FORUM on Corrections Research

Federal trends and outcomes in conditional release

Parole is not a new concept in Canada. For almost a century, legislation has been in place to permit release from prison before expiry of sentence. In fact, 1999 will mark 100 years of parole in Canada.

Legislative framework

The evolution of conditional release legislation in Canada is marked by three major milestones. It began in 1899 with the Ticket of Leave Act, which allowed release from prison based on factors related to the offence, the offender's character and the likelihood that the offender would commit another offence.

The Parole Act came into force in 1959, creating the National Parole Board and providing it with the authority to grant release on parole, set the conditions of release, and revoke the release of federal offenders or provincial and territorial offenders in jurisdictions without parole boards.

In 1992, the Corrections and Conditional Release Act (CCRA) replaced the Parole Act (and the Penitentiary Act), and provided a comprehensive framework for corrections and conditional release policy, training and operations.

The CCRA is scheduled for review by a committee of Parliament in 1998. This review, and the government response, are expected to shape corrections and conditional release as Canada enters the new millennium.

Conditional release over the last 20 years

Over the past 20 years, conditional release has played an important role in federal corrections. Each year, 4 of every 10 (37­44%) federal offenders served their sentence in the community on parole or mandatory supervision (MS), which became statutory release (SR). The CCRA replaced MS, which was community release based on remission of sentence for good behaviour in the institution, with SR, which is community release required by law after an offender has served two thirds of his or her sentence.

Closer review of the conditional release population illustrates that each year about 7 of 10 offenders (66­75%) under supervision were in the community on parole, about half of them on full parole. MS/SR accounted for about 3 of every 10 offenders under supervision each year.

Although SR is based on proportion of sentence served, parole is based on review of the case and assessment of risk by the National Parole Board. There are two types of parole:

• day parole, which is intended to prepare offenders for full parole or statutory release -- offenders on day parole must return nightly to a halfway house or a penitentiary; and
• full parole, which is intended to give offenders the opportunity to work and live in the community -- offenders may serve up to two thirds of their sentence in the community under supervision.

In the last 20 years, the Board granted day parole in 6 or 7 of every 10 day parole cases (58­71%) it considered. By comparison, the Board granted full parole in about 3 or 4 for every 10 cases (30­42%) it considered.

In 1992­1993, the CCRA introduced accelerated parole review (APR); first-time, non-violent federal offenders must be directed to release on full parole if the Board determines that they are not likely to reoffend violently before their warrant expires. Rates of directed release (80­89%) for accelerated parole cases have been considerably higher than grant rates (19­27%) for regular full parole cases.

Effectiveness of conditional release

Conditional release trends demonstrate that parole and SR are important strategies for federal offender population management. Annually, about 40% of the federal offender population is supervised in the community, which obviously affects accommodation needs and costs.

For the Board, however, the effectiveness of conditional release is measured against its contribution to public safety. In fact, conditional release is based on the theory that gradual release of offenders enhances community safety. Data on the outcomes of conditional release appear to validate this theory.

The Board assesses the effectiveness of conditional release with three indicators. The first indicator is rate of success, which includes:

• completed releases in which the offender remains in the community for the entire period of supervision -- for full parole and SR the period of supervision would end at warrant expiry; and
• revocations for breach of release conditions (often called technical violations); these revocations are considered positive interventions to reduce risk to the community by preventing further criminal activity.

Failure is defined as any conditional release that results in revocation for a new offence (recidivism).

The second indicator is number of charges for serious offences committed by offenders while on release in the community, by release type, in eight offence categories that emphasize violence against the person:

• murder;
• attempted murder;
• sexual assault;
• major assault;
• hostage-taking;
• unlawful confinement;
• robbery; and
• sensational incidents such as arson.

The third indicator is recidivism for federal offences after offenders' warrants have expired. Factors influencing the behaviour of offenders after warrant expiry are beyond the Board's capacity to manage. Nevertheless, the Board tracks this type of recidivism as an indicator of long-term effectiveness.

Success Rates for conditional release

Information for the period from 1992­1993 to 1996­1997 indicates that success rates have been high for all types of release, but there have also been noteworthy differences by release type.

Success rates for day parole (see Figure 1) have been higher than rates for full parole or SR. In fact, day parole success rates increased from 92% to 96%, including a 17% increase in the proportion of offenders who complete their day parole without return to an institution, and a 13% decrease in the proportion of offenders who are revoked for a breach of condition (e.g., abstain from intoxicants). Annual rates of reoffending by day parolees (recidivism) declined from 8% to 4%, with non-violent reoffending falling from 6% to 3%, and violent reoffending dropping from 2% to 1%.

Figure 1

Success rates for regular full parole (87% to 92%) have approximated rates for day parole (see Figure 2). Successful completions ranged from 62­73%, with highest rates in 1995­1996 and 1996­1997. Rates of revocation for breach of conditions varied from 17­24%. Recidivism rates for regular full parole declined from 13% to 8% during the review period, as non-violent reoffending dropped from 11% to 6% and violent reoffending remained stable at 2%.

Figure 2

APR was introduced in 1992­1993. Information on success for this type of release during 1992­1993 and 1993­1994 should, therefore, be considered with caution, as the program was in its developmental phases. In subsequent years, however, APR matured to full program status.

Table 1

Charges for Serious Offences
Offence	1992-1993	1993-1994	1994-1995	1995-1996	1996-1997	1997-1998
Murder	21	17	16	15	10	9
Attempted murder	17	9	13	15	7	9
Sexual assault	43	43	49	22	31	23
Major assault	34	27	25	21	28	34
Hostage taking	1	4	5	3	1	1
Unlawful confinement	19	9	3	5	4	6
Robbery	88	102	113	71	90	79
Other sensational	3	29	32	13	24	27
Total	226	240	256	165	195	188

Success rates for APR have improved in recent years (85% in 1996­1997); however, they remain lower than day parole or regular full parole (see Figure 3). Completion rates (56%) for APR remain lower than rates for regular full parole (72%), while rates for breach of conditions remain higher (29% compared with 20%). Recidivism rates for APR have also remained higher (15%), but have been improving. Recidivism for APR cases has involved primarily non-violent offences, with the violent recidivism rate at 1% in the past two years.

Figure 3

Success rates for SR have ranged from 82­89% (see Figure 4) lower than rates for day parole or regular full parole, but similar to rates for APR cases for both successful completions and revocations for breach of conditions. Rates of reoffending for SR cases declined over the past five years (18% down to 13%), but remain higher than rates for day or regular full parole.

Figure 4

Charges for serious offences

Information on success illustrates that violent recidivism has declined for all types of release. Analysis of charges for serious offences reinforce these findings.

From 1992­1993 to 1994­1995, the number of charges for serious offences increased from 226 to 256, or by 13%. In 1995­1996, these charges dropped to 165, 36% or 91 fewer charges than in 1994­1995. The number of charges for serious offences rose to 195 in 1996­1997 or by 18%, and then declined to 188 in 1997­1998.

Table 2

Charges for Serious Offences by Release Type
Release Type	1992-1993	1993-1994	1994-1995	1995-1996	1996-1997	1997-1998
Day parole	73	68	64	15	12	26
Full parole	55	79	69	43	50	37
Statutory release	98	93	123	107	133	125
Total	226	240	256	165	195	188

When charges are examined by release type, the number of charges for serious offences involving day parolees decreased sharply, falling from 73 charges in 1992­1993 to 12 in 1996­1997 (an 84% drop), and then rose to 26 in 1997­1998 (see Table 2). Since 1995­1996 to 1997­1998 charges involving full parolees have fluctuated but remain well below the 1992­1993 to 1994­1995 levels. Charges against offenders on SR grew considerably, with the highest totals in 1996­1997 (133) and 1997­1998 (125). During the review period, offenders on SR were involved in more than 50% of all charges for serious offences. By comparison, day parolees accounted for 20% of charges, while full parolees accounted for 26%.

Numbers of charges, however, are only part of the picture. Rates of charges for serious offences per 1,000 offenders in the community, by release type, also provide a meaningful indicator (see Figure 5).

Figure 5

Annual rates of charges for serious offences per 1,000 offenders on SR have ranged from 45 to 62. Rates per 1,000 full parolees have ranged from 9 to 15. That is, offenders on SR have been four to five times more likely to be charged with a serious offence than offenders on full parole. Before 1995­1996, rates of charges per 1,000 day parolees (37 to 48) approximated rates for SR. Beginning in 1995­1996, however, rates for day parole dropped sharply compared with SR rates.

Figure 6

Post-warrant Expiry Recidivism

Information on federal recidivism after warrant expiry indicates that offenders reaching warrant expiry on SR are three to four times more likely to be readmitted to a federal institution than offenders who complete their sentence on full parole. Calculations are based on the status, as of March 31, 1997, of offenders who were released on SR or full parole annually from 1986­1987 to 1996­1997. As expected with this method of calculation, recidivism rates are higher for groups of offenders who have been in the community for longer periods, regardless of the type of their release.

These data appear to support two key conclusions. First, conditional release is a critical aspect of federal strategies for offender population management and cost control. Second, conditional release supports public safety, particularly in situations where the Service prepares the case and the National Parole Board assesses risk regarding the decision to release on parole.

---

1. 340 Laurier Avenue West, Ottawa, Ontario K1A 0P9.