RMS incorporated data from its asset condition assessments for the first time in the valuation methodology which improved the valuation outcome. Overall, we were satisfied with the valuation methodology and key assumptions, but we noted some deficiencies in the asset data in relation to asset component unit rates and old condition data for some components of assets. 

Also, a bypass and tunnel were incorrectly excluded from RMS records and valuation process since 2013. This resulted in an increase for these assets’ value by $133 million.

The valuation inputs for Wetlands and Moorings were revised this year to better reflect the assets' characteristics resulting in a $98.0 million increase.

Majority of this write-off was recognised by RMS, with $199 million relating to the write-off of existing assets which have been replaced during the year. 

Recommendation: Agencies should reduce IT risks by:

• assigning ownership of recommendations to address IT control deficiencies, with timeframes and actions plans for implementation
• ensuring audit and risk committees and agency management regularly monitor the implementation status of recommendations.

Conclusion: Agencies can more effectively manage IT vendor contracts by developing policies and procedures to ensure vendor management frameworks are kept up to date, plans are in place to manage vendor performance and risk, and compliance with the framework is monitored by:

• internal audit focusing on key contracting activities
• experienced officers who are independent of contract administration performing spot checks or peer reviews
• targeted analysis of data in contract registers.

Performance management
Eighty-six per cent of agencies meet with vendors to discuss performance. 

Only 24 per cent of agencies sought assurance about the accuracy of vendor reporting against KPIs, yet sixty-seven per cent of the IT contracts allow agencies to determine performance based payments and/or penalise underperformance.

Conclusion: Agencies are monitoring IT vendor performance, but could improve outcomes and more effectively manage under-performance by:

• a more active, rigorous approach to both risk and performance management
• checking the accuracy of vendor reporting against those KPIs and where appropriate seeking assurance over their accuracy
• invoking performance based payments clauses in contracts when performance falls below agreed standards.

Transitioning services
Forty-three per cent of the IT vendor contracts did not contain transitioning-out provisions.

Where IT vendor contracts do make provision for transitioning-out, only 28 per cent of agencies have developed a transitioning-out plan with their IT vendor.

Conclusion: A contract register helps to manage an agency’s compliance obligations under the Government Information (Public Access) Act 2009 (the GIPA Act). However, it also helps agencies more effectively manage IT vendors by:

• monitoring contract end dates and contract extensions, and commence new procurements through their central procurement teams in a timely manner
• managing their contractual commitments, budgeting and cash flow requirements.

Recommendation: Agencies should ensure their contract registers are complete and accurate so they can more effectively govern contracts and manage compliance obligations.

User access administration
Seventy-two deficiencies were identified related to user access administration, including:

• thirty issues related to granting user access across 43 per cent of agencies
• sixteen issues related to removing user access across 30 per cent of agencies
• twenty-six issues related to periodic reviews of user access across 50 per cent of agencies.

Recommendation: Agencies should:

• review the number of, and access granted to privileged users, and assess and document the risks associated with their activities
• monitor user access to address risks from unauthorised activity.

Only 57 per cent of agencies linked reporting on performance to their strategic objectives.

The use of targets and reporting performance over time was limited and applied inconsistently.

Conclusion: There is significant disparity in the quality and consistency of how agencies report on their performance in their annual reports. This limits the reliability and transparency of reported performance information.

Agencies could improve performance reporting by clearly linking strategic objectives to reported outcomes, and reporting on performance against targets over time. NSW Treasury may need to provide more guidance to agencies to support consistent and high-quality performance reporting in annual reports.

There is no independent assurance that the performance metrics agencies report in their annual reports are accurate.

Prior performance audits have noted issues related to the collection of performance information. For example, our 2016 Report on Red Tape Reduction highlighted inaccuracies in how the dollar-value of red tape reduction had been reported.

Conclusion: The ability of Parliament and the public to rely on reported information as a relevant and accurate reflection of an agency's performance is limited.

The relevance and accuracy of performance information is enhanced when:

• policies and guidance support the consistent and accurate collection of data
• internal review processes and management oversight are effective
• independent review processes are established to provide effective challenge to the assumptions, judgements and methodology used to collect the reported performance information.

Agency reporting on major projects does not meet the requirements of the annual reports regulation.

Forty-seven per cent of agencies did not report on costs to date and estimated completion dates for major works in progress. Of the 47 per cent of agencies that reported on major works, only one agency reported detail about significant cost overruns, delays, amendments, deferments or cancellations.

NSW Treasury produce an annual report checklist to help agencies comply with their annual report obligations.

Recommendation: Agencies should comply with the annual reports regulation and report on all mandatory fields, including significant cost overruns and delays, for their major works in progress.

The information the annual reports regulation requires agencies to report deals only with major works in progress. There is no requirement to report on completed works.

Sixteen of 30 agencies reported some information on completed major works.

Conclusion: Agencies could improve their transparency if they reported, or were required to report:

• on both works in progress and projects completed during the year
• actual costs and completion dates, and forecast completion dates for major works, against original and revised budgets and original expected completion dates
• explanations for significant cost overruns, delays and key project performance metrics.

• all agencies maintained purchasing card registers
• seventy-six per cent provided training to cardholders prior to being issued with a card
• eighty-nine per cent appointed a program administrator, but only half of these had clearly defined roles and responsibilities
• thirty-two per cent of agencies place merchant blocks on purchasing cards
• forty-seven per cent of agencies place geographic restrictions on purchasing cards.

Agencies have designed and implemented preventative controls aimed at deterring the potential misuse of purchasing cards.

Conclusion: Further opportunities exist for agencies to better control the use of purchasing cards, such as:

• updating purchasing card registers to contain all mandatory fields required by TPP17–09
• appointing a program administrator for the agency's purchasing card framework and defining their role and responsibility for the function
• strengthening preventive controls to prevent misuse.

Detective controls
Ninety-two per cent of agencies have designed and implemented at least one control to monitor purchasing card activity.

Major reviews, such as data analytics (29 per cent of agencies) and independent spot checks (49 per cent of agencies) are not widely used.

Agencies have designed and implemented detective controls aimed at identifying potential misuse of purchasing cards.

Conclusion: More effective monitoring using purchasing card data can provide better visibility over spending activity and can be used to:

• detect misuse and investigate exceptions
• analyse trends to highlight cost saving opportunities.

• a further 41 per cent of agencies have not reviewed their policies by the scheduled revision date, or do not have a scheduled revision date
• more than half of all agencies’ policies do not offer alternative travel options. For example, only 36 per cent of policies promoted the use of general Opal cards.

• limit the circumstances where taxi use is appropriate
• offer alternate, lower cost options to using taxis, such as general Opal cards and rideshare.

Prevention systems
Ninety-two per cent of agencies have a fraud control plan in place, 81 per cent maintain a fraud database and 79 per cent report fraud and corruption matters as a standing item on audit and risk committee agendas.

Only 54 per cent of agencies have an employment screening policy and all agencies have IT security policies, but gaps in IT security controls could undermine their policies.

Conclusion: Most agencies have implemented fraud prevention systems to reduce the risk of fraud. However poor IT security along with other gaps in agency prevention systems, such as employment screening practices heightens the risk of fraud and inappropriate use of data.

Agencies can improve their fraud prevention systems by:

• completing regular fraud risk assessments, embedding fraud risk assessment into their enterprise risk management process and reporting the results of the assessment to the audit and risk committee
• maintaining a fraud database and reviewing it regularly for systemic issues and reporting a redacted version of the database on the agency's website to inform corruption prevention networks
• developing policies and procedures for employee screening and benchmarking their current processes against ICAC's publication ‘Strengthening Employment Screening Practices in the NSW Public Sector’
• developing and maintaining up to date IT security policies and monitoring compliance with the policy.

Studies have shown data monitoring, whereby entire populations of transactional data are analysed for indicators of fraudulent activity, is one of the most effective methods of early detection. Early detection decreases the duration a fraud remains undetected thereby limiting the extent of losses.

Conclusion: Data monitoring is an effective tool for early detection of fraud and is more effective when informed by a comprehensive fraud risk assessment.