Electronically monitored curfew as a condition of bail report of the pilot

Electronically monitored curfew as a condition of bail report of the pilot by Jennifer Airs Robin Elliott Esther Conrad

Electronically monitored curfew as a condition of bail report of the pilot by Jennifer Airs Robin Elliott Esther Conrad

Additional copies of this report can be obtained from the Home Office, Communications and Development Unit, Research, Development and Statistics Directorate, Room 201, 50 Queen Anne's Gate, London SW1H 9AT. Telephone 020 7273 2084 The Research, Development and Statistics Directorate RDS is part of the Home Office. The Home Office's purpose is to build a safe, just and tolerant society in which the rights and responsibilities of individuals, families and communities are properly balanced and the protection and security of the public are maintained. RDS is also a part of the Government Statistical Service (GSS). One of the GSS aims is to inform Parliament and the citizen about the state of the nation and provide a window on the work and performance of government, allowing the impact of government policies and actions to he assessed. Therefore - Research Development and Statistics Directorate exists to improve policy making, decision taking and practice in support of the Home Office purpose and aims, to provide the public and Parliament with information necessary for informed debate and to publish information for future use. The views expressed in this report are those of the authors, not necessarily those of the Home Office (nor do they reflect Government policy). Crown copyright 2000 Application for reproduction should be made to the Communications and Development Unit, Room 201, Home Office, 50 Queen Anne's Gate, London, SW1H 9AT. ISBN 1 84082 505 7

Acknowledgements We would like to thank the many people who assisted in the production of this report. In particular, we are very grateful to all the court staff who assisted in data collection: especially David Scanlan at Manchester City magistrates court and David Cupit and Helen Chatten at Norwich magistrates court. We are also indebted to the monitoring companies, particularly Andy Homer and Nicky Bickle at Premier Monitoring Services, and Patricia Walmsley and Maurice Skinner at Securicor. We are grateful that so many people agreed to be interviewed: court staff, magistrates, defence lawyers, crown prosecutors, police officers, the electronic monitoring companies, probation officers, police officers, offenders and their families. We would like to thank Marie Howes, Jane Woodrow and Ruth Stanley who interviewed defendants and their families on our behalf, and all the members of the Steering Group who oversaw the research and the Interagency Groups which smoothed the path of the project. Within the Research, Development and Statistics Directorate we would like to thank Karl Johnston for costing the pilots, Stephen Niven and Ruth Lewis for their help in collecting and entering data, and Ruth Lewis and Ron Haggar for helping to edit interview transcripts. We are also grateful to the students who assisted in the project, particularly with the onerous and routine but essential task of data collection from court records: Natalie Wood, Stefan Webb, Jessica Kruger, Rosie Woolley, Caitlin Garland, Angela Shaw, and Claire Easton. Jennifer Airs Robin Elliott Esther Conrad i

ii

Contents Acknowledgements Summary 1 Introduction 1 Legislation 1 Electronically monitored bail trials 3 Learning from the past 4 Bail in Manchester and Norfolk 6 2 The bail snapshot 9 Background to the snapshot 9 Types of bail used 9 Conditional bail 14 Young defendants 20 Conclusions 21 3 The bail pilot 23 Bail in Manchester and Norfolk 23 Relationships in the pilot sites 30 Defendant characteristics 31 An alternative to custody? 36 Young defendants 40 Interviews with defendants and their families 43 Breaches and violations 46 Operational implications 48 Problems 49 Summary 53 4 Costs 55 Cost assumptions 55 Savings 56 Costs of a national roll-out 57 5 Conclusions 59 Low take-up may be due to target population 59 An alternative to custody for more than half those curfewed 59 Many criminal justice agencies and defendants thought bail curfew was a successful alternative to custody 60 More than five per cent of defendants with bail curfews absconded 60 Some defendants violated their curfew and were not reported for breach action 60 Operational difficulties show that better procedures are needed 61 Where next for bail curfew? 61 Page i v iii

Appendix 1 Methodology 63 Appendix 2 Steering Group and Inter-agency Group membership 65 References 67 iv

Summary Introduction The presumption of innocence is central to our criminal justice system and is reflected in the statutory right to bail enjoyed by unconvicted defendants. However, there is also a need to ensure i) that any criminal proceedings progress swiftly and without interference; and ii) that the public is protected from any danger posed by the defendant awaiting trial. These two factors need to be weighed in the balance by the police and the courts who determine whether a defendant is released or detained. The police must make this decision whenever they charge a suspect and if they decide to release him/her on bail they may attach conditions to ensure that the suspect does not present a bail risk. If they decide to detain the suspect they must bring him/her swiftly to court, usually within 24 hours. The court must then, and at each stage of proceedings thereafter, decide whether the defendant should be remanded on bail or in custody. As with the police, the court may decide to attach conditions to bail. Bail in Manchester and Norwich This study looks at electronically monitored curfew (bail curfew) as an alternative to court remand in custody. It built on an earlier study in 1989 and 1990 (Mair and Nee 1990) where there were severe problems with equipment. The new pilot ran at two sites, Manchester and Norwich, between April 1998 and August 1999. A sample of 9,000 remand decisions (affecting 5,200 defendants) was collected from Manchester City and Norwich magistrates courts to provide a snapshot of remand in the courts at the heart of the study. Various characteristics of defendants receiving the main types of remand (in custody, conditional bail, and unconditional bail) were identified to provide a yardstick for bail curfew. These characteristics were used to measure the likeness/unlikeness of defendants granted bail curfew with other groups of defendants. We collected information from courts and the electronic monitoring contractors. We also interviewed many representatives of criminal justice agencies, 24 offenders and 31 members of their families. In all, 198 bail curfews were made on 173 individuals. Very few women were curfewed. Most men curfewed were aged between 17 and 35. Take-up was lower than planned: partly because the measure was not used by all magistrates and judges in the pilot areas, and partly because there were comparatively few defendants who could meet the residence requirements and be considered sufficiently trustworthy to be released from custody. New violation procedures were developed to ensure that early action was taken against defendants who broke or violated their curfew in some way (for example, by damaging the equipment or by being absent during the curfew period). Two-thirds of defendants violated their curfews: 95 had v

lesser violations recorded (these were typically a few minutes absence at the start of the curfew period) and these did not trigger breach action. Forty-two defendants had breach action taken against them: in all 24 were remanded in custody. We know that three defendants had serious violations recorded against them but were not reported for breach action and that a further seven were arrested by the police on suspicion of re-offending. A total of 11 defendants absconded. In all, 142 defendants completed their bail curfew some after being breached and having their bail curfew continued. A number of problems arose during the period of the study, many of them because of lack of interagency communication. It is clear from some of the problems that have arisen that it has not been easy for criminal justice agencies to apply or administer this new measure. We also found inequalities in use by magistrates and judges: including variable application and use of the measure, and treatment of breach. The police were concerned that defendants, who would otherwise have been in custodial remand, were again within police jurisdiction. They were also concerned that they had no right of access to a defendant s home when a curfew violation was reported. Other criminal justice agencies had concerns about the potential for domestic violence not being identified during the limited curfew assessment produced for the remand decision we found no evidence of domestic violence although there were reports of aggression. Electronic monitoring contractors had concerns about retrieving equipment from defendants after the bail curfew ceased. The bail curfew was generally popular with defendants and their families (until they found themselves sentenced to custody and regretted losing the remission that custodial remand would have attracted). The role of women was found to be crucial to the successful operation of many of the bail curfews. It is clear from comments made during interviews that many women made personal sacrifices for the family as a unit and there would have been fewer successful bail curfews, particularly for young defendants, if women had not done so. There were reports of aggression and increased tension within families, although some reported improved relationships. There were some problems with the equipment although these were less severe than those of the first bail pilot. An alternative to custody? There was no consistent evidence that a bail curfew provided a true alternative to custodial remand. There were strong indications that it had been used in place of custodial remand for some defendants, but there were equally clear indications that bail curfew had been used as an additional bail condition for others. After weighing up the evidence we concluded that bail curfew had been used as a true alternative for over half those curfewed under the trial. Costs Costs of the new measure will depend on how widely it is used. Estimates based on the use made of it in Manchester and Norwich provided a base to calculate likely take-up in metropolitan and non-metropolitan areas. The calculations for bail curfews are based on a fixed price per curfew since curfew orders are currently charged at a fixed rate regardless of their length. vi

We have compared the cost of bail curfew with the cost of remand in custody, taking into account the likelihood of a longer period on bail curfew than in custody, and the lack of remission for any sentence from bail curfew. Overall we estimate that savings to the prison population will be outweighed by the cost of adding electronically monitored bail to the list of conditions available to courts. Assuming that only half the estimated number of electronically monitored bail orders made were for defendants who would otherwise be remanded in custody (about 2,500 at magistrates courts and about 100 at the Crown Court), and the remaining 2,500 from magistrates courts and 100 at the Crown Court would otherwise have been granted conditional bail the cost of a national roll-out (to England and Wales) would be 1.53 million. We also estimate that if this measure were used only for defendants who would otherwise be remanded in custody, then the saving from a lower unsentenced prison population would outweigh the cost of bail curfew orders. These figures are volatile and do not included full costs of breach action because not all the information needed was available. Conclusions We concluded that: Low use of the measure might be due, at least in part, to a small target population because few defendants have the type of accommodation needed for the equipment, or because of the nature of the alleged offence, or because of the defendants previous record. Bail curfew had been used as a true alternative to custodial remand for at least half those curfewed although there was some evidence of net widening with bail curfew being used as an additional bail condition for some defendants. Bail curfew worked as an alternative to custodial remand. Although some defendants absconded, this was a lower proportion (6%) than recorded as absconding from conditional bail in available national and local statistics. National statistics show 12 per cent of bailed defendants abscond from magistrates court and nine per cent from Crown Court. Breach action was not correctly taken in all cases. This was partly due to poor differentiation between bail curfew and curfew as a sentence by the contractors. Good procedures were developed by all agencies for the trial but were found to fall short in some areas because of staff turnover, insufficient training, overlooked or forgotten procedures, and unforeseen circumstances. There are three options for bail curfew: to extend the pilot (which has now finished), to roll-out nationally, or to do nothing. vii

viii

1 Introduction Legislation A balance is needed between the needs of the community and the rights of the individual when making remand decisions. The community expects an offender to receive justice. The community also expects to be protected from intimidation by alleged offenders attempting to avoid justice, and from offending by those awaiting trial. At the same time, an unconvicted defendant is entitled to be presumed innocent and the courts must balance this right against the need to protect the community. The bail legislation provides the framework for making such decisions. The Bail Act 1976 provides a general right to bail for unconvicted defendants, and the court must consider granting bail whenever a bail application is made and/or whenever the defendant appears before it and a remand decision is made. Remand decisions Magistrates courts may remand a defendant in custody or on bail for any offence: on adjourning a case for any reason; on committing a case to the Crown Court; or on an appeal against the magistrates court s decision being lodged. The Crown Court may remand a defendant in custody or on bail: on adjourning a case for any reason; on an appeal against the Crown Court s decision being lodged; or where the defendant has been remanded in custody by the magistrates court. Before the court makes a remand decision, the Crown Prosecutor may give their opinion of the appropriateness of bail and the defence solicitor may make representations on behalf of their client. In making a remand decision the court is attempting to predict whether defendants will present themselves to court when required, whether they will attempt to contact or intimidate prosecution witnesses, and whether they will commit further offences. This decision is thus very different from other decisions made by the court. Morgan and Henderson (1998) found evidence that magistrates do remand defendants in custody when they are more likely to offend on bail. Grounds and reasons for remand decisions Grounds for refusing bail to a defendant are referred to in some detail in Chapters 2 and 3 and are listed here. (i) Grounds which apply in all cases 1

The defendant need not be granted bail if: a) the defendant has previously been released on bail in the same proceedings and has been arrested for absconding while on bail or breaking his conditions of bail; b) the court is satisfied that the defendant should be kept in custody for his own protection or if he is a child or young person for his own welfare; c) the defendant is in custody in pursuance of a sentence of a court. (ii) Grounds which apply in imprisonable offences only If the court is satisfied that there are substantial grounds for believing that the defendant, if released on bail (whether subject to conditions or not) would: a) fail to surrender to custody, or b) commit an offence whilst on bail, or c) interfere with witnesses or otherwise obstruct the course of justice; bail may be refused. In addition the defendant need not be granted bail if: a) the offence is an indictable or triable either way offence; and b) it appears to the court that the defendant was on bail in criminal proceedings on the date of the offence. (iii) Grounds which apply in non-imprisonable offences only The defendant need not be granted bail if: a) the defendant has previously failed to answer bail; and b) the court believes that if released he would fail to surrender to custody. (iv) General ground if no information In imprisonable offences the defendant need not be granted bail if the court is satisfied that it has not been practical since the proceedings were instituted to obtain sufficient information to make the bail decision. (v) Bail in certain serious cases Where a defendant is charged with (attempted) homicide or rape and has a previous conviction for one of these offences, the court may grant bail only where exceptional circumstances apply. When deciding whether or not the grounds have been substantiated, the court must consider, among other relevant issues: the nature and seriousness of the offence and the probable method of dealing with it; the character, antecedents, associations and community ties of the defendant; the defendant s previous record under grants of bail; the strength of evidence against the defendant. If, having considered these factors, the court is of the opinion that there are no substantial grounds for believing that if released on bail the accused would, either: a) fail to surrender, or b) commit an offence, or 2

c) interfere with witnesses or obstruct the course of justice, the court must grant the accused unconditional bail. Conditions may be imposed if it appears to the court that this is necessary to prevent failure to surrender, the commission of an offence on bail, or interference with witnesses or obstruction of justice. The Divisional Court has ruled that conditions may be imposed where there is a real risk of an offence being committed and there do not have to be substantial grounds for believing that conditions are necessary. Courts are required to explain and record bail decisions. Most courts use a multi-part bail form to record reasons for remand decisions, the decisions themselves, and any conditions attached to bail. One part of the form is normally handed to the defendant so that they have a permanent record of the decision. Young defendants Young defendants (generally those under 18) remanded in custody might be remanded into local authority care under the Children and Young Persons Act 1969. Since 1 June 1999, certain young people could be remanded to secure accommodation in certain circumstances. No such remands were made in respect of persons given a bail curfew under this pilot. Electronically monitored bail trials Conditional bail with curfew (where the conditions include residing at a particular address between specified hours) has been in use for some time. An extension of this set of conditions is to use technology to monitor the curfew that is, to use electronically monitored curfew as a condition of bail. The aim of using such conditional bail is to replace some custodial remands with conditional bail and thus reduce the size of the prison population. Remand prisoners account for about a fifth of the prison population but, because their average time spent in custody is low 1, they form a larger proportion (about three-fifths) of the number of people received into prison. The number of remand prisoners received into prisons between 1988 and 1998 in England and Wales is shown in Figure 1.1. 1 The average time spent in custody during 1998 by male remand prisoners was 47 days, compared to 15.4 months for sentenced adult males. Similarly, the average time spent in custody during 1998 by female remand prisoners was 34 days, compared to 11.3 months for sentenced adult females. 3

Figure 1.1: Remanded prisoners received into prison, 1988 to 1998 75,000 Numbers of remand prisoners 60,000 45,000 30,000 15,000 0 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 Source: Prison statistics England and Wales 1998 The number of remand prisoners fluctuated from one year to another. Overall, the number fell from 59,000 in 1989 to 50,000 in 1992 before rising to 65,000 in 1998. The first trial of electronically monitored curfew as an alternative to remand in custody ran in Nottingham, North Tyneside and Tower Bridge in London in 1989 and 1990 (Mair and Nee, 1990). There were several technical problems with the equipment and take-up was too low to permit detailed analysis or conclusions to be drawn. There was early adverse publicity and magistrates lacked confidence in the new measure. Poor communication between the various criminal justice agencies exacerbated the problems. The system was not cost-effective, but this was not surprising given the low level of use and high initial setting-up costs. This, the second trial, ran at larger courts in two cities: Manchester and Norwich. Crown Courts in these cities are able to make electronically monitored bail decisions, but the majority of such decisions (as expected) were made at the magistrates court. Cost considerations limited the trial to 200 cases. In practice, 198 orders were made. Learning from the past Lessons were learnt from the first bail trial and efforts made to avoid similar problems. In the first trial all defendants had been remanded in custody and were fitted with a Personal Identification Device (PID) while still in prison. They were then released to a home which had a Home Monitoring Unit (HMU) already set up. Many of these defendants were curfewed for 24 hours a day and it became apparent that this was unworkable for most defendants because they needed to leave the house occasionally to buy food or sign on for benefit. Particular problem areas from the first trial are looked at in more detail below. These include equipment and technical problems, low take-up rates, communication and costs. 4

Technical problems There were frequent false readings from the tamper alert mechanism of the monitoring equipment in the first trial. More minor problems, such as blown fuses in the HMU, also occurred. The equipment was simply not reliable enough. After the trial concluded, the Home Office kept a watching brief on electronic monitoring developments in other countries, particularly in the United States of America. Home Office officials developed a specification for a further British trial of electronically monitored curfew orders as a sentence under the Criminal Justice Act 1991 (CJA) (Mair and Mortimer 1996, and Mortimer and May 1997). The specification, although based on existing systems, could only be met by modifying and developing an existing system. Evidence from abroad suggested that advances in technology shifted problems to the operational, rather than technical, aspects of electronic monitoring (Mair and Mortimer 1996). The specification produced by the Home Office for the latest trials covered technical matters but also required minimum standards in such areas as staff training, company organisation, documentation and administration. Modern monitoring technology has proved reliable and is now operating throughout the country for Home Detention Curfews (Dodgson and Mortimer, 2000) and curfew orders made under the CJA. More than 20,000 people have been subject to electronic monitoring since January 1999 and few problems have been reported. Low take-up rates Low take-up rates were a particular problem with the first bail trial, not least because sentencers and courts were unfamiliar with the measures. Orders were only permitted for defendants accused of violent offences and this resulted in a smaller number of tagged 2 defendants than might have been expected because most of these were young men who were less likely to have living arrangements suitable for the measure. In addition, there was slow growth in the number of curfew orders made. One way of minimising low take-up for this trial was to remove the restriction on type of offence. Another was to pilot the new measures in areas where electronic monitoring had been in use for some time under the CJA. Take-up rates in Greater Manchester and Norfolk were very low at first under the CJA pilots, but eventually reached acceptable levels. Take-up was expected to be lower for bail than for the other curfew pilots in any event. Few defendants held in custodial remand would be suitable for bail with curfew. Such people would not only need to be considered unlikely to re-offend, but would also need to have stable and suitable accommodation to which they could be tagged. Special training sessions about the new measures were arranged for all magistrates in both magistrates courts, and for judges in Manchester. Regular newsletters have been circulated to courts and others as a reminder about the trial. 2 Curfew devices are referred to as tags. 5

Communication Another problem with the first bail trial was a lack of effective inter-agency communication. Local, inter-agency communication was a high priority in this trial. Liaison committees with representatives from all local criminal justice agencies (which had been set up for the earlier CJA pilot) were used in both Manchester and Norfolk as the primary loci for communications. The groups met regularly with Home Office officials, but also met without them and developed a strong team spirit. This spirit helped to prevent problems by ensuring that potential difficulties were discussed in advance, both inside and outside meetings. Strategic decisions about the trials were made by a Steering Group, which included a representative from each local committee. This ensured that local groups had a hand in managing and guiding the trials and were aware of strategic decisions. Costs The first bail trial, having high start-up costs and very low usage, was not cost-effective. Running this trial in areas where electronically monitored curfews are regularly made under the CJA reduced start-up and running costs. Home Detention Curfews and CJA curfews are now available nationally and these will result in certain economies of scale. Possible future costs are explored in Chapter 4. Bail in Manchester and Norfolk The bail trial started on 1 April 1998 in Norwich and on 1 July 1998 in Manchester. The start was staggered to allow the electronic monitoring contractors time to overcome any start-up difficulties with the Crime (Sentences) Act (CSA) pilots 3 before starting the bail pilot. Other measures under the CSA are described in Elliott and Airs, 2000. The bail pilot was limited to Manchester City magistrates court, Manchester (Crown Square) Crown Court 4, Norwich magistrates court and Norfolk Crown Court for two reasons. These magistrates courts had higher case loads than other courts in their areas and would be more likely to make such bail orders than smaller courts. The possible cost of the measure also had to be considered. The Crown Courts were included because the magistrates courts committed more cases to them than to other Crown Courts. This would enable defendants to continue their bail curfew once committed to the Crown Court. Magistrates were asked to ensure that electronically monitored bail was used only for defendants who would otherwise have been remanded in custody. Sentencers were advised that eligible defendants would need to meet certain requirements, including: being unlikely to breach the order; having a stable home suitable for the equipment (a telephone line could be provided as part of the order, if needed); and having the agreement to the order by other parties involved in the home (family, landlord etc). 3 The CSA pilots had started in Norfolk and Manchester City magistrates courts on 1 January 1998 and other Greater Manchester magistrates courts on 1 April 1998. 4 All electronically monitored bail cases that started at Manchester City magistrates court and were committed to Crown Court were committed to Manchester (Crown Square) Crown Court. 6

The use of electronically monitored curfew as a condition of bail is reported in Chapter 3. The costs of bail curfew are examined in Chapter 4, and our conclusions appear in Chapter 5. Types of bail used One important aspect of the bail trial was to identify differences in the pattern of bail under the trial, to see how the new measure was used. Samples of all bail decisions made at Manchester City magistrates court and Norwich magistrates court for part of the summer of 1997 and part of the summer of 1998 were collected for this purpose. Patterns in these samples are reported in Chapter 2. 7

8

2 The bail snapshot Background to the snapshot One important aspect of the bail trial was to identify whether defendants given electronically monitored bail curfew would otherwise have been remanded in custody. To do this we needed to examine whether they were like those remanded in custody, or more like those granted conditional bail. This chapter examines characteristics of bailed defendants in the two pilot areas to enable these comparisons to be made. All bail decisions made at Manchester City magistrates and youth courts and Norwich magistrates and youth courts for part of the summers of 1997 and 1998 were collected for this purpose we call this the bail snapshot. Cases appearing at youth courts are included with those shown below for the magistrates court. Manchester City and Norwich magistrates courts use pre-printed bail forms to record remand decisions. Manchester City magistrates court records exceptions to the right to unconditional bail found by the court, grounds for finding exceptions to the right to unconditional bail and conditions attached to bail. Norwich magistrates court records grounds for withholding bail, reasons for applying grounds, conditions attached to bail and reasons for applying conditions. A copy of the form is given to the defendant at each remand hearing. Table 2.1 shows that 9,000 records were collected in the bail snapshot. Many individuals made multiple appearances during the sampled periods and the 9,000 records represent 5,200 individuals approximately 2,200 in each year from Manchester City magistrates court and 400 in each year from Norwich magistrates court. This reflects their comparative caseload. Table 2.1: Records collected for the bail snapshot 1997 1998 Total snapshot Decisions Defendants Decisions Defendants Decisions Defendants Manchester City magistrates court 4,081 2,218 3,096 2,172 7,177 4,390 Norwich magistrates court 1,068 438 771 403 1,839 841 Both courts 5,149 2,656 3,867 2,575 9,016 5,231 Types of bail used As explained in Chapter 1, there are three possible types of remand decision. The individuals captured in the bail snapshot were: remanded in custody (18% of defendants) or care (1%) 9

bailed with conditions (53%), or unconditionally bailed (28%). Defendants remanded in custody appeared in the snapshot most often (an average of 1.9 appearances in the periods sampled) and those on unconditional bail the least (1.6). Defendants given bail with conditions appeared an average of 1.8 times and young defendants remanded to care appeared 1.7 times (although there were few such cases and this may not be representative). Figure 2.1 shows the proportionate use of bail and remand in both courts in each year. There were very few remands to care at either court. The proportions of custodial remand were similar at both courts although higher in 1997 at Manchester City magistrates court (22%, as against 15% in 1998) and higher in 1998 at Norwich magistrates court (19%, as against 16% in 1997). These evened out to 19 per cent over the two years from Manchester City magistrates court and 18 per cent from Norwich magistrates court. The proportion of defendants granted conditional bail at Manchester City magistrates' court was lower in both years (50%) than the proportion at Norwich magistrates' court (67% in 1997, 68% in 1998). Unconditional bail was used more frequently at Manchester City magistrates court (27% in 1997, 33% in 1998) than at Norwich magistrates court (16% in 1997, 12% in 1998). Figure 2.1: Types of custody and bail recorded in the 1997 and 1998 bail snapshots 100% Percentage of all records 80% 60% 40% 20% 0% 1997 Manchester 1998 Both 1997 Norwich 1998 Both 1997 Total 1998 Both Source: Court records in the pilot remanded to care remanded to custody conditional bail unconditional bail Custodial remand Table 2.2 shows the number of custodial remands captured in the snapshot by court and year. There were about 1,000 such remands the majority of which were from Manchester City magistrates' court. 10

Table 2.2: Custodial remands captured in the bail snapshot 1997 1998 Total Manchester City magistrates court 481 336 817 Norwich magistrates court 71 77 148 Both magistrates courts 552 413 965 Grounds for the remand decision As explained in Chapter 1: grounds exist for refusing bail where imprisonable offences are concerned: the court may still grant bail if these grounds are found to be unsubstantiated; and the reasons why grounds are substantiated are recorded and a copy given to the defendant. Figures 2.2 and 2.3 show the grounds recorded for each spell of custodial remand at the two courts. Each defendant may warrant any number of grounds for refusing bail and the figures show percentages of defendants with each ground recorded. The most common grounds were: nature of offence (56% at Manchester City magistrates court, 55% at Norwich magistrates court); character/antecedents (56% at Manchester City magistrates court, 39% at Norwich magistrates court); and previous record (21% at Manchester City magistrates court, 43% at Norwich magistrates court). A significant proportion had no grounds given recorded against them (15% in 1997 and 10% in 1998 at Manchester City magistrates court, 8% in 1997 and 4% in 1998 at Norwich magistrates court). Some of the grounds appear in shortened form on the figures: a full listing can be found in Chapter 1. Figure 2.2: Grounds for custodial remand in Manchester evidence of self-injury serving sentence unlikely to co-operate in reports strength of evidence other behaviour towards witness no grounds given previous record nature of offence character/antecedents 0% 20% 40% 60% 80% 1997 1998 both Source: Court records in the pilot area. 11

Figure 2.3: Grounds for custodial remand in Norwich evidence of self-injury serving sentence unlikely to co-operate in reports strength of evidence other behaviour towards witness no grounds given previous record nature of offence character/antecedents 0% 20% 40% 60% 80% 1997 1998 both Source: Court records in the pilot scheme Reasons for the remand decision Figures 2.4 and 2.5 show the reasons recorded for each spell of remand in custody at the two courts. Each defendant may warrant any number of reasons for refusing bail and the figures show percentages of defendants with each reason recorded. The most common reasons at both courts were would fail to surrender (67% of remands at Manchester City magistrates court, 49% at Norwich magistrates court), and would commit offence (62% at Manchester City magistrates court, 60% at Norwich magistrates court). The third most common reason would interfere with witness was given much less frequently, for only 16 per cent of remands at either court. Other reasons were used in comparatively few cases. Some of the reasons appear in shortened form on the figures. Figure 2.4 shows arrested, s7 Bail Act 1976 as a reason for custodial remand. This means that the defendant was arrested because they failed to surrender; or they absconded; or they broke a bail condition; or there is reason to believe that one of these may occur; or a person standing surety either withdrew their surety or notified the court that the defendant was likely to abscond. 12

Figure 2.4: Reasons for custodial remand in Manchester serving custodial sentence custody for own protection/welfare arrested s7 Bail Act 1976 insufficient info for bail decision would interfere with witness would commit offence would fail to surrender 0% 20% 40% 60% 80% Source: Court records in the pilot scheme 1997 1998 both Figure 2.5: Reasons for custodial remand in Norwich impractical to obtain report insufficient info for bail decision previously absconded custody for own protection serving custodial sentence would interfere with witness would fail to surrender would commit offence 0% 20% 40% 60% 80% Source: Court records in the pilot scheme 1997 1998 both There are links between the grounds and reasons shown above. For example, there is an obvious relationship between the ground behaviour towards/proximity to prosecution witness (given in 9% of custodial remand cases in the snapshot) and the reason would interfere with witnesses/obstruct the course of justice (given in 16% of custodial remand cases in the snapshot). The discrepancy between the number of defendants where this ground for refusing bail applied and the number with the reason for substantiating bail refusal might be because the magistrates did not make their deliberations clear to the person completing the form or because the form was incorrectly completed. There are also obvious links between the ground record of accused under previous grants of bail (24%) and the reason previously absconded in any criminal proceedings (5% at Norwich magistrates court). The difference between these figures is logical since we would not expect the ground to be substantiated in every case. 13

Types of offence Figures 2.6 and 2.7 show the types of offences alleged to have been committed by those defendants in the bail snapshot who were remanded in custody. Both Manchester City and Norwich magistrates courts have offending profiles dominated by violence, theft, burglary, and (to a lesser extent) drugs offences. Manchester City magistrates court also has a significant number of other indictable cases: these include causing an affray, perverting the course of justice, and possessing firearms without a certificate. Figure 2.6: Custodial remand in Manchester types of offence Violence Sex Burglary Robbery Theft Fraud Criminal damage Drugs Other indictable Indictable motoring Summary non-motoring Summary motoring Breach of bail Breach of sentence 0 40 80 120 160 Number of defendents Source: Court records in the pilot scheme 1997 1998 Figure 2.7: Custodial remand in Norwich types of offence Violence Sex Burglary Robbery Theft Fraud Criminal damage Drugs Other indictable Indictable motoring Summary non-motoring Summary motoring Breach of bail Breach of sentence 0 10 20 30 40 Number of defendents 1997 1998 Source: Court records in the pilot scheme 14

Conditional bail Table 2.3 shows the numbers of defendants granted conditional bail and captured in the snapshot. There were about 2,800 such cases, the majority of which were from Manchester City magistrates court. Table 2.3: Defendants granted conditional bail captured in the bail snapshot 1997 1998 Total Manchester City magistrates court 1,104 1,080 2,184 Norwich magistrates court 294 275 569 Both magistrates courts 1,398 1,355 2,753 Conditions attached to bail Multiple conditions can be attached to bail. Figures 2.8 and 2.9 show conditions attached to bail at the two courts. A small number of electronically monitored bail curfews (ten in all) were caught in the 1998 sample. The most common condition reside in a particular place was attached to bail in almost three-quarters (73%) of cases at Manchester City magistrates court (78% in 1997, 69% in 1998) and two-thirds (66%) of cases at Norwich magistrates court (70% in 1997 and 62% in 1998). The second most common condition, other, covers a wide range of non-standard conditions such as do not enter [name] public house. The third most common condition, do not communicate/interfere with prosecution witnesses was used in more than a third (35%) of cases at Manchester City magistrates court and more than two-fifths (41%) of cases at Norwich magistrates court. Report to the police (about 28% at both courts) and reside between specified hours (13% in Manchester, 18% in Norwich) were also used frequently. Figure 2.8: Conditions attached to bail in Manchester provide a surety provide a security reside between specified hours report to the police do not communicate/interfere with prosecution witnesses others reside in a particular place 0% 10% 20% 30% 40% 50% 60% 70% 80% Percentage of defendents 1997 1998 both Source: Court records in the pilot scheme 15

Figure 2.9: Conditions attached to bail in Norwich curfew with electronic monitoring reside between specified hours report to the police do not communicate/interfere with prosecution witnesses others reside in a particular place Source: Court records in the pilot scheme 0% 10% 20% 30% 40% 50% 60% 70% 80% 1998 both There was a significant reduction in the number of curfews without electronic monitoring (that is, those with the conditions reside in a particular place and reside between specified hours ) granted by Norwich magistrates court. The proportions were 26 per cent (75 defendants) in 1997 and 11 per cent (31 defendants) in 1998. There was a small fall at Manchester City magistrates court in such cases: 15 per cent (169 defendants) in 1997 and 14 per cent (155 defendants) in 1998. 1997 Grounds for the remand decision Norwich magistrates court does not usually record grounds for conditional bail: nearly half of the Norwich cases in the snapshot had no grounds recorded, and as many showed no grounds given. Only a very small number had any grounds recorded and these have not been included. Figure 2.10 shows the grounds recorded for each spell of bail with conditions at Manchester City magistrates court, again using the shortened forms. As explained earlier, a defendant may have multiple grounds recorded. The two most common grounds at Manchester City magistrates court were, as for custodial remand, character/antecedents/associations/lack of community ties (41%) and nature and gravity of offence and probable sentence (36%). A ground that did not figure significantly for custodial remand, but did appear in a significant proportion of conditional bail cases was behaviour towards/proximity to prosecution witness (32%). 16

Figure 2.10: Grounds for granting conditional bail in Manchester unlikely to co-operate in obtaining reports no grounds given Source: Court records in the pilot scheme others previous record behaviour towards witness nature of offence character/antecedents Reasons for the remand decision 0% 10% 20% 30% 40% 50% 1998 both Figures 2.11 and 2.12 show the reasons recorded for granting conditional bail. Each defendant may warrant any number of reasons for refusing bail and the figures show percentages of defendants with each reason recorded. There are fewer categories used than for custodial remand and the shortened forms are not needed on these figures. The two most common reasons were again would fail to surrender (52% of conditional bail at Manchester City magistrates court, 47% at Norwich magistrates court), and would commit offence on bail (47% at both courts). The third most common reason would interfere with witnesses/obstruct the course of justice was used much more frequently for conditional bail than for remand (34% at Manchester City magistrates court, 37% at Norwich magistrates court). This is what we would expect given the more frequent use of the corresponding ground for conditional bail. Other reasons were used far less frequently. 1997 Figure 2.11: Reasons for bail with conditions in Manchester insufficient info for bail decision arrested, s7 Bail Act 1976 would interfere with witness would commit offence on bail would fail to surrender 0% 10% 20% 30% 40% 50% 60% 1997 1998 both Source: Court records in the pilot scheme 17

Figure 2.12: Reasons for granting conditional bail in Norwich insufficient info for bail decision impractical to obtain report would interfere with witness would fail to surrender would commit offence on bail 0% 10% 20% 30% 40% 50% 60% 1997 1998 both Source: Court records in the pilot scheme Types of offence Figures 2.13 and 2.14 show the types of offences alleged to have been committed by those defendants in the bail snapshot who were granted conditional bail. As for cases of custodial remand in the snapshot, both Manchester City and Norwich magistrates courts have offending profiles dominated by violence and theft. Unlike custody cases, there are fewer conditional bailees alleged to have committed burglary or drugs offences. There are larger numbers of summary cases at both courts than were seen for custody cases. Figure 2.13: Conditional bail in Manchester types of offence Violence Sex Burglary Robbery Theft Fraud Criminal damage Drugs Other indictable Indictable motoring Summary non-motoring Summary motoring Breach of bail Breach of sentence 0 100 200 300 400 500 Number of defendents 1997 1998 Source: Court records in the pilot scheme 18

Figure 2.14: Conditional bail in Norwich types of offence Violence Sex Burglary Robbery Theft Fraud Criminal damage Drugs Other indictable Indictable motoring Summary non-motoring Summary motoring Breach of bail Breach of sentence 0 50 100 150 200 Number of defendents 1997 1998 Source: Court records in the pilot scheme Unconditional bail Table 2.3 shows the total numbers of defendants granted unconditional bail and captured in the snapshot. There were about 1,500 defendants, and again the majority of them were from Manchester City magistrates court. Table 2.3: Defendants granted unconditional bail captured in the bail snapshot 1997 1998 Total Manchester City magistrates court 609 726 1,335 Norwich magistrates court 72 49 121 Both magistrates courts 681 775 1,456 Unconditional bail cases in the snapshot have no details of grounds, reasons or conditions. Types of offence Figures 2.15 and 2.16 show the types of offences alleged to have been committed by those defendants in the bail snapshot who were granted unconditional bail. The alleged offending profiles differ from those for custodial remand and conditional bail. The main difference in the Manchester profile is the reduction in the number of alleged serious offences violence, sex, burglary, and robbery. Theft still dominates the unconditional bail profile, but summary offences have sprung to far greater prominence. The profile for Norwich does not show the same clear differences, possibly because of the small number of defendants (fewer than 50 in 1998). The profile is similar to both the custodial remand and conditional bail profiles for Norwich, with the exception of the prominence of alleged drugs offences in the profile. 19

Figure 2.15: Unconditional bail in Manchester types of offence Violence Sex Burglary Robbery Theft Fraud Criminal damage Drugs Other indictable Indictable motoring Summary non-motoring Summary motoring Breach of bail Breach of sentence 0 50 100 150 200 250 300 350 Number of defendents Source: Court records in the pilot scheme 1997 1998 Figure 2.16: Unconditional bail in Norwich types of offence Violence Sex Burglary Robbery Theft Fraud Criminal damage Drugs Other indictable Indictable motoring Summary non-motoring Summary motoring Breach of bail Breach of sentence 0 10 20 30 Number of defendents Source: Court records in the pilot scheme Young defendants 1997 1998 The snapshot includes a small proportion of young defendants (11%). There are 153 15-year-olds, 183 16-year-olds and 238 17-year-olds and these are shown in Table 2.4 (data have been amalgamated for both years of the snapshot). In common with national figures (in Chapter 1), very few of these young defendants were remanded to care. Most were granted conditional bail, although many young defendants were granted unconditional bail or remanded in custody. There were too few cases to say whether the number of defendants given a care remand was statistically significant by either age or against the other types of bail. We did find, however, that the pattern of use of custodial remand, unconditional and conditional bail was statistically significant for this age group. Fewer young defendants were remanded in custody than expected when measured against all defendants, and more were granted conditional bail. This may reflect the increased number of community ties we expect in this age group. 20

Table 2.4: Young defendants by court (1) and type of remand Care remand Custodial remand Conditional bail Unconditional bail Age 15 Manchester 8 27 66 38 Norwich 2-12 - Age 16 Manchester 14 25 94 23 Norwich 1 1 23 2 Age 17 Manchester - 45 112 56 Norwich - 5 19 1 (1) Includes cases appearing in both magistrates and youth courts. Conclusions We found that nearly a fifth (18%) of the defendants in our snapshot were remanded in custody, more than a half (53%) were granted conditional bail, and more than a quarter (28%) were granted unconditional bail. There were many similarities between cases attracting custodial remand in Manchester and Norwich. The three most frequently used grounds for refusing bail were the same at both courts, although there were differences in ranking. The same was true for the reasons for refusing bail. Alleged violence, burglary and theft offences dominated the type of offence profile of both courts. There were also similarities between grants of conditional bail in Manchester and Norwich. The similarities were strong when the conditions attached to bail and the reasons for granting conditional bail were compared, but less strong when type of offence profiles were compared. Comparisons could not be made between grounds for refusing (unconditional) bail because Norwich magistrates court did not record these for conditional bail. There was a significant reduction in the number of curfews without electronic monitoring granted by Norwich magistrates court in the second of the snapshot years. This was not repeated at Manchester City magistrates court. The only data available for comparisons on unconditional bail cases was type of offence profile and these showed few similarities. Some differences were found between defendants refused bail and defendants granted conditional bail. Data on grounds for refusing bail for both types of defendant were only available from Manchester City magistrates court, but these showed behaviour towards witness was used more often for conditional bail than for custodial remand. This difference was also apparent in reasons for refusing unconditional bail in both courts would interfere with witness was used more often for conditional bail than for custodial remand. There were also differences in the type of offence profile: alleged burglary, robbery, and drugs offences took a more prominent role in custodial remand cases. 21