percent
Less than 3 months 226 31.5
3 to 6 months 205 28.6
6 to 12 months 224 31.2
More than 12 months 63 8.8
Number of cases 718 100.0
Table A-2: Number offences during current and previous sentences.
Number of offences Pre-amendment Post-amendment
Referred for detention Referred for detention SR-residency SR suspended-residency
One 11% (32) 14% (14) 12% (118) 3% (16)
Two to three 31% (90) 15% (15) 20% (199) 13% (63)
Four to five 19% (57) 15% (15) 20% (198) 14% (67)
Six to ten 26% (75) 32% (32) 24% (235) 28% (132)
More than ten 13% (39) 25% (25) 25% (247) 41% (195)
Mean 5.9 7.9 7.8 11.1
Number of cases 293 101 997 473
Table A-3: Percentage of offenders in each group with convictions for selected types of offences in current and previous sentences.
Offence type Pre-amendment Post-amendment
Referred for detention Referred for detention SR-residency SR suspended-residency
Murder/ Attempted Murder/ Manslaughter 10% (29) 12% (12) 5% (48) 7% (32)
Sex offences 51% (153) 43% (43) 31% (314) 16% (74)
Assault 39% (117) 39% (39) 30% (302) 40% (191)
Robbery 25% (75) 37% (37) 42% (420) 64% (302)
Other violent offences 20% (60) 14% (14) 17% (172) 15% (71)
Drug offences 9% (27) 10% (10) 17% (173) 23% (109)
Number of cases 298 101 997 473

Note: An offender may have been convicted of more than one type of offence and all different types convictions are included in the table. Therefore the offence categories are not mutually exclusive. Non-violent offences are not included in the table.

Table A-4: Number of Scheduled (I and II) offences before residency decision on current sentence
Number of offences Pre-amendment Post-amendment
Referred for detention Referred for detention SR-residency SR suspended-residency
None 4% (11) 6% (6) 10% (99) 8% (40)
One 40% (116) 30% (30) 38% (381) 30% (144)
Two to three 43% (125) 45% (46) 37% (371) 35% (166)
Four to five 10% (29) 12% (12) 10% (99) 15% (69)
Six to ten 4% (12) 6% (6) 4% (43) 10% (48)
More than ten 0% (0) 1% (1) 0.5% (4) 1% (6)
Mean 2.1 2.4 2.0 2.7
Number of cases 293 101 997 473
Table A-5: Number of Scheduled (I and II) offences before residency decision date on current and previous sentences
Number of offences Pre-amendment Post-amendment
Referred for detention Referred for detention SR-residency SR suspended-residency
None 4% (13) 5% (5) 6% (64) 7% (31)
One 23% (68) 21% (21) 27% (274) 19% (90)
Two to three 50% (149) 40% (40) 40% (397) 34% (162)
Four to five 15% (45) 22% (22) 17% (167) 23% (107)
Six to ten 8% (23) 10% (10) 9% (89) 15% (73)
More than ten 0% (0) 3% (3) 1% (6) 2% (10)
Mean 2.7 3.1 2.7 3.1
Number of cases 293 101 997 473
Table A-6: Number of Schedule I offences before residency decision on current sentence
Number of offences Pre-amendment Post-amendment
Referred for detention Referred for detention SR-residency SR suspended-residency
None 5% (15) 7% (7) 12% (117) 10% (47)
One 41% (120) 31% (31) 39% (393) 32% (150)
Two to three 41% (119) 45% (45) 35% (352) 35% (166)
Four to five 10% (28) 11% (11) 9% (91) 13% (63)
Six to ten 4% (11) 6% (6) 4% (40) 9% (41)
More than ten 0% (0) 1% (1) 0.4% (4) 1% (6)
Mean 2.0 2.3 1.9 2.5
Number of cases 293 101 997 473
Table A-7: Number of Schedule I offences before residency decision on current and previous sentences
Number of offences Pre-amendment Post-amendment
Referred for detention Referred for detention SR-residency SR suspended-residency
None 3% (10) 6% (6) 8% (79) 8% (37)
One 24% (71) 21% (21) 29% (292) 21% (98)
Two to three 51% (149) 42% (42) 39% (388) 35% (164)
Four to five 14% (41) 19% (19) 16% (157) 21% (98)
Six to ten 7% (22) 10% (10) 8% (76) 14% (67)
More than ten 0% (0) 3% (3) 0.5% (5) 2% (9)
Mean 2.6 3.0 2.5 3.2
Number of cases 293 101 997 473
Table A-8: Most serious offence type in current and previous sentences
Offence type Pre-amendment Post-amendment
Referred for detention Referred for detention SR-residency SR suspended-residency
Murder/ Attempted Murder/ Manslaughter 10% (29) 12% (12) 5% (48) 7% (32)
Sex offences 50% (148) 43% (43) 31% (312) 15% (71)
Assault 25% (76) 26% (26) 24% (235) 33% (158)
Robbery 6% (19) 10% (10) 27% (271) 36% (170)
Other violent offences 5% (16) 3% (3) 6% (58) 2% (8)
Other 3% (10) 7% (7) 7% (73) 7% (34)
Number of cases 298 101 997 473
Table A-9: Most serious offence type in current sentence (revised version with robbery more serious than assault)
Offence type Pre-amendment Post-amendment
Referred for detention Referred for detention SR-residency SR suspended-residency
Attempted Murder / Manslaughter 8% (24) 10% (10) 4% (39) 6% (27)
Sex offences 48% (143) 43% (43) 29% (291) 12% (57)
Robbery 14% (42) 24% (24) 34% (344) 55% (261)
Assault 18% (55) 13% (13) 15% (153) 15% (69)
Other violent offences 6% (18) 2% (2) 6% (56) 2% (10)
Other 5% (16) 9% (9) 11% (114) 10% (49)
Number of cases 298 101 997 473
Date Modified: 2015-03-05

Top of Page
Important Notices