Paper 4: Remand in Custody
Proposal
1 This paper outlines current work and further proposals to reduce:
- the proportion of defendants remanded in custody pre-trial (prior to conviction); and
- the time spent on remand in custody.
Executive summary
2 The remand population has steadily increased since 1996 and at a slightly accelerated rate since 2001 (partly due to the Bail Act 2000 coming into force). The increase in the remand population has principally been the result of an increase in:
- the number of prosecutions;
- the proportion of defendants remanded in custody for some types of offence, such as non-cannabis drug offences; and
- the average time spent on remand.
3 One or more of these drivers must be reversed to reduce the remand population. However, there is a risk that any reduction in the remand population may simply result in an increase in the sentenced population. This is because any time spent on remand is taken into account in determining parole eligibility and release dates. To have any real effect on the prison population, any changes must prevent remand in custody of defendants who will either:
- not subsequently be found guilty;
- receive non-custodial sentences; or
- receive custodial sentences shorter than time spent on remand.
4 While the proportion of the remand population that falls into these categories is estimated to be 20 percent at the most, the proportion is likely to be much lower. This is because some defendants must be remanded in custody because of the risk they pose, regardless of the outcome of the case.
5 Two current initiatives are expected to have a significant impact on the remand population over the next few years. First, Police are preparing to roll out electronic monitoring as a condition of bail. Second, the Ministry of Justice and other agencies are working on ways to make pre-trial court processes more efficient.
6 Once the estimated savings from these initiatives are taken into account, there is not a great deal of scope left for further reductions. Radical changes to remand are not recommended because the size of any possible gains would be outweighed by the inevitable increase in the risk of offending on bail and absconding.
7 Instead, it is recommended that the current system be improved by amending the Bail Act 2000 to clarify the threshold for remand in custody, and the relevance of breaches of bail conditions to the decision whether to remand. This will promote consistency in decision-making and is expected to have a modest effect on the prison population, avoiding an estimated total of 40 beds by 2011.
8 It is also recommended that further work be done in the following areas where there is potential to reduce the remand population:
- availability of support services: initial work suggests that some defendants may have difficulty accessing support services (especially housing), which prevents them getting bail; and
- appeals from decisions of justices of the peace (JPs): initial work suggests that if appeals from bail decisions of JPs were heard by a District Court judge, instead of the High Court, this might encourage earlier applications for bail and result in less time spent on remand in some cases.
9 It should also be noted that, to the extent that the proposals coming out of the in this suite of papers result in less imprisonment, they will also likely result in a drop in the remand population. The likely sentence to be imposed is an important consideration when deciding whether to remand in custody or on bail.
Background
10 The Bail Act 2000 arose from a review of the bail system in the late 1990s. The review was in response to:
- growing public, political and media pressure for change; and
- Ministry of Justice research indicating a sizable increase in the rate of offending on bail (especially violent offending) between 1986 and 1994.
11 The Bail Act 2000 brought together a large amount of the existing statutory and common law on bail with two main changes. The Act:
- increased the range of circumstances in which a defendant was subject to a
- reverse onus" of proof (the defendant has to prove he or she should be bailed, rather than the prosecution having to prove he or she should not); and
- provided for breaches of bail conditions to be recorded by the courts, and to be available for subsequent decisions on bail.
12 It is likely that the Bail Act 2000 has had some effect on the proportion of defendants remanded in custody. In particular, the recording and subsequent availability of information on a defendant's breach of bail conditions is likely to have resulted in more defendants being remanded in custody. However, other factors will also have had an effect. For instance, the re-classification of methamphetamine from a Class B to Class A restricted drug, will have made remand in custody more likely for those charged with such offending.
Comment
Introduction
13 Defendants charged with an offence may be remanded on bail with conditions, remanded in custody, or remanded at large (unconditional release with a specified court date). Ninety percent of defendants are remanded at large or on bail.
14 The inherent tension in the pre-trial bail system is finding the appropriate balance between: (a) allowing a defendant to remain in the community, which recognises the fundamental principle that a person is innocent until proven guilty; and (b) incapacitating those who are likely to pose a risk to public safety and/or the administration of justice. Different considerations apply when a defendant has been convicted and remanded in custody awaiting sentence or appeal.
15 The number of defendants remanded in custody has continuously increased since 1996 and at a slightly accelerated rate since the Bail Act 2000 came into force. In June 1999 the average daily custodial remand population (both pre-trial and those awaiting sentence or appeal) was 761, which increased to 1,404 in June 2005 and 1,595 in February 2006.
16 The percentage of people who offended on bail decreased between 1994 and 2003. In particular, there was a drop in violent and property offending. The drop in offending coincided with the increase in the proportion of defendants remanded in custody. This does not necessarily mean that there is a causal relationship between the two, but does highlight the risk that any reduction in custodial remand might result in an increase in offending on bail.
17 To reduce the remand population there must be a reduction in:
- the number of prosecutions; and/or
- the proportion of defendants remanded in custody; and/or
- the average time spent on remand.
18 The relative contributions of these three factors to the remand population are not currently known.
19 Time spent on remand is taken into account in determining parole eligibility and release dates. This means that a reduction in the remand population may simply result in an increase in the sentenced population. To have any real effect on the prison population, any changes to bail must prevent remand in custody of defendants who will:
- not subsequently be found guilty;
- receive non-custodial sentences; or
- receive custodial sentences shorter than time spent on remand.
20 An indicative analysis suggests the number of defendants that fall into these categories is no more than 20 percent (319 beds on February 2006 figures). This takes into account the relative time defendants in these categories spend on remand and controls for problems in the underlying data. Previous estimates of between 45 to 50 percent did not control (or sufficiently control) for these variables.
21 It is likely that the real potential to reduce the prison population through decreased use of remand in custody is actually much less than 20 percent. Even if a particular defendant may not subsequently be sentenced to imprisonment, it may still be necessary to remand him or her in custody because of the risk he or she poses to the community and the administration of justice at the time.
22 The conclusion from this analysis is that, unless there are significant changes that result in fewer people being sentenced to imprisonment (discussed further below), there is not a great deal of scope for further reductions once the estimated beds to be avoided by current work (electronic monitoring as a condition of bail and improving pre-trial processes) are taken into account. Even if the Government were willing to accept the inevitable increases in the risk of offending and absconding that a radical change would entail, the gains would be small.
Reducing the proportion of defendants remanded in custody
Electronic monitoring
23 On 16 March 2005 the Cabinet Policy Committee requested a number of reports on options to reduce future prisoner numbers, including one on electronic monitoring as a special condition of bail ("electronic monitoring"). On 19 December 2005, the Cabinet Business Committee (CBC) agreed that this report be considered as part of the Effective Interventions in the Criminal Justice System project [ s.9(2)(f)(iv)].
24 In accordance with a CBC decision in December 2005 [s.9(2)(f)(iv)], Police will be managing the phased introduction of electronic monitoring in late 2006/early 2007. Police have indicated that the roll-out will begin in September 2006.
25 The national roll-out of electronic monitoring is expected to have a substantial effect on the remand population. Based on figures from the Department of Corrections, Police estimate that up to 120 beds per year could be avoided by the use of electronic monitoring.
26 Police are currently developing strategies to minimise the risks associated with the use of electronic monitoring, particularly around net-widening and offending on bail. To limit net-widening, only those defendants remanded in custody can apply to be monitored. Community safety will be a key consideration in the assessment process.
27 Twelve months after electronic monitoring is rolled out, the Ministry of Justice and Police will report back to justice sector Ministers on whether legislation is desirable to ensure appropriate targeting, and whether Police are the best agency to assume primary responsibility for the regime [s.9(2)(f)(iv)].
The effect of the proposals coming in this suite of papers
28 The proposals in this suite of papers are designed to reduce crime, re-offending and/or the use of imprisonment. To the extent that the proposals result in less imprisonment, it is likely they will also result in a drop in the remand population. This is because the likely sentence to be imposed is an important consideration when deciding whether to remand in custody or on bail. For example, the court will generally only remand a defendant who is unlikely to go to prison if he or she poses a very serious risk of offending, absconding, or interfering with the course of justice.
The legislative threshold for remand in custody
29 There is some confusion among the judiciary about the current test for remand in custody, which in some cases may result in the courts applying a lower threshold than Parliament intended.
30 A defendant who is not bailable as of right must be released on reasonable terms and conditions unless the court is satisfied that there is just cause for continued detention. In considering whether there is just cause for continued detention, section 8(1) of the Bail Act 2000 provides that the court must take into account whether there is "a risk" that the defendant may not answer to bail, interfere with witnesses or other evidence, or offend while on bail.
31 Subsection 8(2) lists a number of other considerations that the court may take into account, such as the strength of the evidence, the likely penalty, and the length of time before trial.
32 Subsections 8(1) and (2) were intended to reflect the common law position, which required the court to weigh the risk posed by the defendant against any matters that made it unjust to detain them. However, the way that these subsections are currently worded has given rise to confusion. A number of High Court decisions are divided on the issue, and the author of Hall's Sentencing has stated that "[c]learly this divergence of views needs urgent clarification".
33 It is recommended that the position be clarified by amending section 8 of the Bail Act 2000 to reflect the original common law position.
34 There is also a further issue with section 8: the Bail Act 2000 does not elaborate on what constitutes "a risk" for the purposes of subsection 8(1).
35 In R v Hines (CA384/02, 29 November 2002) the Court of Appeal implicitly agreed with the High Court that to amount to just cause for continued detention, the risk that a defendant may interfere with a witness should be "more than nebulous and insignificant and should be a real and significant risk". In line with Hines, it is recommended that the threshold in section 8(1) be clarified by providing that "a risk" of the defendant not answering bail, interfering with witnesses or other evidence, or offending while on bail, means a "real and significant risk".
The relevance of breaches of bail conditions
36 In R v Ropiha (CA325/05, 28 September 2005) the Court of Appeal considered an appeal against a decision not to readmit to bail after a breach of conditions. After dismissing the appeal on the facts, the Court made the following observation:
[T]he allowance of bail, on conditions, is on the footing that those conditions must be strictly observed. Arguments about the "degree" of breach will not generally avail a party who breaches those conditions. That is not to say that in unusual cases the breach might be excused. But in the absence of a truly compelling excuse, the conditions must be faithfully adhered to, and, if they are not, the bailee can expect to be remanded in custody.
37 On its face, this could be interpreted as requiring defendants to be remanded in custody if they breach their bail conditions, unless there is a compelling excuse for the breach. This interpretation is inconsistent with the policy underlying the use of conditions.
38 Conditions are imposed to mitigate the risk that a defendant will abscond, offend, or interfere with witnesses or evidence. To the extent that the breach of a condition is relevant to assessing these risks, it is open to the court to take the breach into account in determining whether the defendant should be remanded in custody. While it is important that court orders be obeyed, the court should not remand a defendant simply because he or she has breached a condition, unless the breach indicates that he or she now poses an unacceptable risk.
39 The amendments to section 8 of the Bail Act 2000 outlined above make this point clearer. However, it is recommended that an "avoidance of doubt" provision (i.e. a subsection indicating that breaches of conditions are only relevant to future risk, rather than requiring punishment through loss of liberty) also be included in section 8 to remove any remaining ambiguity.
40 The three amendments to section 8 will promote consistency in decision-making and are expected to have a modest effect on the prison population, avoiding an estimated total of 20 to 60 beds (with a mid-point of 40) by 2011.
Bail outcomes of Māori and Pacific defendants
41 A higher proportion of Māori and Pacific defendants than European defendants spend at least some time remanded in custody. The bail outcomes are shown in the table below.
Bail outcomes of Māori, Pacific and European defendants 2001 to 2005
| Ethnic Group | Custody only (%) | Bail and custody (%) | Bail only (%) | Other* (%) |
| Māori | 5.9 | 9.7 | 50.3 | 34.0 |
| Pacific peoples | 4.3 | 7.8 | 50.2 | 37.7 |
| European | 3.1 | 6.0 | 47.3 | 43.6 |
*The majority of the "other" category are those remanded at large. Source: Ministry of Justice 2006
42 The difference in the relative rates is partly explained by the fact that Māori and Pacific peoples are over-represented in the number of people charged with certain offences that are more likely to result in remand in custody, such as aggravated robbery and, for Māori, burglary. However, this does not entirely account for the difference. A lack of access to appropriate support services may be another reason why a higher proportion of Māori and Pacific defendants than European defendants spend at least some time remanded in custody (see below).
Availability of support services
43 The inability of defendants to access support services, such as appropriate housing, may result in remand in custody rather than bail.
44 Adequate support, including stable and appropriate accommodation, is an important factor in bail decisions as it helps mitigate the risk that a defendant may abscond or offend. Bail is more likely to be opposed, and remand in custody is a more likely outcome, if defendants do not have access to appropriate and adequate support in the community. Adequate support is also important to help bailees comply with any conditions of their bail.
45 Informal practices have developed in some courts. For instance, in the Wellington District Court, Salvation Army staff often assist defendants to find accommodation. However, it is not known if such practices are widespread, or whether they are sufficient to meet the needs of defendants.
46 Better information is required on: (a) what support services are available to defendants; (b) whether defendants can and do access them; and (c) how this affects their ability to get bail. It is recommended that the Ministry of Justice include, in the programme of research to be considered arising from the proposals in this suite of papers, an investigation of defendants' access to support services, with a particular focus on Māori and Pacific defendants. Resourcing for the programme of research will be sought through the Budget process.
47 If an unmet need is identified, options that could be considered include bail hostels, and dedicated staff to assist defendants to access services, such as the Credit Bail Support Program in Victoria, Australia.
Reducing the time spent on remand
48 One of the key drivers of the remand population is the delay before cases are brought to trial or resolved pre-trial.
49 A number of agencies have work under way to improve the practice and efficiency of pre-trial processes that should help reduce the time defendants spend on remand, including:
- the Ministry of Justice's Service Improvement Program (SIP);
- the Legal Services Agency's Review of Initial Criminal Legal Services; and
- the work being done by government agencies in response to the Law Commission's report Criminal Pre-Trial Processes - Justice Through Efficiency.
50 As outlined in a report to CBC on 14 September 2005 [s.9(2)(f)(iv)], it is not possible to give an accurate estimate of the likely effect of improving pre-trial court processes on the remand population and prison population as a whole. However, it is known that the areas where savings can be made are in the summary jurisdiction and, to a lesser extent, cases going to depositions and jury trials.
51 The Ministry of Justice will report to the Ministers of Justice, Finance and State Services, and the Minister for Courts (who collectively have oversight of the SIP) by s.9(2)(f)(iv) on the results of a closer analysis of the causes of unnecessary delay and a work programme to give effect to changes.
Appeals process
52 Appeals against bail decisions of a community magistrate are heard by a District Court judge and appeals against decisions of a justice of the peace ("JP") and District Court judge are heard by the High Court. Having appeals against decisions of JPs heard by a District Court judge could encourage early bail applications and could reduce the time offenders spend on remand.
53 It is recommended that the Ministry of Justice investigate whether it is desirable for District Court judges to also hear appeals from decisions of JPs, bearing in mind the Government's response to the recommendations around appeal structures in the Law Commission's report, Delivering Justice For All. It is recommended that work on this issue be progressed in line with the report back to justice sector Ministers on electronic monitoring (see above), so that the same legislative vehicle can be used if required.
54 Having appeals against decisions of JPs heard by a District Court judge would have resource implications for the District Court. A possible risk would be an increase the number of unmeritorious appeals. There is a also a political risk that the proposal could be seen as downgrading the decisions of JPs, which in turn could be perceived as contrary to the spirit of the recently introduced Justices of the Peace Amendment Bill.
Recommendations
55 It is recommended that the Cabinet Policy Committee:
1 note that improvements to the current remand system are proposed, rather than radical reform;
2 note that Police will be managing the phased introduction of electronic monitoring as a special condition of bail in late 2006/early 2007 (Police have indicated that the roll-out will begin in September 2006) and that it is estimated that this will save up to 120 beds per year;
3 agree that section 8 of the Bail Act 2000 be amended to clarify:
3.1 the status of the considerations relating to just cause for continued detention in subsections 8(1) and (2), to align with the original common law position;
3.2 the threshold for just cause for continued detention in section 8(1), to align with the Court of Appeal's decision in R v Hines;
4 agree that an "avoidance of doubt" provision on the relevance of breaches of bail conditions to the decision whether to remand in custody be included in section 8;
5 direct the Ministry of Justice to include in the programme of research to be considered from the Effective Interventions proposals, an investigation of defendants' access to support services, with a particular focus on Māori and Pacific defendants. s.9(2)(f)(iv)
6 note that Ministry of Justice and other agencies are working on ways to make pre-trial court processes more efficient, including implementing those recommendations in the Law Commission's report on pre-trial processes that were accepted by the Government;
7 note that the Ministry of Justice will report to the Ministers of Justice, Finance and State Services, and the Minister for Courts, by s.9(2)(f)(iv), on the analysis of the causes of unnecessary delay in the pre-trial court processes, and a work programme to give effect to changes;
8 direct the Ministry of Justice to do further work on whether appeals from bail decisions of justices of the peace should be heard by District Court judges (instead of the High Court). Work on this issue will be progressed in line with the report back to justice sector Ministers on electronic monitoring as a condition of bail, 12 months after roll-out.
Hon Mark Burton
Minister of Justice
On behalf of:
Hon Rick Barker, Minister for Courts
Hon Damien O'Connor, Minister of Corrections