Supporting the District Criminal Court

Media release

The Auditor-General for New South Wales, Margaret Crawford, released a report today on whether the Department of Communities and Justice (the department) effectively supports the efficient operation of the District Criminal Court system.

The audit found that in the provision of data and technology services, the department is not effectively supporting the efficient operation of the District Criminal Court system. The department has insufficient controls in place to ensure that data in the system is always accurate.

The department is also using outdated technology and could improve its delivery of technical support to courts.

The audit also assessed the implementation of the Early Appropriate Guilty Pleas reform. This reform aims to improve court efficiency by having more cases resolved earlier with a guilty plea in the Local Court. The audit found that the department effectively governed the implementation of the reform but is not measuring achievement of expected benefits, placing the objectives of the reform at risk.

The Auditor-General made seven recommendations to the department, aimed at improving the controls around courts data, reporting on key performance indicators, improving regional technical support and measuring the success of the Early Appropriate Guilty Pleas reform. 

Executive summary

The District Court is the intermediate court in the New South Wales court system. It hears most serious criminal matters, except murder, treason and piracy. The Department of Communities and Justice (the Department) provides support to the District Court in a variety of ways. For example, it provides security services, library services and front-desk services. This audit examined three forms of support that the Department provides to the District Court:

  • data collection, reporting and analysis - the Department collects data from cases in its case management system, JusticeLink, based on the orders Judges make in court and court papers
  • technology - the Department provides technology to courts across New South Wales, as well as technical support for this technology
  • policy - the Department is responsible for proposing and implementing policy reforms.

Recent years have seen a worsening of District Court efficiency, as measured in the Productivity Commission's Report on Government Services (RoGS). Efficiency in the court system is typically measured through timeliness of case completion. There is evidence that timeliness has worsened. For example, the median time from arrest to finalisation of a case in the District Court increased from 420 days in 2012–13 to 541 days in 2017–18.

As a result, the government has announced a range of measures to improve court performance, particularly in the District Court. These measures included the Early Appropriate Guilty Pleas (EAGP) reform. One of the objectives of EAGP is to improve court efficiency, which would be achieved by having more cases resolve with a guilty plea in the Local Court.

This audit assessed whether the Department of Communities and Justice effectively supports the efficient operation of the District Criminal Court system. We assessed this with the following lines of inquiry:

  • Does the Department effectively collect, analyse and report performance information relevant to court efficiency?
  • Does the Department effectively provide technology to support the efficient working of the courts?
  • Does the Department have effective plans, governance and monitoring for the Early Appropriate Guilty Pleas reform?

The audit did not consider other support functions provided by the Department. Further information on the audit, including detailed audit criteria, may be found in Appendix two.

Conclusion
In the provision of data and technology services, the Department is not effectively supporting the efficient operation of the District Criminal Court system. The Department has insufficient controls in place to ensure accurate data in the District Criminal Court system. The Department is also using outdated technology in significant numbers and could improve its delivery of technical support to meet agreed targets.
The Department effectively governed the implementation of the Early Appropriate Guilty Pleas reform. However, it is not ensuring that the benefits stated in the business case are being achieved, placing its objectives at risk.
The impact of inaccurate court data can be severe, and the Department does not have sufficient controls in place to ensure that its court data is accurate. Recent Bureau of Crime Statistics and Research reviews have identified data inaccuracies, and this demonstrates the Department needs strong controls in place to ensure that its court data is accurate.
The Department does not have a policy for data quality and has not formally assigned responsibility for data quality to any individual or branch. The Department also does not have a data dictionary outlining all the fields in its case management system. While the Department validates the highest risk items, such as warrants, to ensure that they are accurate, most data is not validated. The Department has recently commenced setting up a data unit for the Courts, Tribunals and Service Delivery branch. It is proposed that this unit will address most of the identified shortcomings.
The Department did not provide timely technical support to the court system in 2017 and is using outdated technology in significant numbers. The Digital and Technology Services branch of the Department had agreed a Service Level Agreement with the rest of the Department, outlining the expected speed of technical support responses. The branch did not meet response times in 2017. Performance improved in 2018, though DTS fell short of its targets for critical and moderate priority incidents. Critical incidents are particularly important to deal with in a timely manner as they include incidents which may delay a court sitting.
Requests for technical support rose significantly in 2018 compared to 2017, which may be related to the number of outdated pieces of technology. As at April 2019, the whole court system had 2,389 laptops or desktop computers outside their warranty period. The Department was also using other outdated technology. Outdated technology is more prone to failure and continuing to use it poses a risk of court delays.
The Department is not measuring all the expected benefits from the Early Appropriate Guilty Pleas reform, placing the objectives of the program at risk. The Early Appropriate Guilty Pleas business case outlined nine expected benefits from the reform. The Department is not measuring one of these benefits and is not measuring the economic benefits of a further five business case benefits. Not measuring the impact of the reform means that the Department does not know if it is achieving its objectives and if the reform had the desired impact.

1. Key findings

The Department has insufficient controls in place to ensure accurate data

The Department has few controls in place to ensure that its data is accurate. It also does not know how often data is entered incorrectly or where there are persistent data entry errors. There are numerous risks arising from inaccurate court data, the most severe of which are that an individual receives the wrong bail conditions or is placed in detention for an incorrect length of time, either longer or shorter than their sentence. This audit did not seek to determine the extent to which this data was accurate, but instead considered the controls which the Department has in place to prevent these inaccuracies.

Recent Bureau of Crime Statistics and Research reviews have identified data inaccuracies and this demonstrates that it is critical for the Department to have strong controls in place to ensure that data is entered accurately.

We found that the following controls supporting data quality are missing:

  • there is no policy governing data quality
  • responsibility for data quality is not assigned
  • a data dictionary is not available
  • data validation only occurs for some high‑risk items such as warrants.

The Department does not ensure that its staff enter data into JusticeLink consistently. In addition to the risks noted above, inaccurate and inconsistent data limit the Department's ability to undertake data analysis to understand the performance of the court system or identify inefficiencies.

The Department has recently commenced setting up a data unit for the Courts, Tribunals and Service Delivery branch. It is proposed that this unit will have stewardship over data quality, the production of reports and data analysis. The Department has also advised that it intends for this unit to develop and manage a data quality framework.

The Department only has one performance indicator for its court support activities

The Department only has one performance indicator relating to the support it provides in the District Court. To ensure that it can evaluate its own performance, the Department needs performance indicators which relate to its own court support functions, including its impact on efficiency.

The Department measures the cost per finalised matter in the District Court, which it has some control over because cost reductions on the Department's side will be reflected in a lower total cost per finalisation. There are other performance indicators the Department could use to measure its own activities, such as the number of full time equivalent staff per 1,000 finalisations, the accuracy of case files and the accuracy of data entered in JusticeLink.

The Department is using outdated technology in significant numbers

As of April 2019, the whole court system had 2,389 laptops or desktop computers outside their warranty period. The court system also had 786 printing devices outside their normal warranty period. The Department advised that many of its audio transcription machines are also out of date. Using outdated technology poses a risk to the court system as older equipment may be more likely to break down, potentially delaying courts or slowing down court services.

Technical support requests from all courts or court services increased significantly from 4,379 technical support incidents in 2017 to 9,186 in 2018. The number of technical issues is likely impacted by using outdated technology, which is more prone to failure.

The Department did not meet all court technical support targets in 2017 and 2018

The Digital and Technology Services (DTS) branch of the Department had agreed a Service Level Agreement (SLA) with the court system, both court support services and the courts themselves, outlining the expected speed of technical support responses. Incidents are classified as low, moderate, high or critical priority. DTS failed to reach its target for all incident priorities in 2017. Performance improved in 2018, though the Department fell short of the target for critical and moderate priority incidents. Critical incidents include incidents which may delay a court sitting.

Technical support response times were marginally slower in regional areas in 2018, particularly for moderate and high priority incidents. All technical support staff are based in Sydney, meaning the Department is reliant on third party vendors being available to provide support in regional areas, which are not bound by the same SLA as the Department. There is no provision for in‑house on‑the‑ground technical support in regional areas, exposing regional courts to the risk of significant delays in having technical issues resolved.

The Department rolled out Audio‑Visual technology to improve efficiency by reducing defendant transport

As part of a series of business cases, the Department installed 244 Audio‑Visual Link (AVL) facilities in courts between 2014 and 2018, either for the first time in those courts or as replacements for older models of AVL. As a result, 90 per cent of courtrooms in courthouses which sit as District Courts now have AVL facilities.

AVL facilities allow court users, including defendants held in prison on remand, to appear remotely either in court or at other points of the court process, such as when conferencing with their lawyer. For this reason, AVL facilities can result in cost savings for the justice system as prisoners do not need to be transported from their prison to the court and back. While the Department is not directly responsible for the usage of AVL, it is the Department's responsibility to ensure that AVL facilities are available to the court to take advantage of any potential efficiencies that can occur.

The Department is not following good practice in benefits realisation management for Early Appropriate Guilty Pleas

Benefits realisation management is the process of organising and managing a project or program so that potential benefits are actually achieved. The Department is not following good practice in benefits realisation management in the implementation of the Early Appropriate Guilty Pleas (EAGP) reform.

The Department calculated nine benefits for the EAGP business case. It is not measuring one of these at all and it is not measuring the economic benefits derived from a further five business case benefits. Not measuring the expected benefits stated in the business case means that the Department does not know if the reform is achieving what it was designed to achieve. This also means that the Department does not know if it must take corrective action to ensure that the program achieves the stated benefits. These two things put the overall program benefits at risk.

The Department has not assigned responsibility for the realisation of each benefit stated in the business case. Good practice is to assign a responsible party for each benefit at the business unit level, who is responsible for monitoring the benefit and taking corrective action as appropriate. Not having individual responsibility for this means that responsibility and accountability are not clearly defined and corrective action may not be undertaken.

2. Recommendations

By June 2020, the Department of Communities and Justice should:

  1. develop a strategic framework for improving and managing court data, including:
    • a strategy for court data
    • a policy governing court data quality
    • assigning formal responsibility for data quality
    • proposed actions to improve the quality of court data, including the development of a data dictionary for JusticeLink
    • proposed actions to improve the use of courts data
  2. formalise the responsibilities of the courts data team and ensure that it is appropriately resourced to carry out its responsibilities
  3. align its internal benefits realisation guidance with the NSW Government's Benefits Realisation Management Framework
  4. measure all benefits stated in the Early Appropriate Guilty Pleas business case.

By December 2020, the Department of Communities and Justice should:

  1. investigate additional key performance indicators to measure its support activities, such as the accuracy of its data entry and its own contribution to the efficiency of the court
  2. report performance in the court system annually against its key performance indicators in its annual report
  3. evaluate options for improving the delivery of technical support to regional and rural courts and commence implementation of the preferred option.

1. Introduction

1.1 The New South Wales criminal justice system

New South Wales court jurisdictions and hierarchy

Legal cases in New South Wales can be divided into two jurisdictions: civil and criminal. Civil cases can be brought to redress a private wrong, such as the breach of a contract. Criminal cases decide whether a person who has been charged is guilty of a crime or other offence. This audit considered only the support provided to the criminal jurisdiction.

The New South Wales criminal court system comprises a number of different courts arranged in a hierarchy. A simplified version of this hierarchy can be seen in Exhibit 1.

Image description - hierarchy of criminal court. Please email communications@audit.nsw.gov.au for a more accessible version
Exhibit 1: The New South Wales criminal court hierarchy

The roles of some of the courts in the criminal jurisdiction are as follows:

  •  Local Court ‑ most criminal cases begin in the Local Court, which can deal with matters of less complexity and 'commit' more complicated matters to the District Court or Supreme Court. The Local Court can sentence up to two years for one offence, or a total of five years cumulatively in limited circumstances.
  • Children's Court ‑ hears and determines matters where a child or young person has been charged with a criminal offence.
  • District Court ‑ largely hears matters which have been committed from the Local Court and appeals against Local Court decisions. Most matters which have been committed will go to trial in the District Court. The District Court can deal with most serious criminal matters, except murder, treason and piracy, which are dealt with by the Supreme Court.
  • Drug Court ‑ ensures that drug dependant offenders receive treatment.
  • Supreme Court ‑ the highest court in the New South Wales hierarchy. It hears the most serious criminal matters, such as murder or offences where the prosecution seeks a sentence of life imprisonment. 

The Court of Criminal Appeal, which hears appeals from the District Court, the Drug Court and the Supreme Court sits above these. At the top of the Australian hierarchy is the High Court of Australia, which can hear appeals from the State's court system and the decisions of which are binding on all courts throughout Australia.

This audit focused on the support provided to the criminal jurisdiction of the District Court, but due to the overlapping nature of some of these support services, it has been necessary at times to consider services which overlap with the other courts in the hierarchy. This is because it is common for the District Court to sit in the same building as the Local Court and sometimes to utilise the same courtrooms.

New South Wales justice agencies

There are a range of government agencies which play a key role in the New South Wales justice system. The Department of Communities and Justice provides support services to the court system. It is discussed in more detail below. Other key agencies include:

  • NSW Police Force ‑ Police officers can issue Court Attendance Notices, which set out the offence a defendant is to be charged with and the time and location when the defendant must attend court. Police investigate alleged crimes and put together briefs of evidence for cases. Police Prosecutors also handle most crimes in the Local Court.
  • Office of the Director of Public Prosecutions (ODPP) ‑ ODPP is the independent prosecuting authority of New South Wales. It is responsible for the prosecution of all serious offences committed against the laws of the State. This means that ODPP deals with most criminal matters in the District Court and higher.
  • Legal Aid NSW (Legal Aid) ‑ Legal Aid is responsible for delivering and coordinating legal services, including defence representation, for disadvantaged people. This includes representing people in criminal cases in the District Court. 

The activities of these agencies have not been considered as part of this audit, however it has sometimes been necessary to comment on activities which they may have been involved in. This is particularly the case for the Early Appropriate Guilty Pleas reform, discussed below.

The role of the Department of Communities and Justice

The Department of Communities and Justice (the Department) was formed on 1 July 2019, as part of the ‘Administrative Arrangements (Administrative Changes ‑ Public Service Agencies) Order 2019’. The Courts, Tribunals and Service Delivery (CTSD) branch of the Department is responsible for delivering support services to the court system. Before 1 July 2019, the Courts and Tribunal Services (CaTS) branch of the former Department of Justice delivered these services.

CTSD provides support services to the 164 courts and tribunals across New South Wales. Registries are a key element of these support services. Each Local Court has a registry attached which assists the District Court if the District Court sits in that location. Registry staff perform a range of public‑facing functions, such as witnessing court documents and assisting with applications for apprehended violence orders. In addition, registries provide support services to the court to which they are attached. For example, registries record Judicial decisions in the Department's case management system, JusticeLink. The Department is responsible for analysing the information held in JusticeLink to identify potential efficiencies. These data functions are discussed in Chapter two.

In addition to supplying registry services, the Department provides support to courts in other ways. The Department, largely through the Information and System Design branch, provides technology and technical support to the court system. This is discussed in Chapter three. The Department is also responsible for proposing policy reforms and potential efficiency gains for the system. The audit examined one reform; Early Appropriate Guilty Pleas. This reform is discussed further below and in Chapter four.

The Department's roles must be understood in the context of providing support services to the Judiciary. The Judiciary are a separate branch of government and it is important that their decision making is independent, and that the Department does not interfere with this independence. This means that the Department is limited in what it can do in some cases because it does not wish to breach Judicial independence. Some decisions, such as case management in the court system, operate according to Judicial decision‑making, therefore the Department is limited in the direct influence it can have over the efficiency of the court system. Exhibit 2 delineates in simple terms some key roles of the Department and the Judiciary.

Exhibit 2: Some roles of the Judiciary and Department
Judiciary's role Department's role
Put in place court procedures and practices Inform individuals about court procedures and practices
Make orders in court Record the Judicial officer's orders in JusticeLink
Decide how court lists will be managed Manage court lists in line with Judicial directions
Decide when and how technology is used in court in line with legislation Provide technology to the courts and Judiciary
Interpret legislation Inform policy and legislative development
Source: Audit Office of NSW analysis.

This audit did not examine the conduct, processes, procedures or operations of courts that are under the purview of the Judiciary.

Performance of the court system

The timeliness of proceedings is a key indicator of efficiency in the court system. The Bureau of Crime Statistics and Research (BOCSAR) and the Productivity Commission's Report on Government Services (RoGS) provide measures for the court system's timeliness. Examples include the median time taken from arrest to finalisation of a case, the percentage of cases finalised within 12 or 24 months and the number of cases in the court backlog. It is important to note that the Department is not wholly responsible for the performance against these indicators because, as noted above, the Judiciary are responsible for many aspects of the court system.

While performance in the criminal jurisdiction of the Local and Supreme Courts has been stable in recent years, some indicators of court performance in the District Criminal Court have worsened. The median time from arrest to finalisation of a case in the District Court has risen from 420 days in 2012–13 to 541 days in 2017–18. Preliminary data from 2018–19 indicates that this has now plateaued at 527 days. This can be seen in Exhibit 3. The percentage of cases that are dealt with in the District Court in under 24 months has fallen from 64.8 per cent in 2015–16 to 55.4 per cent in 2017–18.

A court's backlog is a measure of how many cases remain as part of the court's active pending caseload. The New South Wales District Court saw its pending trial caseload rise from 1,026 in June 2011 to 1,574 in July 2019 with a peak of 2,110 in June 2017. The government's response to the District Court backlog is discussed below.

There are other ways to measure efficiency in the court system. RoGS includes measures for the cost per case, the number of Judicial officers per 1,000 finalisations and the number of full time equivalent staff per 1,000 finalisations.

Please email communications@audit.nsw.gov.au for an accessible version of this graph
Exhibit 3: Median time from arrest to finalisation of a case in the District Court
Source: Bureau of Crime Statistics and Research.

1.2 Responses to the court system's performance

Government's responses to the District Court backlog

The government has implemented several responses to the increase in the District Court backlog. These include:

  • the appointment of five new District Court Judges in 2016 and seven more in 2018
  • the use of short‑term 'call‑overs', which aim to clear a large number of cases in targeted courts in the course of one or two week
  • increasing the number of offences which can be dealt with in the Local Court
  • implementing the Early Appropriate Guilty Pleas reform, discussed below.

The Early Appropriate Guilty Pleas reform

One of the responses to the District Court backlog is the implementation of the Early Appropriate Guilty Pleas (EAGP) reform. This reform has a wide impact on stakeholders across the justice sector. One of the objectives of EAGP is to improve court efficiency, which would be achieved by having more cases resolve with a guilty plea in the Local Court. The reform commenced in April 2018. Given that there is a time lag between the commencement of cases and finalisation in the District Court, it is currently too early to determine the impact of EAGP.

EAGP consists of five main elements:

  • Early disclosure of evidence from NSW Police Force ‑ Police provide a simplified summary of the evidence to the prosecution and defence early in the process. This is to ensure that both sides are informed about the strength of evidence from an early stage.
  • Early certification of what the accused is going to be charged with to minimise changes ‑ a senior prosecutor reviews the evidence and confirms the charges at an early stage to ensure that charges are not withdrawn or changed late in the process. This is to prevent defendants from delaying their guilty plea in the hope that they will later be charged with a lesser crime.
  • Mandatory criminal case conferencing between the prosecutor and accused's representation ‑ this provides an opportunity for early dispute resolution and for the defendant to plead guilty early.
  • Changes to Local Court case management ‑ Magistrates perform a changed case management function.
  • More structured sentence discounts ‑ discounts from a guilty plea are provided based on how early a plea is entered, with the highest discount provided to those who plead guilty in the Local Court.

These reform elements are anticipated to have three key effects:

  • accelerate the timing of guilty pleas
  • increase the overall proportion of guilty pleas
  • decrease the average length of contested trials.

Further anticipated benefits of the reform are outlined in Exhibit 12 in Chapter 4.

2. Data collection and reporting

2.1 Data collection

Accurate data is critical in the court system. There are numerous risks arising from inaccurate data, the most severe of which are that an individual receives the wrong bail conditions or is placed in detention for an incorrect length of time, either longer or shorter than their sentence. Given these risks, it is important for the Department to have controls in place to ensure it manages data well. Quality data also allows the Department to conduct analysis to improve its understanding of the operations of the court system, and to identify potential efficiencies.

The Department's case management system duplicates paper case files

The Department collects court data in its case management system JusticeLink, which keeps a record of decisions made in court. These court decisions are made by a judicial officer, such as a Magistrate or Judge, and then their decision is recorded in JusticeLink. In addition to operating as a case management system, the Department uses information from JusticeLink for performance reporting and data analysis, making it a key data collection system.

The court system is largely paper‑based, meaning that a physical court file is maintained for each case. The orders from Judicial officers are kept in a physical form as part of this paper file. Registry staff or Judge's Associates transcribe these orders into JusticeLink. As a result, JusticeLink relies on manual data entry. Manual data entry carries a risk that data will be entered incorrectly and requires a good control environment to ensure that data is accurate.

The Department has insufficient controls in place to ensure accurate data

The Department has few controls in place to ensure that JusticeLink data is accurate and does not know how often data is entered incorrectly or where there are persistent data entry errors. Incorrect data in the court system can have severe impacts on individuals and it is important for the Department to have strong controls in place to ensure that data is accurate. Given the Department shares its information with other justice sector agencies, inaccurate data can also impact the activities of other organisations. In addition, the Department relies on accurate data in its system to ensure that it can conduct data analysis, which may include identifying potential efficiencies.

This audit did not seek to determine the extent to which this data was accurate, but instead considered the controls which the Department has in place to prevent these inaccuracies.

The Department relies on third parties to inform it if information is missing from JusticeLink. The Bureau of Crime Statistics and Research (BOCSAR) uses an automated process to check the information in JusticeLink to determine if key information is missing or if contradictory information has been entered. Once BOCSAR identifies these issues, it sends a list to each registry with identified issues. Registry staff revisit the original case files and upload the information onto JusticeLink but correcting this can sometimes take several months and some errors must be re‑reported as a result. BOCSAR conduct monthly reviews of data. BOCSAR only look for missing or contradictory information, meaning that they will not pick up all errors in JusticeLink.

BOCSAR has identified data issues in each of its reviews, demonstrating that there are inaccuracies in the JusticeLink data. Common data issues identified by BOCSAR include papers not being uploaded, pleas not being recorded and the total duration of a sentence not matching the start and end date entered. The Department does not keep records on the percentage of cases which have data issues identified, though the number of cases where issues are raised is low. As noted above, BOCSAR does not check all data for accuracy meaning the Department does not know how widespread data errors are beyond what BOCSAR check.

A recent review of data quality in the civil jurisdiction of the District Court also identified other data inaccuracies. The findings of this review align with our consultation with various stakeholders which identified data quality issues in JusticeLink for the criminal jurisdiction. This demonstrates a need to improve the controls around the quality of data in JusticeLink.

The Department also relies on other justice partners to inform it if key information has not been entered. Bail decisions in court impact on the work of Corrective Services NSW and the NSW Police Force. The Department relies on Corrective Services NSW to alert registries if it has not received information on an individual's bail outcomes. Corrective Services NSW will also alert the Department if it identifies inaccurate sentences. Relying on these organisations does not mitigate the risk of missing information, but only moves it to a different part of the justice sector.

While these processes allow the Department to learn of any missing data, there are few controls over the accuracy of data in JusticeLink. Exhibit 4 demonstrates the Department's controls against some better practice controls for managing data quality.

Exhibit 4: Better practice controls over data accuracy
Control Department's result
Training and manuals are available yes
A data dictionary is available no
Responsibility for data quality is assigned no
There is a policy governing data quality no
Validation of data, such as data audits, occurs no
Source: Controls adapted from NSW Government Standard for Data Quality Reporting, October 2015. Audit Office of NSW analysis.

 

The Department provides training courses to staff and manuals explaining how to perform each function in JusticeLink. Staff also report informal on the job training occurs in registries, particularly for new staff or staff who have persistent errors in BOCSAR data audits. There is no data dictionary for JusticeLink defining the fields and data tables. One of the purposes of a data dictionary is to provide a single source of truth for all fields in a database.

The Department has not assigned responsibility for the quality of its data to any individual or branch. The Department also does not have a policy governing its data quality setting out responsibility, controls and expectations for the quality of data. The Department does not audit the quality of its data. Conducting audits of data would allow the Department to understand how often data is entered inaccurately and identify persistent problems. The Department advised it does not do this due to resource limitations.

Twice per year, the Department undertakes stocktakes of all the cases in JusticeLink to ensure they have the right stage listed and to update cases which can be finalised. Registry staff check the status of each case and update any inaccuracies. This means the Department has some assurance over the quality of this part of its data only. The Department also double checks some of the highest‑risk items, such as warrants, to ensure that these have been filled out correctly, but this is not done for most forms of documentation. This means that there are few controls in place to ensure that data is entered correctly in the first instance for most forms of documentation. 

A lack of assurance over data quality can lead to issues with program‑level data collection and analysis. Exhibit 5 demonstrates this for the Early Appropriate Guilty Pleas reform.

Exhibit 5: Data collection problems in Early Appropriate Guilty Pleas

Poor data quality in JusticeLink has limited the ability of the Department to collect useful information for the Early Appropriate Guilty Pleas (EAGP) reform. This may impact on BOCSAR's ability to measure the effectiveness of the reform. The reform introduced two new forms of documentation, charge certificates and case conference certificates. The Department made changes to JusticeLink to allow registry staff to indicate whether these documents had been filed as part of the case and have provided user guides to staff explaining what they are required to do for the changes. BOCSAR was initially planning to identify which cases were EAGP cases based on whether a charge certificate had been filed. However, persistent under‑reporting of charge certificates meant that BOCSAR had to change how it identified EAGP cases.

In March 2019, BOCSAR advised the reform's Steering Committee that a 'high proportion' of EAGP cases were being committed to the District Court without proper data being recorded in relation to these documents. This included 19 cases where data not permissible under law was entered. The Department added a prompt to JusticeLink in March 2019 to ask staff whether these documents had been lodged. However in June 2019 BOCSAR noted this had not addressed the identified problems and incorrect data was still being entered.

BOCSAR advised the Steering Committee that the ongoing problems with data quality may impact the quality of its post‑implementation evaluation. If the evaluation is undertaken with poor quality data, there is a risk that the evaluation will not accurately reflect the impact of the reforms. The Department advised that further JusticeLink changes are planned for December 2019 which should address data accuracy concerns.

Source: Audit Office of NSW analysis.

The Department does not ensure consistent and timely data entry

In addition to accurate data, it is important for information to be entered consistently between registries to ensure that data is collected in the same way across the State. Consistent data can assist with data analysis. The Department does not ensure that information is collected consistently. The Department also does not check that information is uploaded in a timely manner.

As noted above, the Department provides staff with manuals and training. These manuals provide some guidance to staff designed to ensure consistent data entry. The Department does not monitor how often this advice is followed for consistency.

The Department does not have a data dictionary which defines the fields and data tables in JusticeLink. Data dictionaries can ensure that there is a single source of truth for all fields and tables in a database. A 2017 review of JusticeLink data in the civil jurisdiction of the District Court recommended that the Department develop a data dictionary.

The Department provides advice to staff on how quickly data must be entered after it is decided in the courtrooms. Information which is critical for other justice partners, such as Corrective Services NSW, must be uploaded more quickly than lower‑priority information. The Department does not monitor how often registry staff comply with these time standards and is reliant on these partners to advise when data has not been lodged.

JusticeLink allows free text entry, making inconsistent data entry more likely

Some information in JusticeLink is collected through staff entering information in free text entry boxes. Free text entry can make data more difficult to analyse due to differences in how staff phrase information and the degree of detail input by staff. Free text entry is difficult to analyse and for this reason it reduces the Department's ability to analyse the efficiency of the courts.

The Department provides some standardised text which staff can select to fill the free text entry box, however there is no requirement to use this. This means that even in situations where the text could be filled by one of the Department's standardised text options, staff are still able to manually enter the information. The Department does not monitor how often the standardised text is used and how often it could have been used but was not.

The Department advised that free text entry is important for its purposes due to the complexity of laws and the need to ensure that their systems can account for rare or unusual cases.

JusticeLink does not capture all data useful for analysis and identifying efficiencies

In 2015, the Bureau of Crime Statistics and Research (BOCSAR) released the report 'Trial court delay and the NSW District Criminal Court', in which the authors noted there were some statistics which did not exist, but which would be important for diagnosing causes of trial delay in the District Court. The most important of these were:

  • the percentage of trials that proceed on the date they are first listed
  • the cause of any failure to proceed when listed
  • trial duration broken down by offence type.

BOCSAR advised that since 2015 only the third of these has become available and that this indicator requires additional manual data collection at the registry level.

Stakeholders advised the audit team that there are other pieces of information that could be helpful, such as whether a person was legally represented. JusticeLink can capture if a defendant is represented and the name of their representation, however it is not mandatory for staff to enter this. This means that information on whether defendants were legally represented may be inconsistently captured.

2.2 Data reporting and analysis

As the organisation responsible for court data collection, the Department plays a role in reporting performance both within the Department and externally. This includes reporting information about efficiency, even in situations where the Department is not wholly responsible for the activities being reported. The Department is also responsible for data analysis to identify potential areas of efficiency, as the policy arm of the justice sector.

The Department only has one performance indicator for its court support activities

The Department only has one performance indicator relating to the support it provides in the District Criminal Court. The Department operates in an environment where many aspects of performance are outside their control. To ensure it can evaluate its own performance, the Department needs performance indicators which relate to its performance, including its impact on efficiency.

The Department's indicators for the District Criminal Court are shown at Exhibit 6. These performance indicators are the ones the Department uses internally or which it reports to NSW Treasury for budget purposes. Of these, the only performance indicator the Department can impact directly is 'cost per finalisation', as savings at the registry level can cause an improvement in this result. The Department advised, however, that its ability to impact on costs is limited because each Judge has a certain number of staff and other associated costs which the Department must supply.

Exhibit 6: The Department's District Criminal Court performance indicators and targets
Performance indicator 2018–19 Target
Finalisations 11,460
Clearance ratio (ratio of finalisations to registrations) 100%
Cost per finalisation $7,200
Judicial officers per 100,000 population 0.7
Rate of clearance within 12 months (non-appeal) 90%
Source: Department of Communities and Justice, 2019.

The Department reports additional information in the Productivity Commission's 'Report on Government Services' which could be useful for creating a measure for the Department's efficiency. The Department provides information on the number of full time equivalent staff per 1,000 finalisations as part of this framework. Not setting a target for this indicator means the Department is not determining if their performance is in the desirable range or not.

Another potential indicator which the Department could use is found in the International Consortium for Court Excellence's Global Measures of Court Performance. This document contains advice for how to set up an indicator for 'court file integrity', which relates to the accuracy and completeness of the information in a case file. By measuring and setting a target for this indicator, the Department could understand the effectiveness of its own functions.

The BOCSAR data error reports provide an opportunity for the Department to report on the completeness of its JusticeLink data. The Department could set a target for the percentage of reports with identified incomplete data and report against this target. This could be expanded to include any future data validation activities undertaken by the Department.

In addition, the Department measures other key performance indicators but has no targets attached to them. An example is the 'Pending trials' indicator. The Department's target for this is to reduce it to a level where the clearance target can be consistently met. The Department provides no commentary on what this level may be, nor does it set an explicit target. The Department advised that it is not possible to set targets for this indicator as stakeholders and other jurisdictions do not have a clear idea of what constitutes desirable performance.

The Department reports some performance indicators internally but does not report its performance in its annual report

Each month, the Department produces a performance report for each registry. The Department produces other performance reports for the District Court and a specific performance report for the District Court backlog. While the Department produces these internal reports, the Department does not report its performance in its annual report.

The monthly performance reports produced for each registry contain key information such as the number of new registrations and disposals, the speed of disposals and the number of pending trials. Registry staff consulted by the audit team found these reports helpful, though some noted that having more commentary as part of their report would be helpful for explaining the results to key stakeholders. JusticeLink cannot automatically produce these monthly reports. Producing individual registry reports requires manual processes which must be performed each month.

The performance reports produced monthly for the District Court and the District Court backlog report on a number of key performance indicators with some commentary included.

The Department does not report on the court's performance in its annual report. Instead, the Department relies on other documents to report on its performance. There are three main reports for this:

  • Productivity Commission's 'Report on Government Services' (RoGS)
  • BOCSAR's annual statistical report
  • District Court Annual Review. 

All three report aspects of court performance. The Department provides data to facilitate reporting, though the results differ in these cases as they all use different counting rules. The Department uses RoGS results to report to NSW Treasury for budget purposes.

While using these reports means that the Department's performance is publicly available, none of the above are tied to the Department's performance indicators. By reporting against its performance targets in its annual report, agencies are able to demonstrate and explain progress against key performance indicators to parliament and the public. The Department advised that it does not report externally due to a risk of misinterpretation as a result of the different counting rules used internally by the Department and externally in RoGS. 

The Department's data collection systems are disconnected from each other, impacting on its ability to analyse performance

In JusticeLink, the Department collects only information written down by the Judicial officer and does not collect information beyond that. This means that staff only enter what is necessary for case management and the Department may miss an opportunity to collect information beyond this, such as whether a person was represented.

The Department collects some data outside JusticeLink which can be useful for understanding the performance of the court system. The Department measures how long each court has been sitting. The Department also captures data regarding when an Audio‑Visual Link (AVL) system has been booked.

While these pieces of information could help with analysing aspects of court performance, these data sets are not easily connected to the information in JusticeLink. This limits the Department's ability to undertake detailed data analysis combining JusticeLink data and data held in other systems. The reason for this disconnect is that JusticeLink was designed solely as a case management system.

The Department is developing a court‑specific data unit

In recent years, the Department did not have a dedicated branch responsible for data analysis in the courts, which would be useful for understanding the causes of inefficiency in the court system. The Courts, Tribunals and Service Delivery (CTSD) branch only undertakes data analysis concerning the District Court on an ad hoc basis, such as when requested by the Chief Judge. CTSD rely on two other branches to provide data analysis: BOCSAR and the Performance and Analysis Branch (PAB). Both BOCSAR and PAB perform data analysis roles across the entire justice sector and this limits their ability to provide support to CTSD.

PAB have worked with CTSD to develop a costing model. This model allows the user to model the cost of different policy and legislative reforms to the court system. This is a useful tool for understanding the cost of policy changes. When putting forward potential changes to policy or legislation, the Department can create Justice Impact Assessments (JIAs), outlining the cost to the whole justice system based on changes to one part of the system. The Department has used the costing model in two recent JIAs to demonstrate the impact of policy changes on the cost of the courts system. The Department can use this model to inform business case development for policies aimed at increasing efficiency.

Following an internal review of its data capability, the Department has recently commenced setting up a data unit for CTSD. It is proposed that this unit will have stewardship over data quality, the production of reports and data analysis. The Department has also advised that it intends for this unit to develop and manage a data quality framework. The Department has not yet formalised the responsibilities of this unit.

3. Technology and technical support

The Department is responsible for providing technology to the courts, which can improve the efficiency of court operations by making them faster and cheaper. The Department is also responsible for providing technical support to courtrooms and registries. It is important that technical support is provided in a timely manner because some technical incidents can delay court sittings and thus impact on court efficiency. A 2013 Organisation for Economic Co‑operation and Development report emphasised the importance of technology and digitisation for reducing trial length.

While the Department may provide technology to the courts, they are not responsible for deciding when, how or if the technology is used in the courtroom.

The Department is using a significant amount of outdated technology, risking court delays

As of April 2019, the whole court system had 2,389 laptops or desktop computers out of warranty, 56.0 per cent of the court system's fleet. The court system also had 786 printing devices out of their normal warranty period, 75.1 per cent of all printers in use. The Department also advised that many of its court audio transcription machines are out of date. These machines must be running for the court to sit and thus it is critical that they are maintained to a high degree. The then Department of Justice estimated the cost of aligning its hardware across the whole Department with desired levels at $14.0 million per year for three years. Figures for the court system were not calculated but they are likely to be a significant portion of this figure.

Using outdated technology poses a risk to the court system as older equipment may be more likely to break down, potentially delaying courts or slowing down court services. In the court system throughout 2018, hardware made up 30.8 per cent of all critical incidents reported to technical support and 41.9 per cent of all high priority incidents. In addition, 16.2 per cent of all reported issues related to printing devices or printing.

From 2017 to 2018, technical support incidents from courts or court services increased. There were 4,379 technical support incidents in 2017, which increased significantly to 9,186 in 2018. The Department advised that some outside factors may have contributed to this increase. The Department was rolling out its new incident recording system throughout 2017, meaning that there would be an under‑reporting of incidents in that year. The Department also advised that throughout 2018 there was a greater focus on ensuring that every issue was logged, which had not previously been the case. Despite these factors, the use of outdated technology has likely increased the risk of technology breakages and may have contributed to the increase in requests for technical support.

Refreshing technology on a regular basis would reduce the risk of hardware failures and ensure that equipment is covered by warranty.

The Department did not meet all court technical support targets in 2017 and 2018

The Digital and Technology Services branch (DTS) was responsible for providing technical support to the courts and the Courts and Tribunal Services branch prior to July 2019. DTS provided technical support in line with a Service Level Agreement (SLA) with the Department. In 2017, DTS did not provide this support in a timely manner. Performance improved in 2018, though DTS fell short of its targets for critical and moderate priority incidents. Exhibit 7 outlines DTS' targets under the SLA.

Exhibit 7: Digital and Technology Services' Service Level Agreement
Priority Target resolution time Target percentage in time (%)
1. Critical 4 hours 80
2. High 1 day 80
3. Moderate 3 days 85
4. Low 5 days 85
Source: Department of Communities and Justice, 2019.

Critical incidents are particularly important for the Department to deal with in a timely manner because these include incidents which may delay a court sitting until resolved or incidents which impact on large numbers of staff. Some of the critical incidents raised with DTS specifically stated that they were delaying a court sitting, often due to transcription machines not working. High priority incidents include those where there is some impact on the functions of the business, which may in turn affect the efficiency of the court system. High priority incidents also include those directly impacting on members of the Judiciary. 

This audit examined DTS' performance against its SLA in the 2017 and 2018 calendar years across the whole court system, not just the District Court. The total number of incidents, as well as critical and high priority incidents, can be seen in Exhibit 8.

Exhibit 8: Number of incidents in 2017 and 2018
Priority 2017 2018
All 4,379 9,186
1. Critical 48 91
2. High 128 315
Source: Audit Office of NSW analysis of Department of Communities and Justice data, 2019.

The Department's results against its SLA in 2017 and 2018 are shown in Exhibit 9.

Graph showing Digital and Technology Services' results against SLA by incident priority
Exhibit 9: Digital and Technology Services' results against SLA by incident priority
Source: Audit Office of NSW analysis of Department of Communities and Justice data, 2019.

DTS failed to reach its target for all priority incidents in 2017. Performance improved in 2018, though the Department fell short of the target for critical and moderate priority incidents.

The Business Information Services (BIS) unit provides business support to courts covering less serious issues, including support for JusticeLink. BIS can triage more serious issues it receives and escalate them to DTS. The Department measures results for technical support provided to JusticeLink users across all responsible branches, however BIS does not measure its own performance beyond a raw count of the number of requests for technical support it receives. There is an opportunity for BIS to better understand the support it provides to the court system by formalising an SLA with the Department and measuring its performance against this.

Technical support times were marginally slower in 2018 in regional areas

Technical support response times were marginally slower in regional areas in 2018 for most incident categories compared to metropolitan or greater metropolitan areas. Results against the Department's SLA in 2018 are shown at Exhibit 10, broken down by location. Location has been determined according to the Court Services region which each courthouse belongs to. Under this structure, there are five regions:

  • Metropolitan
  • Greater Metropolitan (primarily Western Sydney)
  • Hunter-North
  • West/South-West
  • Illawarra-South.

The three regional areas have been combined for the analysis in Exhibit 10. Caution should be taken when interpreting the 'critical' incident data, as this is only a small sample size for each region, making results more volatile.

Graph showing DTS' results against SLA by incident priority and location - 2018
Exhibit 10: DTS' results against SLA by incident priority and location - 2018
Source: Audit Office of NSW analysis of Department of Communities and Justice data, 2019.

Exhibit 10 demonstrates that response times are marginally slower in regional areas, particularly for moderate and high priority incidents. Technical support times in the metropolitan and greater metropolitan areas are better than regional areas, but remain below the SLA for some incident categories.

Results among the three Court Services regions outside Sydney were similar in 2018. This indicates that slower technical support is not isolated to one particular region, but occurs across all regional areas.

During this period, DTS had seven staff to deliver technical support across the State. All seven staff are based in various locations across Sydney. All staff in the Courts, Tribunals and Service Delivery branch across New South Wales are able to receive remote technical support, either over the phone or via video conferencing. Some regional staff we interviewed said this was often useful, and most incidents could be addressed this way. This is reflected in Exhibit 10, which indicates that the results for 'low' priority incidents are similar across metropolitan and regional areas.

However, some more complicated issues posed a challenge for DTS as demonstrated by the results for 'high' priority incidents in Exhibit 10. Given that all support staff are based in Sydney, the Department is reliant on third party vendors being available to provide support in regional areas, which are not bound by the same SLA as the Department. The Department does not have provisions for in-house on-the-ground technical support in regional areas. This poses a risk that if a serious issue arises, third party vendors may not resolve the issue within the SLA's timeframe and courts may be delayed. The Department has early plans to improve technical support delivery in regional areas, but these plans are not well advanced.

The Department does not systematically consult with court users in each registry regarding technology needs

The Department does not systematically consult with frequent users of courts to ensure that it is meeting their technology needs. The Department advised that it consults with court users at a senior level on an ad hoc basis, however it does not systematically consult with court users at a registry level. Each registry conducts court users' fora, which allow for court users to inform the Department of technology needs, however the feedback from these fora are not systematically captured. This means that the Department cannot know if it is meeting the needs of core users. The Department advised that it does not consult with users at a local level because it cannot afford to have customised technology in each courthouse.

The Department meets with the heads of the Local, District and Supreme Courts twice per year in a technology forum convened by the Judiciary. These allow the Department to gauge, at a high level, the needs of Judicial officers. However, there is no formal consultation with Judicial officers at a registry level which would allow the Department to understand the needs of its stakeholders and inform future technology investments required.

The Department has rolled out Audio-Visual technology in significant numbers

As part of a series of business cases, the Department installed 244 Audio-Visual Link (AVL) facilities in courts between 2014 and 2018, either for the first time in those courts or as replacements for older models of AVL. As a result, 90 per cent of courtrooms in courthouses which sit as District Courts now have AVL facilities.

AVL facilities allow court users, including defendants held in prison on remand, to appear remotely either in court or at other points of the court process, such as when conferencing with their lawyer. For this reason, AVL facilities can result in cost savings for the justice system as prisoners do not need to be transported from their prison to the court and back. While the Department is not directly responsible for the usage of AVL, it is the Department's responsibility to ensure that AVL facilities are available to the court to take advantage of any potential efficiencies that can occur.

A BOCSAR evaluation in 2018 identified significant cost savings for Corrective Services NSW in Local Courts with AVL facilities. It is unclear what the total dollar value savings attached to the Department's broader AVL roll-out are, but they are likely to be significant. AVL facilities are useful not just as a potential efficiency gain, but can also allow the Department to make its reforms operate more effectively, as demonstrated for EAGP in Exhibit 11.

Exhibit 11: The use of AVL in EAGP

AVL is widely used as part of the Early Appropriate Guilty Pleas (EAGP) reform for case conferencing. Case conferencing involves the prosecutor and the accused's legal representative meeting to discuss the case and negotiate. The prosecutor and the accused's legal representative can either meet in person or via AVL to discuss the case. The accused is also able to attend the case conference via AVL.

For this reason, the Department rolled out 13 additional AVL facilities in Legal Aid offices and 12 additional AVL facilities in correctional centres. By having these facilities in correctional centres, an accused person on remand can appear remotely at case conferences held in Legal Aid offices, meaning that they would not have to be transported to the location where the conference was being held.

Source: Audit Office of NSW analysis. 

4. Early Appropriate Guilty Pleas

The Early Appropriate Guilty Pleas (EAGP) reform consists of five main elements:

  • early disclosure of evidence from NSW Police Force to the prosecution and defence
  • early certification of what the accused is going to be charged with to minimise changes
  • mandatory criminal case conferencing between the prosecutor and accused's representation
  • changes to Local Court case management
  • more structured sentence discounts.

More detailed descriptions of each of these changes can be found in the Introduction. These reform elements are anticipated to have three key effects:

  • accelerate the timing of guilty pleas
  • increase the overall proportion of guilty pleas
  • decrease the average length of contested trials.

Improving District Court efficiency is one of the stated aims of EAGP, which would be achieved by having more cases resolve in the Local Court and having fewer defendants plead guilty on the day of their trial in the District Court. The reform commenced in April 2018 and it is too early to state the impact of this reform on District Court efficiency.

The Department is responsible for delivering EAGP in conjunction with other justice sector agencies. They participated in the Steering Committee and the Working Groups, as well as providing the Project Management Office (PMO).

The Department is not measuring the economic benefits stated in the EAGP business case

The business case for EAGP listed nine quantifiable benefits which were expected to be derived from the achievement of the three key effects listed above. The Department is not measuring one of these benefits and is not measuring the economic benefits for five more, as shown in Exhibit 12.

Benefit Economic benefit (over ten years) Being measured?
Accelerated timing of guilty pleas $54.6m yellow circle with minus in the center
Increased guilty plea rate $90.7m yellow circle with minus in the center
Decreased average trial length $27.5m yellow circle with minus in the center
A reduction in the delay of indictable matters proceeding to trial N/A check circle mauve
Increase the number of finalised matters per annum N/A check circle mauve
Reduction of the current backlog of criminal trials in the District Court N/A check circle mauve
Reduction in bed pressure on the correction system due to reduced
average time in custody
$13.7m Exclamation circle red
Productivity improvements due to reduction in wasted effort $53.3m yellow circle with minus in the center
Bankable cost savings due to jury empanelment avoided $2.5m yellow circle with minus in the center

 

Exhibit 12: The Department's measurement of quantifiable benefits
Key check circle mauve Measuring yellow circle with minus in the center Not measuring economic benefit Exclamation circle red Not measuring
Source: Audit Office of NSW analysis.

While it is too early to comment on the overall impact of EAGP, better practice in benefits realisation involves an ongoing effort to monitor benefits to ensure that the reform is on target and determine whether any corrective action is needed.

The Department is measuring the number of finalised matters per annum and while the Department is not measuring the reduction in the backlog as part of this program, this measure is reported as part of the Department's internal reporting framework. The Department is not monitoring the reduction in delay of indictable matters proceeding to trial directly as part of this reform, but this does form part of the monthly Operational Performance Report which the Department sends to the EAGP Steering Committee.

The Department is not monitoring any of the economic benefits stated in the business case. These economic benefits are a mixture of bankable savings and productivity improvements. This amounts to a total of $242.3 million over ten years which was listed in the business case as potential economic benefits from the implementation of this reform against the total cost of $206.9 million over ten years. The Department is collecting proxy indicators which would assist in these calculations for several indicators, but it is not actively monitoring these savings. For example, the Department is monitoring average trial length, but is not using this information to calculate economic benefits derived from changes in trial length.

The Department is also not collecting information related to the average length of custody as part of this program. This means that it is unable to determine if EAGP is putting less pressure on the correctives system and it is not possible for the Department to calculate the savings from this particular benefit.

While stakeholders are optimistic about the impact of EAGP, not measuring the expected benefits stated in the business case means that the Department does not know if the reform is achieving what it was designed to achieve. Further, the Department does not know if it must take corrective action to ensure that the program achieves the stated benefits. These two things put the overall program benefits at risk.

The Department has not assigned responsibility for the realisation of each benefit, potentially risking the success of the program

The Department has not assigned responsibility for the realisation of each benefit stated in the business case. The Department holds the Steering Committee responsible for the realisation of all benefits. Benefits realisation is the process which ensures that the agency reaches benefits as stated in the business case. Assigning responsibility for benefits realisation to the Steering Committee rather than individuals is not in line with good practice.

Good practice benefits realisation involves assigning responsibility for the realisation of each benefit to an individual at the business unit level. This ensures there is a single point of accountability for each part of the program with knowledge of the benefit and the ability to take corrective action if it looks like that benefit will not be realised. This responsibility should sit at the operational level where detailed action can most easily be undertaken. The role of a Steering Committee in benefits realisation is to ensure that responsible parties are monitoring their benefits and taking appropriate corrective action.

The Department advised that it believes the Steering Committee should have responsibility for the realisation of benefits due to the difficulty of attributing the achievement of each benefit to one part of the reform alone. Given the Steering Committee meets only quarterly, it is not well placed to take action in response to variances in performance.

A BOCSAR evaluation is planned, however data errors make some of the information unreliable

BOCSAR are planning to undertake an overall evaluation of EAGP which is planned for release in 2021. Undertaking this evaluation will require high quality data to gain an understanding of the drivers of the reform. However, data captured throughout the first year of EAGP has proven unreliable, which may reduce the usefulness of BOCSAR's evaluation. These data issues were discussed in Exhibit 5 in Chapter 2, above. Access to accurate data is vital for conducting any program evaluation and inaccurate data raises the risk that the BOCSAR evaluation will not be able to provide an accurate evaluation of the impact of EAGP.

In addition to the BOCSAR evaluation, the Department had plans for a series of 'snapshot' evaluations for some of the key elements of the reform to ensure that they were operating effectively. These were initially delayed due to an efficiency dividend which affected EAGP. In August 2019, the Department commissioned a review of the implementation of several key success factors for EAGP.

There was clear governance throughout the implementation of EAGP

The implementation stage of EAGP had clear governance, lines of authority and communication. The Steering Committee, each Working Group and each agency had clear roles and responsibilities, and these were organised through a Project Management Office (PMO) provided by the former Department of Justice. The governance structure throughout the implementation phase can be seen at Exhibit 13.

The Steering Committee was established in December 2016 and met regularly from March 2017. It comprised senior members of key government agencies, as well as the Chief Judge and the Chief Magistrate for most of the duration of the implementation period. The Steering Committee met at least monthly throughout the life of the program. The Steering Committee was responsible for overseeing the delivery of EAGP and making key decisions relating to implementation, including spending decisions. The Chief Judge and the Chief Magistrate abstained from financial decisions. The Steering Committee updated the governance and membership of the Steering Committee as appropriate throughout the life of the reform.

Infographic showing EAGP governance structure and roles from March 2017 to April 2018
Exhibit 13: EAGP governance structure and roles March 2017 to April 2018
Source: Adapted by Audit Office of NSW from Department of Justice, 2017. Note that JSP stands for Justice Strategy and Policy, part of the then Department of Justice.

Each Working Group had a specific area of responsibility which required cross-agency interaction. For this reason, each Working Group had representatives from several agencies who were able to work collaboratively. The Working Groups had detailed terms of reference and objectives, as well as timelines for key deliverables. Each individual agency also had detailed workplans and objectives.

Each Working Group and agency had an implementation plan and reporting template, which they used to report to the PMO weekly. The PMO monitored these weekly reports and aggregated them to report to the Steering Committee at the monthly meetings. The EAGP Working Group comprised of senior representatives from each agency and met with the PMO weekly to support the PMO in their monitoring of implementation.

In May 2018, following implementation of EAGP, the Steering Committee made changes to the program governance. The Steering Committee has moved to quarterly meetings and a single Working Group, comprising representatives from a wide range of stakeholders. This Working Group also meets quarterly to align with the Steering Committee.

The Department consulted widely with stakeholders and involved them in the Steering Committee and Working Groups

EAGP was a wide-ranging reform to the criminal justice system and impacted on the operations of many government and non-government stakeholders, most notably NSW Police Force, ODPP and Legal Aid. The Department consulted widely with stakeholders across government and outside government before the business case went to Cabinet and throughout the implementation phase of the reform. Stakeholders continue to be involved in the Steering Committee and Working Groups.

In February 2016, prior to the business case going to Cabinet, the Department created a Steering Group which reported to the Court Reform Steering Committee and the Criminal Justice Transformation Board. The Steering Group aimed to inform the development of EAGP's business case based on consultation with stakeholders. Membership in this group included senior staff from key stakeholders. The Steering Group consulted with other stakeholders, including those outside government, to provide input for the business case.

Consultation with government stakeholders included discussions of the resourcing impact of EAGP. The resourcing impact on some stakeholders, such as ODPP, has been significant and it was important for the Department to understand this prior to the creation of the business case.

Stakeholders were widely involved in the implementation of the reform. The Steering Committee consisted of key stakeholder agencies, as well as the Chief Magistrate and Chief Judge throughout most of the implementation phase. Partner agencies were also involved in the Working Groups, which gave them a way to impact the reform in detail. Consultation informed the EAGP legislation, as well as key policies and procedures throughout the implementation phase.

Stakeholders remain involved in EAGP implementation. The current Steering Committee includes key stakeholders from across government, as well as members of the Judiciary including the Chief Magistrate and Chief Judge. The Working Group which supports the Steering Committee also contains a wide range of stakeholders, including stakeholders from outside government. This provides a forum for the Department to hear the needs of the whole sector more directly.

The Department created implementation plans but reporting against these was inconsistent

All agencies and each Working Group for the reform had an implementation plan. These plans outlined the tasks which must be completed, key milestone dates, project risks and interdependencies with other agencies or Working Groups. Implementation plans were largely met, though reporting on variances against these implementation plans to the PMO was inconsistent.

Each agency or Working Group reported to the PMO on a weekly basis against their implementation plan, giving each activity a rating of 'on track', 'at risk' or 'off track'. The reporting of these activities was inconsistent. In some cases, tasks which were several months overdue were marked as 'on track', creating a risk that progress was seen to be more positive than was actually the case. The Department advised that the PMO held a weekly meeting with each agency and Working Group at which it raised any concerns it had with the reporting and provided advice about appropriate levels to set tasks at. Implementation plans were not always updated in a timely manner following these discussions.

The PMO provided weekly updates to the Attorney-General's Office based on the agency and Working Group reports. From August 2017 until the launch of the program in April 2018, this also included additional commentary on most items listed as 'at risk' or 'off track' and items which were marked as 'on track' or 'completed' where the PMO believed that they were not. This commentary included comments on where the Department disagreed with the agency's rating, as well as an outline of mitigation which was occurring to bring the item back on track.

The PMO kept the Steering Committee well informed about the program

The PMO reported to the Steering Committee each month on risks, interdependencies between agencies and progress against the budget. They produced a dashboard which set out all this key information on a page, as well as information on the progress made by each agency and Working Group. Each agency or Working Group could provide an additional briefing to the Steering Committee to explain key items or when seeking a decision. This allowed the Steering Committee to effectively perform its role of overseeing the implementation of the reform.

As with reporting against the implementation plans, discussed above, agencies and Working Groups had an inconsistent approach to risk reporting. Risk reporting was focused largely on those risks particular to the agency or Working Group. As a result, in November 2017 and April 2018, the PMO produced its own overview of the high-level risks to the program to better inform the Steering Committee. This overview was not provided on a monthly basis throughout the implementation phase, which would have allowed the PMO to better inform the Steering Committee. However, the PMO continued to report monthly as part of the dashboard about the highest risk items.

Appendices

Appendix one – Response from agency
 
Appendix two – About the audit 

Appendix three – Performance auditing 

 

Copyright Notice

© Copyright reserved by the Audit Office of New South Wales. All rights reserved. No part of this publication may be reproduced without prior consent of the Audit Office of New South Wales. The Audit Office does not accept responsibility for loss or damage suffered by any person acting on or refraining from action as a result of any of this material.

Parliamentary Reference: Report number #329 - released 18 December 2019