Overview of the JobKeeper Payment and scheme

1.1 The JobKeeper Payment was announced by the Prime Minister and Treasurer on 30 March 2020 as part of the Australian Government’s response to the coronavirus disease 2019 (COVID-19) pandemic.

1.2 The announcement stated that the JobKeeper Payment was a wage subsidy to businesses that would keep more Australians in jobs through the outbreak. The announcement outlined that:

• the payment would be made to employers for up to six months for each eligible employee who was on the employer’s books on 1 March 2020 and was retained or continued to be engaged by that employer;
• employers would receive a payment of $1500 per fortnight per eligible employee² and every eligible employee must receive at least $1500 per fortnight from this business, before tax;
• eligible employers included businesses structured through companies, partnerships, trusts and sole traders, and not-for-profit entities, including charities; and
• the program would commence on 30 March 2020, with the first payments to be received by eligible businesses in the first week of May as monthly arrears from the Australian Taxation Office (ATO).

1.3 Other countries also introduced wage subsidy or ‘job retention’ schemes in response to the COVID-19 pandemic, including the United Kingdom, New Zealand and Canada.³

Legislative framework

1.4 The legislative framework for the JobKeeper Payment mainly comprises:

• the Coronavirus Economic Response Package (Payments and Benefits) Act 2020 (CERP Act), which received Royal Assent on 9 April 2020; and
• the Coronavirus Economic Response Package (Payments and Benefits) Rules 2020 (‘the Rules’), which is a legislative instrument made under the CERP Act by the Treasurer.⁴

1.5 The Rules were first made by the Treasurer on 9 April 2020 and were updated a number of times during the course of the JobKeeper scheme.

1.6 The Rules establish the ‘JobKeeper scheme’ and set out the eligibility requirements for the JobKeeper Payment, payment arrangements and administration matters. The JobKeeper scheme is administered by the Commissioner of Taxation, who also has the general administration of the CERP Act.

1.7 The JobKeeper scheme was originally legislated to operate from 30 March to 27 September 2020. Following an announcement by the Australian Government on 21 July 2020, the JobKeeper scheme was extended to 28 March 2021. Under the Rules, the Commissioner of Taxation must not make any JobKeeper payments after 31 March 2022, except in specified circumstances.⁵

Rapid implementation

1.8 As with other COVID-19 economic response measures administered by the ATO in 2020, the JobKeeper scheme was characterised by rapid implementation. The first monthly claim period under the scheme opened on 4 May 2020, five weeks after the JobKeeper Payment was announced.

1.9 In his opening statement to the Parliament of Australia’s Senate Select Committee on COVID-19 on 7 May 2020, the Commissioner of Taxation articulated the ATO’s approach to administering the economic response measures, including JobKeeper, as follows:

In response to COVID-19, the ATO has pivoted our focus to ensure the efficient rollout of the five key measures, in particular the JobKeeper payments, early release of superannuation, cash flow boost, increasing the instant asset write-off, and accelerated depreciation. Our priority has been to deliver on the government’s commitment to get millions of Australians access to financial support quickly and as easily as possible during this difficult time.⁶

JobKeeper Payment

1.10 As of 15 August 2021, ATO data indicated that net payments⁷ totalled $88.82 billion ($69.97 billion in the original period and $18.85 billion in the extension period). A total of 1,068,856 entities had applications processed under the scheme.⁸ An average of 3.6 million individuals were estimated to have received payments each month in the original period⁹ and 1.8 million unique individuals received payments in the extension period.¹⁰ A monthly breakdown of net payments, the number of entities receiving payments and the number of individuals covered by payments over the course of the scheme is provided in Figure 1.1.

Figure 1.1: Net payments, entities with processed applications, individuals with processed applications — 30 March 2020 to 28 March 2021

Notes: The first ‘JobKeeper fortnight’ was 30 March to 12 April 2020.

The August and December 2020 net payments are higher due to three JobKeeper fortnights being included in these periods.

Source: ANAO, reproduced from the ATO’s JobKeeper data (as at 15 August 2021).

1.11 Other summary data produced by the ATO on the JobKeeper scheme, including analysis by entity type, market segment and jurisdiction, is provided at Appendix 3.

Administration arrangements for the JobKeeper scheme

1.12 The ATO assisted the Commissioner of Taxation in the day-to-day administration of the JobKeeper scheme, while Treasury has responsibility for JobKeeper policy and evaluation.

1.13 The ATO’s administration arrangements for the JobKeeper scheme changed over time.

• For the first six months, compliance activities were undertaken by four business lines within the ATO’s Client Engagement Group. ATO documentation noted that this reflected the rapid implementation and the ATO’s ability to leverage its existing structures and processes.
• In October 2020, the ATO established the Economic Stimulus Branch within the Client Engagement Group to centralise administration of the JobKeeper scheme (as well as the JobMaker Hiring Credit scheme and the Single Touch Payroll program). The Economic Stimulus Branch comprised three work streams: Engagement and Assurance (initially 556 staff), Advice and Guidance (212), and Program Governance and Management (74, of which 12 worked on JobKeeper and 62 on the Single Touch Payroll program).
• The Economic Stimulus Branch was disbanded in June 2021 and responsibility for JobKeeper functions, including compliance work, was assigned to the Superannuation and Employer Obligations business line.

1.14 The cost of administering the first six months of the JobKeeper scheme was funded from the ATO’s existing budget in 2019–20. The ATO received additional funding of $305.9 million over four years in the 2020–21 Budget to deliver the extension period and the JobMaker Hiring Credit scheme. The bulk of this funding ($256.2 million) was allocated for 2020–21.

1.15 The ATO advised the ANAO that the cost of administering the JobKeeper scheme from March 2020 to 30 June 2021 was $286 million. Figure 1.2 shows the number of ATO staff involved in administering the JobKeeper scheme (as their primary role) from May 2020¹¹ to June 2021.

Figure 1.2: Number of ATO staff administering the JobKeeper scheme, May 2020 to June 2021

Note: Staff worked in Client Engagement Group, Enterprise Solutions and Technology, and Law Design and Practice.

Source: ANAO, based on the ATO’s data.

1.16 Treasury’s responsibilities for JobKeeper policy were managed by the JobKeeper Division, which was later renamed the Labour Market Policy Division, located within Fiscal Group. An average of 22 staff worked in the division between April 2020 and June 2021. Staff also had responsibilities for the JobMaker Hiring Credit scheme and other labour market policy matters. The Tax Analysis Division within Treasury’s Revenue Group also had dealings with the ATO on fiscal and economic updates and reporting.

Rationale for undertaking the audit

1.17 An audit of the JobKeeper scheme is part of phase two of the ANAO’s multi-year strategy that focuses on the effective, efficient, economical and ethical delivery of the Australian Government’s response to the COVID-19 pandemic.¹²

1.18 The JobKeeper scheme was a key measure in the Australian Government’s economic response to the COVID-19 pandemic. Following the phase one audit that examined the ATO’s management of risks related to the rapid implementation of six COVID-19 economic response measures (including the JobKeeper scheme)¹³, this audit focused on the ATO’s administration of the JobKeeper scheme. This audit also examined the ATO’s and Treasury’s strategies for evaluating the JobKeeper scheme and disseminating lessons learned.¹⁴

Audit approach

Audit objective, criteria and scope

1.19 The objective of the audit was to assess the effectiveness of the ATO’s administration of the JobKeeper scheme.

1.20 To form a conclusion against the audit objective, the following high-level criteria were adopted:

• Has the ATO effectively administered the legislative rules for the JobKeeper scheme?
• Has the ATO implemented fit for purpose arrangements to protect the integrity of JobKeeper payments?
• Has the ATO effectively monitored and reported on the operational performance of the scheme?

1.21 Under the third criterion, the scope of the audit included an examination of Treasury’s arrangements for evaluating the JobKeeper program and policy.

Audit methodology

1.22 The audit methodology included:

• examination of ATO documentation;
• analysis of JobKeeper data to assess timeliness of JobKeeper payments and correct payment rates;
• examination of Treasury documentation relating to the evaluation of the JobKeeper program and policy; and
• meetings with ATO and Treasury staff.

1.23 The audit considered feedback from members of the National Tax Liaison Group¹⁵ on aspects of the ATO’s administration of the JobKeeper scheme, and six submissions received through the citizen contribution function on the ANAO website.

1.24 The audit was conducted in accordance with ANAO auditing standards at a cost to the ANAO of $963,033.

1.25 The team members for this audit were David Willis, Samuel Painting, Evan Lee, Connor McGlynn, Chay Kulatunge, Matt Rigter, Omer Shaikh, Peta Martyn and Christine Chalmers.

2.1 To assess the effectiveness of the ATO’s administration of the legislative rules governing the JobKeeper scheme, this chapter examines whether the ATO:

• established effective processes to administer the legislative rules on entitlement, including changes to those rules during the course of the scheme;
• appropriately updated systems and processes when JobKeeper payment rates changed; and
• made JobKeeper payments in accordance with the timeframes set out in the legislative rules.

Did the ATO establish effective processes to administer the legislative rules on entitlement?

2.2 The ANAO examined whether the ATO established effective administrative design and decision-making processes to give effect to the rules on entitlement determined by the Treasurer. The legislative rules on entitlement for a JobKeeper payment are set out in part 2 of the Rules and are specified under three headings:

• Entitlement based on paid employees — an employer’s entitlement for an employee;
• Entitlement based on business participation — a business’ entitlement for an individual who is not an employee but is actively engaged in operating the business¹⁶; and
• Entitlement based on paid religious practitioners — a registered religious institution’s¹⁷ entitlement for a minister of religion or a full-time member of a religious order.

2.3 Part 2 of the Rules also lists the types of entities that do not qualify for the JobKeeper scheme.¹⁸ These include Australian government agencies and local governing bodies.¹⁹

Self-assessment approach and administrative design

2.4 The ATO administered the JobKeeper scheme, including the legislative rules on entitlement, on a self-assessment basis — in line with its approach to administering other parts of the taxation and superannuation systems.

2.5 The main features of the ATO’s self-assessment approach for the JobKeeper scheme were:

• applicants (or their intermediaries²⁰) were responsible for assessing their eligibility to receive JobKeeper payments by answering questions on forms developed by the ATO (with questions tailored based on information already held by the ATO);
• the ATO provided information and guidance in its forms, on its website and in direct stakeholder communication to support applicants to provide the correct information;
• applicants were not required to provide evidence upfront to support their self-assessment — they were required to declare that the information provided was true and correct; and
• the ATO undertook pre-payment and post-payment checks and compliance activities to support the integrity of the application and payment processes for the scheme — these included processes to ‘block’ entities that were ineligible under the Rules.

2.6 The decision-making process that led to the ATO administering the JobKeeper scheme on a self-assessment basis was not evident from the ATO’s records. Treasury advised the ANAO that the decision to adopt self-assessment was implicit in the original policy decision made by government that the ATO would be the key delivery entity leveraging its existing mechanisms.

2.7 The ATO’s early planning arrangements included establishing a ‘core design team’ to develop the legislation, processes and data management. The core design team was chaired by an Assistant Commissioner and was comprised of representatives from different program ‘streams’ such as compliance, marketing and communications, advice and guidance, internal readiness, IT and data, and rapid dispute resolution. The core design team was focused on the ATO’s readiness for JobKeeper applications to be made and the development of ‘experience pathways’ for applicants.²¹

2.8 An internal briefing pack, which the ATO advised incorporated the JobKeeper measure on 9 April 2020, indicated that the following key design decisions had been made by this date.

• Eligible employers to apply online, self-assess eligibility through a reduction in turnover and identify eligible employees.
• Employees to complete an approved nomination notice that is kept by the employer.
• Single Touch Payroll (STP) employers to have some prefilled employee data.²²
• Non-STP employers and the self-employed to have a more manual claim process.

2.9 The JobKeeper Program Board — the ATO’s key governance body for the scheme — held its first meeting on 14 April 2020²³, with the focus on the ATO’s ‘Day 1 Readiness Program’. A paper on ‘experience pathways’ for different JobKeeper parties (employers, intermediaries, employees) across the initial phases of the scheme was considered (Figure 2.1).

Figure 2.1: JobKeeper experience pathway, as of 14 April 2020

 

 

Note: Generic for all employees excluding non-digital.

Source: Reproduced from the ATO’s records.

2.10 The ATO’s intended approach was to focus on online forms and to use its existing online service channels as the main mechanism for accessing the scheme. Applicants would apply through one of three channels: ATO Online²⁴; Business Portal²⁵; or Online Services for Agents.²⁶ ATO documentation shows that 87.6 per cent of applications were made through the three online channels.²⁷

2.11 The paper considered by the JobKeeper Program Board on 14 April 2020 set out a number of principles for administering the JobKeeper scheme. This included the intention for payments to be made ‘as timely and seamlessly as possible.’ A 28 April 2020 paper set out the guiding principles for compliance, including that the ATO sought to make it easy for JobKeeper payments to be made to eligible applicants, and stated that strategies to detect suspected ineligible applicants would be developed.

2.12 The ATO continued to consider design issues over the course of the scheme. A JobKeeper Design Board was established in July 2020 following the announced extension of the scheme.²⁸ The Design Board was renamed the Economic Stimulus Design Board in September 2020. At its 5 August 2020 meeting, the Design Board decided to maintain the existing administrative arrangements for the JobKeeper extension period.

Administering the decline in turnover tests

2.13 The decline in turnover test was a key eligibility requirement for the JobKeeper scheme (Appendix 4). The Rules provided for four types of decline in turnover tests:

• a basic test — met by having projected, or experienced, a threshold reduction in turnover, which varied by entity type;
• an alternative test — met by satisfying one of eight alternative tests determined by legislative instrument²⁹;
• a modified test for certain group structures — met by having projected, or experienced, a specified reduction in turnover, when the turnover of each member of a consolidated, consolidatable or GST group is combined; and
• an actual decline in turnover test — introduced in the JobKeeper extension period and met by having experienced the threshold reduction in turnover, which varied by entity type.

2.14 JobKeeper applicants were required to satisfy the relevant test, along with other requirements³⁰, in order to be eligible for a JobKeeper payment.

Online application form

2.15 The online application form developed by the ATO for the original period of the scheme required applicants to determine the relevant turnover threshold and declare whether they had experienced, or were likely to experience, that reduction for a nominated month (Appendix 5). Questions were tailored for the applicant based on the information previously provided to the ATO.

2.16 The form did not require the applicant to specify:

• which relevant decline in turnover test they used in self-assessing their eligibility;
• whether the ‘turnover test period’ was for a month or quarter as provided for in the Rules and the Explanatory Statement to the Rules, instead asking applicants to nominate a month only;
• whether the claimed reduction in revenue was informed by a projection or on the basis of actual revenue data — as was considered to be an important distinguishing factor in the ATO’s subsequent compliance activities on the scheme³¹; or
• whether a cash or accrual accounting method was used in calculating their decline in turnover — the two options provided for in the ATO’s ‘administratively binding’ Law Companion Ruling on its approach to administering the decline in turnover test.

2.17 The ATO advised the ANAO that the forms for the original period of the scheme did not ask applicants which test they used to satisfy the decline in turnover test because of the speed in which the ATO had to implement the application process for the JobKeeper scheme.

2.18 An internal ‘Strategy finalisation report’ dated 23 October 2020 noted that in relation to the risk population of ‘self-preparers’ and ‘agents and intermediaries’:

not capturing the basic or alternative test information on the application form made it difficult to correctly determine the risk. In many instances the Agent had used actual turnover figures to determine eligibility … Many clients had also appropriately used the alternative test.

2.19 This same document also noted that not asking applicants to specify whether the cash or accrual method was used limited the use of Business Activity Statement data as a reliable indicator of whether the decline had eventuated.

2.20 The JobKeeper Design Board considered the question of whether the ATO should capture additional information in the JobKeeper extension period to assist in assurance over eligibility. A paper considered at the 14 August 2020 meeting recommended that the JobKeeper extension form capture details on the alternative decline in turnover tests. The ATO revised its online application form for the extension period to capture these details (Appendix 5).

Reflecting other rule changes in administrative processes

2.21 In addition to the actual decline in turnover test introduced for the JobKeeper extension period, other changes to the rules on entitlement were made over the course of the JobKeeper scheme. The ATO reflected these other rule changes in its online application forms, associated guidance and employee nomination notices (Table 2.1).

Table 2.1: Other rule changes reflected in the ATO’s administrative processes for JobKeeper

Note a: Explanatory Statement, Coronavirus Economic Response Package (Payments and Benefits) Amendment Rules (No. 2) 2020 (Cth), p.22.

Note b: Explanatory Statement, Amendment Rules (No. 7) 2020 (Cth), p.3.

Note c: Explanatory Statement, Amendment Rules (No. 8) 2020 (Cth), p.4.

Source: ANAO, based on the ATO’s records and public records.

Processes to block ineligible entities

2.22 The ATO’s general approach to block ineligible entities involved the compilation of an Australian Business Number (ABN) ‘exclusion list’ and the use of system-based exclusion rules to prevent ineligible entities from accessing the ATO’s online enrolment form and claiming a JobKeeper payment. The exclusion list drawn from ATO-held information was developed in April 2020 and included entities that do not qualify for the JobKeeper scheme listed in part 2 of the Rules, including entities who registered for an ABN after 12 March 2020.³²

2.23 The ATO identified that it was not able to effectively develop a population of all the exceptions provided for in subsection 7(2) of the Rules. Wholly-owned government bodies (as per paragraph 7(2)(d) of the Rules) had not been excluded because of difficulties in precisely identifying these entities from the source data available. Also, the ATO identified that its population of exclusions would be restricted to those entities that meet a tight definition of liquidation and bankruptcy.

2.24 The ATO reported in a compliance update to the Treasurer on 7 July 2021 that 69,500 ineligible JobKeeper enrolment attempts had been prevented. The ATO advised the ANAO that it did not maintain a report on the types of excluded entities that made up the total 69,500 enrolment attempts or which entities triggered the ‘blocks’ set up through its system access rules.

Have the ATO’s systems and processes been appropriately updated when payment rates have changed?

2.25 Section 13 of the Rules sets out the amount of the JobKeeper fortnightly payment.

• For the original period of the scheme, the payment rate was $1500 per fortnight for each eligible individual. The Explanatory Statement to the Rules noted that the $1500 payment provided the equivalent of approximately 70 per cent of the national median wage. Treasury’s three-month review of the JobKeeper Payment noted that the $1500 JobKeeper rate provided an ‘income transfer payment’ to some individuals.³³ That is, the $1500 rate was higher than some individuals’ ordinary wages.³⁴
• For the extension period, the JobKeeper payment rate was ‘stepped down’ in two stages and two payment ‘tiers’ were introduced — a higher rate and a lower rate (Table 2.2).

Table 2.2: JobKeeper payment rates, 28 September 2020 to 28 March 2021

Source: ANAO, based on section 13 of the Coronavirus Economic Response Package (Payments and Benefits) Rules 2020 (Cth).

2.26 Entitlement to the Tier 1 rate generally depended on whether an individual satisfied the relevant 80-hour threshold over the specified ‘reference period’³⁵; otherwise, the Tier 2 rate applied. The Australian Government’s announcement of the JobKeeper extension noted that the two-tiered payment system was designed to better align the payment with the incomes of employees before the onset of the COVID-19 pandemic.

2.27 The ANAO tested a population of more than 8.8 million JobKeeper transactions³⁶ to determine whether the ATO’s payment systems and processes used the correct payment rates for different periods of the JobKeeper scheme.³⁷ The transactions covered the period from 3 May 2020 to 28 February 2021, and involved net JobKeeper payments and receipts of over $84.4 billion. Of this total, $72.1 billion in transactions were system-automated without any manual intervention by ATO staff. The balance of the transactions, $12.3 billion (15 per cent), involved manual intervention. Manual interventions could include entering the complete transaction as the result of a telephone call, reversing or amending a transaction as a result of compliance procedures or amending a transaction at the request of the entity.

2.28 The ANAO tested the transaction amount through a combination of automated evaluation and sample testing.³⁸ The testing took account of the relevant claim period and rate, the number of employees, and employees’ respective payment rates as reported by entities in monthly JobKeeper claims. Thirteen per cent of the 8.8 million transactions were updating non-financial data and had no financial effect. The remaining 7.7 million transactions were found to be materially correct.³⁹

Were JobKeeper payments made in accordance with required timeframes?

2.29 Section 15 of the Rules sets out the timeframe in which the Commissioner of Taxation (‘the Commissioner’) must pay JobKeeper payments.

The Commissioner must pay the jobkeeper payment no later than the later of:

(a) 14 days after the end of the calendar month in which the fortnight ends; and

(b) 14 days after the requirements in section 14 for the Commissioner to make the payment are met.⁴⁰

2.30 In relation to section 14 of the Rules, the Explanatory Statement notes that if the Commissioner is satisfied that an employer or business is entitled to a JobKeeper payment for a fortnight, the Commissioner must pay the employer or business the JobKeeper payment. The ATO advised the ANAO that the practical application of section 14 is that when an entity lodged its monthly JobKeeper declaration, the Commissioner could choose to accept the statements and be satisfied on the basis of those statements to make the payment to the entity. Alternatively, the Commissioner may not be satisfied and could undertake checks before making a payment. The ATO further advised that the Rules did not prescribe when the Commissioner must become satisfied. That is, the Rules did not set a time limit on how long the ATO could take to determine whether an entity was entitled to a JobKeeper payment.

2.31 Automated testing of 7,329,699 JobKeeper payment transactions for 3 May 2020 to 28 February 2021 was performed by the ANAO to identify the number of payment transactions that were disbursed by the ATO within 14 days of a JobKeeper application form being received by the ATO.⁴¹ A total of 7,265,348 payment transactions (99 per cent) were paid within 14 days (Figure 2.2). The balance of the tested population (64,351 payment transactions) were paid after 14 days. The average timeframe for making a JobKeeper payment was four days.

Figure 2.2: Distribution of JobKeeper payments made within 14 days

Source: ANAO, based on the ATO’s data.

2.32 The ATO’s goal was to make timely payments whilst also checking the integrity of payments. A 27 April 2020 paper considered by the ATO recommended a four-day rather than three-day payment cycle, allowing the ATO an additional day to run ‘risk rules’ and undertake fraud checks.

2.33 In relation to the number of payments processed within two days, the ATO advised the ANAO that risk cases were still able to be identified and removed prior to any payment file being generated.

2.34 The ANAO analysed the time distribution of the 64,351 payments made after 14 days, nearly half of which were paid within 24 days (Figure 2.3). The longest timeframe for disbursement of a JobKeeper payment was 306 days.

Figure 2.3: Distribution of JobKeeper payments made after 14 days

Source: ANAO, based on the ATO’s data.

3.1 In administering the legislative rules for the scheme, the Commissioner of Taxation (the Commissioner) and the ATO had responsibility, and broad discretion, for determining and implementing measures to protect the integrity of JobKeeper payments.⁴²

3.2 To assess whether the ATO implemented fit for purpose arrangements for protecting the integrity of JobKeeper payments, this chapter examines whether the ATO:

• developed a compliance framework and strategy for the JobKeeper scheme;
• implemented key compliance and integrity measures as intended;
• conducted ‘decline in turnover’ reviews in accordance with its internal procedures; and
• exercised discretion on overpayments in accordance with its internal policies and procedures.

Did the ATO develop a compliance framework and strategy for the JobKeeper scheme?

3.3 The expectation that the ATO develop a compliance framework and strategy for the JobKeeper scheme is based on the materiality of the scheme and on the principle of a ‘risk-based’ approach to inform and promote the proper use and management of public resources under the Public Governance, Performance and Accountability Act 2013.⁴³

Compliance strategy for the original period of the scheme

3.4 At the first meeting of the JobKeeper Program Board on 14 April 2020, it was stated that ‘the highest level of assurance and integrity is maintained to monitor and act to identify and prevent apparent misuse and fraud’. A ‘JobKeeper Payment: Compliance Risk Framework’ was included in the Program Board papers. The draft framework set out the ATO’s top eight risks for JobKeeper payments and the intended payment action for each risk: ‘Payments should be stopped’ or ‘Pay but review before second payment’ (Figure 3.1). The draft framework also outlined the ATO’s intended mitigation measures for the top eight risks.

Figure 3.1: The ATO’s top eight risks from its draft JobKeeper Payment: Compliance Risk Framework (14 April 2020)

 

 

Source: ANAO, based on the ATO’s records.

3.5 Separate to the draft compliance risk framework, the ATO progressively developed a compliance strategy. Six iterations of the strategy were presented to the Program Board between 21 April and 3 July 2020. The Program Board suggested ways to improve the strategy, including ensuring the identified risks are examined from a ‘whole of ATO compliance lens’. The Program Board noted that the risks identified for the JobKeeper scheme could be applicable to the other stimulus measures being administered by the ATO.

3.6 The compliance strategy was presented to the ATO Executive Committee⁴⁴ on 30 June 2020. A companion paper⁴⁵ noted that the intent was to:

• support an economic stimulus package with the principal policy aims of keeping people in business and employment;
• pay JobKeeper payments to businesses as quickly as possible with minimal friction points and high levels of certainty for those that are entitled; and
• identify and treat those who are not entitled, have made mistakes or who intentionally defraud the system.

3.7 The Commissioner noted that the compliance program was both comprehensive and collaborative. The Commissioner emphasised the importance of obtaining and retaining the learnings and ensuring that the capability was systemised and held within the ATO for future initiatives.

3.8 The final version of the compliance strategy, presented to the Program Board on 3 July 2020, included the objective ‘To provide confidence that we will make timely payments while maintaining integrity and fairness and dealing with fraud to the system’. The strategy outlined assurance approaches and specific risk tolerances for 10 identified risks covering entity eligibility, employee eligibility, JobKeeper obligations and decline in turnover declarations (Appendix 6).

3.9 As with the draft compliance risk framework, a feature of the compliance strategy was a combination of pre-payment and post-payment verification. Pre-payment verification included the option of ‘suppressing’ (stopping) payments that triggered pre-set risk rules. At the enrolment stage from 20 April 2020 onwards, the ATO’s intended strategy was to block entities identified in the Rules as being ineligible for JobKeeper payments from progressing to the application stage of the scheme.

3.10 For the original period of the scheme, the ATO developed ‘treatment plans’ for eight payment risks (Table 3.1), which set out the ATO’s detailed intentions about how to manage the specific payment risks.

Table 3.1: Risk-specific treatment plans prepared by the ATO for the original period

Note a: Previously called Public and Multinational Businesses.

Source: ANAO, based on the ATO’s records.

3.11 There was broad alignment between the payment risks identified in the compliance risk framework, compliance strategy and risk treatment plans.

3.12 The ATO’s governance arrangements for the JobKeeper scheme did not specifically identify which person or body was responsible for approving the risk treatment plans. The June 2020 JobKeeper Program Board Charter did not include a specific role to review and approve the risk treatment plans for the original period of the scheme. The JobKeeper and JobMaker Hiring Credit Program Board Charter, which was formalised during the extension period, also did not include a specific role to review or approve the risk treatment plans.

3.13 The ATO advised the ANAO that three business lines were responsible for the development and execution of the treatment plans for their relevant areas of focus⁴⁶, in concert with the ‘stream lead’ for compliance.⁴⁷ The three business lines and stream lead for compliance were responsible for reporting to the Senior Responsible Officer and the Program Board. The ATO was unable to demonstrate that all eight risk treatment plans had been formally approved within the relevant business lines or by the stream lead for compliance.

Compliance strategy for the extension period

3.14 The ATO prepared a new compliance strategy for the extension period. A ‘JobKeeper Extension–Compliance Approach’ was provided for information to the Program Board at its 18 November 2020 meeting. The strategy reflected learnings from the original period (Figure 3.2).

3.15 The ATO designed a seven-step compliance model for the extension period based on a ‘whole of client’ view of risk across an entity (Figure 3.3).

Figure 3.2: ATO learnings from JobKeeper into JobKeeper extension

 

 

Note: ‘1:many’ refers to ATO bulk communications and ‘1:1’ refers to ATO engagement with individual taxpayers.

Source: Reproduced from the ATO’s records.

Figure 3.3: Overview of the ATO’s compliance model for the extension period

 

 

Source: Reproduced from the ATO’s records.

3.16 The extension period compliance model was designed to calculate an overall ‘Client Level Risk Score’ for each JobKeeper recipient, based on the likelihood of ineligibility and financial consequences of ineligibility (such as the number of employees being claimed for). The likelihood was to be derived from the client’s demographic characteristics (for example, type of industry) and six risks models, comprising: ‘Eligible business participant eligibility’, ‘Employee eligibility and (payment) tiers’, ‘Identity take over’, ‘Signs of life’, ‘Turnover’, and ‘Agents of threat’.

3.17 The ATO prepared treatment plans for seven payment risks in the extension period (Table 3.2). None of the seven risk treatment plans were dated and there was no evidence of the plans being approved in writing.

Table 3.2: Risk-specific treatment plans prepared by the ATO for the extension period

Source: ANAO, based on the ATO’s records.

Have key compliance and integrity measures been implemented as intended?

3.18 To assess whether key compliance measures had been implemented as intended, the ANAO selected one or more compliance measures from each of the ATO’s 15 risk treatment plans. A total of 22 compliance measures were selected.⁴⁸ The selection was principally based on compliance measures that the ATO had directed at higher-risk populations or behaviours, which were typically intended to involve direct engagement with JobKeeper recipients. Evidence was sought from the ATO on whether it had implemented the selected measures as intended or implemented any approved changes to the measures (Appendix 7).

3.19 The ATO provided ‘case lists’ of individual compliance activities extracted from its case management system, Siebel. The case lists represented the ATO’s initial claims about the implementation of intended compliance measures. The ATO provided lists of compliance cases for each of the 22 compliance measures selected by the ANAO. From a total of 41,031 compliance cases in the case lists across both the original and extension periods, 735 cases were selected for testing. For each compliance case, the ANAO sought evidence that the intended action had been undertaken.

Implementation of compliance measures

3.20 Of the 22 compliance measures tested, two were partly implemented as intended, seven largely as intended and eight fully as intended. For five measures comprising 202 compliance cases, the ANAO was unable to conclude on whether the intended compliance measures were implemented (Table 3.3).

• A total of 30 compliance cases with substantive exceptions – that is, where the ATO did not evidence that its intended compliance action had been fully undertaken or that an eligibility outcome had been recorded – were identified across nine of the 17 compliance measures that could be tested.
• There were 41 compliance cases across 16 of the compliance measure populations where issues were identified with the reliability of the ATO’s case lists. These included compliance cases that were created in error and closed without compliance action, and apparent duplicates in the case lists.

3.21 The highest number of exceptions (15 among the 49 cases tested) was for the Wage Condition treatment plan for the original period, which was intended to ensure that employers receiving JobKeeper payments were paying their employees. The 15 exceptions comprised:

• five compliance cases that were closed together with no further attention given to the individual cases in February 2021 — six months after a review was opened — with no indication that the wage condition risk had been addressed;
• five compliance cases for which there was not sufficient evidence that the risk had been addressed (for example, no notes recorded);
• three compliance cases that were closed due to being ‘likely low risk’ although there was no confirmation that the risk had been addressed; and
• two compliance cases in which a case officer recorded that the risk had been addressed through the completion of an eligibility checklist; however, verifying the wage condition was not included on the checklist.

3.22 In relation to the ATO’s JobKeeper/JobSeeker risk treatment plan, JobKeeper data was regularly shared with Services Australia from May 2020 to March 2021. The principal aim of the data sharing was to identify persons who were in receipt of both JobKeeper and JobSeeker payments — which was not permitted under the JobSeeker payment rules.

3.23 The five compliance measure populations where testing was not able to determine whether they had been implemented as intended involved the application of the ATO’s ‘Action Differentiation Framework’ (ADF) — a key element of the extension period compliance model. The intent of the ADF was to align the treatment approach to the identified level of risk. Each entity was assigned to one of four ADF categories:

• Detect and monitor — monitoring groups of entities for risk through data analysis;
• Detect, educate and encourage — promoting compliance across a group of entities through outbound telephone calls, emails and letters to entities;
• Detect and review — reviewing high-risk payments without applying a suppression; or
• Detect, stop and review — suppressing the highest-risk payments prior to review.

3.24 During the ANAO’s initial testing of the five populations, inconsistencies were identified in all five populations between the action recorded in the ATO’s case lists and the compliance action that was evident from the ATO’s case documentation. There were 106 inconsistencies among the 202 sampled compliance cases in these five populations. These included 47 compliance cases where ‘detect, stop and review’ was recorded in the case list, but there was no evidence of payments being suppressed before the review commenced, or at all. A further eight cases were identified where either ‘detect, stop and review’ or ‘detect and review’ was recorded in the case list, but there was no evidence of a review being completed. Following testing, the ATO advised the ANAO that the ADF category recorded in the case lists was not a reliable indicator of the ATO’s intended action:

The ADF categories were not always final actions undertaken by the ATO. Each risk was monitored on an ongoing basis … Where the action taken does not match the ADF, a decision would be made based on a number of factors… The ADF rating in these cases would not be solely relevant in selection. The data showing the ADF category does not get updated if we [the ATO] decide to undertake a review on the client. The ADF is based on automated risk models and is not manually changed if for example a risk manager decides to review a ‘monitor’ ADF case.

3.25 On this basis, no finding could be made on whether the ATO’s compliance measures over the five affected populations were implemented as intended.

Oversight of compliance measures

3.26 The ATO’s governance arrangements for the JobKeeper scheme did not specifically identify which person or body was responsible for overseeing the implementation of the risk treatment plans for the original period of the scheme. The ATO advised the ANAO that its existing business lines were responsible for the development and execution of the treatment plans for their relevant areas of focus. No document was provided to support this advice or to set out the respective responsibilities of business lines and the Program Board.

3.27 The Program Board received a range of reporting on the ATO’s compliance strategies, activities and outcomes. In addition, there was evidence of progress and outcomes reporting being provided within the business lines for some of the risk treatment plans.

• A ‘Strategy Finalisation Report’ (dated 23 October 2020) on the Small Business Decline in Turnover Risk treatment plan provided to an Assistant Commissioner in the Small Business line.
• An ‘Evaluation Report’ provided on 1 December 2020 to the Strategic Management Committee within the ATO’s Public Groups and International business line on the decline in turnover compliance measures undertaken within this business line.
• A ‘Risk Insights Report’ provided to the ATO’s stream lead for compliance (Assistant Commissioner level) on 5 October 2020 outlining compliance activities undertaken within the Private Wealth business line for the decline in turnover risk.

3.28 The reporting to the Program Board did not typically provide a clear ‘line of sight’ back to the intended compliance measures from the ATO’s risk treatment plans.

Did the ATO conduct ‘decline in turnover’ reviews in accordance with its internal procedures?

3.29 The decline in turnover test was a key eligibility requirement for the JobKeeper scheme and an area of focus for the ATO’s compliance activities. The ATO’s published guidance emphasised the need for an entity’s projected decline in turnover to be ‘reasonable’ and for the entity to maintain records to demonstrate that a reasonable approach was taken.⁴⁹

3.30 The extension period compliance strategy stated that over 1000 ‘decline in turnover reviews’ had been undertaken by business lines on applications received during the original period of the scheme and that over 97 per cent of the cases reviewed were found to be eligible. A similar result was reported publicly. On 10 September 2021, the Commissioner advised the Senate Economics Legislation Committee that:

We have undertaken a comprehensive review of cases that forecast a decline in turnover and found the vast majority of taxpayers undertook the projected decline in turnover test in good faith. From our review of more than 1,600 entities across all markets, including 480 large businesses, we found more than 95% were eligible.⁵⁰

Decline in turnover reviews

3.31 The ANAO asked the ATO to provide a list of the decline in turnover reviews that were referenced in the compliance strategy. From a list of 1619 decline in turnover reviews provided by the ATO in March 2021⁵¹, a sample of 40 reviews was selected by the ANAO for testing. The sample comprised 30 cases listed as ‘eligible’ and 10 cases listed as ‘ineligible’. Reviews were undertaken by three business lines — Public Groups and International, Private Wealth, and Small Business.

3.32 The ATO developed a range of internal procedures and guidance material to support the decline in turnover reviews. This included a ‘risk guide’, first issued on 10 July 2020, designed to assist staff when actioning activities relating to entities identified as potentially not meeting the decline in turnover test. The risk guide set out four main tasks for ATO staff to complete and included different guidance for the three business lines conducting the decline in turnover reviews.

• Profile the client to understand the risk.⁵²
• Understand the product that is being recommended.
• Contact the client.
• Determine the outcome.

3.33 For Public Groups and International, the options for conducting a decline in turnover review included sending a ‘nudge’ email to the taxpayer where they were of a ‘lower consequence’ (fewer than 25 employees) or a Significant Global Entity that was applying the 30 per cent decline in turnover rate and had not experienced an actual decline of 50 per cent or more (Appendix 4).⁵³ The risk guide stated that the nudge approach was designed to prompt entities to reconsider their eligibility for JobKeeper payments by checking for any errors in their decline in turnover test. The ATO advised the ANAO that 113 nudge emails were sent by the Public Groups and International business line.⁵⁴

3.34 The risk thresholds that underpinned the nudge email for Public Groups and International were not required to be applied by the two other business lines — Private Wealth and Small Business. The ATO advised the ANAO that the different thresholds and treatments across the three business lines was attributed to three factors:

• ATO policy to segment the taxpayer population by different client experiences;
• the differences in the severity of behaviours that caused risks to manifest; and
• the risks that were being treated.

3.35 Public Groups and International also had the option of issuing a formal request for information, seeking detailed source documents from the taxpayer. The risk guide stated that this approach was to be used by exception and only after a telephone questionnaire had been used. The ATO advised the ANAO that no formal requests for information were issued.

Implementation of decline in turnover reviews

3.36 The main issues identified by the ANAO through testing concerned the nature of the evidence sought and received by the ATO to gain assurance that the entity had satisfied the decline in turnover test.

3.37 A requirement in the ATO’s risk guide, which applied to all three business lines, was the need to obtain documents to substantiate the entity’s decline in turnover claims. Where documentation was directly sought from entities in line with this requirement, there was variability in the nature and quality of the records provided. Some entities provided a spreadsheet including only a decline in turnover calculation, which provided no indication of when the calculation was completed. The ATO acknowledged the general variability of the documentation and advised the ANAO that ‘there was no definition of what records were required to be kept or how to present the calculation, so we received large variations between clients.’

3.38 In three cases where direct contact was made (which were all assessed to be eligible), no documentation was included in the ATO’s records. The ATO advised the ANAO that these three cases were part of a private group comprising 29 employer entities. In such cases, the ATO’s risk guide allowed the Private Wealth business line to determine which employer entities to request working papers from, having regard to factors such as the number of employees and the expected impact on the relevant industry. The ATO’s records indicate that working papers were requested from six of the 29 entities.

3.39 The ANAO’s testing sought to identify whether the ATO obtained evidence from entities on when the decline in turnover projection was made and who prepared the projection. This first aspect reflected a principle in the ATO’s published guidance that the decline in turnover projection needed to be a reasonable assessment of what was likely at the ‘point in time’ an entity calculated the test.⁵⁵ The ATO was generally accepting of entity assertions with regard to the manner in which the decline in turnover test was completed and did not seek to verify the responses provided, by, for instance, requesting primary documentation to evidence the date on which the projections were produced (such as emails or board papers). This reduced the intended assurance that entities had completed their decline in turnover projection before they submitted their application.

3.40 In conducting the reviews, the ATO typically sought to identify which decline in turnover test the entity used to establish their eligibility — the basic test, the alternative test or the modified alternative test for group employers. As discussed in Chapter 2 (paragraph 2.16), the ATO’s online form for the original JobKeeper scheme did not require entities to specify which test was used.

3.41 The sampled reviews included one case, categorised as eligible, where a nudge email was sent to the entity. The ATO’s profiling document noted that the entity’s reported revenue increased by 152 per cent, whereas a decline of 50 per cent had been nominated on the entity’s application form. The entity was categorised as ‘low consequence’ (fewer than 25 employees), in line with the ATO’s risk guide for the Public Groups and International business line. The nudge email encouraged the entity to review its eligibility for JobKeeper payments and contact the ATO to discuss any issues. The ATO’s records state that no response was received. Based on the information gathered by the ATO, the potential overpayment was up to $360,000.

3.42 The ATO’s risk guide provided the three business lines with different options for profiling the client and recording the outcome. The ATO created profiling templates and an ‘analysis and decision template’, but these were not required to be used across the three business lines. The ANAO observed variability in the ATO’s documentation in relation to the entity profiling step and recording the outcome of the review.

3.43 For the 10 sampled reviews that were assessed by the ATO as ‘ineligible’, the ATO provided sufficient evidence that the proposed decisions were referred for approval in line with the requirements for each business line set out in the risk guide.

Has the ATO exercised discretion on overpayments in accordance with its internal policies and procedures?

Legislation, policy and guidance on the exercise of discretion on overpayments

3.44 The JobKeeper legislation provides the Commissioner with the discretion to make a written determination that an entity is not liable to repay an amount of a JobKeeper payment where the entity was either not entitled to the payment or the amount paid was more than the amount to which the entity was entitled.⁵⁶ The Explanatory Memorandum to the Coronavirus Economic Response Package (Payments and Benefits) Bill 2020 stated that:

This ensures that the Commissioner has flexibility to address issues that might otherwise arise where entities may have made an honest mistake and not retained any personal benefit from a payment they have received.⁵⁷

3.45 On 10 September 2021, the ATO publicly reported that it had identified a total of $470 million worth of JobKeeper overpayments, of which $180 million had been waived following the exercise of discretion.⁵⁸

3.46 The ATO developed a range of internal guidance material to support the exercise of the Commissioner’s discretion on JobKeeper overpayments. This included the ATO’s policy position on the practical application of the discretion, approved by the Senior Responsible Officer for the JobKeeper scheme on 16 June 2020, as well as four ‘Practice Notes’ to support decision-making by ATO officers.⁵⁹

3.47 In considering the practical circumstances under which the Commissioner should exercise the discretion to waive repayment, the ATO determined, based upon the Explanatory Memorandum, that the discretion is to be exercised broadly, when:

• there has been an honest mistake, including where there is reasonable reliance on a statement made by the other entity in their nomination notice; and
• there has been no direct financial benefit retained by the entity that received the JobKeeper payment.

3.48 In addition, the ATO’s policy position included ‘concessional treatment’ for overpayments identified by the ATO in the first four JobKeeper fortnights.

• For the first two fortnights, discretion could be exercised solely on the basis that an honest mistake was made by the applicant, regardless of whether or not a financial benefit had been retained. The ATO’s stated basis for taking this approach included the rapid implementation and complexity of the scheme and the fact that ATO guidance materials were still being developed after JobKeeper applications had opened.
• For the second two fortnights, other factors were to be used in determining whether discretion should be exercised, including whether the ATO had previously contacted the entity in relation to eligibility issues.
• For all four fortnights, the ATO determined that where a JobKeeper recipient had independently contacted the ATO to repay an overpaid amount, the ATO would process the repayment and not offer to waive it.

3.49 For the extension period, concessional treatment was expanded to apply to the entire original period (the first 13 JobKeeper fortnights). The ATO’s rationale for this change was that it would be unreasonable to recover these amounts after the conclusion of the original period. The ATO also considered that exercising discretion would provide closure for ineligible recipients by indicating that the ATO would not later seek to recover past overpaid amounts. Under this policy, JobKeeper recipients found ineligible could keep all overpaid amounts from the original JobKeeper period, provided that ATO was satisfied that they had made an honest mistake.

3.50 The Practice Notes broadened the ATO’s earlier decision that the application of the Commissioner’s discretion was to be exercised broadly by adding ‘including’ when there has been an honest mistake and no financial benefit retained. The Practice Notes state that:

Ultimately, it will be a question of whether it is reasonable to not require the repayment of the overpayment of JK in all the circumstances. This requires consideration of the interests of both the entity and the broader community. The circumstances surrounding the rapid implementation of the JK scheme will also be relevant, including the need for quick decisions about eligibility, the fact that the provisions were concerned with getting stimulus into the economy as fast as possible and retaining jobs as an emergency measure, the fact that JK was intended to have a response effect in those receiving it, and that the recovery of good faith claims may leave employers in a worse position than if they didn’t claim at all.

3.51 The Practice Notes also stated that, in considering whether it is reasonable to not require repayment, a significant factor would be the ‘honesty of the entity’s mistake’, and another significant factor would be ‘the extent to which receipt of JK payments has provided a financial benefit.’ The Practice Notes explained the ATO’s concept of an ‘honest mistake’ and ‘financial benefit’ (Appendix 8) and provided illustrative examples of each.

3.52 The Practice Notes did not include specific guidance on how ATO staff were required to support their decisions if the significant factors of honest mistake and retention of financial benefit were determined not to apply, or if other factors were considered more material to the exercise of discretion.

3.53 The ATO established overpayment review and advisory panels to provide a consistent view over overpayment decisions, particularly for higher value and more complex cases. The exercise of discretion based on considerations other than the two ‘significant factors’ listed in the Practice Notes was one of the reasons that would require mandatory referral to the panels.

Implementation of the Commissioner’s discretion in relation to overpayments

3.54 The ATO’s understanding of the law was that the Commissioner’s discretion could not be strictly limited by its policies and procedures.⁶⁰ The ATO advised that ‘While the concepts of honest mistake and retention of [financial] benefit were primary considerations, they were not conclusively determinative in the final decision as to whether the overpayment discretion was exercised’, and further advised that:

The final decision on each matter, regardless of the content of the Practice Notes, was wholly determined by the consideration of the relevant decision-maker as to whether the exercise of the discretion (in whole, in part, or not at all) was reasonable based upon the circumstances of the case.

3.55 In relation to the question of whether decisions to exercise discretion on JobKeeper overpayments were made in accordance with its internal guidance, the ATO advised that: ‘Unless the discretion decision made was established to be unreasonable in the eyes of the relevantly delegated and experienced decision-maker, then it was made in accordance with the Practice Notes’.

3.56 Despite this discretion, there is a requirement under the Public Governance, Performance and Accountability Act 2013 for the accountable authority of a Commonwealth entity to promote the proper use and management of public resources. The general expectation is that entities provide transparency over and accountability for the use of public funds.

3.57 In this context, ANAO testing focused on whether the ATO had clearly documented the basis for its overpayment decisions — in particular, how the decisions took account of the two ‘significant’ factors listed in its Practice Notes (honest mistake and retention of financial benefit) and whether any other factors relevant to the decision were explicitly identified. The expectation that decisions are clearly documented does not limit the exercise of discretion.

3.58 The population of overpayments as of 8 June 2021 comprised 32,514 unique clients who had received overpayments totalling $433 million. Sixty-three JobKeeper overpayments, totalling $57.8 million, were sampled for testing using a stratified sampling approach based on three decision outcomes — full write-off, partial write-off and full repayment. The testing targeted the highest value cases in the full write-off and full repayment categories.

3.59 Of the 63 overpayments examined, 40 decisions were explicitly based on the significant factors of ‘honest mistake’ and ‘financial benefit’ as set out in the Practice Notes. Nine cases were assessed as ‘not applicable’ for a range of reasons.⁶¹ In the remaining 14 cases, the ATO’s documentation did not clearly set out how the overpayment decisions took account of the two significant factors (Table 3.4).

Table 3.4: Implementation of the Commissioner’s discretion in relation to overpayments where the two significant factors were not documented

Source: ANAO, based on the ATO’s records.

3.60 Three of the five overpayment decisions that were based on factors other than the two significant factors did not include evidence of referral to an overpayment review and advisory panel — as was required under the ATO’s internal guidance.

Source: ANAO, based on the ATO’s records.

4.1 The ATO and Treasury had responsibilities for monitoring and reporting on the performance of the JobKeeper scheme — reflecting their respective roles for administration and policy. This chapter examines whether the ATO:

• implemented sound arrangements for monitoring scheme performance; and
• reported on the scheme in a timely and informative manner.

4.2 In addition, this chapter examines Treasury’s arrangements for evaluating the JobKeeper program and policy.

Has the ATO implemented sound arrangements for monitoring scheme performance?

4.3 As the entity responsible for administering the JobKeeper scheme, the ATO was able to determine what arrangements would be appropriate for monitoring scheme performance.⁶² It outlined its intended approach to administering the scheme in a paper provided to the first meeting of the JobKeeper Program Board on 14 April 2020. The paper noted the ATO’s intentions that:

• a strong and robust governance framework is maintained;
• the coordination of the various streams of work are in place, which need to come together to enable the program to succeed; and
• the monitoring and reporting of key performance indicators and milestones are in place to support the policy objective.

4.4 The ATO advised the ANAO that the governance arrangements determined for the JobKeeper scheme were the principal means by which the ATO monitored scheme performance.

Governance framework

4.5 A proposed governance structure for the JobKeeper scheme⁶³, and related arrangements, were set out at the 14 April 2020 meeting of the JobKeeper Program Board (Figure 4.1).

Figure 4.1: The ATO’s proposed governance structure for the JobKeeper scheme (14 April 2020)

Source: ANAO, based on the ATO’s records.

4.6 The Program Board and the Core Design Team operated from April 2020. The JobKeeper Oversight Committee was not established and its intended role was instead absorbed into the Program Board.

4.7 Following a review of the JobKeeper accountability and governance structure by PricewaterhouseCoopers, the ATO established the Economic Response Oversight Committee in August 2020 to ‘provide strategic direction and overall program assurance of the ATO’s management of the COVID-19 Economic response measures’, which included the JobKeeper scheme.⁶⁴ The Economic Response Oversight Committee comprised three ATO Second Commissioners and was supported by a range of advisers at Deputy Commissioner and Assistant Commissioner level (Senior Executive Service Band 2 and 1). The Economic Response Oversight Committee held its first meeting on 25 August 2020. Material considered included a table of ‘program level tolerance thresholds’ for different aspects of program management (such as cost, resources and risk) above which an ‘exception report’ was to be raised with the Committee. This requirement was reflected in the charter for the JobKeeper and JobMaker Hiring Credit Program Board, which superseded the JobKeeper Program Board in October 2020.

4.8 Other material provided to the Program Board on 14 April 2020 included:

• a program structure — comprising an ‘oversight level’ by a Senior Responsible Officer, a ‘program level’ by a program lead, program management directors and a ‘stream level’ involving seven different functions⁶⁵;
• key program accountabilities — involving a reporting structure; and
• the planned schedule and content of governance meetings and reporting — including reporting to Treasury and ministers’ offices.

4.9 Different governance structures were considered and implemented by the Program Board over the course of the JobKeeper scheme. Common features of the governance arrangements included a key role for the Program Board and for the Senior Responsible Officer.

4.10 Following the establishment of the JobKeeper and JobMaker Hiring Credit Program Board in October 2020, the Senior Responsible Officer became the sole decision maker for the JobKeeper program. Previously, the JobKeeper Program Board’s role was to make ‘significant decisions’. The Program Board’s initial charter did not delineate the Senior Responsible Officer’s and the Program Board’s respective decision-making authorities.

4.11 The JobKeeper Program Board and Senior Responsible Officer were supported by other bodies. This included the JobKeeper Design Board, which was responsible for making design decisions. There were three iterations of an ‘administrative’ group responsible for working through ‘critical priority issues’ to advise the Program Board.

4.12 The ATO’s documented governance arrangements for the Program Board and the Economic Response Oversight Committee included intentions about how often these bodies should meet and the type of reporting they should receive. From October 2020, both bodies were expected to meet on a fortnightly basis⁶⁶ and to receive a JobKeeper ‘program status report’ that included a program summary, key decisions, a risk/issue summary, payment statistics and stream updates. Review and analysis of the agenda papers for both bodies showed that the intended meeting frequencies and reporting arrangements were met.

4.13 The provision of program status reports to the Program Board was a key mechanism through which the ATO monitored the various streams of work in place for the JobKeeper scheme. The reporting typically provided a summary and statistics on current activities and an overview of upcoming activities.

4.14 In October 2020, the Economic Stimulus Branch was established, with the JobKeeper Senior Responsible Officer appointed as its head. All monthly meetings of the Economic Stimulus Branch executive included stream performance reports with updates from the three branch streams.

4.15 The Program Board’s governance role to escalate risks that exceeded tolerance thresholds to the Economic Response Oversight Committee did not eventuate in the manner intended. No ‘exception reports’ were escalated. The Committee was informed of emerging risks and issues in other ways such as through project status updates, meeting papers on JobKeeper identity security and fraud, and court and tribunal decisions related to JobKeeper. From February 2021, the JobKeeper status report included a ‘Risk Profile Summary’ setting out risk levels (High, Medium, Low) against different risk areas (Client experience, Compliance, Advice, and Data).

4.16 At its final meeting on 10 June 2021, the JobKeeper and JobMaker Hiring Credit Program Board discussed ongoing governance and management arrangements for the JobKeeper scheme. A paper asked the ATO’s Executive Committee to note that:

• specific advice or compliance action for both the JobKeeper and JobMaker programs would thereafter be managed within the ATO’s Superannuation and Employer Obligations business line;
• the Economic Stimulus Branch, which was established in October 2020 to centralise administration of the JobKeeper scheme for the extension period, would be disbanded (effective from 19 June 2021); and
• ongoing Senior Executive Service level accountability, governance and reporting arrangements for the JobKeeper scheme were being finalised.

4.17 From July 2021, the Charter for the newly established JobMaker Hiring Credit Steering Committee included oversight of residual work being undertaken for the JobKeeper scheme.

Performance indicators and milestones

4.18 The 14 April 2020 JobKeeper Program Board paper stated that the ATO would monitor and report on key performance indicators and milestones to track whether the policy objective of the scheme was being met. The paper did not define the policy objective and described the outputs as: ‘a “wage subsidy” paid in arrears to “eligible employers” for their “eligible employees” in accordance with the law governing this program from May – September 2020.’

4.19 The ATO monitored and reported internally on performance in different reports, including:

• Program Board status reports — initially weekly reports that included an update from each of the ‘stream’ leads and performance information, such as key decisions, risks and issues, dashboard reporting on applications and payments, and data on community interactions including ‘tip-offs’ and complaints;
• Economic Response Oversight Committee status reports — initially fortnightly reports that included a list of performance indicators and controls covering items such as total budget, eligible applicants, complaints received, objections and fraud detected, as well as an assessment of key risks;
• Corporate project status reports — monthly reports that included a summary of JobKeeper applications and payments, a ‘traffic light’ assessment of delivery status, budget information, risks and issues, and a status update on deliverables.

4.20 The ATO produced iterations of an internal document titled ‘JobKeeper Payment Program Outcomes’ (Figure 4.2). The outcomes were provided to the JobKeeper and JobMaker Hiring Credit Program Board on 13 November 2020 for the original period and on 10 June 2021 for the extension period. The outcomes were also provided to the Economic Response Oversight Committee on 1 February 2021 for the original period and on 11 June 2021 for the extension period.

Figure 4.2: The ATO’s summary of JobKeeper Payment Program Outcomes (as at 30 April 2021)

 

 

Source: Reproduced from the ATO’s records.

4.21 On 18 February 2021, a paper to the JobKeeper and JobMaker Hiring Credit Program Board outlined, for both the JobKeeper and the Cash Flow Boost programs, what effective ATO administration would look like and proposed five performance criteria (Box 2).

Source: ANAO, based on the ATO’s records.

4.22 The Program Board was advised that the five proposed measures had been approved by the Senior Responsible Officer for JobKeeper on 7 February 2021 and that measurement methodologies for all five measures would be completed for the JobKeeper project closure report.⁶⁷

4.23 The five performance measures were first reported on, in draft, to the JobMaker Hiring Credit Steering Committee on 27 July 2021, four months after the JobKeeper scheme had closed.⁶⁸ The draft paper provided a provisional assessment of performance against the five measures, which was positive (Box 3). The five performance measures were not included in the project closure report provided to the JobKeeper and JobMaker Hiring Credit Program Board on 10 June 2021.

Source: ANAO, based on the ATO’s records.

Has the ATO reported on the scheme in a timely and informative manner?

4.24 The legislative framework for the JobKeeper scheme does not impose external reporting obligations on the ATO or other entities. The nature and extent of reporting were shaped by arrangements reached between the ATO and Treasury and by existing scrutiny and reporting mechanisms for Australian Government entities, such as parliamentary committees and annual reporting requirements.

Reporting to government entities

4.25 For the JobKeeper scheme, the ATO’s principal reporting relationship was with Treasury, which was responsible for JobKeeper policy. In turn, Treasury’s policy role included reporting JobKeeper data to other government entities. This included providing JobKeeper data to the Department of the Prime Minister and Cabinet (PM&C) for forwarding to state and territory agencies.⁶⁹

4.26 The ATO’s and Treasury’s records provide evidence of regular engagement between the entities on reporting. On 30 April 2020 the ATO provided Treasury with a ‘mock up’ of its proposed reporting. Treasury’s response on 5 May 2020 provided a list of data items to be provided by the ATO on a daily, weekly or monthly basis. Treasury advised the ATO that the data items encompassed requests from the Treasurer’s office, PM&C and the states and territories.

4.27 In August 2021, Treasury advised the ANAO that there was an agreement made between Treasury, the ATO and the Treasurer’s office that reporting on JobKeeper would be daily from the commencement of the scheme in the form of a ‘daily dashboard’. Treasury advised the ANAO that it also made additional data requests to the ATO outside of the daily dashboard.

4.28 In December 2020 the ATO and Treasury formalised a ‘reporting schedule’ for the scheme during the extension period. The schedule set out the type of JobKeeper reports to be produced by the ATO, the frequency of those reports, and a distribution list of Treasury staff who would receive the reports.

• Daily JK2 — a daily dashboard of high-level statistics for the extension period.
• Extended JK2 — a weekly report of key data (such as applications, payments, and individuals by entity type, jurisdiction, market segment and industry).
• Combined JK1 and JK2 — a weekly report that provided a comparative breakdown of key data for both the original and extension periods of the JobKeeper scheme. This report provided cumulative totals across both periods.
• JK1 — a weekly report that provided a breakdown of key data for the original period.

4.29 The ANAO selected 12 reports from the two periods and asked the ATO to evidence the reporting to Treasury.⁷⁰ The ATO either provided evidence of the reports being provided to Treasury within the agreed timeframe or satisfactory explanations for any changes to the agreed timeframe.

4.30 Reporting arrangements for the JobKeeper scheme were also considered through the JobKeeper Data and Reporting Working Group — one of four Working Groups established under an interdepartmental committee for JobKeeper.⁷¹ The meeting minutes and records kept for the Data and Reporting Working Group show that a range of data and reporting issues were discussed, and that the members of the Group received access to a range of JobKeeper data. This included copies of the data releases prepared for state and territory agencies.

Data sharing arrangements

4.31 The ATO received various requests for JobKeeper data from Commonwealth entities and other external parties such as state government entities. Such requests could be made directly to the ATO or through Treasury or the Data and Reporting Working Group.

4.32 Division 355 of Schedule 1 of the Taxation Administration Act 1953 provides that the disclosure of information about the tax affairs of a particular entity is prohibited, except in certain specified circumstances. On 3 September 2020, the Taxation Administration Act 1953 was amended to add an exception (under the category of ‘disclosure for other government purposes’) for information that relates to the JobKeeper scheme.⁷²

4.33 The ATO maintained a register of external data requests relating to the JobKeeper scheme (or other COVID-19 measures being managed by the ATO). Among the details recorded was whether the request was for ‘unit level’ or ‘aggregate’ (de-identified) data. As of 28 February 2021, the register listed 72 data requests.

4.34 The ATO developed procedural documentation to administer its data sharing arrangements with government agencies. The ATO’s process involved four main steps — the first step of which was to determine whether the data sharing was lawful (Figure 4.3).

Figure 4.3: The ATO’s four-step data sharing process with government entities

 

 

Source: Reproduced from the ATO’s records.

4.35 To test whether the ATO addressed Step 1, the ANAO selected six (of 28) data sharing arrangements that were categorised by the ATO as ‘outbound requests’ that had been finalised. The sample comprised three requests for aggregate level data and three requests for unit level data, and were made by Commonwealth, state, and local government bodies, and one non-government organisation. The ANAO found that the data was shared in a manner that was consistent with what was recorded in the data sharing register. The legal basis for sharing the data was assessed by ATO staff beforehand.⁷³

4.36 The ATO provided JobKeeper data to the Australian Bureau of Statistics (ABS), for the ABS’s ‘DataLab’.⁷⁴ The ABS facilitates access to the DataLab for approved researchers and other parties. Provision of this data supports the undertaking in Treasury’s July 2020 report to make de-identified program administration data available for research purposes (paragraph 4.41).

Public reporting

4.37 The ATO’s public reporting on the JobKeeper scheme has been provided through its website and submissions to parliamentary committees, including the Senate Select Committee on COVID-19 (COVID-19 Committee), which was established to inquire and report on the Australian Government’s response to the COVID-19 pandemic. The ATO appeared before the COVID-19 Committee on six occasions between 7 May 2020 and 11 February 2021. The ATO also provided details on the JobKeeper scheme to the Senate Economics Legislation Committee.

4.38 The ATO prepared a ‘live’ communications document, called a ‘Strategic Communications Overview’, for the original and extension periods. Both documents outlined ‘key messages’ for the affected audiences (such as employers, employees, tax professionals, sole traders and digital service providers) and listed communication activities in different channels (webinars, podcasts, social media, newsletters for industry sectors and web content in different languages).

4.39 The ATO’s 2019–20 Annual Report, which covered the first three months of the JobKeeper scheme, included high level information on the scheme’s operation such as the number of enrolments on the first day, the amount of payments made and the number individuals covered by these payments. In the Annual Performance Statement section of the Annual Report, JobKeeper (along with the Cash Flow Boost payment) is listed under the Administered Program ‘1.19 Economic Response to the Coronavirus’. The 2020–21 Annual Report, which covered the remaining period of the JobKeeper scheme, provided various details including: total payments ($88.8 billion); total overpayments ($470 million); and the total of JobKeeper payments stopped, recovered or prevented through compliance activities ($1.1 billion).

4.40 The ATO is providing input to Treasury’s evaluation of the JobKeeper policy and program, which Treasury intends to be a public report.

Has an evaluation strategy for the scheme been developed?

4.41 Treasury undertook a three-month review of the JobKeeper Payment in May and June 2020 and reported its findings on 21 July 2020.⁷⁵ The review found that JobKeeper had met its three objectives — supporting business and job survival, preserving the employment relationship, and providing needed income support.⁷⁶ The report stated that:

Beyond this review, it is imperative that a program of this magnitude and novelty should be studied and evaluated very closely. The Treasury, the Australian Taxation Office and the Australian Bureau of Statistics will work with the academic community and others to make de-identified program administration data available for research purposes. An independent evaluation should be conducted at the completion of the program.⁷⁷

4.42 Treasury published a further review of JobKeeper on 11 October 2021: Insights from the first six months of JobKeeper.⁷⁸ The review noted that an evaluation would be completed in 2022 to provide a detailed assessment of JobKeeper against its policy objectives and lessons learned for future policy makers.

Evaluation planning

4.43 Treasury prepared a draft work plan for the evaluation in March 2021. The draft work plan and a related briefing note indicated that the evaluation would be conducted by Treasury’s Labour Market Policy Division, which was responsible for JobKeeper policy. Treasury’s planning was based on a nine-month delivery timeframe for the final evaluation report.

4.44 Treasury’s planning included engaging an external provider with experience in conducting program performance evaluations. Treasury advised the ANAO that a desktop review and research process was undertaken to identify suitable providers within the ‘Strategy and Policy’ category of the Australian Government’s Digital Marketplace in late March 2021. Six providers were requested to provide a proposal and quote, four of whom responded.

4.45 Following evaluation of the four proposals, Treasury selected Value Management Consulting Pty Ltd. A ‘work order’ was executed between the two parties on 11 May 2021.⁷⁹ The requirements in the work order included the preparation of:

• an agreed policy and program evaluation framework for Treasury to use for JobKeeper, and also for potential re-use for other existing and future programs; and
• a strategy for delivery of the JobKeeper policy and program performance evaluation.

Evaluation framework and governance

4.46 A draft Evaluation Framework was discussed at a meeting of Treasury’s Executive Board on 25 August 2021. The draft document detailed the evaluation purpose, scope, timing, governance, evaluation questions, methodology and stakeholders.

4.47 The minutes of the Executive Board meeting included the statement that:

The Board discussed the merits of the evaluation being governed within Labour Market Policy Division, another Division within the Department or an independent person. The Board agreed to further consider the approach to governance and have a further conversation about the approach and return for a further discussion.

4.48 The Executive Board asked the head of the Labour Market Policy Division, in consultation with the head of Treasury’s Strategic Coordination and Communications Division, to review the proposed JobKeeper evaluation governance approach and return to the Executive Board.

4.49 Treasury provided a ministerial submission to the Treasurer on 16 November 2021, recommending that the Treasurer agree to developing terms of reference for the Productivity Commission to conduct an inquiry into the design and economic outcomes of JobKeeper. The submission noted that the inquiry was intended to commence in February 2022 and be completed by the end of 2022. The submission also noted that should the Treasurer not support an inquiry by the Productivity Commission, Treasury would establish departmental governance arrangements to complete an evaluation by the end of 2022. On 20 December 2021, Treasury advised the Treasurer’s office that, following a discussion by Treasury’s Executive Board, consideration of the ministerial submission was no longer required.

4.50 A briefing paper was provided to the Executive Board ahead of its meeting on 14 December 2021, which outlined proposed departmental governance arrangements in the event that the Treasurer did not support an inquiry by the Productivity Commission (Figure 4.4).

Figure 4.4: Treasury’s proposed governance arrangements for the JobKeeper evaluation (10 December 2021)

Source: Reproduced from Treasury’s records.

4.51 The briefing paper recommended that, in order to provide a whole-of-department perspective to the JobKeeper evaluation and enhance separation of responsibilities for program delivery and evaluation, the Deputy Secretary of Treasury’s Corporate and Foreign Investment Group assume the role of project sponsor with accountability for evaluation oversight. The briefing paper also recommended that the First Assistant Secretary of Treasury’s Strategic Coordination and Communications Division (SCCD) lead the evaluation process. The briefing paper noted that there would not be spare capacity at Senior Executive Service level within SCCD until mid-2022.

4.52 On 14 December 2021, Treasury’s Executive Board determined that the JobKeeper evaluation would commence in early June 2022 and be undertaken within Treasury in line with the arrangements recommended in the briefing paper. The Executive Board also approved the JobKeeper Evaluation Framework (which had been revised since the Board’s previous meeting). The agreed purpose and objectives of the evaluation are set out in Box 4.

Source: ANAO, based on Treasury’s records.

4.53 The Evaluation Framework indicates that the final evaluation report containing analysis of outcomes for public release, updates to early insights analysis and broader analysis, will be finalised by late 2022. Treasury’s documentation indicates that JobKeeper data is expected to be included in the ATO’s 2022 release of Taxation Statistics. The ATO is also expected to provide updated JobKeeper microdata to the ABS (for uploading to the DataLab) in early 2022.

The ATO’s ‘closure report’ and other internal reports on the JobKeeper scheme

4.54 The preparation of a closure report is a requirement under the ATO’s project management framework. A project closure report records the formal termination of a project and includes information on a range of matters including budget and funding, project risks and issues, records management, law conformance, and lessons learned. The ATO completed a project closure report on its administration of the JobKeeper scheme as well as preparing other documents detailing its experience and insights in administering the scheme.

4.55 The closure report for the JobKeeper scheme was provided to the final meeting of the JobKeeper and JobMaker Hiring Credit Program Board on 10 June 2021. The closure report stated that the project was delivered within allocated budget for the extension period while the original period was absorbed within the existing ATO budget; mandated milestones were delivered on time as were changes announced by the Australian Government; scope was modified and extended by the Australian Government during the project and changes were delivered on time; and risk and issues were managed within tolerance. The closure report documented lessons learned across different aspects of the project, including what the ATO would do differently next time. The report also set out the challenges and practical solutions and noted that the JobKeeper scheme was unlike any other measure delivered by the ATO.

4.56 The Program Board was advised that residual risks were being handed to business-as-usual arrangements, which included ongoing management of JobKeeper risks within the ATO’s Superannuation and Employer Obligations business line.

4.57 For the last Program Board meeting on 10 June 2021, the ATO also prepared a ‘JobKeeper Experience’ report. This report documented the ATO’s experience and insights from administering the JobKeeper scheme, with the aim of informing future programs across different aspects of scheme administration such as design, governance, leadership and integrity. The ATO’s key observations from this report are set out in Box 5.

Source: ANAO, based on the ATO’s records.

Appendix 1 Entity responses

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 2021–22 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. The ANAO did not observe any changes in the ATO’s administration of the JobKeeper scheme. The scheme closed in March 2021, which was early in the fieldwork stage of the audit.

Appendix 3 ATO analysis of JobKeeper data

1. The ANAO has reproduced the following graphs (and notes) from data presented in the ATO’s ‘Combined JobKeeper 1.0 and 2.0 Payment Report’. The data presented is cumulative across both periods of the JobKeeper scheme as of 15 August 2021.

Figure A.1: Applications processed and net payments by entity type

Figure A.2: Applications processed and net payments by market segment

Figure A.3: Applications processed and net payments by state/territory

Figure A.4: Applications processed and net payments by industry

Notes: Data are for unique entities (unique Australian Business Numbers, ABNs) with a processed application for JobKeeper. Entities include employers and eligible business participants.

Net payments is the payment disbursements after repayments from entities. Total for all periods may not balance due to rounding.

State/territory is taken from the location of the entity’s address for the registered ABN. Not all employees are based in the state/territory where the ABN is registered.

Industry is based on ANZSIC codes registered with the Australian Business Register.

Appendix 4 Overview of the decline in turnover tests for the JobKeeper scheme

Note a: Explanatory Statement, Coronavirus Economic Response Package (Payments and Benefits) Rules 2020 (Cth), p.6.

Note b: An eighth alternative decline in turnover test was introduced, by legislative instrument, on 10 October 2020 covering businesses that had temporarily ceased trading during the relevant comparison period. In addition, some changes were made to the existing seven alternative decline in turnover tests from 27 September 2020 onwards.

Note c: Explanatory Statement, Coronavirus Economic Response Package (Payments and Benefits) Rules 2020 (Cth), p.9.

Note d: For the original period of the JobKeeper scheme the test period was a month or quarter between 30 March 2020 and 1 October 2020. For the JobKeeper extension period the test period was the quarters ending 30 September 2020 and 31 December 2020.

Note e: Explanatory Statement, Coronavirus Economic Response Package (Payments and Benefits) Amendment Rules (No. 2) 2020 (Cth), p.4.

Note f: Explanatory Statement, Coronavirus Economic Response Package (Payments and Benefits) Amendment Rules (No. 8) 2020 (Cth), p.1.

Source: ANAO, based on public records and the ATO’s records.

Appendix 5 Examples of the online forms used for the original and extension periods of the JobKeeper scheme

Figure A.5: Example eligibility information required in the online form for the original JobKeeper period^a

Note a: Figure A.5 reproduces how the eligibility information in the enrolment form was presented for an employer. Responses to the first two questions would modify the turnover threshold (to either 15%, 30% or 50% as required) for the final two questions.

Source: Reproduced from the ATO’s original JobKeeper scheme online enrolment form.

Figure A.6: Example turnover test information required in the online form for the JobKeeper extension period

Source: Reproduced from the JobKeeper extension ‘check decline in turnover’ stage of the online application process, as released by the ATO on 1 October 2020.

Appendix 6 Summary of the ATO’s assurance approach for risks listed in the JobKeeper Compliance Strategy (3 July 2020)

Appendix 7 ANAO sampling methodology for testing ATO compliance measures

1. The ANAO’s methodology for testing the ATO’s implementation of compliance measures for the JobKeeper scheme involved a two-stage sampling process.

Stage 1: Selecting compliance measures for testing

2. The ATO prepared 15 risk treatment plans for the JobKeeper scheme. The risk treatment plans contained many compliance measures, which represented the ATO’s intentions about how it would mitigate the payment risks identified in the risk treatment plans.

3. The ANAO used a targeted sampling methodology to select which compliance measures to test. Twenty-two compliance measures were selected. In addition, the ANAO examined the ATO’s arrangements for sharing data with Services Australia, as outlined in the JobKeeper/JobSeeker risk treatment plan.

4. The selection was principally targeted at compliance measures that the ATO had directed at JobKeeper populations deemed by the ATO to be higher risk, and which typically involved direct engagement with JobKeeper recipients.

5. Each compliance measure comprised individual compliance cases. For each of the 22 compliance measures selected for testing, the ATO provided lists of individual compliance cases which the ATO claimed to have undertaken.

Stage 2: Selecting individual compliance cases for testing

6. There was a total of 41,031 compliance cases across the 22 selected compliance measures.

7. The eight compliance measures set out in Table A.1 each comprised fewer than 100 cases. Census testing was performed.

Table A.1: Compliance measures subject to census testing

8. For the 14 compliance measures which each comprised 100 or more individual compliance cases, random sampling was used to select from the populations of compliance cases.

9. Where random sampling was used, the ANAO sought to achieve a margin of error of plus or minus five per cent at a 90 per cent confidence level.

10. The sample sizes selected depended on the expected exception rate set out in Table A.2. Given that the compliance measures selected for testing were directed by the ATO at higher risk populations, the ANAO expected that those measures would be implemented by the ATO with few exceptions. An ‘exception’ was where the ATO’s documentation did not evidence that the intended compliance action was undertaken. Whether an estimated exception rate of two per cent or four per cent was selected depended on the ANAO’s professional judgment as to the complexity of the compliance measure.

Table A.2: Compliance measures subject to random sampling

11. The observed exception rate exceeded the expected exception rate for some of the compliance measures subject to random sampling. An increase in sample size for the affected compliance measures would have been necessary to achieve a margin of error of plus or minus five per cent at a 90 per cent confidence level.

12. The ANAO did not sample additional cases for compliance measures where observed exception rates exceeded expected exception rates. The additional testing would not have materially affected the audit conclusion that the ATO did not evidence the full implementation of the selected compliance measures.

Appendix 8 The ATO’s Practice Note explanations of ‘honest mistake’ and ‘financial benefit’

Source: ANAO, based on the ATO’s records.