THE GAMA FOUNDATION AND THE CONTROLLER AND AUDITOR- GENERAL OF NEW ZEALAND
[2024] NZHC 2735
•20 September 2024
IN THE HIGH COURT OF NEW ZEALAND WELLINGTON REGISTRY
I TE KŌTI MATUA O AOTEAROA TE WHANGANUI-A-TARA ROHE
CIV-2022-485-789
[2024] NZHC 2735
UNDER the Judicial Review Procedure Act 2016 IN THE MATTER
of an application for judicial review of decisions of the Controller and Auditor-General
BETWEEN
THE GAMA FOUNDATION
Applicant
AND
THE CONTROLLER AND AUDITOR- GENERAL OF NEW ZEALAND
Defendant
Hearing: 11–12 March 2024 Counsel:
T Mijatov and L E Kenner for Applicant
K McDonald KC and N Whittington for Respondent
Judgment:
20 September 2024
JUDGMENT OF ISAC J
[Application for judicial review]
THE GAMA FOUNDATION v THE CONTROLLER AND AUDITOR-GENERAL OF NZ [2024] NZHC 2735
[20 September 2024]
TABLE OF CONTENTS Para no.
Introduction [1]
The pleadings, evidence, and procedural issues [9]
The Auditor-General and his statutory functions [14]
The wage subsidy scheme [23]
Managing a high trust payment model [28]
The Deloitte review [31]
The Auditor-General’s involvement with the wage subsidy scheme [33]
Audit NZ’s financial audit for the year ended 30 June 2020 [33]
The Auditor-General’s 2021 performance audit [41] The Auditor-General’s draft performance audit recommendation [48] The draft performance audit recommendation is changed [54]
MSD’s response to Recommendation 4 [57]
Continued reporting [59]
The Auditor-General’s continued reporting to the Finance and
Expenditure Committee [59]
Audit NZ’s continued reporting in 2022 [66]
First issue: does parliamentary privilege attach to the documents
in issue? [69]
A preliminary issue — are recommendations by public bodies
capable of review? [72]
The Auditor-General’s argument on parliamentary privilege [77] Parliamentary privilege and the Parliamentary Privilege Act 2014 [85] Consideration [99]
The documents in this case in light of the privilege [126]
Conclusion on parliamentary privilege [137]
Second issue: did the Auditor-General breach his statutory duty of
independence or otherwise fetter his discretion? [140] The duty of independence and applicable audit standards [142] A lack of evidence to support the amendment to Recommendation 4? [149] Has the Auditor-General unlawfully fettered his discretion? [170]
Third issue: did the Auditor-General act unreasonably and has
there been an unreasonable delay? [176]
Relief [178]
Conclusion and result [179]
Introduction
[1] As part of the Government’s economic response to the COVID-19 pandemic, Cabinet agreed to implement a wage subsidy scheme for a 12-week period, beginning on 16 March 2020. Te Manatū Whakahiato Ora | The Ministry of Social Development (MSD) was appointed to lead the implementation of the scheme, which operated on a “high trust” model, whereby MSD approved payments based only on a declaration by applicants that they were eligible. Various iterations of the scheme eventually led to approximately $18.8 billion in wage subsidy payments being made to businesses during the outbreaks of COVID-19 in 2020 and 2021.
[2] The applicant for review in this case, the Gama Foundation (Gama), is a charitable trust and philanthropic organisation concerned with Government decision making, wealth inequality in New Zealand and responsible spending of public money. Since May 2020, Gama has been examining the Government’s implementation of the COVID-19 wage subsidy scheme to determine whether it was lawful.
[3] In 2021, Gama commenced judicial review proceedings against MSD, alleging that MSD prosecuted only ineligible recipients in very limited circumstances despite public information suggesting the scheme had been abused. In November 2023, McQueen J dismissed Gama’s challenge, finding no such “unwritten unlawful practice” existed.1
[4] In these proceedings, Gama seeks judicial review of decisions of the Tumuaki o te Mana Arotake | the Controller and Auditor-General of New Zealand (the Auditor-General) who has, as part of his statutory function, monitored and reported on MSD’s management of the wage subsidy scheme over the last four years.
[5] The focus of Gama’s claim is a draft recommendation by the Auditor-General that MSD seek written confirmation from all wage subsidy recipients of their eligibility to receive the payments. Following feedback from MSD about the proposed recommendation, the Auditor-General eventually recommended these enquiries
1 The Gama Foundation v Chief Executive of the Ministry of Social Development [2023] NZHC 3098 at [102]–[122].
“could be targeted towards larger or risk-indicated applicants”. Gama argues that the Auditor-General acted unlawfully when he altered the draft recommendation. In particular, Gama says there was no evidence to support the amendment and, in changing the recommendation, the Auditor-General breached his obligation to act independently. It also contends that the decision was unreasonable in a public law sense.
[6] Against that backdrop, three issues arise for determination. The first is whether parliamentary privilege precludes the Court from considering the lawfulness of the Auditor-General’s reports and recommendations. Parliament and the Finance and Expenditure Committee have reviewed the Auditor-General’s work on the wage subsidy scheme, and the Committee has written its own report on it. The Auditor-General says to question the validity of its recommendations to Parliament is to question proceedings in Parliament. In response, Gama says that the discharge of the Auditor-General’s statutory functions, including auditing MSD’s management of the wage subsidy scheme, does not infringe parliamentary privilege.
[7] Second, Gama argues the Auditor-General’s recommendation was unlawful because he did not comply with a statutory duty of independence and otherwise fettered his discretion. The recommendation was unsupported by evidence and the result of negative feedback from MSD to the draft report. In response, the Auditor-General argues he acted independently and was satisfied the final recommendation was appropriate. The recommendation was supported by evidence and was not simply based on MSD’s opinion.
[8] The final issue is whether the Auditor-General acted unreasonably in failing to recommend that MSD write to all Wage Subsidy payment recipients, and whether his continued failure to make such a recommendation to MSD constitutes an unreasonable delay. Gama says that such a recommendation was the only one open to the Auditor-General. The Auditor-General disagrees. He submits the final recommendation was grounded in the available evidence. Gama’s issue is one of opinion, not one that gives rise to a reviewable error.
The pleadings, evidence, and procedural issues
[9] Gama advanced four grounds of review, but at the hearing its primary focus was on one of them: that the Auditor-General’s decision to change his draft recommendation—that MSD write to all wage subsidy recipients—breached a duty of independence. This claim was closely connected to a second ground of review, that the Auditor-General’s “removal” of the recommendation from three reports meant in substance he acted at the direction of MSD and thereby unlawfully fettered his discretion. The remaining grounds of review—the first and third—were also closely related to each other: that the Auditor-General’s final recommendation was unreasonable, and his delay in making an appropriate recommendation was also unreasonable. I address these related claims together later in this judgment.
[10] In addition, the Auditor-General raised several procedural and jurisdictional points:
(a)First, objection is taken to a “supplementary bundle” of documents provided by the applicant. While the papers were made available by the Auditor-General in discovery, they are not part of the evidence filed by Gama or the Auditor-General. The respondent says that many of the documents lack any context, and Gama’s failure to adduce them in its own evidence has deprived him of the ability to provide that context in evidence in reply. These documents should not be admitted accordingly.
(b)Second, the Auditor-General, as an Officer of Parliament, argues the principle of non-interference enshrined in the Parliamentary Privilege Act 2014 prevents the Court (and Gama) from calling into question the contents of the Auditor-General’s reports and his recommendations.
(c)Finally, the respondent argues it is doubtful audit recommendations are capable of being the subject of judicial review, as they do not amount to a determination, decision or order in the conventional sense.
[11] For the reasons that follow, I have upheld the Auditor-General’s claim to parliamentary privilege. However, I have not accepted his argument that recommendations in his reports are not capable of review. I have also considered Gama’s case at its highest, which includes its reliance on the documents contained in the supplementary bundle documents subject to objection.2 Had I found that those documents were material to the outcome in a way adverse to the Auditor-General’s position, I would have excluded them as a matter of fairness.
[12] Evidence was provided by six deponents supported by an extensive documentary record. For the applicant, Mr Grant Nelson, one of Gama’s trustees, addressed Gama’s concern that “billions of dollars” of wage subsidy payments were made to businesses that were operating profitably and were not entitled to the payment. Mr Nelson says that the subsidy triggered “a house and investment asset price bubble” which greatly increased wealth inequality within the community, adversely affecting poorer members and exacerbating child poverty. Mr Patrick Hoy also provided expert evidence for the applicant. For 25 years he was a senior performance auditor employed in the Audit Office, which later became the Office of the Auditor-General. Mr Hoy provided a detailed analysis of the process and evidence leading to the final recommendation, and what he identified as its deficiencies.
[13] The evidence for the Auditor-General was provided by four staff members who held senior roles in the teams responsible for undertaking the relevant performance and financial audits. Their evidence was directed to the process followed during the conduct of the relevant investigations, the applicable audit standards, and the evidence provided by MSD which supported the ultimate change to the recommendation concerning MSD’s approach to the wage subsidy recipients.
2 I note that it is not clear how the supplementary material was selected from the total discovered documents. For example, a document titled “Record of Evidence” was provided in the supplementary material. It appeared to be based on other analytical work material of the Auditor-General that was likely to have also been discovered but was not provided in evidence. This suggests some documents seem to be included in the supplementary materials but not others which may have provided further context.
The Auditor-General and his statutory functions
[14] The Auditor-General is an “Officer of Parliament”.3 Unlike other officers of Parliament, the Auditor-General exercises Parliament’s “control” functions as a check on actions of executive government.4 Their functions, duties and powers are prescribed by the Public Audit Act 2001. As an Officer of Parliament, the Auditor-General promotes executive accountability to Parliament,5 and to the public through Parliament.
[15] Auditors-General are appointed by the House of Representatives and swear an oath of office administered by the Speaker.6 They are not in the service of the Crown.7
[16] The Auditor-General acts as a public “watchdog”, ensuring good stewardship of public money in New Zealand. In his auditing function, the Auditor-General carries out annual financial report audits8 (often called annual audits) for all public organisations.9 Annual audits typically focus on the financial statements, accounts, and other information that the public organisation is required to have audited. Annual audits have two outputs: an audit report, and a management letter, which contain findings and recommendations from the auditor to the governance body and management of the audited entity. The audit report for a ministry of the Crown is included in the ministry’s annual report presented to the House of Representatives.
[17] Under s 16 of the Public Audit Act, the Auditor-General can also carry out performance audits at any time. Performance audits examine whether a public entity is carrying out its activities efficiently and effectively, complying with statutory obligations, and whether it is doing so with the appropriate attention to probity,
3 Public Finance Act 1989, s 2(1) definition of “Office of Parliament”. Before the Public Audit Act 2001, the Auditor-General was previously an officer of the Crown appointed on the recommendation of the government.
4 Philip Joseph Joseph on Constitutional and Administrative Law (5th ed, Thomson Reuters, Wellington, 2021) at [13.1] [Joseph on Constitutional Law].
5 Public Audit Act, ss 20, 36 and 37.
6 Public Audit Act, sch 3 cl 3.
7 Public Audit Act, sch 3 cl 6. Reinforcing this, the Office’s funding is the responsibility of the bipartisan Officers of Parliament Committee, chaired by the Speaker. See Joseph on Constitutional Law, above n 4, at [13.2]–[13.3].
8 Public Audit Act, s 15(1).
9 Public Audit Act, s 14.
financial prudence, and waste.10 In the resulting performance reports, the Auditor-General reports on its findings, makes recommendations, and provides guidance on best practice.11 It follows that while investigation and reporting under s 16 is referred to as an “audit”, the nature of the inquiry and the report itself are quite different from a financial audit.
[18] Performance reports are usually presented to the House of Representatives under s 20 of the Act. Before these reports are presented to the House, the Auditor-General presents a draft report to the auditee. The auditee is then typically given two weeks to provide comments, and the audit findings are checked for factual accuracy.
[19] While in undertaking a performance audit the Auditor-General may make recommendations to a ministry or public organisation, he is unable to direct how the audited entity is to act, and his recommendations are not binding. Ms Leeanne McAviney, an Assistant Auditor-General responsible for performance audits over the relevant period, gave evidence that:
The Auditor-General is unable to direct any public organisation to act in a certain way or to make different decisions to the ones it has made. Rather, the Auditor-General’s role is to report what he or she sees, make recommendations, and provide guidance about good practice so that public organisations can make improvements. Whether a public organisation implements a recommendation, follows guidance, or takes any action in response are decisions for it to make.
[20] However, the potential for the Auditor-General’s recommendations (and whether there has been implementation of them) to face subsequent scrutiny by Parliament mean that the Auditor-General’s recommendations generally carry weight, in much the same way as those of other Officers of Parliament, such as the Ombudsman.
10 Public Audit Act, s 16(1).
11 Auditors-General complete annual financial audits under their “controller” function set out in the Public Finance Act, ensuring that the spending of government departments is in keeping with the relevant appropriations by Parliament. Audits of a public entities’ performance in meeting its statutory obligations is done by the respondent under his “Auditor-General” function.
[21] The work of the Auditor-General is performed by two business units, Audit New Zealand (Audit NZ) and the Office of the Auditor-General, being the financial audit and performance audit teams respectively. The financial auditing of public sector entities is carried out by an “appointed auditor” from Audit NZ (alongside private auditing firms).
[22] Pursuant to s 18 of the Public Audit Act, the Auditor-General can also inquire into any matter concerning a public entity’s use of its resources. This can be done on request or on its own initiative.12 A report similar to that which would be provided after a performance audit is drafted and provided to the relevant public entity. Under s 21 of the Act, the Auditor-General is empowered to report to a Minister, a committee of the House of Representatives, a public entity, or indeed any person on any matter arising out of the performance and exercise of its function.
The wage subsidy scheme
[23] On 16 March 2020, Cabinet approved the first wage subsidy scheme.13 Under the scheme, applicants who met the eligibility criteria were paid $585.80 per week per full-time employee, and $350 per week per part-time employee.14 Cabinet specified several qualifying criteria for employers seeking payments. The principal criterion, however, was that an employer was required to have suffered a revenue loss due to the COVID-19 outbreak of at least 30 per cent.
[24] The key aim of the scheme was to minimise the adverse economic impact of the lockdowns by ensuring businesses continued to be able to pay their employees. This aim would be frustrated if payments could not be made quickly. When Cabinet considered the establishment of the scheme, it considered this question. Prioritising speed over eligibility checks, Cabinet adopted what became known as a “high trust model”. In the accompanying minute, Cabinet recorded that:15
12 Public Audit Act, s 18(1).
13 Further detailed background is also set out in the Judgment of McQueen J, above n 1, at [7]–[16].
14 Payments were initially capped at $150,000 per business, but this was removed by Cabinet on 23 March 2020.
15 Cabinet Minute of Decision “COVID-19: Overview of the Government’s Response: Economic Package” (16 March 2020) CAB-20-MIN-0108 at [47.3] (redaction in original).
MSD [Redacted] will have the ability to later audit the application, verify information with other agencies and refer possible instances of fraud for investigation.
[25] In keeping with the aim of ensuring money continued to flow to businesses and employees, payments were made to employers based on a declaration by applicants. The eligibility criteria were amended over time by MSD but, generally, applicants were required to declare that:
(a)the applicant’s business was registered and operating in New Zealand and the employees named in the application were legally employed in New Zealand;
(b)the business had experienced a minimum decline (30 and later 40 per cent) in actual or predicted revenue attributable to COVID-19 over the period of any month from January 2020 to the end of the scheme when compared to the same month the previous year;
(c)the recipient had taken active steps to mitigate the financial impact of COVID-19 on their business activities, including engaging with their bank and (from 27 March 2020) drawing on cash reserves;
(d)if anything affecting their eligibility changed, the recipient would inform MSD; and
(e)all information provided was true and correct.
[26] Applicants were required to certify that they would repay the subsidy if they were not entitled to it or if they stopped being entitled to it.16 Applicants also acknowledged that they could be subject to civil proceedings for the recovery of any amount they receive that they were not entitled to, as well as prosecution for offences
16 This was amended from 21 August 2020, to require employees to declare they would repay any subsidy if they were not (or stopped being) entitled to the subsidy, including where any predicted decline in revenue over the relevant period was not realised. This was further amended from 4 March 2021 to require applicants to declare they would repay any subsidy in circumstances where they predicted they would meet the revenue decline test but did not in fact.
under the Crimes Act 1961 (including for the provision of false or misleading information).
[27] The Government publicly announced the start of the wage subsidy scheme on 17 March 2020. MSD began accepting applications the same day. The subsidies were available to all businesses, including self-employed persons, contractors, sole traders, registered charities, incorporated societies, and post-settlement governance entities. Between 2020 to 2021, the scheme was implemented over five periods.17
Managing a high trust payment model
[28] MSD paid very substantial sums of public money to applicants at a break-neck speed, with the average payment taking just three-and-a-half days after application.18 As a high trust model requiring applicants to declare their own eligibility for payments, the scheme was considered at the outset vulnerable to both fraud and error.19 Within a month of the first subsidy payments, there were public reports of applicants abusing the scheme.
[29] To manage these risks, from 30 March 2020, MSD implemented a system of desk-based pre-payment review for applicants with 80 or more employees. These reviews generally consisted of phone calls by MSD officials to employers, designed to advise recipients of their obligations before payments were made. No evidence of eligibility was required in these pre-payment reviews. Until automation of the process on 11 June 2020, MSD staff were also required to manually confirm an applicant’s basic details with Inland Revenue.
[30] MSD also carried out 6,639 post-payment reviews. No substantiating evidence was collected as part of this process either, which involved another desk-based review confirming the legitimacy of the business and confirmation that the subsidy was paid
17 It is not necessary to go into detail regarding these stages, however, they were: the Original Wage Subsidy (17 March 2020 to 9 June 2020); the Wage Subsidy Extension (10 June 2020 to 1 September 2020); the Resurgence Wage Subsidy (21 August 2020 to 3 September 2020); the Wage Subsidy March 2021 (4 March 2021 to 21 March 2021); and the Wage Subsidies August 2021 (20 August 2021 to 9 December 2021).
18 At least as at the date the Auditor-General issued his performance audit report in May 2021.
19 Tumuaki o te Mana Arotake | Controller and Auditor-General Management of the Wage Subsidy Scheme (Office of the Auditor-General, May 2021) at [1.37] [Performance audit report].
correctly. At 30 October 2020, refunds were required in 15.4 per cent of all post-payment checks.
The Deloitte review
[31] Of some relevance in these proceedings is a report commissioned by MSD. Deloitte was engaged to carry out an independent “integrity focussed risk assessment” of the wage subsidy scheme, encompassing fraud, corruption, waste and error.
[32] By the end of April 2020, Deloitte had provided an interim report with 14 recommendations, nine of which MSD agreed to implement. In its final report issued in July 2020, Deloitte’s first recommendation was that MSD “quickly consider how best to communicate with [wage subsidy scheme] recipients to remind them of their obligations and to maximise the potential of early voluntary repayments”. Deloitte said this could either be a blanket communication or a more “targeted approach”, involving MSD profiling applications to identify sub-groups that presented a voluntary repayment opportunity. It also recommended that this action should be taken quickly.
The Auditor-General’s involvement with the wage subsidy scheme
Audit NZ’s financial audit for the year ended 30 June 2020
[33] MSD’s management of the wage subsidy scheme became a key focus for Audit NZ’s appointed auditor, Mr John Whittal. In a draft interim financial audit report for the year ended 30 June 2020, Mr Whittal highlighted MSD’s lack of follow-up communication with wage subsidy recipients after application and payment. Mr Whittal then went on to recommend that positive confirmation of eligibility should be sought from payment recipients, saying:
The Ministry should consider seeking a positive confirmation from recipients that they have assessed their situation and confirm that they have subsequently met the requirements of the scheme.
[34] Two points need to be made about the draft recommendation. First, while it might be implicit, the recommendation did not explicitly require, as Gama contends, that MSD should write to all wage subsidy recipients. More importantly, the draft
recommendation merely suggested MSD “should consider” seeking positive confirmation. There is no suggestion in the language used that such enquiries were in fact necessary, or ought to occur.
[35] The draft financial audit report was sent to MSD for consideration and comment. On 5 August 2020, MSD’s General Manager of Risk and Assurance provided the Auditor-General’s office with the Ministry’s response. MSD said its implementation of the scheme had gone “above and beyond” the eligibility criteria and checks required by Cabinet. For instance, it had not been required to but nevertheless carried out pre-payment checks against Inland Revenue records relating to every subsidy-claiming employer. MSD also noted that every employer who received the wage subsidy was sent a letter regarding their need for the wage subsidy extension. That letter “reminded employers of their status of entitlement”. MSD also noted it had commissioned the Deloitte report on gaps in its integrity processes, and that it had “implemented several enhancements” in response. Finally, MSD had also:
… modified our audits depending on the risk profile of the client groups, and we completed random sampling across the [wage subsidy scheme] payments and are now conducted targeted assessments across payments of higher risk profiles.
[36] Following these comments, Audit NZ prepared a second draft of the financial audit report for the 2020 year. The recommendation that MSD should consider communicating with recipients was altered to provide:
We recommend that the Ministry seek positive confirmation (this could be targeted towards larger or risk indicated applicants) from recipients that they have assessed their situation, and confirm that they have subsequently met the requirements of the scheme.
(emphasis added)
[37] This is a much firmer recommendation than that contained in the first draft. While it reflects MSD’s apparent preference for targeted communications with wage subsidy recipients, the use of parentheses indicates that the opening words were intended to leave open the possibility that MSD might undertake enquiries with all recipients. Second, the amendment changed the passive recommendation that MSD “should consider” seeking positive confirmation to far more direct language, involving a positive injunction to do so.
[38] In response to the second draft, MSD’s management highlighted the steps taken for “larger” recipients, and that it would consider contacting a wider pool:
Service Delivery are seeking positive confirmation from larger recipients through our audit processes. We are proactively contacting larger businesses prior to payment to seek this positive confirmation and contacted a number of other businesses through post payment auditing to do the same. The Ministry will consider the feasibility of contacting a broader range of clients.
[39] Audit NZ’s recommendation that confirmation “could be targeted towards larger or risk indicated applicants” was carried through to another draft interim report dated 19 August 2020, and the report on the final audit for the year ended 30 June 2020. It is the inclusion of the words in parenthesis—providing MSD with discretion to use a targeted rather than blanket approach to follow-up—that Gama argues rendered the recommendation unlawful.
[40] As part of the final report, Audit NZ identified what it considered to be a gap in the process followed by MSD in relation to its written communications to wage subsidy recipients concerning repayments:
Ministers have made clear public statements setting expectations about repayments. We have reviewed written communications that have been sent to recipients of the scheme. These in our view have not covered the expectation the recipients assess their eligibility regularly and make repayment when required. There appears to be a gap, as due to the nature of the scheme, we consider it is likely that some applicants would have initially considered their eligibility under conditions of uncertainty or that for others their situation could have changed over the period of the Scheme.
The Ministry is currently considering what additional process it could use to ensure that recipients have adequately considered whether any repayment is required and confirm this assessment.
The Auditor-General’s 2021 performance audit
[41] As early as April 2020, a performance audit was mooted within the Office of the Auditor-General as a means to provide assurance to Parliament about the amount of public money being spent on the wage subsidy scheme.
[42] An audit proposal was developed over following months and, on 14 September 2020, the Auditor-General publicly announced his intention to conduct a performance audit. One of the issues to be addressed was how well the integrity of the scheme was
managed, including recovery of money from anyone who might deliberately or accidentally have received more than they should have. The aim was to complete the initial audit investigations and provide interested Government agencies, including MSD, with a draft report and recommendations before the end of 2020.
[43] In order to manage the overlapping work of the financial and performance audit teams, a Wage Subsidy Scheme Steering Committee was established comprising members of both units. On a number of occasions, the Steering Committee discussed whether it was appropriate for MSD to seek retrospective confirmation of eligibility from wage subsidy recipients. For instance, at a Steering Committee meeting on 24 September 2020, the appointed financial auditor Mr Whittal reported that MSD had “decided not to do this at this stage”.
[44] On 19 October 2020 Mr David Press, an auditor assisting with the performance audit, and officials from MSD met to discuss MSD’s approach to eligibility declarations and enforcement processes. The minutes of the meeting record that:
MSD has not used any specific tools to recover money. MSD has found that simply writing to businesses requesting repayments has been successful. MSD also believes that media coverage, the published list of recipients and coverage by third parties such as business advisors, have all influenced people to repay the subsidy where they have decided they are no longer eligible or have a moral obligation to repay.
[45] The Steering Committee then met again on 22 October 2020. In the context of the financial audit, Mr Whittal reported that MSD “seem[ed] to be coming around to the idea [of seeking positive confirmation] for a limited cohort of recipients”. At the same meeting, the Auditor-General, Mr John Ryan, questioned whether “we should escalate the matter of MSD sending a positive confirmation letter, on eligibility and using the subsidy to pay staff, to all recipients again”.
[46] At the Steering Committee meeting on 24 November 2020, Mr Whittal noted there had been “some difficulty” with a recommendation that MSD “carry out further investigations into those who received [wage subsidy] payments”, and the “better recommendation” was for MSD to “do the positive confirmation exercise”.
[47] The financial audit team met with MSD officials again on 1 December 2020. Notes of the meeting record that there had been “some discussion within MSD about writing to all recipients”. However, an MSD official pointed out that:
… this isn’t straightforward because, for example, it would generate additional calls to service centres which would take up some resources. They are also waiting to see what [Audit NZ] say in their [annual financial] report and are talking to [Inland Revenue] to see what else [Inland Revenue] could do, given [Inland Revenue] collects revenue and profit information.
The Auditor-General’s draft performance audit recommendation
[48] On 11 December 2020, the performance audit team sent MSD comprehensive “preliminary audit findings and conclusions” as part of the performance audit process. A covering letter from the team’s lead, Ms McAviney, explained that the team would “use [MSD’s] comments about the preliminary findings to make necessary changes to our findings and conclusions”. The letter went on to record:
… we welcome your comments about whether the preliminary findings and conclusions are accurate, balanced and fair. We also appreciate that things change over time and welcome evidence of recent changes and developments.
[49] An undated draft of the performance audit team’s preliminary findings contained a proposed recommendation concerning MSD “enforcement work”, which is the focus of Gama’s application for judicial review. The recommendation was in the following terms:
iii.MSD prioritises and resources the remaining enforcement work necessary to take appropriate action where there has been abuse of the Wage Subsidy. As a minimum this should include:
•seeking written confirmation from applicants (who have not refunded payments to date) of their compliance with the eligibility criteria and obligations;
•testing the reliability of a sample of the findings from its post payment assurance work (referred to as “audits” by MSD) by obtaining and using documentary evidence where the post payment assurance work primarily relied on verbal discussions with employers and employees; and
•taking prosecution action where possible and necessary to recover Wage Subsidy funds and/or to hold people to account for criminal behaviour.
[Adjustments may be made here depending on MSD’s final management responses to the 2019/20 annual audit management letter].
(emphasis added)
[50] The preliminary findings also noted that implementation of the wage subsidy scheme had “stretched MSD’s resources”, and that “use of staff from across MSD to support Wage Subsidy work impacted on its capacity to perform other work”. Doing so would divert resources from hardship and income support services and increase time frames for processing applications and making payments. This in turn would “create hardship for some MSD clients”.
[51] Later in the document, the draft performance audit report recorded in a series of headings that its investigation expected to find that MSD had:
… appropriate policies and procedures … in place to support post-payment actions … [including] procedures to effectively recover money where necessary, and to identify fraudulent applications.
We also expected that actions were taken to manage or mitigate post-payment risks to the extent possible within the circumstances.
We also expected resourcing of those actions to be risk-based.
[52] At a Steering Committee meeting on 26 January 2021, Mr Whittal reported that there was a “gap” in the assurance processes MSD were using, “and a positive confirmation exercise is the best way to fill this gap”. The performance audit team also met with MSD on 26 January 2021 before MSD provided its written feedback in respect of the draft report. MSD had sought “clarification” around the wording of the recommendations and the audit team agreed to “use Audit NZ wording if there was any inconsistency”. MSD officials also noted that they did not have the “powers” to seek written confirmation of eligibility for the subsidy, or to seek information to test the reliability of assessment work.
[53] On 3 February 2021, an MSD official sent an email to Mr Whittal. It noted the apparent divergence between Audit NZ’s financial audit report and management letter, and recorded that “some confusion” had arisen from the wording of the performance audit team’s recommendations. The email went on to say that MSD had been “of the understanding that the recommendations made by Audit NZ and the [performance
audit team] on the wage subsidy would align”. A staff member from the audit team responded that he would need to look into it further, but reminded MSD that the annual financial audits and performance audit were separate, so while the recommendations would not contradict each other, they would not necessarily be the same.
The draft performance audit recommendation is changed
[54] Following the engagement with MSD (and other state sector agencies) in relation to the preliminary findings, the performance audit team decided to alter its draft recommendation regarding written confirmation of eligibility from applicants. The revised wording was provided to MSD for comment under cover of a letter dated 10 March 2021 with a copy of a further draft performance audit report. The letter sought MSD’s feedback on whether the draft recommendations were “relevant, reasonable, and practicable”. The amended “Recommendation 4” now aligned with that contained in the 2020 interim financial annual audit report of Audit NZ and said MSD could, if it wished, target written confirmation “towards larger or risk-indicated applicants”.
[55] The same wording was ultimately adopted in the final performance audit report, which issued on 4 May 2021. The recommendation recorded:20
Recommendation 4
In relation to the Wage Subsidy Scheme, we recommend that the Ministry of Social Development prioritise remaining enforcement work, including:
•seeking written confirmation from applicants (which could be targeted towards larger or risk-indicated applicants) of compliance with the eligibility criteria and the obligations of receiving the subsidy; and
•pursuing prosecutions to recover funds and/or to hold businesses to account for potentially unlawful behaviour.
(emphasis added)
[56] The final performance audit report also made four other recommendations, including that MSD “test the reliability of a sample of the post payment assurance work it carried out against documentary evidence held by applicants”
20 Performance audit report, above n 19, at [4.90].
(Recommendation 3),21 and, alongside other public sector agencies, carry out a timely evaluation of the wage subsidy scheme to inform future schemes (Recommendation 5).22
MSD’s response to Recommendation 4
[57] MSD responded to Recommendation 4 by sending an email seeking written confirmation of eligibility to 1,000 wage subsidy recipients between 29 March and 22 May 2020. The sample was weighted towards larger businesses, with 500 of the sample having at least 80 employees. No evidence of eligibility was required in response to the email, and the sample did not include recipients who had been paid before 27 March 2020.
[58] Over a nine-month period, MSD sent two reminder emails to members of the sample who did not respond. Eight-hundred and sixty-four recipients responded.23 As at 18 October 2021, MSD had received 535 replies. Of these, 532 employers confirmed they met the eligibility criteria and had complied with their obligations. Three applicants who responded indicated they might not meet the criteria. Two ultimately paid back the subsidy in full or in part, and it was determined the third was eligible.
Continued reporting
The Auditor-General’s continued reporting to the Finance and Expenditure Committee
[59] In August 2021 the Finance and Expenditure Committee reported to Parliament on the performance audit report, having heard evidence from the Auditor-General. The Committee recommended to the House that it take note of the report and urged MSD to continue working to improve its management of the wage subsidy scheme. In terms
21 Performance audit report, above n 19, at [4.72].
22 Performance audit report, above n 19, at [5.18].
23 In response to Recommendation 3, from July 2021, MSD requested documentary evidence of eligibility from a limited sample of 339 applicants previously subject to random post-payment checks who had not been required to make repayment. This comprised an initial phone conversation with the applicant to assess which documents to request, provided the applicant said they were available, followed up with a written request.
of its future expectations of the Auditor-General, the Committee concluded by saying:24
Because of the scheme’s significance, and the out-sized effect that this single item has had on the Government’s fiscal position, we expect that the Auditor-General will continue to prioritise auditing of the scheme. If appropriate, we expect that the [Office of the Auditor-General] will conduct a follow-up audit of the management scheme, with particular focus on MSD’s implementation of the recommendations from the Auditor-General’s report.
[60] The Auditor-General’s office continued to monitor MSD’s implementation of the performance audit’s recommendations. This work involved preparing a detailed letter to the Finance and Expenditure Committee following further enquiries with MSD and other agencies.
[61] In November 2021, Mr Press from the performance audit team emailed the appointed financial auditor, Mr Whittal, about MSD’s response to the Auditor-General’s recommendations. Replying to a query from Mr Press about MSD’s implementation of Recommendation 3, Mr Whittal explained that MSD:
… targeted the earlier applications (first iteration of the Scheme), in particularly [sic] applicants with a small number of employees (the random checks were anyway initially were over smaller applicants) – their (MSD) rationale is that the larger applicants had better system and capability and therefore the risk is with smaller entities and sole traders. MSD also judged that 1000 risk identified and targeted checks held better information, were focused and overall provided better assurance.
[62] The same month, in an internal paper, the Auditor-General commented on the draft reporting letter to the Committee, saying the key point to be made is “the still undetected level of non-compliance”. Then in December 2021, in comments made on the draft letter to the Finance and Expenditure Committee, the Auditor-General described MSD’s proposal to send reminder emails to the sample of 1,000 recipients who had not replied as “pathetic”, and queried why MSD was not writing to everyone “given the high success rate and such a low cost”. In a later draft of the letter, the Auditor-General appears to have expressed surprise that MSD had not done more to validate payments given the scale and significance of the scheme.
24 Finance and Expenditure Committee Report of the Finance and Expenditure Committee – Report of the Auditor-General, Management of the Wage Subsidy Scheme (August 2021) at 10 [Finance and Expenditure Committee report].
[63] The final version of the reporting letter was sent to the Finance and Expenditure Committee by the Auditor-General on 21 December 2021. The letter recorded that “[o]verall there has been good progress with implementing the recommendations in [the performance audit report]”. The Auditor-General noted that MSD had implemented Recommendation 4 by “seeking written confirmation” and “pursuing prosecutions”. Once again, the Auditor-General did not recommend that MSD should extend the written confirmation exercise to all recipients of the wage subsidy.
[64] As to MSD’s response to Recommendation 3, by 27 October 2021, 186 of the sample of 339 from whom MSD had sought documentary evidence had responded. The other 45 per cent had not. Of the responses, 14 recipients (7.5 per cent) had been required to make a full or partial refund. The Auditor-General’s letter to the Finance and Expenditure Committee made the following comments on the response:
Although these may appear to be small numbers, it is important this information is carefully considered to determine what further work needs to happen to protect the integrity of the whole Scheme, and the significant public funds involved.
MSD does not yet know whether the findings from its follow-up work are indicative of wider non-compliance with the requirements of the Scheme. If they are, this could involve a substantial amount of public funds. This reinforces the importance of MSD knowing more about the extent of non- compliance across all iterations of the Scheme and taking action to recover funds paid to applicants who did not meet the criteria or their obligations.
MSD told us that it intends to analyse the results of all integrity and assurance work to inform a risk-based assessment of next steps, and what further integrity work needs to be completed, if any, to strengthen the integrity of the Scheme. MSD has said that this analysis will happen in 2022. It is important that this work is progressed.
We will continue to monitor the progress made with implementing our recommendations through our annual audit and ongoing engagement our sector managers have with the audited agencies, in particular with MSD.
When MSD’s testing work is completed, we expect MSD to consider whether further additional checks or changes to integrity work are required to protect the integrity of past, current, and future iterations of the Scheme. Even small levels of non-compliance across the Scheme as a whole involves significant public funds.
[65] The Auditor-General also suggested that the Finance and Expenditure Committee ask MSD to quantify the “most recent estimate of likely non-compliance”.
Audit NZ’s continued reporting in 2022
[66] Audit NZ continued to monitor MSD’s integrity work on the wage subsidy scheme through its annual financial audits. In a draft report to MSD’s Chief Executive on the annual audit for the year ended 30 June 2022, Audit NZ concluded that MSD’s sample of 1,000 wage subsidy recipients had received a “reasonable response rate” of 70 per cent.25 The draft went on to state:
We recommend that the Ministry significantly extends the number of confirmations it is seeking urgently to cover all cohorts of recipients. The confirmation should include that the Ministry has evidence of recipients not paying back the subsidy when required and request that they confirm their eligibility or repay. This will provide additional assurance of the level of compliance overall for the scheme.
We will continue to follow up with the Ministry on the progress made on our recommendations as part of next year’s audit.
[67] MSD responded in an email to Audit NZ on 13 December 2022. It did not consider the draft recommendation was appropriate given a number of factors:
In regard to the Ministry confirming with all recipients their eligibility for the wage subsidy we did question the value of this in light of what we know to date and the significant effort that would be required – for the first scheme there were around 400,000 applicants, with around 88% of these being sole traders or employers with five or fewer employees who received around 31% of the value of the scheme.
…
Overall, we have reasonably targeted our integrity efforts for the nature of the scheme, which was broad economic support delivered through a high trust model. We believe any further work the Ministry undertakes needs to be balanced with its expected effectiveness as well as the ongoing impacts on our core duties. Once we have the findings of the Deloitte assessment, and have considered them along with the results above, we will be making recommendations as to appropriate next steps. It is unlikely that writing to every recipient and requesting and assessing documentary evidence is practical, but there may be other opportunities we can explore.
[68] It became clear to Mr Whittal that MSD was considering scaling down its proactive work on the wage subsidy scheme. Mr Whittal did not consider this appropriate. As a result, in the final financial audit report for the year end 30 June 2022
25 This opinion was retained in Audit NZ’s final report.
to the Chief Executive, the draft recommendation that MSD “urgently” extend its confirmation exercise was replaced with a much less prescriptive or exacting proposal:
The Ministry needs to determine what additional steps are necessary to reach an acceptable level of understanding about how much may have been overpaid, and what it should do to recover overpayments.
First issue: does parliamentary privilege attach to the documents in issue?
[69] Two weeks before the hearing of the application for judicial review, the Auditor-General filed an amended statement of defence which raised for the first time an affirmative defence based on parliamentary privilege. The pleading asserts that all steps undertaken by the respondent and his employees “for the purposes of or incidental to” the preparation of the reports and recommendations in issue are “Proceedings in Parliament” as that term is defined in s 10 of the Parliamentary Privilege Act.
[70] If the various reports that are the focus of Gama’s challenge fall within that definition, by operation of s 11 of the Parliamentary Privilege Act, Gama would be prevented from offering them in evidence or making submissions on them for the purpose of drawing inferences or conclusions, or proving any fact “necessary for, or incidental to, establishing any liability”, or to support any “order remedy or relief” sought in this proceeding.26
[71] Before turning to address that question, however, I first deal briefly with another jurisdictional issue raised by the Auditor-General.
A preliminary issue — are recommendations by public bodies capable of review?
[72] Counsel for the Auditor-General, Ms McDonald KC, argued as a preliminary point that there is authority to suggest the courts do not usually review a recommendation made by public bodies, and it was therefore doubtful that the recommendations of the Auditor-General in this case are capable of judicial review.27
26 Parliamentary Privilege Act 2014, s 11(c)-(e).
27 Relying on O’Regan v Lousich [1995] 2 NZLR 620 at 629 and Gazley v Wellington District Law Society, SC Wellington A48/78, 16 June 1978.
[73] The authorities relied on by the Auditor-General to support the argument are not apt in the present case. Gazley v Wellington District Law Society concerned Mr Gazley’s challenge by way of judicial review of a circular issued by the respondent containing advice about the responsibilities of legal practitioners conducting domestic proceedings.28 Mr Gazley disagreed with the views expressed and sought to challenge them. In striking out the proceeding, the Supreme Court’s decision turned principally on its conclusion that the Wellington District Law Society had not exercised a statutory power within the scope of s 3 of the Judicature Amendment Act 1972.29 Similarly, in O’Regan v Lousich,30 Tipping J considered an application for certiorari of an adverse comment made by the Māori Land Court of a witness in a decision.31 An application for judicial review under the Judicature Amendment Act had not been pursued.32 While the Court noted that “actions which amount only to recommendations will not normally be susceptible to certiorari”,33 the Court went on the find that its jurisdiction was engaged, and made an order quashing the Māori Land Court’s “finding” about the witness.34 Both cases reflect a view prevalent at the time they were decided that the exercise of a public power that does not determine an applicant’s rights or interests is beyond the Court’s supervisory jurisdiction.
[74] The present case is quite different. There should be no real question that the Auditor-General’s exercise of powers under the Act are statutory powers subject to judicial review. As the Supreme Court has noted, in principle, all exercises of public power are reviewable, whether the relevant power is derived from statute, the prerogative or any other source.35 While there are limits on the scope of review, if in the course of exercising statutory functions under the Public Audit Act the Auditor-General acts unlawfully—even if that involves making a recommendation as Gama alleges—his decision will generally be amenable to judicial review, unless the
28 Gazley v Wellington District Law Society, above n 27, at 1.
29 At 10.
30 O’Regan v Lousich, above n 27.
31 At 624.
32 At 625.
33 At 629.
34 At 632.
35 Ririnui v Landcorp Farming Ltd [2016] NZSC 62, [2016] 1 NZLR 1056 at [1].
Court’s ability to do so is otherwise prevented by parliamentary privilege, or for some other reason.36
[75] I am reinforced in this view because Parliament did not include in the Public Audit Act an ouster of the Court’s jurisdiction as it has done for the Ombudsman.37 I regard the distinction drawn between the two officers of Parliament as an expression of different parliamentary intention. While it is clear the Ombudsman’s exercise of powers is beyond the scope of judicial review, the fact Parliament did not exempt the Auditor-General from the Court’s supervisory jurisdiction is an indication that it wished to retain the benefit of the court’s assistance in ensuring its officer acts lawfully and otherwise within the scope of the powers Parliament has conferred.
[76] I will return later in this judgment, however, to the relevance of the Auditor- General’s status as an Officer of Parliament when considering the availability of relief.
The Auditor-General’s argument on parliamentary privilege
[77] The Auditor-General’s position is that, given his work as an Officer of Parliament, and the fact the Finance and Expenditure Committee engaged with and reported to Parliament on his monitoring of the wage subsidy scheme, Gama is prevented from questioning the efficacy of the Auditor-General’s monitoring or the recommendations he made because to do so amounts to a challenge to proceedings in Parliament.
[78] Ms McDonald for the Auditor-General argues the principle of comity requires that the Court exercise caution before reviewing the performance of an Officer of Parliament. As indicated, performance audit reports are often presented to the House and referred to a Select Committee, and annual reports of government departments (which draw on the annual financial audits by Audit NZ) are required to be tabled in
36 As Lord Diplock said in R v Inland Revenue Commissioners, Ex parte National Federation of Self-employed and Small Businesses Ltd [1982] AC 617 at 644: “It is not, in my view, a sufficient answer to say that judicial review of the actions of officers or departments of central government is unnecessary because they are accountable to Parliament for the way in which they carry out their functions. They are accountable to Parliament for what they do so far as regards efficiency and policy, and of that Parliament is the only judge; they are responsible to a court of justice for the lawfulness of what they do, and of that the court is the only judge.”
37 Ombudsmen Act 1975, s 25.
the House by the relevant Minister.38 Ms McDonald says that what is before one branch of government should not be evaluated by another, and this principle operates as a legal limit controlling the Court’s approach to Gama’s claim. She argues that it is also inappropriate for the Court to second-guess the expert recommendations of the Auditor-General acting within his core function, especially where that function is complementary to the Courts’ own check on the executive through judicial review.
[79] As to the second strand of argument, Ms McDonald points to Gama’s use of four documents in particular: the financial audit reports for the 2019-2020 and 2021-2022 financial years; the performance audit report; and the Auditor-General’s letter of response to the Finance and Expenditure Committee’s own review of the performance audit report. The Auditor-General considers these documents fall within the definition of “proceedings in Parliament”, so relying on the documents to support Gama’s judicial review means that Gama is infringing the privilege. Questioning the Auditor-General’s work necessarily impeaches parliamentary treatment of it now that various reports have been tabled in the House and the Finance and Expenditure Committee has responded with its own consideration of at least one of them. To the extent Gama’s case seeks to do so, Ms McDonald says that the Court must ignore the evidence and argument. She submits this includes “all preparatory or incidental steps”, such as internal emails or drafts of the reports.
[80] Ms McDonald emphasises that parliamentary privilege does not operate such that the Court is never able to refer to or address in evidence material relating to proceedings in Parliament. For example, the Court would be able to legitimately use material to demonstrate a factual or legislative history. It is the use to which the privileged material is put in Court that is crucial. Using the material to call into question the relevant proceedings in Parliament—as in the present case—infringes the privilege.
[81] In response, Mr Mijatov emphasises that the nature of the connection with the House’s business is the “touchstone” in deciding whether the Auditor-General’s work or a particular document falls within the definition of “proceedings in Parliament”.
38 Public Finance Act, s 44.
[82] To begin, Mr Mijatov relied on the authority from the Supreme Court of the Australian Capital Territory in ACT v SMEC Australia Pty Ltd to argue that the Auditor-General was not immunised from judicial review as a result of their office.39 In relation to a claim to parliamentary privilege, Associate Judge McWilliam explained that the focus of the enquiry was the purpose for which the document was created, or the act was done. This was a question of fact requiring “an assessment of the subjective purpose of the actor in question”. A document prepared for a non-parliamentary purpose is not transformed into a document forming part of or incidental to the transacting of the business of the House, even if it is subsequently used in Parliament.40 Gama stresses this last point in particular.
[83] As to the four documents whose use is challenged by the Auditor-General, Gama does not accept they are capable of attracting privilege anyway. Financial audit reports are prepared as part of auditing a government department and are addressed to the Chief Executive of the relevant state agency. Audit NZ does not present those reports to Parliament, the relevant Minister does through their own annual report. Similarly, the performance audit of MSD and the resulting performance audit report were prepared by the Auditor-General on his own motion. There was nothing to suggest the audit process had a connection to specific business of the House at the time the report was being prepared. As for the reporting letter to the Finance and Expenditure Committee, Gama says there can be no impermissible use of the document because it does not contain the impugned recommendations, only factual context.
[84] In any case, there is no impermissible use of those documents here. Mr Mijatov submits that Gama is not seeking to challenge the core or essential business of Parliament, or any speech or debate in the House or its Committees, but rather respond to the actions of the Auditor-General, who is not a member of the House over which privilege automatically extends. Gama’s position is that this judicial review does not seek to quash or set aside the documents, nor does it invite the Court to make findings about parliamentary treatment of them after their receipt or publication.
39 Australian Capital Territory v SMEC Australia Pty Ltd [2018] ACTSC 252, (2018) 337 FLR 390.
40 At [48]–[49].
Parliamentary privilege and the Parliamentary Privilege Act 2014
[85] Parliamentary privilege encompasses certain special rights, powers and immunities held by members of the House of Representatives that allow them to effectively discharge their functions and duties.41 The privileges, powers and immunities held by the House are adapted to the needs or purposes of the legislature.42 Generally recognised privileges include (among others) freedom of speech, freedom of debate, exclusive control of the House’s own proceedings, control of reports of the House’s proceedings, and the power to punish for contempt anyone who commits a breach of Parliament’s privileges or interferes with its members in the execution of their duties.43
[86] Parliamentary privilege was born out of the laws and customs of parliamentary practice. By the time representative government was established in New Zealand, the United Kingdom’s House of Commons had evolved its rules of privilege over nearly six centuries.44 Importantly, this included art 9 of the Bill of Rights 1688, which provides:
Freedom of speech
That the freedom of speech and debates or proceedings in Parliament ought not to be impeached or questioned in any court or place out of Parliament:
[87] The Parliamentary Privileges Act 1865 was passed to confer on the New Zealand House of Representatives the privileges equivalent to those enjoyed by the House of Commons.45 As is discussed below, the nature, scope and extent of these privileges was further reaffirmed in New Zealand by the Parliamentary Privilege Act 2014.
[88] Although parliamentary privilege is today an important constitutional norm, considered essential to the public interest in an effective functioning parliamentary system, it is also part of the general law of New Zealand and remains a creature of
41 David Wilson (ed) Parliamentary Practice in New Zealand (5th ed, Clerk of the House of Representatives, Wellington, 2023) at [57.1].
42 Wilson, above n 41, at [57.2.2].
43 Wilson, above n 41, at [57.3].
44 Wilson, above n 41, at [57.2].
45 Wilson, above n 41, at [57.2.1]–[57.2.2].
statute. As such, it may be abrogated, in whole or in part, by legislation. Such abrogation will be rare though, given the constitutional importance of the privilege, and only clear and unambiguous legislation will do.46 It is also possible for the otherwise wide privileges of the House to be tempered by other legislation, such as the New Zealand Bill of Rights Act 1990 (NZBORA). In short, the privilege is not limitless, and as David Wilson puts it in Parliamentary Practice in New Zealand:47
… the existence of other interests that may be infringed or abridged by the operation of parliamentary privilege justifies restricting the privileges to activities with a real connection to the operation of the legislature, and confining their scope so as not to trespass on other rights unnecessarily.
[89] Of particular importance in this proceeding is Parliament’s right of “exclusive cognisance”. In other words, its exclusive jurisdiction over how its proceedings are to be conducted. The courts recognise that internal decisions of the House rest within that sphere. In this way, the right of exclusive cognisance is tied closely to the principles of comity and separation of powers. Exclusive cognisance is considered broader than the privilege contained in art 9 of the Bill of Rights 1688 and is seen to embrace it.48 As this case demonstrates, there may be some difficulty in establishing the precise limits of exclusive cognisance, and to what (if any) extent the protections of parliamentary privilege apply to matters ancillary to the core work of the House but that remain necessarily connected to it.
[90] The Parliamentary Privilege Act is now the operative legislation that defines the nature and scope of parliamentary privilege in New Zealand. The Act reaffirms the privilege, but it is not intended to codify it.49 Section 3(1) provides that the Act has two main objects, being to:50
(a)reaffirm and clarify the nature, scope, and extent of the privileges, immunities, and powers exercisable by the House of Representatives, its committees, and its members; and
(b)ensure adequate protection from civil and criminal legal liability for communication of, and of documents relating to, proceedings in Parliament.
46 Wilson, above n 41, at [57.5].
47 Wilson, above n 41, at [57.1] (footnote omitted).
48 R v Chaytor [2010] UKSC 52, [2011] 1 AC 684 at [13].
49 Parliamentary Privilege Bill 2013 (179–1) (explanatory note).
50 Parliamentary Privileges Act, s 3(1).
[91]Parliament was also clear that the Act must be interpreted in a way that:51
…
(b)promotes the principle of comity that requires the separate and independent legislative and judicial branches of government each to recognise, with the mutual respect and restraint that is essential to their important constitutional relationship, the other’s proper sphere of influence and privileges; and
(c)ensures privileges, immunities, and powers of the House of Representatives, its committees, and its members are exercisable for the purpose stated in section 7.
[92] The Act also sets out Parliament’s understanding of the purpose of the privilege:
7 Purpose of parliamentary privilege
The privileges, immunities, and powers exercisable in accordance with the rest of this Act by the House, committees, and members, are exercisable to—
(a)uphold the integrity of the House as a democratic legislative assembly; and
(b)secure the independence of the House, committees, and members, in the performance of their functions.
[93] The statute was passed as a response to what Parliament perceived to be a “significant shift in the interpretation of the scope of privilege” following the Supreme Court’s decision in Attorney-General v Leigh (Leigh).52 Leigh was a defamation case concerning the then Minister for the Environment, the Hon Trevor Mallard. The Minister had requested a briefing from the Chief Executive of the Ministry for the Environment on the circumstances giving rise to a question he had been preparing for oral answer in the House. This task was delegated to a Mr Gow, a deputy secretary in the Ministry, who briefed the Minister both orally and in writing, and the Minister used the information supplied to him to answer the question in the House. The respondent, Ms Leigh, issued defamation proceedings claiming that Mr Gow had defamed her in what he had told the Minister. The Minister and Mr Gow argued that the communications had taken place on an occasion of absolute privilege, and the
51 Section 4.
52 Parliamentary Privilege Bill (explanatory note); in reference to Attorney-General v Leigh [2011] NZSC 106, [2012] 2 NZLR 713 [Leigh].
defamation claim should be struck out. Both the High Court and the Court of Appeal held that the occasion was instead one of qualified privilege and therefore could not be struck out. The Supreme Court agreed.53
[94] After the Leigh decision, the Privileges Committee drafted a report entitled “Question of privilege concerning the defamation action Attorney-General and Gow v Leigh”.54 This was presented to the House in June 2013.
[95] The Committee recommended the House pass legislation to restore the scope of parliamentary privilege to Parliament’s “previous understanding”.55 Specifically, the Committee recommended that the House disagree with the Supreme Court’s application of a “necessity test” in Leigh when ascertaining the scope of the privilege of freedom of speech.56 The Committee also recommended that the definition of “proceedings in Parliament” is provided for the avoidance of doubt.57
[96] The Committee explained that “the principle of exclusive cognisance reflects the importance of ensuring no overlap between the courts and Parliament”.58 Parliamentary privilege and the powers it confers are “not a bare minimum to allow the House to operate—they ensure the House can carry out its functions fully”. This meant that the restriction of privilege to what is deemed “necessary” was seen to “severely limit the way Parliament evolves to remain relevant and to operate in the public interest”.59
[97] The House adopted the Committee’s recommendations. It did so by enacting a definition in s 10 of “proceedings in Parliament” as follows:
10Proceedings in Parliament defined
(1)Proceedings in Parliament, for the purposes of Article 9 of the Bill of Rights 1688, and for the purposes of this Act, means all words
53 Leigh, above n 52, at [1].
54 Privileges Committee “Question of privilege concerning the defamation action Attorney-General and Gow v Leigh” [2011–2014] AJHR I.17A at 5.
55 Parliamentary Privilege Bill (explanatory note).
56 Privileges Committee, above n 54, at 31.
57 At 34.
58 At 16.
59 At 19.
spoken and acts done in the course of, or for purposes of or incidental to, the transacting of the business of the House or of a committee.
(2)The definition in subsection (1) must be taken to include the following:
(a)the giving of evidence (and the evidence so given) before the House or a committee:
(b)the presentation or submission of a document to the House or a committee:
(c)the preparation of a document for purposes of or incidental to the transacting of any business of the House or of a committee:
(d)the formulation, making, or communication of a document, under the House’s or a committee’s authority (and the document so formulated, made, or communicated):
(e)any proceedings deemed by an enactment to be (or a thing said or produced, or information supplied, in an inquiry or proceedings, if an enactment provides the thing or information is privileged in the same way as if the inquiry or proceedings were) for those purposes proceedings in Parliament.
(3)In determining under subsection (1) whether words are spoken or acts are done for purposes of or incidental to the transacting of the business of the House or of a committee, words spoken or acts done for purposes of or incidental to the transacting of reasonably apprehended business of the House or of a committee must be taken to fall within subsection (1).
(4)In determining under subsection (1) whether words are spoken or acts are done for purposes of or incidental to the transacting of the business of the House or of a committee, no necessity test is required or permitted to be used.
(5)Necessity test includes, but is not limited to, a test based on or involving whether the words or acts are or may be (absolutely, or to any lesser degree or standard) necessary for transaction of the business.
(6)Subsections (2) and (3) do not limit subsection (1).
(7)This section applies despite any contrary law (including, without limitation, every enactment or other law in the decision in Attorney General v Leigh [2011] NZSC 106, [2012] 2 NZLR 713 (SC)).
[98] By applying to “all words spoken and acts done”, particular documents or materials can themselves meet the definition of “proceedings in Parliament”. The way
that evidence of proceedings in Parliament may be used in court is then limited by the operative provision, namely s 11:
11Facts, liability, and judgments or orders
In proceedings in a court or tribunal, evidence must not be offered or received, and questions must not be asked or statements, submissions, or comments made, concerning proceedings in Parliament, by way of, or for the purpose of, all or any of the following:
(a)questioning or relying on the truth, motive, intention, or good faith of anything forming part of those proceedings in Parliament:
(b)otherwise questioning or establishing the credibility, motive, intention, or good faith of any person:
(c)drawing, or inviting the drawing of, inferences or conclusions wholly or partly from anything forming part of those proceedings in Parliament:
(d)proving or disproving, or tending to prove or disprove, any fact necessary for, or incidental to, establishing any liability:
(e)resolving any matter, or supporting or resisting any judgment, order, remedy, or relief, arising or sought in the court or tribunal proceedings.
Consideration
[99] It is important for courts to interpret the scope of parliamentary privilege by reference to the Act’s purpose and context.60 This is especially so because, while the separation of powers and the principle of non-interference—or comity—underlies the privilege, if the privilege is drawn too widely it will undermine access to justice and the constitutional role of the courts to ensure power is exercised in accordance with the law. Comity is recognised as a placing obligations of restraint and respect on both the judicial and legislative branches of government.61 So in asserting the privilege, and interpreting the Act, Parliament should not be readily taken to have intended to trench upon the constitutional role of another branch of government.62 This is expressly recognised in s 4(1)(b), which requires the Act to be interpreted in a way that promotes the principle of comity:
60 Legislation Act 2019, s 10(1).
61 Skerret-White v Minister for Children [2024] NZCA 160, [2024] 2 NZLR 493 at [105]–[108] and [115].
62 Skerret-White v Minister for Children, above n 61, at [108]; citing Joseph on Constitutional Law, above n 4, [17.14.4].
… that requires the separate and independent legislative and judicial branches of government each to recognise, with mutual respect and restraint that is essential to their important constitutional relationship, the other’s proper sphere of influence and privileges.
[100] If the words of the Act were applied literally and without reference to context and purpose, a great number of cases routinely coming before the Courts would no longer be capable of consideration given the expansive language of s 10’s definition of “proceedings in Parliament”. The sterilising effect of applying the privilege to every document provided to a member of the House, or a select committee, immediately reveals the need for a functional relationship between a document or act, and the business of Parliament. For this reason, recent decisions of the higher courts that have considered the principle of non-interference are helpful when drawing the proper boundaries of the privilege.
[101] In Ngāti Whātua Ōrākei Trust v Attorney General (Ngāti Whātua), Ngāti Whātua challenged the proposed Treaty settlements between the Crown and other iwi which were soon to be implemented by legislation.63 The central issue was whether Ngāti Whātua’s claim was properly characterised as a challenge to legislative proposals, and therefore prevented by parliamentary privilege. The Supreme Court allowed Ngāti Whātua’s appeal in part, allowing it to pursue broadly worded declaratory relief of concerning its rights, but held that it could not seek declarations expressly impugning the proposed settlements because the settlements were to inevitably become legislation.64 The majority of the Court did not find it necessary to resolve the “exact metes and bounds” of the principle of non-interference in parliamentary proceedings, sounded a note of caution at the extent to which the principle has been held to apply to decisions “somewhat distant” from the introduction of a bill to the House.65 It would be “overbroad” to suggest that the fact a decision may be the subject of legislation means that the process leading up to the decision is out of reach of the courts’ supervision.66
63 Ngāti Whātua Ōrākei Trust v Attorney General [2018] NZSC 84, [2019] 1 NZLR 116 [Ngāti Whātua].
64 At [66].
65 At [46].
66 At [46].
[102] In a separate, partially dissenting judgment, Elias CJ considered that, notwithstanding art 9 of the Bill of Rights, it is “for the courts to determine what matters are within the statutory and common law privileges”.67 The principle of non-interference is “not a matter of etiquette or deference” but one “concerned with the function of parliament”68 It reaches matters which are so connected to Parliament’s function that they too must be similarly privileged.69 So long as the court has been asked to determine a present right, and does not seek to preclude parliamentary consideration, a claim would not interfere with proceedings in Parliament.70 The Chief Justice considered that where the relief sought is discretionary (such as declaratory relief), the fact the court’s determination or the subject of the litigation is under the active consideration by Parliament may be relevant in determining whether to grant the relief, but did not preclude the Court’s jurisdiction.71
[103] In Attorney-General v Taylor,72 the Court of Appeal considered the Parliamentary Privilege Act in the context of the nascent jurisdiction to make declarations of inconsistency in terms of the NZBORA. A particular issue that fell for consideration was whether the use of the Attorney-General’s report for Parliament prepared under s 7 of the NZBORA breached parliamentary privilege.
[104] The Court dismissed a submission that to rely on parliamentary proceedings is necessarily to consider whether the views expressed in them were correct.73 Giving the reasons of the Court, Wild and Miller JJ explained that courts will be inhibited and “perhaps disabled” from assessing legislative intent, meaning or incompatibility with the Bill of Rights if they cannot consider anything falling within the definition of proceedings in Parliament.74 In holding that proceedings in Parliament were not inadmissible in incompatibility proceedings per se, the Court noted that s 11 of the Parliamentary Privilege Act means it is necessary to ask “why and how the court uses such material”, and “[a]ttention must be paid to the purpose of the privilege and the
67 At [106].
68 At [108].
69 At [108].
70 At [115].
71 At [112].
72 Attorney-General v Taylor [2017] NZCA 215, [2017] 3 NZLR 24.
73 At [124].
74 At [125].
circumstances”.75 They held that, generally, courts do not impeach parliamentary proceedings “merely by describing parliamentary processes or making a finding about the same subject matter, so long as it is careful not to make a finding on ‘the parliamentary treatment of the matter’”.76
[105] Similarly, in Hata v Attorney-General (No 2) (Hata), this Court was required to determine whether the applicant’s claim (again in the context of Treaty settlements) should be stayed on the basis of non-interference with parliamentary proceedings.77 There, a Treaty claim forming part of the claimant’s application for judicial review had been the subject of a settlement bill introduced before Parliament, but had not been passed before Parliament was dissolved prior to the 2023 general election. The principle of non-interference was raised by the Crown in opposition to the proceeding. Justice Cooke considered that: 78
… the real dividing line is identified by distinguishing between legal questions that are for the court to decide, and political questions that are for Parliament.
[106] Justice Cooke rejected the Crown’s argument based on s 11 of the Act that the proceeding should be stayed. As to the effect of the Parliamentary Privilege Act on the applicant’s proceeding, Cooke J said:79
The fact that proceedings before the court, and proceedings before parliament, generally involve the same subject matter does not eliminate the constitutional role of either body. All that the Parliamentary Privilege Act confirms is that the court has no role to adjudicate on, or otherwise call into question what is happening in parliament. If a party seeks a remedy in relation to what has happened, or is happening in parliament this is a breach of the Act and the court will not address such matters. It will strike out proceedings that invite it to do so. But it will exercise its jurisdiction in relation to matters that do not call into question what is before parliament, including the jurisdiction to determine legal rights that exist entirely independently of what may be taking place in parliament.
[130] However, the position is very different when it comes to the performance audit report. In response to the report, the Auditor-General gave evidence before the Finance and Expenditure Committee. Then, in August 2021, the Committee issued its own report to the House.120 There can be no debate that the Committee’s report to the House is covered by parliamentary privilege. In its report, the Committee not only recommended that the performance audit report should be noted by the House, but it went on to comment on the substantive findings and recommendations of the Auditor-General, and importantly, the Committee concluded with its own response to performance audit report, requiring further action by both MSD and the Auditor-General. The Committee said:121
Given that the effectiveness of the Wage Subsidy Scheme relies on the ability to distribute funds to employers quickly, we believe the scheme necessitates a high-trust model. This has the benefit of saving jobs and providing certainty to business owners.
Therefore, we also believe it is imperative that the agencies responsible for the scheme establish clear operating procedures and guidelines which incorporate the Auditor-General’s recommendations. These guidelines should seek to achieve a balance between retaining the high-trust model, and minimising the potential for abuse of the system.
We therefore urge MSD to continue working to improve its management of the scheme. We believe that this is important, and should be done with a high degree of urgency. This work is important not just for recovering public funds,
120 Finance and Expenditure Committee report, above n 24.
121 Finance and Expenditure Committee report, above n 24, at 10–11.
but would also help to support the integrity of all current and future government assistance programmes—particularly those that operate on a high-trust model. The risks posed to the New Zealand economy by the COVID-19 pandemic remain substantial. We are concerned that, should a new outbreak of COVID-19 necessitate a return to higher alert levels, the Wage Subsidy Scheme might need to be reintroduced before MSD had fully implemented the Auditor-General’s recommendations.
Because of the scheme’s significance, and the out-sized effect that this single item has had on the Government’s fiscal position, we expect that the Auditor General will continue to prioritise auditing of the scheme. If appropriate, we expect that the [Office of the Auditor-General] will conduct a follow-up audit of the management of the scheme, with a particular focus on MSD’s implementation of the recommendations from the Auditor-General’s report. We look forward to having future discussions with both the [Office of the Auditor-General] and MSD about the management of the Wage Subsidy Scheme.
[131] The Committee’s consideration of the Auditor-General’s report led to a follow-up line of enquiry by the Auditor-General into MSD’s response to his recommendations, and a further detailed reporting letter to the Select Committee in December 2021.
[132] As a matter of fact, then, I am satisfied that the performance audit report, and its subsequent reporting letter, are proceedings of Parliament in terms of s 10 of the Parliamentary Privilege Act. There is an obvious link between the report, its receipt by the House, and its incorporation into business of the House. Further, in terms of the requirement in s 11 for a link between the documents and the purpose to which they are put in this proceeding, I am also satisfied that Gama has sought to rely on the performance audit report to invite a conclusion on the efficacy of Recommendation 4 within it (contrary to s 11(c) of the Act), and to support a claim for a judgment, order or relief (contrary to s 11(e)). I am also satisfied that the Auditor-General’s December 2021 reporting letter to the Committee is also subject to the privilege.
[133] Those conclusions have implications for the status of the recommendations in the two financial audit reports. That is because the recommendations that Gama seeks to impugn within them is the same as that in the performance audit report. Given my conclusion that Recommendation 4 and its treatment by the House is a proceeding in Parliament, I have also concluded that the same recommendation in the financial audit reports are “words spoken and acts done … incidental to the transacting of the business of the House”. It would deprive the privilege of its efficacy and undermine the public
interest it protects to permit Gama to collaterally challenge in substance the same recommendation merely because it is repeated in reports that have not been the subject of separate, substantive consideration by the House.
[134] If I am wrong in this conclusion, I would have found that the engagement by Parliament with the performance audit report, and in particular Recommendation 4, removes that part of Auditor-General’s financial audit reports from the Court’s supervisory jurisdiction, and in any case would suggest the grant of relief would be inappropriate, even if a ground of review was made out.
[135] I have also considered whether, in keeping with the view of Cooke J in Hata, it might be open to consider judicial review of the performance audit and related reports is compatible with Parliament’s engagement with them, and its business. However, the context is very different here. Hata concerned a claim to existing customary rights and interests under Te Tiriti ō Waitangi that were soon to be extinguished by an act of Parliament. As Cooke J recognised, the legal problem that had been brought to the Court for determination predated the introduction of the settlement legislation to the House. In those circumstances, provided the form of relief the Court granted did not impinge on the business of the House, there was no reason on which the applicant’s should be denied access to the Court. It is an essential function of the Court to answer questions in respect of the legal bounds of a claimant’s rights. In the present case, Gama’s rights and interests are not directly affected by the Auditor-General’s recommendation. There can be no real suggestion that giving effect to the privilege in this case will undermine the constitutional role of the Court. For these reasons, the principle of non-interference is not as acute as it was in cases such as Ngāti Whātua, Wairarapa Moana Ki Pouākani Inc, or Hata where, notwithstanding the passage of settlement legislation, declarations by the courts would continue to have practical benefit for the parties in relation to their ongoing Treaty relationship.
[136] Finally, Gama’s present difficulty is in large part the result of its delay in pursuing a challenge to the Auditor-General’s performance audit report. The report issued in May 2021. The Finance and Expenditure Committee did not issue its response until August 2021. This proceeding was not filed until December 2022, over a year later. Had Gama filed its proceedings sooner—and critically before the Finance
and Expenditure Committee’s engagement with the performance audit report—it is possible that a very different outcome would have been reached in terms of parliamentary privilege. It also follows that Gama’s delay in pursuing its proceeding would be a further reason making relief in this case inappropriate.
Conclusion on parliamentary privilege
[137] Given my conclusions, and the effect of s 11 of the Parliamentary Privilege Act, there is an argument that it is not only unnecessary but in fact inappropriate to consider the substance of Gama’s judicial review challenge, to the extent it is reliant on the content of the Auditor-General’s reports. However, two factors have led me to consider it is appropriate to consider the merits of Gama’s claim. First, unlike other forms of privilege, there is no suggestion the documents in issue are confidential or otherwise not available to the public. To address them in a public judgment dealing with the merits of Gama’s application for review is in substance no different from dealing with them—as I have—in relation to the respondent’s claim to privilege.
[138] Second, the parties addressed the merits of the application for judicial review, and if I am wrong in my conclusion on parliamentary privilege, it is helpful to them to have an answer to the underlying issues.
[139]I therefore turn to consider them.
Second issue: did the Auditor-General breach his statutory duty of independence or otherwise fetter his discretion?
[140] The second issue for determination relates to Gama’s second and third pleaded grounds of review, and was the applicant’s principal focus at the hearing. Gama says the Auditor-General’s decision to “amend or delete” recommendations that MSD write to all recipients of the wage subsidy to seek positive confirmation of eligibility was a breach of the Auditor-General’s statutory duty of independence and constitutes an unlawful fettering of his discretion.
[141] There are three elements to Gama’s claim. The first is whether the change in the draft recommendation was supported by evidence consistent with the requirements
of the relevant audit standards. Related to this is the second question, namely, what is the scope and effect of the Auditor-General’s final recommendation, and whether it meant, as Gama alleges, that MSD should only write to “larger or risk indicated applicants”. The third question is whether the evidence establishes the Auditor- General fettered his discretion.
The duty of independence and applicable audit standards
[142] Section 9 of the Public Audit Act sets out the Auditor-General’s statutory duty of independence:
9 Duty to act independently
The Auditor-General must act independently in the exercise and performance of the Auditor-General’s functions, duties, and powers.
[143] The requirements of the duty of independence in any given case is a question of fact and law for the court. That said, there are several audit standards that may assist the Court’s assessment of the practical requirements of independence, and both parties referred to them in the course of submissions.
[144] First, the Office of the Auditor-General is a member of the International Organisation of Supreme Audit Organisations, which sets standards for public sector auditing internationally. Of relevance is principle 3 of the INTOSAI Standard P-10, which was emphasised by the applicant as being relevant to determining the scope of the statutory duty to act independently. Principle 3 provides that:122
SAIs [Supreme Audit Institutions] should not be involved or be seen to be involved, in any manner, whatsoever, in the management of the organizations that they audit.
SAIs should ensure that their personnel do not develop too close a relationship with the entities they audit, so they remain objective and appear objective.
[145] This standard has not been adopted by the Auditor-General.123 However, in his affidavit evidence, Mr Beardsworth for the respondent considered that certain
122 International Organisation of Supreme Audit Institutions INTOSAI-P 10 Mexico Declaration on SAI Independence (Declaration endorsed in 2007, Preamble amended in 2018).
123 Section 23 of the Public Audit Act requires the Auditor-General to publish the auditing standards that it will apply when conducting audits, inquiries, and providing other auditing services.
INTOSAI standards are reflected in the auditing standards published by the Auditor- General. Additionally, a requirement that independence is seen to be achieved is also reflected in the Auditor-General’s Code of Ethics:124
The perception of independence (independence in appearance) is a vital component of independence, which those who carry out work on behalf of the Auditor-General must consider when assessing their independence. Such an assessment requires considering how a particular situation would look from the perspective of a reasonable and informed third party with only publicly available information.
[146] In 2020, the Auditor-General’s standards on independence were contained in “AG PES-1 – Code of ethics for assurance practitioners”.125 Under this standard, the concepts of integrity, objectivity and professional scepticism are emphasised, alongside the concept of independence being seen to be achieved. The relevant section on independence provides:
120.12 A 1 Assurance practitioners are required by International Independence Standards (New Zealand) to be independent when performing audits, reviews, or other assurance engagements. Independence is linked to the fundamental principles of objectivity and integrity. It comprises:
(a)Independence of mind – the state of mind that permits the expression of a conclusion without being affected by influences that compromise professional judgement, thereby allowing an individual to act with integrity, and exercise objectivity and professional scepticism; and
(b)Independence in appearance – the avoidance of facts and circumstances that are so significant that a reasonable and informed third party would be likely to conclude that a firm’s or an audit, review or assurance team member's integrity, objectivity or professional scepticism has been compromised.
(emphasis in original)
[147] In the context of performance audits, the duty of independence is supplemented by the Auditor-General’s Auditing Standard 5 (AG-5). AG-5 establishes principles and
124 Controller and Auditor-General Auditor-General’s Code of Ethics (Office of the Auditor-General, March 2023) at [12].
125 Controller and Auditor-General AG PES 1 – The Auditor-General of New Zealand’s Standard on Professional and Ethical Standard 1 International Code of Ethics for Assurance Practitioners (Including International Independence Standards) (New Zealand) (Office of the Auditor-General, issued March 2020, updated September 2020).
requirements, as well as providing guidance for auditors carrying out their performance audit functions under s 16 of the Public Audit Act.126
[148] The version of AG-5 operating at the time of the performance audit relevantly sets out that an auditor shall “obtain sufficient appropriate evidence” and shall “evaluate the evidence obtained … as a basis for clearly expressing the auditor’s findings, recommendations, and, where appropriate, conclusions”.127 Auditors are also to approach their work with “professional scepticism”.128 The relevant provisions of AG-5 on which Gama relies state:
20.The auditor shall obtain sufficient appropriate evidence on which to base the auditor’s findings, recommendations, and, where appropriate, conclusions.
21.The auditor shall evaluate the evidence obtained from the procedures performed as a basis for clearly expressing the auditor’s findings, recommendations, and, where appropriate, conclusions.
22.Where appropriate, and whenever the auditor is proposing to criticise a person or organisation, the auditor shall provide relevant parties with sufficient information in order to seek comments on:
(a)the factual accuracy of the auditor’s findings, recommendations, and, where appropriate, conclusions;
(b)whether any facts material to the auditor’s findings, recommendations, and, where appropriate, conclusions have been omitted; and
(c)the balance and fairness of the auditor’s findings, recommendations, and, where appropriate, conclusions.
23.The auditor shall consider any feedback received from relevant parties from whom comments have been sought and, if necessary, make such changes to the auditor’s findings, recommendations, and, where appropriate, conclusions that the auditor considers necessary to achieve a report that is factually accurate and also balanced and fair.
A lack of evidence to support the amendment to Recommendation 4?
[149] The first strand of Mr Mijatov and Ms Kenner’s argument is that the change from the initial recommendation to the final form of Recommendation 4 was not a
126 Controller and Auditor-General Auditor-General’s Auditing Standard 5 – Performance Audits, Other Auditing Services and Other Work Carried Out By, Or On Behalf Of, the Auditor-General (Office of the Auditor-General, March 2020).
127 Cls 20–21.
128 Cl A20.
change that was supported by evidence. Mr Mijatov says that the impetus for the change—which introduced in parentheses that MSD could choose to send targeted communications to certain groups of subsidy recipients—was that Audit NZ had changed its original recommendation in its second draft interim financial report for the 2019/2020 financial year, and the performance audit team had simply agreed with MSD to mirror the same wording. However, counsel for Gama argued the amendment to the recommendation first made by Audit NZ itself made no sense. In addition, MSD had simply pushed back consistently at the suggestion that it should be writing to all subsidy recipients, but in doing so it had either not raised resourcing implications, or if it had, it failed to provide evidence to support its claims sufficient to meet the requirements of AG-5. The available internal documents on which Gama relies suggest, according to Mr Mijatov, that rather than being persuaded by MSD’s feedback on the draft recommendation, the Auditor-General and his office remained distinctly unpersuaded that the recommendation should be watered down, but went ahead and did so anyway.
[150] Ms Kenner argued that the requirements of independence are mandatory. She accepted that draft recommendations made by the Auditor-General can change, but that can only occur where evidence supports the amendment, or for “error correction”. The first draft recommendations in the various reports did not “come out of nowhere”. In counsel’s submission, to make the original recommendation requiring communication with all wage subsidy recipients, the Auditor-General must have based the recommendation on an assessment of evidence, as suggested in the evidence of Mr Hoy. Those recommendations are presented to an audited agency for their comment and will be final subject only to the feedback received. If an auditor was simply to reconsider the evidence and “change their opinion after comment”, that would require thorough documentation which shows why the evidence no longer supported the original recommendation. Against those process requirements, there is simply no evidence, according to Gama, that would justify the changes in the draft.
[151] Ms Kenner submitted that the first time MSD asserted the draft recommendation would strain its resources was on 1 December 2020, but no evidence to substantiate the claim was ever provided. The Auditor-General’s discovery had not revealed any exploration by his office of what the claimed resource implications were,
Further, the respondent’s reasons for changing the recommendation need to be verified with sufficient evidence, and there was “no record that has occurred in this case”.
[152] By implication, the inference the Court was invited to draw was that the Auditor-General and his officers agreed to water down the recommendation once faced with stern and consistent resistance from MSD and as a result of indifference to their professional obligations. In short, the Auditor-General had chosen the path of least resistance.
[153] Counsel for Gama argued that all of this indicates a lack of independence on the part of the Auditor-General. Mr Mijatov’s principal point was that, based on the evidence before it, the performance audit team knew the most effective recommendation would be for MSD to contact all recipients of the wage subsidy to confirm their eligibility, made that recommendation, and retreated from it unlawfully. Gama relied extensively on the affidavit of Mr Hoy to suggest that Recommendation 4 is inadequate in this way.
[154] Having considered the evidence, I do not agree with Gama that the Auditor- General has failed to act independently or has otherwise acted unlawfully. There are three reasons for this conclusion.
[155] First, Gama equated a claimed lack of evidence to support the change to the recommendation with a failure to act independently. However, while a complete lack of evidence may give rise to a challenge for irrationality, it does not follow logically that there has been a lack of independence—and a breach of s 9—in making the decision.129 At a more fundamental level, Gama’s interpretation of the requirements for “evidence” in AG-5 in the context of a performance audit enquiry lacks a sense of realism. A performance audit is a very different creature to a financial audit. Performance audits are not designed to uncover evidence that appropriated public moneys have been expended strictly in accordance with the appropriation. Rather, while it is called an “audit”, its purpose is to investigate and report on whether the expenditure of public funds has been effective in achieving desired outcomes.
129 In the same way that one may fail to act independently but nevertheless make a well-reasoned and evidentially supported decision.
Recognising the different nature of the enquiry, while AG-5 refers at cl 20 to the need for a performance auditor to obtain “adequate evidence”, cl 23 recognises that “feedback” rather than “evidence” from an audited entity is an appropriate reason to modify recommendations and findings:
The auditor shall consider any feedback received from relevant parties from whom comments have been sought and, if necessary, make such changes to the auditor’s findings, recommendations, and, where appropriate, conclusions that the auditor considers necessary to achieve a report that is factually accurate and also balanced and fair.
(emphasis added)
[156] In undertaking a performance audit, natural justice and the need for efficacy of the proposed recommendations require the Auditor-General to invite the audited entity to provide its comments. This does not require the Auditor-General, as Gama suggested, to undertake a detailed forensic examination of every assertion or submission made by senior officials of government agencies to test their veracity.
[157] Given there were almost 500,000 wage subsidy recipients, and that MSD’s core business is not operating an integrity check for the scheme, it does not require much beyond common sense to conclude that resourcing enquiries of all recipients might create a strain on the Ministry’s resources, and result in a public trade-off in terms of welfare and public benefit outcomes.
[158] Second, it is clear the Auditor-General’s office did have evidence indicating MSD might not be able to follow up with all recipients of the wage subsidy scheme without compromising on its other work. This in turn supported the respondent’s conclusion that a recommendation regarding post-payment follow up checks could acknowledge a targeted approach depending on MSD’s available resources.
[159] The evidence indicates that in relationship meetings between the Office of the Auditor-General and MSD staff, MSD representatives outlined that writing to all recipients would not be “straightforward” because, for example, it would generate additional calls to service centres which would take additional resourcing. MSD representatives told Office staff that there had been targeted and random audits to check for potential fraud, and “Random weren’t bringing up much”. MSD also advised
Office staff that they “Can never investigate everyone committing fraud in the system
… So need to be targeted … If you looked at everything, you’d be going forever”.
[160] Importantly, the performance audit team’s preliminary findings noted that the use of staff from across MSD to support wage subsidy work had impacted its capacity to perform other work, which included diverting resources from hardship and income support services. Use of staff for wage subsidy work also meant that some existing investigations into benefit fraud had to be put on hold.
[161] In addition, as the performance audit report disclosed, 7052 post-payment reviews had been conducted by MSD. 3751 of these were random and 3301 had been targeted. The performance audit team received evidence that 1412 refunds were required, mainly as a result of the targeted rather than random reviews.130 So MSD’s feedback was that targeted enquiries had been more effective than random ones. This approach was also consistent with the Deloitte review’s recommendations, that MSD increase communication with wage subsidy recipients and stakeholders, which could be done through “driv[ing] appropriately messaged and targeted proactive communication with subsidy recipients …”.
[162] Finally, I find Mr Press’ evidence on the level of information available to the performance audit team and its relevance to amending the recommendation telling. His evidence was that targeting a combination of low risk but high consequence applicants (in this case, larger applicants) as well as high risk but lower consequence applicants (for example, sole traders) was a reasonable approach for MSD to take. Early on in the process, it was clear that there were particular groups of payment recipients who MSD and Inland Revenue considered were at “risk” of abusing the wage subsidy scheme. MSD’s own internal work indicated multiple times that sole- traders were a particular area of concern, as there had been instances of applicants deliberately registering as a sole trader just to receive a benefit. Other particular “risk” grounds included those who had been referred to MSD for potential abuse by another agency, through complaints, or by some other risk indicator such as multiple applications by the same employer, or applicants that had previously repaid a subsidy
130 Performance audit report, above n 19, fig. 10.
payment.131 The reason why MSD staff conducted pre-payment reviews of applicants with over 80 employees was because of the risk associated with the high value of applications from larger employers; larger entities attracted a greater proportion of funding as a group.132
[163] As Mr Press’ evidence confirms, MSD sought changes to other draft recommendations. The performance audit team did not make them because it did not consider they were appropriate. Rather than deferring to MSD’s preferences, the evidence satisfies me that the approach of the Auditor-General’s office was robust and plainly independent. Mr Press also confirmed that MSD’s resourcing “had been a regular item of discussion during the audit” and had been discussed ahead of the “preliminary findings discussion”.
[164] It follows that I cannot accept Gama’s submission that the Auditor-General did not have the evidence to support Recommendation 4. I accept Mr Mijatov’s point that auditors must approach evidence with a level of “professional scepticism” but, as explained by Mr Press, the general approach for auditors is to triangulate evidence (including testimonial evidence) against other sources. Where this is not possible, the audit team uses their professional judgement and knowledge of wider context to decide how much reliance to place on a particular piece of evidence. Against this background, there is nothing which raises concerns that the performance audit team failed to test MSD’s claims, lacked professional scepticism, or failed to act independently in terms of s 9.
[165] Third, I accept Ms McDonald’s submission that the evidence of Mr Hoy—and indeed Gama’s case more broadly—mischaracterises Recommendation 4. As Mr Press deposed, Recommendation 4 does not preclude MSD from writing to all recipients of the wage subsidy if it wishes to do so. It also does not require MSD to write to only larger recipients, or only risk-indicated recipients. Mr Press explained, and I accept, that Recommendation 4 is drafted in the way it is to achieve a particular balance between the Auditor-General’s position and the fact that the Auditor-General must allow MSD discretion in the operationalisation of any recommendations. It is not for
131 Performance audit report, above n 19, at [4.24] and [4.83].
132 Performance audit report, above n 19, at [4.20].
the Auditor-General to seek to prescribe how a state agency should achieve his recommendations, because those recommendations are primarily directed to substantive outcomes, rather than the process by which they are achieved.
[166] I also consider that Gama has misstated the nature of the amendment to Recommendation 4. While the original recommendation in the draft performance audit report did not contain the parenthetical qualification on the pool of recipients to whom MSD might send an eligibility enquiry, there was much more to the recommendation than just that. The original recommendation in the preliminary audit report sent to MSD for its comment was in the following terms, which I repeat again for ease of reference:
iii.MSD prioritises and resources the remaining enforcement work necessary to take appropriate action where there has been abuse of the Wage Subsidy. As a minimum this should include:
•seeking written confirmation from applicants (who have not refunded payments to date) of their compliance with the eligibility criteria and obligations;
•testing the reliability of a sample of the findings from its post payment assurance work (referred to as “audits” by MSD) by obtaining and using documentary evidence where the post payment assurance work primarily relied on verbal discussions with employers and employees; and
•taking prosecution action where possible and necessary to recover Wage Subsidy funds and/or to hold people to account for criminal behaviour.
[167]The final recommendation was in the following terms:
Recommendation 4
In relation to the Wage Subsidy Scheme, we recommend that the Ministry of Social Development prioritise remaining enforcement work, including:
•seeking written confirmation from applicants (which could be targeted towards larger or risk-indicated applicants) of compliance with the eligibility criteria and the obligations of receiving the subsidy; and
•pursuing prosecutions to recover funds and/or to hold businesses to account for potentially unlawful behaviour.
[168] A comparison of the two versions reveals, first, that the primary recommendation—that MSD should prioritise remaining enforcement work—remains largely unchanged. Second, while the words in parentheses that have sparked Gama’s challenge to the Auditor-General’s work appear in the final draft, importantly, the second bullet-point in Recommendation 4 is in much stronger terms than the original version of the recommendation. Initially, the Auditor-General had stated that MSD should only take prosecution action “where possible and necessary”. This qualification is removed from the Recommendation 4, which provides that MSD should pursue prosecutions “to recover funds and/or to hold businesses to account”. So, overall, it is not accurate for Gama to suggest that the Auditor-General’s final recommendation was “watered down”, demonstrates a lack of independence, or evinces a lack of fortitude when dealing with an audited agency.
[169] Mr Mijatov also argued that, even if there was a lawful basis to amend the draft recommendation, the Auditor-General should have at least recorded the preliminary view that MSD should write to all recipients. To the extent Gama makes this submission on the basis that a “refusal to implement a stronger recommendation should be memorialised and publicly aired”, I do not agree. The evidence of Mr Press and Mr Whittal is that there is no benefit in the office of the Auditor-General making a recommendation that it believes MSD is unable to operationalise. That would amount to the Auditor-General making a recommendation that is not grounded in evidence. MSD is not required to comply with any recommendation made by the Auditor-General. Recommendation 4 does not preclude MSD from writing to all recipients of the wage subsidy, so there is still the opportunity for the public and Parliament to ask whether MSD has done so and, if not, why not. Recommendation 4 does not limit MSD’s accountability for its implementation of the wage subsidy scheme. And, where Gama seeks to make public the circumstances surrounding the change in wording for Recommendation 4, these proceedings in themselves will suffice to meet this end.
Has the Auditor-General unlawfully fettered his discretion?
[170] The distinction between legally permissible constraints on the exercise of a discretion and those that cross the line and become an unlawful fetter has been described by the Court of Appeal as follows:133
When a statute confers a discretion on a particular person, it cannot be altered by means other than statutory amendment. The adoption of policy guidance might be administratively convenient for a decision maker, and can advance rule of law values such as consistency and certainty in decision making. However, a policy which guides the exercise of a discretion will inevitably fetter that discretion to some extent. If that policy guidance crosses the line between legally acceptable limits on the exercise of discretion and those which are not legally acceptable, it “fetters” the discretion and is unlawful.
[171] There are three categories generally accepted in practice as unlawful constraint on decision making. Only one is relevant to the present case: that is “acting under dictation”, where the decision-maker allows someone else in substance to exercise or control a discretion.134 The essential feature of this form of illegality is that the authority with the power to decide did not exercise the power. In this case, the appellant alleges that the Auditor-General fettered his discretion by acting under the dictation by MSD.
[172] Gama argues the Auditor-General simply deferred to the wishes of MSD in changing the wording of the initial recommendation, which itself shows an unlawful fettering of discretion. This claim fails on the facts for the same reasons that have led me to dismiss Gama’s claim that the Auditor-General failed to act independently.
[173] As noted by Mr Mijatov himself, staff in the Office of the Auditor-General, including Mr Ryan, at various points expressed scepticism about MSD’s advice that it would not be able to write to all recipients without resource implications, and considered a targeted approach was a more effective use of public resources.
[174] For instance, in feedback provided during discussions on the scope of the financial audit for the year ending June 2020, Mr Whittal repeatedly raised with others
133 Criminal Bar Association of New Zealand Inc v Attorney-General [2013] NZCA 176, [2013] NZAR 1409 at [118].
134 Graham Taylor Judicial Review: A New Zealand Perspective (4th ed, LexisNexis, Wellington, 2018) at [15.71].
in the financial audit team that there was a need for a line of inquiry into post-payment communication by MSD about wage subsidy recipient eligibility. At a Steering Committee meeting, the Auditor-General questioned whether it would be possible to escalate within MSD the question of the Ministry sending a positive confirmation letter on eligibility to all recipients of the subsidy. At the same meeting, the Steering Committee resolved to tell MSD that the performance audit team “expects them to do their own work to get evidence of eligibility for any of their entities that claimed the wage subsidy”. In other words, the issue of post-payment confirmation was clearly an important one for members across the two audit teams. It is not accurate to suggest that the exercise of the discretion to make Recommendation 4 was determined by MSD. Mr Press’ evidence confirms otherwise and is supported by the documentary record that not all amendments MSD sought were accepted, but the audit team did accept MSD’s reasons for suggesting other changes.
[175]For these reasons, Gama’s third ground of review must be dismissed.
Third issue: did the Auditor-General act unreasonably and has there been an unreasonable delay?
[176] The remaining issue is whether the Auditor-General acted unreasonably in a public law sense and incorporates two grounds of review. Under the first, Gama claims the decision to alter the draft recommendation was substantively unreasonable. Under the fourth ground of review, Gama argues the continued delay in recommending that MSD write to all recipients of the wage subsidy is unreasonable.
[177] Neither claim is tenable given my conclusion that the decisions to amend the draft recommendations were supported by evidence, the exercise of independent judgment and were open to the respondent to make. Both the first and fourth grounds of review must therefore be dismissed.
Relief
[178] As I have found that none of the grounds of review are made out, it is unnecessary to address relief in any detail. However, given the status of the Auditor-General as an Officer of Parliament who reports to Parliament, and the nature
of the power under review (the making of recommendations to the executive government and the House), I would not consider it appropriate to grant mandatory relief as Gama sought. In particular, while it might have been open to make declarations as to lawfulness, I am not persuaded that it would be appropriate to quash the recommendations in the audit reports or direct reconsideration. In my view, as an Officer of Parliament, it would be for Parliament to determine the appropriate response to the Court’s declaration.
Conclusion and result
[179]The application for judicial review is dismissed.
[180] The respondent would ordinarily be entitled to costs. My preliminary view is that costs on a 2B basis are appropriate. I would certify for second counsel.
[181] I encourage the parties to reach an agreement on costs. If they are unable to do so, brief memoranda may be filed, and I will deal with the issue on the papers.
Isac J
Solicitors:
Macalister Mazengarb, Wellington for Applicant
1
7
1