The Scottish Government

« Previous | Contents | Next »

Listen

Annex 4: Economic Evaluation of Selected Antisocial Behaviour Initiatives

Introduction

This annex describes the methodology and findings of the economic evaluation of 12 selected antisocial behaviour initiatives. The findings from the economic evaluation are summarised in paragraphs 7.24 - 7.30 of the main report. The analysis focused on quantifying the costs and benefits of 12 initiatives in the four selected Scottish local authorities. These included the community wardens schemes and mediation services operating in each of the four local authorities. In addition, the Early Intervention Families project in Edinburgh, the Safer Neighbourhoods Team in Fife, the Night Noise Team in North Lanarkshire and the Freephone Antisocial Behaviour Helpline in the Scottish Borders were also evaluated Where possible, the benefits are identified in terms of the possible cost consequences for various stakeholders. These included increases and decreases in current levels of expenditure and estimates of potential future expenditure that is prevented because of the intervention. Other less tangible benefits are identified where relevant. This analysis focuses primarily on the impacts on public costs ( i.e. payments for services funded from the public purse using money raised from taxation), though broad cost consequences for individuals and local neighbourhoods and communities are also identified were possible. This approach enables local decision makers to reach informed decisions about the costs and benefits of individual initiatives within their local context, whilst taking account of local policies and priorities.

Methodology

The economic evaluation analysed data covering the period from the beginning of each initiative (typically from late 2004/early 2005) until January 2007 (depending on local data collection and availability). Financial data on expenditure was not available for the full 2006/07 year, though some information was provided on expenditure to date within the year and on full year budgets.

A contact person was identified for each of the 12 initiatives included in the economic evaluation. The contact provided locally-generated information and reports on procedures, activity, outcomes and costs, where these were available. In addition, most of the contacts were interviewed to provide further information about the initiative. This approach was particularly useful where contact was maintained with the same individual throughout the evaluation period. However, in some cases initiatives experienced managerial changes and/or directly involved several agencies, which made maintaining contacts more difficult. The amount and types of data collected varied considerably between each initiative, even for those with broadly similar aims and objectives. This made it difficult to undertake meaningful comparisons between the local authorities of the costs of delivery of similar initiatives and of their value for money.

Community Wardens

Description of Services

A wide range of warden schemes operate in Scotland. The key functions of warden schemes fall into one or more of the following four broad categories:

• community development,
• crime prevention,
• environmental improvements, and
• housing management.

The four community warden schemes analysed in this evaluation had similar overall objectives: reducing crime and the fear of crime, reducing antisocial behaviour, and improving the amenity of local neighbourhoods and the quality of life of residents. However, each scheme worked in different ways that reflected locally-identified needs and other local service provision.

The City of Edinburgh Council was one of the first local authorities in Scotland to set up a team of community wardens (also known as Community Safety Wardens or Community Safety Concierges) within its Housing Department. The first team of six staff was set up in the Broomhouse estate (one of our case study neighbourhoods) in 2000. Since then, the Council has used Scottish Executive funding and its own resources to fund five teams, deploying a total of 45 wardens across specific areas of the city. These wardens wear distinctive maroon uniforms and work in small groups, patrolling their designated area seven days a week, both during the daytime and in 'out of hours' evenings. The wardens work closely with local residents to identify and act upon priority issues and to encourage community development. As well as being a highly visible Council presence and the first point of contact for residents, the wardens' day-to-day functions include on-the-spot environmental improvements ( e.g. keeping neighbourhoods clean and graffiti removal) and reporting required repairs and abandoned vehicles. The wardens work with the police and other agencies to help reduce crime, the fear of crime and all types of antisocial behaviour. The City of Edinburgh Council also employs Environmental Wardens (through its Environmental and Consumer Services Department) and Park Rangers (through its Culture & Leisure Services Department).

Community wardens have been introduced in five areas of Fife (including Abbeyview in Dunfermline, another of our case study neighbourhoods). Wardens were deployed in four areas in the late summer and autumn of 2004 and began operating in a fifth neighbourhood in April 2005. These localities were selected due to their high levels of antisocial behaviour and crime and poor local reputations. By the summer 2006 there were 22 full-time and two part-time community wardens in post in Fife.

Community wardens have operated in three areas of North Lanarkshire (Airdrie, Bellshill and Coatbridge) since November 2004. In addition, a mobile warden service was established in November 2005 and the community warden service was extended to include Motherwell and Wishaw in June 2006, by which time a total of 41 uniformed community wardens were employed. Their job description states that their role is:

• to proactively improve the quality of the local environment within the designated area and assist where appropriate in promoting community safety, and
• to carry out regular foot patrols of all residential areas, neighbourhood shopping centres and public places, acting as the eyes and ears of the local community, reporting damage defects, vandalism and anti-social acts.

They are also expected to respond to a variety of complaints about antisocial behaviour, including loud music, street football, and youth disorder.

The initial proposal for funding for community wardens made by Scottish Borders Council to the Scottish Executive was based upon a model where wardens would have a roaming presence within each of the Council's five Area Committee localities. This was not supported by the Scottish Executive which argued that this approach would not achieve the levels of community engagement required for an effective service. The proposals were consequently revised, with funding approved for community warden services in Galashiels and Kelso. This funding was used to employ five community wardens. In addition, a Warden/Antisocial behaviour Co-ordinator was recruited to oversee the development of the service and to assist with wider antisocial behaviour interventions. The community wardens work in teams of two. According to their job description and other local documents, their main role is to make the Scottish Borders a safer place to live by providing high visibility patrols that aim to improve the quality of life for residents by assisting in reducing antisocial behaviour, crime and fear of crime. In the course of conducting such patrols the community wardens are envisaged as acting as the 'eyes and ears' of local communities.

Methodological Challenges

The most significant methodological challenge encountered in this element of the economic evaluation was linking the activities undertaken by the wardens to specific outputs and outcomes. This was a problem for each of four warden services, and was exacerbated by the fact that each service had different aims and objectives and recorded its activity in different ways. Each local authority identified varying sets of statistical indicators to compare with baseline indicators and it was often unclear if an increase in reported levels of crime and antisocial behaviour was due to increases in the number of actual incidents, or because residents were more aware of antisocial behaviour activities and/or more willing to report them. It was not possible to identify appropriate comparator areas without wardens in order to identify the specific impact of the presence and activities of community wardens on levels of antisocial behaviour. The nature of the tasks undertaken by the wardens themselves also varied across the four local authorities. In addition, wardens do a great deal of community development work which is very hard to quantify. It is also difficult to place a financial value on such work. Furthermore, because each community warden service responded to locally identified requirements and reported its outcomes in different ways, it was not possible to identify comparable unit costs for similar activities.

Activities and Outcomes

The local authorities were required to submit to the Scottish Executive 12- month and 24- month reports on the activities and outcomes of their community warden schemes. Although these reports were expected to follow a common format, including collecting statistical indicators and perception/survey-based indicators for comparison with baseline indicators, the actual indicators used were selected locally. These returns generally included quantitative information about the numbers of incidents reported by the wardens, along with descriptions of other activities undertaken within their designated neighbourhoods. However, the amount of information included in the returns varied considerably, as did the amount and types of other relevant data provided directly for this research.

Key Performance Indicator ( KPI) data was provided by the City of Edinburgh Council for KPI 2 (Crime/ ASB) and KPI 3 (Vehicles). However, although the 24-month report identified the locally-selected statistical and perception/survey-based indicators, no specific numerical details were included due to local problems with recording this data.

In Fife the statistical indicators selected for inclusion in the Scottish Executive reports focused on reducing violent crime, antisocial behaviour, vehicle crime, domestic housebreaking, vandalism, and void properties. Fife Council's 24-month report showed considerable reductions compared with baseline indicators for most measures, although not all of the improvements can be attributed solely to the wardens, given all of the other local interventions in place to reduce antisocial behaviour and crime. Nevertheless, the available data indicates that community wardens in Fife are having a positive impact in the neighbourhoods where they are operating.

North Lanarkshire Council provided detailed information on the activities undertaken by their community wardens as reflected by the incident reports they made to other services. This data indicated that almost half of the reports made by wardens related to cleansing streets and other public areas. The statistical indicators included for Years 1 and 2 in North Lanarkshire Council's 24-month report, which focus on reducing levels of crime, are presented in Table A4 1. The table also provides a breakdown for the three areas where warden services have been operating since the end of 2004 and shows the percentage changes in recorded criminal activity over this period.

Table A4.1: North Lanarkshire's statistical indicators for years 1 and 2

* A: Airdrie; B: Bellshill; C: Coatbridge.
** The percentages for Year 2 reflect the change from Year 1, not from the baseline.

The data shows varied success in achieving reductions in offences, with some increases as well as decreases. Furthermore, there does not seem to be any consistency in trends across the three areas. For example, there was a large reduction in offences related to motor vehicles in Airdrie but a significant increase in these types of offences in Coatbridge. In many situations it is also unclear if an increase in the number of recorded offences is a manifestation of growing numbers of incidents or, alternatively, an enhanced propensity amongst residents to report incidents. It is also not possible to compare these trends with what would have happened in the three areas had wardens not been operating: therefore this data needs to be interpreted with caution. However, the data suggests that community wardens have had a particularly strong positive effect in Airdrie, where there have been reductions in recorded levels of almost all types of offences during the study period.

Scottish Borders Council's 24-month community wardens report included data on two selected statistical indicators.

• Reduction in noise-related antisocial behaviour
  • Number of complainers contacted and advice given. The target was to increase contact with complainers by 40 per cent from a baseline of 15 contacts. During Year 1 there were 26 contacts ( i.e. an increase of 73%).
  • Number of directed patrol strategies implemented - the target was to implement 12 (from a baseline of 0), but during Year 1 the number implemented was 132, due to the success of a database planning tool which increased the efficiency of directing and targeting patrols.
• Reduction in environmental damage
  • Number of recorded incidences of vandalism. The average year-on-year increase over the previous three years had been seven per cent. The target was to restrict this to a 4% (four per cent?) increase. However, comparisons of January - March 2005 with January - March 2006 showed a reduction of 51% in Kelso and 17% in Galashiels.

Scottish Borders Council also undertook some analysis of local data provided by Lothian and Borders Police to identify changes in antisocial behaviour incidence rates in areas with and without wardens. These are reported in Tables A4.2 and A4 3.

Table A4.2: Changes in numbers of antisocial behaviour incidents for areas in Scottish Borders with and without wardens: 2003/04 - 2005/06

Source: Lothian and Borders Police.

Table A4.3: Percentage changes in antisocial behaviour incidents for areas in Scottish Borders with and without wardens: 2003/04 - 2005/06

Source: Lothian and Borders Police.

These figures indicate that community wardens have had a significant positive impact. For example, whilst incidents of noise nuisance and environmental antisocial behaviour increased by 184% in areas without wardens, they decreased by almost a third (31%) in the areas where wardens were operating. Similarly, reported vandalism declined marginally in the areas with wardens, in contrast to a 15% rise in areas without wardens. However, neighbour disputes increased in areas with wardens (although not to the same extent as areas without wardens) and there was a 53% rise in youth-related calls in the areas where wardens were operating, which was significantly greater than the increase in youth-related calls in areas without wardens. The generally positive comparative impact of the presence of community wardens on rates of antisocial behaviour is particularly noteworthy given that the areas without wardens were considered by the Scottish Borders Council and the police to be less problematic in terms of antisocial behaviour. It is also the case that the wardens may well have increased the propensity of local residents to report incidents. This data therefore suggests that, after 24 months of operation, the community wardens scheme is having a significant positive impact on reducing antisocial behaviour in Galashiels and Kelso. However, it was acknowledged by local practitioners that the community wardens are not the only explanatory variable; for example, levels of youth provision differ throughout the Scottish Borders area.

The information provided by the four schemes for the perceptions/survey-based indicators in the 24-month reports is less robust. This is partly because some follow-up surveys were still to be undertaken. However, where surveys had been conducted, response rates were often very low, thus compromising the reliability of the data. Furthermore, asking people about their perceptions of change is notoriously problematic, as their responses can be influenced by a myriad of subjective factors. Scottish Borders Council undertook a household Community Safety Survey in November 2005, 14 months after the introduction of the community warden service. The survey results showed that there was good awareness of the community wardens service amongst local residents and that the wardens had a high profile within the communities in which they operated, particularly in Kelso. There was a greater positive perception of the difference the wardens have made to communities in Galashiels than in Kelso. Anecdotal evidence from some Scottish Border Council's Elected Members and local residents suggests there is widespread support for the community warden service in the two towns in which it operates.

Further anecdotal evidence from the warden services in the other areas also indicated that these services were valued and popular. For example, in Fife feedback on the community warden schemes from local Community Councils, Tenants and Residents Associations and individual local residents had been positive. The service has been described as 'invaluable' and many local residents reported that their community is now safer, better and cleaner, with one stating that "the Council should have done this years ago."

Costs

It proved difficult to get specific information on budgets and expenditure for all of the community wardens schemes as some of the schemes did not appear to record regular financial information for monitoring or other purposes. However, the initial allocations made by the Scottish Executive for developing the schemes in the four selected local authorities in 2004/05 were available (Table A4.4). It should be noted that additional funds may also have been allocated to the schemes from other local sources and budgets. For example in Edinburgh, additional funds were drawn from the Housing Revenue Account, Quality of Life funding, and Community Safety Partnership resources.

Table A4.4: Allocations by Scottish Executive for community wardens in 2004/05

Source: Scottish Executive (2003): Building Strong, Safe and Attractive Communities: Guidance for Submissions.

Where local budgetary and expenditure information was provided, it was clear that the vast majority of expenditure comprises salary costs. For example, accounts from Fife Council show that almost 95% of the expenditure on their warden service is accounted for by employee-related payroll costs. The remaining five per cent covers premises-related expenditure, transport-related costs, supplies and services, and support service charges. Total expenditure on warden services in the Scottish Borders was estimated to be about £135,000 in 2006/07, with almost 85% of this being accounted for by employee-related costs (including insurance, training and recruitment). Each team has also been provided with a van. Most of the administrative expenditure in the Scottish Borders is on computer equipment and telephones. This data suggests that warden scheme managers have very little opportunity for discretionary spending, which may in part explain why some scheme managers did not seem to receive financial information on a regular basis.

Wardens and Senior Wardens are paid on Scottish Local Authority salary scales for general grade staff and administrative and professional staff. Although their possible scales and grades in Edinburgh run from GS2 - AP1, all of the concierges were on GS2 (£13,893 - 14,577) in 2006/07. The community wardens in Fife were paid on AP1 or AP2 (£15,441 - £17,871) and the Senior Wardens were paid on AP3 (£18,840 - £20,673) during 2006/07. They also received an additional seven and a half per cent for working unsocial hours. The pay scales in 2006/07 for the wardens in Scottish Borders were GS1 with spinal column point (scp) 7 start point ( i.e. £12,438 - £13,893) and AP1 with scp15 start point ( i.e. £15,441 - £16,518).

Although it is possible to calculate unit costs - such as the cost per report made by the community wardens - for most of the schemes, these have not been reported here. While such figures may be useful locally for making comparisons of such costs over a period of time, they should not be used for making comparisons across the services due to their different objectives and data recording mechanisms. Furthermore, focusing only on reports of incidents excludes the vital community development work undertaken by wardens.

Cost Consequences

There is some anecdotal evidence of the cost-effectiveness of community warden schemes, or at least of some elements of them. For example, in Edinburgh an initial assessment of the Broomhouse wardens pilot, which started in 2001, showed increased resident confidence in the area, a reduction in residents' concerns about "young people hanging around", and a saving of £50,000 in costs relating to vandalism and illegal litter dumping over a six-month period. The local, unpublished evaluation of the first year of the project showed an eight per cent increase in resident satisfaction and recognition that the estate was cleaner and felt safer than before the introduction of the wardens. Savings on the cost of vandalism and bulk rubbish removal were estimated at £107k for the period.

Some of the 24-month reports to the Scottish Executive included specific examples of where community wardens had made a significant difference. One example was the improvement in the Magdalene area in Edinburgh. The new concierge satellite office there had resulted in a significant reduction in youth nuisance and crime in an area which, 12 months previously, was one of the city neighbourhoods worst affected by antisocial behaviour. The Magdalene Balconies are now a popular place for families to live, in contrast to their previous reputation as a 'no go' area. Much of this improvement was attributed locally to the hands-on commitment and dedication of the Community Safety Concierge. These improvements will have had considerable indirect cost consequences for local residents, the City of Edinburgh Council and the police. For example, the Council is likely to have higher rental income due to a reduction in void properties and will also have to pay less money on boarding-up and monitoring empty properties and responding to the vandalism of such properties.

Future Potential Methodologies

One approach to determining the cost-effectiveness of community wardens is to consider their financial impact on other services. For example, they may generate additional work for other services, for example by reporting environmental problems that would otherwise have been ignored. However, they may also lead to a quicker response from other services to some problems, thus preventing an escalation, for example by reporting damage to unoccupied property or new graffiti. Some work, such as litter picking, may also be diverted to the wardens from other services units. The economic analysis undertaken in England for the evaluation of the Neighbourhood Wardens Scheme by the Neighbourhood Renewal Unit (Office of the Deputy Prime Minister, 2004) concluded that the effects of warden activity on the costs and savings of other services could be assumed to balance out.

Another approach is to try to assign values to the benefits of the warden programmes, which may include preventing costs that would otherwise have been incurred. The benefits of the warden programmes fall into two categories - quantifiable and non-quantifiable. Non-quantifiable benefits are associated with subjective outcomes, such as improved quality of life, reduced fear of crime, community development, and improved satisfaction with the neighbourhood as a place to live. These benefits, which are enjoyed by local residents, are important and highly relevant, but monetary values cannot be assigned to them.

The main way in which benefits may be quantified is through assigning monetary values to local reductions in crime. Some of the published estimates are presented in Tables A4.5 and A4.6.

Table A4.5: Costs associated with generic antisocial behaviour

Source: Whitehead et al. (2003).

Table A4.6 Costs associated with specific types of antisocial behaviour

Source: Whitehead et al. (2003)

Other possibilities include identifying prevented costs associated with void properties due to lost rent (of, say, £70 per week) and costs of boarding-up empty properties. If voids are reduced, along with the numbers of tenants requesting transfers to properties in other, less problematic areas, then local authority housing departments and housing associations will benefit from higher rental income and less transfer-related work.

Another, more aggregated approach, is used in the Neighbourhood Renewal Unit's evaluation of wardens ( NRU, 2004). The researchers drew on data provided by Home Office economists (Brand and Price, 2000) which suggested that the benefits of preventing the 'average' offence is around £2,000 (at 2000 prices). This value includes the Exchequer consequences ( e.g. as borne by the Criminal Justice System, the NHS and local authorities) as well as those borne by the individual (in terms of the physical and emotional impact) and by society (in terms of lost output). The researchers then calculated the net present value ( i.e. the present value of the benefits minus the present value of the costs over the time period for the evaluation) to determine the estimated net benefits attributable to the impact of the Neighbourhood Wardens programme on crime. They recognised that it would be unrealistic to attribute all of these benefits to the warden schemes, but nevertheless were able to conclude that:

Even if no more than 10% of the crime impacts in the Neighbourhood Wardens areas are attributable to the Neighbourhood Wardens schemes, then the estimated total costs of the Neighbourhood Wardens Programmes are exceeded by the estimated benefits. Taking account of the important but uncosted additional benefits in terms of wardens' impacts on residents' perceptions, for example, about the improved quality of life in scheme areas further strengthens the conclusions that the Neighbourhood Wardens schemes represent value for money.

Dubourg and Hamed (2005) updated the work of Brand and Price on the economic and social costs of crimes against individuals and households in England and Wales. Recent research on the cost of exclusion (Prince's Trust, 2007) uses this data to estimate that the average cost per crime committed by young people aged 10-21 years (weighted by the prevalence of the types of crime among young people) was almost £4,600 in 2004.

Local authorities wishing to get a clearer picture of the economic consequences of their warden services need to consider their local expenditure on this service and determine the net numbers of criminal and antisocial behaviour incidents that the wardens have prevented locally (or, alternatively, compare local trends in similar areas with and without wardens). They can then use the cost data presented in this section to estimate the broad cost consequences of their warden service, either using a representative average cost per incident prevented or estimated costs for specific activities. They will need to determine the extent to which they believe the wardens (rather than other locally-operating initiatives) have been responsible for these cost savings. They will also need to decide if they want to focus purely on the consequences for public costs or also to include the costs borne by individuals experiencing criminal activities.

With regard to the economic impact of the community warden schemes in Scotland, the evidence suggests that, overall, these are having a positive impact on reducing criminal activity, even though some of the activity data is hard to interpret. Given the relatively modest costs associated with the schemes and the high costs associated with many of the criminal and antisocial behaviour incidents that are likely to have prevented by the community wardens, these schemes are likely to reduce public costs by more than the costs of providing the service. Furthermore, wardens tend to be popular in the areas where they work and they have positive, but intangible, impacts such as helping to increase perceptions of safety, improving the physical amenity of neighbourhoods and encouraging greater community involvement, which in turn improve the quality of life of local residents.

Mediation Services

Description of Services

Mediation is a process whereby an impartial third party helps two (or sometimes more) disputing parties resolve their disagreement, with the disputing parties - rather than the mediator - deciding the terms of the agreement. In the selected local authorities, two of the mediation services were provided in-house and two were provided through Service Level Agreements ( SLAs) with SACRO (a voluntary organisation specialising in conflict resolution and providing support and rehabilitation to offenders).

In each local authority, the mediation services were provided by a small team of employed staff assisted by trained volunteers, as shown in Table A4.7. Mediation Staff work in pairs with the disputing parties to try to achieve a resolution to a dispute.

Table A4.7: Staffing of mediation services (December 2006)

Previous Research

Previous major research reports have evaluated the comparative costs of mediation services in the UK. Dignan et al. (1996) compared the cost-effectiveness of mediation and alternative ( e.g. legal action or re-housing) approaches to resolving neighbour disputes in England and Wales. They focussed on identifying and, where possible, quantifying the full range of costs that might be incurred by agencies whose responsibilities encompass handling neighbour disputes. They adopted a mixed-method approach which allowed them to assess relative costs of mediation in neighbour disputes. The estimated average cost per mediation case was £252, though this figure was based on timesheet data from only one mediation service. This study was extended to Scotland in 1999 (Dignan and Soreby, 1999), although this work extrapolated costs from the previous English study rather than calculated average costs per case of mediation in Scotland. The second study compared average costs per case and concluded that community mediation was more expensive than informal intervention, but cheaper than the formal intervention that may be required when disputes persist or escalate. The reports stressed that mediation was not suitable for all cases of neighbour dispute, but where it was used appropriately, there were significant cost savings for a number of agencies.

Brown et al. (2003) considered the role of mediation in tackling neighbour disputes and antisocial behaviour using Scottish data, focussing on the effectiveness and costs of mediation compared with legal remedies, such as actions for repossession and Antisocial Behaviour Orders ( ASBOs), for resolving neighbour disputes. The researchers analysed a total of 100 referred cases from two community mediation services and two local authority mediation services and 50 legal proceedings cases (eviction, antisocial behaviour order and interdict) from local authorities. From the 100 mediation cases studies, the researchers found that the average cost of handling a referral to mediation was £121. This rose to an average of £204 when face-to-face or shuttle mediation was involved. The maximum cost was £484 (and was incurred because the antisocial behaviour had escalated to serious proportions). The costs for the local authority mediation services were, on average, slightly lower than those for community mediation (£105 compared with £136), reflecting the higher proportion of cases managed by local authority mediation services where no contact was made with the parties to the dispute.

The costs of the 50 legal cases included in the study ranged from £339 to £13,692 for a very complex eviction case, with an average cost of £3,546. Average costs of ASBOs and repossession actions were approximately £2,250 (range: £500 - £6,500) and £9,000 (range: £6,500 - £14,000), respectively. These figures were broadly comparable with those found in other relevant studies.

A small study based on a sample of seven possession actions and two interdicts (Atkinson et al., 2000) found that the average cost of an action for possession (at 1999 prices) was £3,200, and £1,175 for an interdict. A complex, defended repossession case could cost up to £7,500, although undefended cases cost in the region of £1,500 - £2,000. A survey of local authorities in England (Campbell, 2002) found the average cost associated with an ASBO was £5,350 (range: under £400 - £18,000). These findings led the authors to conclude that

Although mediation will not be sufficient to deal with serious antisocial behaviour, which is associated with alcohol and drug abuse, mental health problems or criminal activity, its cost-effectiveness suggests that there is considerable scope to extend mediation in the area of neighbour disputes.

However, it is crucial to remember that the costs calculated by Brown et al. (2003) are net costs. The authors suggest that total costs including overheads would be likely to be at least 40 - 50 per cent higher than the net costs that they present. Another study of mediation services (Mulcahy and Summerfield, 2001) estimated the average cost per case to be £410, of which £226 was due to specific case tasks, £75 for general mediator tasks such as monitoring and liaison, and £109 for organisational overheads.

Methodological Challenges

Several methodological challenges arose when considering the cost-effectiveness of mediation services during this research. These were:

• different ways of recording activity,
• different ways of defining outcomes ( e.g. success),
• the lack of clear links between activity and outcomes,
• the need to calculate a value for the input provided by volunteers,
• the lack of specific budgets for in-house services,
• the potential omission of organisational overheads in the cost information for in-house services,
• measuring other elements ( e.g. public relations and awareness raising; community development; work with schools) provided by the service, and
• lack of data over the full 2006/07 financial year.

Referrals and Activity Data

All four mediation services received referrals from a range of sources, including local antisocial behaviour teams and units, local authority housing departments, registered social landlords and the police, as well as self-referrals by individuals. The proportions of referrals from each of these sources varied considerably between the four services. For example, no referrals had been received from the police in North Lanarkshire, though they were responsible for more than 10 per cent of referrals in Fife and the Scottish Borders. Individuals involved in referred disputes came from all housing tenures, including owner occupiers. Referrals were also made for a number of different reasons, though many were due to noise of one form or another ( e.g. barking dogs, loud music, family arguments, children). Boundary and property disputes were also common. Many disputes were caused by several factors, and some had lasted for several years. In addition, some conflicts were the result of serious mental health problems or due to addictions; these were often too complex to be resolved through mediation alone.

Although there is no overall consistency in recording activity, the work of each mediation service can generally be sub-divided into two broad elements: assessments and cases. 'Assessments' or 'enquiries' relate to those referrals receiving advice, guidance and assistance (often described as ' AGA'), whereas 'cases' involve parties meeting (or using shuttle mediation) in order to try to identify a mutually acceptable solution to their dispute. Generally only a relatively small proportion of referrals become cases. Table A4.8 summarises the data on referrals provided by the four mediation services and shows their split across assessments and cases (where known).

Table A4.8: Referrals received by the mediation services and their split between assessments and cases

Source: Mediation services in each local authority area.

Outcomes

Two of the biggest problems encountered when considering mediation services are the lack of a consistent link between activity and outcomes, and different views about what constitutes a 'successful' outcome. For example, for some referrals a successful resolution can be achieved simply through providing advice and guidance to one disputant by telephone. In other situations a considerable amount of time may be spent arranging meetings or providing shuttle mediation without a successful resolution being agreed by the disputants. Furthermore, referred clients may define 'success' differently. Some parties may feel that a successful outcome has been achieved if they agree to ignore each another, even though the underlying problem remains. However, in some situations a successful outcome (from the perspective of both parties) may arise only if one or other party moves to a new property. Conversely, even if mediation fails to result in what external parties may deem to be a 'successful' outcome, one or both disputants may have learned a great deal about conflict resolution which may result in fewer future disputes. However, even where seemingly mutually acceptable agreements are reached, there are no figures available about the proportion of these cases where problems re-emerge at a later date.

To illustrate some of the above points, of the 116 referrals in Edinburgh in 2005/06 that proceeded to mediation ( i.e. became cases), full agreement was reached in slightly over two-fifths (42.3%) of these, with improvement in a further 15.5%. However, in over a quarter (27.6%) of these cases one or more parties withdrew from the mediation process and one in ten cases were closed 'due to irreconcilable differences'. Of the 242 referrals classified as enquiries during the same period, the problem was resolved without intervention in just over one in ten of these (11.2%). Comparable data for Fife shows that almost four-fifths (79.6%) of cases reached full agreement.

Costs of Mediation Services

When calculating the costs of providing the four mediation services, three specific methodological issues were encountered.

Identifying relevant costs

Because the two SACRO services were provided according to a SLA, they each had clearly-defined budgets. However, the budgets for the two in-house local authority mediation services were incorporated into the budgets for their host antisocial behaviour team/unit, which meant that the costs of the mediation service had to be built up from the costs of the staff salaries and a proportion of the other costs associated with the team/unit. However, despite having clearly identifiable budgets, calculating costs for the two SACRO services was complicated by the SLAs included other elements (for example work with local schools). Therefore some estimates had to be made to determine the share of the costs attributable to the community mediation service. These included costs associated with specific promotional work undertaken by the Service Manager and the Mediation Officers for the four services, as these are seen as integral parts of the mediation schemes. Further adjustments were subsequently made to the costs of the two SACRO services to reflect the specific requirements to undertake local awareness-raising and capacity building activities as part of their SLAs.

Organisational overheads

Service evaluations which adopt a bottom-up approach to costing and focus only on the amount of time staff spend on various activities fail to take into account the full costs associated with providing a service. In reality, aspects such as staff (and volunteer) training and supervision and time spent on administrative tasks are important elements of the service. Using a top-down approach, where the full costs of providing a service are identified for use when calculating unit costs, provides a much more accurate estimate of the costs of all the resources required to deliver the service. These full costs should also include organisational overheads such as the value of office space and of centrally-provided services ( e.g. central administration, senior management involvement, payroll activities and recruitment). The SLAs include monies for these aspects, but they had to be estimated for the in-house local authority services.

Volunteer time

To ensure that a value is placed on all of the resources used to deliver mediation services, it is necessary to include some estimate for the value of the time provided by the volunteers. This is partly to reflect the fact that their time has an 'opportunity cost' to them, as they could be using it for other uses. In addition, it serves to give a truer picture of what it would cost to provide the service if staff had to be employed to undertake the work performed by the volunteers.

Different approaches were used for estimating the contributions of volunteers to the four mediation services. The volunteers in Edinburgh were felt to be more experienced than those in Fife, so values of £15 per hour and £7.50 per hour, respectively, were assigned to their inputs (£15 per hour equates to a salary of about £24,000 over a full year, which is close to the national average, whereas £7.50 per hour is slightly above the minimum wage). The managers of the in-house local authority services provided estimates of the number of additional Mediation Officers they would need to employ in the absence of their volunteers. The value of the volunteer time for these services has been based on the annual costs of employing a Mediation Officer. These values are inevitably arbitrary, but they do ensure that some recognition and monetary value is given to the work of the volunteers.

The annual costs for the mediation services in Edinburgh and Fife can be determined from the value of the SLA, adjusted to account for the factors outlined above. The information in the expenditure accounts for 2004/05 and 2005/06 showed the proportions of expenditure on salaries, office costs and central administrative overheads. These proportions were then applied to the estimated salary costs for the two in-house services to provide a range of plausible estimates of the value of the resources (excluding and including volunteer time) used to deliver these services. The resulting annual costs are shown in Table A4.9.

Table A4.9: Summary of annual costs of mediation services

Source: Mediation services in each local authority area.

Unit costs

The above cost data was combined with the estimated annual activity data to provide some unit costs. Table A4.10 shows the estimated average costs per referral for each service. Those for Edinburgh and Fife refer to 2005/06, whilst those for the values for the in-house local authority services in North Lanarkshire and the Scottish Borders draw upon data for 2005/06 and 2006/07.

Table A4.10: Estimated average costs per referral received

Source: Mediation services in each local authority area.

It was also possible to estimate more specific unit costs for three of the services. The two SACRO services classify all referrals according to the service provided. Staff working for each of these services were asked to allocate local weights to these classifications to reflect the average amounts of time and resources required for each activity. This information was then applied to the overall costs and the numbers of referrals falling into each category to enable estimates of the average cost of each activity to be calculated. These are shown in Table A4.11.

Table A4.11: Estimated average unit costs for various SACRO activities

Source: Mediation services in Edinburgh and Fife.

Although referrals in North Lanarkshire are only sub-divided into assessments and cases, the calculations were more complex. This was partly due to fact that a range of cost estimates had been calculated (see Table A4.9) due to the lack of specific local budgets for expenditure. The split between assessments and cases also varied over the two years under consideration (see Table A4.8), which meant that calculations were undertaken for two possible scenarios. Two estimates were made to reflect the relative amounts of time required for assessment and cases. The resulting unit cost estimates are shown in Table A4.12.

Table A4.12: Estimated average unit costs for assessments and cases in North Lanarkshire

Source: Mediation service in North Lanarkshire

No such calculations were undertaken for the Scottish Borders as mediation services data was not sub-divided to reflect the numbers of assessments and cases.

The average cost per referral varied from about £450 to £820 (including the value of volunteer time). The information provided by SACRO showed that the average cost of an assessment (including the value of volunteer time) was about £40 for the provision of basic information, £80-£90 if advice and support were provided by telephone, and around £600 for a home visit. The cost of an assessment provided by North Lanarkshire's in-house local authority services was estimated as £35-£145. For the three services where the average cost per case could be calculated, this was generally in the order of £1,200-£1,500. However, it was not possible to estimate the cost per successful outcome due to the different interpretations of how such an outcome should be defined.

Potential cost consequences

Although other studies have compared the costs of mediation services with those of legal remedies, it seems unlikely that many of the referrals to the mediation services included in this evaluation would have pursued legal solutions in the absence of a local mediation service. Thus mediation services are not expected to result in preventing significant costs for legal services. It is more likely that those in dispute would have endured the problem or possibly adopted evasive action by moving (or requesting to move) elsewhere. These costs would tend to fall on the individuals and/or on the local authority housing department.

Neighbour disputes are linked with deteriorating health, which may not return to its previous status after a dispute has been settled. A survey was undertaken in Fife in 1999 of 262 clients of Fife Community Mediation Service (Marshall, 1999). It achieved a 37% response rate, 74% of whom had suffered adverse effects on their health during the dispute. Sleeplessness affected about two-thirds of respondents and almost half suffered from depression. Two-fifths of respondents reported visiting their doctor as a consequence of the neighbour dispute and almost half reported taking medication because of it. Just over half of those whose health was affected reported an improvement in their health following their involvement with community mediation, though many people felt that their health did not return to its pre-dispute level. This suggests that there may be considerable benefits for both the individuals concerned and also (albeit to a lesser extent) for the NHS if neighbour disputes can be resolved before they have a strong negative impact on people's health status. There may also be effects on people's workplace performance, or even on their ability to work, which again has consequences for both the person themselves and for wider society (in terms of lost output and benefits paid).

Most of the benefits of preventing neighbour disputes from escalating (and, indeed, of preventing them from occurring by teaching people how to manage neighbour relations) will be felt by individuals rather than by publicly-funded services. However mediation services can have a considerable impact on local residents and communities, especially if they are targeted appropriately.

Edinburgh Early Intervention Families Project

Description of Service

The Case Management Early Intervention Project ² (aka the Early Intervention Families Project) was established in Edinburgh in August 2005 with initial funding for two years. It works with vulnerable families in specific parts of the city (including the two case study Edinburgh neighbourhoods) where there is at least one child aged 12 years or under and the behaviour of one or more children is giving cause for concern. Each family is allocated a Case Manager who works holistically with them and with all of the other agencies in contact with the family to co-ordinate these services and to identify and fill any gaps. One of the aims of the Project is to improve the functioning of the family as a unit, which should reduce the subsequent development of antisocial behaviour.

The project is similar in some respects to the Dundee Families Project (Dillane et al., 2001) and others provided by Councils and voluntary agencies with a focus on working with vulnerable families (usually at risk of eviction) involved in antisocial behaviour (see, for example, Nixon et al., 2006 for an evaluation of one Council-run and five NCH-run Intensive Family Support Projects in northern England) ³. The Edinburgh Project is less intensive than these other models and has a stronger focus on prevention through early intervention and on families with at least one young child. It also draws on a New Zealand model known as Strengthening Families, which encourages families to develop and draw on their own resources.

The project team comprises one Project Manager, two Case Managers (each with a caseload of about 10 families) and an Administrator. The project's aim is to intervene when problems are starting to become apparent ( e.g. when a child is truanting or behaving badly at school) but before a crisis point ( e.g. school exclusion) has been reached. However, it can be difficult to get families to engage with the project until they are prepared to recognise the seriousness of their situation ( e.g. they are facing imminent eviction).

Methodological Challenges

Several methodological challenges were encountered during the evaluation of the project, including:

• the lack of information on expenditure over a full financial year,
• projects of this type take 18-24 months to become established and achieve 'steady state',
• the Council does not levy organisational overheads during the pilot stage of projects,
• the lack of a realistic comparator to reflect what may have happened in the absence of the intervention,
• many of the potential benefits would arise in the longer-term, but only if the positive effects of the intervention can be sustained, and
• there is a paucity of specific Scottish cost information to use when examining the potential cost consequences.

Activity and Outcomes Data

During the period August 2005 to December 2006, the project worked with 37 families (16 from North Edinburgh and 21 from West Edinburgh). Two-thirds of these families were headed by a lone adult parent and each family had an average of 3.9 children (range: 1 - 9). A total of 97 families (44 from Broomhouse and Sighthill in West Edinburgh and 52 from North Edinburgh) were referred to the project during this period, meaning that around a third were accepted.

As well as including one or more children whose behaviour was causing concern, 60% of accepted families were affected by parental mental health problems and substance misuse, over half by significant poverty and debt, and 40% were affected by poor parental physical health or learning disabilities. Slightly over three-quarters (77%) of accepted families had been involved in, or had generated complaints about their involvement in, antisocial behaviour. Four-fifths of families working with the project had had problems relating to the attendance or behaviour of their children at school and two-thirds of families were involved with the police and social workers at the time of referral or previously. Two-thirds of accepted families were living in City of Edinburgh Council tenancies. At least 40% of these were seriously overcrowded. The remaining third of families were either living in the private rented sector or in homeless accommodation. Over a third of the families accepted by the project were homeless or threatened with homelessness at the point of referral.

Table A4.13 summarises the project's activity since its inception in August 2005 to December 2006. Two families had worked with the project for 17 months ( i.e. since it started in August 2005) and almost half (10/21) of the families working with the project in December 2006 had done so for 12 months or longer.

Table A4.13: Summary of project activity: August 2005 - December 2006

Source: Edinburgh Early Intervention Families Project

Given that families are referred to the project for a wide variety of reasons, it can be difficult to capture the specific outcomes, as these will vary from family to family. However, cases are not closed until a family's behaviour and functioning has improved (unless the family chooses to cease engaging with the project before this point is reached). The project also monitors each family's ability to function effectively during their time with the project using the Family Development Matrix ⁴. This is completed regularly (usually before each action planning meeting) and although there is some variation between families, most showed an improvement in their functioning over their time with the project.

Costs

Project expenditure to the end of December 2006 is shown in Table A4.14. The table shows that salaries and other employee-related costs comprised about three-quarters of expenditure in each period. However, a fifth of expenditure during the 2005/06 financial year was on the external evaluation, whereas almost a quarter of the expenditure to date in 2006/07 has been on buying in other services and support needed by the families.

Table A4.14: Expenditure on the Early Intervention Project

Source: Edinburgh Early Intervention Families Project

The forecast at the end of December 2006 was for project expenditure of about £270,000 in 2006/07, which included £25,000 for the second year of the external evaluation and a total of £90,000 for 'other expenses'. However, it should be noted that due to its status as a pilot project, City of Edinburgh Council did not levy any overhead payments for premises, central support and other services. If the project continues with funding from mainstream sources after the pilot phase, these overheads are expected to amount to about £50,000 per year. Given that the annual allocation of £90,000 for 'other expenses' seems to be relatively high (an estimate of about £45,000 may be more realistic), the above figures suggest that the project should cost about £250,000 per year after the pilot phase (assuming no external evaluation after the pilot phase and making no adjustment for inflation).

Costs per Unit of Outcome

This evaluation adopts similar measures for units of outcome as those used in the evaluation of the Shelter Inclusion Project (Jones et al., 2005) and the evaluation of six projects in England focusing on providing intensive support to families at risk of losing their tenancies due to antisocial behaviour (Nixon et al., 2006). The two key units of outcome used for costing purposes in those projects were the average cost per client month ⁵ and the average total cost per closed case ⁶. In addition, the average cost per family member month ⁷ has been calculated for this project.

These unit costs (which have been calculated excluding the evaluation costs) are shown in Table A4.15. The average cost per client ( i.e. per family) per month fell from £888 in 2005/06 to £646 in the first nine months of 2006/07. The average cost per family member per month fell from £179 to £123 over the same period. These reductions occurred as the project built up its caseload of clients. However, the average total cost per closed case rose from £4,292 to £4,670. This was because average contact duration for closed cases increased (see Table A4.13), partly because families with more complex problems take longer to reach closure. Table A4.15 also shows that the maximum cost to date for a closed case is less than £10,000.

Table A4.15: Average unit costs for the Family Intervention Project

Given the above information, it is possible to derive some estimates of future unit costs. These are shown under a variety of potential scenarios in Table A4.16 and are based on the assumption that the project has an average caseload of 20 families each month ( i.e. it provides 240 client contact months each year). If the total annual expenditure on the project is £250,000 (as suggested above, when overhead payments are included) and the average duration of contact with the project is 12 months ⁸, the average total cost per closed case would be £12,500. However, it would only be £9,375 if the average duration of contact was nine months. These costs are similar to previous evaluations of similar projects (Jones et al., 2006; Nixon et al., 2006).

Table A4.16: Average total cost per closed case in various scenarios

Note: Based on an average caseload of 20 families per month

Potential Cost Consequences

Unlike most of the other initiatives included in this economic evaluation, there is already a considerable amount of published literature on the potential cost consequences for publicly-funded services of not tackling children's and families' antisocial behaviour ⁹. Some of these are short-term, such as the costs of tenancy failure, foster and residential care, and youth justice. Some of the key cost consequences that may be prevented by a project such as the Early Intervention Families Project are presented in Table A4.17 below. The Table shows clearly that the costs can be very high, particularly if specialist provision is required.

Table A4.17: Some potential short-term public-cost savings

An in-depth discussion of these cost consequences is provided in Nixon et al. (2006). Although it was not possible to determine the costs for publicly-funded services that would have been incurred by these families in the absence of the intervention, it includes an illustrative case study showing how a family with four children can easily generate public costs of more than £330,000 over a 12-month period if their antisocial behaviour is not addressed. This sum exceeds the estimated annual cost of running the entire Early Intervention Families Project by about £80,000.

Some of the potential cost consequences, such as those of poor educational attainment and social exclusion, have lifelong and even inter-generational impacts for families and for society as a whole. One of the main strengths of the Edinburgh Early Intervention Families Project is that it focuses on families with children at primary school that are experiencing problems. There are many programmes and initiatives that focus on pre-school age children and on youths, but children aged between 5 and 12 years tend to be overlooked - yet these years are crucial to a person's development.

Although school attendance does not guarantee success in passing examinations and gaining qualifications, there are strong correlations between them. There will also be substantial longer-term savings if the children are prevented from being NEET (not in education, employment or training) when aged 16 - 18 years (and above). Time spent in primary school lays the foundations for future educational attainment, employment and lifetime earnings. The personal costs of not addressing childhood problems with antisocial behaviour are also very high, as people tend to become 'trapped out of opportunity' (Evans and Eyre, 2004). A study of the financial cost of social exclusion due to antisocial behaviour in childhood (Scott et al., 2001) showed that children and adolescents with persistent and pervasive patterns of ASB cost society an average of over £60,000 ¹⁰ more by the age of 28 than those not displaying such problems in childhood. It concluded that:

Antisocial behaviour in children is a major predictor of how much an individual will cost society. The cost is large and falls on many agencies, yet few agencies contribute to prevention, which could be cost-effective.

Those lacking knowledge and skills generally face a lifetime of low-income jobs and/or unemployment (see, for example, Freud (2007) and Prince's Trust (2007), which both illustrates the benefits for individuals and economic productivity of reducing benefit dependency and social exclusion by encouraging young people to gain qualifications and skills).

Although there is no guarantee that initiatives such as the Early Intervention Families Project will succeed for all families (especially in the longer-term), changing the behaviour of a few families will generate considerable public cost savings through preventing the need for other expensive services in both the short-term and the longer-term. Furthermore, the potential personal and economic benefits for family members are considerable. The Early Intervention Families Project offers an effective and potentially highly cost-effective approach to preventing childhood antisocial behaviour from escalating. It also helps to improve family functioning, which may help reduce family breakdown and its associated public costs.

Fife Safer Neighbourhoods Team

Description of Service

The creation of a Safer Neighbourhoods Team ( SNT) was an integral part of Fife's antisocial behaviour strategy. The primary function of the Team is to implement the provisions of the Antisocial Behaviour etc. (Scotland) Act 2004 in a co-ordinated, multi-agency manner.

The SNT is split into two elements. The SNT Co-ordination Team comprises representatives from a wide variety of organisations, including Fife Constabulary, Fife Fire and Rescue Service, Fife Council Housing Investigation Team, Victim Support, an antisocial behaviour analyst, and administrative support. It is co-located with the Noise Nuisance Team. Two additional staff members were also recruited to work within the Force Contact Centre to deal with all calls relating to antisocial behaviour. Calls that are 'trivial' in policing terms are referred on to the appropriate service. The SNT has approximately 60 staff attached to it.

The SNT Operations Team comprises four Police Constables, who target small geographical areas experiencing antisocial behaviour problems with the remit of implementing the provisions of the Antisocial Behaviour etc. (Scotland) Act 2004. It was initially envisaged that the Operations Team would work within an identified locality for a prolonged period of time (possibly up to 18 months). This stemmed from the premise that a problem-solving partnership approach to dealing with crime and antisocial behaviour would be implemented within the selected area. However, following commencement of Team operations, it was decided that a more fluid, intelligence-led method of Team deployment would be advantageous. The medium- and long-term aspects of the problem-solving approach need to be carried forward by local officers and associated agencies following redeployment of the SNT. It should be noted that the SNT (as a whole) is viewed within Fife as a strategy rather than as a team. It encompasses all partners and agencies with an interest in antisocial behaviour and brings various agencies on board with varying levels of involvement as appropriate.

Illustrating the Modus Operandi of the SNT- Deploying the SNT in Methil

The work of the SNT and how it inter-acts with other agencies in Methil (one of the Fife case study neighbourhoods) illustrates its activities. To ensure that the SNT was deployed in Fife in an intelligence-led manner, the SNT Antisocial Behaviour Analyst analysed police call data for the three months of October - December 2005. This identified that the Methil area generated the highest number of calls relating to antisocial behaviour in Fife and it was therefore decided to deploy the SNT Operations Team in this area, which includes 1,008 residential properties, 39 businesses (mainly retail and services) and one secondary educational establishment. The SNT Operations Team was deployed between 7 February 2006 and 14 July 2006.

The specific objectives of the SNT activity were to:

• reduce the level of antisocial activity occurring in the target area,
• target repeat offenders with antisocial behaviour legislation and other enforcement, prevention and intervention methods, and
• reduce the impact of antisocial behaviour on the quality of life of local residents.

The Team started their work by gathering information from a diverse range of sources within the area, including a household survey. This approach provided significant amounts of information about the area and an early insight into the ten main community concerns, from which a Problem Profile was created. This analytical document provided a clear picture of local concerns about antisocial behaviour within the target area. Its purpose was to enable managers to prioritise problems, identify appropriate resources, and determine tactics for problem resolution. It also included recommendations for prevention and enforcement activity.

The Problem Profile, which subsequently provided the basis for all SNT activity within the target area, identified local hotspots associated with antisocial behaviour and areas of environmental degradation. It also provided a comprehensive list of offences, areas or individuals to be targeted for intervention within several categories, namely:

• noise nuisance,
• disruptive neighbours,
• individuals under 16,
• individuals over 16,
• complainers about youth disorder,
• victims of malicious antisocial behaviour,
• areas of environmental degradation and
• vehicle nuisance.

Activity and Outcomes

After this initial "fact finding" work, a multi-agency group was established to take forward the response phase of the SNT's involvement in Methil. At the initial meeting of this group, which was chaired by the Safer Neighbourhoods Manager, tasks were allocated to relevant agencies. For example, dealing with concerns about dog excrement (the highest ranked local priority) was allocated to Fife Council's Environmental Services Department, whilst problems with youths gathering were allocated to the Community Services Detached Youth Workers. A Task Review Group was established (chaired by the Locality Manager) which met regularly to ensure that relevant actions were progressed.

The SNT also carried out extensive operational activities within the target area. Many of these were enforcement-related activities, including:

• issuing ASBO warnings to relevant individuals,
• deploying the mobile CCTV Unit,
• seizure of vehicles under antisocial behaviour legislation, and
• referrals to SACRO for relevant juvenile offenders.

Others activities included maintaining high visibility foot patrols in hotspot areas and providing reassurance visits to local businesses and individuals from ethnic minority backgrounds. Because of the multi-faceted and multi-organisational activities undertaken in Methil no quantitative data on specific activities was provided.

To determine the impact of the SNT in Methil, a number of measures were used to judge its success or failure. These included:

• the number of calls made to the police regarding antisocial behaviour in the target area,
• whether the antisocial behaviour of targeted repeat offenders had reduced, and
• improvement in the public perception of the local residents as measured by a comparison of 'before and after' surveys.

Three types of calls to the police were monitored and the following changes were identified when the August 2005 - January 2006 period was compared with February - August 2006:

• antisocial behaviour calls: 17% reduction (between January 2006 and August 2006),
• noise calls: 21% increase, and
• street disorder type calls: 34% reduction.

It is difficult to interpret these figures because the public had also been encouraged to phone the police to report problems with antisocial behaviour. However, the falling trend in calls relating to street disorder (including youth-related nuisance and disturbances) does seem to suggest that the SNT and the overall approach had a significant local impact, especially as calls relating to antisocial behaviour and street disorder would be expected to increase in the summer months.

Youth disorder was an endemic problem in the target area during the period of SNT deployment. The Problem Profile had identified 33 youths aged under 16 who were involved in low-level crime and disorder. The most prolific offenders were targeted with enforcement and intervention measures, many of which were successful. For example:

• 10 youths were referred to SACRO, with 5 showing a dramatic improvement in their behaviour,
• 2 youths who refused to interact with SACRO were referred to the local Family Intervention Service,
• 1 youth who was referred to SACRO entered into an Acceptable Behaviour Agreement, and
• 13 ASBO warnings were issued, with 10 of these individuals involved in no further incidents.

The targeting of repeat offenders was the most successful and unambiguous outcome of SNT activity in Methil. In addition to the successes presented above, 11 of the 13 addresses targeted for noise and disruptive behaviour showed no further incidents, while 4 of the 5 addresses issued with tenancy warnings showed no further incidents.

The findings of the 'before and after' household surveys suggest that there was a minor reduction in the proportion of residents feeling that antisocial behaviour had a significant or fairly significant impact on their quality of life (from 51.7% before deployment of the SNT to 48.2% after deployment). They also show that there was an increase in the proportion of residents feeling that antisocial behaviour had no impact on their quality of life (from 11.0% to 17.9%). However, these figures should be treated with extreme caution, because only 56 responses were returned for the 'after' survey, a response rate of only 5.6% (compared with 30.5% for the 'before' survey) and there is no way of knowing the extent to which the views of the 56 respondents were representative of the local community.

Methodological Challenges

The SNT is an integral part of local multi-agency working and it was not possible to isolate its specific activities and their associated costs and benefits. Activities are undertaken by several agencies and cannot be linked directly to outcomes. Services are funded through a variety of budgets across different organisations. The use of 'before and after' surveys to measure the perceived impact of the approach is fraught with problems, especially when response rates are very low. Also, given that a number of initiatives focusing on reducing antisocial behaviour and crime were already in place, it is not possible to link any observed benefits specifically to the work of the SNT. Although it is possible to attribute a proportion of the observed benefits to the SNT, this estimated value would inevitably be arbitrary.

Costs

It was not possible to collect meaningful data on the costs of the SNT because of its multi-organisational composition. It was also not possible to calculate any unit costs for specific activities or outcomes. Furthermore, although the Operations element of the SNT undertook a variety of specific tasks when working in the Methil area, some of its work involved co-ordinating the required services, which are provided by local agencies, making it an integral part of local multi-agency working. In many ways it fulfils the role of a troubleshooter and case management system as well being a provider of some services. Some of the SNT's activities will have resulted in increased expenditure (at least in the short term) by other agencies, although some potential future expenditure should also have been prevented.

Potential Cost Consequences

Although it is not possible to isolate the specific costs and benefits of the SNT, the available evidence suggests that the overall approach can deliver considerable benefits, especially in terms of reducing disturbances and repeat offences (particularly those involving young people). However, the identified outcomes cannot be attributed specifically to the SNT due to the contributions of all of the other agencies working to achieve the stated objectives. Nor is it possible to attribute a specific proportion of the benefits to the SNT. The identified benefits are likely to be enjoyed by specific individuals and by their communities, at least in the short-term. Follow-up work would be needed to determine if the benefits were sustained after the withdrawal of the SNT Operations Team. Nevertheless, the approach is likely to have resulted in the prevention of some significant amounts of expenditure relating to youth offending. The SNT appears to work effectively. It provides a good example of the benefits that can be achieved through adopting a coordinated multi-agency response in a specific problem area.

North Lanarkshire Night Noise Team

Description of Service

The Antisocial Behaviour etc. (Scotland) Act 2004 gave local authorities across Scotland the option to impose specific restrictions on night noise in their area. North Lanarkshire Council resolved to adopt noise control provisions between 19:00 and 07:00 throughout the year. The Night Noise Team, which started operating in late May 2005, comprises a team of one Senior Environmental Health Officer and three Environmental Health Officers ( EHOs). A member of the Team is on duty every night between 20:00 and 03:30 to respond to complaints made about night-time domestic noise by residents of North Lanarkshire. When a call is received the EHO on duty visits the home of the complainant unless requested by the caller not to do so or the problem can be resolved over the telephone. The team works very closely with the police, who screen all callers to make sure that it is safe for a Team member to visit the address.

Methodological Challenges

Given that the Team has a clearly-defined budget and that all calls about night noise are logged, it is relatively straightforward to calculate the average cost per call received. However, given that in many instances the noise self-resolves (either with or without a visit to the home of the complainant), it is not possible to define a 'successful' outcome relating directly to the Team's intervention.

The analysis faced three other methodological challenges: the lack of activity and cost data over a full financial year, difficulties in placing a value on the possible deterrent effect of the service on levels of night noise, and the difficulty in identifying quantitative and qualitative outcomes linked to the reduction/cessation of night noise in a neighbourhood and subsequently placing an economic value on these benefits.

Activity, Outcomes and Costs

Since the Team's inception, the numbers of calls received has tended to fluctuate depending on the time of year and day of the week (calls are generally highest on Fridays and Saturdays). An average of about 52 calls per month was received during ten full months of operation in 2005/06. This fell to an average of about 47 calls per month during the first nine months of 2006/07. Almost 80% of the calls related to loud music, and 10% to domestic animals.

Approximately a quarter (24.1%) of the 523 calls received in 2005/06 were resolved over the telephone, with visits being made to the other three-quarters of complainants. Of these cases requiring a visit most were resolved by verbal warnings to alleged perpetrators, with Warning Notices being issued in one-in-twelve (8.1%) visits ( i.e. Warning Notices were issued for 6.1% of calls received). Data for the first three quarters of 2006/07 gives a somewhat different picture, as over half (53.3%) of the 363 complaints were resolved over the telephone. Just under four-fifths (78.2%) of those complaints resulting in a visit were resolved by verbal warnings, with over one-in-five visits leading to the issue of a Warning Notice ( i.e. 10.2% of calls received led to the issue of a Warning Notice). Therefore the data shows that, although a lower proportion of complaints were being visited in 2006/07 than in 2005/06, a higher proportion led to the issue of a Warning Notice. Since the Team started operating, only one Fixed Penalty Notice has been issued and no seizures of equipment have been required.

Because expenditure data is not available over a full financial year, the costs should be interpreted with some caution. Expenditure during May 2005 - March 2006 was almost £328,000. Staff payroll costs accounted for just over three-fifths (61%) of this, and central overheads (including office accommodation) were responsible for slightly more than one-eighth (13%). The service cost around £245,400 between April and December 2006. This excludes the annual payment towards administrative support and service overheads, which accounted for 20% of expenditure in 2005/06. Staff costs in the period of analysis during 2006/07 comprised over three-quarters (77.4%) of this amount, and publicity and advertising - at almost £16,400 - accounted for 6.7% of expenditure. Each year almost £32,000 was paid for police support (including providing safety checks).

These activity and cost data give an average cost per call received by the Team in 2005/06 of £627 and an equivalent cost of £609 for the first nine months of 2006/07 (though this is likely to be an underestimate of the final cost for the year because of the exclusion to date from the expenditure accounts of the internal cost transfers for administrative support and service overheads).

Potential Cost Consequences

The benefits to local residents of reducing night noise are likely to be considerable (including, for example, improved sleep and better daytime functioning and fewer disputes with neighbours). However, such benefits are hard to quantify financially. There are also cost savings to other services, including housing and the police, that may otherwise have needed to intervene in a case.

However, the effectiveness of the service is dependent upon residents being aware of it and knowing how they can access it. Despite considerable expenditure on promoting the service in a variety of ways, calls have been much lower than anticipated. It is not clear if night noise is less of a problem in North Lanarkshire than initially thought, or if the knowledge that the Team exists has a strong deterrent effect. The Senior EHO who manages and promotes the service believes that the unit cost of the service could be reduced by using trained Noise Officers rather than EHOs and/or by extending the Team's responsibilities to include night noise from commercial premises.

Scottish Borders Freephone Helpline

Description of Service

Scottish Borders' Anti-Social Behaviour Unit ( ASBU) became operational at the end of January 2006. At this time a freephone helpline was launched for the provision of general advice or to point callers to the most appropriate support. It is answered by the four members of the ASBU team and operates from 08:45 to 17:00 from Mondays to Thursdays and from 8:45 to 15:45 on Fridays. (Calls received outwith the above working hours go to an answer service where the caller can leave their name and contact number for follow-up contact the next working day). Staff on the Scottish Borders Council's switchboard also direct relevant calls to the helpline. The helpline number has been publicised extensively throughout the Council area.

Where possible, the caller is helped directly. If the caller lives in an area served by the community warden service (Kelso or Galashiels), a community warden will be sent round to discuss the problem face-to-face and, if necessary, to collect some evidence. Callers living in other parts of the Scottish Borders will be sent incident diary sheets to record times and dates of problems and will be telephoned again by a member of the ASBU team within 28 days. Not all of the calls made on the helpline relate to ASBU problems: if the ASBU team cannot help, the caller's details are taken and these are forwarded to the relevant Council service or local organisation, such as Technical Services or an Environmental Health Officer. Where appropriate, callers are encouraged to talk to other relevant parties, such as their landlord, when they are sent incident sheets.

Methodological Challenges

Calculating the costs associated with the freephone helpline are relatively straightforward, given that costs can be readily identified for the equipment, the allocation of the freephone number and the calls received. The publicity budget for the ASBU and the helpline can also be identified. The costs of the time spent by ASBU staff handling calls can be ignored if it is assumed that answering telephone queries is part of their job (and therefore already covered by their salaries), and that the freephone number does not increase this element of their work to a significant extent. The average cost per call received can easily be calculated.

However, two specific methodological issues were encountered:

• The absence of any baseline for comparison ( e.g. of the numbers of call received), given that the ASBU unit and the freephone helpline became operational at the same time.
• The knock-on cost consequences for other services as a consequence of calls being made to the helpline - these costs could increase if issues are raised that otherwise may not have required a response or decrease if the service enables more timely responses to be provided due to problems being identified quickly.

Activity, Outcomes and Costs

During the first year of operation the ASBU received a total of 302 calls on the helpline, which gives an average of about 25 per month. However, by the end of 2006 the Unit was receiving about 35 such calls each month, as knowledge of the service built up. ASBU staff started to monitor the number of out-of-hours call from 16 October 2006 with 42 such calls being received between 16 October 2006 and early February 2007. During most weekends between two and six such calls are received, suggesting that a significant proportion of calls relating to antisocial behaviour are made outwith standard working hours.

A budget of £5,000 was allocated to the service during 2006/07. Out of this, £1,210 was spent on the equipment and the allocation of the freephone number. A further £334 was spent to the end of January 2007 on the received calls; these were expected to cost about £500 to the end of March 2007. The remainder of the budget ( i.e. about £3,300) was expected to be used on further advertising the work of the ASBU, including the freephone number.

The above costs suggest that the freephone number was expected to cost a total of about £1,700 during 2006/07 ( i.e. £1,200 for the equipment and £500 for the received calls). If an average of 35 calls is received each month, this would amount to a total of 420 over a year, giving a cost of about £4 per call made to the freephone number. This would rise to an average of almost £12 per call if the remainder of the £5,000 budget was used to advertise the freephone number. The longer-term costs associated with this service will depend on the numbers of calls received and the amount spent on advertising it.

Potential Cost Consequences

The impact of the freephone helpline on the number of calls made to the Scottish Borders Council relating to antisocial behaviour is not known, though it will improve equity of access to services relating to antisocial behaviour for those living in rural areas and in towns without community wardens. It is also too early to tell if the helpline is having an impact on the workload of other services and whether this ultimately increases or decreases the costs of these services. The fact that the service is provided free of charge to users will prevent callers being deterred by the cost of making the call. Thus the service offers benefits to both the Council and local residents, though such benefits cannot readily be quantified.

Conclusions

This annex has used information provided by the 12 evaluated initiatives and, where relevant, drawn from previously published research to identify and quantify the associated costs and short-term and longer-term benefits of each initiative. The interpretation and use of this information will depend upon local policies and priorities, as an initiative that is considered to offer good value for money in one area may not be felt to do so in another neighbourhood or local authority. Although local decision making can never be an exact science, it will be better informed where it takes account of the tangible costs and considers the intangible consequences for local stakeholders - some of which may be experienced over a considerable period of time. The research has also highlighted a number of methodological challenges in undertaking economic evaluations of local antisocial behaviour interventions.

REFERENCES

Anti-Social Behaviour Unit (2003) The one day count of anti-social behaviour. London: Anti-Social Behaviour Unit/Home Office

Atkinson, R., Mullen, T. and Scott, S. (2000) The Use of Civil Legal Remedies for Neighbour Nuisance in Scotland. Edinburgh: Scottish Executive Central Research Unit

Audit Commission (2004) Youth justice 2004 - A review of the reformed youth justice system. London: Audit Commission

Brand, S. and Price, R. (2000) The economic and social costs of crime. Home Office Research Study 217. London: Home Office Research, Development and Statistics Directorate

Brown, A.P., Barclay, A., Simmons, R. and Eley, S. (2003) The Role of Mediation in Tackling Neighbour Disputes and Anti-social Behaviour. Edinburgh: Department of Applied Social Science, Stirling University. Scottish Executive Development Department Social Research

Campbell, S. (2002) A Review of anti-social behaviour orders. Home Office Research Study 236. London: Home Office Research, Development and Statistics Directorate

Crowther, C. and Formby, E. (2004) An Evaluation of the Costs of Responding to and Preventing Anti-Social Behaviour in Rotherham MBC. Sheffield: Research Centre in Community Justice and Centre for Social Inclusion, Sheffield Hallam University

Dignan, J., Sorsby, A. and Hibbert, J. (1996) Neighbour Disputes: comparing the cost-effectiveness of mediation and alternative approaches. Sheffield: Centre for Criminological and Legal Research, University of Sheffield

Dignan, J. and Sorsby, A. (1999) Resolving Neighbourhood Disputes through Mediation in Scotland. Edinburgh: Scottish Office Central Research Unit

Chapter Nine: Dillane, J., Hill, M., Bannister, J., Scott, S. (2001) Evaluation of the Dundee Families Project. Edinburgh: Dundee City Council, Scottish Executive, NCH Scotland

Dubourg, R, and Hamed, J. (2005) Estimates of the economic and social costs of crime in England and Wales: Costs of crime against individual and households, 2003/04. London: Home Office, Section 1 of Online Report 30/05
( http://www.homeoffice.gov.uk/rds/pdfs05/rdsolr3005.pdf)

Evans, M, and Eyre, J. (2004) The opportunities of a lifetime: Model lifetime analysis of current British social policy. Bristol: The Policy Press

Freud, D. (2007) Reducing dependency, increasing opportunity: options for the future of welfare to work (an independent report). London: Corporate Document Services, Department for Work and Pensions

House of Commons (2004) Select Committee on Public Accounts Fortieth Report.
( http://www.publications.parliament.uk/pa/cm200304/cmselect/cmpubacc/307/30702.htm)

Hunter, C., Nixon, J. and Shayer, S. (2000). Neighbour Nuisance, Social Landlords and the Law. Coventry: Chartered Institute of Housing in association with the Joseph Rowntree Foundation

Jill Dando Institute of Crime Science (2003) Measuring the Impact of MVRIb Initiatives on Abandoned Vehicles. London: University College London

Jones, A., Pleace, N., Quilgars, D. and Sanderson, D. (2006) Addressing antisocial behaviour: An independent evaluation of Shelter Inclusion Project. London: Shelter

Marshall, A. (1999) The Effects of Neighbour Disputes on Health. A Report by RFC Total Quality Consultancy on behalf of Fife Community Mediation. Fife: Fife Community Mediation

Mulcahy, L. with Summerfield, L. (2001) Keeping it in the Community: an Evaluation of the Use of Mediation in Disputes Between Neighbours. Norwich: The Stationery Office

Neighbourhood Renewal Unit (2004) Research Report 8 Neighbourhood Wardens Scheme Evaluation. London: Neighbourhood Renewal Unit, Office of the Deputy Prime Minister

Nixon, J., Hunter, C., Parr, S., Whittle, S., Myers, S. and Sanderson, D. (2006) Anti-social Behaviour Intensive Family Support Projects: An evaluation of six pioneering projects. London: Department for Communities and Local Government

Pawson, H., Flint, J., Scott, S., Atkinson, R, Bannister, J., McKenzie, C. and Mills, C. (2005) The Use of Possession Actions and Evictions by Social Landlords. London: Office of the Deputy Prime Minister

Prince's Trust (2007) The Cost of Exclusion - Counting the cost of youth disadvantage in the UK. London: The Prince's Trust

Reading Borough Council (2003) Spotlight on Abandoned Vehicles. Reading: Reading Borough Council

Scott, S., Knapp, M., Henderson, J. and Maughan, B. (2001) 'Financial cost of social exclusion: follow up study of antisocial children into adulthood' . British Medical Journal, 323; 194-7

Social Exclusion Unit (2000) National Strategy for Neighbourhood Renewal - Report of Policy Action Team 8: Antisocial behaviour. London: Social Exclusion Unit

Whitehead, C., Stockdale, J. and Razzu, G. (2003) The Economic and Social Costs of Anti-Social Behaviour: A Review. London: London School of Economics

« Previous | Contents | Next »