• the development of a management model for long-term sentences;
• staff training to meet the specific needs of long-term offenders; and
• the development of programming for long-term offenders.

Supporting these recommendations, recent research has underscored the unique nature of this offender group and has argued that innovative methods of service delivery and programming opportunities are necessary to meet the needs of long-term offenders adequately.

This article strives to provide a clearer picture of the characteristics of the long-term offender population, based on an overview of available statistical information. It should be noted that this is a diverse group of offenders. In this article, the term "long-term offenders" refers to offenders serving life sentences, indeterminate sentences and determinate sentences of 10 years or more. A sampling of criminal offences yielding long-term sentences includes:

• life sentences - first-degree/capital murder, second-degree/non-capital murder, manslaughter, attempted murder, etc.;
• indeterminate sentences - dangerous offender, dangerous sexual offender and habitual criminal designations and commitments on a Lieutenant Governor's Warrant; and
• determinate sentences of 10 years or more - aggravated sexual assault, robbery, kidnapping and abduction, etc.

Long-term offenders comprise about one quarter (25.3%) of the total federal-offender population (including incarcerated offenders and those on some form of release). On 31 January 1992, there were 22,121 offenders under the jurisdiction of the Correctional Service of Canada; of these, 5,595 were serving long-term sentences.

Almost three out of five long-term offenders (57.5%) were incarcerated, while the rest were on some form of release.

Almost three of every five long-term sentences are either life sentences for first- or second-degree murder, or life or indeterminate sentences for some other offence.

Specific types of long-term sentences, in decreasing order of frequency, are:

• determinate sentence of 10 years or more -41.4% of long-term offenders (10.5% of the total offender population);
• second-degree/non-capital murder -37.9% of long-term offenders (9.6% of the total offender population);
• other life and indeterminate sentences - 10.8% of long-term offenders (2.7% of the total offender population); and
• first-degree/capital murder -9.8% of long-term offenders (2.5% of the total offender population).

Just over one quarter of the incarcerated offender population (27.8%) is serving a long-term sentence (excluding those on day parole). On 31 January 1992, 3,449 inmates were serving long-term sentences under the jurisdiction of the Correctional Service of Canada.

Most long-term inmates (90.7%) are housed in either medium- or maximum-security institutions:

• maximum security -45.9% of all long-term offenders;
• medium security -44.8%;
• minimum security -6.3%; and
• community correctional centres or provincial jails - 3%.

About one of every five offenders on conditional release (22%) is a long-term offender.

On 31 January 1992, there were 2,146 long-term offenders on conditional release. The offence or sentence breakdown of these offenders, in decreasing order of frequency, is:

• determinate sentence of 10 years or more -46.7% of long-term offenders on conditional release (of these, 18.6% are on day parole, 64.4% on full parole and 17% on mandatory supervision);
• second-degree murder - 39% (of these, 21.6% are on day parole and 78.4% on full parole);
• other life and indeterminate sentences - 10.2% (of these, 14.7% are on day parole, 81.4% on full parole and 3.9% on mandatory supervision); and
• first-degree murder -4.2% (of these, 4.7% are on day parole and 95.3% on full parole).

There are marked differences in the distribution of long-term offenders across the regions. Quebec and Ontario have larger proportions of long-term offenders than other regions.

However, when we compare the proportion of long-termers with the proportion of short-termers (i.e., those serving a sentence of less than 10 years) in each region, we see that Quebec and the Pacific region have proportionately more long-term offenders, and the Atlantic and Prairie regions have proportionately fewer long-termers (see Figure 1).

• Atlantic -5.4% of long-term offenders versus 9.7% of all offenders (proportionately fewer long-termers);
• Quebec - 34.6% of long-term offenders versus 29.8% of all offenders (proportionately more long-termers);
• Ontario -27.7% of long-term offenders versus 26.6% of all offenders (approximately the same proportion of each);
• Prairies - 15.6% of long-term offenders versus 20.9% of all offenders (proportionately fewer long-termers); and
• Pacific - 16.6% of long-term offenders versus 12.9% of all offenders (proportionately more long-termers).

As Figure 2 shows, there was a 41.5% increase in the number of long-term offenders under the jurisdiction of the Correctional Service of Canada from 1981 to 1991. During this same period, however, the total incarcerated population grew in a similar manner. On 31 December 1981, there were 2,672 long-term offenders in federal institutions, representing 26.4% of the inmate population. Ten years later, on 31 December 1991, long-termers still represented 26.4% of all inmates; however, the total number of long-termers had grown to 3,782.

From 1981 to 1991, the number of annual admissions of long-term offenders increased from 382 admissions to 453. However, it appears that proportionately fewer long-term offenders are being admitted to federal institutions than 10 years ago: in 1981, long-term offenders represented 7.3% of a total 5,248 admissions but by the end of 1991, this percentage had decreased to 6.4% of a total 7,021 admissions.

From 1981 to 1991, the number of annual releases of long-term offenders increased dramatically from 265 to 448. In fact, it appears that we are releasing proportionately more long-term offenders: in 1981, long-term offenders represented 5.6% of a total 4,754 releases, but by 1991, this percentage had increased to 7% of a total 6,392 releases.

The number of previous federal incarcerations of long-term offenders and short-term offenders (i.e., those serving sentences of less than 10 years) is strikingly similar. The majority of both groups have no previous federal incarcerations.

On 31 January 1992, the federal incarceration history of our offenders was as follows:

• no previous federal incarcerations -58% of long-termers versus 60.1% of short-termers;
• one previous federal incarceration -18.9% of long-termers versus 16.1%
• of short-termers; and
• two or more federal incarcerations -23.1% of long-termers versus 23.8% of short-termers.

However, as Figure 3 shows, when we divide long-term offenders into those serving life or an indeterminate sentence and those serving a determinate sentence of 10 years or more, we find that those serving a life sentence or an indeterminate sentence are less likely to have two or more previous federal incarcerations.

A group of 294 long-term offenders released in 1986 was followed over a five-year period. Of every five of these long-term offenders released in 1986:

• three did not have any readmissions or reconvictions (58%);
• one was readmitted for technical violations of parole (19.7%); and
• one was readmitted with new offences (22.8%).

The readmission rate for long-termers serving determinate sentences (i.e., sentences of 10 years or more) was higher than for long-termers serving life or indeterminate sentences (50.8% versus 28.1%). Of the 75 murderers who were released, only 11 (14.6%) were subsequently reconvicted of a criminal offence. Although some of these offenders were convicted of serious crimes such as aggravated sexual assault and attempted murder, none received subsequent convictions for either murder or manslaughter.

On 31 January 1992, there were 491 female offenders under the jurisdiction of the Correctional Service of Canada, representing 2.2% of the total federal-offender population. Just under one quarter of these women (24%) were serving a long-term sentence.

The proportion of males and females serving long-term sentences is almost identical:
·long-term sentences -24% of female offenders versus 25.3% of male offenders;
·sentences of less than 10 years -76% of female offenders versus 74.7% of male offenders.

However, as Figure 4 shows, it seems that female offenders are more likely than males to be serving sentences for murder, whereas males are more likely to be serving other life and indeterminate sentences or determinate sentences of 10 years or more.

On 31 December 1991, the average long-term offender was almost 38 years old. The oldest long-term offender was 80 years old and the youngest was 17. Interestingly, long-term offenders as a group appear to be aging: between 31 December 1981 and 31 December 1991, the average age of long-term offenders increased by almost three years, from an average of about 35 years to almost 38 years of age.

As Figure 5 illustrates, while the average age of offenders sentenced to other life and indeterminate sentences has remained relatively unchanged, the average age of offenders convicted of first-degree murder, second-degree murder and other determinate sentences has increased. In particular, the average age of offenders convicted of first-degree (capital) murder has increased from about 32 years to 38 years.

On 31 January 1992, the overwhelming majority of long-term offenders (almost 85%) were Caucasian. The remaining group consisted of inmates from diverse ethnic backgrounds, including native people, Asiatics, blacks and others. The distribution of long-term offenders across ethnic backgrounds is similar to the ethnic distribution for the total offender population, although there was a somewhat higher proportion of Caucasians and a somewhat lower proportion of native people in the long-term offender group.

The ethnic breakdown, in decreasing order of frequency, is:

• Caucasian - 84.7% of long-termers versus 80.2% of the total offender population;
• native -6.7% of long-termers versus 9.8% of the total offender population:
• black - 2.5% of long-termers versus 3.7% of the total offender population;
• Asiatic - 1.4% of long-termers versus 0.9% of the total offender population; and
• other (including not specified) -4.7% of long-termers versus 5.4% of the total offender population.

This distribution suggests that members of diverse ethnic groups (i.e., non-Caucasians) are not overrepresented in the long-term offender population.

Available data indicate that about half of incarcerated offenders reported their marital status as single. Sentence length appeared to have no bearing on this finding. However, long-term offenders appeared somewhat less likely than other offenders to be involved in common-law relationships.

The breakdown of marital status, in decreasing order of frequency, is:

• single - 50.5% of long-term inmates versus 47.5% of short-term inmates;
• common-law -21.5% of long-term inmates versus 28.6% of short-term inmates;
• married- 13.5% of long-term inmates versus 11.9% of short-term inmates;
• separated or divorced - 11.1% of long-term inmates versus 10.6% of short-term inmates; and
• other (including not specified) -3.4% of long-term inmates versus 1.4% of short-term inmates.

About one quarter of the total federal-offender population is serving a long-term sentence (i.e., 10 years or more). This is true of both male and female offender populations. Three out of five long-termers are incarcerated and two out of five are on some form of conditional release.

Quebec and the Pacific region have proportionately more long-term offenders, while the Atlantic and Prairie regions have proportionately fewer. Ontario has a more equitable proportion of long-termers.

During the past 10 years, the number of long-term offenders under federal jurisdiction increased by the same proportion as the number of federal offenders in general. During this same period, federal corrections admitted proportionately fewer long-termers, and released proportionately more long-termers, than offenders in general.

The vast majority of long-termers are Caucasian. About half of all long-term offenders are single, while about one in three is married (includes common-law). During the past 10 years, the average age of long-term offenders has increased by almost three years and is now about 38 years. Offenders serving life sentences for first-degree murder as a group, show the most dramatic increase in age.

Long- and short-term offenders have similar histories of federal incarceration, with the majority of both groups having no previous federal incarceration. After a five-year follow-up, only about one in five long-term offenders had been reconvicted of a criminal offence, while none of the 75 released offenders serving life sentences for murder had been subsequently reconvicted of murder.