Evaluation of Community Payback Orders, Criminal Justice Social Work Reports and the Presumption Against Short Sentences

1 Introduction

1.1 Research has consistently shown that community penalties are more effective than short prison sentences in reducing offending, and that they can play an important role in addressing some of the individual and social harms associated with crime[1]. Nevertheless, in Scotland - as elsewhere - attempts to maximise the potential of such disposals have faced two significant, and inter-related, challenges. The first has been the practical challenge of developing a coherent, flexible and effective framework for the imposition and delivery of community penalties. The second has been to persuade members of the public and sentencers alike that such disposals represent an appropriate, meaningful and, again, effective response to the problems of lower level offenders and offending.

1.2 This report examines the implementation and early impacts of three related reforms intended to help address these issues. Two of those - Community Payback Orders (CPOs) and the Presumption Against Short Sentences (PASS) - were the result of legislative change (specifically, the Criminal Justice and Licensing (Scotland) Act 2010). The third was the introduction of a new template for reports provided to the courts by criminal justice social workers in support of the sentencing process. Previously known as Social Enquiry Reports (SERs), from February 2011, these were replaced by a new template - the Criminal Justice Social Work Report (CJSWR).

1.3 This evaluation was conducted by ScotCen Social Research, working in with two external consultants, Simon Noble of Noble Openshaw Limited and Professor Neil Hutton of Strathclyde University. It involved a wide range of data collection activities (detailed later in the report) and was carried out between December 2012 and June 2014.

Policy background

1.4 As part of the Scottish Government's strategic objective of a 'safer and stronger Scotland', there has been a concerted focus in recent years on the problems of crime and offending and a commitment to what has been described by the Cabinet Secretary for Justice as 'a coherent penal policy that uses prison for serious and dangerous offenders but deals with lower-risk offenders in the community' (Scottish Government, 2007, p.1).

1.5 Community penalties have actually been a longstanding feature of the criminal justice system in Scotland, and by the middle part of the last decade were being used by the courts more frequently than prison. This development was consistent with a body of research evidence suggesting that such disposals are more effective than short prison sentences in terms of discouraging reoffending. Nevertheless, the use of imprisonment remains higher in Scotland than in similar jurisdictions elsewhere, and a relatively high proportion of the prison population is accounted for by short-term prisoners, with whom there is limited scope to undertake effective rehabilitative work.

1.6 Against this backdrop, recent years have seen an extended process of policy review and development, marked by several significant staging posts. In November 2007, the Scottish Government published Reforming and Revitalising: Report of the Review of Community Penalties which argued that there is 'scope to make greater use of community penalties in appropriate cases' (2007, p.7) on the basis not only that such disposals are more effective than prison at reducing reoffending, but that they offer greater scope for 'payback' to the community, allow offenders to maintain important links to employment, family and community, and mean that resources within prisons can be concentrated on meaningful rehabilitative work with serious offenders. Nevertheless, the Review suggested that there were a number of issues relating not only to the effectiveness of the existing framework for community penalties in Scotland, but also the credibility and acceptability of such penalties in the eyes of the public and sentencers.

1.7 In terms of effectiveness, the Review argued that community penalties should be high quality, effective, immediate, visible, flexible and relevant. In particular, it was argued that, in order to deliver effective practice, the criminal justice system needed to 'focus its efforts on the core, high volume community penalties, ensuring that these work effectively across the country' (2007, p.9). It also noted that the existing range of community sentencing options was unnecessarily complex and that a more coherent packaging and 'branding' of such measures would improve understanding among sentencers and the public.

1.8 On that theme, research conducted specifically for the Review suggested that members of the public generally had a limited understanding of what community penalties might involve; saw them as appropriate only for low-level and non-violent offenders (and hence not as an alternative to prison); and viewed them as a 'soft option which is perceived to accomplish little in the way of punishing or deterring offenders' (TNS, 2007, p.11). That said, the research also pointed to scepticism about the effectiveness of prison and to greater potential support for community penalties, subject to such penalties retaining a punitive element, and yielding tangible (and visible) benefits to victims and communities.

1.9 In response to this, the Review concluded that a programme of positive change was needed in order to demonstrate that community penalties could be tough, effective (in terms of reducing reoffending) and involve 'meaningful payback' to the community. While acknowledging the importance of public engagement and education, the Review argued that the key to making community penalties more acceptable and credible would be effective delivery: 'The public and sentencers will have confidence in community penalties if they see them working effectively in their communities' (ibid, p.9).

1.10 The plan of action proposed by the Review was centred around four main themes - Reparation and Payback; Rehabilitation and Reintegration; Quality and Enforcement; and Community Engagement. These themes were subsequently developed in the course of two further policy reviews.

1.11 In June 2008, the Scottish Prisons Commission (SPC) published its report on Scotland's penal system, Scotland's Choice. This exercise arrived at the following overarching conclusion:

'The evidence that we have reviewed leads us to the conclusion that to use imprisonment wisely is to target it where it can be most effective - in punishing serious crime and protecting the public.

1. To better target imprisonment and make it more effective, the Commission recommends that imprisonment should be reserved for people whose offences are so serious that no other form of punishment will do and for those who pose a significant threat of serious harm to the public.
2. To move beyond our reliance on imprisonment as a means of punishing offenders, the Commission recommends that paying back in the community should become the default position in dealing with less serious offenders' (Scottish Prisons Commission, 2008, p.3)

1.12 The SPC Report also made 23 specific recommendations, including a number relating to the use of community sentences in place of short prison sentences. It suggested a single community sentence which would allow offenders to pay back to communities - through unpaid work, engaging in rehabilitative work or a combination of both. It also recommended that the Scottish Government legislate for a ban on short prison sentences of six months or less, except in particular circumstances.

1.13 It is worth noting that this report consciously defined 'payback' as being more than simply forcing offenders to contribute through unpaid work or financial compensation, but as encompassing work to rehabilitate offenders. As such, and as McNeill (2010, p.3) has noted, this signalled a clear attempt to move beyond the way in which the term has been used in England & Wales, where it has been specifically associated with the rebranding of community service as more visible and demanding.

'In essence, payback means finding constructive ways to compensate or repair harms caused by crime. It involves making good to the victim and/or the community. This might be through financial payment, unpaid work, engaging in rehabilitative work or some combination of these and other approaches. Ultimately, one of the best ways for offenders to pay back is by turning their lives around.'
(Scottish Prisons Commission, 2008, para. 3.28).

1.14 In December 2008, the Scottish Government published Protecting Scotland's Communities: Fast, Fair and Flexible Justice, setting out its response to the review of community penalties and the SPC's work on the penal system as a whole. The overarching framework for this was the commitment, first articulated in Reforming and Revitalising, to the delivery of justice that is 'immediate, visible, effective, high quality, flexible and relevant'. The report also contained the clearest expression to date of the overall thrust of Scottish Government policy in relation to community penalties:

'There are too many short prison sentences, which achieve nothing beyond warehousing the offender. We need a more effective and publicly acceptable system of community sentences. And the key for broader support for community sentences is to show that they are not only served in the community, but that they are for the community. That means they must be speedy - so victims see action being taken immediately after the court makes an order. And they need to be relevant, making communities safer, either through direct reinvestment by the offender, or by the offender seriously tackling his offending behaviour, or both. And they must be visible to communities.'
(Scottish Government, 2008, p.10. Emphasis in original.)

The reforms

1.15 The ideas contained and developed in and across these three policy documents eventually found legislative expression in the Criminal Justice and Licensing (Scotland) Act 2010. This was a wide-ranging and complex piece of legislation which was subject to extended scrutiny by the Scottish Parliament.[2] In relation to community penalties, it contained two central provisions.

Community Payback Orders

1.16 The first was the introduction of a single community disposal - the CPO - in place of Probation, Community Service and Supervised Attendance Orders. The intention behind this change was summarised in the Policy Memorandum which accompanied the original Bill as follows.

'In bringing together the options for judges, we are highlighting the scope for courts to punish offenders in a way in which also addresses the areas of their lives which need change. Setting out the options in this way also enables us to underline the fact that a community sentence is a punishment and not merely a supportive intervention.'

1.17 Specifically, the CPO was intended to serve the following main objectives:

• Achieve a positive impact on individuals.
• Require individuals to make payback to the community.
• Replace an unnecessarily complex range of community sentences and increase public understanding.
• Ensure the level of intervention matches the level of assessed risk.
• Create a robust and consistently delivered community sentence, which enjoys public confidence and credibility.

1.18 The Act gave courts the scope to impose a CPO with up to a maximum of nine requirements. These were:

• Unpaid Work or Other Activity Requirement (UPWOA)
• Supervision Requirement
• Conduct Requirement
• Programme Requirement
• Residence Requirement
• Drug Treatment Requirement (DTR)
• Alcohol Treatment Requirement (ATR)
• Mental Health Treatment Requirement (MHTR)
• Compensation Requirement

1.19 The CPO retained many of the features of the existing orders such as unpaid work (Community Service Order), supervision, mental health, drug and alcohol treatment and compensation (all Probation Order) and the concept of making reparation to the community (the former Community Reparation Order). But it also introduced some new elements, which echoed some of the core themes of Protecting Scotland's Communities. For example, it introduced an 'other activity' element as part of the Unpaid Work Requirement, with the intention of helping offenders to address their behaviour through educative and rehabilitative work alongside unpaid work - resonating with the notion of 'relevant' punishments. The Act introduced a new Level 1 order, allowing an order of up to 100 hours of unpaid work or other activity to be imposed without a report from social work which, along with shorter timescales for the commencement of orders, could be seen as improving the immediacy of community punishment. It also gave sentencers the ability to set review hearings during the course of the order and put in place more robust sanctions for breaching an order.

1.20 The introduction of CPOs was accompanied by revised practice guidance for social workers as part of the National Outcomes and Standards for Social Work Services in the Criminal Justice System (NOS).

The Presumption Against Short Sentences

1.21 The second significant element contained within the 2010 Act was the introduction of a presumption against prison sentences of three months or less. In its original form, this would have applied to sentences of six months or less but, following opposition from Labour and Conservative MSPs, an amendment during the passage of the Bill lowered the sentence length from six to three months.

1.22 It should be emphasised that PASS represented a presumption against, and not the abolition of, short prison sentences. Following the legislation, Sheriffs remain free to impose such a sanction where they feel it is justified and appropriate. They are, however, required to state their reasons for doing so and to have those formally recorded by the court.

Criminal Justice Social Work Reports

1.23 The introduction of CJSWRs was closely bound up with that of CPOs, both in terms of timing (both initiatives went live in early 2011) and the need to provide sentencers with improved information in order to make informed decisions about the appropriate use of community penalties. The overarching aim of CJSWRs was to combine traditional social work assessment skills with a structured approach to presenting sentencing options for dealing with offending behaviour and risk management and to steer report writers towards a more analytical style based on evidence and reasoned argument. A national template provided the means to deliver this more consistently across the country and simultaneously, the Level of Service/Case Management Inventory (LS/CMI) assessment tool (a full review of which was outside the scope of this evaluation) was intended to provide more robust and consistent additional assessment evidence.

1.24 In addition to planned training for all report writers, the introduction of CJSWRs was supported by new practice guidance[3], which, as well as explaining the format and content of the new template, emphasised the 'change in style of report writing, to a briefer, focused and more concise report'
(Scottish Government, 2011b, p.5).

The phasing in of the new penalties

1.25 Because CPOs can only be imposed for any offence(s) committed on or after 1 February 2011, or in default in payment of a fine imposed for an offence committed on or after 1 February 2011, CSOs, POs and SAOs did not disappear overnight. As Figure 1.1 indicates, in the first full year of operation (2011/12), approximately 10,000 CPOs were imposed, rising to more than 15,000 in the following year (2012/13). There was a corresponding reduction in the number of 'legacy' orders over the same period. Further detail about the number and type of CPOs imposed since introduction can be found in Chapter 3 (para 3.2).

Figure 1.1: Use of community penalties in Scotland, 2004-2013

The evaluation

Overall aims

1.26 The evaluation was loosely structured around a logic model showing the relationship between the three reforms and highlighting the 'theory of change' through which specific inputs and activities were expected to lead to particular short, medium and long-term outcomes. This is shown in Figure 1.2.

1.27 The overall aim of the evaluation was to examine whether the reforms have been implemented as intended (and outlined as 'activities' in the logic model); the extent to which the short and medium term outcomes have been realised; and the potential obstacles to realising those and longer-term goals.

Figure 1.2: The logic model for the reforms

1.28 Among the specific research questions flowing from this and highlighted in the original specification for the evaluation were the following:

• Whether and how the introduction of CJSWRs has led to overall quality and consistency; the inclusion of specific and bespoke sentencing recommendations; and to Sheriffs having confidence in the reports and using them to tailor sentences to individual need.
• Whether and how the use of the various CPO requirements has varied over time and across Scotland, and the possible reasons for - and potential impacts of - this.
• Perceived and actual levels of compliance with CPOs and the different kinds of requirements and the possible reasons for any variation.
• Potential barriers to the realisation of short- and medium-term outcomes, the source and impact of these, and how they might be addressed.
• The impact, if any, of PASS on sentencing practice.

1.29 It is worth noting, however, that some of the themes that were prominent in the original policy documents - e.g. in relation to the clarity, relevance, immediacy and visibility of community punishments - are less explicit within the logic model and in the specification for the evaluation. In practice, however, these too have formed part of the framework of expectations against which the reforms have been assessed and the question of the relationship between them and the original logic model is returned to in the conclusions.

Methods

1.30 In recognition of the complex and multi-faceted character of the reforms and the research aims, the evaluation involved a wide range of methods. At the heart of the study was a series of qualitative interviews with key actors - including criminal justice social workers and other service providers, Sheriffs and court staff, and offenders - in four case study areas. These were selected on the basis of geographic location and included two 'city' authorities, one 'other urban'/small town authority and one predominantly rural area; but the choice of areas also drew on quantitative data relating to the implementation of the reforms, and information gathered from orientation interviews with stakeholders about local criminal justice relationships and culture.

1.31 This was complemented by analysis of existing monitoring data and bespoke surveys of Sheriffs and criminal justice social work managers, allowing the evaluation to examine the implementation and operation of the reforms across Scotland as a whole, and to balance the more detailed picture emerging at a local level from the qualitative work.

1.32 More specifically, the main elements of the evaluation were as follows. (A fuller account of the evaluation methods can be found in Appendix A.)

1.33 Analysis of national monitoring data. CPO completions and terminations for the whole of Scotland were explored using unit level data from the Scottish Government. This included an investigation of variation by demographics and area.

1.34 National surveys of the Judiciary and criminal justice social work managers. All 32 Local Authorities were contacted to take part in the survey of Criminal Justice Social Work Managers and 30 took part. All 141 permanent and floating Sheriffs then in post in Scotland were also asked to fill out a short survey - just over half (n=72) completed and returned it. Results from both of these survey exercises are presented as raw numbers only, because of the relatively small total sample sizes. Some open text responses to particular questions are also drawn on in parts of the report. For reasons of confidentiality, these are not attributed to individuals or particular local authority areas.

1.35 Qualitative interviews in each of the four case study areas. A more detailed picture of the implementation and operation of the reforms was provided by qualitative interviews in four case study areas, conducted over two periods of fieldwork (in late 2013 and early summer 2014). Across the four areas as a whole, interviews were conducted with the following types of interviewee:

• Criminal Justice Social Work Managers (n=13)
• Criminal Justice Social Work and Unpaid Work Staff (n=25)
• Other practitioners (e.g. Addictions workers, Mental Health professionals, professionals from third sector Drug and Alcohol Services, Sheriff Clerks, Managers of charity shops where some individual unpaid work placements take place) (n=23)
• Sheriffs (n=21)
• Offenders (n=15)

1.36 It was agreed by the Research Advisory Group that protecting the anonymity of the research participants was of greater importance than a detailed exploration of geographical differences in the presentation of findings. Throughout the report, therefore, extracts from these interviews are referenced by interviewee type and by a sequential interviewee number - for instance, as CJSW interview 2 - but not by reference to case study area.

1.37 A participative audit of CJSWRs was also conducted within the four case study areas. This involved the small teams of CJSW staff from each area auditing a sample of CJSWRs from another area, using an agreed quality assessment instrument. This involved some factual information (e.g. about timings, or sources cited in the report) but also some more subjective assessments by auditors of the quality of the reports. In total, 144 reports were examined as part of this exercise. Results from this exercise are presented as both raw numbers and percentages.

1.38 Towards the end of the evaluation, a validation event was held, bringing together practitioners from a range of professional backgrounds and geographic locations. The principal aim of this was to provide an opportunity for sharing and 'sense-checking' emerging findings - and, in particular, to explore the extent to which experiences within the four case study areas appear to have been common to other areas. However, it also provided an opportunity for practitioners to suggest additional topics or questions for the final report and to contribute to a discussion about possible next steps. The discussions at this event have, therefore, helped to shape the final report and are reflected at various points in the text.

Limitations

1.39 A case study approach has important advantages when no single perspective can provide a full account or explanation of the effectiveness of an intervention or reform, and where understanding of implementation and operation needs to be holistic, comprehensive and contextualised. Nevertheless, it also has limitations - not least, that it carries a risk of unwarranted generalisation. While the case study areas were selected to provide a range of characteristics and experiences in relation to the implementation of CPOs and the related reforms, some of the features and experiences that were evident in those four areas were not necessarily shared across Scotland as a whole. The validation event allowed such divergence to be identified to some extent, although the scope of the project did not allow these to be explored in detail.

1.40 The long-term outcomes identified in the logic model were determined at the outset to be outside the scope of the evaluation for two main reasons. The first relates to timescales and the impossibility of determining long-term impacts within a couple of years of the implementation of an intervention. The second is methodological: put simply, identifying and attributing long-term changes in individual behaviour is extremely complex and would require a bespoke research project, capable of controlling for other potential influences. Such studies have been implemented in other jurisdictions and should (as we argue in the conclusion) potentially form part of the long-term evidence base in Scotland, though it should be noted that they are very costly to implement and often generate results that are equivocal.

1.41 There are particular challenges in trying to understand judicial decision-making. Like many such processes, this cannot be observed directly and the research was, therefore, dependent on retrospective accounts which may, of course, have contained elements of post-hoc rationalisation, faulty recall or socially acceptable responses. We had limited access to Sheriffs in order to conduct the research, as a result of other demands on their time, and it is also possible that those who chose to participate will - because of the focus of the research - have been more positively disposed to the use of community penalties in general.

1.42 Accessing the views and experiences of offenders is difficult. Such individuals are often reluctant to participate in research, viewing it as an extension of the contact they have with criminal justice social work and other agencies. They can also be difficult to contact and liaise with about practical arrangements because of the often complex nature of their day to day lives. As a result, there is a long-recognised problem in research involving offenders that those who are willing to engage with researchers are also likely to exhibit a greater degree of cooperation with the criminal justice system (Taxman and Belenko, 2012, p.215). The voices of the hard to reach and those with negative attitudes towards the criminal justice service, therefore, are often muted or missing. This evaluation is also subject to that limitation. Although significant efforts were made to ensure that the views of a wide range of offenders were included, this was within the context of limited resources and an already complex and wide-ranging study. To address this issue, views about the extent and other reasons for non-compliance were also sought from offenders who had complied with their orders and from CJSW staff and other professionals working closely with relevant groups. Nevertheless, it has been difficult to address the offender-related short- and medium-term outcomes as fully as some others.

The report

1.43 The report is broadly structured around the three reforms examined as part of the evaluation. It begins by looking at the introduction of CJSWRs (Chapter 2). It is worth noting, however, that what might be considered the 'core' reform - the introduction of CPOs - accounts for the bulk of the analysis (contained in Chapters 3 to 6). This reflects not only the importance of this element of the work, but also its sheer complexity. As the research progressed - and even at the validation event towards its conclusion - it was clear that there were many important issues at play and that the process of mapping and unravelling these would be time (and page) consuming.

1.44 By contrast, and despite the debate that surrounded its introduction, the Presumption Against Short Sentences (PASS) proved relatively straightforward to examine. Consequently, this has been embedded in a broader discussion about judicial decision-making in relation to community penalties in general and CPOs in particular. This can be found in Chapter 7. Conclusions are presented in Chapter 8.

1.45 Despite this apparently discrete discussion of the different reforms, it is worth noting the degree of overlap between them. The introduction of CPOs, for example, has been critically dependent on the availability of timely, high quality and accessible reports to sentencers, and on the willingness and ability of sentencers to use such disposals. These three broad domains have acted back upon each other in a range of other ways, too. The views of Sheriffs (and, indeed, the perceived views of Sheriffs) have had, for example, a potential influence on the style and content of CJSWRs; and both CJSWRs and judicial decision-making play a key role in determining the actual form and use of CPOs on the ground (see Figure 1.3).

1.46 As a result of these overlaps, we have had to make some relatively arbitrary decisions about where some specific issues (such as judicial views of the preferred sentencing options presented by social workers) are reported. These are, however, signposted and cross-referenced within the text.

Figure 1.3: Relationship between the three main elements of the reforms

1.47 It is worth noting that there are also two other important themes here which cut across the three main parts of the report and are illustrated in the above diagram. The first, of course, is the way that offenders are perceived to respond to CPOs (by sentencers, CJSW staff and others) and the way that they actually do. These topics are explored at various points in the report, but especially within Chapter 6 which examines the issues of engagement, compliance and breach - in part from the offender's perspective - and Chapter 7, which explores the views of sentencers. The second is the broad arena of public opinion, knowledge and expectation - again, perceived and actual - within which criminal justice operates. While a direct examination of the latter issues lies outside the scope of the evaluation, the views of the public cannot be wholly ignored, given the implicit and explicit references to notions of community. Indeed, the wider public is ever present in discussions of community penalties as the entity which has experienced harm and to which offenders are required to 'pay back'; the community is the setting within which such reparation is intended to occur; and the 'public acceptability' of particular disposals is a critical factor in the long-term viability of any sentencing framework, and particularly one that consciously departs from the 'populist punitiveness' (Bottoms, 1995) that has characterised much public debate in Scotland and elsewhere in the UK in recent decades. This issue is also touched on in the context of sentencers' views in Chapter 7, and returned to in the concluding chapter.

1.48 Across the three sets of reforms, but particularly within the context of the implementation and effectiveness of CPOs, the evaluation has had to grapple with a wide range of issues and sources of data - many of which could have been the subject of separate studies and reports in their own right. As a result, for reasons of space and scope, the analysis contained within the report is necessarily at a fairly high level and differences between case study areas are highlighted where especially illuminating rather than explored systematically in relation to each issue. Moreover, data from the various sources are drawn on judiciously to illustrate arguments rather than presented in full. Some additional depth is provided by the report's appendices, which include a more detailed description of the study methods and supporting data for some of the quantitative analyses in Chapters 3 and 6; any remaining questions should be addressed to the authors.