Sentencing Conference 2010 Canberra 6 & 7 February 2010 VARIATIONS IN FEDERAL SENTENCING by DAVID ADSETT1 and MARK PEDLEY2 SUMMARY3 This paper looks at the relationship between the non‐parole periods and head sentences imposed on federal offenders around Australia in the 3 years between 2006/7 and 2008/9 across a total of 439 sentences. In each case the head sentence was more than 3 years as under federal law for a non‐parole period to be imposed the head sentence must be more than 3 years. The sentences are analysed by type of offending and on a State/Territory basis. The survey of sentences reflects a significant degree of consistency in that relationship for drug sentences across Australia. The survey also reveals a very similar pattern in non‐parole periods relative to the head sentence as between drug, money laundering offences and people smuggling offences. In other offence types there is some variation of approach across the various jurisdictions. Nevertheless there is a reasonable level of consistency bearing in mind the approach reflected in the Judiciary Act 1903 (C’th) is that federal sentencing is carried out by State/Territory courts and so State/Territory practice has a bearing on the sentences imposed. 1
Deputy Director, Commonwealth DPP, Brisbane Office Deputy Director, Commonwealth DPP, Melbourne Office. 3
The authors acknowledge the contribution and assistance of Ms Averil Templar LLB, B.Com (Hons), Commonwealth DPP Canberra, in extracting the statistics presented in this paper. 2
PART A. CONTEXT OF FEDERAL SENTENCING When considering federal sentencing it is important to bear in mind the context in which federal sentencing occurs. The following features of federal criminal jurisdiction are all important: •
Federal criminal jurisdiction is exercised by State courts Sections 71 and 77 of the Constitution provide that the Commonwealth Parliament may invest State courts with Federal jurisdiction. The Commonwealth Parliament has invested State courts with federal jurisdiction through the Judiciary Act 1903 (C’th). Section 68(2) of the Judiciary Act 1903 (C’th) provides that the State courts exercising criminal jurisdiction have “like jurisdiction with respect to persons who are charged with offences against the laws of the Commonwealth”. • Federal prisoners account for just on 3% of the total prison population in Australia.4 Australian Bureau of Statistics data shows that 3.4% of prosecutions in the higher courts in Australia include one or more federal offence.5 • There are variations in relevant laws and sentencing options applicable to Commonwealth offenders The Commonwealth’s principal sentencing provision, Part 1B of the Crimes Act 1914 (C’th), largely provides a uniform sentencing law for Commonwealth offences, however, it is not an exclusive code on sentencing. It has been noted that the objectives of Part 1B “varied between making exhaustive provision on some subjects and supplementary provision on others”.6 Some State and Territory based variations are adopted for the purposes of Commonwealth sentencing. The following are examples: ‐
‐
‐
the commencement date for sentences;7 some additional alternative sentencing options available in a particular “participating State or a participating Territory” are available in respect of Commonwealth offenders. These sentencing options are State/Territory based and will vary depending on where the sentencing takes place. These may include an order known as a community service order, a work order, a sentence of periodic detention, an attendance centre order, a sentence of weekend detention or an attendance order... or a similar sentence or order or a sentence or order that is prescribed.8 additional forms of sentences not specifically available under Commonwealth law and not inconsistent with specific Commonwealth 4
6
7
8
5
Australian Law Reform Commission Report, “Same Crime, Same Time”, Report 103, April 2006, para 1.62. Australian Bureau of Statistics, 2008‐9 Criminal Courts Australia, Report 4513.0, p 21, table 2.3. Putland v The Queen (2004) 218 CLR 174 at 181, 193 and 215. See s 16E Crimes Act 1914 specifically adopts State law in relation to this aspect of sentencing. s 20AB Crimes Act 1914 (C’th). 2 Variations in Federal Sentencing ‐
sentencing provisions, e.g. a State or Territory law providing for a single global sentence for multiple counts. remissions or reductions of a head sentence. The laws of the State of Territory providing for the remission or reduction of a head sentence apply to a Commonwealth offender.9 Currently there are no State or Territory laws that provide for the remission of head sentences apart from those relating to the impact of prison strikes. •
There are reduced opportunities for comparison between jurisdictions Until January 2003, courts sentencing a defendant for a Commonwealth offence were required to take into account whether the State law provided for remissions to be deducted from a head sentence. If the State law in the State in which the federal offender came to be sentenced did not provide for remissions from the head sentence, the federal sentence had to be reduced to take account of the fact that the federal offender serving his sentence in that State would not be subject to remissions. This had a downward impact on head sentences and a consequential effect on the minimum term.10 This provision, s16G of the Crimes Act 1914 (C’th), was repealed with effect from 17 January 2003. Given the different laws applicable in Australian states during the period 16G was in operation, there was minimal opportunity to compare sentencing data easily.11 •
Applicability of variations in sentencing practice Because those sentencing federal offenders are officers of State courts, it is inevitable that State sentencing practice will affect the approach to Federal sentencing. For example, as is shown below, in those States where there is a fixed statutory formula governing the relationship between the head sentence and the non parole period, there will be a tendency to adopt a similar approach in sentencing a federal offender. In some States courts require detailed submissions from prosecutors, not only on the type of sentence submitted to be appropriate, but also, if a custodial sentence is submitted as appropriate, on the level of sentence. In other jurisdictions, general submissions of this nature are made. In other jurisdictions it was routine until only fairly recently for prosecutors not to make submissions on sentence at all.12 Commonwealth criminal law is enforced in eight independent and separate judicial systems. Although there is limited interchange of judicial personnel, the judicial systems of each State or Territory which are invested with federal jurisdiction, operate independently. Notwithstanding the High Court’s formulation of an appropriate approach to the sentencing for multiple offences in Pearce v The Queen13, there remain variations 9
s 19AA(1) Crimes Act 1914 (C’th). DPP v El Kaharni (1990) 21 NSWLR 370. 11
R v Tran (2007) 171 A Crim R 436 and R v Phillips (2008) 188 A Crim R 133. 12
R v MacNeil‐Brown (2008) 188 A Crim R 403 and DPP (Vic) v Terrick [2009] VSCA 220. 13
(1998) 194 CLR 610. 10
3 Variations in Federal Sentencing within Australian jurisdictions as to practice in relation to the detailed formulation of sentencing outcomes for such offences. For example in Queensland, when sentences are to be imposed for multiple offences, whether committed as part of an episode of criminality or comprising distinct, unrelated offences, the practice is to “fix a sentence for the most serious offence which might be higher than the sentence for that offence had it stood alone, the higher sentence taking into account the overall criminality”.14 This practice has been sanctioned by the Court of Appeal.15 It is the approach Queensland courts traditionally take to sentencing accused charged with numerous offences. One sentence is elevated to reflect the “totality of the criminality”16 and other sentences are made concurrent. This practice appears not to exist in other jurisdictions where there is less reliance on concurrent sentences.17 •
Absence of single appeal court. There is no single appeal court to which all federal sentences are appealed. Rather the scheme of the Judiciary Act 1903 (C’th) is that each State and Territory appeal court determines appeals in respect to federal sentences as it does in relation to State/Territory sentence appeals. So one of the major mechanisms which operates in respect of State and Territory sentences to achieve consistency, namely overview by a single appellate court, does not apply in respect of federal sentences. It is a matter of long standing principle that special leave to appeal to the High Court will not be granted “simply because the sentence appears... to be excessive”.18 Neither will the High Court grant special leave where it is contended a sentence is inadequate.19 It has been noted that review of sentences by the High Court is “not a practical proposition”.20 In Russell v The Queen21 Brennan CJ noted that the Courts best placed to “discern community standards” are those of the States and Territories, his Honour continued: “The level of sentencing is predominantly a matter for the courts of criminal appeal of the States and Territories. It is not this Court’s function to review the standards which are deemed appropriate to local conditions unless they are under the influence of an error in sentencing”. 14
R v Nagy [2004] 1 Qd R 63 at para [39] per Williams JA and at para [66] per Jerrard JA, see also R v Gilles; ex parte Attorney General [2002] 1 Qd R 404 paras [14] to [17] per Pincus JA and para [19] per McPherson JA. 15
See for example R v Svensson; ex parte A‐G (Qld) [2002] QCA 472 and R v Moore [2003] 1 Qd R 205 at para [12] per Williams JA. 16
R v Moore at para [12]. 17
Shen v The Queen [2009] NSWCCA 251 at para [27]. 18
Lowe v The Queen (1984) 154 CLR 606 and 607 per Gibbs CJ, see also White v The Queen (1962) 107 CLR 174 and Radenkovic v The Queen (1990) 170 CLR 623 at 640 per Dawson J. 19
R v Skaf [2006] HCA Trans 10. 20
Hon JJ Spigelman AC, “Consistency and Sentencing”, (2008) 82 ALJ 450 at 456. 21
(1998) 5 Leg Rep C1. 4 Variations in Federal Sentencing • There is no national mechanism for guideline judgments in respect of federal sentences.22 • Variations in appeal practice and law There are variations in practice and law applicable to appeals. For example, in Queensland it is not required that the prosecution show error before the Court of Appeal may intervene in a prosecution appeal against sentence.23 In Victoria24, Tasmania25, Western Australia26, South Australia27 and New South Wales28 the element of double jeopardy on a Crown appeal is now no longer relevant, though there remains an issue as to whether the State legislation, designed to preclude an appeal court from having regard to that element, will apply in each State to federal matters.29 There has been acceptance in both commentary and by the judiciary that there will be some variation in federal sentencing given the policy choice that has been made to invest State courts with federal jurisdiction. In the ALRC report “Same Crime, Same Time”30 it was noted that “Part 1B of the Crimes Act 1914 (C’th) operated as a complete code for sentencing Federal offences while on other topics state and territory laws are picked up and applied”. The Commission noted: Given existing state and territory infrastructure (including courts and corrective services agencies and facilities) the relatively small number of Federal offenders, and the geographic dispersal of offenders across Australia, it is not viable to establish a completely separate federal criminal justice system. This means that the overwhelming majority of Federal offenders will continue to be sentenced in state and territory courts and that the sentences imposed will continue to be administered by state and territory corrective services agencies for the foreseeable future. On this basis, it is the ALRC’s view that it is not possible to achieve complete inter‐jurisdictional equality for federal offenders. ... Broad equality can be achieved while accepting certain differences that arise from Australia’s federal system of government”.31 It has been accepted by the High Court that the reliance on State courts and their procedures will sometimes produce different results but this is a concomitant of the policy choice made under Australia’s constitutional arrangements to invest State courts with federal jurisdiction. 22
Wong v The Queen (2001) 207 CLR 584 R v Lacey; ex parte Attorney‐General (Qld) [2009] QCA 274. 24
s 289(2) Criminal Procedure Act 2009 (Vic). 25
s 402(4A) Criminal Code (Tas). 26
s 41(4)(b) Criminal Appeals Act 2004 (WA). 27
Criminal Law Consolidation (Double Jeopardy) Amendment Act 2008 (SA). 28
s68A Crimes (Appeal and Review) Act 2000 (NSW). 29
In a Tasmanian case the rule prohibiting consideration of double jeopardy was held not to apply to a Commonwealth appeal – R v Talbot [2009] TASSC 107. 30
Australian Law Reform Commission, “Same Crime, Same Time”, Report 103, April 2006. 31
Ibid, paras 3.25 and 3.26. 23
5 Variations in Federal Sentencing In Leeth v The Commonwealth32 the Court considered the federal sentencing law before Part 1B of the Crimes Act 1914 (C’th) came into operation. The sentencing regime for federal offenders then in existence relied to a significantly greater extent on State parole systems than the current arrangements under Part 1B. There were some differences with the way the State parole and remission systems operated in relation to federal offenders sentenced in different States for the same type of offending. One of the offenders appealed to the High Court. The Court noted: “It is obviously desirable that, in the sentencing of offenders, like offenders, should be treated in a like manner. But such a principle cannot be expressed in absolute terms. .... it has long been recognised that sentencing practices may not be uniform from State to State but may be affected by local circumstances. Of course, with many offences, particularly federal offences, local circumstances may, under State sentencing practices, have no bearing upon the appropriate sentence and it may be proper to have regard to sentences imposed elsewhere in Australia. Even if it is accepted for the purpose of argument that any fundamental departure by the legislature from the principle that like offenders should be treated in a like manner may involve the imposition upon a court of a non‐judicial function, it is in our view apparent that to require a court, in the case of a federal offender, to have regard to the sentencing practices of the State in which he is convicted involves no such departure. To require a court to do so does not convert the sentencing process into some process of a non judicial kind but merely reflects the manner in which the Commonwealth, within the means made available to it by the Constitution, has chosen in the administration of its criminal law to operate through the existing State systems.”33 Since the decision in R v Leeth in 1992, the Commonwealth enacted Part 1B of the Crimes Act 1914 (C’th) which includes a separate regime for fixing federal non‐parole periods rather than relying on applied State or Territory legislation. In Putland v The Queen it was noted that one of the objects of Part 1B was to provide a separate regime for fixing federal non parole periods, rather than relying on applied State or Territory legislation. However, despite the attempt at uniformity in some aspects of sentencing, in that case, the High Court recognised that the Commonwealth sentencing regime inevitably allowed for variation. It has been noted that uniformity of treatment of federal offenders was not explicitly stated to be a goal of the legislation embodying the current federal sentencing regime. Gleeson CJ noted: “Section 68 of the Judiciary Act 1903 (C’th) reflects a permissible legislative choice, and one which, for a century, has resulted in some differences in the sentencing of federal offenders according to where they are sentenced.”34 32
(1992) 174 CLR 455. (1992) 174 CLR 455 per Mason CJ, Dawson and McHugh J at 470‐471. 34
Putland v The Queen (2004) 218 CLR 174 at 185. 33
6 Variations in Federal Sentencing That having been said it would be wrong to suggest that the High Court has endorsed inconsistency. There is strong judicial support for a broad consistency in Commonwealth sentencing. As noted above in Leeth the High Court said: “It is obviously desirable that, in the sentencing of offences, like offenders should be treated in a like manner.”35 More recently, in Wong v The Queen, Gleeson CJ noted: “Like cases should be treated in like manner. The administration of criminal justice works as a system; not merely as a multiplicity of unconnected single instances. It should be systematically fair, and that involves, amongst other things, reasonable consistency.”36 However, given the factors influencing the federal sentencing environment, particularly the reliance in important respects on the State sentencing law and the acceptance of State sentencing procedure, some variation will be inevitable. PART B. MAKING RECOGNISANCE RELEASE ORDERS AND FIXING NON‐PAROLE PERIODS – LEGAL PROVISIONS In general the Commonwealth sentencing regime, set out in Part 1B of the Crimes Act 1914 (C’th), does not provide any detailed guidance as to the length of the non‐parole period and there is no general Commonwealth statute that requires the non‐parole period to bear any proportion in relation to the head sentence. The common law provides that guidance.37 Exceptions exist in relation to federal offences of terrorism, treachery, treason or espionage. Section 19AG the Crimes Act 1914 (C’th) provides that an offence of this type is a minimum non parole offence” and the sentence is to be served as the non‐parole period”.38 For people smuggling offences section 233C of the Migration Act 1958 (C’th) specifies the relationship between the head sentence and the non‐parole period depending on the circumstances. For an offender without a prior conviction for the offence, the head sentence must be at least 5 years with a non‐parole period of at least 3 years.39 The approach to determining the relationship of the non‐parole period to the head sentence imposed for a federal offence is not wholly consistent throughout the States and Territories and varies to some degree. Given the factors referred to above which operate in relation to federal sentencing, this is not surprising. 35
Leeth v The Commonwealth (1992) 174 CLR 455 at 470 and see also Lowe v The Queen (1984) 154 CLR 606 at 610‐611. 36
(2001) 207 CLR 584 at 591, see also Hon JJ Spigelman AC, “Consistency and Sentencing”, (2008) 82 ALJ 450. 37
R v Selim [1998] NSWSC 165 at [8] ‐ “more often than not the non‐parole period is more than 50% of the head sentence and is ordinarily in the order of 60% ‐ 66⅔% of it”. 38
See Lodhi (2007) 179 A Crim R 470. 39
Migration Act 1958 s 233C. Note that although the Migration Act contains provisions setting a mandatory relationship between the non‐parole period and the head sentence for people smuggling offences, there is no similar provision in the Criminal Code people smuggling provisions which applies to Australian citizens and residents committing people smuggling offences outside Australia. 7 Variations in Federal Sentencing Statutory provisions in relation to fixing non‐parole periods and making recognisance release orders are found in 19AB to 19AK of the Crimes Act 1914 (C’th) Part 1B. If the head sentence exceeds three years or, in aggregate exceeds 3 years, the Court is then to fix a non‐parole period or make a recognisance release order.40 If the sentence is 3 years or less than 3 years, the Court must make a recognisance release order.41 The total sentence must be set first taking into account the requirement to “impose a sentence or make an order that is of a severity appropriate in all the circumstances of the case “ as directed by section 16A(1) and taking into account the factors identified in section 16A(2) “as are relevant and known to the court”.42 Under Commonwealth law it is not appropriate to impose the non‐parole period first and then determine the head sentence. The NSW Court of Criminal Appeal has noted that, “it is only after a proportionate sentence is imposed that the Court is to fix a non–parole period or make a recognisance release order”.43 As the Queensland Court of Appeal has noted with respect to Part 1B: “There are no criteria by which the non parole period is to be determined. Nor are there any parameters set for it, other than the requirement in s19AF that it not extend beyond the end of the sentence, as reduced by any remissions.”44 The High Court has noted however that, “except to the extent stated in Sections 16A and 16B of the Crimes Act 1914 (C’th), general common law and not peculiarly local or State statutory principles of sentencing are applicable”.45 So, because there is no general law on the proportion of the non‐parole period or recognisance release order to the head sentence, the common law will apply. In a series of cases the High Court has said that at common law the yardstick for fixing the time an offender is to serve before he is released on parole (the non‐parole period) is the “minimum time that a Judge determines justice requires that [the offender] must serve having regard to all the circumstances of his offence”.46 In the landmark New South Wales case of DPP (C’th) v El Karhani, the CCA was required to interpret the then newly enacted Part 1B of the Crimes Act 1914 (C’th). The Court referred to the established practices under the former law in relation to setting non parole periods and considered a bench mark for sentencing under Part 1B. The Court of Appeal noted that: “... in the general run of non‐parole periods a ratio of approximately 60 percent to the head sentence was not unusual under the former law. A ratio of 75 percent... was formerly confined to the most serious of cases. We consider that an appropriate ratio in this case is two thirds of the adjusted head sentence...”47 40
s 19AB Crimes Act. s 19AC Crimes Act, Lodhi (2007) 179 A Crim R 470. 42
s 16A(1) and (2) and Wong v The Queen (2001) 207 CLR 584 at para [71]. 43
R v Lodhi (2007) 179 A Crim R 470 at para [258] per Price J. 44
R v Mokoena [2009] QCA 36 at [9]. 45
Johnson v The Queen (2004) 205 ALR 346 at para [15]. 46
Power v The Queen (1974) 131 CLR 623 at 629, Deakin v The Queen (1984) 54 ALR 765 and Bugmy v The Queen (1990) 169 CLR 525 at 531. 47
DPP (C’th) v El Karhani (1990) 21 NSWLR 370 at 386. 41
8 Variations in Federal Sentencing PART C. SETTING THE NON‐PAROLE PERIOD– APPROACHES BY COURTS The various approaches of the States/Territories on the relationship of non‐parole period to head sentence are set out below. What follows is largely based on a recent survey of approaches throughout Australia by the Western Australian Court of Appeal:48 •
New South Wales49 In New South Wales the “Commonwealth norm”50 or “general guide”51 in terms of the relationship of non‐parole period to head sentence is ordinarily in the range of 60%‐66% but it can be outside this range as the setting of the non‐parole period is a matter of judicial discretion not a statutory ratio mandated by Parliament except in relation to terrorism and people smuggling (see above). A sentence could be well outside that range where most unusual factors justified it. This approach reflects to some degree the influence of state law and practice. Currently s44 of the Crimes (Sentencing Procedure) Act 1999 (NSW) requires in effect that where imprisonment is imposed for a state offence that in the absence of special circumstances the non‐parole period be at least 75% of the head sentence.52 •
Victoria In Victoria there is no state legislation that determines the relationship of any non‐parole period to the head sentence other than the requirements that the non‐parole period be at least six months less than the head sentence.53 The Victorian Court of Appeal has said that there is no standard non‐parole period, standard proportion or standard gap as the setting of a non‐parole period is a matter of judicial discretion.54 This approach applies in practice to both Commonwealth and State matters. In 2006 Justice Redlich of the Victorian Court of Appeal noted that the ALRC reported that case law recognised that the non‐parole period is generally set at 60% to 66.6% of the head sentence with the non‐parole period increasing to 75% in the worst category of case.55 He noted that this accorded with the observations of Callaway JA that “[I]n the majority [of cases] the proportion is between two‐thirds and three quarters but both shorter and longer periods are found”.56 48
Bertilone v R (2009) 31 FLR 383. R v Bernier (1998) 102 A Crim R 44 at 49; R v Stitt (1998) 102 A Crim R 428 at 432; R v Sweet (2001) 125 A Crim R 341 at 346‐347; R v Martinsen [2003] NSWCCA 144 at [14]; Bick v R [2006] NSWCCA 408 at [13]; Ly v R [2007] NSWCCA 28 at [16]; Studman v R (2007) 175 A Crim R 143 at [9]‐[11]. 50
Shen v R [2009] NSWCCA 251 at [11]. 51
James v R [2009] NSWCCA 62 at [15]. 52
In 2004 research by the Judicial Commission of NSW revealed that “special circumstances” must have been found in 87.1% of District and Supreme Court matters where imprisonment was imposed – see R v Fidow [2004] NSWCCA 172 at [20]‐[22]. 53
s 11(3) Sentencing Act 1991 (Vic). 54
R v Ngui and Tiong [2000] 1 VR 579 at 583; R v Harkness [2001] VSCA 87 at [24] and the cases cited therein. 55
R v Tran [2006] VSCA 222 at [27]‐[28]. 56
R v Bolton & Barker [1998] 1 VR 692 at 699. 49
9 Variations in Federal Sentencing An observation has been made by the Queensland Court of Appeal that drug sentences across Australia including Victoria reflect the “norm” or “general guide” of the non‐
parole period being about 60%‐66% of the head sentence as outlined in the New South Wales cases below.57 A number of Victorian decisions agree with that statement.58 •
Queensland In Queensland there is no generally applicable State legislation that determines the relationship of the non‐parole period to the head sentence in State matters. Courts sentencing State offenders may fix a parole release date or fix a parole eligibility date.59 In the absence of an order a prisoner will be eligible for parole after serving half his sentence.60 For certain categories of State offenders however, namely serious violent offenders, sexual offenders and those sentenced to imprisonment exceeding 10 years, parole eligibility must not be ordered before 80% of the head sentence has been served.61 Outside these categories it is not uncommon for State offenders with significant mitigation to receive a recommendation that they be eligible for parole after serving about ⅓ of their sentence. This has evolved from the statutory presumption above and an influential 1986 decision, R v Pickett,62 to become an accepted practice in State matters. It is now accepted as “common sentencing practice” and the “usual” sentence for State matters on a plea of guilty.63 The reasons for this practice were noted by Holmes JA in R v Mokeona [2009] QCA 36 at [12]: ... in Queensland for many years, eligibility for parole for a prisoner not serving a life term, or (since 1997) not subject to a serious violent offence declaration, commenced, in the absence of an order, after the prisoner had served half of his term of imprisonment.64 That regime still applies, albeit to a more limited class of prisoner.65 As a result, the tendency has been to recognise mitigating factors such as co‐operation with the authorities by setting a non‐parole date earlier than that statutory half‐way mark. But that statutory position has no application, direct or indirect, to the process of fixing a non‐parole period for a Federal offence.66 The setting of the non‐parole period is, of course, a matter of individual discretion; but at 57
R v Mokoena [2009] QCA 36 at [10]. Drug sentences for federal offences tend to follow that practice – eg R v Phong (2005) 12 VR 17; R v Thomas [1999] VSCA 204 and R v Ngui and Tiong [2000] 1 VR 579. However it does not seem that any similar uniformity has developed in respect of Commonwealth offences that do not involve drugs – R v Mokoena [2009] QCA 36 at [10]. 59
ss 160B to 160D Penalties and Sentences Act 1992 (Qld). 60
s 184 Corrective Services Act 2006 (Qld). 61
s 182 Corrective Services Act 2006 (Qld). 62
[1986] 2 Qd R 441 and JM Roberston, “Plea Bargaining and the Decision in R v Pickett”, (1987) Queensland Law Society Journal 275. 63
R v Hoad [2005] QCA 92 at para [31], R v Kitson [2008] QCA 86 and R v Mara [2009] QCA 208 at para [28]. 64
s 166(1)(d) Corrective Services Act 1988 (Qld); s 135(2)(e) Corrective Services Act 2000 (Qld). 65
s 184 Corrective Services Act 2006 (Qld). 66
R v To and Do; ex parte Director of Public Prosecutions (C’th) [1999] 2 Qd R 166 at 170. 58
10 Variations in Federal Sentencing the same time, in exercising Federal jurisdiction, consistency with decisions in other jurisdictions is, as this Court observed in R v Tran,67 desirable. The State practice of ordering release on parole at a point less than the halfway point of a sentence (usually after a third) has, as Holmes JA noted, no application to Commonwealth offenders. As is apparent from the statistics below however the widespread State practice has undoubtedly affected federal sentencing outcomes in Queensland until very recently. In February 2009 the Queensland Court of Appeal stated that “the norm” for Commonwealth non‐parole periods is generally considered to be after the offender has served 60%‐66% of the federal head sentence and that a sentence well outside that range would have to have most unusual factors to justify it.68 In February 2010, after this paper was delivered, the Court of Appeal revised this view and held that, setting the time to serve “is the result of the sentencing judge’s discretionary determination of both the appropriate sentence of imprisonment and the appropriate terms of the recognisance release order after taking into account all of the circumstances of the offence, rather than by applying or making adjustments to any rule of thumb”.69 In that decision the Court of Appeal confirmed that State sentencing practice in Queensland continued to have no application to federal sentencing.70 •
Western Australia The Western Australian Court of Appeal recently stated that a review of cases reveals in general the non‐parole periods for Commonwealth drug importation and related drug offences have usually been about 60%‐66% of the head sentence.71 In Western Australia the ratio between non‐parole period and the head sentence imposed for state offences is fixed by s93(1) of the Sentencing Act 1995 (WA) which requires that: −
−
if the term served is 4 years or less the person is to be released when he or she has served one half of the term; or if the term served is more than 4 years the person is to be released when he or she has served 2 years less than the term. 67
(2007) 172 A Crim R 436 at 443. R v CAK and CAL [2009] QCA 23 per Atkinson J at [18] (Muir JA and P. Lyons J agreeing); see also R v Mokoena [2009] QCA 36 and R v Mara [2009] QCA 208. 69
R v Ruha, Ruha and Harris; ex parte C’th DPP [2010] QCA 10 at para [57]. See also the subsequent decisions of R v Chandler [2010] QCA 21 at para [19], R v Fidler [2010] QCA 25 at paras [27] to [29] and R v Marshall [2010] QCA 29 paras [29] to [42]. 70
R v Ruha above at para [55]. 71
Bertilone v R (2009) 231 FLR 383. 68
11 Variations in Federal Sentencing •
South Australia In South Australia a mandatory minimum non‐parole period of four‐fifths (80%) of the head sentence is prescribed for a number of state offences namely “a serious offence against the person” which includes “a major indictable offence (other than murder)” that results in the death of the victim or the victim suffering total incapacity.72 South Australian cases have not identified a particular “ratio” or general guide as to what proportion of the head sentence the non‐parole period should be and this approach applies to both State and Commonwealth sentencing. •
Tasmania Tasmania has no State legislation that governs the relationship of any non‐parole period to the head sentence. No particular “ratio” or general guide as to what proportion of the head sentence the non‐parole period should be and this approach applies to both State and Commonwealth sentencing. •
Northern Territory The Northern Territory has legislation that determines the relationship of any non‐parole period to the head sentence. The scheme is that the non‐parole period must be at least 50% of the head sentence but cannot be less than 8 months. For murder the non‐parole period must be either 20 years or at least 25 years depending on the circumstances. For sexual offences and certain offences against children the non‐parole period must be at least 70% of the head sentence.73 The Northern Territory Court of Criminal Appeal has recently stated that there are sound reasons for suggesting that the norm for non‐parole periods for Commonwealth offences are in the range of 60%‐66% of the head sentence but that does not apply to the fixing of the length of the pre‐release period under a recognisance release order where the time to be served is not determined by reference to a norm.74 PART D. RESEARCH ON FIXING NON‐PAROLE PERIODS AND MAKING RECOGNISANCE RELEASE ORDERS We surveyed Commonwealth sentences imposed with a view to identifying trends with respect to the relationship between the custodial portion of sentences (the non‐parole or pre‐release period) and the head sentence. The research , undertaken, was based on an analysis of all federal sentences [439 in total] imposed across Australia in the last 3 years 2006‐7 to 2008‐9 where the head sentence exceeded 3 years. The analysis was done by State/Territory on the basis of where the sentence was imposed and by crime type irrespective of where the sentence was imposed. 72
ss 32A and 32 Criminal Law (Sentencing )Act 1988 (SA) and s 5(3) Summary Procedure Act 1921 (SA). ss 53A, 54, 55 and 55A Sentencing Act (NT). 74
R v Woods (2009) 24 NTLR 77 at [25]. 73
12 Variations in Federal Sentencing Sentences by State: all The graph shows sentences [439 in total] for all federal sentences where the defendant was sentenced to a head sentence exceeding 3 years in the 3 years between 2006‐7‐2008‐9. Life sentences (3) were excluded. The horizontal axis is the proportion of the non‐parole period (time to be served in custody) of the overall sentence. The gradations along the horizontal axis are in 10 percent increments. The vertical axis is the proportion of all the sentences in excess of 3 years which have that particular percentage of non‐parole period as a percentage of the head sentence. Of significance is that: 13 Variations in Federal Sentencing •
•
•
88%[387] of the federal sentences overall had a non‐parole period of 50% of head sentence or higher. Close to half [211] of the non‐parole/pre‐release periods are in the range 60‐69% of the head sentence The New South Wales pattern had a significant bearing on the national pattern as in the 3 years surveyed the New South Wales sentences accounted for 60% [268] of all relevant sentences. The sentences imposed in New South Wales, Victoria and Queensland account for 90% [391] of all relevant sentences so in effect in combination determine the national pattern 14 Variations in Federal Sentencing New South Wales • Not surprisingly the NSW pattern closely resembles the national pattern as the NSW sentences account for well over half of the relevant sentences. • Approximately 60% [167] of NSW sentences are in the band where the non‐
parole/pre‐release period is 60‐69% of the head sentence. • Approximately 30% [81] of NSW sentences are in the band where the non‐
parole/pre release period is 50‐59% of the head sentence. • So 90% [248] of NSW sentences fall within a band where the non‐parole/pre release period is between 50‐69% of the head sentence. • There is a strong reflection in the NSW graph of a structured approach to the setting of non‐parole/pre release periods in NSW for federal offences even though there is no general statutory requirement to impose a mandatory minimum non‐parole period or one that is a specified percentage of the head sentence‐terrorism sentences and people smuggling sentences are the exception where there is a statutory requirement to impose a mandatory minimum non‐parole period. • The NSW pattern in the graph reflects the consistent statements from NSW Court of Appeal from 1990 recognising a “commonwealth norm” or “general guide” that a federal non‐parole period is usually about 60‐66% of the head sentence. 15 Variations in Federal Sentencing • It is probably also relevant in terms of context that under state law in NSW there is a rebuttable presumption (albeit one often rebutted) that a non parole period be 75% of the head sentence. 16 Variations in Federal Sentencing Victoria •
•
•
•
•
•
Consistently with the NSW pattern only a small percentage of head sentences had a non‐parole/pre‐release period of less than 50% of the head sentence. 87% [59] of the Victorian sentences had a non‐parole/pre‐release period which was half or more of the head sentence. Just over 60% [42] of the Victorian sentences had a non parole period/pre‐release period where this was 60% or more of the head sentence. The corresponding figure for NSW was that 65% [176] of non‐parole/pre‐release periods were in that range. In contrast with the NSW pattern the spread of non‐parole/pre‐release periods was wider particularly in the band where the non‐parole/pre‐release period was between 40% and 79% of the head sentence. The Victorian sentences include a number of terrorism sentences where federal law requires that a mandatory minimum non‐parole of 75% of head sentence be imposed. 30% [21] of the Victorian sentences had a non parole period/pre‐release period which was 70% or more of the head sentence. In Victoria there is no state law, unlike NSW, nor judicial presumption that alters the common law approach to setting the non‐parole period and the wider spread of non‐parole periods in Victoria is not surprising in that context. 17 Variations in Federal Sentencing Queensland •
•
•
•
Consistently with NSW and Victoria over half, but just over half in the case of Queensland, of the sentences had a non‐parole/pre‐release period of 50% or more of the head sentence. In contrast to NSW and Victoria close to half [27] of the Queensland sentences had a non‐parole/pre‐release period of less than 50% of the head sentence and just over one half [28] have a non‐parole period/pre‐release period of at least 50% of the head sentence. That is unlike in NSW and Victoria where there only relatively few sentences with non‐parole/pre‐release period under 50% of the head sentence. In this respect the Queensland pattern reflects state practice described above in the paper. Again unlike NSW and Victoria just less than a quarter [13] of the Queensland sentences had a non‐parole/pre‐release period 60% or above the head sentence. Under state practice in Queensland it is common for state offenders with substantial mitigation who plead guilty to receive a non‐parole/pre‐release period of around 1/3 of the head sentence. 18 Variations in Federal Sentencing •
•
In February 2009 the Queensland Court of Appeal confirmed that this state practice had no application to the process of setting the non‐parole period for federal offences. The pattern of federal sentences in Queensland over the last three years in some important respects does not accord with the national pattern. 19 Variations in Federal Sentencing Western Australia •
•
•
•
Again and consistently with NSW, Victoria and Queensland, over half the sentences had a non‐parole/pre‐release period of 50% or more of the head sentence. Very similar to NSW and unlike Victoria 86%[20] of the WA sentences have a non‐
parole/pre‐release period in the range 50‐69% of the head sentence. The non‐parole/pre‐release period for the WA sentences are clustered in the range 40‐
69% of the head sentence. None were lower or higher than that range. As a matter of context under Western Australian sentencing practice there is a requirement that the non‐parole period for state sentences of 4 years or less be half of the head sentence. 20 Variations in Federal Sentencing South Australia •
•
•
•
Again similar to all the other states over half of the non‐parole/pre‐release periods were 50% or more of the head sentence. In SA just over 90% [12] had a non‐parole/pre‐
release period of 50% or more of the head sentence. A distinctive feature of the SA pattern is that it has two peaks rather than one. Close to 40% [5] of the non‐parole/pre‐release periods were in the range 50‐59% of the head sentence and another 40% [5] were in the range 70‐79% of the head sentence. For major state indictable offences other than murder, a mandatory minimum non‐
parole period of 80% is required in SA. Like Victoria, but unlike NSW, the SA cases have not recognised a norm or general guide as to what proportion of the head sentence the non‐parole period should be. 21 Variations in Federal Sentencing Northern Territory •
•
•
•
•
Again in keeping with all the states covered so far more than half of the sentences had a non‐parole/pre‐release period which was more than 50% of the head sentence. The pattern in NT resembles NSW where the sentences are clustered in a range where the non‐parole/pre‐release period is between 50% and 69% of the head sentence. Recently the NT Court of Appeal stated that there were sound reasons for suggesting that the norm for federal non parole periods (but not pre‐release periods under a recognisance release period) are in the range 60‐66% of the head sentence. As a matter of context Territory legislation in respect of territory sentences required that the non‐parole period must be at least 50% of the head sentence and for sexual offences and certain offences against children the no parole period must be at least 70% of the head sentence. Interestingly in just over half of NT sentences the release mechanism was a recognisance release order rather than a non‐parole period even though the federal head sentence exceeded 3 years. This may reflect territory sentencing practice as under Territory law if the sentence is 5 years or less the sentencing options available include a partially suspended sentence which is the Territory equivalent of a recognisance release order under the Crimes Act 1914 (C’th). 22 Variations in Federal Sentencing Statistics by offence type: all: This graph shows a significant degree of consistency in the pattern for some offence types (drugs, money laundering and people smuggling) and variations on others. 23 Variations in Federal Sentencing Drugs75 This graph shows a significant degree of consistency in the patterns for the total sentences and the drug sentences. The majority of the sentences (70% or 221) occurred in NSW – which we note has a consistent practice of setting the non–parole period at around 2/3 of the head sentence for Commonwealth matters. The strong adherence to that practice in NSW will affect the results. As a matter of context there is a statutory presumption for state offences in NSW for a non‐parole of at least ¾ of the head sentence. However there were a number of drug sentences (30% or 103) outside NSW. If there were variations evident in sentencing practices for drug offences with regard to setting the non‐
parole period in other jurisdictions outside NSW one would expect that to have been evident in the graph but there is little variation. The high level of consistency in federal drug sentences around Australia bears out the observations of the Queensland Court of Appeal76 and the Western Australian Court of Appeal77 and the Victorian Court of Appeal78 to that effect. 75
Drug offences included those under ss 302.2, 307.1, 307.2 and 307.5 Criminal Code (C’th) and ss 233B and 233BAA Customs Act 1901 (C’th). 76
R v Mokoena [2009] QCA 36 at [10]. 77
Bertilone v R (2009) 231 FLR 383. 78
R v Thomas [1999] VSCA 204, R v Ngui and Tiong [2000] 1 VR 579 and R v Phong (2005) 12 VR 17. 24 Variations in Federal Sentencing We noted that 88% [285] of the drug sentences had a non‐parole/pre‐release period of between 50% and 69% of the sentence with over 30% [110] in the range 50%‐59% and over 50% [175] in the range 60‐69% of the head sentence. 25 Variations in Federal Sentencing Money laundering79 Our observations are: •
•
The pattern of sentences for money laundering conforms closely to the pattern in respect of the total sentences and also for drug sentences. In particular, the sentences for money laundering are clustered in the band where the non‐parole/pre release period is between 50% and 69% of the head sentence. 95%[20] of the money laundering sentences are in that band and as noted above close to 90%[285] of drug sentences are also in this band. It is often the case that money laundering is committed along with drug offences and this may explain the strong correlation between the pattern for drug offences and money laundering offences. 79
Money laundering offences included those under ss 400.3, 404.4 and 400.5 Criminal Code (C’th). 26 Variations in Federal Sentencing People smuggling80 •
•
•
There is a statutory mandatory minimum non‐parole period required for people smuggling offences under the Migration Act 1958 (C’th). For those offences an offender must serve a minimum of 5 years imprisonment with a minimum NPP of 3 years imprisonment (i.e. if minimum mandatory sentence imposed, a 60% ratio) or 8 years imprisonment with a NPP of 5 years if a repeat offender (i.e a 62% ratio if minimum imposed).81 The effect of these requirements is evident in the graph for people smuggling offences as all the sentences [7] are clustered in the range where the non‐parole/pre‐release period is between 50% and 69% of the head sentence. It should also be noted that the number of relevant sentences (7) was relatively small. 80
81
People smuggling offences included those under ss 232A and 233 Migration Act 1958. s233C Migration Act 1958 (C’th). 27 Variations in Federal Sentencing Fraud82 •
•
By comparison to the previous graph displaying drug sentences the sentences for fraud/corporate offending are more dispersed. While over 40% [23] of the relevant sentences had a non‐parole/pre‐release period of between 50%‐69% of the head sentence 36% [19] are in the range below 50% of the head sentence. As well 13% [7] were higher than the cluster in the range 50‐69% and were in the range 70‐79% of the head sentence. This is in contrast to the drug sentencing pattern where they are concentrated in the band 50 – 69% and where less than 5% [14] are 70% or more of the head sentence. The wider spread in the pattern for fraud/corporate offending may reflect the wide range of offences within that group of offences and a wider variety of circumstances of offending than for drug offences. 82
Fraud offences included those under ss 134.1, 134.2, 135.1, 135.4 and 145.1 Criminal Code (C’th) and s 29D Crimes Act 1914. 28 Variations in Federal Sentencing Child sex offending83 •
•
The pattern in respect of child sex crimes does not conform to national pattern. In particular 35% [5] of sentences had a non‐parole/pre release period that was less than half the head sentence. Consistently with the national pattern for all sentences there was also a cluster of 57% [8] of the sentences in the range 50‐69% of the head sentence. It should be noted that there were only a relatively small number of relevant sentences in the offence type of child sex crime and the offence type includes a variety of offences only some involving contact offences with children. 83
Child sex offences included those under ss 474.19 and 474.26 Criminal Code (C’th) and ss 50BA and 50BC Crimes Act 1914. 29 Variations in Federal Sentencing Terrorism84 •
•
As noted above there is a requirement that the non‐parole period for a terrorism offence be at least 75% of the head sentence.85 As a result of the pattern for terrorism offences is that the non–parole period is clustered in the range of 75% and above. In respect of all terrorism sentences (11) the non‐parole period was 75% or more of the head sentence. In over a third of the sentences (4 in total) the non‐parole period was 80% or more of the head sentence. 84
85
Terrorism offences included those under ss 101.4, 101.5, 101.6 and 102.3 Criminal Code (C’th). s19AG Crimes Act 1914 (C’th). 30 Variations in Federal Sentencing Human trafficking86 •
•
The pattern of sentences for human trafficking has a broader spread than the national pattern for all federal sentences. Consistently with the national pattern for all sentences there is a cluster of 57% [4] of the sentences where the non‐parole/pre‐release period is in the range beyond 50% and up to 69% of the head sentence. There were also 28% [2] of the sentences where the non‐parole/pre‐release period was in the range between 40 to 49% of the head sentence. It should also be noted that the number of relevant sentences (7) was relatively small. 86
Human trafficking offences included those under ss 270.3, 270.6 and 271.2 Criminal Code (C’th). 31 Variations in Federal Sentencing PART E: CONCLUSION From the survey of 439 federal sentences conducted over the 3 years between 2006/7 and 2008/9 our main conclusions are: • There is a reasonable level of consistency across Australia in the pattern of non‐parole periods when compared to head sentences. For example in all jurisdictions half or more of the sentences had a non‐parole/pre‐release period that was 50% or more of the head sentence. • That there is a reasonable level of consistency across Australia, and not total consistency, is not surprising given the context in which federal offenders are sentenced particularly that this takes place in State/Territory courts and that State/Territory sentencing practices and approaches most likely play a role in the outcome. • There is a significant degree of consistency for drug offences in the pattern across Australia of non‐parole periods/pre‐release periods when compared to head sentences. The observations by the Queensland, Western Australian and Victorian Courts of Appeal to that effect are made out by the research. • There is also a very similar sentencing pattern in the relationship of non‐parole/pre‐
release periods to head sentence for drug, money laundering people smuggling sentences. • State/Territory sentencing practice in the period reviewed appears to have influenced the sentencing pattern in each jurisdiction in respect of federal sentences. • The State sentencing practice in Queensland tends to result in a wide gap between the head sentence and non‐parole period. The application of that state practice to federal sentencing in Queensland has resulted in close to half of the federal sentences in Queensland having a non‐parole/pre‐release period that was less than 50% of the head sentence. State sentencing practice in Queensland appeared to have had a pronounced effect on federal sentencing in Queensland. 32 Variations in Federal Sentencing