1.1 The Australian Government has outsourced employment services to largely non-government owned providers since 1998.¹ Workforce Australia is the Australian Government’s current employment services program. It commenced on 1 July 2022 and is administered by the Department of Employment and Workplace Relations (DEWR or the department). It replaced the predecessor program, jobactive, which had been in place since July 2015.

1.2 Jobactive was initially established to operate for five years between July 2015 and June 2020. By April 2019, deeds between DEWR and the jobactive providers were extended for a further two years to June 2022 to, among other things, ‘enable continuation of employment services while the new employment services model [was] trialled.’

1.3 There were around 630,000 job seekers in jobactive as at 29 February 2020, prior to the onset of the COVID-19 pandemic.² This increased to a peak of 1.49 million by 30 September 2020, and by December 2021, had reduced to 895,000. As at August 2023, there were 604,000 job seekers in Workforce Australia employment services.³

1.4 Employment service providers are contracted by DEWR to deliver employment services and support job seekers and employers, including assisting job seekers receiving certain income support payments to manage their mutual obligations requirements.⁴ Under Workforce Australia, mutual obligations are reflected in a ‘Job Plan’ and involve agreed tasks and activities to increase job readiness, taking account of job seekers’ personal circumstances.⁵

New Employment Services Model

1.5 Between May 2018 and May 2021, the Australian Government invested funding of over $6 billion across the forward estimates for the delivery of ‘a new approach to employment services that is digitally driven, tailored and flexible.’ Details announced in the context of Federal Budget processes included:

• five announcements between January 2018 and October 2020 relating to the development of ‘Online Employment Services’, commencing with a trial in July 2018. The trial was expanded progressively until online self-servicing was implemented for all ‘job-ready job seekers’ in April 2020.⁶ In November 2023, the department advised the ANAO that the collective net efficiencies recorded against these measures totalled $1.3 billion;⁷
• an April 2019 announcement for ‘$249.8 million over five years from 2018–19 (including $25.7 million in capital funding over four years from 2018–19) to pilot key elements of a new employment services model’ in two employment regions;
• an October 2020 announcement of $295.9 million over four years from 2020–21 (including $150.7 million in capital funding) to ‘establish a new digital employment services platform that will be available to all Australians’⁸;
• an announcement in May 2021 that an efficiency of $860.4 million over four years from 2021–22 would be achieved by transitioning from jobactive to the ‘New Employment Services Model’ (the new model, or NESM) from 1 July 2022.

1.6 In October 2021, the NESM was renamed as ‘Workforce Australia’ and was to introduce two pathways of support for job seekers, comprising:

• Enhanced Services — a network of employment services providers to deliver tailored and intensive case management support to higher-risk job seekers; and
• Online Employment Services — also known as ‘Workforce Australia Online’, for job seekers assessed as lower-risk, considered job ready, and able to self-manage their mutual obligations and job search activities online.⁹

1.7 The new model was to be a ‘transformational change’ from the previous program. In contrast to jobactive, where all job seekers were referred to the provider network for face-to-face assistance, only higher-risk job seekers would be required to access employment services through a provider under the NESM.¹⁰ Net efficiencies from the reduction in face-to-face servicing arrangements were to be ‘reinvested in the employment services system to provide a more intensive, targeted and tailored service for those who need extra help in addressing their barrier to getting a job.’¹¹

1.8 Development of the NESM commenced in early 2018. Detailed aspects of the new model were developed between 2018 and 2021, which was informed by a ‘New Employment Services Trial’ pilot from July 2019 and a range of financial viability analyses conducted from mid-2020 and early 2021. The department engaged KPMG to undertake these analyses in May 2020 to inform its refinement of the ‘micro policy’ settings for the program and to provide assurance that providers would remain viable under the new model when servicing the smaller pool of job seekers in Enhanced Services. This work was a key input to the design of the NESM procurement process and tender documents (discussed from paragraph 1.10). Figure 1.1 illustrates the reduction of the job seeker caseload between jobactive and the new model.

Figure 1.1: Changes in the job seeker caseload between jobactive and Workforce Australia by state and territory

Note a: The jobactive caseload data is taken from Australia Bureau of Statistics (ABS) reporting in June 2020.

Note b: Indicative Enhanced Services caseload numbers were those published in the NESM Request for Proposal.

Source: ANAO analysis of ABS and DEWR records.

Procurement process

1.9 At least seven procurement processes for a range of employment assistance and related services were conducted either in parallel or in close succession throughout 2021–22. One of these processes was for identifying providers to deliver Enhanced Services under the new program, which was also used to establish the Workforce Australia Services Panel. The panel was established to ‘streamline’ any future procurement processes required over the life of the new program, such as when filling an Enhanced Services provider ‘gap’ in a specific employment region. Figure 1.2 illustrates the timing of these procurements.

Figure 1.2: Timeline of the NESM and other employment services procurements in 2021–22

Note a: Savings announced as part of the 2023–24 Budget process included $22.8 million by ceasing the Entrepreneurship Facilitator Program from 1 July 2023 and $111.6 million by reducing place allocations for the Self-Employment Assistance Small Business Coaching program. See: Australian Government, Budget 2023–24 Budget measures: Budget Paper No. 2, p. 106, available from https://budget.gov.au/content/bp2/download/bp2_2023-24.pdf [accessed 3 October 2023].

Source: ANAO analysis of departmental records.

1.10 On 8 September 2021, proposals for the following three ‘service areas’ were invited through a single open tender process:

• Enhanced Services — a range of intensive, individualised and tailored services for vulnerable job seekers delivered by a network of employment service providers.
• Employability Skills Training (EST) — a targeted training service to enhance work readiness through intensive pre-employment training.
• Career Transition Assistance (CTA) – designed to help mature age job seekers aged 45 and over to build their confidence and skills to become more competitive in their local labour market.

1.11 The EST and CTA service areas are support services for job seekers, which were also available under the jobactive program. Under jobactive, providers were able to refer job seekers to these services according to their personal needs. While this remains the case under Workforce Australia, Enhanced Services providers are now subject to a ‘referral cap’ and job seekers in online services can enrol themselves if they are eligible.¹²

1.12 The relevant information for each service area was provided within the published ‘Request for Proposal for the New Employment Services Model 2022’ (RFP), which included a consolidated request for proposal document, draft deeds or deed of standing offer, and a range of factsheets on program aspects. Proposals for all three services were to be received by 3pm (Canberra time) on 22 October 2021. The assessment processes for each service area were conducted separately but in close succession, with successful providers notified in March 2022.

1.13 Work orders with successful Enhanced Services providers were executed during July 2022. In total, 155 providers submitted 2111 proposals across 51 employment regions. As outlined by Table 1.1, over half of these proposals were for Enhanced Services.

Table 1.1: Tender submissions to the NESM procurement (by service area)

Note a: Total number of providers does not equal the sum of individual providers as some bid for multiple service areas.

Source: ANAO analysis of departmental records.

1.14 On 11 March 2022, the department notified 104 providers that they had been appointed to the national panel¹³, and from those providers, 45 were offered 181 ‘licences’ to deliver Enhanced Services within one or more employment regions.¹⁴ Licences were ‘issued’ by executing work orders under a Deed of Standing Offer (DoSO).¹⁵ Of the 45 providers invited to sign a work order in March 2022, 43 had provided written acceptances by 13 July 2022. After the commencement of Workforce Australia from 1 July 2022, the panel consisted of 99 providers, inclusive of the 43 that accepted offers under the DoSO.¹⁶ The total estimated value of the licences awarded, as reported on AusTender, was $3.27 billion.

Recent reviews and inquiries into employment services

1.15 On 2 August 2022, the House Select Committee on Workforce Australia Employment Services was established to ‘inquire into and report on matters related to Workforce Australia Employment Services’.¹⁷ The Committee tabled its Interim Report in ParentsNext in February 2023, and as at mid-November 2023, the Committee was due to table its final report by 30 November 2023.

1.16 On 25 September 2023, the Australian Government released the White Paper on Jobs and Opportunities.¹⁸ The White Paper set out the Government’s ‘roadmap’ for the labour market and included eight ‘guiding principles to strengthen employment services in Australia’.¹⁹

Previous audits

1.17 Previous ANAO performance audits relating to the department’s administration of the previous employment services program include:

• Auditor-General Report No. 31 2021–22 Jobactive – Integrity of Payments to Employment Service Providers; and
• Auditor-General Report No. 4 2017–18 Jobactive: Design and Monitoring.

Rationale for undertaking the audit

1.18 This audit was conducted to provide independent assurance to the Parliament on the design and establishment of the panel arrangements for the program, including whether the procurement process conducted was effective and consistent with the Commonwealth Procurement Rules.²⁰

Audit approach

Audit objective and criteria

1.19 The objective of the audit was to assess the effectiveness of the establishment of the Workforce Australia Services panel.

1.20 To form a conclusion against the objective, the following high-level criteria were considered.

• Was the design of the procurement consistent with policy objectives and achieving value with money?
• Was the procurement conducted in accordance with the published process?
• Did the results of the evaluation process appropriately inform the establishment of the panel?

Audit methodology

1.21 The audit methodology included examination and analysis of entity records and meetings with relevant departmental officials and stakeholders.

1.22 As the procurement process was conducted during periods of COVID-19 health restrictions in early 2022, departmental records for the procurement process comprised online meeting recordings of the Tender Review Committee (TRC) deliberations on proposals in each the 51 employment regions. The audit approach included reviewing the recordings for at least 42 of those regions.

1.23 The audit was conducted in accordance with ANAO Auditing Standards at a cost to the ANAO of approximately $652,200.

1.24 The team members for this audit were Swatilekha Ahmed, Dr Adam Reddiex, Jay Banpel, Kathryn Longstaff, Jude Lynch, Josh Carruthers, Michelle Page and Amy Willmott.

2.1 The Australian Government has outsourced the delivery of employment services since the ‘Job Network’ program was introduced in May 1998. While this model has evolved over time in successor programs, its underlying elements have remained largely unchanged.²¹

2.2 In November 2017, the Australian Government agreed to pursue ‘fundamental reform’ to improve those services. That reform was to be guided by seven foundational principles, which included a system design that would maximise job seeker and employer engagement; and ensure efficiency and value for money in policy design and service delivery.²² The ANAO examined whether these aspects were reflected in the design of the procurement process.

Were lessons learnt from the previous program used to inform the new design, including any stakeholder input?

Employment Services Expert Advisory Panel

2.3 To inform the development of future employment services, the Australian Government established an Employment Services Expert Advisory Panel (the expert panel) on 22 January 2018.²³ The expert panel’s work was informed through public consultation of over 1400 job seekers, employers, employment services providers and community groups.²⁴ It reported its findings in October 2018 and made 11 recommendations, with key findings including that:

• a new digital approach should be adopted, with capable job seekers allowed to self-serve online and providers able to focus on supporting job seekers needing the most help; and
• a provider licensing framework and improved payment and performance model should be adopted to allow for greater competition and diversity between providers, without compromising market stability.

Licensing arrangements

2.4 A licensing framework for employment services was contemplated by the Productivity Commission in its review of the Job Network in 2002. It noted that competitive tendering is complex and expensive for providers and disruptive to services. It recommended that licensing of providers be adopted to enable free entry into the market by accredited agencies — subject to ongoing assessment of quality.²⁵

2.5 Similarly, the expert panel recommended a ‘managed approach to increasing competition’ in the employment services market by issuing provider licences for a minimum of five years and, ‘at least initially, guaranteed minimum and maximum market shares’. The panel envisaged that:

• entry into the market would be easier, with flexibility in the number of licences available and the timing of their distribution;
• providers would be required to complete a licence application, with that process being simpler and less costly than tender processes; and
• licence extensions would be available for consistently high performing providers and loss of licences before expiry for consistently poor performers.

2.6 The expert panel recommended that the number of licences issued in each of the 51 employment regions be capped, which differed from the Productivity Commission model. This was to ensure an appropriate balance between market sustainability and providers’ capacity to meet job seeker and employer needs. The licensing approach is discussed further at paragraph 2.16.

Senate Education and Employment References Committee inquiry into jobactive

2.7 In August 2018, the Australian Senate referred an inquiry into ‘the appropriateness and effectiveness of the objectives, design, implementation and evaluation of jobactive’ to the Education and Employment References Committee (the committee). The committee’s findings were published in February 2019.²⁶ It found that jobactive was not meeting its intended objectives and, among other things, noted that:

• the program was not providing sufficient or appropriate support for disadvantaged job seekers; and
• the funding model incentivised providers to churn people through short term work, rather than securing longer-term employment.

2.8 The committee made 41 recommendations to improve the employment services system. The Australian Government’s response to the report was tabled on 20 May 2020, which outlined a range of features to be explored as part of a new model, including separate ‘streams’ for job-ready and disadvantaged job seekers, and intensive case management for the job seekers most in need.

Policy intent for the design of the New Employment Services Model

2.9 On 11 December 2018, the Australian Government agreed that a range of fully costed options for the NESM be developed for implementation following the expiry of jobactive contracts on 30 June 2020. Consistent with the expert panel’s recommendations, the following ‘key transformational changes’ from the jobactive model were to underpin the design of the NESM:

• self-servicing through digital services for job-ready job seekers and a digital and data ecosystem that will reduce red tape;
• more intensive support for disadvantaged job seekers;
• more effective, tailored and flexible job seeker activity requirements that maintain mutual obligations; and
• a licensing framework to lower barriers to entry and exit, more effectively drive quality outcomes, and reduce the cost and disruption of procurement processes.

Was the policy intent reflected in the new design, including the procurement approach?

2.10 Following government agreement in December 2018, the Department of Employment and Workplace Relations (the department) worked until early February 2019 to develop policy options for the NESM, for consideration in the 2019–20 Budget process.

2.11 The department established six ‘sprint teams’ to focus on refining the options and policy settings for key aspects of the new model.²⁷ On 11 January 2019, the department advised the Minister for Jobs and Industrial Relations (the Minister)²⁸ that:

• the employment services reforms represented significant fiscal and reputational risk, and given the tight timeframes for the April 2019 Federal Budget, the department’s capacity to manage these risks by engaging with stakeholders on the detailed design and implementation of the model was ‘severely’ limited; and
• instead of the full implementation of a new model on 1 July 2020, it recommended that an iterative implementation process for the new model be adopted, consisting of:
  • rolling over the existing jobactive contracts; and
  • commencing with a pilot of the reforms from July 2019 and a national rollout from July 2021.²⁹

2.12 The Minister’s proposal to government was consistent with this advice. On 5 March 2019, the Australian Government agreed that a fully costed model for the NESM be developed in accordance with the Minister’s preferred option. The model was to be informed by evidence from a pilot in two employment regions: Adelaide South in South Australia (SA) and Mid North Coast in New South Wales (NSW).³⁰ The Minister announced the reform process on 20 March 2019 at a jobactive CEO Forum and provided further details in another announcement on 2 April 2019.

New Employment Services Trial (NEST)

2.13 The New Employment Services Trial (NEST) pilot was implemented in phases between July 2019 and June 2021 to test key design elements including the provider payment framework, ‘points-based activation system’ for mutual obligations, and job seeker registration and referral processes (for both online and face-to-face services). The contractual licensing model and the provider performance framework were not tested as part of the trial.³¹

2.14 On 2 April 2019, the eight pre-existing jobactive providers in the two pilot regions were invited to submit expressions of interest (EOIs) to participate in the NEST through a limited tender process.³² All eight providers participated in the pilot.

2.15 The final evaluation report covering the period from July 2019 to June 2021 was delivered in November 2022, around four months after Workforce Australia commenced.³³ Learnings from the pilot, including analysis undertaken by KPMG (discussed at paragraphs 2.24 to 2.31), informed the refinement of policy settings. This included changes to the proposed provider payment framework and simplification of the new model by removing the ‘Digital Plus’ stream and consolidating the ‘tiers’ of support within Enhanced Services.³⁴

Licensing framework

2.16 The initial NESM program design in July 2019 was largely consistent with the ‘transformational changes’ agreed by government in December 2018. The key point of difference was that the NESM was to be established through a procurement process, rather than through a licensing application process and framework (see paragraphs 2.4 to 2.6). When examining potential options in January 2019, the department recorded that there were two approaches available: ‘statutory licensing’ and ‘contractual licensing’, and also considered whether the number of providers in the market should be capped (each with allocated ‘market share’ percentages) or uncapped. It noted that a:

• [statutory] licencing model would require primary legislation to establish it and is subject to rejection or amendment by Parliament. This creates a risk that a licencing model may not pass Parliament, or that is amended by Parliament, that could result in a model inconsistent with the intent and objectives of the overarching employment services model; … and
• … contractual licencing model is a procurement administered under the Commonwealth Procurement Rules (CPRs) and in practice is not materially different to the current procurement processes used….

2.17 The department’s analysis was that while an uncapped statutory licensing model most closely met the government’s objectives, the ‘risks associated with progressing with this option [were] considered too high.’³⁵ Contractual licensing through a ‘streamlined procurement model’ was identified as the preferred approach and recommended to government in late January 2019.³⁶

2.18 A risk identified in a procurement approach was that it may be seen as disadvantaging small and specialist providers.³⁷ To mitigate this and give effect to the policy objectives, the department noted that there were a number of options available to streamline, simplify and differentiate the upcoming procurement process from previous processes. It further noted that a number of changes could be made to reduce barriers to entry, particularly for small and niche organisations. This included:

• allowing ‘part region’ tendering, to encourage local, regionally/locally focused providers to tender³⁸;
• different tender requirements for generalist/specialists/part region providers; and
• re-weighting of the selection criteria (for example, reducing the weighting for past performance and increasing it for local community connections and feedback).

2.19 These changes were factored into the evaluation criteria for the procurement and are examined further from paragraph 2.45.

National panel and regional sub-panels

2.20 In a panel arrangement, suppliers are appointed to supply goods or services for a set period of time under agreed terms and conditions. Once a panel has been established, an entity may then purchase directly from the panel by approaching one or more suppliers.³⁹ A ‘national panel’ was established through the NESM Request for Proposal (RFP) process to, among other things, streamline the entry and exit of providers in the market and to fill gaps within regions as needed.⁴⁰

2.21 The RFP required providers to give tailored responses for each employment region they had bid for. Sub-panels were to be established for each of the 51 employment regions. Where providers were found ‘suitable’ for one or more regions, they were to be appointed to the national panel and the sub-panels for those regions.⁴¹ The national panel was established for an initial period of six years through a Deed of Standing Offer (DoSO). After that period, panel members may be extended for up to an additional four years at the department’s discretion.

2.22 Successful providers were identified from the sub-panels and offered ‘licences’ for those regions through the execution of work orders under the DoSO.

Provider payment model and financial viability

2.23 The NESM program design was based on the premise that by lowering providers’ caseloads, more intensive and tailored services could be provided to the most disadvantaged job seekers. Under jobactive, the average job seeker to consultant (caseload) ratio was 150:1.⁴² The NESM policy objective was to achieve a caseload ratio of 80:1.

2.24 The department engaged KPMG in May 2020 to conduct financial viability analysis, based on jobactive and NEST pilot data. While this work was initially due for completion by the end of July 2020, additional analysis was requested by the department. The analysis progressively examined the feasibility of the policy underpinning the NESM — with a focus on assessing the financial viability of providers under the new policy settings — to identify the appropriate number and types of licences for each employment region. In total, three sets of modelling were undertaken by KPMG in August 2020, January 2021 and May 2021.⁴³ The results of these analyses informed the detail included in the RFP.⁴⁴

August 2020 financial viability analysis

2.25 Between May and June 2020, KPMG conducted interviews with departmental subject matter experts (across seven employment services programs) and engaged with providers directly to source financial and operating data and, where relevant, seek their perspectives and experiences from participating in the NEST. In total, four providers contributed data for the financial viability analysis and eight provided insights during interviews. The department also provided NEST and jobactive caseload and payment data to KPMG.

2.26 Initial findings provided to the department on 13 July 2020 indicated there would be ‘significant viability challenges for providers’ if the current NEST settings were included in the NESM, with 65 per cent of employment regions unable to support a single provider over two years.⁴⁵ In response to feedback from the department, KPMG drew ‘more deeply on historic jobactive data as an additional input’ to its analysis.⁴⁶ The updated results were reflected in the final August 2020 ‘New Employment Services Model Financial Viability Analysis’ report (August 2020 KPMG report), which included the following observations.

• Most scenarios are viable over both a three and five year period – … Scenarios with higher commencement numbers, higher caseload to staff ratios, lower fixed cost structures and higher outcome rates tended to increase profitability over three and five years. Scenarios with a combination of negative factors – notably lower caseload, high cost structures and low outcome rates – were loss making over the period examined.⁴⁷
• Caseload to staff ratios of 80:1 were not viable under any of the scenarios examined – No scenarios examined were viable under the high cost structure assumption. This assumption included caseload to staff ratios of 80:1. However, providers were viable at a ratio of 120:1.
• Jobseeker volumes and cost structures are key factors driving viability – … analysis indicates that, in order of importance, the key drivers of viability are caseload to staff ratios, jobseeker volumes, provider fixed costs and outcome rates. Staff ratios and fixed costs are clearly factors determined by providers.
• Upfront cash flow for providers supports initial viability – In all scenarios examined, providers experience net monthly losses in the first year of operation.
• Market share allocations may need to change compared with jobactive – The scenario analysis suggests that providers are unlikely to be viable based on historic market share allocations and some reduction in provider numbers may be necessary to maintain viability. [Emphasis in original]

2.27 In this latter respect, KPMG noted that ‘[w]hile this could reduce competition within some regions … the department could consider a mixed model where fewer large and medium providers are contracted but competition within a region is maintained through adding a number of smaller, low cost providers.’ This was, in part, to achieve a balance between two competing policy objectives of providing intensive and tailored services to job seekers and increasing job seeker choice of providers within employment regions.

January 2021 analysis on updated provider payment framework

2.28 The department re-engaged KPMG in November 2020 to ‘undertake supplementary financial analysis’ on caseload ratios using updated Treasury forecasts and data. This analysis was to test the impact of changes to provider viability using an updated payment model and examine policy settings that had not been settled at the time of the first analysis.

2.29 The NEST pilot policy settings were adjusted, and supplementary analysis was undertaken by KPMG based on a broader range of fixed cost assumptions for providers, a revised payment model and updated inputs from the department.⁴⁸ The supplementary analysis was finalised in a January 2021 KPMG report, which outlined that the analysis was undertaken to ‘investigate payment settings which aim to increase the likelihood of provider viability when caseload to staff ratios of 80:1 are applied.’ The market share allocations were not increased and the caseload ratio of 80:1 was maintained (as the intent was to achieve provider viability at that caseload ratio). The range of fixed cost assumptions used for the January 2021 analysis are outlined in Table 2.1.

2.30 The supplementary analysis was finalised in January 2021, finding that a provider operating with a revised payment model and caseload ratio of 80:1 would be ‘viable across a 10-year contract period’, with the new payment structure having a positive impact on viability. It was noted that while there ‘would be months within the first year when losses would be recorded, in aggregate and across the 10-year period, the modelled provider is viable’.

2.31 The department advised government in April 2021 that the provider payment model was critical to the success of the NESM and KPMG’s analysis had found that the revised payment model will support provider viability at caseload ratios of 80:1. The department did not outline that the January 2021 report noted that at least two and up to 12 employment regions could not support a single Enhanced Services provider under the NESM policy settings (as shown in Table 2.1).

Table 2.1: Number of viable Enhanced Services providers by region and provider fixed cost base compared with contracted jobactive providers

Source: ANAO representation of departmental records.

Market share and diversity of providers

2.32 Reflecting that some policy settings were introduced or modified to support smaller providers and encourage more diversity in the market, the published RFP stated that:

• assistance would be provided for eligible small organisations through a new Capacity Building Fund⁴⁹; and
• the total business share of an organisation licensed to deliver Enhanced Services was not to exceed 20 per cent of the national market (inclusive of related entities).

2.33 In respect to the latter, this represented a doubling of the initially agreed 10 per cent market share cap determined by the department’s procurement interim steering committee in June 2021. This change was made as a result of a request from the Minister in August 2021. The previously agreed market share cap had been communicated to stakeholders and providers through the RFP Exposure Draft in June 2021.

May 2021 analysis of specialist licences

2.34 The RFP outlined that ‘Specialist Enhanced Services’ licences would be allocated in some regions to providers with expertise in supporting refugee, culturally and linguistically diverse (CALD), Indigenous or ex-offender job seekers (specialist cohorts). In regions with no specialist services, generalist providers were expected to ensure appropriate support for all job seekers on their caseload.

2.35 The November 2020 work order with KPMG was extended in early April 2021 for additional analysis to identify whether particular job seeker cohorts could be viable in each employment region, and whether specialist licences would have an impact on the viability of generalist providers in those regions.⁵⁰ The final report was provided to the department on 10 May 2021.

2.36 The analysis indicated that specialist providers were unlikely to be viable in most employment regions. This was based on the assumption that a specialist provider would hold 40 per cent of the market share for the relevant cohort in the region, as this was required for viability. Table 2.2 outlines the findings of that analysis.

Table 2.2: Number of viable specialist cohorts across employment regions

Note a: The analysis was based on ‘representative’ provider sizes. For example, a medium to large provider was assumed to have fixed costs of approximately $2.4 million and would therefore need a gross margin of more than this to operate viably.

Source: ANAO representation of departmental records.

2.37 This analysis informed the ‘indicative’ number of licences and average caseload numbers that were included in the NESM RFP for each employment region. The maximum number of licences included in the RFP for each region exceeded the number of licences identified as being viable by KPMG.⁵¹ The department’s rationale for suggesting more providers could be viable was ‘promoting job seeker choice and market diversity.’⁵² This resulted in the RFP stating that between one and three licences could be awarded in 12 regions identified by KPMG as unlikely to be able to support a single provider (see Table 2.1).

Public release of financial viability reports

2.38 The department kept the Minister progressively informed of KPMG’s findings and sought legal advice on at least three occasions between September 2020 and May 2021 regarding its obligations for the public release of the financial viability reports.⁵³ The advice sought was in respect of the department’s obligations under the Freedom of Information Act 1982 (FOI Act) and deeds with the providers participating in the KPMG analysis.

2.39 Following a request from the Minister, the department made arrangements in early August 2021 for all three KPMG reports to be provided to government, to inform the government’s policy considerations. As outlined at paragraph 2.40, Australian Government policy approval for the final NESM design was provided on 11 May 2021. In response to ANAO queries in late September 2023, the department advised that it ‘has not had an FOI request for the release of these reports so further consideration has not been required’ with respect to their public release. The department advised that ‘[t]o better support tenderer’s response to the NESM [RFP], the department provided a summary of the findings from the KPMG report as well as the assumptions that underpinned the KPMG analysis’ as attachments to the RFP exposure draft.

Final policy approval

2.40 Policy approval for the final NESM design was provided by the Australian Government on 11 May 2021, with remaining details to be finalised via correspondence between the Minister and the Prime Minister. Details finalised by correspondence included: the Enhanced Services Provider Performance Framework; final Enhanced Services provider payment model; and Employability Skills Training (EST) payment model adjustment.

2.41 Letters were exchanged between the Minister and the Prime Minister on 23 August 2021 and 6 September 2021, respectively. The Prime Minister noted the final details and agreed the proposed changes, with costs to be agreed with the Department of Finance (Finance).⁵⁴ Finance’s agreement was received on 13 September 2021. The RFP was released on 8 September 2021.⁵⁵

Was request for proposal documentation made publicly available, including appropriate evaluation criteria?

2.42 In accordance with the Commonwealth Procurement Rules (CPRs), the department published the NESM procurement on the AusTender website on 8 September 2021. Potential respondents were invited to submit proposals across three ‘service areas’ under the NESM. Multiple responses were required for those bidding to deliver more than one service and/or in more than one of the 51 employment regions. Proposals were to be tailored to address the local needs of each region and received on or by 22 October 2021. The three service areas were:

• Enhanced Services — a range of intensive, individualised and tailored employment services for vulnerable job seekers delivered by generalist and specialist providers.⁵⁶ The procurement process for Enhanced Services was used to establish the ‘Workforce Australia National Panel’.
• Employability Skills Training (EST) — a targeted training service to enhance work readiness by providing intensive pre-employment training.
• Career Transition Assistance (CTA) — designed to help mature age job seekers aged 45 and over to become more competitive in their local labour market.

2.43 The relevant information for each service area was included in the ‘Request for Proposal for the New Employment Services Model 2022’ (RFP), which was available on the department’s website and on AusTender. In addition to the RFP document, the approach to market materials included a range of other supporting attachments, such as frequently asked questions (FAQs) and information sheets.

2.44 Generally, where tender documentation is amended after it has been released, addenda are published to ensure all potential respondents are made aware of any additional information and clarifications. The NESM Probity Plan required the external probity adviser to be consulted prior to addenda being issued. Eight addenda were published on the department’s website between 15 September and 18 October 2021, with majority for the release of FAQs. One addendum was reviewed by the probity adviser prior to being issued.⁵⁷

Published evaluation criteria

2.45 Relevant evaluation criteria should be included in request documentation to enable the proper identification, assessment and comparison of submissions on a fair, common and appropriately transparent basis.⁵⁸ Request documentation must include a complete description of the evaluation criteria to be used when assessing submissions and, if applicable, the relative importance of those criteria.⁵⁹

2.46 The selection criteria were grouped into three ‘areas of capability’ in the RFP. Respondents were required to address the first criterion only once, irrespective of the number of services and regions being bid for. Tailored responses for the other two criteria were required for each service and/or employment region being bid for. The selection criteria are outlined in Figure 2.1.

Figure 2.1: NESM selection criteria and sub-criteria questions (and respective weightings)

Source: ANAO analysis of departmental records.

Evaluation criteria weightings

2.47 As illustrated by Figure 2.1, each of the three high-level criteria comprised between four and nine sub-criteria questions.

2.48 The qualitative rating scale to be used for the NESM was developed throughout September 2021 and programmed into the IT system used to assess proposals — the Procurement and Licence Management System (PaLMS).⁶⁰ These qualitative ratings, or ‘evaluation descriptors’, were mapped to numerical values in PaLMS, enabling assessment results to be calculated and orders of merit to be produced. ANAO review of PaLMS identified that:

• the questions were of equal value and importance in the calculation of the overarching criteria scores, as sub-criteria weightings were evenly distributed across the relevant questions; and
• between two and four evaluation descriptors were to be selected for the assessment against each question.

Evaluation component descriptors

2.49 The NESM Assessment Guide, approved on 5 November 2021, was a key reference document for staff conducting assessments against the evaluation criteria.⁶¹ It explained that:

EDs [evaluation descriptors] are specific indicators used to rate aspects of the Respondent’s submission against the selection criterion questions. Assessors must consider the criterion question responses and select the relevant ED as they analyse how and to what degree the Respondent’s response supports its claims against the selection criterion.

2.50 The number of aspects, or ‘evaluation component descriptor elements’, to be assessed for each selection criterion question comprised between two and four of the seven elements listed in Table 2.3. Appendix 3 provides further detail, including the combination of elements adopted for each criterion and the respective spread of weightings.

2.51 User acceptance testing (UAT) for the PaLMS was conducted over a two-week period between late July and early August 2021, after delays caused by the IT build and competing priorities across multiple concurrent procurements. The combination of evaluation descriptors to be used for the NESM procurement was determined in parallel with the PaLMS platform build and finalised after the UAT period, on 12 October 2021. Therefore, the department’s ability to fix any system issues or analyse any unintended effects on the assessment process or outcomes was limited.

Table 2.3: Distribution of selection criteria weightings across the evaluation component descriptor elements adopted for the Enhanced Services criteria

Note a: These columns identify the number of times this ‘evaluation component descriptor element’ was used across the sub-criteria questions for this selection criterion. Appendix 3 provides further details.

Source: ANAO analysis of departmental records.

Providers with previous experience with the department

2.52 Respondents were to provide information on their ‘demonstrated performance’ under sub-criteria 1.8 and 1.9. Consistent with departmental advice in early 2019, these sub-criteria represented 4.45 per cent of the total score. In contrast, the equivalent criteria for the 2014 jobactive procurement was 30 per cent of the total score. This shift in approach for the NESM was adopted to reduce barriers to entry, particularly for small or new providers and alleviate administrative burden (see paragraph 2.18). Guidance in the RFP stated that:

• for existing jobactive providers — ‘the department will use current performance and other quantitative data held by the department. Existing Providers will have the option to NOT provide additional information in response to this question’ [emphasis in original]; and
• respondents not contracted by the department — ‘should ensure they describe current or past performance in delivering similar services for another organisation(s) and/or different services targeted to similar Participants. These Respondents must provide details of referees who can verify the Respondent’s specific claims.’

2.53 The implications of this approach during the assessment process are discussed from paragraph 3.40.

Was an appropriate evaluation process established and supported by appropriate guidance?

2.54 The department’s procurement policy requires the assessment criteria and processes included in publicly released RFP documents to be ‘sufficiently thorough and well planned to enable the department to make purchasing decisions that represent best value for money.’ The respective evaluation processes must be finalised and approved by the Delegate prior to the closing date of the relevant procurement process.

Published NESM evaluation process

2.55 The NESM RFP provided a basic overview of a five-stage process to be undertaken for the evaluation of proposals.⁶² Those stages comprised:

• Stage 1 – Receipt and screening of responses (including conformance assessments);
• Stage 2 – Assessment of responses against areas of capability and selection criteria;
• Stage 3 – Right Fit For Risk Cyber security assessment;
• Stage 4 – Value for money assessment and recommendations; and
• Stage 5 – Negotiations with respondents and final decisions.

2.56 The RFP outlined that each proposal was required to pass Stage 1 before being assessed against the published selection criteria to ‘determine the Respondent’s eligibility’ for appointment to the national panel for Enhanced Services under a Deed of Standing Offer (DoSO).⁶³ The RFP did not outline how those assessment results would be used to determine a respondent’s ‘eligibility’ for the national panel (that is, a score of more than four out of 10, or 40 per cent, against criterion one). Potential respondents were also not:

• advised that the assessment process involved a ‘business allocation’ stage to be conducted by staff with local area and employment region knowledge (outlined from paragraph 3.66); or
• provided details of the Stage 5 methodology, which involved a Tender Review Committee (TRC) recommending successful proposals to the Delegate (the Deputy Secretary) (outlined from paragraph 4.6).

2.57 It was noted several times throughout the RFP that each response to individual criteria would be ‘read in isolation as a discrete response.’ Respondents were advised that they ‘should answer the criteria and relevant questions directly, [and] avoid repetition and cross-referencing of other criterion responses.’ This indicated that early planning of the evaluation process had been undertaken prior to the RFP release, as assessors were allocated responsibility for specific criteria.

NESM procurement policy framework and related guidance

2.58 The department had established specific internal policies and guidance for the procurement of employment services. An overview of those relating to the NESM are set out in Table 2.4.

Table 2.4: Approval of NESM-specific guidance documents

Note a: This version applied to the design and implementation of the NESM procurement.

Note b: Version 2 was approved on 17 December 2021 and Version 3 on 1 February 2022.

Note c: If there was an inconsistency between these and the NESM 2022 Purchasing Plan, the terms of the Purchasing Plan were to prevail to the extent of the inconsistency.

Source: ANAO analysis of departmental records.

Changes to the internal policy framework

2.59 The ‘Employment Services Purchasing Guidelines’ (ESP guidelines) were introduced in 2017 to consolidate ‘the framework, approach and processes for conducting major procurement or grant processes’ for employment services programs. Arrangements for individual processes were to be documented in separate ‘Purchasing Plans’ approved by the Delegate before the relevant RFP closing date.⁶⁴ This approval is required because the purchasing plans:

• include the ‘specific evaluation approach for the relevant program, as well as the assessment and [BA] requirements to be used for the conduct of the procurement’; and
• used by assessment staff and the TRC to undertake the tender evaluation process.

2.60 The ESP guidelines were updated in late September 2021 to ‘better reflect the current more streamlined approach to the procurement of employment services’ and limit the ‘need for repetition’.⁶⁵ Those changes to the ESP guidelines included:

• reducing its scope to procurement processes only; and
• halving the length and details included, with a view to moving more content from the guidelines into the relevant purchasing plans.⁶⁶ This was to include: the assessment guidance, the BA guide, staffing roles and responsibilities, and communication protocols.

2.61 The NESM 2022 Purchasing Plan was approved on 22 October 2021. This met the policy requirement to obtain the Delegate’s approval before the relevant RFP closing date (paragraph 2.59 refers). However, it did not set out a range of required guidance.⁶⁷ Rather, the purchasing plan:

• either indicated that those details would be developed separately in other guidance materials (that were subsequently approved by the Project Sponsor, rather than the Delegate); or
• referred back to the ESP guidelines as the primary source for more detailed information. This created a circular reference, with each document referencing the other.⁶⁸

2.62 The guidance omitted was included to varying degrees across the NESM Business Allocation Guideline, NESM Assessment Guide or other guidance documents developed throughout November and December 2021. These were developed during the first stages of the assessment process and were not made available to all assessment staff. The assessment guide was of particular importance because it was a key reference point for a predominantly contracted workforce of assessors (see Appendix 6). The impact of the delayed guidance on the assessment process is discussed from paragraph 3.26 to 3.28.

2.63 In September 2023, the department advised the ANAO that it had reviewed its ESP Guidelines and Purchasing Plan template after the NESM process in February 2023 and ‘a number of changes were made with the reintroduction of certain items.’

Were appropriate probity arrangements in place commensurate with the risk and scale of the procurement?

2.64 The CPRs require that procuring entities act ethically throughout the procurement, including by recognising and dealing with actual, potential and perceived conflicts of interest and seeking appropriate internal or external advice when probity issues arise.⁶⁹

External probity adviser

2.65 The department’s policy framework requires that internal legal advice be sought for spending proposals greater than $1 million (GST inclusive) and for an independent, external probity adviser to be appointed for the procurement of employment services and related programs.⁷⁰

2.66 A procurement process for an external probity adviser for ‘employment services tenders’ was conducted during June 2020. Six firms were invited on 15 June 2020 to submit quotes, with one of those (Maddocks) being the incumbent adviser since August 2015. Two providers, Ashurst and Sparke Helmore, were appointed on 2 July 2020 to ‘deliver independent probity and procurement advice’, as required. Ashurst primarily provided advice for the NESM and Transition to Work (TtW) procurements, while Sparke Helmore provided advice on aspects of TtW, Local Jobs Program, Self-Employment Assistance and the Entrepreneurship Facilitators Program. The previous probity adviser, Maddocks, continued to provide services for other matters.

Independence of the probity adviser

2.67 Probity experts should be independent and free from all conflicts of interest and have sound knowledge of all relevant government policies and procedures.⁷¹ Department of Finance (Finance) guidance outlines the following three probity expert roles: procurement process advisers, probity advisers, and probity auditors. While some overlap between roles is acceptable, entities should carefully consider the effects of this on the probity of the procurement process.

2.68 The NESM external probity adviser role was outlined in the request for quotes, which included both ‘probity services’ and ‘procurement services.’ While Ashurst had a defined role in the NESM probity plan, a large component of the work undertaken related to the provision of procurement or legal advice. The ANAO observed that in practice, Ashurst assisted with tasks such as the drafting of procurement planning and RFP documentation, including the NESM probity plan, materially contributing to discussions about the merits of proposals during TRC deliberations (which is expanded upon in paragraph 3.93), and providing probity sign-off on the TRC deliberations. This approach can represent a risk to the independence of the probity adviser as there is not a clear separation between the tasks or responsibilities undertaken.

2.69 Ashurst had been engaged by the department since January 2019 for up to eight separate services, including as the probity adviser across several other procurements. Over that period, 60 per cent of the department’s external legal services (which includes probity advice) have been provided by three firms, including Ashurst and Sparke Helmore.⁷²

Probity requirements for NESM personnel

2.73 The Employment Services Procurement (ESP) Guidelines set out that all project personnel and consultants involved in procurement activities, must execute a general ‘Procurement Activities Confidentiality and Conflict of Interest (CCoI) Deed Poll’. In addition, specific probity requirements were set out in the NESM Probity Plan, approved by the Delegate on 8 January 2021. The purpose of the plan was to ‘set out the principles and protocols that apply to managing probity in relation to the NESM Project.’ Project personnel were required to:

• attend a probity briefing regarding the NESM project;
• attend probity training;
• complete and provide a NESM probity undertaking, attached to the NESM Probity Plan;
• complete a NESM Conflict of Interest Declaration.

Probity registers

2.74 According to the probity plan, the external probity adviser was responsible for maintaining both the probity undertakings register and the probity issues register. The completion of probity undertakings, signed confidentiality deeds and attendance at probity briefings were to be recorded in the probity undertakings register, with any significant probity issues, and the respective mitigation strategies, to be captured in the issues register.

2.75 All ‘project personnel’ were to complete conflict of interest (COI) declarations through an online process and were to update those declarations as required.⁷³ A departmental ‘Internal Probity Coordinator’ was responsible for maintaining a register of all COI declarations and escalating material issues to the project manager (the branch head responsible for the procurement).

Communication with potential suppliers

2.76 A range of protocols were outlined for project personnel needing to engage with potential employment service providers. This included principles to be applied during the conduct of the NESM procurement and business-as-usual activities. A range of other potential interactions that could have probity implications were covered, including attendance at employment services related conferences and events; offers of any gifts or hospitality from potential suppliers; and pre-existing social relationships with or any offers of employment by potential suppliers. The department’s management of post-separation employment issues are discussed from paragraph 3.85.

3.1 Following the close of submissions for the NESM procurement on 22 October 2021, the department was to assess compliant proposals against the published evaluation criteria. Assessment results were to inform the Tender Review Committee’s (TRC’s) identification of the proposals offering the best value for money and recommendations to the Delegate for each of the 51 employment regions.⁷⁴

Were tender proposals received in accordance with the method and date specified in the request documentation?

3.2 Responses to the NESM tender process were to include all forms provided in the RFP response pack and be submitted via the department’s 360Pro system. This was accessible via a link on the department’s website. Responses were to include written claims against the published selection criteria, along with any supporting documentation.

3.3 Prior to the closing date and time, respondents could re-open their submitted proposals to make changes and re-submit their amended response. Once applications had closed, only responses with a ‘submitted’ status were to be considered by the department.

3.4 In total, 155 respondents lodged complete responses on time, with 117 submitting proposals for Enhanced Services, 51 for Career Transition Assistance (CTA) and 76 for Employability Skills Training (EST). Only proposals received through the 360Pro system were accepted and assessed by the department.

3.5 Between 22 and 23 October 2021, the Tender Opening Committee and witnesses, including the external probity adviser, oversaw the opening of the 360Pro tender box.⁷⁵ The opening process involved downloading an initial report from 360Pro containing the list of respondents, followed by the tender submission documents. All responses had been downloaded by the afternoon of 23 October 2021. A quality assurance process was undertaken of the transfer of data from 360Pro to the department’s Procurement and Licence Management System (PaLMS) on 25 October 2021 (see paragraph 3.12).

3.6 Three respondents requested permission to make late submissions on the basis of technical issues.⁷⁶ After conducting further inquiries in relation to two of these, the department notified all three that it would not accept late responses.⁷⁷

Were non-compliant proposals removed from consideration prior to the evaluation process?

3.7 The NESM Request for Proposal (RFP) set out a range of mandatory ‘conditions for participation’ and ‘minimum content and format requirements’ that needed to be met by respondents. Compliance with these were assessed by the department in the first stage of the assessment process, with any non-compliant responses to be excluded from continuing in the evaluation process. These requirements included, among other things:

• submitting a complete proposal, including addressing all selection criteria, completing all forms and providing all required supporting information;
• a copy of a Statement of Tax record (STR) or a receipt demonstrating that the STR(s) had been requested from the Australian Taxation Office (ATO)⁷⁸; and
• not be named as not complying with the Workplace Gender Equality Act 2012; or named in Regulation 40 of the Charter of the United Nations (Dealing with Assets) Regulation 2008 (the UN Charter).

3.8 If any non-compliance was identified as being due to an unintentional error of form, the department had discretion to give those respondents the opportunity to correct or clarify, provided it was not a material alteration or addition to the response. If this discretion was exercised, the same opportunity was to be given to all respondents in the same position.

Conformance assessment

3.9 The department conducted Stage 1 of the assessment process between 27 October and 16 November 2021. This involved the receipt and screening of responses and a ‘conformance assessment’ against the mandatory requirements outlined in the RFP for all NESM proposals.

3.10 A system control within 360Pro was relied on to prevent the submission of incomplete applications. However, this system control was limited to incomplete free text responses. This resulted in proposals being submitted which either had incorrect attachments or no attachment where one was required.⁷⁹

3.11 In total, 155 respondents submitted complete proposals across the Enhanced Services, EST and CTA service areas, which were migrated into PaLMS. The total number of proposals submitted for Enhanced Services comprised 1222 responses from 117 respondents out of which, 1192 proposals from 113 respondents were assessed as being conformant.

3.12 Following migration, a spreadsheet was used to check that the manual export of proposals out of 360Pro had been successful.⁸⁰ A manual public domain check of each respondent was also to be conducted as part of the conformance and checking process.⁸¹ Through these processes, the department identified:

• five respondents with public domain issues;⁸²
• 45 respondents with incorrect or no statement of tax record; and
• 12 respondents with out of date or incorrect letters of compliance with the Workplace and Gender Equality Act (WGE Act).⁸³

Conformance with Statement of Tax Record policy

3.13 In accordance with the Shadow Economy Procurement Connected Policy, respondents were required to provide a statement from the ATO showing that they have a satisfactory tax record.⁸⁴

3.14 Shortly before the NESM RFP was published in early October 2021, the department identified that several submissions for its Transition to Work (TtW) tender process were non-compliant with the STR requirements.⁸⁵ After seeking advice from its external probity adviser, the department took a consistent approach to STR compliance across both procurements. This involved requesting STRs from respondents where one had not been provided or where an STR receipt was provided.

3.15 Following conformance assessments for all NESM respondents, the department issued letters in early November 2021 to 45 respondents requesting that the correct STR documents be provided.⁸⁶ Out of the 45, 20 respondents had submitted a receipt confirming their request for an STR, two had submitted incorrect documents, and 23 had nominated themselves or their partners as requiring but had not provided an STR. Following contact from the department:

• 25 respondents provided valid STRs issued within the required four-day timeframe; and
• 17 claimed an error of form stating that STRs requirements for partners, trustees or consolidated tax groups were not applicable to them, which was accepted by the department.

3.16 Six of the contacted respondents (which included four Enhanced Services respondents) did not provide STRs dated within the required four-day timeframe and were therefore deemed non-conformant and excluded from further assessment.⁸⁷

Were compliant proposals assessed in accordance with the published evaluation process?

3.17 As outlined from paragraph 2.55 to 2.57, the RFP provided a high-level overview of the evaluation process. The detail of that process was developed iteratively between July and December 2021. This was documented in the department’s internal guidance materials, which were approved incrementally between late September and November 2021 (see paragraph 2.62).⁸⁸

Tender Assessment Centres

3.18 Planning for concurrent assessment processes for the TtW and NESM procurements commenced in July 2021 and continued to run in parallel with the RFP and assessment processes. A ‘decentralised assessment model’ was adopted, which involved establishing several locations for state and territory ‘Tender Assessment Centres’ (TACs).⁸⁹ Each TAC was to have a co-located team of managers, assessors and moderators throughout the evaluation period. Assessments against criterion one were to take place in Canberra and criteria two and three were to be assessed by TAC staff with local region knowledge and expertise. In advice to the Delegate in September 2021, the department outlined that this decentralised model, in conjunction with a ‘highly flexible staff recruitment, onboarding, and training model’ and the implementation of PaLMS, would mitigate, to the extent possible, any delays to the delivery of procurement outcomes within the Government’s timeframes. The Government had committed to the roll out of the NESM from 1 July 2022, which involved the department running seven procurement processes between August 2021 and April 2022.⁹⁰

3.19 By early September 2021, fully staffed TACs were established in Canberra, Brisbane and Adelaide, with Melbourne partly staffed as a moderation-only centre.⁹¹ By the end of September 2021, a number of other smaller assessment and moderation ‘hubs’ were established in Bendigo, Newcastle, Perth, Darwin and Hobart.⁹²

Staffing and roles

3.20 By mid-September 2021, the department identified that limited subject matter experts were available internally and it would be reliant on a larger than expected pool of inexperienced or contracted assessors.⁹³ As this reduced the amount of local area knowledge of the TACs, the department took steps to offset this by commencing development of employment region-specific guidance materials. Appendix 6 provides an overview of the various roles held by departmental and contracted staff throughout the procurement.

3.21 A moderation process was to occur at least once for each assessment conducted.⁹⁴ Moderators were responsible for overseeing the quality, consistency and completeness of assessments undertaken by their assigned assessors. Each moderator was assigned oversight of between three and five assessment staff. Moderators were not to participate in or independently undertake any assessments.⁹⁵ An overview of the assessment process from tender submission through to the execution of work orders is provided in Figure 3.1.

Assessment of proposals against the selection criteria

3.22 Stage 2 of the assessment process involved the assessment of conformant proposals against the three published selection criteria (or ‘areas of capability’). Stage 2 commenced in Canberra on 16 November 2021 involving the assessment of 149 responses to the first selection criterion. These assessments related to all proposals received across Enhanced Services, EST and CTA, as providers were required to respond only once for that criterion (discussed at paragraph 2.46).

3.23 As criterion one assessments were finalised, the individual responses for Enhanced Services proposals against criterion two (1176 in total) and criterion three (972 responses) were allocated to state/territory TACs for assessment.⁹⁶ All assessments for the 1176 Enhanced Services proposals across the 51 employment regions were finalised by 18 January 2022, prior to the commencement of the Business Allocation stage.⁹⁷

Figure 3.1: Overview of the assessment process for Enhanced Services proposals

Source: ANAO analysis.

3.24 All notes and assessment results were to be recorded in electronic format in PaLMS. Assessors were prompted to select the most appropriate qualitative rating, or ‘evaluation descriptor’, for each aspect of the response being assessed (see paragraphs 2.49 to 2.50).⁹⁸ Ratings were then to be moderated, and in the case of any disagreement, sent to an adjudicator.⁹⁹ In total, 34 out of 1176 criterion two assessments (2.9 per cent) and 24 out of 972 criterion three assessments (2.5 per cent) were adjudicated.

3.25 These ratings formed the basis for system-generated scores, which were mapped to numerical scales in PaLMS (Figure 2.1 provides an overview of these mappings). This design was to enable qualitative assessments to be converted into quantitative scores for the sub-criteria. These were then weighted and aggregated by PaLMS to generate total scores and orders of merit for each employment region, and were not visible to staff. Scores were not visible to staff involved in the criteria assessment process. The delayed delivery of PaLMS and its impact on the quality of the assessment process is discussed further from paragraph 3.29.

Training and guidance

3.26 The delivery of comprehensive training was recorded in the NESM risk plan on 3 September 2021 as a possible mitigation for the risks of running concurrent procurement processes (see Figure 1.2). Development of the training commenced in August 2021 for both the TtW and NESM procurements. Training packages were to be delivered before the assessment process commenced.

3.27 Training for the TtW procurement was prioritised, as it was the first to commence assessments on 27 September 2021. By 26 October 2021, delays in the TtW process had resulted in an overlap with the NESM process, reducing the originally planned training period from around four weeks down to a few days. The delivery modes were also reduced from face-to-face training and including multiple modules on probity, assessment theory, case studies, to self-directed and on-the-job training. Key guidance, including the NESM Assessment Guide, was finalised on 5 November 2021, around a week after Stage 1 assessments had commenced (see Table 2.4). Similarly, guidance for the Business Allocation stage was developed shortly before or in parallel with that stage of the process, which began on 22 January 2022 (see paragraph 3.66 and Appendix 4).

3.28 In mid-December 2021, the department conducted a review to identify the reasons for the slower than expected progress for the NESM assessments. The feedback from assessors and moderators reflected two main themes:

• insufficient training and limited access to guidance materials and experienced supervisors; and
• ongoing inefficiencies due to system limitations and technical difficulties with PaLMS (discussed from paragraph 3.31).

Procurement and Licence Management System (PaLMS)

3.29 In October 2020, the Australian Government announced a $295.9 million investment for the delivery of the first tranche of the ‘New Employment Services Model digital platform’ (the digital platform).¹⁰⁰ The digital platform was considered a ‘fundamental element’ of the NESM and essential for achieving the expected outcomes and benefits from the reforms. Tranche 1 was to be delivered progressively between January 2021 and June 2022, and was to include, among other things:

• base capabilities for digital service channels, including web and mobile applications;
• foundational digital service channels for users of the platform, including tailored and targeted messaging and a Digital Services Contact Centre¹⁰¹; and
• an end-to-end procurement and contract management solution to replace the department’s current systems.¹⁰²

3.30 The planning and design process for the procurement and contracting system commenced in early 2021. The department advised Ministers in April 2021 that the system — later named PaLMS — was to be delivered by 30 June 2021 to support the NESM approach to market. Risks relating to the implementation timeframe and the inter-dependencies between PaLMS and the upcoming procurement assessment processes were recorded in the NESM Risk Plan.¹⁰³

3.31 By early June 2021, the delivery of PaLMS had been delayed and the pre-existing tender system, 360Pro, was included in the exposure draft and final version of the NESM RFP. Following the close of the RFP, submission data was exported from 360Pro into PaLMS. This process was conducted over two days, with not all data able to be transferred. Some of these data fields were accessed separately during later stages of the process as they were identified as missing by assessment teams.¹⁰⁴

3.32 As implementation risks with the broader digital platform project were realised in November 2021, a ‘Tiger Team’ was established ‘to review and prioritise the critical deliverables for achievement of the [Minimum Acceptance Criteria]’. Parallel to this, NESM assessment staff raised concerns with the functionality of PaLMS through focus groups conducted in December 2021 (see paragraph 3.28). Moderators and assessors noted the following system issues had caused delays and disruptions.

• Details, such as tender responses and feedback from moderators, did not always appear upon opening in PaLMS, requiring assessors to exit the browser multiple times.
• Logging in at the start of each day took between 20 to 40 minutes and using the ‘back’ function did not always navigate to the correct pages.
• Despite expressly saving work, comments were not always saved, leading to loss of time and effort.

3.33 To work around these issues, Microsoft Word was used first before pasting results into PaLMS. This approach reduced the probity, efficiency and security benefits to be gained from using PaLMS, as documented by the department in May 2021.

3.34 In late February 2022, the department provided the government with a yearly progress update on the implementation of the digital platform. In respect to PaLMS, it reported that:

• the first key deliverable, the Tender Management System release one, was delivered on 30 June 2021¹⁰⁵;
• progressive releases of functionality of the Tender Management System were delivered by 31 December 2021; and
• a fully integrated workflow for the Tender Management System and delivery of the related Contract and Licence Management system was on track for completion by 30 June 2022.¹⁰⁶

3.35 As at June 2022, work to transfer the remaining data fields from 360Pro to PaLMS remained ongoing. Concerns that the PaLMS deliverables had been overstated were raised by the area responsible for the NESM procurement on at least five occasions between July 2021 and June 2022. This included advising the Senior Responsible Officer for the digital platform in late March 2022 that:

… the tender submission element of PaLMS had not been completed, and that the data connection between PaLMS tender management and PaLMS contract management had not been completed. Which, for the department meant that the front end, and the connection in the middle of the end-to-end capability, had not been built.

3.36 The extent of the outstanding deliverables was also examined in a post-procurement review of PaLMS in June 2022. Through this review the department identified that PaLMS was either partially or fully underdeveloped from the business allocation stage to the end of the procurement process.

Specialist proposals outside of the published categories

3.37 As outlined at paragraph 2.34, the RFP provided indicative numbers of licences to be offered, including how many should go to cohort specialists and in which regions. While respondents could submit proposals for specialist services for a cohort outside of those listed and in regions not earmarked by the RFP, they needed to demonstrate that an ‘unmet demand’ existed. Where this was not demonstrated ‘to the department’s satisfaction’, proposals would ‘not progress to evaluation’. A short guidance document for the assessment process for these proposals was developed in late November 2021.¹⁰⁷

3.38 In total, 19 ‘Specialist–Other’ proposals were submitted by the RFP closing date and reviewed in early December 2021. In contrast to other proposals, this step was conducted outside the PaLMS workflow. One such proposal was deemed to have demonstrated an unmet need in the region and progressed to further evaluation. The remaining 18 proposals from eight respondents were excluded from further consideration and deactivated in PaLMS, and at the time of this audit, these proposals had an ‘inactive’ status within PaLMS¹⁰⁸, and were not provided to the Delegate for consideration.

3.39 The sole Specialist-Other proposal progressing to further evaluation stages aimed to address an unmet need for Indigenous specialist services in that region.¹⁰⁹ The same provider had also submitted a proposal for the Indigenous cohort in the adjoining employment region, which, according to the RFP, had a higher average caseload for Indigenous job seekers and similar characteristics.¹¹⁰ That application was assessed as not identifying or addressing any unmet need in the region.¹¹¹

Demonstrated performance of existing providers

3.40 Two of the nine questions for the first criterion were designed to assess the ‘demonstrated performance’ of potential providers in respect to the ‘quality of service, contractual compliance, and achieving measurable outcomes for disadvantaged cohorts.’ As discussed at paragraph 2.52, the RFP outlined that the department would use ‘current performance and other quantitative data held by the department’ for assessing proposals from existing jobactive providers and that existing providers ‘will have the option to NOT provide additional information in response to this question’ [emphasis in original].

3.41 In this context, the NESM Assessment Guide outlined to assessors that ‘Assessment results will be generated for selection criterion Questions 1.8 and 1.9 for Respondents currently contracted by the department, based on performance data held by the department.’ In practice, this involved using existing providers’ star ratings for assessing their past performance.¹¹²

3.42 The department relied on calculated ‘average’ star ratings (averaged nationally), which were used for all proposals submitted by existing providers irrespective of whether they were tendering for a new employment region or one in which they were already delivering services. This methodology was not published in the RFP. It was developed in November 2021 and involved the calculation of a national average star rating for each provider. That average rating was used to generate a score out of 10 for sub-criterion 1.8, which represented 11 per cent of the first criterion and two per cent of the overall score.

3.43 Generally, the more sites and employment regions serviced by a provider, the greater the range of its individual site star ratings. This affected the organisational performance scores for providers’ sites. Accordingly, providers most impacted by this methodology were large organisations with between 75 and 204 sites across 10 to 28 employment regions.¹¹³ Out of the 274 proposals submitted by incumbent providers, 22 per cent were under-rated, and 36 per cent were over-rated.¹¹⁴ By way of example, one provider that submitted proposals in 18 employment regions received an average star rating of 2.66. For one of its proposals, the provider’s actual star rating for that region was 5 stars. Applying the average star rating resulted in that proposal scoring 5 out of 10 for demonstrated performance in this region, whereas it would have scored 8.75 out of 10 if the region-specific rating had been applied.

Compliance against the department’s IT security framework

3.44 Stage 3 of the evaluation process comprised the ‘Right Fit For Risk’ (RFFR) assessments, which were conducted as a discrete activity outside the PaLMS workflow between late October 2021 and late January 2022.

3.45 In accordance with the RFP, respondents were required to complete the RFFR form to demonstrate their capacity to comply with the obligations under the department’s External Systems Assurance Framework (ESAF).¹¹⁵ Pre-existing providers already accredited under the framework were not required to complete the RFFR form as they were to be required to maintain their accreditation.

3.46 The department’s Digital Information Assurance Section reviewed 153 RFFR forms, from which 56 respondents (35 per cent) achieved RFFR accreditation and were assessed as ‘low risk’.¹¹⁶ It is not clear that the RFFR assessment results were used to inform the appointment of providers to the national panel and DoSO arrangements. For example, out of the remaining unaccredited respondents:

• 39 were assessed as ‘medium risk’ with indications that the provider would achieve accreditation (five received licences);
• 53 were assessed as ‘medium risk’ and requiring additional caveats relating to their information security practices (26 received licences); and
• five assessed as ‘high risk’ and not meeting minimum security requirements (no licences).¹¹⁷

3.47 No licences were issued to the five respondents assessed as ‘high risk’, one was appointed to the sub-panels for three employment regions, with no caveats or mitigation strategies recorded.¹¹⁸ In November 2023 correspondence to the ANAO, the department noted that the NESM RFP outlined that ‘[a]ll offers of business made by the department will be conditional on progress against and achievement of accreditation under the ESAF within the timeframe advised by the department.’

Financial viability assessments

3.48 Stage 4 of the evaluation process comprised financial viability assessments, which occurred in parallel with Stages 1 and 2 from 25 October 2021 to 14 January 2022.¹¹⁹ During this time, financial viability assessments of 164 NESM respondents were conducted.¹²⁰ The results were compiled outside of the PaLMS workflow in a spreadsheet and provided to staff conducting the Business Allocation (BA) processes in mid-January 2022.

Quality assurance process undertaken by department

3.49 A well-developed quality assurance process can enhance the robustness of an assessment process. The quality assurance approach adopted for the NESM had limited ability to identify global or systemic issues until after the completion of the assessment process.¹²¹ The NESM Purchasing Plan outlined that quality assurance (QA) activities would be scoped throughout the procurement by the QA Team.¹²² To this end, the development of a ‘Quality Assurance Strategy’ remained underway as at November 2021.

3.50 The strategy focused on alignment and consistency between assessors’ written rationales (or ‘strengthening statements’) with the selected evaluation descriptors. The QA of individual assessment ratings commenced on 25 November 2021, with reviews conducted:

• for the first assessment completed by each assessor for each of the three criteria; and
• for up to 10 per cent of criteria two and three assessments, including a combination of: all reassessed assessments; assessments identified as outliers due to having ‘very low or very high scoring results’; assessments by assessors/moderators with immaterial score variances; and reviews of assessments requested by the National Assessment Manager.¹²³

3.51 In a February 2022 minute to the chairs of the three NESM Tender Review Committees, the following were reported as the ‘most significant’ issues: evaluation descriptors being unsupported by analysis or the respondent’s evidence; strengthening statements being insufficient; and justifications not supporting the chosen evaluation descriptor.

Assessment issues identified during Business Allocation

3.52 During the Business Allocation process, unusual variations were identified between scores awarded to proposals submitted by the same provider in multiple employment regions. These employment regions were considered to have similar labour markets and job seeker demographics. Specifically, business allocation teams identified and requested quality assurance checks for assessments undertaken in two states.

3.53 In one of those states, three providers were identified across multiple regions, where the scores for their proposals appeared inconsistent with the quality of their proposals or indicated some misalignment in relation to the assessment. For example, one tenderer had two proposals in neighbouring regions, where one proposal had received 4.97 (ranked twentieth) and another had received 7.85 (ranked first) for near-identical responses. Had the score for this lower-scoring proposal been more closely aligned with the higher scoring proposal, it would have ranked higher than another provider in that region ranked sixth with a score of 6.91 and receiving a licence. The score for the lower-scoring proposal was not re-assessed due to an oversight. With respect to another tenderer with multiple proposals, the business allocation team noted that it was ‘difficult to see and understand how [the provider] was awarded such high descriptors and scores’ given their generic response to the criteria.

3.54 In the other state in which this issue was raised, the business allocation team requested adjudication to review the large variation between the score for one of the proposals compared with the other six regions for which the tenderer had applied. ¹²⁴ As the initial adjudication request was rejected on the basis that ‘a robust QA process has been undertaken,’ the team subsequently performed its own analysis on the proposal in question, finding that there were inconsistencies around the moderated scores for this tenderer being higher compared to scores received by the same tenderer in other regions. An additional issue was identified in relation to another tenderer receiving a lower score in this region compared to its other proposals, meaning that in this region, it was ranked thirteenth (in contrast to being ranked first in two other regions).¹²⁵ Following a second request from the team in the context of the ‘national consistency review’ (this process is explained at paragraph 3.73), adjudication was undertaken, confirming the BA team’s analysis.¹²⁶ A further request for the adjudication of scores in this region was made by the TRC in mid-February 2022. This was based on the TRC’s observations that some scores did not reflect the quality of proposals.¹²⁷ This region is discussed further at paragraph 4.30.

3.55 Consistent with the department’s observations through its QA processes, the ANAO’s analysis of assessment scores for the 47 respondents that applied in more than five employment regions indicated considerable variances in scores. Noting that some variation is to be expected between proposals across employment regions, large score variances occurred where responses were largely or almost the same.

Were the results of the evaluation process used to populate assessment reports and compare the merits of competing proposals?

3.56 The ‘Business Allocation’ (or BA) process has been a feature of procurement processes for employment services for at least nine years.¹²⁸ The BA process was designed to be conducted by State Office staff with local region knowledge after the merit assessments of all compliant proposals. The objective of the BA process during the NESM procurement was to produce preliminary recommendations at the employment region level for consideration by the TRC. This was to inform the TRC’s deliberations and assist with the development of the TRC’s advice and recommendations for the Delegate.

Business Allocation Guideline

3.57 The October 2021 NESM Purchasing Plan stated that when compiling recommendations, Business Allocation teams (BA teams) ‘must refer to the New Employment Services Model 2022 Business Allocation Guide’. While the guide had not yet been developed, it was to ‘be approved by the Project Sponsor’ (who was also the chair of the TRC for Enhanced Services). The ‘Business Allocation Guideline’ (BA Guideline) was subsequently approved on 10 November 2021, by the Senior Procurement Officer (or Project Manager).

3.58 In accordance with the BA Guideline, BA teams first determined which respondents would be invited to enter into a Deed of Standing Offer (DoSO) to deliver Enhanced Services or a Deed for the delivery of EST and/or CTA services. For Enhanced Services, respondents receiving more than 40 per cent (that is, ‘suitable’ or better) against selection criterion one were considered eligible for appointment to the ‘national panel’ and the sub-panel for the respective employment region. BA teams were then to compare the merits of competing proposals and recommend:

• which of those providers on the panels should be awarded a licence; and
• the percentage of market share to be awarded to licensed providers.

3.59 Recommendations for the award of licences were to be informed by two main principles outlined in the BA Guideline as follows.

• Principle #1: Preliminary allocation of business will be based on the Respondent’s assessment scores, with preference given to higher ranked Respondents.
• Principle #2: Higher ranked Respondents will be allocated a greater share of the business than lower ranked Respondents (subject to consideration of the ranges of higher ranked Respondents).

3.60 A ‘comparative assessment’ was to be produced as part of a ‘Business Allocation Report’ (BA Report) for each region, with the rationale as to why a respondent was ‘offered business’ to be documented. This included recording where the proposed offer of business was:

[f]or example, according to the initial ranking of Respondents, or where higher ranked Respondents have been skipped over in favour of allocating business to lower ranked Respondents and the reason for this decision.

3.61 In any instances where lower ranked respondents were to be recommended, the BA team was to ‘demonstrate the basis for [the] recommendation through the comparative assessment’ based on the ‘relative strengths of each Respondent’s offer in contributing to value for money, including meeting the program/service objectives.’ In relation to the scores already allocated to proposals through the merit assessment process, the BA Guideline outlined that:

For the absence of doubt, assessment scores and rankings are fixed and must not be adjusted or tampered with during the BA process and/or comparative assessment process.

Identifying ‘Special Conditions’

3.62 BA teams were also responsible for identifying any ‘Special Conditions’ offered within proposals and determining whether they should be included in the BA recommendations.¹²⁹ The TRC was to undertake a ‘final review’ of any special conditions identified through the BA process to determine ‘whether they constitute a material impact on awarding licence(s).’¹³⁰

3.63 Following the BA process, ‘business allocation reports’ were drafted to capture the outcomes of those processes. These were subject to a QA process before provision to the TRC. Special conditions were identified in the reports for 19 proposals in eight employment regions. However, these were not discussed in TRC proceedings (see further at paragraph 4.19), identified in the report to the Delegate (see paragraph 4.37), or included in the executed work orders (see paragraphs 4.40 and 4.41). Work undertaken by the department in 2022 and 2023 to monitor the delivery of special conditions by providers is discussed from paragraph 4.46.

Appointment to the panel

3.64 Once appointed to a sub-panel, the BA Team could then recommend providers be awarded a licence to deliver Enhanced Services in all or part of that employment region. As outlined at paragraph 3.58, the threshold for determining whether a respondent ‘suitable’ for panel membership was a score of more than 40 per cent against criterion one. The scores achieved by respondents against the second and third criteria did not inform the membership of the panels. While criteria two and three were weighted at 40 per cent each, criterion one was worth 20 per cent of the overall score.¹³¹ Each criterion was awarded a score out of 10, which was then adjusted within PaLMS for the relevant weightings.

3.65 While this methodology was not outlined in the RFP, it was initially foreshadowed in the September 2020 licensing framework, which noted ‘that any organisation assessed as suitable to deliver services in at least one Employment Region would be offered a place on the national panel’. In July 2021, the department published an Exposure Draft of the RFP, which outlined that the ‘organisational capacity’ criterion would determine eligibility for appointment to the national panel. The outcomes of this approach are discussed at paragraphs 4.14 and 4.15 and illustrated by Figure 4.1.

Business Allocation process

3.66 The BA process commenced across state offices on 22 January 2022 and was completed by 5 February 2022. BA reports were endorsed by the respective state managers by 11 February 2022. As outlined in Appendix 4, guidance for the BA process was developed shortly before or in parallel with the process. Online training was delivered during the two days prior to commencement on 20 and 21 January 2022, with the necessary guidance provided incrementally between 22 and 31 January 2022.¹³²

3.67 The delayed guidance led to teams having to clarify issues or raise requests for clarity raised through email and Microsoft Teams messages, demonstrating a lack of adequate planning and communication within the geographically dispersed assessment environment.¹³³ Examples of questions raised throughout training and during the BA process included:

• whether the number of licences included in the RFP could be exceeded, with guidance limited to relying on a licence viability tool for recent caseload figures¹³⁴;
• questions around how far down the order of merit to go and if there were any broad principles to follow;
• requests for clarity around the TRC’s expectations for allocating business share to specialist providers; and
• the availability of and corrections to templates, and notifications for and updates on recently identified or known issues.

3.68 The Business Allocation Information Tool (BAIT) was a key spreadsheet provided to BA teams to develop and record recommendations. It comprised an export of the tender submission data from PaLMS for each state and region, along with rankings, scores and information on site locations. As discrepancies were identified throughout the process, concerns over the reliability of the BAIT were raised by BA teams. The BAIT was revised and re-issued at least three times between 22 and 25 January 2022 due to these issues.¹³⁵

Ranking of generalist and specialist proposals

3.69 A key matter requiring clarification was the approach adopted for the ranking of generalist and specialist proposals against one another in the same employment region. The Senior Procurement Officer was advised of this on 28 January 2022 by one state Business Allocation team, who observed that specialist providers were ranked substantially lower than generalist providers in three employment regions.¹³⁶ It noted that ranking these proposals against each other may result in no specialist proposals being recommended in these regions. This was inconsistent with the NESM policy intention to make specialist providers more accessible for specific job seeker cohorts.

3.70 While no clear resolution was reached in respect to this issue at the business allocation stage, the TRC’s recommendations resulted in 80.65 per cent of regions earmarked as viable for specialist services having one or more licensed specialist providers. In total, 138 generalist licences and 43 specialist licences were recommended across the 51 employment regions.

National consistency review

3.73 The BA Guidelines outlined that a ‘national consistency review’ was to be undertaken by BA teams where respondents had submitted proposals in multiple regions. The purpose of the review was to ensure consistency and balance in the allocation of business to providers across state and territory boundaries. Where any adjustments were made to preliminary recommendations as a result of the national consistency review, these were to be documented and justified. This was the extent of the guidance provided to business allocation staff.¹³⁷

3.74 The inconsistences between assessment scores discussed between paragraphs 3.52 and 3.55 became more apparent during the national consistency review, and later, during TRC deliberations. BA staff were asked to not question specific assessment scores when concerns were raised around the scoring variations or inconsistencies. Therefore, the national consistency review did not contribute towards rectifying any score outliers or errors. Where assessment scores can be relied upon, the national consistency review process would assist to strengthen the recommendations provided to the TRC.

Were procurements conducted ethically and in compliance with internal probity frameworks, including identification and management of any conflicts of interest?

3.75 Well-defined probity arrangements are important for the ongoing and effective management of probity risks in complex procurements. As outlined from paragraph 2.73 to 2.76, the department established specific arrangements through the NESM probity plan, which was first approved by the Delegate in January 2021.¹³⁸ Throughout the NESM procurements, the external probity adviser, Ashurst, and the ‘Internal Probity Coordinator’ were to work together to manage probity issues in accordance with the plan. The ANAO examined whether these arrangements were effective in identifying and managing the relevant probity risks.

Probity registers

3.76 As outlined at paragraph 2.74, the probity undertakings and the probity issues registers were to be established to track attendance at probity briefings and the completion of relevant declarations by all ‘project personnel’. These personnel were defined as either APS employees or external service providers with involvement in and/or access to any confidential information relating to the NESM project.¹³⁹ While these arrangements did not explicitly apply to ministers or their staff, the department sought to provide probity briefings to its portfolio ministers and their staff (discussed further from paragraph 3.90).¹⁴⁰

3.77 The department’s administration of its probity registers deviated in some respects from the requirements in the probity plan. This included the department maintaining the probity registers instead of Ashurst, and the establishment of a fourth ‘Consultation Register’ not identified in the probity plan. The four registers were each maintained by the department.

Recording of probity issues

3.78 While the ANAO identified completeness and accuracy issues across all four probity registers, this was particularly the case for the probity issues register. The probity plan required the ‘External Probity Adviser [to] record the occurrence of all meetings involving Potential Suppliers which relate to the NESM Project in the Probity Issues Register.’¹⁴¹ However, these were recorded separately in the consultation register.¹⁴² The external probity adviser was required to make detailed notes of those meetings or, if absent, project personnel in attendance were to make those notes.¹⁴³ The department advised the ANAO in November 2023 that:

The addition of the fourth ‘consultation register’ was necessitated by the number of consultations and meetings being undertaken across the department and externally where NESM could potentially be raised. This was to mitigate the risk of not capturing these, being able to identify them easily and a more efficient method to manage as they were not probity issues as such.

3.79 The consultation register was populated manually after the submission of a completed ‘meeting record’ form via email to the probity team. This process relied on all departmental officials being aware of and complying with the relevant procedures.¹⁴⁴ While the consultation register had 229 entries as at April 2022, the meeting records for 31 of these were not filed. There was no indication whether the external probity adviser had attended any of the meetings in the register. The ANAO also identified completed forms for three more meetings not listed in the register. Of the 229 meetings recorded:

• In relation to discussion of the NESM project,
  • 74 (32 per cent) were for meetings with the NESM project recorded as discussed (47 of these included potential suppliers);
  • 148 (65 per cent) were meetings for business as usual purposes, with the NESM project not discussed; and
  • 7 (three per cent) had no recorded purpose;
• In relation to which office organised or held the meeting,
  • 140 (61 per cent) were for meetings held by state or regional office officials; and
  • 89 (39 per cent) were for meetings held by national office officials.

3.80 As at June 2022, the probity issues register comprised 28 issues or events that had occurred between February 2021 and June 2022. This was notably less than the 47 meetings with potential suppliers (at which the NESM was discussed) listed in the consultation register and the 177 actual or apparent/perceived COIs declared by project personnel from the 322 declarations recorded. The 28 matters on the probity issues register comprised:

• eight instances where procurement information was incorrectly or inadvertently shared;
• six instances of assessment staff being currently or previously employed by respondents;
• five instances where respondents had approached departmental officials through non-official channels about the procurement;
• two instances where assessors were able to access information in the PaLMS platform where that access should have been denied; and
• nine other events related to the reporting of a COI by the Senior Procurement Officer, the loss of a building and tender floor pass and other queries or issues rated as ‘low risk’.¹⁴⁵

3.81 The information recorded in the department’s registers was not analysed or used to inform strategic reporting, which would have enabled senior management to be appropriately informed of probity risk. While emerging probity issues were dealt with promptly, the ad hoc approach adopted resulted in some issues not being documented as required. For example:

• Mandatory COI declarations and probity undertakings were dispersed across three separate and mandatory processes.¹⁴⁶ Out of 711 project personnel recorded:
  • 209 (29 per cent) were recorded as having completed all three processes;
  • 127 (18 per cent) were recorded as completing at least one of the three¹⁴⁷; and
  • 375 (53 per cent) were recorded as not completing any of the three processes.¹⁴⁸
• Registers were not often updated to reflect why roles had been terminated where actual COIs had been identified. For example, 51 of the 79 assessors (eight of which had completed probity undertakings) recorded as having their contracts terminated had assessed at least one proposal prior to their contract termination. Out of these, 15 had COIs identified and 36 did not have a rationale recorded for the termination;
• meeting records being incomplete or inaccurate. For example, the records for an August 2021 meeting involving state office officials did not identify which providers were in attendance or reflect that the Minister and Assistant Minister were both present; and
• at least three probity issues where the department sought advice from Ashurst but had not recorded the details on the issues register as required, comprising:
  • one inadvertent disclosure of information to potential respondents in August 2021 (whereas eight similar incidences were recorded on the register); and
  • two ex-departmental officials that had accepted employment offers from providers prior to and during the NESM RFP period (six similar but lower risk incidences were recorded).

3.82 The department advised the ANAO in November 2023 that its range of COI declarations processes were being examined and a ‘streamlined process’ was under development. In respect to the range of declarations required for the NESM process, the department further advised that the:

direction for staff was that a single COI required completion not all three as they were all the same, that is, the general COI was the same as the NESM COI and the SES [Senior Executive Service] COI.

3.83 In this latter respect, and consistent with broader Australian Public Service (APS) policies, the department requires its SES employees to submit an annual Declaration of Personal Interests.¹⁴⁹ The department’s Conflict of Interest Policy states that the ‘completed declarations and the disclosure register are secured in the office of the Branch Manager, People Culture and Capability, with access limited to the Secretary and staff with a business requirement to access this information.’ The evidence available to the ANAO indicated that the status of relevant declarations for the 2021–22 period was sought by the internal probity coordinator on 14 July 2022, after Workforce Australia had commenced.¹⁵⁰

3.84 The size and scale of the NESM procurement combined with the limited access to annual SES declarations indicates there is merit in activity specific COI declarations being maintained for staff directly involved in procurement in addition to any general declarations required as part of their employment.¹⁵¹

Post-separation employment issues

3.85 As outlined at paragraph 2.76, any offers of employment made by potential respondents were to be reported as soon as practicable and recorded in the probity issues register. Additional restrictions apply to the employment of ex-SES APS officials, including a 12-month period of restrictions on lobbying activities for any matters on which they had official dealings as public servants.¹⁵² Where offers were accepted, the external probity adviser was to be consulted and consideration given to restricting the individual’s access to confidential information. During the NESM process, the department become aware of two such offers to project personnel as follows:

• an executive level official, working in the NESM area advised in writing of their plan to leave the Australian Public Service (APS) and accept a part-time role in 2021 with a potential NESM respondent; and
• without prior notification — as required by the department’s internal policy framework — a senior executive service (SES) official that had left the public service in late 2021 and commenced employment with a large jobactive provider within a couple of weeks of resigning.

3.86 The department promptly sought advice on both occasions from its external probity adviser. In respect to the first ex-official, the department implemented a post-separation arrangement in line with the mitigations proposed by the probity adviser and in the ex-official’s written notice. This involved reducing IT access; recusing themselves from any further work involving the provider and its subsidiaries; and no further involvement in the development of the NESM Performance Framework.¹⁵³ While post-separation arrangements were established, the department did not record this matter in its probity registers (see paragraph 3.80).

3.87 The department was made aware of the post-separation employment of the second ex-official during a business-as-usual discussion between the provider and state office officials in late 2021. The provider outlined that the ex-official was recently engaged to provide consultancy services and had been given NESM RFP response material to review. While the external probity adviser did not provide an overall risk rating for the incident, it advised the department that:

The post-separation employment of department personnel with potential suppliers creates a probity risk to the actual or perceived fairness of the RFP.

This risk arises as the former department employee’s knowledge of confidential information relating to the NESM or the department, obtained through their work for the department, could be used to provide their new employer an unfair advantage in the RFP process.

3.88 After exchanges of at least four letters at senior levels between the department, the provider and the ex-official in late 2021, the department accepted written assurances from both parties that they understood and accepted their respective ethical duties and confidentiality obligations.¹⁵⁴

3.89 The department also remained aware that an ex-SES official was employed, at least since early 2019, by a large jobactive provider.¹⁵⁵ The ex-official was a key contact point for the department’s contractual arrangements with that provider. At least six NESM project personnel remained in contact with the ex-official since their last role in the department ended in 2017. While each of the personnel completed at least one type of COI or probity declaration during the NESM project period, only one identified this relationship as a potential COI.¹⁵⁶ This was documented in an annual COI declaration in April 2022, after the Delegate’s decisions in March 2022. Noting that work on the NESM had not commenced in 2017, the department advised the ANAO in March 2023 that it was not aware of any further mitigations or post-employment arrangements being in place.

Other probity arrangements

3.90 Probity risks associated with lobbying activities by potential suppliers are inherently increased in the lead up to and throughout a high-value procurement process. Demonstrating an awareness of this risk, the department arranged, from time to time, for probity information to be provided to its portfolio ministers, although this was not required by the probity plan. For example, probity briefings with the external probity adviser were arranged:

• in February 2021 for the Assistant Minister for Youth and Employment Services and relevant ministerial staff; and
• in April 2021 for the relevant staff and advisers to the Minister for Employment, Workforce, Skills, Small and Family Business (the Minister). At that stage, the Minister was unable to attend and was therefore not briefed by Ashurst.

3.91 Following a disclosure by the responsible Minister at a public forum in early May 2021, the department sought to provide a probity briefing for the Minister, which was delivered by senior departmental officials on 27 May 2021.¹⁵⁷ Consistent with the probity plan, the department recorded this disclosure on the probity issues register and obtained advice from Ashurst on 18 May 2021 prior to arranging the briefing. While Ashurst advised the department that the consequences of this event were ‘low and acceptable’, it was recommended that to reduce the consequences of the incident, and the risk of further incidents, the department arrange a briefing for the Minister (to be delivered by Ashurst) ‘as soon as possible’ and that current providers be advised that the number of providers to be in the new market had not been settled.¹⁵⁸

3.92 All ministerial staff in attendance at probity briefings were asked to complete probity undertakings and declare any COIs in writing. By 9 June 2021, completed forms had been returned for 10 of the 17 staff members and documented by the department.¹⁵⁹

External probity adviser’s role in Tender Review Committee and probity sign-off

3.93 During TRC proceedings, the external probity adviser contributed to or provided advice on matters that would ordinarily be expected as being outside the role of an external and independent probity adviser. Most of this advice was ‘procurement advice’, which was a consequence of the department’s decision to engage Ashurst as both a probity and procurement adviser (outlined in paragraph 2.68). Out of the 42 regions reviewed, the external probity adviser provided procurement advice to the TRC in 12 cases around licence allocation or value for money considerations, out of which the adviser’s input contributed to outcomes in at least five cases. For example, during the review of one employment region, the probity adviser provided input on the allocation of business share in that region, influencing the market share providers were awarded. In another region the probity adviser’s input contributed towards the TRC’s decision to rank a respondent first.¹⁶⁰

3.94 Ashurst provided a probity sign-off letter on 27 February 2022. While this followed the completion of the TRC deliberations for Enhanced Services on 23 February 2022, the assurances provided by the letter did not extend to a review of the written Delegate’s report, which was still under development on 27 February 2022. This was at odds with the arrangements set out in the probity plan, which stated that the external probity adviser would ‘review and provide comments on the final Evaluation Report, prior to its submission to the Delegate’. Departmental records examined by the ANAO indicate that Ashurst was not provided with an evaluation report for review or comment.¹⁶¹

3.95 The Delegate was first provided with the report on 2 March 2022, which was signed off on 7 March 2022 (the advice to the Delegate is discussed from paragraph 4.37).

4.1 Achieving value for money is the core rule of the Commonwealth Procurement Rules (CPRs) and is critical in ensuring that public resources are used in the most efficient, effective, ethical and economical manner. In assessing value for money, officials must consider a range of financial and non-financial factors.¹⁶² Value for money in this context involves the achievement of the government’s policy objectives for the New Employment Services Model (NESM), including among other things, the ‘intensive and tailored level of servicing desired for more disadvantaged job seekers’.

4.2 The ANAO examined the extent to which these assessments informed the recommendations to, and decisions taken by the Delegate for the Enhanced Services component of the NESM procurement, which was used to establish the Workforce Australia Services panel.

Were recommended tenderers those assessed as providing the best value for money?

4.3 A panel arrangement is a way to procure goods or services regularly acquired by entities. In a panel arrangement, suppliers have been appointed to supply goods or services for a set period of time under agreed terms and conditions.¹⁶³ Panels are established following a competitive value for money assessment process and involve a contractual relationship for a set period of time.

4.4 Department of Finance (Finance) guidance indicates that value for money is to be assessed at two stages for panel procurements. Firstly, when selecting suppliers as part of the initial establishment of the panel arrangement; and secondly, each time an individual panel supplier is engaged from the panel.¹⁶⁴ Each purchase from a panel represents a separate procurement process. When accessing a panel, entities must demonstrate that value for money has been achieved for each engagement.

4.5 In addition to considering the results of the evaluation process, Stage 4 of the NESM procurement was to involve an ‘overall value for money assessment to inform recommendations for national panel membership and … the allocation and issuance of licences within employment region sub-panels.’

Tender Review Committee

4.6 The Tender Review Committee’s (TRC’s) role and membership was set out in the NESM Purchasing Plan.¹⁶⁵ The TRC was responsible for considering ‘the assessment results, Business Allocation recommendations, financial viability assessments, and RFFR [Right Fit for Risk] assessment outcomes to determine the final recommendations to the Delegate.’ The TRC was to ‘conduct an overall value for money assessment’ during its deliberations, which were held from 14 to 23 February 2022. The purchasing plan outlined that this assessment may include, but was not limited to:

• the assessment of responses against the areas of capability categories and Selection Criteria;
• considering the diversity, scope, and viability of the Provider market to encourage innovation in the delivery of the services;
• breadth of coverage to maximise access for individuals, employers and stakeholders;
• consideration of the effectiveness of the proposed Indigenous Participation Plan to strengthen Indigenous economic participation in the supply chains of contractors to government;
• consideration of the benefit of the Respondent’s offer to the Australian economy;
• consideration of the risk to the Australian Government, including – but not limited to – the national business share of each provider¹⁶⁶;
• a desire to obtain the highest possible quality in service delivery.

4.7 No further details as to how the TRC was to make its assessment were included in any of the NESM-specific policy and guidance documents outlined in Table 2.4.

Business allocation recommendations

4.8 The TRC was primarily guided by proposal assessment results and the orders of merit, largely accepting the recommendations from the business allocation (BA) teams.¹⁶⁷ Consistent with the framework:

• all 104 respondents recommended for panel membership were agreed by the TRC. While this was in respect to 1095 proposals from those respondents, the TRC selected a further eight proposals¹⁶⁸, taking the total to 1103 that were recommended to the Delegate for panel membership; and
• of the 45 respondents (with 189 associated proposals) recommended for licences, the TRC agreed to 43 of those and selected two others. Of the 189 proposals recommended by the BA teams:
  • the TRC agreed to 168, disagreed to 21, and selected 13 others, taking the total to 181 proposals recommended to the Delegate for licences.

4.9 Out of the 1176 conformant proposals reviewed by the TRC¹⁶⁹:

• no changes were made to the licence and business share allocation for 1076;
• the business share was increased for 51, resulting in 13 additional licences being awarded; and
• the business share was decreased for 49 proposals.

Appointment of providers to national and regional panels

4.10 With the exception of the Queensland team, the BA teams relied on criterion one assessment results to identify the respondents eligible for appointment to the panel (see paragraphs 3.58 and 3.64).¹⁷⁰ The TRC accepted the BA team recommendations for panel appointments, and was primarily concerned with the allocation of licences and market share during its deliberations, and it was in this context that the value for money factors were subsequently discussed.

Discrepancies between PaLMS records and recommendations to the Delegate

4.11 The Queensland BA team recommended that nine proposals (from seven providers across two employment regions) not be appointed to the panel. While each had scores of more than 40 per cent against criterion one, the BA team noted concerns involving inadequate scores achieved across the other criteria and high financial viability and cyber security risks. These outcomes were consistently recorded in both the Procurement and Licence Management System (PaLMS) and the BA reports.

4.12 Following consideration of the BA recommendations in respect to these proposals, the following decisions were made by the TRC and accurately recorded in PaLMS:

• TRC agreed that two proposals from one provider in the same region should not be appointed to the panel¹⁷¹; and
• did not agree in respect to the other seven proposals — from six providers in one region and one provider in another region — recommending them for panel membership.

4.13 The ANAO’s analysis of the proposals recommended to the Delegate identified some discrepancies. Specifically, in the advice to the Delegate, all nine proposals with criterion one scores of more than 40 per cent were recommended for appointment to the panel and approved by the Delegate. There was no rationale documented as to why these changes were made in the period between the TRC’s deliberations and the report to the Delegate.

Panel appointment

4.14 Consistent with internal guidance and as a result of achieving scores of more than 40 per cent against criterion one alone, a total of 104 (92 per cent) of the 113 Enhanced Services providers assessed as conformant were recommended for appointment to the national panel. Noting that the department’s threshold score for panel appointment was 40 per cent against criterion one, the ANAO’s analysis was that 16 (15 per cent) of those 104 providers recommended for the panel had received scores of between 40 and 50 per cent.¹⁷²

4.15 Those providers had submitted 1105 (94 per cent) of the 1176 conformant proposals and were recommended for appointment to one or more of the regional sub-panels. Of the 1105 proposals, 181 (16 per cent) were recommended for licences, and 924 (84 per cent) did not receive licences but were recommended for appointment to the regional sub-panels. Of those 924 proposals recommended for a sub-panel, 376 (41 per cent) of these from 88 providers had scored less than 50 per cent against one or more criteria. Table 4.1 provides an overview of the spread of scores across the 924 proposals without licences.

Table 4.1: Analysis of scores for unlicensed proposals recommended for the panel

Source: ANAO analysis of proposal scores.

4.16 A ‘Panel Maintenance Guide’ to inform the use of the national panel and regional sub-panels, including for gap-filling requirements, was developed between late July 2022 and early February 2023. The final guide was endorsed by the Deputy Secretary on 6 February 2023.¹⁷³ The panel guide outlined that panel members (with and without licences) would be issued with ‘declaration forms’ on an annual basis to confirm their willingness to maintain their capacity to provide services in the relevant employment regions and comply with IT security requirements.

4.17 Following notification of the procurement outcomes in March 2022, five providers declined the department’s offer (see paragraph 4.43 and Appendix 7). This left a total of 893 proposals remaining across the sub-panels, of which 362 proposals (41 per cent) from 84 providers had scored less than 50 per cent against at least one criterion and remain available for selection for gap-filling requirements in any of the 51 employment regions in accordance with the department’s panel maintenance guide.

4.18 According to the panel maintenance guide, the first step during a panel refresh or gap filling process involves the department approaching the providers on the relevant regional sub-panels. If ‘no suitable organisation exists’, any of these 84 providers across the sub-panels can be approached through a limited tender process and potentially awarded a licence in any employment region.¹⁷⁴ Figure 4.1 provides an overview of the distribution of proposal scores by TRC recommendation.

Figure 4.1: Distribution of criterion 1 assessment ratings by TRC recommendation

Note a: 54 proposals recommended for the panel had scores of below 50 per cent against criterion one. Out of these, one had been recommended for a licence.

Note b: 10 proposals receiving a score of more than 40 per cent in criterion one (from one provider) were not recommended for the panel due failing the RFFR assessment due to its storage of data offshore.

Note c: Six proposals receiving criterion one scores of more 60 per cent (assessed as ‘very suitable’) against criterion one (from one provider) were not recommended because the provider was the subject of an ongoing compliance investigation for alleged non-compliant outcome payment and wage subsidy practices.

Source: ANAO analysis of departmental records.

Tender Review Committee deliberations

4.19 The ANAO reviewed the majority of the meeting recordings of TRC deliberations to identify how value for money was considered by the TRC. In addition to the assessment scores and preliminary recommendations from the BA process, the TRC discussed the following aspects, first in the context of each individual employment region, and then nationally: the quality of proposals¹⁷⁵; increasing job seeker/participant choice; the necessary amount of specialist licences; and ensuring sufficient provider coverage in and across regions. Viability considerations were limited to discussions on geographical coverage (expanded upon from paragraph 4.28) and did not consider the KPMG findings on provider viability (outlined from paragraph 2.26). The number of licences recommended by the TRC exceeded KPMG’s recommendations in 92 per cent of regions. Depending on employment market conditions, this heightens risks around provider viability and continuity of services in those regions.

Quality of proposals and service offering

4.20 Key factors for the TRC in determining the quality of the proposals and services offered were the evaluation scores and the features of proposals, such as unique service approaches or a comparatively lower caseload ratio.¹⁷⁶

4.21 The procurement was designed such that the quality of proposals would be reflected in overall assessment scores. These were heavily relied upon through the business allocation (BA) principles discussed at paragraph 3.59, which involved allocating the most market share to the highest ranked providers. The TRC queried the robustness and reliability of the assessment scores in several regions. For example, it was noted that high-scoring providers in one employment region had not provided sufficient detail on their service model or plans to provide accessible services where there was a high level of socioeconomic disadvantage and low internet and transportation connectivity. Following initial committee proceedings for that region on 16 February 2022, the Senior Procurement Officer requested re-adjudication and subsequent re-assessment for that region.¹⁷⁷

Increasing participant choices

4.22 Reflective of the TRC’s considerations when assessing value for money, scores did not always prevail when balanced against other factors, such as ensuring adequate coverage of services and job seeker choice. For example, in one region in Western Australia, the BA team recommended licences for two proposals, each with 50 per cent market share, noting that no providers had adequately addressed providing services to the CALD community in that region. The TRC reallocated the market share between three providers, including the third ranked provider, to increase job seeker choice.¹⁷⁸ Increasing choice was for a key consideration for the TRC in allocating licences in at least 14 other regions, resulting in some business share allocated to lower ranked providers.

Caseload ratio

4.23 In the absence of a requirement in the RFP to provide caseload ratio details in proposals, not all providers included this information. Where ratios were included, the TRC took this into account when making its decisions.¹⁷⁹ Where a respondent did not identify the ratio but referred to a ‘low’ consultant to participant ratio, this was also noted during the TRC proceedings.¹⁸⁰

Allocation of specialist licences

4.24 The department’s methodology for the TRC’s allocation of specialist licences had not been developed in advance of deliberations and the TRC did not refer to the relevant KPMG report on specialist provider viability (see paragraphs 2.34 to 2.37). Rather, the TRC discussed the demographics of each employment region and agreed the specialist licences to be awarded on a region-by-region basis. Compared with KPMG’s analysis, there was no identifiable correlation between the size of the specialist cohort caseloads and the amount of business share awarded to a specialist provider. However, the TRC’s recommendations resulted in the awarding of fewer specialist licences compared with the overall KPMG analysis (34 per cent less). Instead of a set methodology, the following decisions were observed.

• Where the specialist cohort was considered large enough, market share was allocated to generalist providers in at least eight regions.¹⁸¹ This was on the basis that generalist providers were expected to service the region’s specialist cohort, or that the proposals from those providers reflected a strong understanding of the specialist cohorts in the region. For example, no CALD licences were allocated in the region with the highest CALD proportion nationally, and a high refugee caseload of 20 per cent (see Appendix 5).¹⁸²
  • In the same region where the ex-offender caseload was nine per cent of the total, an ex-offender licence (with five per cent of the market share) was allocated to the highest ranked tenderer.¹⁸³
• In regions where the specialist job seeker cohort was comparatively small, the TRC agreed that specialist providers would be unviable, choosing to instead allocate a generalist licence.¹⁸⁴
• In five regions where the department had identified in the RFP that specialist licences would be allocated, the TRC did not allocate any specialist licences. These are outlined at Appendix 5.

Was provider coverage for employment regions appropriately addressed in advice for the delegate?

4.25 The TRC had responsibility for ensuring appropriate provider coverage across regions was addressed in the advice for the Delegate. The relevant state and territory BA teams with local area knowledge were to provide key input into those recommendations and assist the TRC by attending and presenting at committee proceedings.

4.26 The BA teams provided advice on the coverage requirements of the region, including relevant caseload statistics; appropriate service models; proposed site numbers and types, locations; and accessibility of services. In forming this advice, the BA teams considered the combined geographic coverage of the region offered through different combinations of the competing proposals, rather than each proposal on its individual merits. This was to achieve optimal coverage for the region.

4.27 As outlined at paragraph 3.31, issues emerged with the PaLMS system when importing site location details. As a result of this, and inaccuracies in scribe notations, a manual process was required to confirm site locations.

Provider coverage

4.28 The TRC’s approach to provider coverage was largely influenced by two competing objectives: achieving adequate coverage across sites in each region; and ensuring that providers were able to viably operate within the employment region. While deliberations aimed to strike a balance, provider coverage was at times impacted by TRC decisions to remove or consolidate sites to ensure viability. In doing so, the TRC considered factors such as the number of providers already operating in the specific location, the types of sites proposed by providers (full-time, part-time and outreach), and whether the providers operating were generalist or specialist. Examples of these decisions are outlined below.

Region A and B

4.29 The RFP outlined that, for these two regions, the indicative number of licences to be awarded ranged between two and five. The BA team recommended four providers, with up to four sites in the main population centres and between one to three providers servicing remote locations from part-time or outreach sites. Three of those providers were ranked first, second and fourth, and proposed to service the entire region through several fulltime sites and additionally provide coverage of remote areas. The third ranked provider offered multiple sites around one city centre, but not for the remote areas. The TRC recommended that the four city centre sites (two full-time and two outreach sites) proposed by one provider be reduced to two sites (one outreach and one full-time), due to viability concerns with having four providers in the region. While four providers had operated in this region under jobactive, this was previously a larger caseload.

Regions B, C and D

4.30 A number of proposed sites were modified to prevent coverage overlaps and mitigate risks to viability in three regions. In Region B, where a provider had a location-specific proposal, the TRC agreed that this site would not contribute additional value to the combination offered by other providers or improve service access in that region. Further, the TRC agreed it would introduce viability concerns and it was therefore not recommended for a licence.

Note a: The two top-ranked proposals were a specialist (Indigenous) and a generalist proposal, both from the same provider. The BA team’s rationale was that there was an insufficient Indigenous caseload in the region to meet the provider’s minimum market share requirement of 16 per cent.

Note b: The business allocation team noted that the relevant LGAs were ‘the most difficult locations to service in terms of disadvantage and probability of achieving employment outcomes’.

Full-time, part-time and outreach sites

4.31 In their submissions, providers proposed site locations (either proposed or pre-existing), which included full-time, part-time, and outreach sites.¹⁸⁵ During deliberations, the TRC considered the type of site and coverage achieved, mainly in relation to preventing over-saturation of services within a specific location. The proportion of contracted full-time, part-time and outreach sites based on departmental data from 2 June 2022 indicated that most sites were full-time.¹⁸⁶ Out of 350 outreach sites contracted by the department, 270 (77 per cent) had no detail on what frequency or manner of outreach services would be provided, describing the provision of services on an ‘as needs’ or ‘as required’ basis. This was not queried by the TRC.

4.32 As illustrated by Figure 4.2, New South Wales (NSW) and Tasmania have comparatively higher numbers of outreach sites, with the number of outreach sites in NSW being more than a quarter of the total sites. Within NSW, the highest numbers of outreach sites were:

• 28 sites in New England and North West;
• 22 sites in Central West;
• 12 sites in Sydney Greater West; and
• 17 sites each in Murray Riverina and Sydney East Metro.

4.33 Out of the ten employment regions with the highest number of outreach sites, five were located in NSW, two in Victoria and one each in Queensland, South Australia and Tasmania.

Figure 4.2: Proportion of full-time, part-time and outreach sites

Source: ANAO analysis of departmental records.

4.34 The TRC’s consideration of site types and coverage, were not informed by an established framework or methodology. However, a common consideration was provider viability within the region. For example, the TRC:

• Did not accept the BA team’s recommendation in Region A to allocate a full-time site in a difficult-to-service locality.
• Allocated business share to a provider proposing to service 14 outreach sites (in a region where 48 per cent of sites were outreach, 41 per cent were full-time and 11 per cent were part-time).
  • In response to the Delegate’s query on the anticipated frequency of service for the outreach services, the chair of the TRC advised that the ‘respondent indicates it will provide outreach at the identified locations on an ‘as required’ basis so the TRC considers the Respondent has assessed that demand for outreach services may be intermittent/low and made its offer accordingly’;
• Allocated business share and licences in a manner which left one town centre (located 18 km from another town with a full-time site) without any dedicated site locations apart from outreach services.
  • When this recommendation was queried by the Delegate, the chair advised that all recommended providers had full-time sites in the town centre 18 kms away, and all respondents had alternative servicing strategies including digital platforms and mobile services, such as a mobile bus.¹⁸⁷
• Skipped the fifth-ranked provider in favour of another provider as it did not propose a site in a location noted as a future hotspot for employment service needs.
• Reduced one recommended specialist provider’s proposed 24 sites to 18 sites on the basis that there was not a substantial specialist cohort caseload in those regions to justify the provider’s resources being established there and due to potential overlaps with other service providers.

National Business Review conducted by TRC

4.35 The TRC conducted a national business review following seven days of deliberations for all 51 regions. The purpose of this review was for the TRC to revisit their interim recommendations on a case-by-case basis made in the days prior and was not informed by any established framework or methodology.¹⁸⁸ Largely, the TRC did not revise its recommendations, and only adjusted their recommendation for five employment regions.

Were final procurement decisions appropriately informed by departmental advice?

4.36 The NESM purchasing plan set out that the Delegate report was to contain: the recommendations; a summary of responses assessed, including details of TRC deliberations, scores and rankings; any respondents who have been excluded and why; and details of any applicable dissenting positions. The Delegate could seek further information from TRC members prior to deciding whether to approve the recommendation.¹⁸⁹

Delegate’s approval of recommendations

4.37 After the TRC proceedings concluded on 23 February 2022, the Senior Procurement Officer prepared the Delegate recommendation report based on the information recorded in PaLMS by the scribes and input from the TRC chair and members. The Delegate was provided with draft recommendations for four states on 1 March 2022 (NSW/Australian Capital Territory, Northern Territory, Tasmania and Victoria) and a complete report comprising recommendations for all states on 2 March 2022.

4.38 The Delegate made several inquiries about the underlying value for money factors and associated risks for some of the TRC’s licence recommendations. This included analysis undertaken for:

• risks presented through providers’ financial statements, subcontracting arrangements and delivery model; and
• the financial viability of providers in the context of the size of the caseload and site coverage considerations for the applicable region.¹⁹⁰

4.39 Following receipt of further information from the TRC chair and the Senior Procurement Officer, the Delegate approved the TRC’s recommendations on 7 March 2022. The Minister was provided with a ‘noting brief’ on the same day, outlining the decisions taken by the Delegate and the procurement outcomes. The Minister signed the brief on 14 March 2022 without further annotation.

Final procurement outcomes

4.40 Following the Delegate’s approval, the department notified 104 organisations on 11 March 2022 that they had been successful for appointment to the national panel.¹⁹¹ Out of these, 45 organisations were offered 181 licences to deliver Enhanced Services under the NESM.

4.41 Following initial notifications in March 2022, 100 providers accepted offers to be appointed to the national panel and 44 accepted 179 licences. Following the termination of one provider’s Deed of Standing Offer (DoSO) in late May 2022 (outlined in Table 4.2), the department issued work orders to 43 providers to deliver Enhanced Services with 178 licences.¹⁹²

Table 4.2: Notification and acceptance of panel membership and licence offers

Note a: Of the 104 organisations offered panel membership, two proposals were from the same provider, one of which was as part of a proposed joint venture, to be formed if awarded business share. Both proposals were appointed to the panel upon the commencement of Workforce Australia, with one of those issued a work order to deliver services.

Note b: One provider appointed to the panel and awarded two licences (Generalist and Specialist) rejected the offer upon receipt, with three other providers who had been appointed to the panel but not awarded a licence rejecting the offer at this stage.

Note c: One ex-offender specialist provider with one ex-offender licence had acknowledged the deed at initial notification in March 2022, but then requested termination in May 2022, which was actioned by the department in June 2022. This is reflected in the final work orders and licence numbers issued in July 2022.

Source: ANAO analysis of departmental records.

4.42 Nine out of 113 providers (submitting 71 proposals) were not appointed to any one of the regional sub-panels as:

• one provider (submitting six proposals), rated very suitable against criterion one, was undergoing a compliance investigation at the time of the procurement;
• one provider (submitting 10 proposals), rated suitable against criterion one, did not fulfill the department’s cybersecurity requirements in relation to offshore data storage; and
• seven providers (submitting 55 proposals) were rated below suitable for criterion one.

Gap filling exercise following declined offers

4.43 In total, five providers declined the offer from the department (see Appendix 7). In May 2022 the department undertook gap-filling exercises after two providers rejected their licences in two regions. The department’s gap-filling exercise also aimed to address any gaps left from the exit of three providers appointed to regional sub-panels but not in receipt of a licence.

4.44 For the gap left in the first region due to the exit of one provider, the business share was offered to other successful respondents in the employment region. In addition to holding a generalist licence, the exiting provider was one of two Indigenous specialist providers on the panel, with a total of 29 per cent market share between the two licences. There was one other Indigenous (specialist) provider on the panel with a licence, with a market share of six per cent. The business share was redistributed by reallocating the available business share from the first ranked generalist respondent down to the last ranked respondents already with a licence in the region.

4.45 For the provider gap in the second region, the department reallocated that provider’s full business share (five per cent) to the first ranked generalist licence holder, on the basis that other panel members had higher minimum business share requirements and there was no other ex-offender panel member available.¹⁹³

Department’s approach to monitoring key service offers (‘special conditions’)

4.46 A procurement process does not achieve value for money if a contract is not negotiated or managed with a view to achieving performance.¹⁹⁴ Effective management is supported by an appropriate contractual framework against which supplier performance can be monitored.

4.47 The department’s approach to capturing the ‘key service offers’ proposed by respondents for Enhanced Services differed from the approach under jobactive, where providers had service delivery plans (SDPs) and service offer commitments (SOCs).¹⁹⁵ Discussions at the August 2021 Employment Steering Committee (ESC) meeting noted that SDPs were ‘not useful, inefficient, and [did] not achieve the intended purpose of holding providers to account for the commitments made in their tender response.’ The ESC agreed to shift the approach to ‘special conditions’, noting that:

For the NESM, if the intention is to ensure providers follow through with their tender commitments on the basis of which business is being awarded to them, a more efficient and contractually stronger approach would be to capture a provider’s key tender commitments as Special Conditions attached to each licence. This approach would formalise the current SOC activity, noting that responsibility for monitoring compliance with these Special Conditions is yet to be determined and may form part of licence review.

4.48 While special conditions were identified for 19 proposals across eight regions during the business allocation process in late January and early February 2022, these were not discussed during TRC deliberations and did not appear in the Delegate recommendations.¹⁹⁶ Consequently, no special conditions were captured in the July 2022 work orders.

4.49 In late August 2022, the department sought internal legal advice on reflecting tender commitments under the ‘special conditions’ clause in the DoSO. This was to provide contract managers with visibility of providers’ commitments for compliance purposes. As providers had already accepted the work orders by that stage, the department noted risks around providers being potentially unwilling to accept changes, particularly if they had amended their service offerings since the conclusion of the procurement:

As this more narrow approach was then used in how the work orders were prepared and issued to the 43 licence holders it will be too late to update these (providers will ask why the change and may not accept these changes)…

… licence holders can still be held to account to their commitment under the Deed and successful tender bids are available for state office staff to access and identify all relevant commitments.

4.50 In January 2023, the department decided that all key service offer commitments from providers’ responses would be identified and formalised retrospectively via letters to providers. The basis for enforcing the key service offer commitments were provisions contained in the DoSO around service guarantees and contractual commitments from providers to carry out their services in accordance with representations made through the RFP process. This approach meant that it was not necessary to use the ‘special conditions’ annexure, where special conditions were to be appended to individual deeds, if relevant.

4.51 Over March and April 2023, the department shifted its approach to instead identifying and capturing ‘the reasons why the provider was awarded one or more licences’, rather than identifying all tender submission commitments. The rationale for this shift was to reduce the workload for state office staff, and to provide ‘more flexibility for providers to deliver individualised services’ while remaining responsive to policy or market changes.¹⁹⁷ The department identified a number of risks with this approach, including that there was a risk inherent in not managing all tender commitments in a systematic way. Additionally, this approach meant that the original tender submissions would still need to be referred to by contract managers to ensure that providers were delivering on what they committed to in their proposals.

4.52 The department advised the ANAO that it had undertaken consultation with 42 providers to confirm the contents of the key tender representations from late June to mid-August 2023, with the letters containing the key service offer commitments for those providers sent out over that same time period.

Provider exit in the Broome employment region

4.55 In December 2022, the sole provider in the Broome region requested to terminate their Workforce Australia deed, stating lack of financial viability.¹⁹⁸ The department explored potential options in February and March 2023 to address the potential service gap for 743 affected job seekers, and a future approach to providing employment services in the region. The department noted a ‘timing imperative’ and a ‘real risk of an imminent servicing gap’ since the provider could not continue until 31 October 2023 (given its financial position). Initial options considered by the department for future service delivery in the region included:

• issuing a new Workforce Australia licence through the existing regional sub-panel and a limited tender process;
• implementing a ‘modified’ (or ‘hybrid’) services model involving a potential new provider supported by APS staff^199; or
• removing Broome from Workforce Australia and redesignating as a Community Development Program (CDP) region.

4.56 A new Workforce Australia licence through the regional sub-panel was not considered suitable for addressing the local or cultural issues associated with viability.²⁰⁰

4.57 In light of the provider viability issues, the department explored changes to the existing provider payment model. Of particular note in this respect, was that the Broome region was consistently found not viable throughout the KPMG analyses in 2020 and 2021.²⁰¹ By 2023, the department had noted that specific characteristics of the Broome caseload reduced opportunities for providers to earn both the initial engagement and larger outcome payments currently in place in Workforce Australia.²⁰²

4.58 In late March 2023, in line with departmental recommendations, the Australian Government agreed $5.6 million for the department to:

• support the exiting provider to continue services until 31 October 2023; and
• conduct a limited tender process using the Mandatory Set Aside arrangements of the Indigenous Procurement Policy to select a new First Nations provider to deliver new servicing arrangements from 1 November 2023.²⁰³

4.59 In April 2023 the department approved the termination of the exiting provider’s existing DoSO from 31 October 2023 and the provision of in-kind departmental support until then. On 9 May 2023, the Australian Government announced new arrangements to ‘better target the department’s activities and programs, including ensuring the continuity of employment services in the Broome employment region and transitioning to a culturally appropriate, place-based model.’²⁰⁴

4.60 The department developed a new Purchasing Plan and Limited Request for Tender (RFT) for the ‘new Broome Employment Services’, which opened for submissions on 6 July 2023 and closed on 3 August 2023. The RFT documentation contained the same selection criteria as the NESM RFP, but differed by including specific requirements for the successful provider in Broome and a new payment model.²⁰⁵

Analysis of final procurement outcomes

4.61 The procurement process resulted in 12 employment regions (24 per cent) experiencing a total turnover of providers, with no previous jobactive providers being awarded licences.²⁰⁶ Eight regions had no new providers awarded licences.²⁰⁷ Figure 4.3 and Figure 4.4 illustrate the distribution of licences by employment region.

Figure 4.3: ANAO analysis of licence allocation

Source: ANAO analysis of departmental records.

Figure 4.4: Provider turnover across the 51 employment regions

Note: This data comprises all licences that were approved by the Delegate and offered to providers, including those that declined the department’s offer. For the NEST regions (Adelaide South, SA and Mid North Coast, NSW), NEST providers were included in the jobactive provider numbers. This data excludes instances where providers were offered multiple licences within one employment region. Therefore, the number of licences in this figure do not always align with the actual number of licences offered.

Source: ANAO analysis of departmental data and recommended licences.

4.62 One of the policy objectives for the NESM was to increase the number of smaller providers in the market. The percentage of small providers on the national panel has notably increased from jobactive. However, this increase is smaller when comparing against those that are NESM licence holders.²⁰⁸ Figure 4.5 provides a comparison by organisation size between jobactive and NESM providers.

Figure 4.5: Providers in NESM and jobactive based on organisation size

Note: NESM provider numbers exclude those that declined the department’s offer.

Source: ANAO analysis.

4.63 As shown by Figure 4.6, the proportion of incumbent providers was lower on the panel and higher within the population of licensed providers.

Figure 4.6: Proportion of new and incumbent providers in NESM

Source: ANAO analysis of departmental records.

Was timely and appropriate feedback made available to respondents following the procurement process?

4.64 The CPRs outline that debriefings must be made available, on request, to unsuccessful tenderers outlining the reasons their submissions were unsuccessful.²⁰⁹ Effective feedback provides transparency and accountability and assists tenderers to improve their submissions in future processes.²¹⁰

4.65 Departmental officials involved in debriefing sessions were instructed to adhere to a debriefing script when providing feedback to respondents. A number of these officials were not involved in the procurement process, and in accordance with the script, did not always provide respondents with feedback that was sufficiently tailored or consistent with the assessment process undertaken. For example, the debriefings did not address:

• how some incumbent providers had been assessed for demonstrated performance by averaging their star ratings across their sites nationally²¹¹;
• where providers were unsuccessful due to requirements relating to cybersecurity, (for example, the storage of their data outside Australia) they were not informed that this was the reason for being unsuccessful; and
• that assessments against criteria were being undertaken largely by contracted assessors in state and territory hubs.

4.66 Debriefing officers were also instructed to not disclose to respondents their scores against the selection criteria and to explain that while proposals may have been assessed as ‘very good’, the procurement ‘was a competitive environment and the contract was awarded to a higher ranked Respondent or a lower ranked Respondent that offered a better value for money proposition.’ This approach provided limited information to providers on the comparative quality of their proposals which could be used by providers to improve their proposals for future procurements.

Debriefings provided by the department following NESM procurement

4.67 Following the initial notification of procurement outcomes to respondents, 52 Enhanced Services respondents sought debriefings. Out of those requesting debriefings:

• 43 respondents had been appointed to the panel, out of which:
  • 20 had been awarded at least one licence and were on the panel; and
  • 23 had been appointed to the panel but had not been awarded a licence; and
• nine respondents had not been appointed to the panel and had not received a licence.

4.68 Respondents were notified of the procurement outcomes on 11 March 2022 and debrief requests were to be submitted prior to 11 April 2022. To meet the two-month timeframes the department had allocated for the completion of debriefings, the requested debriefings were to occur on or before 11 June 2022. The department provided debriefs within 52 days to 135 days of requests being made, with an average response time of 84 days.

4.69 Respondents seeking a debriefing completed a ‘debriefing form’ where most respondents (50 out of 52) included specific areas of query. These queries largely reflected aspects of the evaluation process that were not explained in detail in the RFP documentation, such as:

• how existing performance data for the ‘demonstrated performance’ sub-criterion had resulted in the loss of business in regions where providers had high star ratings (see paragraph 3.40); and
• how the weightings allocated to aspects of the evaluation criteria had impacted assessments (see paragraph 2.47).

4.70 Where debriefing officers could not answer questions during the sessions, these were taken on notice, and in 23 cases, the department addressed these after the debriefing sessions.

Appendix 1 Entity response

ANAO comments on the Department of Employment and Workplace Relations’ response

1. Department of Finance guidance on panel procurements highlights that for a supplier to be appointed to a panel, value for money should have been demonstrated through the evaluation process used to the establish the panel.²¹² As outlined at paragraphs 6, 9, 26 and 27 of this audit report, the ANAO:
   • concluded that the design of the procurement process was largely consistent with the Australian Government’s policy objectives. The evidence relied upon by the ANAO is outlined in paragraphs 2.16 to 2.19; and
   • found that while licences were awarded to the providers assessed as representing the most overall value for money, the establishment of the Workforce Australia Services Panel was not informed by an appropriate value for money assessment. The evidence relied upon by the ANAO is detailed in paragraphs 3.58, 3.64 to 3.65, 4.3 to 4.5, and 4.10 to 4.16.

Appendix 2 Improvements observed by the ANAO

1. The existence of independent external audit, and the accompanying potential for scrutiny improves performance. Improvements in administrative and management practices usually occur: in anticipation of ANAO audit activity; during an audit engagement; as interim findings are made; and/or after the audit has been completed and formal findings are communicated.

2. The Joint Committee of Public Accounts and Audit (JCPAA) has encouraged the ANAO to consider ways in which the ANAO could capture and describe some of these impacts. The ANAO’s corporate plan states that the ANAO’s annual performance statements will provide a narrative that will consider, amongst other matters, analysis of key improvements made by entities during a performance audit process based on information included in tabled performance audit reports.

3. Performance audits involve close engagement between the ANAO and the audited entity as well as other stakeholders involved in the program or activity being audited. Throughout the audit engagement, the ANAO outlines to the entity the preliminary audit findings, conclusions and potential audit recommendations. This ensures that final recommendations are appropriately targeted and encourages entities to take early remedial action on any identified matters during the course of an audit. Remedial actions entities may take during the audit include:

• strengthening governance arrangements;
• introducing or revising policies, strategies, guidelines or administrative processes; and
• initiating reviews or investigations.

4. In this context, the below actions were observed by the ANAO during the course of the audit. It is not clear whether these actions and/or the timing of these actions were planned in response to proposed or actual audit activity. The ANAO has not sought to obtain assurance over the source of these actions or whether they have been appropriately implemented.

• Endorsement and approval by the Deputy Secretary of a Panel Maintenance Guide, endorsed in February 2023, which sets out the department’s approach to managing panel membership and approaching the panel when needed.
• Issuing ‘key tender representation’ letters to providers in June 2023. The department intended to ‘facilitate the monitoring’ of provider’s compliance with the deed through listing and providing to providers the list of tender representations made during the tender response process.

Appendix 3 Assessment evaluation descriptors

Table A.1: Criterion 1 — Distribution of selection criteria weightings by evaluation component descriptor element

Source: ANAO analysis.

Table A.2: Criterion 2 — Distribution of selection criteria weightings by evaluation component descriptor element

Source: ANAO analysis.

Table A.3: Criterion 3 — Distribution of selection criteria weightings by evaluation component descriptor element

Source: ANAO analysis.

Figure A.1: Evaluation descriptor rating scale included in the NESM Assessment Guide

Source: Extract from departmental records.

Appendix 4 Key guidance documents for the Business Allocation process

Table A.4: Key guidance documents for the Business Allocation process

Note a: Guidance documents were placed on a shared network drive accessible to NESM assessment staff.

Note b: The BAIT underwent at least three revisions over this period.

Source: ANAO analysis of departmental records.

Appendix 5 Regions with no specialist licence where specialist was indicated as possible in the RFP

Table A.5: Licence allocation for regions with CALD cohort above 40 per cent

Note a: Indicative Enhanced Services caseloads published in the NESM RFP.

Note b: Specialist cohorts could overlap with each other, therefore did not necessarily add up to the total indicative caseload figures.

Note c: All of these regions had been published in the RFP as having potential for a CALD specialist licence.

Source: ANAO analysis of departmental records.

Table A.6: Regions with no specialist licence where specialist was viable, published in the RFP

Note a: Specialist cohorts could overlap with each other, therefore did not necessarily add up to the total indicative caseload figures.

Note b: In the case of one of these regions, the business allocation team noted that the CALD provider’s claims did not propose a strong service offering for the CALD cohort.

Note c: For Darling Downs and Townsville, the numbers include one respondent which was recommended and offered licences, but declined the offer.

Source: ANAO analysis of departmental records.

Appendix 6 Staffing for the NESM evaluation process

Table A.7: Split between APS and Non-APS staff for a range of key roles for the NESM evaluation process

Note a: Staff were able to be both adjudicators and moderators in the same process. They were not, however, able to be involved in the assessment of the responses being moderated or adjudicated.

Note b: The business allocation process was undertaken in state/territory departmental offices after all relevant proposals had been assessed against the selection criteria. Proposals were reviewed at the employment region level at this step.

Note c: Includes largely contracted staff and a small number of staff identified as ‘other’.

Note d: Totals in this row do not equate to the 277 staff due to overlap in roles. On average, individuals performed two roles each.

Source: ANAO analysis of departmental records.

Appendix 7 Respondents that declined the department’s offer

Table 4.8: Rationale provided by respondents that declined the department’s offer

Note a: This respondent had been appointed to 12 regional sub-panels, out of which they were offered two licences in one region. The outcome of this procurement process meant that this respondent would have to exit from three of these regions where they had been providing services under jobactive.

Note b: This respondent declined the offer but requested a meeting to discuss alternative options, including a right in the DoSO to assess the terms of a work order. This request was not granted by the department. The respondent also noted that the documentation under the deed, including chapter 1 and chapter 5, was not included.

Note c: This respondent had initially accepted the department’s offer in March 2022.

Source: ANAO analysis of departmental records.