Conditional Release of Federal Offenders Convicted of Criminal Organization Offences

Acknowledgements

The authors would like to thank Public Safety Canada for their funding and collaboration on this project.  This research would not have been possible without the data development skills of Ian Broom and the editorial suggestions of Rick Ruddell.

Executive Summary

Past research on criminal organization offenders has typically centred on the nature of the offences committed and profiling those offenders. This study extended the extant knowledge of criminal organization offenders by updating past profiles, focusing on community outcomes while on conditional or statutory release, and identifying risk factors related to re-offending for these offenders.

Overall, 451 offenders were identified in the Correctional Service of Canada’s (CSC) Offender Management System (OMS) as being convicted of a criminal organization offence, as outlined in Sections 467.11 to 467.13 of the Criminal Code of Canada (CCC), between April 25, 1997 and March 31, 2009. This included 418 non-Aboriginal males, 19 Aboriginal males, and 14 women offenders, with an average sentence length of 5.2 years.

Most offenders convicted of a criminal organization offence had some prior involvement with the criminal justice system, with 21.5% having served a previous adult term in a federal penitentiary. Along with their current criminal organization conviction, offenders were most commonly also convicted of drug offences (59.6%) or attempted murder (8.2%).

Examination of criminogenic risk, need and reintegration potential found that the typical criminal organization offender was assessed as being “medium” risk (58.1%) and “high” need (45.9%), with “high” reintegration potential (68.8%). Domain-level analyses of need illustrated that criminal organization offenders were significantly more likely to have some or considerable need in the areas of criminal attitudes and criminal associates than a matched sample of CSC offenders.

Of the 451 offenders who were convicted of a criminal organization offence, 332 (73.6%) had been released to the community. The majority were released on day parole (51.8%) or statutory release (44.9%). Most (76.4%) had been employed at some point during release, and 14.8% of those released were participating in some sort of community intervention program, with the most common programs including education, Counter-Point, and living skills programs.

Of those who were released, 12.7 % (42) were re-admitted to a federal institution. Most had their release revoked without a new offence (76.2%), while 14.3% (n=6) were convicted of a new offence.

Survival analyses conducted to determine the risk of failure upon release found that those convicted of criminal organization offences were significantly less likely than the matched group to be returned to custody. Risk factors found to be especially predictive of readmission or re-conviction included age at release and type of release, with younger offenders and those on statutory release more likely to fail than those released at an older age or released on day or full parole.

Introduction

One of the largest challenges confronting the Canadian justice systems is the inordinate amount of harm caused by a relatively small percentage of offenders who are involved in organized crime.  Organized criminal activity has had a pervasive impact upon the economic and social well-being of Canadian society, breeding corruption in political systems, compromised personal security, and contributed to weakness in global and domestic markets (Royal Canadian Mounted Police, 2005).  In response to this threat, the Government of Canada has made several amendments to the Criminal Code of Canada (CCC) which target offenders with various degrees of involvement with criminal organizations as well as increasing the penalties for those offences.  In addition to profiling these offenders, this study investigates the community outcomes for those convicted of involvement in organized criminal activities.

The presence of criminal organization activity in Canadian society was evident even before Canada became an independent nation.  As noted in Schneider (2009), the trademarks of criminal organization activity were observed as far back as the pirate and privateer activities in the Atlantic region during the sixteenth and seventeenth centuries.  Relying on violence, black markets, powerful connections, and a code of secrecy, the pirates and privateers off the coast of Newfoundland capitalized on a profitable industry (fishing) for financial gain.  Since this time, other criminal enterprises have established themselves in Canadian culture.  From the emergence of the Mafia in the 1930’s, the expansion of outlaw motorcycle gangs in the 1970’s, to the growth of Asian Triads and Columbian Cartels in the 1970’s and 1980’s, criminal organization activities have had a growing presence in Canadian society. To date, 750 criminal organizations have been identified as being active in Canada (Criminal Intelligence Services Canada, 2009).

The presence of these criminal organizations has had serious repercussions for Canadians.  Their involvement in activities such as environmental and financial crime, contraband tobacco, illicit drugs and firearms, intellectual property crimes, human smuggling, and metal theft costs Canadians billions of dollars each year (Criminal Intelligence Services Canada, 2009), with the Royal Canadian Mounted Police (2009) estimating that organized crime cost Canadians five billion dollars a year in economic-related crimes alone.  Coupled with less tangible socio-economic effects (such as fear and loss of trust in the justice system), concern over the impacts of organized crime have led to the introduction of legislation aiming to curb the effect of this activity on Canadian society.

Organized crime legislation was established in 1997 with the introduction of a new offence, “participation in a criminal organization”, coupled with increased powers of investigation and protection for the police.  Recognizing that participation in a criminal organization was a multi-faceted issue, the Federal Government amended the CCC in 2001 to allow for three offences in regards to criminal organizations, specifically:

1. Participation in the activities of a criminal organization (Section 467.11);
2. Commission of an offence for a criminal organization (Section 467.12); and
3. Instructing the commission of an offence for a criminal organization (Section 467.13).

In addition, the 2001 amendments improved levels of protection and accountability for those involved in prosecuting organized criminals while simplifying the definition of “criminal organization” to mean a group, however organized, that:

1. is comprised of three or more persons in or outside Canada; and
2. has as one if its main purposes or main activities the facilitation or commission of one or more serious offences that, if committed, would likely result in the direct or indirect receipt of a material benefit, by the group or by any of the persons who constitute the group (Section 467.1).

This new definition delineated that a criminal organization did not include a group of persons that forms randomly for the immediate commission of a single offence.

More recently, Bill C-9, the Conditional Sentencing and Reform Act (May 31, 2008) and Bill C-14, Organized Crime and Protection of Justice System Participants (June 23, 2009) have further increased the accountability for those convicted of criminal organization offences.  Bill C-9 eliminated the availability of conditional sentences for anyone convicted of an indictable offence punishable by a maximum of ten or more years, which includes those convicted of organized crime offences.  Bill C-14, by contrast, created new offences related to drive-by shootings and assaults against peace officers, while making homicides connected to organized crime activities automatically prosecuted as first-degree murder and therefore subject to a mandatory life sentence without parole eligibility for 25 years.

Most Canadian research on criminal organizations centres on the nature of the offences committed by these organizations (Criminal Intelligence Service Canada, 2008), with less focus on those convicted of organized crime offences.  Prior empirical work has profiled those offenders who had been convicted under the expanded, three offence definition of criminal organization (Motiuk & Vuong, 2005; 2006), however few have systematically investigated the community outcomes of those on conditional release after incarceration for a criminal organization offence.  This research addresses the gap in the literature by examining federal offenders convicted of criminal organization offences since the CCC amendments in 1997, with a specific focus on community outcomes such as employment and program activity upon conditional release, and the risk factors that are associated with failure during release (e.g., parole revocations or reconvictions) in this population.

Method

Participants

This study examined the 451 offenders who were identified in the Correctional Service of Canada’s (CSC) Offender Management System (OMS) as being convicted of a criminal organization offence, as outlined in Sections 467.11 to 467.13 of the CCC, between April 25, 1997 and March 31, 2009.  Readmissions to a federal institution were considered until January 8, 2010.

A matched, general offender sample was also drawn from the OMS in order to compare the profile and outcomes of those with a criminal organization offence to the general offender population.  The matched offender group included all warrant of committal admissions between April 25, 1997 and March 31, 2009, not including those admitted for a criminal organization offence.  Based on the methodology of previous studies (Motiuk & Vuong, 2006; Nafekh & Stys, 2004), each federal offender with a criminal organization offence was matched with an offender without such an offence based on race (Aboriginal/non-Aboriginal), offender status (archived vs. active), gender, and sentence length (less than 3 years, 3-6 years, 6-10 years, 10+ years, life/indeterminate).

Procedure

For the purposes of this study, all available data for federally sentenced offenders were extracted from the OMS.   Characteristics of criminal organization offenders were identified by comparing the proportion of offenders convicted of a criminal organization offence to the matched sample group. Contingency table (chi-square) and analyses of variance (ANOVA) were used to draw comparisons between the two groups and between the three different criminal organization offences on a number of demographic and criminogenic factors assessed at intake, including race, age at admission, sentence length, criminal history, region of admission, and need profiles.  All multiple comparisons for significant omnibus tests were conducted at adjusted power levels using the Bonferroni correction.^{Footnote 1}

Community outcomes for the criminal organization offenders were also compared between the criminal organization offences and against the comparison group, with a focus on employment and program activity.  Finally, survival analyses were conducted to identify the predictors of conditional release success for those convicted of criminal organization offences as well as those in the comparison group.

Measures

Criminal Organization Offenders

Participants were considered criminal organization offenders if they were admitted for one of the criminal organization offences under the Criminal Code of Canada, specifically:

1. Participation in activities of criminal organization (Section 467.11);
2. Commission of offence for criminal organization (Section 467.12); and
3. Instructing commission of offence for criminal organization (Section 467.13).

Those offenders who were convicted under the previous legislation governing criminal organization offences (Participation in a Criminal Organization) were combined into the Participation group of offenders to allow for continuity.^{Footnote 2} For the purposes of this study, seriousness of offence will be considered in the order outlined in the Criminal Code of Canada.  According to the sentence lengths outlined for these offences, participation is considered to be the least serious of the three offences (liable to a term of imprisonment not exceeding five years), with commission the next most serious (liable to a term of imprisonment not exceeding 14 years), and instructing being most serious (liable to life imprisonment). 

The Offender Intake Assessment (OIA)

The Offender Intake Assessment (OIA; Commissioner’s Directive 705-6, 2007; Motiuk, 1997) is a comprehensive evaluation of an offender conducted at the time of admission to the federal correctional system by institutional parole officers.  Consisting of two core components, the Static Risk Assessment and Dynamic Factors Identification and Analysis (DFIA), the OIA involves the collection and analysis of information on each offender’s criminal and mental health history, social situation, education, and other factors relevant to determining criminal risk and identifying offender needs.

Criminal Organization (Gang Member) Designation

Participants were considered members of a criminal organization if they were designated as such during the OIA.  According to CSC, a member of a criminal organization is a person associated to or involved with a criminal organization which includes supporters, sympathizers, strikers, affiliates, hangarounds, prospects, associated members and persons aspiring to be members (Commissioner’s Directive 568-3, 2008).

Dynamic Factors Identification and Analysis (DFIA)

The Dynamic Factors Identification and Analysis (DFIA) assesses a variety of contributing criminogenic factors grouped into seven domains (measured by several indicators): associates/social interaction (11 indicators), attitudes (24 indicators), community functioning (21 indicators), employment/education (35 indicators), marital/family (31 indicators), personal/emotional (46 indicators), and substance abuse (29 indicators). Scores on these indicators place an offender on a four-point scale of need to address that domain, ranging from "factor seen as an asset to community adjustment" to "considerable need for improvement" (Commissioner’s Directive 705-6, 2007; Motiuk, 1997)^{Footnote 3}.

The Statistical Information on Recidivism - Revised 1 (SIR-R1)

The SIR-R1 Scale is a statistically derived tool for predicting recidivism.  The scale combines measures of demographic characteristics and criminal history in a scoring system that estimates the likelihood of recidivism.  The 15 item scale yields a total score ranging from -30 (poor risk) to +27 (very good risk) which are then collapsed into five SIR-R1 groupings ranging from very good (4 out of 5 offenders do not re-offend within 3 years of release) to poor (1 out of 3 do not re-offend within 3 years of release).  The SIR-R1 has not been validated for use with women or Aboriginal offenders (Commissioner’s Directive 705-6, 2007).

The Custody Rating Scale (CRS)

The Custody Rating Scale is an empirically derived actuarial tool comprised of 12 items that generate security designations upon an offender's admission. Scale items are grouped into two subscales, the Institutional Adjustment subscale (5 items) and the Security Risk subscale (7 items). Items within each subscale are summed to provide a total score. The resulting security classification level (minimum, medium or maximum security) increases as scores on either subscale increase (Commissioner’s Directive 705-6, 2007; Nafekh & Stys, 2004).

Reintegration Potential Profile (RPP)

For non-Aboriginal male offenders, the Reintegration Potential Profile (RPP) is a computer generated value based on the results of the OIA Overall Static and Dynamic Factor Assessments, the SIR-R1 risk grouping and the CRS security level designation. For Aboriginal and women offenders, RPP is derived using the OIA overall Static and Dynamic factor ratings and the CRS security level designation.  A rating of low, moderate or high potential is assigned to each offender, describing the likelihood that offenders will be able to successfully reintegrate back into the community upon release.  An offender’s RPP dictates, among other things, program referrals and other risk management strategies (Commissioner’s Directive 705-6, 2007).

Results

Characteristics of Criminal Organization Offenders

Admission Trends

In total, 451 federal offenders were convicted of a total of 557 criminal organization offences between April 25, 1997 and March 31, 2009.  In that time period, half of the convictions were for committing an offence for a criminal organization (50.1%, n=279), and 43.3% were for participating in the activities of a criminal organization (n=241).  Only 37 of the 557 criminal organization offences (6.6%) were for instructing the commission of an offence for a criminal organization.  Annual admission figures for criminal organization offences are presented in Table 1.

Table 1. Annual CSC Admissions for Criminal Organization Offences (1997 – 2009)

Note.  Where offenders were admitted for more than one criminal organization offence (n=31), the most serious offence was selected.  Data for 2009 only includes admissions from January through March.

Demographic Information

Table 2 presents a summary of the demographic characteristics of criminal organization offenders by type of criminal organization offence (participating, committing, and instructing).  The most serious offence was considered for those offenders admitted for more than one criminal organization offence (n=31).  The majority of criminal organization offenders were male (96.9%, n=437), Caucasian (82.9%, n=374), convicted in Quebec (83.1%, n=375), and they were, on average, 38 years old (SD=9.9).

Table 2 : Demographic Characteristics of Criminal Organization Offenders as a Percentage of the Criminal Organization Offence Types

Two-way contingency table analysis examining differences in demographic variables (gender, race, and region of conviction) across offence types (participating, committing, or instructing) could not be interpreted as the small number of cases did not provide sufficient cell counts for valid results.  In addition, a one-way analysis of variance (ANOVA) and an independent sample t-test conducted to evaluate the difference in age between the offence types (participating, committing, or instructing) and the criminal organization and matched sample group failed to find significant age differences between the groups.  However, two-way contingency table analyses examining differences in region of offence (Pacific, Prairies, Ontario, Quebec, or Atlantic) across offender groups (criminal organization offender or matched sample) found statistically significant group differences (χ² (4, N = 872) = 381.28, p < .001).  Multiple pair-wise comparisons of independent proportions found that those who were admitted for a criminal organization offence were significantly more likely than the matched group to have committed their crime in the Quebec region and significantly less likely than the matched group to have committed their crime in all other regions (see Table 3).

Table 3. Criminal Organization and Matched Sample Offenders Compared on Region of Offence

Note.  ***p < .001.  Region information was missing for n=30 matched sample offenders.

Past and Present Offence Information

Table 4 presents the past and present offence history for the three criminal organization offence types.  Of the 437 offenders convicted of a criminal organization offence for which past offence data is available, 79.6% (n=359) had some previous criminal record history, with 21.5% (n=94) having served a previous adult custody period in a federal penitentiary.  When looking at individual criminal organization offence types, two-way contingency table analyses examining differences in past youth  or adult records, as well as prior  youth  or adult federal custody (yes or no) across offence types (participating, committing, or instructing) found no statistically significant differences between groups.  However, a two-way contingency table analysis examining differences in past adult provincial custody (yes or no) across offence types (participating, committing, or instructing) found significant group differences (χ² (2, N = 436) = 7.83, p < .05).  Multiple pair-wise comparisons of independent proportions found that those who were convicted of committing a crime for a criminal organization were significantly more likely to have served a previous adult provincial term than those convicted of participating in a crime for a criminal organization (65.3% vs. 51.6%; z = 2.60, p < .01).

Two-way contingency table analyses examining differences in past youth or adult records, prior youth or adult federal custody (yes or no) across offender groups (criminal organization offender or matched sample) found no differences between groups on adult sentences.  However, significant group differences were found for previous youth record and previous youth custody (χ² (1, N = 883) = 28.04, p < .001 and (χ² (1, N = 874) = 14.41, p < .001, respectively).  Multiple pair-wise comparisons of independent proportions found that those who were admitted for a criminal organization offence were significantly less likely than the matched group to have a youth record (24.1% vs. 40.8%; z = -5.30, p < .001) or to have served a previous youth term (10.9% vs. 20.3%, z = 3.80, p < .001).

In addition to their criminal organization offence, almost all offenders (97.3%, n=439) were admitted with another conviction in the same sentence.  Criminal organization offenders were most commonly also convicted of drug offences (59.6%, n=269) and “other” criminal code offences (including break and enter (B&E) and other property crimes, conspiracy to commit an indictable offence, possession of weapons offences, and criminal organization offences; 16.2%, n=73).  Two-way contingency table analyses examining differences in current offences (homicide, attempted murder, sex offence, robbery, assault, other violent, drugs, and other criminal code) across offence type (participating, committing, and instructing) could not be reliably conducted due to inadequate cell sample sizes.  However, of interest is the fact that 90.3% of those convicted of an instructing offence were also convicted of a drug-related offence, while the majority of the violent offences were conducted by those convicted of participating offences. 

Two-way contingency table analyses examining differences in current offences (homicide, attempted murder, sex offence, robbery, assault, other violent, drugs, and other criminal code) across offender groups (criminal organization offender or matched sample) revealed statistically significant group differences (χ² (1, N = 902) = 258.79, p < .001).  Multiple pair-wise comparisons of independent proportions found that those who were admitted for a criminal organization offence were significantly less likely than the matched group to also be admitted with homicide (2.0% vs. 9.1%; z = -4.66, p < .001), sex (0.2% vs. 14.9%; z = -8.32, p < .001), and robbery offences (3.8% vs. 23.1%; z = -8.50, p < .001).  However, those admitted for a criminal organization offence were significantly more likely than the matched group to also be admitted with attempted murder (8.2% vs. 2.9%; z = 3.49, p < .001) and drug offences (59.6% vs. 18.6%; z = 12.62, p < .001).

Excluding those offenders with a life sentence (n = 8), aggregate sentence length for those convicted of a criminal organization offence ranged from two to 26.8 years, with an average sentence length of 5.2 years (SD=3.3). 

Table 4. Criminal Offence History as a Percentage of the Criminal Organization Offence Types

Note.  “Other Criminal Code” offence includes B&E and other property crimes, conspiracy to commit an indictable offence, possession of weapons offences, and criminal organization offences. *p<.05.

Gang Membership

Gang membership information for those convicted of criminal organization offences are presented in Table 5.  Overall, 396 of the 451 offenders convicted of a criminal organization offence (87.8%) were identified as being part of a gang during the OIA process.  Gang affiliation was almost equal across the participation, commission, and instructing offences (89.2%, 87.0%, and 87.1% respectively).  Data outlining the specific gang affiliation of members were available for 380 of the 451 (84.3%) criminal organization offenders.  The majority (51.6%, n=196) of those convicted of a criminal organization offence were identified as being part of a motorcycle gang, with 40.0% (n=152) belonging to a traditional organized crime group.

Two-way contingency table analyses examining differences in gang membership types (Aboriginal, Asian, motorcycle, street, and traditional) across offence type (participating, committing, and instructing) and across offender groups (criminal organization offender or matched sample) were not conducted due to the small number of cases in some of the subgroups.  

Table 5. Gang Affiliation as a Percentage of the Criminal Organization Offence Types

Note.  Traditional organized crime refers to organized crime groups such as the Mafia.

Criminogenic Risk, Need, and Reintegration Potential

Examination of OMS data found that of those criminal organization offenders for whom risk data was available (98.4%, n=444), 58.1% (n=258) were assessed as being a “medium” risk rating, with a quarter of offenders (27.0%, n=120) being rated as “high” risk.  The remaining 14.9% (n=66) of the criminal organization offenders were considered “low” risk.  Two-way contingency table analyses examining differences in overall risk ratings (low, medium, and high) across offence type (participating, committing, and instructing) found statistically significant group differences (χ² (4, N = 444) = 15.67, p < .01).  Multiple pair-wise comparisons of independent proportions revealed that offenders rated as “high” risk were significantly more likely to be convicted of participating in a crime for a criminal organization than committing a crime for a criminal organization (37.0% vs. 21.6%; z = 3.39, p < .001).  Interestingly, two-way contingency table analyses examining differences in overall risk ratings (low, medium, and high) across offender groups (criminal organization offender or matched sample) also revealed significant group differences (χ² (2, N = 895) = 45.66, p < .001).  Multiple pair-wise comparisons of independent proportions found that criminal organization offenders were significantly more likely to be rated as medium risk (58.1% vs. 35.9%; z = 5.54, p < .001), and significantly less likely to be rated as high risk (27.0% vs. 44.8%; z = -6.65, p < .001), as compared to the matched sample group. 

Further, of those criminal organization offenders for whom need data were available (98.4%, n=444),  45.9% (n=204) were assessed as having a “high” overall need level rating, with slightly fewer offenders (44.4%, n=197) obtaining a overall rating of “medium”.  Only 9.7% (n=43) of the criminal organization offenders received a need level rating of “low”.  Two-way contingency table analyses examining differences in overall need ratings (low, medium, and high) across offence type (participating, committing, and instructing) found significant group differences (χ² (4, N = 444) = 18.35, p < .01).  Multiple pair-wise comparisons of independent proportions found that offenders rated as “high” need were significantly more likely to be convicted of participating in the activities of a criminal organization than committing a crime for a criminal organization (59.1% vs. 38.2%; z = 4.11, p < .001).  Two-way contingency table analyses examining differences in overall need ratings (low, medium, and high) across offender groups (criminal organization offender or matched sample) found no significant group differences.

Table 6 outlines a more detailed, domain level, summary of criminogenic needs as assessed in the OIA.  For ease of presentation, the four criminogenic need levels (factor seen as an asset to community adjustment, no immediate need for improvement, some need for improvement, and considerable need for improvement) were collapsed into two categories:  “Factor seen as asset/no immediate need” and “Some/considerable need for improvement”.  As illustrated in the table, almost all criminal organization offenders (99.3%, n=441) were admitted with some or considerable need in the areas of criminal associates and most (88.3%, n=392) were admitted with some or considerable need in the criminal attitudes domain.  

Two-way contingency table analyses examining differences in the collapsed criminogenic need levels (factor seen as asset/no immediate need and some/considerable need for improvement) for each domain across offence type (participating, committing, and instructing) found no statistically significant group differences for the majority of the domains, however, significant group differences in the personal/emotional domain existed across the offence types (χ² (2, N = 444) = 32.40, p < .001).  Multiple pair-wise comparisons of independent proportions found that offenders convicted of instructing the commission of or committing a criminal organization offence were significantly more likely to score well on the personal/emotional domain (asset or no considerable need) than those convicted of participation offences (z = 5.03, p < .001 and z = 3.78, p < .01, respectively).

Table 6. Criminogenic Needs, by Domain, as a Percentage of the Criminal Organization Offence Types

 Note. ***p < 001.  Information was missing for n=7 criminal organization offenders

Two-way contingency table analyses examining differences in the collapsed criminogenic need levels (factor seen as asset/no immediate need and some/considerable need for improvement) across offender groups (criminal organization offender or matched sample) found significant group differences across all domains (see Table 7).  Multiple pair-wise comparisons of independent proportions found that offenders convicted of a criminal organization offence were significantly less likely than those in the general matched sample to have some/considerable need in the employment (z = -13.41, p < .001), family (z = -12.68, p < .001), substance abuse (z = -14.97, p < .001), community functioning (z = -10.51, p < .001), and personal/emotional (z = -15.48, p < .001) domains.  However, offenders convicted of a criminal organization offence were more likely than the matched sample group to have some/considerable needs in the areas of criminal associates (z = 12.73, p < .001) and criminal attitudes (z = 9.06, p < .001).

Table 7. Criminal Organization and Matched Sample Offenders Compared on Need Domains

Note.  ***p < .001.  Information was missing for seven criminal organization offenders.

Examination of an offender’s reintegration potential profile (RPP) as assessed at intake found that very few (8.9%, n=39) of the criminal organization offenders for which RPP data were  available (n=439) were assessed as having a low reintegration potential profile (or a high risk to re-offend upon release). Notably, over two-thirds of offenders (68.8%, n=302) convicted of a criminal organization offence were rated as having high reintegration potential, and thus being low risk to re-offend upon release (see Table 8).  Two-way contingency table analyses examining differences in RPP (low, medium, and high) across offence type (participating, committing, and instructing) found statistically significant group differences (χ² (4, N = 439) = 16.20, p < .01).  Multiple pair-wise comparisons of  independent proportions found that offenders with a low reintegration potential profile were significantly more likely to be convicted of participating in the activities of a criminal organization than committing a crime for a criminal organization (z = 3.15, p < .01).

Two-way contingency table analyses examining differences in RPP (low, medium, and high) across offender groups (criminal organization offender or matched sample) also produced significant group differences (χ² (2, N = 816) = 45.92, p < .001).  Multiple pair-wise comparisons of independent proportions found that criminal organization offenders were significantly more likely than the matched sample to have a high reintegration potential rating (68.8% vs. 47.7%; z = 6.10, p < .001), and that the matched sample group was significantly more likely than the criminal organization offenders to be rated as having low reintegration potential (23.3% vs. 8.9%; z = 5.68, p < .001).

Table 8. Reintegration Potential Profile (RPP) as a Percentage of the Criminal Organization Offence Types

Characteristics of Release to Community

Release Type and Community Engagement

Of the 451 offenders who had been admitted to the CSC with a criminal organization offence between April 25, 1997 and March 31, 2009, 73.6% (n=332) were released to the community.  The majority were released on day parole (51.8%, n=172) or statutory release (44.9%, n=149), with 3.3 percent released on full parole (n=11).  A breakdown of release types for the three criminal organization offence types is presented in Table 9^{Footnote 4}.  Two-way contingency table analyses examining differences in release type (day parole, full parole, and statutory release could not be interpreted due to insufficient cell sample sizes.  Two-way contingency table analyses examining differences in release type (day parole, full parole, and statutory release) across offender groups (criminal organization offender or matched sample) found no significant group differences.

Table 9. Release Type as a Percentage of the Criminal Organization Offence Types

Table 10 details the activity of those convicted of criminal organization offences, specifically their employment and program activity once released to the community.  Of those released criminal organization offenders who had community employment details in their files (94.3%, n=313), over three quarters were employed at some point during their release (76.4%, n=239), with only 17.6% (n=55) being unemployed throughout their release.  In addition, 14.8% of criminal organization offenders had participated in some sort of community intervention program during their release (n=49), with the most common programs including education, Counter-Point (an intervention designed to alter pro-criminal attitudes and associations with antisocial peers) and living skills programs. Interestingly, those convicted of instructing offences did not participate in many community programs, however every instructing offender participated in the Counter-Point program and most (75%) participated in educational programs in the community.

Table 10. Post-Release Activity as a Percentage of the Criminal Organization Offence Types

Note.  Program participation proportions will not sum to 100% as participants could participate in more than one community program.

Two-way contingency table analyses examining differences in employment status using release (employed, unemployed, student) across offence type (participating, committing, and instructing) and across offender group (criminal organization offender or matched sample) found no significant group differences between the types of criminal organization offenders, however, statistically significant group differences were found between the criminal organization offenders and the matched group (χ² (2, N = 568) = 12.78, p < .01).  Multiple pair-wise comparisons of independent proportions revealed that criminal organization offenders were significantly less likely than the matched sample to be unemployed during their release (17.6% vs. 29.0%; z = 3.24, p < .01).  Most analyses examining differences in program participation at release across offence type and across offender group could not be conducted due to inadequate sample sizes.  However, statistically significant differences were found between the criminal organization offenders and the matched group when compared on participation in substance abuse programs (c2 (1, N = 93) = 39.88, p < .001).  Multiple pair-wise comparisons of independent proportions found that criminal organization offenders were significantly less likely than the matched sample to participate in substance abuse programs in the community (4.1% vs. 65.9%; z = -6.31, p < .001).

Predictors of Release Outcomes

In order to examine the predictors of release outcome, it is first necessary to establish rates of failure upon release for the group of criminal organization offenders.  Of the 332 criminal organization offenders who were released to the community, 12.7% (n=42) were re-admitted to a federal institution by the end of the follow-up period (see Table 11).  The majority (76.2%, n=32) had a revocation of their release without an offence.  Few of these offenders were re-admitted with a new offence (14.3%, n=6).  None of the criminal organization offenders were re-admitted for a new criminal organization offence.

Table 11. Reasons for Re-admission as a Percentage of the Criminal Organization Offence Types

Next, a survival analyses using the Cox proportional hazards regression model was conducted to determine the risk of failure upon release and to estimate the effect of risk factors associated with a breach of the conditions of their release or new offence. Of particular interest was whether group membership (criminal organization offender or matched sample), gender, sentence length, age at release, type of release (day/full parole or statutory release), or offence type (participation, commission, or instruction) impacted the risk of failure.

Overall, 609 of the 902 offenders examined in the study had been released to the community during the follow-up period.  Of those offenders, 17.7% (108) were returned to federal custody, 38.9% (42) of which were criminal organization offenders and 61.1% (66) of which were members of the matched sample.  Of those 66 members of the general population, none were returned with a new offence, compared to six of the criminal organization offenders.  Figure 1 illustrates the survival curves for criminal organization offenders and those in the matched sample.  While the mean time to revocation of release for both groups was approximately the same (11 months for the matched sample and 10.6 months for the criminal organization group), the matched sample failed on release significantly more quickly than the criminal organization group.  It was 15 days before the first revocation of release for the matched sample, compared to three months for the criminal organization offenders.  At the mean revocation time (approximately 11 months), 8.7% of the criminal organization offenders had experienced a revocation of their release, compared to 16.0% of the matched sample group.  By the time the two groups had been on release for two years, these recidivism rates had grown to 17.3% of the criminal organization offenders versus 25.4% of the matched sample group.

Figure 1. Survival Functions at Mean of Covariates

Table 12 presents the results of the survival analyses.  Of the risk factors that were examined in this study, three were significantly related to failure upon release.  Belonging to the group of offenders with criminal organization offences was found to significantly decrease the hazard rate for failure upon release, as did being released at an older age, and being released on day or full parole rather than statutory release.  In other words, those offenders who were convicted of criminal organization offences or released later in life were less likely to fail during their release period, and those who were released on statutory release were more likely to fail during their release period.    

Table 12. Cox Regression Hazard Ratios for Risk Factors Related to Failure Upon Release

Note. **p < .01, ***p < .001.

Discussion

This research extends the existing literature surrounding criminal organization offenders by updating prior profiles while supplementing the body of knowledge in this area with new information on the conditional release of these offenders.  The updated profile replicates and extends prior work by Motiuk and Vuong (2005; 2006), and Nafekh and Stys (2004).  Similar to those studies, this research found that, compared to a matched sample, criminal organization offenders were significantly more likely to:

• come from the province of Quebec;
• have convictions for drug offences and attempted murder;
• have high needs in criminal associates and criminal attitudes domains;
• be low risk and have a high reintegration potential profile;

and were significantly less likely to have served a term of youth incarceration.

Further, this study illustrated the differences between those convicted under the three offences related to criminal organization activity: participation, commission, and instruction.  Specifically, those who were convicted of the least serious of the offences (participation) were found to have significantly higher risk and needs than those convicted of commission offences.  They were also more likely than those convicted of commission offences to have a low reintegration potential profile, and displayed higher need in the personal/emotional domain than did the other two criminal organization offence groups.  Interestingly, those who committed an offence for a criminal organization were significantly more likely than those who participated to have served a previous term of adult incarceration in a provincial or territorial facility.

The differences between those convicted of participation and commission offences under the criminal organization legislation warrants consideration for correctional operations.  While the case has been made in the past for specific programs aimed at the unique needs of “gang members” (Nafekh & Stys, 2004), the results of this research indicated that the programmatic and security requirements for criminal organization offenders vary depending on their status (or role) in the criminal organization.  Therefore, a single “criminal organization” approach or intervention would not meet the needs of all criminal organization offenders.  A focus on those convicted of participating offences, however, would be beneficial, especially in light of the finding that five of six of those re-admitted with a new offence after release were participation offenders.

Unique to this study was the focus on community outcomes for criminal organization offenders.  Interestingly, it was found that these offenders experience similar release patterns as the matched sample, with the majority being released on day parole or statutory release.  However, criminal organization offenders were significantly less likely than the matched sample to be unemployed after their release.  This is not surprising, as criminal organization offenders were also found to be significantly less likely to have some or considerable needs in the employment domain of the DFIA than did the matched sample.

Despite the fact that criminal organization offenders were found to be significantly less likely to have some or considerable needs in almost all domains of the DFIA as compared to the matched sample, analyses of program participation illustrated that criminal organization offenders were participating in a variety of community-based programs that might not necessarily address their domain needs as indicated by the DFIA.  For instance, 26.5% of criminal organization offenders were participating in living skills programs, although 97.1% of these offenders were found to have no need in the community functioning area. 

While it is promising that these offenders are participating in correctional programs that will aid in their reintegration into society, the risk principle (Andrews & Bonta, 2007) suggests that offenders who participate in unnecessary programming are actually being placed at greater risk for future re-offending.  This issue should be closely considered to evaluate whether the value of program participation outweighs the risk of future re-offending due to “over-programming” of an offender.  One area where correctional programs appeared to be appropriately used however, the Counter-Point program.  Analyses of need domains found that 99.3% of criminal organization offenders had some or considerable need in the area of criminal associates, and 88.3% had some or considerable need in the area of criminal attitudes.  These needs are specifically targeted in the community-based Counter-Point program, an intervention designed to alter an offender’s pro-criminal attitudes and associations with antisocial peers (Reintegration Programs Division, 2004).  Program referral for criminal organization offenders in this regard seems to have been successful to this point, and the results reported above suggest that should continue.

Community outcomes were also assessed through survival analyses in order to examine the risk factors for failure on release for criminal organization offenders.  Results illustrated that of the risk-factors examined, three were significantly related to failure on release: being younger at the time of release, being on statutory release, and being part of the matched sample group of CSC offenders.  In other words, those offenders who were released at an older age, those who were on day or full parole, and those who were convicted of a criminal organization offence were more likely to be successful upon release.  There are several reasons why criminal organization offenders may be less likely to fail upon release.  First, the innately secretive nature of organized crime may make re-offending more difficult to detect.  Second, criminal organization offences are often targeted as a part of long-term crackdowns of a particular gang or organization.  These sweeps are labour-intensive and require extensive planning.  As a result, it is possible that some of the follow-up periods after release did not allow for adequate time for such law enforcement initiatives, including subsequent arrests, to be reflected in the data.  Third, criminal organization offenders do not possess the same personal, emotional, employment, and family needs that place the matched sample at greater risk of re-offending.  With a greater community support system as well as higher levels of personal and emotional health, criminal organization offenders are simply not as high a risk to re-offend or be returned to custody for a breach of the conditions of their parole as the general offender population.  A fourth possible reason for the reduced recidivism of organized crime offenders is that programmatic interventions, such as the Counter-Point, are successful with this population.

One additional possible explanation for a significantly lower failure rate for criminal organization offenders is the changing face of criminal organization activity.  Organized crime   is increasingly making a transition from illegal activities (such as the gun or drug trade) to the infiltration of the legitimate economy.  Activities such as involvement in construction and financial markets has made organized crime more difficult to detect, track, and penetrate.  Coupled with a growing international approach to conducting organization crime enterprises and a more fluid, horizontal organizational network, the features of organized crime are evolving more quickly than the methods used for its control (Savona, 2009).  Thus, some criminal organization offenders who are returned to the community may be engaging in quasi-legitimate activities that are unlikely to result in detection, arrest, or prosecution. 

It is important to note that the criminal organization offenders examined in this study are not necessarily reflective of the Aboriginal, Asian, or street-gang population that is present in many Canadian institutions.  While 87.8% of the criminal organization offenders were identified during the OIA as being part of a gang, many of the gang members in Canadian institutions have not been convicted of criminal organization offences. As such, while the two populations are somewhat related, the profile of gang members and the profile of criminal organization offenders vary greatly, and one should be careful not to interchange the operational and personal needs of one group with another.  

Another limitation of this study is that the data on which these analyses are based originate from the OMS, a live database tracking the day-to-day activity of the Correctional Service of Canada.  Although rich with information and generally highly accurate, the OMS relies on the diligence of data entry personnel to ensure its accuracy, and sometimes suffers from the issues inherent in such an approach. 

Conclusions

This research contributed to the field of criminal organization research by updating the extant information on this population while examining the conditional release of these offenders.  Valuable insight was gained regarding the community activity of criminal organization offenders while identifying risk factors for conditional or statutory release failure.  The findings reflect an operational requirement to target the specific needs of criminal organization offenders, both in the institution and in the community, with a specific focus on those offenders convicted of participation offences.

Future research in this area should examine the post-release experiences of the organized crime offender population in a more in-depth way using data not found in the OMS.  While the OMS contains a wealth of information on offenders in the Federal penitentiaries, it has less information about offenders post-release.  A follow up study that interviewed those on release, their parole officers, as well as utilizing other community-based information systems (e.g., law enforcement) would further inform our knowledge of this offender population.  In addition, correctional programs and operations could benefit from future research examining the difference between gang-affiliated criminal organization offenders and those organized crime or gang-affiliates who have not been convicted of a criminal organization offence.  Likewise, future research on the emerging avenues of criminal involvement (e.g., the infiltration of legitimate businesses) would be beneficial for policy and program planning, and may aid in our understanding of criminal organization offenders.

Criminal organization offenders will always be a presence in Federal institutions.  With the introduction of more stringent and punitive legislation to sanction the activities of these individuals, the Correctional Service of Canada will likely see an increase in the number of criminal organization offender admissions, and can expect them to remain incarcerated for longer periods of time.  This study and future research can aid in preparing for this phenomenon through the identification of specific needs of this population, and the management of the risk that they pose to the institutions and the community.

References

Andrews, D. A., & Bonta, J.  (2006).  The psychology of criminal conduct – 4th ed.  Cincinnati, Ohio: Anderson Publishing. 

Correctional Service of Canada. (2007). Commissioner’s directive 705-6: Correctional planning and criminal profile. Ottawa, ON: Correctional Service of Canada.

Correctional Service of Canada. (2008). Commissioner’s directive 568-3: Identification and management of criminal organizations.  Ottawa, ON: Correctional Service of Canada.

Criminal Code of Canada, R.S. (1985), c. C-46 s. 467.1 – 467.13.

Criminal Intelligence Service Canada. (2008). 2008 report on organized crime.  Ottawa, ON:  Author.

Criminal Intelligence Services Canada. (2009).  Annual report 2009.  Ottawa, ON: Author.

Lieberman, B., ed. (1971). Contemporary problems in statistics. NY: Oxford.

Motiuk, L. L. (1997). Classification for correctional programming: The Offender Intake Assessment (OIA) process. Forum on Corrections Research, 9 (1): 18-23.

Motiuk, L. L., & Vuong, B. (2005). Federal offenders with criminal organization offences: A profile.  Research Brief B-38.  Ottawa, ON: Correctional Service of Canada.

Motiuk, L. L., & Vuong, B. (2006). Prison careers of federal offenders with criminal organization offences: A follow up.  Research Brief B-42.  Ottawa, ON: Correctional Service of Canada.

Nafekh, M., & Stys, Y. (2004).  A profile and examination of gang affiliation within the federally sentenced offender population.  Research Report R-154.  Ottawa, ON: Correctional Service of Canada.

Reintegration Programs Division. (2004).  Program description tables. Ottawa, ON: Correctional Service of Canada

Royal Canadian Mounted Police. (2005). Impacts of organized crime.  Retrieved Oct. 13, 2009, from  http://www.rcmp-grc.gc.ca/oc-co/impact-eng.htm

Royal Canadian Mounted Police. (2009). Organized crime.  Retrieved Oct. 13, 2009, from  http://www.rcmp-grc.gc.ca/oc-co/index-eng.htm

Savona, E.U. (2009, November).  Combining macro and micro approach to organized crime. Paper presented at the Crime Research Workshop on Best Practices to Combat Organized Crime, Ottawa.

Schneider, S. (2009).  Iced: The story of organized crime in Canada.  Mississauga, ON: John Wiley & Sons Canada, Ltd.