Roberts, Julian --- "Reducing the Use of Incarceration: a Review of Strategies" [2005] ALRCRefJl 4; (2005) 86 Australian Law Reform Commission Reform Journal 15

---

Reform Issue 86 Winter 2005

This article appeared on pages 15 – 18 & 71 of the original journal.

Reducing the use of incarceration: A review of strategies

By Julian Roberts*

Constraining rising—or reducing stable—prison populations remains a challenge confronting most Western nations.

Throughout the 1990s, prison populations rose in many common law jurisdictions, particularly England and Wales and the United States. This paper reviews some of the solutions that have been proposed or adopted around the world. Reducing prison populations in jurisdictions that employ a sentencing guidelines matrix is relatively straightforward; it consists of either moving more offences into the community sanctions zone of the grid, or reducing the sentence lengths prescribed by the guidelines. Matters are more complicated in common law countries that do not employ formal sentencing guidelines such as those found across the US. The following strategies have been developed to constrain the size of prison populations.

1. Statutory directions

Sentencers in common law jurisdictions have traditionally enjoyed wide discretion at sentencing, guided solely by direction from appellate courts. Over the past decade a number of legislatures have moved to curb this discretion. The most common attempt to restrict the use of custody has been to place certain principles on a statutory footing. Placing the principle of parsimony, or restraint, on a statutory footing serves a dual purpose. First, it should serve to inhibit judges from incarcerating offenders unless no community-based sanction will suffice. This, of course, is the primary purpose of the principle. However, if the legislature places its imprimatur on the principle of restraint it will be harder to introduce mandatory sentences of imprisonment at a later point, as these sentences clearly violate the principle.

The principle of restraint with respect to the use of custody has been codified in a number of countries. For example, in Canada, sections 718.2(d) and (e) of the Criminal Code state that:

‘An offender should not be deprived of liberty, if less restrictive sanctions may be appropriate in the circumstances; and all available sanctions other than imprisonment that are reasonable in the circumstances should be considered for all offenders, with particular attention to the circumstances of aboriginal offenders.’

Other jurisdictions such as England and Wales and New Zealand have also recently placed the principle of restraint on a statutory footing.¹ The language used in the New Zealand statute is particularly directive. Courts are instructed that:

‘When considering the imposition of a sentence of imprisonment for any particular offence, the court must have regard to the desirability of keeping offenders in the community as far as that is practicable and consonant with the safety of the community.’²

Thus codifying a general direction to sentencers regarding the parsimonious use of custody represents the most frequently adopted attempt to curb the size of the prison population. The principle of restraint is clear enough, but this step alone will prove insufficient, otherwise the problem of rising custody rates would be easily solved. Indeed, the experience in England and Wales illustrates this point well. The restraint provision was introduced in the Criminal Justice Act 1991. However, between 1991 and 2001, the custody rate in that jurisdiction rose significantly, as did the size of the custodial population.³

Codifying proportionality in sentencing

Many jurisdictions, including Canada, England and Wales, Finland and New Zealand have placed the principle of proportionality on a statutory footing. For example, in Canada, parliament has designated the following principle as fundamental: ‘A sentence must be proportionate to the gravity of the offence and the degree of responsibility of the offender.’ In Florida, the wording is: ‘The penalty imposed is commensurate with the severity of the primary offense and the circumstances surrounding the primary offense.’ The wording in the Finnish statute is comparable: ‘The punishment shall be measured so that it is in just proportion to the harm and risk involved in the offence and to the culpability of the offender manifested in the offence.’ By placing desert-based limits on the severity of the sentence that may be imposed, this principle will help to restrict the use of incarceration by preventing judges from employing harsher sentences (that is, more and longer terms of custody) in an attempt to curb rising crime rates.

2. Establish criteria for custodial sentences

A more forceful way of constraining the number of offenders sent to prison involves creation of specific criteria that must be fulfilled before a term of custody may be imposed. In Canada, the recent youth justice statute does exactly this. According to section 39 of the Youth Criminal Justice Act, a youth court may send a young offender to prison only if one or more of four criteria are met:

‘A youth justice court shall not commit a young offender to custody unless the young offender has:

• committed a violent offence; or

• failed to comply with previous non-custodial sentences; or

• committed an offence for which an adult is liable for a term of imprisonment greater than 2 years and has a history that indicates a pattern of findings of guilt; or

• ‘in exceptional circumstances’, the young offender has committed an indictable offence, the circumstances of which mean that the imposition of a non-custodial sanction would be inconsistent with the purpose and principles of sentencing.’

The youth justice reforms were introduced in Canada in 2003.⁴ Since then, there has been a significant decline in the volume of young persons admitted to custody. Although these criteria apply only to young offenders, there is no reason why criteria could not be created to restrict the imprisonment of adult offenders in a similar fashion.

3. Require justification of a custodial term

A weaker approach to creating specific criteria that must be fulfilled before an offender can be committed to custody consists of requiring judges to provide reasons for sentence.⁵ Many countries have created a statutory obligation on judges to provide reasons for the sentences that they impose. Such a requirement facilitates appellate review and is in the interests of the administration of justice. However, requiring judges to justify a term of custody may also help to lower the proportionate use of custody; judges may be less likely to impose a sentence that requires specific justification. Once again, the Youth Criminal Justice Act in Canada provides a useful illustration. Section 39(9) of the Act creates a duty for a youth court judge who imposes a term of custody to provide reasons why he or she ‘has determined that a non-custodial sentence is not adequate’ to achieve the purpose of sentencing ascribed to the youth court system.

4. Create alternative forms of custody

An obvious way to decrease the use of imprisonment as a sanction is to offer judges more sentencing options. This strategy has been embraced by many jurisdictions. One limitation on alternatives to imprisonment is that they are not as severe, or are not perceived to be as severe, as a term of custody. This limits the extent to which alternative sanctions can be substituted for terms of imprisonment. One solution is to develop less destructive forms of imprisonment, two of which are periodical (intermittent) custody, and home confinement.

The ‘lost weekend’: Intermittent sentences

Part-time or weekend custody is by no means a recent penal innovation. For example, periodic (that is, intermittent) detention has been part of the penal landscape in New Zealand since 1962. Many countries permit offenders to serve terms of imprisonment on a part time rather than continuous basis, this usually (but not always) means on weekends. All jurisdictions place a limit on the length of sentence that can be served on a part-time basis, with 90 days being the most common limit. The terminology varies from jurisdiction to jurisdiction, with ‘intermittent sentence of imprisonment’ and ‘periodic detention’ being the most common.

Home alone: Confinement sanctions

The search for meaningful alternative sanctions has led to the creation of another variation on imprisonment: home confinement, or community custody.⁶ The purpose of this sanction is to isolate the offender, rupture criminogenic associations, and promote steps to rehabilitation—only the first of these objectives is easily accomplished in prison. Home confinement regimes vary widely—some are quite punitive in nature. As well, the ambit of these sanctions varies considerably. Usually, the sanction is used to replace relatively brief periods of institutional confinement.

Community custody has long been a feature of the Finnish criminal justice system. Called conditional imprisonment, it has proved a success in that jurisdiction, and has played an important role in reducing the use of incarceration as a sanction. Over the past 50 years, the volume of conditional sentences imposed in Finland has increased dramatically. In 2000, 13,974 such dispositions were imposed, representing just under two thirds of all prison sentences. Finland remains the jurisdiction that has employed community custody sentences to the greatest extent. In 1950, conditional imprisonment accounted for 2,812 sentences, less than one third of all sentences of imprisonment. The volume increased steadily and exceeded the number of immediate custody sentences by 1980. This total rose steadily to a high of 17,000 in 1990.⁷ The most recent jurisdiction to introduce a community custody sanction is England and Wales. The Criminal Justice Act 2003 created a sentence called a suspended sentence of imprisonment.

Assignment to home detention in New Zealand requires a two-step approval process involving the judiciary and an administrative body. The consequence is that only a small minority of eligible offenders are released to serve their sentences at home. Only some offenders will be eligible for home confinement. Of these, only a minority will be granted leave to apply, and many will ultimately be turned down by the parole board. Less than one third of prisoners who have applied for home confinement have been granted release to the program.⁸ An even smaller percentage of all prisoners within the range of sentence length will serve part of the sentence in the program. In 2001, only 10% of offenders sentenced to a prison sentence of two years or less (and, therefore, within the ambit of the home detention regime) were actually released to serve their sentences at home.

In Canada, however, the home confinement sentence (known as a conditional sentence of imprisonment) may replace most sentences of custody of up to two years in duration.⁹ This embraces fully 96% of custodial sentences imposed. The introduction of this new (for Canada) sentence has achieved a reduction in the number of offenders admitted to custody. Analyses of admission statistics demonstrate that within three years, a 13% reduction in admissions was directly attributable to the new sanction. This represents about 55,000 offenders who served their sentences of imprisonment at home, rather than in a correctional facility. In addition, the success rate—the proportion of orders completed without violation of conditions—appears relatively high.¹⁰

5. Restrain ‘penal escalation’

Many judges follow what might be termed a sentencing strategy of ‘penal escalation’. If an offender receives a non-custodial sanction and is subsequently re-convicted, judges tend to gravitate towards a more severe disposition on the second or subsequent occasion. This is a form of recidivist premium; the judicial logic underlying the strategy is that if a community based sanction did not ‘work’ on the first occasion (as evidenced by the offender’s re-appearance before the court), perhaps custody is the answer on the second.

Restricting this tendency by courts represents a way of containing the number of prison sentences imposed. A provision in the Youth Criminal Justice Act in Canada is intended to discourage judges from escalating the severity of the sentence in response to subsequent offending. Having imposed an alternative to custody for one offence, some judges shift to custody if a youth re-appears before the court, reasoning that the first sentence was insufficient to discourage the offender. Section 39(4) addresses this judicial reasoning, providing:

‘The previous imposition of a particular non-custodial sentence on a young person does not preclude a youth justice court from imposing the same or any other non-custodial sentence for another offence.’

While s 39(4) does not prohibit judges from following the ‘step principle’ logic at sentencing, the provision makes it clear that the same alternative may be imposed on separate occasions.

6. Create a sentencing commission

Attempting to achieve an important policy goal through legislative interventions alone is far from easy. The process of amending existing statutes is time consuming and susceptible to political pressures. For this reason, a number of jurisdictions have established temporary or permanent sentencing commissions to guide the reform process. These commissions have much to contribute to the policy goal of reducing admissions to custody. First, they create a policy ‘buffer’ between the legislature and the sentencing process. For example, a commission comprised of judges, criminal justice professionals and scholars is unlikely to pursue polices such as mandatory sentencing in response to populist pressures.

The principal cause of the creation of mandatory sentencing laws in Western nations during the 1990s would appear to be the influence of penal populism: politicians responding to pressure to ‘get tough’ and ‘do something about crime’.¹¹ Second, a commission can more rapidly access and utilise relevant statistical information, for example pertaining to the prison population. Third, a commission can draw upon the expertise of professionals, whereas parliament is guided by standing committees composed of politicians who do not necessarily have the necessary experience to evaluate draft legislation pertaining to the functioning of the sentencing process.

Sceptics may argue that independent commissions are somehow undemocratic, and that reform of the sentencing process should remain firmly within the grasp of parliament. If properly constructed, however, sentencing commissions supplement and enhance the work of parliament, without usurping parliamentary authority. Sentencing commissions exist throughout the United States at the state and federal levels. In addition, a number of other countries such as Canada and Belgium have created temporary commissions to review the sentencing process and create proposals for reform.

Conclusions

I conclude this brief review by summarising the elements that comprise a successful strategy to reduce the use of incarceration as a sanction.

(1) A codified statement of the purpose of sentencing that privileges a retributive rather than a crime control orientation to sentencing.

(2) Creation of a permanent sentencing commission to propose and evaluate specific statutory and non-statutory reforms.

(3) A clear statutory statement of sentencing principles that includes the principle of restraint regarding the use of custody.

(4) Statutory directions to discourage ‘penal escalation’, namely the imposition of a disproportionate sanction to reflect the fact that the offender has a previous conviction that resulted in imposition of a non-custodial sanction. In addition, courts need to be discouraged from ‘cumulative sentencing’, that is the practice of imposing progressively more severe sanctions to reflect the number and seriousness of the offender’s previous convictions.

(5) Statutory directions should direct judges to follow the principle of the ‘progressive loss of mitigation’ regarding prior convictions. Enhanced sentences for recidivists may be allowed for offenders who by virtue of their records are ‘exceptional’ and, therefore, not subject to the conventional sentencing standards.

(6) A wide range of alternative sanctions that offers judges real alternatives to the use of custody.

(7) Specific criteria that must be met before an offender may be sentenced to a term of imprisonment.

(8) A home confinement sanction that is sufficiently punitive to serve as a replacement for terms of custody up to six months.

(9) Courts should be required to give reasons for the sentences that they impose, or at least to give reasons whenever a term of custody is imposed.

* Professor Julian Roberts is a Reader in Criminology at the Centre for Criminology, University of Oxford.

This paper is an abridged version of a paper presented at the 2005 Conference of the Association of Law Reform Agencies of Eastern and Southern Africa. A longer version is available from the author at: [e-mail].

Endnotes

1. See J Roberts, ‘An Analysis of the Statutory Statement of the Purposes and Principles of Sentencing in New Zealand’ (2003) 36 Australian and New Zealand Journal of Criminology 249.

2. Sentencing Act 2002 (NZ), s 16(1).

3. M Hough, J Jacobson and A Millie, The Decision to Imprison: Sentencing and the Prison Population (2003).

4. See J Roberts, ‘Sentencing Juvenile Offenders in Canada: An Analysis of Recent Reform Legislation’ (2003) 19 Journal of Contemporary Criminal Justice 413.

5. In some American sentencing guideline systems judges are obliged to give reasons only when they impose a sentence outside the range of sentence prescribed by the guidelines.

6. For a review see J Roberts, The Virtual Prison: Community Custody and the Evolution of Imprisonment (2004).

7. T Lappi-Seppala, The Principle of Proportionality in the Finnish Sentencing System (2002).

8. A Gibbs & D King, ‘Home detention with electronic monitoring: the New Zealand experience’ (2003) 3 Criminal Justice: The International Journal of Policy and Practice 199.

9. There are other criteria that must be fulfilled before a court can impose this sentence. Imposition of the sentence must be consistent with the codified purpose and principles of sentencing; the offence must not be one that carries a minimum term of imprisonment (few such crimes exist in Canada) and the court must be confident that the offender does not represent a threat to the community. See J Roberts, The Virtual Prison: Community Custody and the Evolution of Imprisonment (2004).

10. See, ibid. These are the ‘official’ statistics; clearly an unknown proportion of offenders serving their sentences in the community violate their conditions without being detected. The true success rate must be somewhat lower than these data suggest.

11. See J Roberts et al, Penal Populism and Public Opinion: Lessons from Five Countries (2003).