6.1 This chapter looks at the question of whether existing methodologies and principles that guide the exercise of judicial discretion in sentencing are adequate, or need reform or improvement. These techniques vary considerably in the degree of restraint they impose upon judicial discretion. They range from initiatives such as sentencing information (which aims to improve the information to be used within the existing sentencing process), to complex guidelines which substantially remove the scope for discretion to be exercised. Discussion of the various methods which can be used to guide judicial discretion cannot be isolated from the discussion of general sentencing principles in Chapter 3. For example, principles such as imprisonment is a sanction of last resort operate as a guide to sentencing discretion independently of the methods discussed here.

6.2 The interest in techniques for guiding judicial discretion in sentencing stems largely from a perceived failure to achieve consistency in sentencing, one of the objectives to which the sentencing system is dedicated.¹ To the extent to which sentencing disparity ought to be a matter for concern (a question which, as we have pointed out in Chapter 2, is extremely controversial)² it is of course desirable to identify approaches to sentencing which may improve sentencing consistency.

6.3 The Commission’s general approach to evaluation of the techniques in this Chapter is to balance the supposed improvement of consistency of sentence outcome against the ability to account for the many factors which are relevant to the determination of sentences in individual cases.³ Our tentative conclusion is that the methods of guiding sentencing discretion which have been suggested, or in some cases trialled in overseas jurisdictions, generally fail to account adequately for the relevant factors which must be taken into account in determining sentences in individual cases. The premise is, of course, that judicial discretion remains an invaluable method of ensuring justice in individual cases, and this discretion must be retained.

METHODS CURENTLY USED BY THE COURTS TO ACHIEVE CONSISTENCY OF SENTENCING

Sentencing appeals

6.4 The High Court, as the highest court of appeal, has shown itself to be very reluctant to intervene in the sentencing process. Special leave is required before an appeal to the High Court can occur, and the Court has imposed a heavy burden upon a party seeking to obtain it. In Lowe v The Queen, Chief Justice Gibbs stated:

[T]o warrant the grant of special leave to appeal against sentence when there has been no want or excess of jurisdiction, it must appear that the case involves some question of law or principle of general importance or that there has been a gross violation of the principles which ought to govern discretion in imposing sentence.⁴

6.5 The High Court is especially reluctant to interfere with sentencing decisions which have been made under legislative provisions which are peculiar to a particular jurisdiction. The construction of provisions “with no precise counterpart in other States” should usually be determined by the particular State courts.⁵

6.6 This makes the Court of Criminal Appeal the final court of appeal against sentence in the vast majority of cases. The principles of sentencing are developed at common law by appeal to the Court of Criminal Appeal. Generally, the legislature provides only limited guidance to sentencing and appeal courts in the form of statutory maximum penalties for particular offences. Although such maxima are the first point of reference for judges and magistrates,⁶ they provide very little guidance when selecting an appropriate penalty in a particular case.

6.7 Appellate review limits the exercise of sentencing discretion in two ways. First, it provides a mechanism for challenging a particular sentencing decision. Secondly, it generates a jurisprudence of sentencing for guidance in future sentencing decisions.⁷ The primary principle governing appeals against sentence is that the Court of Criminal Appeal “does not simply embark upon the task of sentencing afresh, substituting its own opinion for that of the sentencing judge”.⁸ Appellate interference is warranted only by a material error of law or fact by the sentencing judge. Such an error may appear in the reasons given by the trial judge, or may be disclosed by the manifest excess or inadequacy of the sentence imposed.

6.8 The chief advantages of appellate review as a method of guiding judicial discretion are flexibility, and the fact that judges and counsel are obviously experienced in this procedure.⁹ It must be pointed out that several other strategies for improving consistency in the exercise of sentencing discretion are currently incorporated into appellate review. Judicial education, and a sophisticated Sentencing Information System, both of which are discussed below, are two methods which are currently available to both sentencing and appeal judges in New South Wales.

6.9 The key disadvantage of relying upon sentencing appeals to ensure sentence consistency is that the process is purely reactive.¹⁰ The Court of Criminal Appeal can articulate principles of sentencing (including information about appropriate sentence length) only in cases which are appealed to it. Sentencing jurisprudence is also far less developed than case law on many other legal subjects, providing less practical guidance than decisions in these areas.¹¹ As serious crimes and extreme sentences predominate in such appeals, the coverage of sentencing law can be incomplete, particularly for the many minor offences (and non-custodial sentencing options) dealt with in Local Courts. This poses significant structural difficulties in changing sentencing values.

Sentencing information

6.10 The Judicial Officers Act 1986 (NSW) established the Judicial Commission of New South Wales. Amongst other purposes, the Act was passed “to confer on the Commission functions relating to sentencing consistency”.¹² The relevant provision is s 8:

(1) The Commission may, for the purpose of assisting courts to achieve consistency in imposing sentences:

(a) monitor or assist in monitoring sentences imposed by courts; and

(b) disseminate information and reports on sentences imposed by courts.


(2) Nothing in this section limits any discretion that a court has in determining a sentence.

6.11 These functions are primarily discharged by the introduction and maintenance of the Sentencing Information System (“SIS”), a large computerised database providing judicial officers with a wide range of qualitative and quantitative information about sentencing practices in New South Wales. The SIS contains four separate components:

• Sentencing law;
• Sentencing statistics;
• Sentencing facilities; and
• Sentencing calculator.

Sentencing law component

6.12 Sentencing Law contains narrative text on sentencing principles, case law and legislation. Most of the cases are unreported judgments of the New South Wales Court of Criminal Appeal, although some leading cases from other jurisdictions are also included. The Sentencing Law component contains a powerful full text retrieval system, which allows the user to locate, view on the screen, print out or download any document contained in the database. In addition, documents which are logically related to each other have been electronically linked so that the user can easily move between documents. The component also contains examples of the forms of order which may be appropriate, or tailored to suit, the circumstances of the particular case.

6.13 Sentencing Law is divided into a number of subsets:¹³

1. Full text of judgments. This contains over 2300 unreported judgments of the Court of Criminal Appeal which can be read and downloaded into a word processor or document, or printed as required.

2. Case summaries. This contains almost 1800 short summaries of relevant facts of Court of Criminal Appeal and High Court cases concerning sentencing.

3. Principles of sentencing. This is narrative text written by the Judicial Commission which summarises sentencing law and provides citations to the leading authorities supporting the particular principle.

4. Purple passages. This contains selected pages of judgments carefully extracted as the essence of a proposition of sentencing. They are similar to “quotations”, providing a quick way of seeing whether the particular case is worth further investigation, and are frequently the more “oft-quoted” extracts of judgments.

5. Sentencing options and orders. This is narrative text which discusses each sentencing option and outlines possible uses, constraints, and interactions with other options in all jurisdictions.

6. Current NSW legislation. This contains the full text of 24 pieces of current State legislation. Legislative change is recorded on the SIS to form a historical collection of amendments, allowing access to repealed sections of legislation.

7. Current Commonwealth legislation. This includes full text of 25 pieces of current Commonwealth legislation, also containing historical information about the legislation.

Sentencing statistics component

6.14 This database contains statistical data on sentences imposed in the Local Court, the higher courts and the Children’s Court. Only statistics resulting from the most recent data are displayed, except where there are fewer than five cases of a particular offence. In such cases, data is retained to ensure that statistics relating to unusual and uncommon offences are as comprehensive as possible.

6.15 The user is able to select:

• the relevant jurisdiction;
• the Act and section number of the principal offence;
• whether other offences are to be taken into account;
• whether the offender is an individual or a corporation;
• and the number of counts of the principal offence.

It is also possible to define specific offender characteristics such as the plea, age, prior criminal record and liberty status of the offender at the time of the commission of the offence.¹⁴

6.16 The information is displayed in a number of tables and graphs which indicate the frequencies in terms of:

• the various kinds of dispositions;
• pecuniary value of fines;
• minimum or fixed custodial terms; and
• full custodial terms imposed.

6.17 The Commission is of the view that the statistics database is deficient in two respects:

• The omission of decisions of the Court of Criminal Appeal in the statistics (a deficiency which will be remedial in the next generation of SIS).
• The inability to identify from the statistics graph individual cases (especially those from the higher courts) on which the graph is based.

Facilities component

6.18 The Facilities component provides details as to the availability of drug and alcohol counselling, periodic detention and community service work, cross-referenced by geographic location and service type. The database includes separate directories of adult and juvenile facilities containing about 800 documents. Each document is in a standard format providing the street address, postal address, telephone and fax numbers, operating hours, services offered, intake policy and any other special services. This component was designed in co-operation with a number of government agencies to facilitate the regular and timely updating of information. The agencies include: the Department of Juvenile Justice, the Department of Corrective Services, the Probation Service, the Attorney General’s Department and the Directorate of the Drug Offensive.

Calculator

6.19 The Sentencing Date Calculator is designed to assist judicial officers to calculate dates required when imposing a custodial sentence.¹⁵ The calculator provides the expiry date of the minimum or fixed term, the expiry date of the additional term and the length of minimum term. In a situation where the statutory formula is not followed, the judicial officer is warned that “special circumstances must exist”.¹⁶

Bench books and references

6.20 The Judicial Commission’s on-line database also contains bench books and references. The bench books are electronic versions of Criminal Trial Courts Bench Book, Local Courts Bench Book and Civil Trials - Judicial Essays and Working Papers. The references subset contains works relevant to contemporary judicial issues, including sentencing. For example, a recent article entitled Changes in Evidence Law and Practice - An overview by J D Heydon, has been included. Documents contained in the Bench Books and Reference component are electronically linked to the Court of Criminal Appeal Judgments, and to the New South Wales and Commonwealth legislation subsets where relevant.

Evaluation

6.21 The SIS is designed to inform judges and magistrates about penalties without imposing any particular option or duration of sentence for a particular case. An appeal against sentence is not affected by the availability of the SIS. Appeals are based upon the facts of the case and the application of the relevant law, not on the information presented in the SIS. The SIS merely seeks to encourage sentencing consistency by making an increased volume of reliable empirical information easily available to judges and magistrates. It is designed to “assist and inform sentencers when contemplating future decisions”.¹⁷

6.22 It is important to note that the SIS does not seek to fetter the discretion of the judicial officer in choosing the appropriate penalty or the quantum of penalty. While the SIS displays the range of penalties imposed and the frequency with which any particular penalty is imposed in a manner which is fairly sensitive to the nature of the offence and of the offender, the factors which it builds into the sentencing equation do not exhaust the large number of factors that are relevant when sentencing an offender.¹⁸ For example, the motive for the crime, the range of aggravating and mitigating circumstances, and the health of the offender are not taken into account by the SIS. The assumption is that the exercise of judicial discretion necessarily remains important to the task of sentencing. This accords with the Commission’s premise in evaluating techniques designed to assist judicial officers in sentencing.¹⁹

6.23 The ability of the SIS to improve sentencing consistency is limited by:

• the accessibility of SIS to judicial officers;
• the availability of empirical evidence;
• the extent to which judicial officers use it; and
• the extent to which judicial officers choose to follow the guidance offered by the system.

6.24 Accessibility of SIS. The overwhelming majority (at least 95%) of judges and magistrates sitting in criminal jurisdiction have access to the SIS from court locations throughout New South Wales. The Attorney General’s Department has provided all judges of the Supreme and District Courts with the necessary hardware and software to access the SIS. The Judicial Commission has provided the same equipment to 109 magistrates (95% of the total magistracy in New South Wales). The remaining 5% of magistrates will be similarly equipped by the middle of 1996. Judicial officers can access the SIS in a number of ways. These include direct connection (Ethernet and X25) from some court buildings in metropolitan Sydney or modem access from home or chambers where a direct connection is not available.²⁰

6.25 The SIS is easy to use even for those without much experience of computers. It will become even easier to use when the next generation of SIS comes on line in the middle of 1996. This generation moves from the current keyboard-driven menu system to a Windows format, utilising internet technology to allow users to retrieve information in a more intuitive manner using hypertext links. The graphical user interface will allow users to view cases and text material simultaneously with statistics, bench books or other relevant material.

6.26 Availability of empirical evidence. SIS is driven by the data which is fed into it. Inevitably, this means that for less common offences, or for less common offender characteristics in relation to common offences, the information-base will be small. In no way does the Commission regard this as the “fatal flaw” of SIS.²¹ First, the database retains, in principle, its value for common offences for which data do exist. Secondly, the enterprise is ongoing and data regarding the less common offences and offender characteristics will, inevitably, be forthcoming.

6.27 Usage of SIS. References to SIS in judgments make it clear that judicial officers and lawyers are making use of the system.²² However, it is difficult to draw any conclusions from usage statistics (which are not limited to judicial officers) except that, at present, far greater use is made of the sentencing law component of SIS than of sentencing statistics. Between July 1994 and November 1995, on-line enquiries to the sentencing law component of the SIS tended to average slightly less than 1,000 per month (though monthly usage is uneven), while enquiries to the statistical component tended to be less than a couple of hundred per month.²³ The Commission is in no doubt that as the database expands and becomes more comprehensive, usage of the SIS will increase correspondingly.

6.28 Usefulness of SIS. References to the SIS in judgments sometimes comment on the usefulness of the system. Thus, in Maguire, Justice Grove said:

[I]nformation of the kind contained in the sentencing information data base ... can often be useful in determining whether a particular sentence should be characterised as manifestly excessive or indeed manifestly inadequate.²⁴

More generally, Justice Carruthers has pointed to the likely greater use of statistical evidence in sentencing hearings in the future. His Honour said:

I daresay that the Court will hear, in the future, a good deal more about the Sentencing Information System and the Sentencing Information Service and about statistical graphs and other statistical material of the kind now before this court.²⁵

6.29 Review of SIS and other matters. The Judicial Commission has commenced a formal evaluation of SIS at two levels - first, its effect on sentencing practice and, secondly, its benefits as expressed by its users. That evaluation will be completed in late 1996. Meanwhile, the Commission is of the view that any serious consideration of mandatory or presumptive sentencing schemes should be suspended until the impact of the SIS on sentencing decisions has been evaluated.²⁶

6.30 More generally, the Commission is in no doubt that the SIS is a valuable addition to the tools available to a judicial officer in arriving at sentence in the instant case. At one level, the system provides the officer with easy access to information which is difficult to find or which is easily overlooked. At a more sophisticated level, SIS serves as a guide to the exercise of judicial discretion without in any way acting as an unacceptable fetter on that discretion. The Commission believes that every effort should be made to maintain and improve what will no doubt eventually become an indispensable aid to sentencing in New South Wales.

Judicial education

6.31 Another technique which may be used to encourage greater consistency in sentencing is judicial education. Judicial education may use a variety of techniques, such as examinations of recent statutory and case law developments; case studies and sentencing exercises; and analysis of the philosophical principles of sentencing.²⁷

6.32 The Judicial Officers Act 1986 (NSW) provides for the Judicial Commission of New South Wales to “organise and supervise an appropriate scheme for the continuing education and training of judicial officers.”²⁸ In organising the scheme of judicial education, the Judicial Commission must:

(a) endeavour to ensure that the scheme is appropriate for the judicial system of the State, having regard to the status and experience of judicial officers;

(b) invite suggestions from and consult with judicial officers as to the nature and extent of an appropriate scheme;

(c) have regard to the differing needs of differing classes of judicial officers and give particular attention to the training of newly appointed judicial officers; and

(d) have regard to other matters as appear to the Commission to be relevant.²⁹

The guiding principle of the Judicial Commission’s education policy is the provision of assistance to judicial officers “by enhancing professional expertise, facilitating development of judicial knowledge and skills and promoting the pursuit of juristic excellence.”³⁰

6.33 The Judicial Commission discharges these functions by providing conferences and seminars, producing a range of publications, and conducting computer training for judicial officers. Sentencing has been one of the most popular topics in the judicial education programmes of many jurisdictions.³¹ During the 1994-95 financial year, the Judicial Commission provided one specialist seminar on sentencing, although sentencing issues were considered in many of the general conferences made available to judicial officers during this period.³² As part of its computer training programme, the Judicial Commission provides both introductory and advanced instruction on the use of the Sentencing Information System (SIS).³³ In addition, the Judicial Commission has recently revised the sentencing law chapters of the Bench Books of both the Criminal Trial Courts and Local Courts in New South Wales.³⁴

6.34 While sentencing issues feature in the educative function of the Judicial Commission, sentencing is only one of many issues that the Judicial Commission must address. If a sentencing education programme designed to improve sentencing consistency were introduced as one of the Judicial Commission’s functions, it would possibly require specialist consideration by a discrete section of the Commission. This is implicit in the suggestion of the Attorney General’s Sentencing Review that a “Sentencing Policy Advisory Council” might become part of the Judicial Commission, if the establishment of such a council were warranted.³⁵ It has been suggested that nothing less than a “permanent infrastructure” is required to support judicial officers in the complex task of sentencing.³⁶

6.35 The belief that sentencing skills may be taught is implicit to judicial education programmes.³⁷ Although the fundamental values which underpin common law principles of sentencing may be effectively taught, it is not certain that this will produce improved consistency in sentencing. Fox has identified three major weaknesses in judicial education as a strategy for addressing sentencing inconsistency. First, the lack of formal assessment makes it difficult to assess how much is learnt, and how long it endures. Secondly, education is unlikely to encourage a fundamental re-assessment of sentencing values or the system as a whole. Thirdly, judicial education lacks the means of ensuring compliance.³⁸

OTHER METHODS FOR REGULATING JUDICIAL DISCRETION

Guideline judgments

6.36 Guideline judgments are judgments of courts of criminal appeal which go beyond the point specifically raised for consideration in the appeal. Their purpose is to suggest the principles (and, possibly, a scale) which ought to apply to sentences for various categories of crime. Such judgments represent a consolidation of advice upon a particular sentencing point.

6.37 Several guideline judgments are issued each year by the English, New Zealand and Canadian courts of appeal. Section 143 of the Sentencing Act 1995 (WA) makes specific provision for guideline judgments to be issued by the Court of Criminal Appeal, which are “to be taken into account by courts sentencing offenders.” Guideline judgments may be given regardless of whether it is necessary for the current proceedings. No mention is made of the principles guiding the content of the judgment, although provision is made for their review, variation or revocation.³⁹

6.38 The Victorian Sentencing Committee made comprehensive recommendations for the adoption of a statutory guideline judgments procedure, accompanying the qualitative and quantitative assistance of a Judicial Studies Board.⁴⁰ The Sentencing Bill 1990 (Vic) contained a provision for guideline judgments, which would have permitted the Full Court to consider statistical material, evidence of relevant public attitudes, and the efficient use of correctional facilities.⁴¹ However, the provision was not enacted. A majority of Supreme Court judges, while not denying their “possible utility”, thought that guideline judgments were unnecessary in the closely knit Victorian legal community, and that the way in which judgments were delivered by the Court of Criminal Appeal should be studied before the fetters on the discretion of the sentencing judge implied in guideline judgments were entertained.⁴²

6.39 These criticisms point to the fact that the innumerable factors which may be taken into account in sentencing any particular offender will necessarily limit the efficacy of a guideline judgments procedure. Guideline sentencing ranges might become so broad so as to provide no useful guidance at all. Alternatively, the sentencing range might remain usefully narrow, though at the expense of excessively refined criteria for determination of the range.

6.40 Guideline judgments will be also be restricted to a fairly narrow cross-section of offences because they also rely upon the standard appeal process. They were criticised as an ineffective means of promoting consistency by the Canadian Sentencing Commission.⁴³ In this respect, they are subject to the same limitations as the current appellate structure, because the Court of Criminal Appeal tends to review sentences for serious crimes, or sentences which are argued to be extremely harsh or lenient. Guideline judgments do not permit a systematic appraisal of the sentencing system, and are unsuitable for debating the overall objectives of that system.⁴⁴ Nor do they permit critical evaluation of penalty severity for one offence relative to other offences.⁴⁵ On the whole, submissions to the Attorney General’s Sentencing Review tended not to favour the introduction of guideline judgments either for unstated reasons or for reasons which have been outlined above.⁴⁶

6.41 The advantage of guideline judgments is that they represent a reform consistent with the nature of the existing appellate process. Judicial officers and counsel could adapt to the changes relatively easily, and the process could be grafted onto existing sentencing appeals processes. The Attorney General’s Sentencing Review suggested that it might be possible for the DPP specifically to request a guideline judgment in a sentencing appeal, and the court should be free to accept or decline such a request.⁴⁷ The Review referred to the need to protect the accused from any additional financial burden that might result from his or her case being made a “test case”.⁴⁸

Sentencing councils

6.42 The prospect of introducing a sentencing council to improve sentencing decisions has been addressed in a variety of forums. Different inquiries and jurisdictions have had quite different understandings of the appropriate functions a sentencing council might exercise. The Victorian Sentencing Committee regarded sentencing councils as a model involving:

• More than one judge imposing a sentence
• A judge with a number of lay assessors determining the sentence for an offender.⁴⁹

This model involves formal consultation between the members of the council or panel, although the ultimate power to impose sentence may reside in one (judicial) member. The rationale for this approach is that a consultative process - even though it may not be binding - will produce consensus increasing the uniformity of sentences.⁵⁰ The major problem with this model is the substantial increase in resources which must be allocated to sentencing decisions made at first instance.⁵¹ The Victorian Sentencing Committee did not recommend the introduction of a sentencing council along these lines.

6.43 Other proposals envisage a less radical function for sentencing councils, combining several of the methods which have been dealt with above.⁵² For example, the Australian Law Reform Commission recommended the establishment of a council whose major function would be the provision of “detailed, comprehensive information to promote consistency in sentencing...”.⁵³ The Victorian Sentencing Committee recommended the establishment of a “judicial studies board” whose aim would be structured education and keeping judicial officers aware of changes in sentencing law.⁵⁴ This recommendation was acted upon by the government with the passing of the Judicial Studies Board Act in 1990.⁵⁵ However, the Board has never appointed any executive officers, and no progress has been made to achieving the functions prescribed by the enabling legislation.⁵⁶ The Judicial Commission of New South Wales largely fulfils the role envisaged for a sentencing council in this paragraph through the maintenance of the Sentencing Information System and the provision of judicial education.

6.44 The introduction of a “Sentencing Policy Advisory Council” whose purpose would be to develop sentencing policy was canvassed by the Attorney General’s Sentencing Review in 1994.⁵⁷ The possibility that the Judicial Commission could incorporate such a council for the purpose of developing sentencing policy was raised. However, it was argued that greater time should be allowed to evaluate the Judicial Commission’s Sentencing Information System before a sentencing council was contemplated.

6.45 Submissions to the Attorney General’s Sentencing Review were divided on the value of introducing a sentencing council in New South Wales. Several agreed with the suggestion made in the Review’s Issues Paper that more time was needed for a proper evaluation of the SIS.⁵⁸ Of those submissions in favour of a council, several were happy for the Judicial Commission to provide the infrastructure for the Council.⁵⁹ The Council of Social Service of New South Wales (NCOSS) submitted that any sentencing council should not be part of the Judicial Commission.⁶⁰ Others were simply opposed to the concept of a sentencing council.⁶¹

Statutory minimum penalties

6.46 Specifying in legislation the minimum penalties that may be imposed for particular offences is one method of providing limits upon judicial discretion in sentencing. As a general rule (as in other common law jurisdictions) the Parliament has not traditionally provided sentencing guidance to judges by imposing minimum penalties in legislation.

6.47 The guidance given by statutory minima is very blunt. No guidance is given for choosing any punishment which would fall within the range established by the minimum and maximum sentence. Sentence disparity might remain, although over a narrower range of sentence. Logically, the extent of this disparity would vary according to the extent of the range of sentence which was available between the two extremes.

6.48 A further criticism of statutory minimum penalties is that they provide little scope for addressing the subjective features of a particular case or offender. Some offenders may be deserving of lesser punishment than other offenders for a number of reasons.⁶² If the sentencing judge would have imposed a sentence of lesser severity than the minimum prescribed penalty, the resulting sentence is unjust.

6.49 A modification to this method of guiding discretion is to provide a “special circumstances” exception.⁶³ This allows the sentencer to impose a sentence below that prescribed by the legislation where special circumstances (which may or may not be specified in the legislation) exist. Whether greater sentencing consistency results from such a system depends largely on how the judiciary interprets the special circumstances requirement. If a broad interpretation evolves, the objective of sentence uniformity is hardly advanced.⁶⁴

6.50 The Australian Law Reform Commission recommended the elimination of all prescribed minimum penalties from Commonwealth law. It was argued that mandatory minimum penalties undermine consistent consideration of offences, since different circumstances are artificially considered in the same way, encourage technical defences and may result in perverse verdicts by juries.⁶⁵ The Commission agrees.

Grid sentencing

6.51 The introduction of statutory grids for determining sentence has been one of the most significant (and controversial) aspects of sentencing reform. They have been used most extensively in United States jurisdictions, where fifteen States have guidelines produced by a standing sentencing commission.⁶⁶ The Minnesota and Federal models are the most widely known, and the following discussion will be largely confined to these two examples of sentencing grid systems. However, it should be remembered that the grids contain differences for a number of important variables. Sentencing grids may be voluntary or mandatory, may or may not abolish discretionary release on parole, and use different formulae for determining criminal history scores. The type of offence to which the grid applies varies across the jurisdictions.⁶⁷ Finally, the grids produced in the various States enshrine different objectives and philosophies of punishment.⁶⁸

6.52 The possibility of introducing a sentencing grid was raised by the Attorney General’s Sentencing Review, but it was argued that serious consideration of such a system should be delayed until the effect of the SIS could be properly evaluated.⁶⁹ There is a vast literature on the American sentencing grids, and much controversy surrounding their value as a sentencing tool. What follows is merely a brief introduction to the concept of sentencing grids. An account of their basic features, and a summary of the main arguments for and against them, is provided.

The Minnesota system

6.53 The basic features of the grid are as follows. The vertical axis of the grid displays the severity levels of the various offences in descending order. Along the horizontal axis, the possible “criminal history scores” are displayed, which refers to the number of the offender’s previous convictions. These two features were selected after preliminary research indicated that the two most important influences upon sentencing were the seriousness of the instant offence and the extent of the offender’s criminal record.⁷⁰ The presumptive sentence lies at the intersection of the points selected on each axis. Where the cell selected is above the “dispositional line” (a heavy black line), the presumptive sentence is a non-custodial order for the number of months indicated by the cell. For all cases appearing below the dispositional line, a sentence of imprisonment is presumed.

6.54 A small range of sentence also appears in each cell falling below the dispositional line (that is, for sentences of imprisonment), which allows for aggravation or mitigation of the guideline sentence. A judge may impose a sentence either above or below this range only if “substantial and compelling” circumstances exist, and reasons are given for the departure from the range.

6.55 Thus, to take an example, an offender convicted of a residential robbery with three previous convictions for a similar offence will presumptively be sentenced to thirty months imprisonment. If the sentencing judge is of the view that mitigating circumstances apply, the offender may be sentenced to twenty-nine months. If the offence is aggravated for some reason, a sentence of thirty-one months may be imposed. A sentence less than twenty-nine months, or longer than thirty-one months, can be imposed only in substantial and compelling circumstances.

6.56 Both the Australian Law Reform Commission and the Victorian Sentencing Committee rejected the introduction of sentencing grids along the lines of the Minnesota model.⁷¹ Both reports indicated considerable concern about the rigidity that would be introduced into the sentencing system if judicial discretion was removed in such a manner.

6.57 The main advantages of a grid system are consistency and certainty of outcome. Apparently, the Minnesota guidelines have resulted in less disparity between offenders on the basis of race and social class.⁷² In 1991, the specified sentencing ranges were departed from in 16% of all cases, three-quarters mitigated and one-quarter aggravated.⁷³ Desirable secondary criteria, such as efficient utilisation of prison space, can be factored into the sentencing formula. Minnesota has been largely successful (especially in relation to other US jurisdictions) in retarding the growth of the prison population.⁷⁴ Authoritative alterations can be made relatively easily where the meeting of policy goals requires the adjustment of sentencing regime.

6.58 However, it has been argued that the improved consistency brought by the grid system is a somewhat artificial advantage. Two elements of sentence information- offence seriousness and criminal record - become the privileged features of the sentencing regime. It has been noted that the offender’s criminal record has played a role of greater importance than originally intended for the system.⁷⁵ Consistency may be increased at the expense of dispassionate consideration of a range of other factors which might be relevant when deciding upon the appropriate punishment.

6.59 Attempts to reduce prison population have had mixed success. While the availability of prison space was factored into the sentencing guidelines, alterations in exercise of prosecutorial discretion (for example, multiple charging) have mitigated the attempts to reduce the prison population. Prosecutors apparently “run-up” criminal history scores by this practice, thereby achieving higher sentences. The grid system has also caused an increase in charge and fact bargaining, as the decision to prosecute (and the choice of charges) becomes more important to sentence outcome.⁷⁶ The most important point to make about this is that the exercise of discretion in the sentencing process is retained, at least to some extent. While the discretion of judges is restricted, discretion now resides with prosecutors, whose role is, of course, less visible and less subject to control.

United States federal guidelines

6.60 The Federal guidelines, introduced by the Sentencing Reform Act 1984, include the same basic concepts as those which apply in Minnesota. Like the Minnesota Sentencing Commission, the federal Sentencing Commission produces guidelines which rely on a sentencing grid. The grid operates by measuring the offence levels and the criminal history of the offender, and a sentencing range is provided at the intersection of each row and column.⁷⁷ Apart from differences in the severity of punishments between the two sets of guidelines, the greater complexity of the federal guidelines distinguishes them from those which operate in Minnesota and other American States. The federal sentencing grid has forty three offence levels, and six categories for the offenders criminal history. To the “base level” of the particular offence, is added additional levels according to the existence of any “special offence characteristics”. To take an example, if an offender is convicted of robbery, the offence level is enhanced by three points if a gun was brandished, and further enhanced on a points scale depending upon the value of money or goods taken in the robbery.⁷⁸ It should be noted that relevant conduct for this purpose extends to activity which is not part of the offence charged, or which constitutes multiple offences charged only as a single offence.⁷⁹ The need for “adjustment” of the offence level is then determined. The fact that the victim is to be regarded as vulnerable would increase the offence level, while remorse (for example) would decrease the level.

6.61 The offender’s criminal history score (which is sensitive to the seriousness, as well as the number of, convictions) is calculated, and the guideline sentence then calculated. In the absence of unusual factors, the guideline sentence must be imposed.⁸⁰ Guidelines and additional “policy statements” are provided as to what may or may not justify departures from the guideline sentence.⁸¹

6.62 Much of the complexity of the federal guidelines is apparently due to efforts to curtail tightly the exercise of judicial discretion, which was previously much less guided by developed case law than other common law jurisdictions.⁸² However, it appears that this complexity has largely nullified the qualified advantages that the Minnesota grid has brought to sentencing in that State. It also appears that the disadvantages brought by the sentencing grid in Minnesota also apply in the federal jurisdiction, and are more severe.

6.63 One ramification of the guidelines’ rigidity is that disparity becomes hidden by the less visible practices which occur at other points of the criminal justice system. For example, prosecutorial discretion in the selection of charges, and negotiation about particular facts to be brought to judicial notice, are practices which appear to be widespread. Such conduct leads to disparities between cases which should (according to the guidelines) be treated in the same way. “Informal non-compliance” is particularly important because the resulting disparities tend to be hidden, and disparity was the particular problem which the guidelines were supposed to address.⁸³

6.64 Not surprisingly, the response to the federal sentencing guidelines has been overwhelmingly negative. Michael Tonry, a leading American sentencing scholar, has written:

The Guidelines ... are the most controversial and reviled sentencing reform initiative in United States history. They are commonly criticised on policy grounds (that they unduly narrowly limit judicial discretion and unduly shift discretion to prosecutors), on process grounds (that they forseeably cause judges and prosecutors to circumvent them), on technocratic grounds (that they are too complex and hard to apply accurately), on fairness grounds (that by taking only offence elements and prior convictions into account, they require that very different defendants receive the same sentence), and on normative grounds (that they greatly increased the proportion of offenders receiving prison sentences and are generally too harsh).⁸⁴

Canadian Sentencing Commission guidelines

6.65 The Canadian Sentencing Commission rejected the type of detailed numerical guidance available in the sentencing grids detailed above. The Commission was in favour of increasing guidance in sentence, and recommended presumptive sentences, of four sentence types only:

1. in custody;

2. out of custody (community sanction);

3. qualified custody (custody presumed unless a minor offence and offender has no record); and

4. qualified out of custody (community sanction unless it is a serious instance of the offence and the offender has a relevant criminal record).⁸⁵

Further, a presumptive range of sentence for each offence normally requiring incarceration would also be set. In all cases, the presumption could be departed from where the judge gives reasons for doing so.

6.66 Such a proposal retains a very wide discretion as compared to other grid approaches to sentencing guidance. The Victorian Sentencing Committee regarded the Canadian model as providing only “limited and imprecise guidance” for the tasks of classifying and combining the variety of information relevant to the task of sentencing.⁸⁶ The Commission’s tentative view is that such limited guidance does not justify departure from the present sentencing model.

QUESTIONS ARISING IN CHAPTER 6

1. Should any changes be made to the Sentencing Information System in New South Wales? If so, what will the benefits of those changes be either in terms of sentence consistency or more generally?

2. Sentencers in Victoria are expressly required to have regard to “current sentencing practices” when sentencing an offender (see Sentencing Act 1991 (Vic) s 5(2)(b)). Should sentencers in New South Wales be required to take current sentencing practice into account by use of the SIS?

3. Is judicial education in sentencing law likely to promote greater consistency in sentencing? If so, how ought that training to be structured to promote this goal?

4. Ought guideline judgments to be introduced in New South Wales?

5. Is there any role for a Sentencing Council in New South Wales?

6. Ought the Legislature ever to prescribe mandatory minimum penalties?

7. Ought consideration to be given to the development of a sentencing grid in New South Wales? If so, how far should the grid be allowed to confine judicial discretion?

1. See paras 3.38-3.40.

2. See paras 2.13-2.21.

3. See Chapter 5.

4. Lowe v The Queen (1984) 154 CLR 606 at 608-09.

5. Crump v The Queen (1995) 69 ALJR 570 at 570 per Brennan CJ, Dawson and Gummow JJ.

6. DPP v El Karhani (1990) 21 NSWLR 370 at 380.

7. A Ashworth, “Criminal Justice, Rights and Sentencing: A Review of Sentencing Policy and Problems” in I Potas (ed), Sentencing in Australia (Australian Institute of Criminology, Seminar Proceedings No 13, 1987) at 36.

8. R v Allpass (1993) 72 A Crim R 561 at 562.

9. R Fox, “Controlling Sentencers” (1987) 20 Australia and New Zealand Journal of Criminology 218 at 226.

10. A Ashworth, “Three Techniques for Reducing Sentence Disparity” in A von Hirsch and A Ashworth (eds), Principled Sentencing (Edinburgh University Press, Edinburgh, 1992) at 283.

11. Ashworth (1987) at 36.

12. Preamble to the Judicial Officers Act 1986 (NSW).

13. Figures used in this paragraph provided by the Judicial Commission of NSW, 20 December 1995.

14. Offender characteristics have deliberately been limited to more “objective” features: see J Chan, “A Computerised Sentencing Information System for New South Wales Courts” (1991) 7(6) Computer Law and Practice 137 at 139-140. Thus no allowance is made for the offender’s sex because the Judicial Commission has been advised that many judicial officers take the view that this is not a factor relevant to the exercise of sentencing discretion: see paras 5.46-5.49.

15. Sentencing Act 1989 (NSW) s 8.

16. Sentencing Act 1989 (NSW) s 5(2). See paras 4.24-4.33.

17. I Potas and S Cumines, “The Sentencing Information System of New South Wales: Using Computers to Inform the Courts” paper presented at the Eighth Conference for Librarians in the Criminal Justice System (Sydney, 2 April 1992) at 1.

18. See Chapter 5.

19. See para 6.3.

20. Information provided by the Judicial Commission of New South Wales, 20 December 1995.

21. See A Lovegrove, “Structuring the Judicial Sentencing Discretion: Some Empirical Considerations on Reforms” in A Kapardis (ed), Sentencing: Some Key Issues (Special issue, Volume 13 No 2 of Law in Context) (La Trobe University Press, 1995) at 150.

22. See R v Baker (NSW CCA, No 60242/92, 26 November 1992, unreported) at 10; R v Mackenzie (NSW CCA, No 60448/93, 15 December 1993, unreported) at 5-6 per Carruthers J; R v Thomson (NSW CCA, No 60307/93, 21 June 1994, unreported) at 5 per Finlay J; R v Jones (NSW CCA, No 60555/93, 30 June 1994, unreported) at 4; R v H (NSW CCA, No 60329/93, 25 July 1994, unreported) at 3; R v Garlick (NSW CCA, No 60038/94, 29 July 1994, unreported) at 11; R v McDonald (NSW CCA, No 60179/93, 18 August 1994, unreported) at 10; R v Alexander (1994) 78 A Crim R 141 at 146; R v Allen (NSW CCA, No 60458/94, 17 February 1995, unreported) at 7-8; R v Gudge (NSW CCA, 60489/94, 3 April 1995, unreported) at 13 per Finlay J; R v K (NSW CCA, No 60401/94, 5 April 1995, unreported) at 7 per Finlay J; R v Yavuz (NSW CCA, No 60102/94, 20 April 1995, unreported) at 9 per James J; R v Lego (NSW CCA, No 60511/94, 24 April 1995, unreported) at 4 per Newman J; R v Griffiths (NSW CCA, No 60723/94, 26 June 1995, unreported) at 14; R v Maguire (NSW CCA, No 60609/94, 30 August 1994, unreported) at 9-10 per Grove J.

23. Information provided by the Judicial Commission of New South Wales, 20 December 1995.

24. NSW CCA, No 60609/94, 30 August 1994, unreported, at 9-10.

25. R v Mackenzie (NSW CCA, No 60448/93, 15 December 1993, unreported) at 5-6.

26. New South Wales, Attorney General’s Sentencing Review 1994 (June 1994) at 59; Weatherburn (1994) at 17.

27. P Sallmann, “In Search of the Holy Grail of Sentencing: An Overview of Some Recent Trends and Developments” (1991) 1 Journal of Judicial Administration 125 at 129.

28. Judicial Officers Act 1989 (NSW) s 9. The Australian Law Reform Commission recommended that the provision of sentencing knowledge and skills programs for judicial officers be developed as one of the functions of the proposed sentencing council: Sentencing (ALRC 44, 1988) at paras 278-282.

29. Judicial Officers Act 1989 (NSW) s 9(2).

30. Judicial Commission of New South Wales, Annual Report 1994-1995 (September 1995) at 62.

31. Sallmann at 129.

32. Judicial Commission of New South Wales, Annual Report 1994-1995 (September 1995) at 9-11.

33. See paras 6.10-6.30.

34. Judicial Commission of New South Wales, Annual Report 1994-1995 (September 1995) at 12.

35. New South Wales, Attorney General’s Sentencing Review 1994 at 59.

36. A Freiberg, “Sentencing and Judicial Administration” (1993) 2 Journal of Judicial Administration 171 at 177.

37. R Fox, “Controlling Sentencers” (1987) 20 Australia and New Zealand Journal of Criminology 218 at 229.

38. Fox (1987) at 230.

39. Sentencing Act 1995 (WA) s 143(3).

40. Victoria, Attorney-General’s Department, Sentencing: Report of the Victorian Sentencing Committee (Melbourne, 1988) Volume 1 at paras 4.20.8-4.20.14.

41. A Freiberg, “Sentencing and Judicial Administration” (1993) 2 Journal of Judicial Administration 171 at 175.

42. Sentencing: Report of the Victorian Sentencing Committee Volume 3 at A9-A10.

43. Canadian Sentencing Commission, Sentencing Reform: A Canadian Approach (1987) at 294-96.

44. Fox at 227; P Sallmann, “In Search of the Holy Grail of Sentencing: An Overview of Some Recent Developments” (1991) 1 Journal of Judicial Administration 125 at 128-29.

45. A Ashworth, “Three Techniques for Reducing Sentence Disparity” in A von Hirsch and A Ashworth (eds), Principled Sentencing (Edinburgh University Press, Edinburgh, 1992) at 284.

46. See Mr Barry Newport QC, Crown Prosecutor, Submission to the Attorney General’s Sentencing Review (22 July 1994); Legal Aid Commission of New South Wales, Submission to the Attorney General’s Sentencing Review (25 July 1994); Law Society of New South Wales, Submission to the Attorney General’s Sentencing Review (25 July 1994); Office of the Director of Public Prosecutions, Submission to the Attorney General’s Sentencing Review (27 July 1994); Bar Association of New South Wales, Submission to the Attorney General’s Sentencing Review (2 August 1994); Public Defenders Chambers, Submission to the Attorney General’s Sentencing Review (12 August 1994).

47. Attorney General’s Sentencing Review at 59.

48. Attorney General’s Sentencing Review at 59.

49. Sentencing: Report of the Victorian Sentencing Committee Volume 1 at 14.15.1.

50. Fox (1987) at 228.

51. Sentencing: Report of the Victorian Sentencing Committee Volume 1 at 14.15.5.; Fox (1987) at 228.

52. Sentencing councils should be distinguished from the North American sentencing commissions which develop sentencing policy and guidelines to be used in conjunction with statutory sentencing grids or presumptive dispositions.

53. Sentencing (ALRC 44, 1988) at para 275.

54. Sentencing: Report of the Victorian Sentencing Committee Volume 1 at para 4.21.5

55. Section 5 of the Act gives the Board fairly detailed functions (exclusively related to sentencing) to: (a) conduct seminars for judges and magistrates on sentencing matters; (b) conduct research into sentencing matters; (c) prepare sentencing guidelines and circulate them among judges and others; (d) develop and maintain a computerised statistical sentencing database for use by the courts; (e) provide sentencing statistics to judges, magistrates and lawyers; (f) monitor present trends, and initiate future developments, in sentencing; (g) assist the courts to give effect to the principles contained in the legislation; (h) consult with the public, government departments and other interested people, bodies or associations on sentencing matters; (i) advise the Attorney-General on sentencing matters. The Board is to consists of seven members, comprising: the Chief Justice of the Supreme Court (or nominee); a Supreme Court judge nominated by the Chief Justice; the Chief Judge of the County Court (or nominee); a County Court judge; the Chief Magistrate (or nominee); two appointees of the Attorney-General, at least one of whom must be an academic: Judicial Studies Board Act 1990 (Vic) s 7.

56. A Freiberg, “Sentencing and Judicial Administration” (1993) 2 Journal of Judicial Administration 171 at 179.

57. Attorney General’s Sentencing Review at 58.

58. See for example Public Defenders Chambers, Submission to the Attorney General’s Sentencing Review (12 August 1994) at 6; Law Society of New South Wales, Submission to the Attorney General’s Sentencing Review 1994 (25 July 1994).

59. See for example Ivan Potas, Submission to the Attorney General’s Sentencing Review (11 August 1994) at 5; Probation and Parole Officers Association of NSW, Submission to the Attorney General’s Sentencing Review (20 June 1994) at 11-12. See also Ministry for the Status and Advancement of Women NSW, Submission to the Attorney General’s Sentencing Review (26 July 1994).

60. Council of Social Service of New South Wales, Submission to the Attorney General’s Sentencing Review (9 August 1994) at 8.

61. See for example Legal Aid Commission of New South Wales, Submission to the Attorney General’s Sentencing Review (25 July 1994) at 4; Office of the Director of Public Prosecutions, Submission to the Attorney General’s Sentencing Review (29 July 1994) at 5.

62. See Chapter 5.

63. See A Ashworth, Sentencing and Penal Policy (Weidenfeld and Nicolson, London, 1983) at 74.

64. Consider, by analogy, the use of “special circumstances” in the Sentencing Act 1989 (NSW) s 5(2): see paras 4.24-4.33.

65. Sentencing (ALRC 44, 1988) at para 58.

66. R Frase, “Sentencing Guidelines in Minnesota and Other American States: A Progress Report” in C Clarkson and R Morgan (eds), The Politics of Sentencing Reform (Clarendon Press, Oxford, 1995) at 171.

67. Frase at 173-74.

68. Frase at 175-76.

69. Attorney General’s Sentencing Review at 46-47.

70. R Fox, “Controlling Sentencers” (1987) 20 Australian and New Zealand Journal of Criminology 218 at 235.

71. Sentencing (ALRC 44, 1988) at para 183; Sentencing: Report of the Victorian Sentencing Committee Volume 2 at 573.

72. Frase at 185.

73. Frase at 186.

74. Frase at 195.

75. A von Hirsch, “Proportionality and Parsimony in American Sentencing Guidelines: The Minnesota and Oregon Standards” in C Clarkson and R Morgan (eds), The Politics of Sentencing Reform (Clarendon Press, Oxford, 1995) at 157.

76. Frase at 188. See also M Tonry and J Coffee, “Enforcing Sentencing Guidelines: Plea Bargaining and Review Mechanisms” in A von Hirsch, K Knapp and M Tonry (eds), The Sentencing Commission and Its Guidelines (Northeastern University Press, Boston, 1987).

77. P Sallmann, “In Search of the Holy Grail of Sentencing: An Overview of Some Recent Trends and Developments” (1991) 1 Journal of Judicial Administration 125 at 133.

78. A Ashworth, “Three Techniques for Reducing Sentence Disparity” in A von Hirsch and A Ashworth (eds), Principled Sentencing (Edinburgh University Press, Edinburgh, 1992) at 288.

79. A Doob, “The United States Sentencing Commission Guidelines: If you Don’t Know Where you are Going, you Might not get There” in C Clarkson and R Morgan (eds), The Politics of Sentencing Reform (Clarendon Press, Oxford, 1995) at 219.

80. S Breyer, “The Key Compromises of the Federal Sentencing Guidelines” in C Munro and M Wasik (eds), Sentencing, Judicial Discretion and Training (Sweet & Maxwell, London, 1992) at 108.

81. Ashworth (1992) at 289.

82. Doob at 203.

83. D Freed, “Federal Sentencing in the Wake of Guidelines: Unacceptable Limits on the Discretion of Sentencers” (1992) 101 Yale Law Journal 1681 at 1754.

84. M Tonry, “Judges and Sentencing Policy - The American Experience” in C Munro and M Wasik (eds), Sentencing, Judicial Discretion and Training (Sweet & Maxwell, London, 1992) at 139-40. For a sustained criticism of sentencing girds by an American judge who felt compelled to resign rather than apply them, see L G Forer, A Rage to Punish (Norton & Co, New York, 1994).

85. Canadian Sentencing Commission, Sentencing Reform: A Canadian Approach (1987) at 296.

86. Sentencing: Report of the Victorian Sentencing Committee Volume 2 at para 14.2.58.