Follow up of Regulating Gambling and Liquor

Tabled: 28 November 2019

2 Licensing industry participants

In the 2017 audit, we found that VCGLR had made limited progress in reorganising the licensing division, training staff and providing improved guidance material to move towards a more risk-based approach to licensing activities.

VCGLR could not demonstrate that it properly examined and assessed all licensing applications in line with legislative provisions before approving them. Those weaknesses were more significant for liquor applications because VCGLR accepted applicants' information at face value. It relied heavily on both the honesty of applicants, and on Victoria Police and potential objectors to raise issues on the suitability of applicants, amenity issues or social harms associated with those applications.

In 2017, we identified instances where VCGLR granted licences without fully identifying and assessing the suitability of applicants and their associates, including cases where applicants had not provided complete information on their associates and past criminal convictions.

The 2017 audit recommended that VCGLR improve liquor licensing processes, adopt a risk-based model for licensing, undertake ongoing checks of liquor licensees, assess licence applications against all relevant legislative considerations, improve the transparency of net detriment decisions and conduct better reporting on licensing activities.

2.1 Liquor licensing process

VCGLR must assess and determine liquor licence applications in accordance with the LCR Act. VCGLR determined 16 227 liquor licence applications in 2018–19—a decrease of 207 from the previous year.

Recommendation 1

We recommend that VCGLR amend its liquor licensing process to:

  • require applicants to provide evidence to show that all directors and associates have been disclosed
  • document its assessments against all relevant legislative considerations when determining applications, including applicant suitability, amenity issues, and risks of misuse and abuse of alcohol.

Key points underpinning past audit recommendation

  • Licence application assessments did not address LCR Act requirements.
  • VCGLR relied solely on others to identify matters against applications.
  • Reasoning behind decisions needed to refer to statutory provisions in more detail.
  • VCGLR did not adequately confirm the accuracy of applicant information.
  • Decision records examined were inconsistent, did not comprehensively address key matters specified in legislation or contain sufficient information to demonstrate why a liquor licence was granted.

Completed.PNG

Progress

VCGLR has implemented the audit recommendation (Figure 2A).

 

Figure 2A
Progress implementing recommendation 1

VCGLR has:

  • analysed the business structures of its applicants to determine the evidence it may obtain. It now requires higher risk-entities, such as companies, to provide additional information on their directors as part of the application process
  • strengthened processes, guidance and training related to applicants providing evidence that they have disclosed all directors and associates
  • updated licence assessment processes, licence application forms, assessment templates (for high‑volume licence types) and decision sheets to enable assessments against all relevant legislation
  • established explicit links between licence application forms, assessment criteria in assessment templates and individual sections in the LCR Act
  • updated relevant internal guidance and training programs, communicated changes to staff and promoted those changes to external parties.

VCGLR has not:

  • addressed the issue of incomplete or incorrect assessment templates. We discuss this issue in Section 2.3.

Lightbulb.PNG

VCGLR could improve implementation by:

  • developing more liquor licence application assessment templates for high-volume licence types. Software with system controls that restrict data entry values can prevent a range of data entry errors and incomplete records being saved.

Source: VAGO.

2.2 Ongoing checks of liquor licensees

Approximately 11 000 companies hold a liquor licence in Victoria. Section 103 of the LCR Act requires a licensee to inform VCGLR within 14 days, in writing, of changes to directors and associates.

Recommendation 2

We recommend that VCGLR undertake ongoing checks of liquor licensees to ensure company changes have been disclosed in line with the Liquor Control Reform Act 1998.

Key points underpinning past audit recommendation

  • VCGLR had insufficient processes to ensure licence holders informed it of changes in company directors and associates after a licence is granted.
  • Compliance officers undertaking routine inspections were not systematically checking for changes in directors or associates due to a lack of consistently applied procedures.
  • VCGLR could conduct random checks on licensees to better monitor whether company details have changed.
  • VCGLR could encourage compliance with the LCR Act by alerting licensees that they may be randomly checked.

InProgress.PNG

Progress

While VCGLR has implemented a system to undertake checks of liquor licensees, the very small sample of checks it has made means it does not provide sufficient assurance of licensee compliance.

VCGLR now undertakes an annual monitoring program of liquor licence holders based on an annual random sample of licensees. Where it finds breaches, it refers them to the compliance team for further investigation and action.

However, Figure 2B shows that the marked decrease in the non-compliance rate reported by VCGLR was based on small samples that are not statistically valid. Without statistically significant sample sizes, it is not possible to determine if non-compliance has reduced.

Figure 2B
Annual monitoring program sample size

Year

Sample size

Non-compliance rate

2016–17

48 (0.4%)

38 (79%)

2017–18

80 (0.7%)

20 (25%)

2018–19

50 (0.5%)

6 (12%)

Source: VAGO, based on VCGLR's data.

Figure 2C summarises progress against recommendation 2 and further suggestions for improvement.

Figure 2C
Progress implementing recommendation 2

VCGLR has:

  • implemented an annual monitoring program of liquor licence holders. Breaches identified through the monitoring program are referred to the compliance team for further investigation and action
  • published on its website the requirement for licensees to notify it of changes to directors and associates.

VCGLR has not:

  • alerted licensees either directly, or publicised on the VCGLR website, that it may conduct random checks to confirm licensees' company details. Promoting the monitoring program would raise awareness of the need to report changes and that VCGLR is actively looking for licensees that are not doing so
  • incorporated checks of associates into the monitoring program. Checks of associates is a more complex and time-consuming task. While VCGLR is not required to monitor checks of associates, doing so provides assurance that licensees are complying with their disclosure requirement.

Lightbulb.PNG

VCGLR could improve implementation by:

  • reviewing the annual monitoring program's sampling methodology. At present, a sample is too small for VCGLR to draw conclusions about the overall compliance rate of licensees. VCGLR should either:
    • increase the sample size to a statistically valid number so it can use results to make observations on the overall compliance rates of licensees, or
    • use a targeted sampling methodology to focus the monitoring program on licensees who present a higher risk to community safety.

VCGLR advised that it will adopt a targeted sampling methodology for its 2019–20 annual monitoring program.

Source: VAGO.

2.3 Guidance on assessment of licence applications

VCGLR administers and makes decisions on 59 types of gambling and liquor licences. Most applications relate to a smaller number of high-volume licence types. Within VCGLR's licensing division, all licensing officers assess and make decisions on the high-volume licence types. Low-volume licence types are handled by licensing officers with specific experience and expertise. Clear guidance on licensing helps keep the application assessment process transparent and consistent with the intent of the GR Act.

Recommendation 3

We recommend that VCGLR improve its guidance on assessment of licence applications, particularly for uncontested applications, and ensure licensing officers use this guidance.

Key points underpinning past audit recommendation

  • VCGLR's 'Wiki' pages were inconsistent and incomplete.
  • VCGLR needed to offer licensing officers more guidance for decisions on uncontested matters to help staff consider relevant information, such as applicant suitability.

While guidance was available for contested applications, licensing officers commented that they did not rely on it, and asked colleagues instead.

InProgress.PNG

Progress

As shown in Figure 2D, VCGLR is implementing the audit recommendation, but incomplete training data makes it difficult to ensure that training is completed by the right staff at the right time. Our review of application assessments found missing data, which prevents scrutiny of how officers made risk ratings or assessment decisions.

Figure 2D
VCGLR progress implementing recommendation 3

VCGLR has:

  • improved and expanded guidance on assessing licence applications, including definitions of contested and uncontested applications in line with the LCR Act
  • emphasised legislative requirements within the assessment process, including the consideration of suitability, amenity, misuse or abuse of alcohol and knowledge of the LCR Act
  • mapped licence application assessment decision records sheets to LCR Act requirements. Licensing officers now retain source documents and reference them in their decision records
  • delivered specific training on the new guidance, decision record sheets and general training focused on better decision-making. The training materials cover key topics and focus on objects of the LCR Act
  • implemented a program of quality assurance reviews. VCGLR reviews a random sample of determined applications, with findings reported to VCGLR's senior executives.

VCGLR has not:

  • maintained a record of all training undertaken by licensing division staff members. The lack of training data limits VCGLR's ability to monitor training activities and consider issues such as the coverage and timeliness of training. This may compromise the competency and consistency of the assessment process. VCGLR advised that it plans to move all training records into its learning management system
  • addressed the issue of incomplete or incorrectly completed assessment templates. We found multiple instances. Specifically:
    • Some template cells were blank or only held instructions on how to populate the cell with the correct data. It was therefore unclear what information was used to support the assessment.
    • While the licence application risk rating is recorded in the licensing information and communications technology (ICT) system, the details that led to the risk rating were often incomplete or not recorded. As a result, it was not possible to independently validate the risk rating or confirm that officers considered all criteria to calculate the risk rating.

VCGLR advised that it is now incorporating the risk rating process into the licensing system.

Source: VAGO.

2.4 Licensing risk‑based model

A risk-based approach helps VCGLR allocate its scarce resources to areas that have the greatest impact on limiting the negative impacts of gambling and alcohol. It also provides VCGLR with a transparent, defensible approach to its regulatory work.

Recommendation 4

We recommend that VCGLR complete implementation of the licensing risk-based model by developing and implementing:

  • a set of risk indicators
  • checklists containing triggers for the escalation of applications within or between teams
  • a risk matrix to be considered through the determination phase.

Key points underpinning past audit recommendation

  • The risk factors considered had not been reviewed since they were established in 2009, and the model did not incorporate any risk factors specifically related to gambling venues or licence types.
  • VCGLR did not systematically review the risk profile of individual applications. It was not possible to determine whether all applications finalised were low-risk or whether VCGLR should have escalated some applications for further review.
  • The audit found no evidence of applications being escalated within teams.
  • VCGLR's approach primarily focused on speed rather than quality, contrary to its claimed risk‑based regulatory approach.

BothBubbles.PNG

Progress

VCGLR is implementing its licensing risk-based model in two parts—liquor and gambling (Figure 2E).

 

VCGLR implemented the liquor licensing risk-based model in July 2018 and risk-based framework for gambling suitability in July 2019. Implementation of the remaining components of the gambling model covering gaming machine types and game applications is due for completion by 30 June 2020.

The completion dates are different due to availability of liquor data compared with the gambling data needed for the risk-based approach. For liquor licences, VCGLR used existing datasets in the model to calculate the risk rating. For gambling licences, VCGLR is considering risk factors such as social impact and looking to source relevant data to input into its risk-based model. It is reasonable for VCGLR to take longer to implement the gambling model.

VCGLR expanded the risk factors in the liquor licensing risk-based model from two to five—trading hours, patron capacity, licence category, venue compliance history and suitability. VCGLR draws data on four of the five risk factors from independent sources. The risk rating methodology aligns with the international standards for risk management and better practice.

Figure 2E
VCGLR progress implementing recommendation 4

VCGLR has:

  • implemented a risk-based model for assessing and determining applications for or relating to liquor licences
  • expanded the risk factors in the liquor licensing risk-based model from two to five. The risk factors link to identified harms such as risky drinking, minors consuming alcohol, antisocial behaviour and detriments to community life. The risk factors also align with the objects of the LCR Act
  • collected data from sources independent of the applicant. This provides a level of independence and minimises the potential for the applicants' data to skew the risk assessment
  • developed internal guidance and provided training that supports the application of the liquor licensing risk-based model. Guidance includes a checklist that outlines when a matter may be escalated between teams.

VCGLR is:

  • developing a risk-based model for assessing and determining applications relating to gambling licences in two components. The first component—assessing suitability of gambling applicants—was implemented in July 2019, with a second component relating to gaming machine types/games due to be implemented by 30 June 2020.

VCGLR has not:

  • consistently captured the details of risk assessments for liquor licence applications in the assessment system. We found that while officers consistently recorded risk ratings in this system, they did not include the details of the assessment. It is therefore impossible to know whether they completed the risk assessment correctly. VCGLR is aware of its system limitations. VCGLR advised that it is updating systems to automatically record the risk assessments before the end of 2019.

Source: VAGO.

Under the GR Act, VCGLR must not amend a venue operator's licence unless satisfied that the net economic and social impact of the amendment will not be detrimental to the wellbeing of the community of the municipal district in which the approved venue is located.

2.5 Assessing electronic gaming machines

The 'no net detriment' test assesses the positive and negative social and economic impacts associated with a venue having electronic gaming machines. A venue is determined to have no net detriment if the positive impacts outweigh the negative. Applying the test on electronic gaming machine licence applications is intended to make sure the net impact of additional or relocated machines on the community is not negative.

The GR Act does not provide guidance on how VCGLR should apply the no net detriment test. Commissioners conduct the test, and it is not delegated to VCGLR staff members. VCGLR publishes the results of no net detriment tests on its website.

 

Recommendation 5

We recommend that VCGLR develop principles or guidance for assessing net detriment and report transparently against them in decisions on applications for electronic gaming machines.

Key points underpinning past audit recommendation

  • VCGLR lacked comprehensive guidance for commissioners on how to assess and calculate net detriment to the local community.
  • There was scope to further improve decision records, as some stakeholders advised that they found it difficult to understand how VCGLR found there was no net detriment.

Completed.PNG

Progress

VCGLR has developed detailed guidance for assessing net detriment. The guidance provides commissioners with a list of factors relevant to the no net detriment test. Commissioners select factors that are relevant to the individual application. Our review of no net detriment tests conducted in 2018 and 2019 found assessments used between 11 and 15 factors.

Commissioners determine the relative importance (weight) of each factor when assessing an individual application. They use a five-point scale to describe the relative importance—nil, marginal, low, moderate and considerable. Commissioners may also use incremental descriptors, such as 'low to moderate'.

Once factors and weightings are established, commissioners then assess the net effect of the economic and social factors on the wellbeing of the community. VCGLR concludes on the application's impact on the wellbeing of the relevant community by determining a net weighting (net weighting = [positive economic – negative economic] + [positive social – negative social]). If the net weighting is neutral or positive, then the application passes the no net detriment test.

VCGLR has implemented this recommendation, as summarised in Figure 2F. However, it could improve the methodology to determine net detriment to provide greater clarity on decisions.

Figure 2F
VCGLR progress implementing recommendation 5

VCGLR has:

  • developed more comprehensive guidance for commissioners in assessing the no net detriment test. The revised guidance, circulated to commissioners on 16May2019, covers aspects such as the key legislative provisions, intent of the no net detriment test and common issues in applying the test. The guidance also contains examples of previous decisions made by VCGLR and the Victorian Civil and Administrative Tribunal
  • revised the assessment methodology to improve the transparency of no net detriment decisions.

VCGLR could improve implementation by:

Lightbulb.PNG

  • limiting the use of weighting descriptors to five options by excluding the use of incremental descriptors. This would provide commissioners with a nine-point scale to rate factors.
  • using a single term such as 'neutral' to describe when a factor is assessed as having neither a positive or negative impact.
  • Simplifying the scale used to rate the relative importance of factors would improve the transparency of the no net detriment test.

Source: VAGO.

2.6 Reporting on licensing activities

Management reports that include timely, relevant data help senior staff to monitor trends and make informed decisions.

Recommendation 6

We recommend that VCGLR broaden its management reporting on licensing activities beyond the speed of processing applications to include quality indicators.

Key points underpinning past audit recommendation

  • Management reporting on licensing activity narrowly focused on the number of applications processed and the speed of processing.
  • VCGLR did not have a monthly report that tracks BP3 performance measures.
  • VCGLR reported separately to its commission and the Department of Justice and Community Safety on a quarterly basis on its progress against BP3 performance measures.

InProgress.PNG

Progress

VCGLR has made progress in implementing this recommendation (Figure 2G). VCGLR recently implemented a new reporting system that extracts data from five core VCGLR ICT systems into a data cube. Staff manually add other datasets, such as proof of age data. A business intelligence tool creates a range of internal and external reports, including reports on BP3 measures. Interactive dashboards provide VCGLR's management and executive with real-time access and ability to customise their reports by applying various filters. Once embedded, the system should significantly improve management reporting capabilities.

Figure 2G
VCGLR progress implementing recommendation 6

VCGLR has:

  • implemented a new reporting system, which is a significant step forward. The system provides management with real-time access to data stored in multiple VCGLR ICT systems
  • developed dashboards in the new reporting system to track progress against BP3 performance measures and sub-measures. Performance data is now available in real time as opposed to manually generated after each reporting period. Section 4.1 further discusses the quality of reporting measures
  • developed other reports that expand the scope of management reports on licensing activities.

VCGLR is:

  • continuing to develop and refine the new reporting system, build staff capability, cleanse data and embed the system into decision-making processes.

Source: VAGO.

Back to Top