The Changing Face of Conditional Sentencing

8. An Empirical Analysis of Conditional Sentencing in British Columbia

8. An Empirical Analysis of Conditional Sentencing in British Columbia*

Dawn North,
Department of Criminology, Simon Fraser University

*This is an abridged version of a report available through the Department of Justice Canada.

8.1 Introduction

Conditional sentence orders were designed as an explicit alternative to incarceration and introduced by way of Bill C-41 (Chapter 22 of the R.S.C. 1995), a package of sentencing reform legislation proclaimed in September 1996. This sanction, which allows offenders to serve terms of imprisonment in the community, has generated considerable scholarly debate. Early research undertaken to monitor the impact of Bill C-41 identified several important issues in terms of the introduction of conditional sentence orders.[259] Later reports, including a ‘final report’ which presented three years of conditional sentencing data from across the country,[260] have provided much needed information regarding the implementation of the new sanction and the perceptions of both the judiciary and the public.[261] While these research projects have contributed greatly to the body of knowledge being generated around conditional sentences, gaps remain which must be addressed in order that a comprehensive assessment of the sanction can be made. Some of the most obvious gaps relate to the need for reliable information regarding: 1)the optional conditions imposed; 2) the number, type and judicial response to breaches; and 3) the relationship between the conditions imposed and the likelihood of breach.

Research as outlined above is especially relevant in light of the recent Supreme Court of Canada judgment in R. v. Proulx. [262] This decision addressed many of the issues raised by the introduction of conditional sentence orders and clarified, among other things, the way in which a conditional sentence should be imposed and the importance of optional conditions in terms of achieving the goals of sentencing – especially denunciation and deterrence. Implicit in the judgment was the opinion that to date, the optional conditions attached to conditional sentence orders had not been sufficiently onerous and the sanction had been unable, therefore, to achieve these sentencing objectives. In order to determine whether or not the judgment has affected sentencing practices at the trial level, it is necessary to have baseline ‘pre-Proulx’ (prior to January 2000) data. One of the purposes of this study, therefore, was to provide a systematic statistical portrait of conditional sentencing prior to the landmark judgment. This will permit a valid evaluation of the effect of the decision in 2001.

Focus of this study

This study provides an in-depth analysis of the use of conditional sentence orders in three specific court locations within British Columbia. It supplies general information regarding the use of conditional sentences (including offence type, sentence length and the nature of optional conditions imposed) and specific information in terms of the number, type and response to breaches. Knowledge regarding the latter is critical in terms of many of the issues raised by the introduction of this sanction (net-widening, prison reduction, public perception, deterrent effect).[263]

Research Questions

In addition to providing much needed baseline information regarding the use of conditional sentence orders in British Columbia, this study addresses the following specific questions:

  • What was the reported breach rate?
  • What types of breaches were reported?
  • What were the judicial responses?

8.2 Findings[264]

Number of Conditional Sentence Orders

In 1998: judges sitting in Vancouver Provincial Court imposed 466 conditional sentence orders (covering 1019 court informations); judges in Burnaby Provincial Court imposed 81 conditional sentence orders (covering 133 ‘informations’); and judges in Port Coquitlam Provincial Court imposed 67 conditional sentence orders (covering 97 ‘informations’). Overall, slightly less than one-half (49.2%) of the orders involved single charges.[265]

Offence Information - Types of Offences (Table I)

  • In all court locations the majority of conditional sentences were imposed for offences against property (MSO): Vancouver (61.6%); Burnaby (65.4%); and Port Coquitlam (50.7%). Overall, offences against property accounted for 60.9% of the orders, persons offences 14.0%, drug offences (CDSA) 19.2%, administrative offences 2.9%, driving offences 1.1% and ‘other’ 1.8%.
  • When the Vancouver data were analyzed by ‘all offences’ (not MSO) the proportions by offence category remained substantially the same, though administrative offences (i.e. fail to appear, breach probation) increased from 2.4% to 12%, apparently at the expense of person and drug offences which were reduced from 11.6% to 8% and 22.7% to 15.0% respectively.
  • Analysis of ‘all offences’ for Vancouver (466 conditional sentences / 1,059 charges) revealed that over one-half (50.2%) of the ‘offences against property’ charges were for theft under $5,000,[266] and 87.2% of the ‘drug offences’ related to trafficking charges.

Sentence Length by Offence (Table II)[267]

  • Of the specific offences considered,[268] the longest average sentences (conditional sentence only) were imposed on property offences: break and enter (9.3 months); and fraud over $5,000 (7.8 months).
  • The average conditional sentence for a charge of theft under $5,000 was for 4.0 months. Sentence lengths for this charge had the greatest range - from one week to two years less a day.[269]
  • Average sentence lengths for the ‘person offences’ examined were as follows: common assault (4.2 months); aggravated assault/assault causing bodily harm (5.5 months).
  • The average conditional sentence imposed for trafficking or possession for the purpose of trafficking (CDSA) was 4.4 months.

Optional Conditions Imposed (Table III)

  • In terms of conditional sentence orders overall (n=614): 7.3% (45) contained only the mandatory conditions; the majority (53.1%) contained from 1 to 3 optional conditions; 30.9% imposed from 4 to 6 optional conditions; and 8.6% had more than 7 conditions. On average there were more conditions imposed on the conditional sentence orders than on the probation orders that were attached to them (conditional sentence: mean=3.2 / median=3.0; probation: mean = 2.7 / median=2.0).[270]
  • Overall, the top three optional conditions imposed (all locations) were related to substance abuse issues. On average, 45.1% of the orders included drug and/or alcohol counselling; 37.4% included ‘residential treatment/recovery house’; and 27.9% referred to ‘obey rules/regulations’ (of recovery house/treatment center).
  • With the exception of community work service (included on 20.6% of the Vancouver orders), conditions considered to be ‘restorative’ in nature were relatively rare: write essay (0.2%); letter of apology (3.2%); and restitution (1.7%).
  • A curfew was included on 17.8% of the conditional sentences; house arrest on 7.3%. 5.6% of the orders required the offender to appear before the sentencing judge for ‘post-sentencing’ reviews.

Models of Sentencing & Optional Conditions

One of the distinctions said to exist between conditional sentence orders and probation orders relates to the underlying purpose of the optional conditions attached to each. In theory, probation conditions are directed towards offender rehabilitation, while conditional sentence order conditions are directed towards preventing recidivism.[271] Classification of optional conditions according to the orientation they are most consistent with (treatment/offender; punitive/protection of public; restorative/community),[272] however, suggests that conditional sentences were used primarily to achieve rehabilitative/treatment goals; then punitive/protection of public; then restorative/community (in that order). This ranking remained unchanged whether the question asked was “how many of the optional conditions imposed related to each orientation?” or “how many orders contained optional conditions related to each orientation?”

  • In terms of the first question, approximately two-thirds (67.3%) of the optional conditions imposed overall (out of 1897 conditions classified) related to a treatment/ offender orientation; 23.2% to a punitive/ protection of the public orientation; and 9.3% to a restorative/community orientation.[273] These proportions were substantially maintained across court locations.
  • When the question was re-phrased as “how many conditional sentence orders contained optional conditions related to each orientation?” the ranking remained the same. Overall, 78.0% of the orders contained optional conditions relating to a treatment/offender orientation; 49.5% to a punitive/protection of the public orientation; and 26.4% to a restorative/community orientation.

While there are obvious weaknesses with this method of establishing sentencing orientation, such an analysis serves to draw attention to the importance of optional conditions in terms of establishing and communicating the purposes and objectives of the sanction. More importantly, perhaps, a similar analysis conducted ‘post-Proulx’ will clarify the impact of the decision in terms of the application of appropriate optional conditions.

Known Breaches (Table IV - VI)

Breach Rate, Process, Response and Findings
  • The overall breach rate for the three court locations was 37.6% (231 orders breached out of a possible 614).[274] In Vancouver there were ‘allegation of breach’ reports filed on 40.6% (189) of the files reviewed; in Burnaby the rate was 34.6% (28); and in Port Coquitlam the rate was 20.9% (14).
  • Breaches of ‘optional conditions’ were the most common in Vancouver (34.7%) and Port Coquitlam (57.1%), multiple breaches in Burnaby (35.7%). Analysis of these figures is complicated, however, due to the high number of allegations which set out ‘multiple’ breaches.[275] See Table IV.
  • Analysis of the Vancouver data showed that the breaches alleged most frequently in the reports were: 1) not ‘reporting to supervisor as directed’ (33.1%); 2) not attending ‘residential treatment/recovery house’ (30.0%); and 3) committing new offences (13.7%). See Table V.
  • Vancouver: There were certain conditions, or types of conditions, that were associated with higher than expected breach rates.[276] Orders which included conditions that suggested drug use, for instance, were more likely to be breached (51.2%) than orders which did not (21.2%).[277] Similar results were obtained with reference, specifically, to the requirement that an offender attend for residential drug treatment. Orders in which this condition was present breached at a rate of 64.7%, as opposed to orders which did not include this conditions (26.3%).[278]
  • Vancouver: 26 out of 189 (13.8%) allegation reports referred to new offences – the breach alleged being that the offender failed to ‘keep the peace and be of good behavior’.[279] This figure is problematic, however, in that breach allegations relating solely to new offences were not regularly processed in 1998.[280]
  • Vancouver: Of the offenders taken into custody as a result of the breach,[281] 27.3% were released pending the hearing, 22.7% were detained, and 50% remained in custody ‘by consent’.
  • Vancouver: Of the 84% (159) that proceeded to the hearing stage, the vast majority (93%) admitted the breach(es). Of the 7% (11) known to have disputed the breach allegation(s), six of the breaches were ‘proven’ and five were ‘not proven’.[282]
Judicial Response
  • ‘Judicial responses’ to breach were known in 90% (208) of the 231 cases studied (the remaining 10% represented breaches that were either withdrawn by crown or not yet dealt with). Overall responses were as follows: 37.5% (78) of the conditional sentences were terminated; 26.9% (56) resulted in amendments to the conditions; 17.8% (37) of the sentences were suspended and a portion was served incarcerated; 13.5% (28) resulted in no action; and 4.3% (9) were dismissed. See Table VI.
  • Vancouver judges terminated a greater percentage of conditional sentence orders than did judges at the other two court locations. The Vancouver conditional sentence termination rate (by judicial response) was 39.8%; Burnaby was 33.3%; and Port Coquitlam was 15.4%.
  • Analysis of the Vancouver breach data suggested that all breaches are not treated equally in terms of judicial response. The breaches most likely to result in the conditional sentence being terminated were: house arrest (60%); new offence (58.3%); residential treatment (51.4%); and curfew (41.7%).

8.3 Discussion

Breach Rates Generally

The findings of this study suggest that previous breach rate estimates for BC (26%)[283] may have been overly optimistic. There are three possible conclusions which could be drawn from the differences observed between this estimate and the relatively high breach rates documented at two of the three court locations studied. First, it may be that the 26% estimate was based on poor or incomplete data – conditional sentence breaches in BC are not reliably tracked in either the court or corrections databases. The relatively large drug offender population in BC, particularly in the downtown Vancouver area, may be contributing to the higher than expected breach rates observed in this study.

An alternate explanation may be related to the differences observed between the ‘urban’ (Vancouver) and ‘suburban’ (Burnaby and Port Coquitlam) breach rates. It’s possible that the high rate reported for Vancouver (40.6%), for instance, may be offset by the lower rates found in less densely populated suburban and rural areas. If this is the case, the 26% overall rate could be accurate for the province.

The final possibility is that the inconsistencies observed are the result of a combination of these factors – incomplete data and variations in rates. What is clear, however, is that there remains a need for further research which collects and analyzes reliable data regarding conditional sentence breach rates. This is particularly critical in terms of responding to suggestions that the sanction has not been successful to date in reducing admissions to provincial institutions.

Though the relatively small number of orders coded (614) precludes the making of definitive statements regarding possible ‘breach predictor’ variables, the findings do suggest that there are some factors which are related to higher breach rates and some that are not. In terms of the latter, race, gender and category of offence appeared to be generally unrelated to the likelihood that a breach report would be filed, though exceptions were noted. In the Burnaby and Port Coquitlam cases, for instance, it appeared that property offences had a higher breach rate than would be expected and person offences had a lower breach rate than would be expected. It is also possible that specific offences within categories (i.e. break and enter) may have high breach rates which are offset by the low rates of other offences within the same category (i.e. fraud). Again, the value of comparing ‘averages’ or ‘categories’ when studying conditional sentence breach rates, becomes questionable.

Several factors were identified which did appear to be related to an increased likelihood that an order would be breached. These would include: prior criminal record; the presence of optional conditions which suggest drug use; and/or requirements that the offender attend for residential treatment, abide by a curfew or be subject to house arrest. While it is obvious that the presence of these conditions did not cause the breaches,[284] the fact that they are associated with relatively high breach rates raises serious issues around assessments of acceptable levels of ‘risk’. Identifying which conditions are most (and least) likely to result in breach by the offender, therefore, is crucial in terms of providing judges with the tools necessary to craft creative and effective sanctions.[285]

In terms of the ‘Sword of Damocles’[286] metaphor associated with conditional sentencing, it would appear that the rope was completely ‘severed’ in approximately one-third (37.5%) of the orders breached overall. Whether this would amount to the degree of certainty referred to in the literature remains a matter for debate. In light of the complex calculations required to determine the time left to be served on a terminated conditional sentence, further research would be required to determine whether, in the cases in which the rope was severed, it was a sword or a butter-knife that fell.

Implications for the Future (Proulx)

Many of the issues relating to the application of conditional sentences were specifically addressed in R. v. Proulx,[287] a recent decision of the Supreme Court of Canada which, among other things, sets out the principles which should govern the use of the new sanction. The court’s interpretation of the legislation has serious implications in terms of conditional sentence orders generally, and breaches of those orders specifically. For instance, although the court situates conditional sentences within a legislative package (Bill C-41) aimed at reducing prison populations, it potentially frustrates the attainment of that goal by: 1) allowing judges to extend the length of conditional sentences beyond the terms of incarceration they replace;[288] 2) suggesting that onerous and punitive conditions (including curfews and house arrest) should be the norm;[289]and 3) creating a presumption of incarceration in situations where an offender has breached a condition without reasonable excuse.[290]

As noted earlier, one of the purposes of this study was to provide ‘baseline’ data regarding the use of conditional sentence orders in specific BC court locations prior to the Proulx decision. The decision is likely to have a significant impact in terms of the application of conditional sentences (specifically decisions relating to sentence length, optional conditions attached and judicial response).[291] Whether the impact will be seen negatively or positively in light of the original goals of the sanction remains to be seen. In terms of breach rates, for instance, it may be that those referred to in this report should be considered as minimum figures.

8.4 Conclusions

The findings of this study will potentially be of interest to many agencies. The Department of Justice Canada has an obvious interest in monitoring the implementation and effect of Bill C-41. Provincial corrections agencies, inasmuch as they have been greatly affected by the introduction of conditional sentences, may be assisted by the general information provided regarding the use of this sanction, and specific information provided in terms of the number of, nature of and response to conditional sentence breach allegations. This study also provides much needed feedback to the judiciary regarding the use of conditional sentences, factors which might increase the likelihood that an offender will breach, and the responses of other judges to specific kinds of breaches.

Conditional sentence orders have become, and will likely continue to be, an important part of the sentencing landscape in Canada. While this new sanction is still in the process of finding its place and purpose within the larger criminal justice system, it is important that its use and impact continue to be systematically monitored.

Table I : CSO by category of offence - CSO - Category of Offence (MS - O) - All Locations (n=614)

Table II - CSO length x offence CSO Length (months) x Offence 2-1
Offence # cases Range Mean Median
from to
fraud/over $5,000 17 3.00 18.00 7.8 6.0
theft/under $5,000 49 0.25 24.00 4.0 3.0
break & enter 63 1.00 24.00 9.3 9.0
Assault 12 0.75 6.00 4.2 4.5
ACBH/aggravated assault 12 3.00 9.00 5.5 5.0
Trafficking (CDSA) 62 1.00 18.00 4.4 3.0

2-1 Refers to orders which cover only one charge:

  • Vancouver = 217 out of 466
  • Burnaby = 44 out of 81
  • Port Coquitlam = 41 out of 67
  • Overall = 302 out of 614

Note: Only offences which had at least 10 cases were included.

Table III : CSO and probation conditions imposed - Vancouver

Table IV : Number and Type of Known Breaches – All Locations

Table V - Specific Conditions Breached - Vancouver
No. Condition # imposed[1] % imposed[2] # breached[3] % of breaches[4]
1 Reporting - as directed5-1 466 21.1% 85 33.1%
2 Residential treatment/recovery house 173 7.8% 77 30.0%
3 Curfew 83 3.8% 24 9.3%
4 Maintain/provide residence5-1 466 21.1% 16 6.2%
5 Obey rules/regulations 144 6.5% 16 6.2%
6 Area restriction (re-offence issue) 100 4.5% 13 5.1%
7 Abstain drugs 94 4.3% 6 2.3%
8 House arrest 34 1.5% 5 1.9%
9 Other - specify 103 4.7% 4 1.6%
10 Review (in court) 26 1.2% 3 1.2%
11 Community work service 96 4.3% 2 0.8%
12 Submit breath/urine/blood demand 87 3.9% 2 0.8%
13 Abstain alcohol 64 2.9% 1 0.4%
14 Counselling - drug/alcohol 206 9.3% 1 0.4%
15 No contact (protection issue) 38 1.7% 1 0.4%
16 Weapons restriction 29 1.3% 1 0.4%
Total 2209 100.0% 5-2257 100.0%
New offence 26 5-313.8%
Table VI - Judicial Response: All locations6-1
Vancouver Burnaby Port Coquitlam Overall
# % # % # % # %
Dismissed 7 4.1% 2 8.3% 0 0.0% 9 4.3%
No action 21 12.3% 3 12.5% 4 30.8% 28 13.5%
Change conditions 49 28.7% 5 20.8% 2 15.4% 56 26.9%
CSO suspended -part to be served I/C 26 15.2% 6 25.0% 5 38.5% 37 17.8%
CSO terminated 68 39.8% 8 33.3% 2 15.4% 78 37.5%
Total 171 100.0% 24 100.0% 13 100.0% 208 100.0%

6-1 Breach allegations which were withdrawn (by crown) or not yet dealt with have been removed.

Classification Guide - Optional Conditions
Treatment/offender oriented conditions:
abstain alcohol T
abstain drugs T
attend residential treatment/recovery house T
counselling - anger management T
counselling - drug/alcohol T
counselling - general/as directed T
counselling - psychiatric/psychological T
education T
maintain/seek employment T
obey rules/regulations (of recovery house) T
review (in court) before sentencing judge T
submit breath/urine/blood upon demand T
Punitive/protection of public oriented conditions:
area restriction (protection issue) P
area restriction (re-offence issue) P
area restriction (unknown) P
curfew P
driving restriction P
house arrest P
no contact (association/co-accused) P
no contact (protection issue) P
no contact (unknown) P
weapons restriction P
Restorative/community oriented conditions:
community work service hours R
letter of apology to victim(s) R
restitution (to victim) R
write and submit essay (judge chooses topic) R

Date modified: