Estimating the costs associated with community justice

Foreword | A range of alternative justice models have been introduced to address the perceived weaknesses of traditional criminal justice processes. While the success of these alternative models is usually assessed in terms of outcomes such as reoffending, they often face scrutiny over the additional costs incurred by government. This paper is one of two AIC publications focused on the Neighbourhood Justice Centre (NJC) in the City of Yarra in Melbourne. It compares the operating costs of the NJC with more traditional court and service delivery models.

The results show that there are areas in which the NJC appears to be more cost-efficient. What these results also show, perhaps unexpectedly, is that community justice models incur additional costs over and above those of normal magistrates’ courts. However, when read alongside the paper by Stuart Ross evaluating the impact of the NJC on order compliance and reoffending, it is apparent that this additional investment contributes to a number of important outcomes. Further work is necessary to determine whether the NJC is cost effective. Nevertheless, having access to data on the relative cost of community justice models like the NJC contributes to a more informed discussion around the cost implications for government.

Chris Dawson APM

In recent years, governments at all levels have been faced with increasingly tight budgets and limited resources, placing greater pressure on service providers—including those involved in crime prevention and criminal justice—to demonstrate value for money. This has resulted in a growing interest in economic analysis as a tool to assess the cost-efficiency and cost-effectiveness of programs and help inform decision-making. This is reflected in the evaluation guides that have been produced by treasury departments to encourage the robust economic analysis of government funded programs (eg Department of Treasury and Finance 2013).

Established in 2007 to address high rates of crime and disadvantage, the Neighbourhood Justice Centre (NJC) provides a range of local justice and social services to the City of Yarra community in Victoria. A major challenge for the NJC is dealing with questions regarding value for money and responding to the criticism that they are an expensive alternative to traditional responses to crime and offending. The development and implementation of new and innovative programs often involves a significant financial investment by government and, in the case of the NJC, the concentration of these resources in one location. Recent examples, such as the abolition of several specialist court models in Queensland, have shown that, even where there is evidence of positive outcomes, an inability to demonstrate cost-efficiency and effectiveness in financial terms can weaken the argument in favour of ongoing funding and support—even where this is not the principal reason for the decision.

The purpose of this paper is to address one dimension of the value for money debate and compare the operating costs of NJC court and client services with similar mainstream programs operating within the Magistrates’ Court of Victoria. Following a brief overview of the NJC and evidence from previous economic studies of community justice models, the methodology used to estimate the costs associated with NJC court and client services is described, along with the results of a cost comparison of the NJC and the Magistrates’ Court of Victoria. Directions for further economic analysis for the NJC and other similar programs are proposed.

Neighbourhood Justice Centre

The NJC provides a range of onsite services and support to accused persons, victims, civil litigants and local communities within the City of Yarra. The NJC is based on a community justice model, which is distinguished from more traditional models of criminal justice by its emphasis on providing services to the local neighbourhood, community engagement and participation in the justice process, therapeutic problem-solving, decentralised authority and accountability, integrated services (a ‘one-stop shop’) and early intervention and resolution (Courts and Tribunals Unit 2009; Karp & Clear 2000; NJC nd). The community justice model is underpinned by two basic assumptions: that communities are different and therefore require tailored responses, and that stable families, positive neighbours and peers and other informal networks of social control are important factors in influencing crime and offending behaviour (Clear, Hamilton Jr & Cadora 2011).

Among other services the NJC includes a multi-jurisdictional court, established to hear matters normally dealt with through the Magistrates’ Court of Victoria, the Children’s Court of Victoria, the Victorian Civil and Administrative Tribunal (VCAT) or the Victims of Crime Assistance Tribunal (VOCAT), but involving a wider range of partners and providing additional support to participants. There are a number of justice-related services located at the centre, including legal aid and dispute resolution services. The NJC also has a dedicated client services team, which provides assessment, treatment and referral services to local people involved in justice-related matters. Some of the services delivered by this team include mental health, drug and alcohol, housing, financial counselling, education and other support services. Consistent with the community justice model, and to better meet the needs of the local community, the NJC utilises a range of service delivery models such as outposting to local agencies, and has developed a range of partnerships with local support agencies.

In addition to court and client services, the NJC invests significant resources in a variety of community-based initiatives targeting the City of Yarra including crime prevention, community development and justice grants. For example, the NJC is involved in programs to reduce family violence in social housing, to support the families of prisoners returning to the community and to provide outreach and access to services to reduce street drinking and associated problems.

The NJC has three overall goals that reflect the unique focus of the community justice approach. The NJC aims to prevent and reduce criminal and other harmful behaviour in the Yarra community, to increase confidence in and access to the justice system for Yarra communities, and to strengthen the NJC community justice model and facilitate the transfer of its practices to other courts and communities. While they are described as separate functions in this paper, the three components of the NJC—court services, client services, and community engagement and crime prevention—work together to achieve these aims.

Community justice models: Effectiveness and efficiency

The NJC is based on community justice models that have been implemented overseas, a number of which have been evaluated. Some of these evaluations have included economic analyses, or at least some attempt to compare the community justice model with mainstream criminal justice processes in terms of overall efficiency. Overall, the findings from these evaluations have been mixed.

For example, the evaluation of the Red Hook Community Justice Center (RHCJC), the first multi-jurisdictional community court in the US, found that there were almost $4,800 in estimated taxpayer savings per misdemeanour defendant, calculated by the avoided victimisation costs relative to similar cases processed in traditional misdemeanour courts (Lee et al. 2013). This came to over $15 million total savings in avoided victimisation costs (Lee et al. 2013). Full data on the overall costs and benefits associated with RHCJC were not available but, based on the known costs and benefits of RHCJC, the results showed a net benefit of over $6.8 million (Lee et al. 2013). These savings exceeded the total costs of operating the RHCJC by almost two to one (Lee et al. 2013).

However, an evaluation of the North Liverpool Community Justice Centre (NLCJC), which was the first community justice centre to be established in the UK, concluded that offenders sentenced at the NLCJC were no less likely to reoffend than a matched group of offenders sentenced in a mainstream court, and were actually more likely to breach the conditions of their order (Booth, Altoft, Dubourg, Goncalves & Mirrlees-Black 2012). This was consistent with the findings of an earlier study into the impact of the NLCJC and a community justice centre at Salford (Joliffe & Farrington 2009). However, the evaluation of the NLCJC did produce a number of positive findings in terms of efficiency, including evidence that the average time from offence to conviction was lower at the NLCJC and that fewer hearings went to trial (Booth et al. 2012). Conversely, there was a higher proportion of cracked trials—trials which do not proceed on the trial date—which was evidence of inefficiency.

There have also been some attempts to evaluate the efficiency and effectiveness of the NJC (Ross 2015; Ross, Halsey, Bamford, Cameron & King 2009). An evaluation of the first two years of operation found a small, non-significant difference in the rate of reoffending between offenders sentenced in the NJC and a comparison group, although this did not take into account potential differences in the risk of reoffending, such as prior offending. There was evidence of improved compliance with community-based orders and an increase in the number of community work hours completed. Consistent with the therapeutic and problem-solving model of operation, the length of hearings was observed to be greater in the NJC for criminal and VCAT matters, but not for family violence intervention orders. Results of the cost-benefit analysis were positive, with a return of between $1.09 and $2.23 for every dollar invested, although there were limitations on this analysis related to the use of an unmatched comparison group and differences in recidivism outcomes that were not statistically significant. More recent analysis of order compliance and recidivism addressed the shortcomings of the earlier research and concluded that the NJC was more effective than normal magistrates’ courts on both outcome measures (Ross 2015). With these latest findings there is scope for more robust economic analysis of both the effectiveness and efficiency of the NJC.

Building on previous research into the operation and effectiveness of the NJC, the AIC was commissioned to develop a cost model for the NJC that would produce a reliable estimate of the costs associated with services delivered by the Centre and also enable comparison with mainstream service delivery models. This is an important first step in being able to determine the cost-efficiency and cost-effectiveness of the NJC.

Estimating the cost of NJC courts and client services

The approach adopted for the current study was to develop estimates of the average cost per unit of service delivered by the NJC and other models of service delivery, which was calculated by dividing the total cost of the service by the quantity of output produced. Information was drawn from the NJC detailed operating statement, administrative data supplied by the NJC and the annual reports of the Magistrates’ Court of Victoria, Children’s Court of Victoria, VOCAT and VCAT. This approach is known as the breakdown method (Moolenar 2009), and has been used extensively, as part of the Report of Government Service Provision, as both a measure of efficiency and a way of comparing the costs of government services between jurisdictions (SCRGSP 2014). While there are benefits associated with bottom-up approaches to calculating costs, whereby the amount of resources involved in an activity is multiplied by the cost of those resources, it was beyond the scope of the current study to collect the additional data for the NJC and the matched comparison group of cases which would be required for this type of analysis.

The first step was to develop an estimate of the average cost associated with each unit of service delivered by the court and client services sections of the NJC. A top-down approach was used, whereby the total operating expenditure associated with court and client services was calculated and then divided by the number of units of services delivered. Separate estimates for the different types of court and client services delivered by the NJC were developed, to account for differences in the amount of resources allocated to different court matters and types of services provided to individuals who come into contact with the NJC.

One of the challenges associated with estimating the cost of a service delivered by an organisation that delivers multiple services is separating the costs of that service from the total expenditure of the organisation. As part of their integrated ‘one-stop shop’, the NJC provides a range of services alongside court and client services, with a significant proportion of time and resources dedicated to their community engagement and crime prevention function. These other functions were not the focus of the current study and the associated expenditure had to be excluded from any cost estimates produced.

It was therefore necessary to start by determining the proportion of the total NJC operating expenditure that was, on average, allocated to the work involved in delivering the court and client services functions. This was achieved by estimating the approximate percentage of time every staff member allocated to the core functions of NJC work (court services, client services, and crime prevention and community engagement and other non-court and client services) on a full-time equivalent (FTE) basis in an average month. This information was supplied by the NJC using a template developed by the AIC.

Importantly, the purpose of the current study was not to determine the overall cost efficiency of the NJC relative to the Magistrates’ Court; rather, it compares the cost efficiency of the two models in utilising their respective operating expenditure. Therefore, only those positions funded by the NJC were included as part of this process; positions allocated to the NJC by other agencies were excluded because their salaries were not funded by the NJC. This is important, because these same positions are not funded within the operating expenditure for other models of service delivery against which the NJC has been compared (ie magistrates’ courts). The assumption underpinning this approach is that the allocation of FTE to different service areas is indicative of the costs associated with providing that service.

Specific guidelines were established to assist with the estimation of staff time. According to these guidelines, a percentage of an FTE position was allocated to court and/or client services where that position was required for the management, administration and/or delivery of that NJC function. For example, to ensure comparability with traditional magistrates’ courts, FTE was allocated to NJC court services for registrars, coordinators and listing staff, and administrative and support staff, which includes human resources, IT, learning and development, finance and administration, contract and corporate management, strategic planning, security, executive and judicial support, specialist courts and services support, and project roles. FTE was allocated to client services for time spent supporting both internal referrals (clients referred to services within the NJC) and external referrals (clients referred to NJC partners operating outside of the centre). The remaining FTE was allocated to crime prevention and community engagement activity, and any other activity not related to the court and client services function.

The results from this exercise are presented in Table 1. All figures are presented in 2012 dollars. This shows that, in 2011–12, 55 percent of all NJC FTE staff time was allocated to court or client services, and the remaining 45 percent was allocated to crime prevention and community engagement activity and other non-court and non-client services. These estimates were then used to determine the approximate total expenditure required for each NJC service area. In 2011–12, the estimated cost associated with NJC court services was $1,067,163 (22% of total NJC expenditure) and the cost associated with delivering client services was estimated to be $1,654,094 (34% of total NJC expenditure).

Table 1 Estimated NJC expenditure by service type, 2011–12

	FTE (%)	Expenditure ($)
Court services	21.7	$1,067,163
Client services	33.7	$1,654,094
Crime prevention and community engagement (and other non-court and non-client services)a	44.6	$2,187,136
Total operating expenditure (excl. magistrates’ & VCAT member salary)	100.0	$4,908,393

a: This includes the daily operation of the centre and its interface with the community. Whilst not directly involved with court and clients, these programs support the operation and intersection of these entities and activities within the NJC as well as achieving the NJC’s broader outcomes

Note: Expenditure based on Actuals reported in the NJC detailed operating statement

Note: Total operating expenditure includes discretionary and non-discretionary expenditure. Excludes expenditure associated with Fitzroy Legal Services and Belgiam Avenue Neighbourhood House

Source: NJC Cost Model, 2014 [computer file]

The next step was to determine the distribution of expenditure associated with court services across the different types of court matters (jurisdictions and list types) heard at the NJC. This was necessary in order for direct comparisons to be made between the NJC and other court models, which have separate budgets and are therefore reported separately. There are also important differences between the various court jurisdictions that have implications for the resources required for their operation and associated expenditure.

The time allocated to different court matters was based on the total number of listing days allocated to the different court jurisdictions (Table 2). This was used to estimate the operating expenditure associated with different list types, as well as the amount of magistrate and VCAT member time spent hearing different matters. A magistrate (1 FTE position, of which 0.9 FTE was allocated to court services) and VCAT member (0.4 FTE position) are assigned to the NJC, but are not funded from the NJC budget. They share responsibility for VCAT matters, although the vast majority of matters were listed before the VCAT member in 2011–12. The salaries and allowances for the NJC magistrate and VCAT member were estimated using the average salary and allowances from across all Victorian court locations. Including these estimates within the cost model was important because they are assumed to act as a proxy indicator for the efficiency of actual court hearings, in that the allocation of magistrates and VCAT members to court listing days is directly related to the number of cases that will be heard.

As shown in Table 2, there were 256 listing days scheduled in the NJC in 2011–12. Forty-one percent of all listing days were dedicated to Magistrates’ Court criminal matters, while a further 41 percent of all listing days were dedicated to VCAT matters (including twilight hearings) and 14 percent of all listing days were dedicated to family violence and personal safety orders (intervention orders) normally heard in a magistrates’ court. A relatively small number of listing days were allocated to Children’s Court and VOCAT matters.

Of the 162 listing days heard by the magistrate, around two-thirds (65%) were dedicated to Magistrates’ Court criminal matters, while a quarter (23%) were dedicated to intervention order hearings. Only six percent of all listing days heard by the magistrate involved VCAT matters. For the VCAT member, around two-thirds (68%) of all listing days (n=94) were assigned to residential tenancy hearings. The distribution of VCAT matters by list type (limited to residential tenancies, guardianship and administration and civil claims in the NJC) was determined using ratios calculated from the published figures in the VCAT annual report (VCAT 2012).

Table 2 NJC listing days, 2011–12

	Magistrate listing days		VCAT member listing days		Total listing days
n	%	n	%	n	%
Magistrates’ Court - Criminal	105	65	0	0	105	41
Magistrates’ Court - Intervention Order	37	23	0	0	37	14
Children’s Court - Criminal	6	4	0	0	6	2
Children’s Court - Intervention Order	1	<1	0	0	1	<1
VOCAT	4	2	0	0	4	1
VCAT - Civil Claims	1	1	29	31	30	12
VCAT - Guardianship and Administration	0	0	2	2	2	1
VCAT - Residential Tenancies	8	5	63	68	71	28
Total	162		94		256

Note: The NJC sits standard court business hours Monday to Fridays, except Thursdays, when VCAT twilight sittings are held from 4:30 pm to 7 pm. There were 237 actual sitting days listed in the 2011–12 financial year. The additional NJC VCAT twilight sittings on Thursdays increase the effective sitting days to 256 days

Note: Community Correctional Order judicial reviews are post-sentence and are not accounted for in court time in above lists (although the judicial reviews would likely be listed on the criminal days)

Note: Total percentages may not add to 100 due to rounding

Source: NJC Cost Model, 2014 [computer file]

The total magistrate salary and allowances, VCAT member salary and NJC operating expenditure for each list type was then calculated by dividing the average salaries reported by the Magistrates’ Court of Victoria (2012) and VCAT (2012) and the total NJC operating expenditure associated with court services by the proportion of court time allocated to each list type (based on listing days). From this, the average magistrate salary and allowances and VCAT member salary, and the average operating expenditure per finalisation for each type of matter dealt with at the NJC, was calculated by dividing the total for each list type by the number of finalised cases in 2011–12. Incidence-based estimates have been calculated, whereby the model includes all costs attributable to cases finalised in the specified period (2011–12), regardless of whether these costs were incurred during this period (Moolenar 2009).

The results are presented in Tables 3 and 4. Based on these estimates, the average magistrate salary and allowances per finalisation ranged from $91 for VOCAT and Children’s Court criminal matters to $202 for intervention order matters (Magistrates’ and Children’s Courts, Table 3). The average across all matter types was $133 per finalisation. The average VCAT member salary was $58 per finalisation, ranging from $49 per finalisation for residential tenancy matters to $98 for guardianship and administration matters (Table 4).

The average operating expenditure per finalisation was $292 in 2011–12. This ranged from $198 per finalisation for Children’s Court criminal matters to $472 for VOCAT matters. The average operating expenditure per finalisation for VCAT matters varied from $240 for residential tenancy matters to $479 for guardianship and administration matters. For all list types, the average magistrate salary and expenses and the VCAT member salary was added to the average operating expenditure to estimate the total cost per finalisation. The average total expenditure per finalisation in the NJC in 2011–12 was $397.

A similar approach was adopted in developing an estimate of the average expenditure per referral for client services. Dividing the total expenditure allocated to client services by the number of referrals in 2011–12, the total client services expenditure per referral was estimated to be $1,848 per referral (Table 5).

Table 3 Estimated magistrates’ salary and allowances, VCAT member salary and operating expenditure, Neighbourhood Justice Centre, by court type, 2011–12

	Mag Court (Criminal)	Mag Court (Int. order)	Ch Court (Criminal)	Ch Court  (Int. order)	VOCAT	VCATa	Total
Caseload
Magistrates’ salary and allowancesc
VCAT member salary
Total expenditure
Finalisations (n)	1,595	342	127	7	79	1,501	3,651
Finalisations (%) – weightedb	42	9	3	<1	5	40	100
Listing days
Court listing days (%)b	40	14	2	<1	3	39	100
Magistrate listing days (%)	65	23	4	<1	2	6	100
VCAT member listing days (%)	-	-	-	-	-	100	100
Total	$197,367	$68,996	$11,544	$1,412	$7,181	$16,853	$303,353
Per finalisation	$124	$202	$91	$202	$91	$129	$133
Total	-	-	-	-	-	$79,158	$79,158
Per finalisation	-	-	-	-	-	$58	$58
Operating expenditured
Total	$430,803	$150,601	$25,199	$3,082	$37,302	$420,176	$1,067,163
Per finalisation	$270	$440	$198	$440	$472	$280	$292
Per finalisation	$394	$642	$289	$642	$563	$344	$397

a: VCAT matters were shared between the magistrate (n=131) and VCAT member (n=1,370). Total for all VCAT matters weighted accordingly

b: Weighted to account for additional time spent preparing for VOCAT matters. Advice from the NJC indicated that registry staff spent, on average 0.2FTE preparing for VOCAT matters. This was equivalent to 2.38 times all other matters

c: Calculated using magistrate listing days

d: Calculated using court listing days

Note: Estimated expenditure and salaries for different court types may not correspond to percentage breakdowns for magistrate and court listing days due to rounding

Note: Estimates for court types with a small number of finalisations should be interpreted with caution as they are more susceptible to outliers. However, given these estimates are based on allocated court time, they are still likely to reflect actual resource allocation

Source: NJC Cost Model, 2014 [computer file]

Table 4 Estimated magistrates’ and VCAT member salary and allowances and operating expenditure, Neighbourhood Justice Centre, VCAT matters by list type, 2011–12

	Civil claims	Guardianship	Residential tenancies	VCAT total
Caseload
Listing days
Magistrates’ salary and allowances
VCAT member salary
Operating expenditure
Finalisations	270	13	1,218	1,501
Finalisations (weighted)	510	26	1,218	1,754
Finalisations (%)	29	1	69	100
Court listing days (%)	11	1	27	39
Magistrate listing days (%)	1	0	5	6
VCAT member listing days (%)	31	2	68	100
Total	$1,525	-	$15,328	$16,853
Per finalisation	$232	-	$123	$128
Total	$24,372	$1,273	$53,512	$79,158
Per finalisation	$93	$98	$49	$58
Total	$122,223	$6,227	$291,726	$420,176
Per finalisation	$453	$479	$240	$280
Total magistrates’ & VCAT member salary & operating expenditure per finalisation	$549	$577	$296	$344

Note:The distribution of VCAT matters by list type (limited to residential tenancies, guardianship and administration and civil claims in the NJC) was determined using ratios calculated from the published figures in the VCAT annual report (VCAT 2012)

Source: NJC Cost Model, 2014 [computer file]

Table 5 Estimated cost of NJC client services, 2011–12

Referrals (n)a	895
Expenditure	$1,654,094
Expenditure per referral	$1,848

a: Excludes clients carried over from previous years

Note: For comparison purposes, the NJC engagement rate adopted the criteria used for the CISP program

Source: NJC Cost Model, 2014 [computer file]

Comparison estimates

The cost model also provides an estimated cost per unit of service for other service delivery models against which the cost of court and client services delivered by the NJC can be compared. This required identifying alternative models of service delivery used elsewhere in Victoria where the NJC does not operate (ie, outside the City of Yarra). These are models that provide a similar type of service and deliver this service to a similar client group, but through a different approach (eg the traditional magistrates’ court model).

In reality, there is no equivalent model of service delivery against which to compare the NJC. The NJC is unique in its design and the only community justice centre operating in Victoria. For this reason, this paper has intentionally avoided using the term ‘benchmark’ to describe comparison cost estimates. Comparison estimates are not presented as the amount that NJC court and client services should cost. They provide a basis against which to compare the relative cost efficiency of providing an alternative model of service delivery.

The methodology used to develop these comparison estimates was the same as the approach used to estimate the average cost of NJC court and client services. This included using consistent inputs and counting rules to maximise comparability between the NJC and comparison estimates. Estimates of the average cost per finalisation differ slightly from those produced by the Productivity Commission in the Report on Government Services, because the operating expenditure and number of finalisations published in the Magistrates’ Court annual report differ from those published in the Report on Government Services (SCRGSP 2013).The Magistrates’ Court of Victoria reports the total operating expenditure and magistrate salary and allowances for all matters (Table 6). The breakdown of these costs by criminal and civil jurisdictions was determined using the number of FTE staff reported in the Report on Government Services (SCRGSP 2013). This was assumed to reflect the distribution of all court staff, and was therefore used to estimate the total operating expenditure and magistrate salaries and allowances for both jurisdictions. Comparison estimates were not available for intervention order or VOCAT matters because the SCRGSP only reports the total number of FTE staff for civil matters.

The total operating expenditure and magistrate salary and allowances for criminal and civil matters was divided by the total number of finalisations in the same period. The same approach was used for the Children’s Court. The results, presented in Table 6, show that the average total expenditure for criminal matters in 2011–12 was $336 for the Magistrates’ Court and $177 for the Children’s Court. It is important to note that there were arrangements in place until July 2012 whereby some of the Children’s Court’s administrative functions were provided by the Magistrates’ Court. This may have implications for the estimated cost per finalisation of both the Children’s Court and the Magistrates’ Court.

The VCAT reports on total operating expenditure, the salaries of full-time and sessional members and expenditure allocation by list type (Table 7). The average cost per finalisation was therefore simply calculated by dividing these figures by the number of finalisations. Not all list types are heard at the NJC; therefore, the estimates in Table 7 are limited to the civil claims ($393 per finalisation), guardianship and administration ($415) and residential tenancies ($208) lists. Separate estimates of the average cost for VCAT member salaries and operating expenditure for individual list types are not available because they are not reported separately in the VCAT annual report.

The total expenditure for selected court support and diversions services was also calculated, using the total court support program expenditure for different programs reported in the Magistrates’ Court of Victoria annual report (Magistrates’ Court of Victoria 2012). As with the cost estimate for the NJC, this was divided by the total number of referrals for each program in 2011–12 to determine the average cost per referral. The results are presented in Table 8. Given the similarities with the NJC in terms of the types of services provided, the Court Integrated Services Program (CISP), which had an average cost of $1,489 per referral in 2011–12, was identified as the most suitable model against which to compare the cost of NJC client services.

Table 6 Estimated magistrates’ salary and allowances and operating expenditure (excluding NJC), 2011–12

	Magistrates’ Court		Children’s Court
	Criminal	Civil	Criminal	Civil
Case load
Magistrates’ salary and allowances
Operating expenditure
Magistrates’ FTE	72.9	34.3	2.0	8.0
FTE (%)	68	32	20	80
Finalised cases	179,136	84,617	11,203	5,841
Total	$24,342,401	$11,453,283	$749,193	$2,996,772
Per finalisation	$136	$135	$67	$513
Total	$35,817,572	$19,335,677	$1,231,891	$4,927,562
Per finalisation	$200	$229	$110	$844
Total magistrates’ salary and allowances and operating expenditure	$336	$364	$177	$1,357

Note: Civil cases reported in the Report on Government Services include civil claims (Magistrates’ Court), family division matters (Children’s Court), family violence and personal safety intervention orders (both Magistrates’ and Children’s Court) and VOCAT matters (Magistrates’ Court). Cost estimates for civil matters are not compared with the NJC because individual estimates for the different court types are not available

Source: SCRGSP 2013; Children’s Court of Victoria 2012; Magistrates’ Court of Victoria 2012

Table 7 Estimated member costs and operating expenditure, VCAT (excl. NJC) , 2011–12

Case load
Member costs
Operational expenditure
Member costs and operating expenditure
VCAT expenditure allocation by List (incl. member costs)
Finalised cases	89,683
Salaries to full-time and sessional members	$14,560,000
Per finalisation	$162
Total	$24,840,000
Per finalisation	$277
Total	$39,400,000
Per finalisation	$439
Civil claims, per finalisation	$393
Guardianship, by finalisation	$415
Residential tenancies list, by finalisation	$208

Source: VCAT 2012

Table 8 Court support and diversions services (Magistrates’ Court), 2011–12

	Court Support Programs Expenditure	Referrals	$ per referral
Court Integrated Services Program (CISP)	$2,830,056	1,900	$1,489
CREDIT Program	$1,111,798	1,409	$789
Bail Support Program	$671,233	1,195	$562

Source: Magistrates’ Court of Victoria 2012

Comparison between the NJC and other models

With equivalent estimates calculated for both the NJC and other models of service delivery, it was possible to then compare the NJC with these other mainstream services and determine the difference in average cost per unit of output for both court and client services. The results of this comparison for court services are presented in Table 9. As mentioned previously, comparison estimates were not available for intervention order or VOCAT matters because the Magistrates’ Court of Victoria only reports the total expenditure for civil matters. Similarly, estimates of the average cost of VCAT member salaries and operating expenditures were not available because they were not reported separately in the VCAT annual report.

The results from the AIC’s cost model showed that, in 2011–12, the average cost of magistrate salaries and allowances per finalised case for Magistrates’ Court criminal matters was lower in the NJC ($124 per finalisation) than it was for mainstream magistrates’ courts ($136 per finalisation). The evaluation of the NJC found that a higher proportion of substantive hearings at the NJC (41%) lasted more than 15 minutes when compared to other court jurisdictions (15 to 18%, Ross et al. 2009). However, the results of the current study suggest the single magistrate and courtroom model may be more efficient in terms of the use of court time, for which magistrates’ salaries and expenditure are proxy measures. This might be due to a reduction in time between matters—that is, in moving from one matter to the next—and the efficiencies that result from having the same magistrate see a matter through from beginning to end. This is supported by further analysis of caseload data, which shows that the magistrate in the NJC dealt with approximately 11 percent more finalised criminal cases (2,726 per FTE position) in 2011–12 than the average magistrate in a normal magistrates’ court (2,457 per FTE position). It might reflect the fact that the NJC is not legislated to hear committals and contested matters, although these represent a very small proportion of all Magistrates’ Court matters. Finally, this result may also reflect savings associated with an increased court appearance rate, which was a key finding of the evaluation of the NJC (Ross et al. 2009), and which may enable the magistrate to deal with cases more efficiently—for example, scheduled court appearances can proceed as planned, time to reschedule appearances is not required and so on. The NJC has a strong focus on procedural fairness and ensuring that accused persons understand the court process and court orders; this has been found to be associated with improved court appearance rates (eg Morgan & Louis 2010).

The findings from the cost model show that the average operating expenditure for criminal court matters in 2011–12 was $70 higher in the NJC than in a normal magistrates’ court ($270 compared with $200). There are several possible explanations for this result.

First, it may be due to the additional level of support offered to accused persons by the NJC and the greater number of participants involved in the court process, which reflects the therapeutic and problem-solving focus of the NJC (Ross et al. 2009). For example, the Neighbourhood Justice Officer—a role unique to the NJC and reflected in the NJC legislation—acts as an important liaison between the magistrate and therapeutic services (Ross et al. 2009), and is estimated to spend at least a third of their time dealing with court-related functions (as distinct from client services and community engagement). Similarly, there are other staff at the NJC who are involved in court functions separate to their normal client services function, as part of the deferred sentencing and judicial monitoring functions of the court. Matters involving deferred sentencing (pre-sentence supervision by the court while under bail conditions) and judicial monitoring (post-sentence supervision while serving some form of court order) are counted within the total number of finalised cases, insofar as they relate to a matter finalised in the relevant counting period. Deferred sentencing and judicial monitoring have been key features of the NJC court model since 2008, and require greater interaction between the magistrate, the accused person, client services support workers and other staff involved with clients. For most other court locations, judicial monitoring was introduced in January 2012 (ie midway through the study period).

Another possible explanation may be that, in order to have a single court that operates independently, it is necessary to have support staff dedicated to that court who might otherwise service multiple magistrates’ courts. While this may contribute to higher costs, it might also be an important factor contributing to improved outcomes for the NJC.

Another explanation may be that there are differences in expenditure for certain cost items that might be due to either individual contracts (as a single court operating in isolation) or due to longstanding arrangements within the Magistrates’ Court of Victoria. For example, the NJC contracts its own security staff whereas, in almost all other Magistrates’ Court locations, Victoria Police provide and pay for police or protective security officers to support the functions of the court. Other cost items that are included in the NJC budget may be included in the budget of the Department of Justice, rather than in the operating expenditure of the Magistrates’ Court of Victoria. A comparison of expenditure on certain budget items, which involved comparing the average expenditure for each of the 53 court locations with the NJC, also revealed some interesting differences. For example, expenditure on property utilities, repairs and maintenance was substantially higher for the NJC than the average expenditure of other court locations—most likely due to contractual differences and economies of scale.

Finally, the results may be due to limitations of the methodology used as part of the current study, which are described at length in a later section of this paper.

Taken together, this means that the average total cost per finalised matter in 2011–12 was $58, or 17 percent higher for the NJC than the mainstream magistrates’ courts. The higher operating expenditure was, in part, offset by the reduced magistrate salary and allowances.

The findings presented in Table 9 also show that the average cost per finalised case for Children’s Court criminal matters was greater in the NJC than in mainstream children’s courts in 2011–12, including for both magistrates’ salaries and allowances and operating expenditure. However, some care needs to be taken with interpreting these estimates. There were a comparatively low number of Children’s Court criminal matters finalised at the NJC in 2011–12 (127), and the estimate is therefore susceptible to outliers. Further, some administrative functions of the Children’s Court of Victoria were provided by the Magistrates’ Court of Victoria until July 2012 (Children’s Court of Victoria 2012), which means that the comparison estimate likely underestimates the cost associated with mainstream children’s courts. This also has possible implications for the estimate for Magistrates’ Court criminal matters.

Finally, the results of a comparison between the NJC and normal VCAT matters indicate that VCAT matters were, on average, more expensive in the NJC than in the normal VCAT (between $88 and $162 per finalisation). This may reflect the additional time spent dealing with individual matters, or it may represent the additional operating expenditure associated with VCAT matters at the NJC, similar to criminal matters (noting that it was not possible to separate VCAT member salaries from operating expenditure for the comparison). This increased cost may also be, in part, accounted for by the NJC’s provision of clinical input and support (eg housing support, mental health support, and alcohol and other drug assessment) to VCAT clients through the multi-jurisdictional operation of the Client Services Team, which is not available through VCAT at other court locations.

The results of a comparison of the NJC client services function and the CISP are presented in Table 10. This shows that the average cost per referral for the NJC ($1,848) was $359, or 24 percent higher than the average cost per referral for the CISP ($1,489). While the CISP is used here as a comparison, there are important differences between the CISP and NJC client services models that will impact on any cost comparison between the two. These include the length of time engaged in the service—NJC clients spend, on average, longer in contact with client services than CISP clients; the range of services offered by the service (as distinct from other agencies to which clients may be referred); the provision of support and input in both the pre- and post-sentence phases; and the proportion of clients who go on to become actively engaged in the service (ie case managed).

Information on engagement rates was available for the current study. According to the Magistrates’ Court of Victoria (2012) annual report, 50 percent of referrals to the CISP were engaged in case management. Using a similar counting rule (referrals resulting in more than one contact, based on advice from CISP staff), data from the NJC showed that 64 percent of referrals were engaged in case management in 2011–12. One possible explanation for the higher engagement rate of the NJC is that NJC client services are largely comprised of services engaged from local service providers through a range of agreements, enabling clients to be directly allocated to ongoing service provision. This is distinct from the approach used for the CISP model, whereby staff are employed by Court Services Victoria (and previously the Department of Justice) to provide case management and intervention and to refer clients to other service providers. The evaluation of the NJC found that the approach used by the NJC allowed for better collaboration between service providers and better integration between the court and client services, which has important benefits for accused persons appearing before the court (Ross et al. 2009).

The average expenditure per engaged referral for both the NJC and the CISP is presented Table 10. This shows that 573 referrals were engaged in case management by NJC court services in 2011–12, at an average cost of $2,888 per engaged referral. During the same period at the CISP, 960 referrals were engaged in case management, at an average cost of $2,979 per engaged referral. The NJC’s expenditure was on average $91, or three percent less than the CISP’s.

Table 9 Average cost per finalised case by jurisdiction, 2011–12

	Magistrate & VCAT member salaries			Operating Expenditure			Total
	NJC	Comp	+/-	NJC	Comp	+/-	NJC	Comp	+/-
Criminal courts
VCAT
Magistrates’ Court	$124	$136	- $12	$270	$200	+ $70	$394	$336	+ $58
Children’s Court	$91	$67	+ $24	$198	$110	+ $88	$289	$177	+ $112
Civil claims	na	na	na	na	na	na	$549	$393	+ $155
Guardianship	na	na	na	na	na	na	$577	$415	+ $162
Residential tenancies	na	na	na	na	na	na	$296	$208	+ $88

na=not available

Comp=Comparison court model

+/- refers to the average difference in cost per finalisation between the NJC and comparison court model

Source: NJC Cost Model, 2014 [computer file]

Table 10 Average cost per referral by jurisdiction, 2011–12

	NJC	CISP	+/-
Expenditure	$1,654,094	$2,830,056	-
Referrals	895a	1,900a	-
Expenditure per referral	$1,848	$1,489	+ $359
Engagement rateb	64%	50%	-
Engaged referrals	573	950	-
Expenditure per engaged referral	$2,888	$2,979	- $91

a: Excludes clients carried over from previous years

b: For comparison purposes, the NJC engagement rate adopted the criteria used for the CISP program

Note: Average cost per engaged referral is used for comparison purposes only. It does not account for the resources expended on referrals that were not engaged in case management and therefore should not be interpreted as the actual cost per engaged referral

+/- refers to the average difference in cost per referral between the NJC and CISP

Source: NJC Cost Model, 2014 [computer file]

Limitations

There are several important limitations to be taken into consideration in interpreting the findings of this cost comparison study. First, as has been made evident throughout this paper, there were limitations to the information that was available for both the NJC and comparison court and service delivery models. For this reason, a number of assumptions about the distribution of time and resources have been made; these are described in the relevant sections above. This process identified a number of gaps in the data held by the NJC, particularly in relation to client services, some of which have been rectified as part of the process of completing this research.

Similarly, estimating the costs associated with a criminal justice program using a top-down approach can be challenging, particularly when relying on published budgets that do not necessarily provide the necessary breakdown of cost items. This was particularly true for comparison estimates, which relied on information on expenditure published in relevant annual reports. There were some differences between the financial reports for the NJC and comparison models in terms of individual cost items; nevertheless, every attempt has been made to ensure that the same cost items have been accounted for in both the NJC and comparison estimates.

The average cost per unit of service can overestimate the cost for each additional unit of service because it does not distinguish between fixed and variable costs. For this reason the use of marginal costs is preferred in economic analysis, particularly in cost-benefit analysis (Henrichson & Galagano 2013). It was not possible to determine fixed and variable costs from the operating expenditure reported in the annual reports used for the current study; care therefore needs to be taken in using the average cost reported in this paper to estimate the financial impact of increased caseloads—that is, in case of additional cost should the number of finalised cases or referrals increase. Nevertheless, calculating the average cost is useful for the purposes of comparing the costs associated with different service types, which was the main purpose of the current study.

Related to this point, the average cost per unit of service for comparison court models was based on the total output and expenditure across all Victorian courts. As well as reflecting the average cost for each individual case, the estimates used in this paper also reflect the average cost across the 53 different Magistrates’ Court locations across Victoria. The variation in costs between these court locations, and how the NJC compares to individual courts, could not be assessed as part of the current study.

While not strictly a limitation, it should be noted that the comparison court and service delivery models are not directly comparable to the NJC because of important differences in the way the NJC operates and adheres to the community justice model. Further—design and operational differences aside—it was not possible to determine whether the cases and referrals in the NJC shared similar characteristics with other court models and the CISP; and if not, what implications this might have for the estimated cost of each. This and the other limitations described above should be taken into consideration when comparing estimates.

Conclusion

This paper presents the estimated average cost associated with court and client services delivered through the NJC in Victoria, alongside estimates for other models of service delivery provided through the normal Magistrates’ Court of Victoria, the Children’s Court of Victoria and the VCAT. The results of this cost model show that in 2011–12 the average cost of NJC client services was approximately three percent lower than those of the CISP, once engagement rates were taken into account. The results also provide evidence that the NJC’s court services are more efficient in the processing of criminal matters in the Magistrates’ Court jurisdiction. Overall, the average cost of NJC court services was higher than that of mainstream court processes. This paper offers a number of possible explanations for this finding, including that additional resources are required to operate an effective community justice model that delivers improved outcomes for accused persons, the Victorian Government and the wider community.

Importantly, and limitations aside, while this study provides some indication as to the cost efficiency of the NJC and also the cost implications associated with providing an alternative model of service delivery, it does not attempt to draw any conclusions as to the benefits associated with this additional expenditure. Ross (2015) found that the NJC has a positive impact on reoffending rates of accused persons as well as better compliance rates for community orders—both of which are likely to result in savings to the NJC and the Victorian criminal justice system. Further work is therefore required to develop a cost-benefit model for the NJC, based on findings from the more recent outcome evaluation. What the current study does offer is a much more valid and robust estimate of the cost inputs for the NJC, which should provide a much better foundation for informed discussions about future resourcing. It also provides a baseline for assessing any future improvements to the NJC and the impact of these improvements on the overall efficiency of the community justice model.

Acknowledgements

The authors gratefully acknowledge the Neighbourhood Justice Centre for funding this research and, in particular, Carolyn Dryden and Cameron Wallace for their assistance in providing access to data and other important information.

References

All URLs correct at August 2015

• Booth L, Altoft A, Dubourg R, Goncalves M & Mirrlees-Black C 2012. North Liverpool community justice centre: analysis of reoffending rates and efficiency of court process. London: Ministry of Justice
• Children’s Court of Victoria 2012. Children’s Court annual report 2011/12. Melbourne: State Government of Victoria
• Clear TR, Hamilton Jr JR & Cadora E 2011. Community justice, 2nd ed. London: Routledge
• Clear TR & Karp DR 2000. Community justice: A conceptual framework. In Friel CM (ed), Boundary changes in criminal justice organisations. Washington, DC: US Department of Justice, Office of Justice Programs, National Institute of Justice: 323–368
• Courts and Tribunals Unit 2009. Victoria’s Neighbourhood Justice Centre: Current initiatives paper 1. Department of Justice (Victoria)
• Department of Treasury and Finance 2013. Economic evaluation for business cases technical guidelines. Melbourne: Department of Treasury and Finance
• Joliffe D & Farrington DP 2009. Initial evaluation of reconviction rates in community justice initiatives. Research Summary 9/09. London: Ministry of Justice.
• Lee C, Cheesman D, Rottman R, Swaner S, Lambson M, Rempel & Curtis R 2013. A community court grows in Brooklyn: a comprehensive evaluation of the Red Hook Community Justice Center. Williamsburg: National Center for State Courts
• Magistrates’ Court of Victoria 2012. Magistrates’ Court annual report 2011/12. Melbourne: State Government of Victoria
• Moolenaar D 2009. Modelling criminal justice system costs by offence: Lessons from the Netherlands. European Journal of Criminal Policy and Research 15: 309–326
• Morgan A & Louis E 2010. Evaluation of the Queensland Murri Court: Final report. Technical and background paper series no. 39. Canberra: Australian Institute of Criminology
• Neighbourhood Justice Centre nd. Strategic Framework 2010–2013. Melbourne: Department of Justice (Victoria)
• Ross S 2015. Evaluating neighbourhood justice: Measuring and attributing outcomes for a community justice program. Trends & Issues in Crime and Criminal Justice no.499.
• Ross S, Halsey M, Bamford D, Cameron N & King A 2009. Evaluation of the Neighbourhood Justice Centre, City of Yarra: Final report. Melbourne: Department of Justice (Victoria)
• Steering Committee for the Review of Government Service Provision (SCRGSP) 2014. Report on government services 2014. Canberra: Productivity Commission
• Steering Committee for the Review of Government Service Provision (SCRGSP) 2013. Report on government services 2013. Canberra: Productivity Commission
• Victorian Civil and Administrative Tribunal (VCAT) 2012. VCAT annual report 2011/12. Melbourne: State Government of Victoria
• Victims of Crime Assistance Tribunal (VOCAT) 2012. VOCAT annual report 2011/12. Melbourne: State Government of Victoria

About the author

Anthony Morgan is the Research Manager of the Crime Prevention and Evaluation Research team at the Australian Institute of Criminology

Dr Rick Brown is the Deputy Director (Research) at the Australian Institute of Criminology