|
|
Home
| Databases
| WorldLII
| Search
| Feedback
University of New South Wales Law Journal Student Series |
A CLOUDED ‘VISION SPLENDID’: THE LIMITS OF ADMINISTRATIVE REVIEW OF EXECUTIVE ACTION IN THE CONTEXT OF THE ROBODEBT SCHEME
FIONA SHAH*
Justice Murphy recently referred to the Robodebt Scheme as a ‘shameful chapter’ in Australian public administration, involving the unlawful raising of welfare debts by the Commonwealth against social security recipients.[1] It was precisely such concern about the deleterious impacts of inadequate checks on executive power that led Australia to devise the 1970s reforms that came to be known as the ‘new administrative law’.[2] This article examines the effectiveness of ‘new administrative law’ accountability and review mechanisms in holding the executive to account in its administration of the Robodebt Scheme. It argues that the Administrative Appeals Tribunal and the Commonwealth Ombudsman, in particular, are given powers to protect against administrative injustice yet, in practice, were unable to hold the executive accountable for an unlawful regime. Part II of this article summarises the history and the illegality of the Scheme. Part III identifies serious deficiencies in the design of merits review performed by the Administrative Appeals Tribunal. Part IV follows with the limitations of the Commonwealth Ombudsman in fulfilling its proactive, ‘system-fixing’ role. Part V turns to examine whether the deficiencies of merits review will be rectified by the Administrative Review Tribunal, the body which will soon replace the Administrative Appeals Tribunal. The article concludes in Part VI with a call to reform the ‘new administrative law’.
I INTRODUCTION
Introduced as part of the Kerr Committee’s ‘new administrative law’ system, the Administrative Appeals Tribunal (‘AAT’) and the Commonwealth Ombudsman (‘Ombudsman’) were lauded as the ‘vision splendid’ of the means by which citizens could test decisions made by the executive.[3] Located within the executive branch of government, these mechanisms practise a ‘self-administered’ form of accountability.[4] Whereas the AAT is equipped with ‘harder-edged’ powers of merits review, the Ombudsman performs an ‘intermediate’ form of accountability as a body vested with ‘soft’ powers of investigation, conciliation and reporting with recommendations.[5]
Close scrutiny of the Robodebt Scheme (‘Robodebt’), however, reveals a glaring failure of these mechanisms to be a sufficient check on executive power; notably, administrative review in ‘theory’ vastly differs from its ‘practice’. The AAT’s narrow jurisdictional remit, ‘two-tier’ structure of review, inconsistent reasoning, and ‘blunt’ effect of remedies in practice undermined the role of merits review as an accessible, fair means of challenging executive decisions. Crucially, accountability mechanisms are only effective if they ‘transform’ the executive culture.[6] While vested with large powers to systemically review executive action, the Ombudsman’s reluctance to exercise these powers eschewed its ‘system-fixing’ potential.
Ultimately, Robodebt’s unlawfulness went unnoticed and unremedied subject to merits review and the Ombudsman alone; the limitations of these mechanisms perpetuated the executive’s deliberate ignorance of the Scheme’s illegality. It is, therefore, apt to raise the question of whether the Kerr Committee’s ‘vision splendid’ has faded.
II HISTORY OF THE ROBODEBT SCHEME
‘We’ll find you, we’ll track you down and you will have to repay those debts and you may end up in prison’ – Alan Tudge, former Minister of the Department of Human Services.[7]
The Online Income Compliance Program, known colloquially as ‘Robodebt’, was an automated system for raising and recovering debts that was proposed and administered by the Department of Human Services (‘DHS’) (now, Services Australia). This Scheme was rolled out in various iterations for four years: the ‘Pay As You Go’ Manual Compliance Intervention, which was the pilot program implemented from early 2015 to 2016; the Online Compliance Intervention, which was run from 2016 to 2017; the Employment Income Confirmation, run from 2017 to 2018; and the Check and Update Past Income, run from 2018 until the suspension of the Scheme in late 2019.[8] Introduced in the 2015–16 Budget, Robodebt was a crucial element of the Coalition government’s widely publicised mission to control the ‘budget black hole’ and to achieve a fiscal surplus by reducing public spending.[9] It was anticipated to recover $1.7 billion over five years from ‘overpayments’ made by Centrelink to social security recipients, in circumstances where those recipients had incorrectly declared their income.[10] The objective of the Scheme was to strengthen the ‘integrity’ of welfare payments and, in essence, to ‘cop people who are going to rort [the social security] system’.[11]
Each iteration of the Scheme used automated data-matching algorithms to raise social security overpayment debts. The system involved:
1. Using ‘Pay As You Go’ income data kept by the Australian Taxation Office (‘ATO’) and evenly apportioning that income over fortnightly increments in the review period to determine the welfare recipient’s notional or assumed fortnightly income;
2. Comparing the notional or assumed fortnightly income with the actual fortnightly income reported by the welfare recipient to the DHS in order to detect discrepancies between these two sets of data;
3. Where a discrepancy arose, a letter was automatically sent to the welfare recipient. The letter acknowledged the discrepancy, required the person to confirm or update their income data and stated that, if they did not do so, the Commonwealth would raise a debt.
Unless the recipient ‘disproved’ the alleged debt by producing the relevant fortnightly pay slips, Centrelink would presume that: the averaged ATO data accurately reflected the recipient’s actual income; given the discrepancy, the recipient underreported their income each fortnight; and, thus, were not entitled to the social security payments received and owed a debt due to the Commonwealth. Ultimately, ‘20,000 debt discrepancy notices per week were generated [using Robodebt], compared to a previous average of 20,000 income data-match discrepancies per year when manual verification processes were employed’.[12]
This automated system, however, was beset with errors. The algorithm produced incorrect, inflated and, in some instances, entirely false debt claims against vulnerable welfare recipients.[13] The case of Deanna Amato, who sought judicial review of Centrelink’s purported exercise of debt collection powers, is demonstrative of this fundamental flaw. Centrelink concluded that Ms Amato had wrongly received social security benefits and, therefore, owed a debt to the Commonwealth in the sum of $2,924.28 – when, in fact, she owed merely $1.48 (a debt which was ultimately waived).[14] Ms Amato’s case, however, was one of many. The use of an automated debt-raising and collection system resulted in systematic miscalculations, amplified by the scale of implementation to hundreds of thousands of debtors.[15] Over the course of the Scheme, the executive unlawfully asserted debts ‘totalling at least $1.763 billion against approximately 433,000 Australians’.[16] The enforcing of wrongful claims of debt gravely affected the livelihood of vulnerable low-socioeconomic debtors, with reports of severe financial and mental distress as well as suicides.[17]
The illegality of the Scheme has been well-documented in the literature.[18] In summary, the core legal deficiencies of Robodebt were twofold. Firstly, Robodebt departed from the correct interpretation of income calculation provisions in the Social Security Act 1991 (Cth) (‘Social Security Act’). The algorithm falsely extrapolated and relied on a fortnightly average of ATO income data. It did not reflect the legislative requirement for entitlements to be calculated based on the precise amount of income earned each fortnight.[19] Ordinarily, it might be expected that the moral and practical gravity of alleging a debt would impose a ‘heightened’ degree of ‘reasonable’ satisfaction, which ‘should not be produced by inexact proofs, indefinite testimony, or indirect inferences’.[20] Yet, as Terry Carney observes, ‘it was precisely such inexactitude of debt, indefinite character of proof, and indirect inference on which the executive relied using Robodebt’.[21] The regime was not based on ‘probative material’ and would have failed to meet the standard of proof required of Centrelink to establish a debt at common law.
Secondly, the Social Security Act casts the burden of proving the existence of a social security debt on the Commonwealth.[22] There is no evidentiary ‘onus of proof’ borne by a welfare recipient;[23] rather, the Commonwealth bears a ‘practical onus’ to prove that a debt exists.[24] If unable to do so, the ‘status quo prevails’; no debt exists in the absence of proof. However, if a welfare recipient could not disprove a supposed overpayment, Robodebt automatically raised a ‘debt’ against them. By requiring citizens to refute the false inference, the Scheme unlawfully ‘reversed’ this onus of proof.
Alarmingly, at the time the Scheme was developed into an New Policy Proposal, approved by Cabinet and announced as part of the Budget measure, the Department of Social Services had already received advice that Robodebt was unlawful.[25] Ultimately, the Scheme rolled on – unexposed and uncorrected – until it was struck down by the Federal Court in Amato v Commonwealth (‘Amato’) as an illegal and invalid automation algorithm, four years after it began.[26] The Minister for Government Services acknowledged liability of $721 million, in respect of 470,000 debts, relying on Amato.[27] However, as Justice French (as his Honour then was) once acknowledged, judicial review is at the far ‘end of the line of official review’ and should not be assumed as the best means of attaining administrative justice.[28] Robodebt’s demise, for all of its seeming inevitability, took far longer than it should have. Robodebt was repeatedly subject to merits review at the AAT as early as 2016, while the Ombudsman launched an inquiry culminating in reports issued in 2017 and 2019. The question, therefore, that this article asks is why these mechanisms failed to uncover the Scheme’s obvious illegality and, in doing so, fail to hold the executive accountable.
III MERITS REVIEW
Whereas the role of courts is to detect and correct errors of law, the AAT reviews the substantive merits of executive decisions.[29] Standing in the original decision-maker’s shoes, the AAT considers the matter ‘afresh’[30] and makes the ‘correct or preferable’ decision.[31] The AAT seeks to resolve disputes expeditiously, without the technicality and formality of courts;[32] it has broad inquisitorial powers[33] and may consider evidence that the primary decision-maker did not.[34] In essence, the AAT was the Kerr Committee’s ‘Trojan horse’;[35] the AAT was conceived to remedy, from within the executive itself, the inadequacies of judicial review as a redress for administrative injustice and, thus, to bring the rule of law to the level of the executive.[36] Nonetheless, these purported ‘strengths’ of merits review failed, in practice, to hold the executive accountable for the unlawfulness of Robodebt.
A Jurisdiction: Inheriting Bond’s Finality Requirement
Tribunals may only hear matters that fall within their statutory remit; pursuant to sections 3(3) and 25(1) of the Administrative Appeals Tribunal Act 1975 (Cth) (‘AAT Act’), the jurisdiction of the AAT is limited to merits review of administrative ‘decisions’ made under an enactment. Notably, very few Robodebt cases exited the DHS’ internal review process and were appealed;[37] only five out of every 10,000 Robodebts were heard by the AAT.[38] One possible reason is that a narrow statutory interpretation of the phrase ‘decision’ has restricted its jurisdictional ambit.[39]
The High Court in Australian Broadcasting Tribunal v Bond (‘Bond’) ruled that reviewable ‘decisions’ under the Administrative Decisions (Judicial Review) Act 1977 (Cth) (‘AD(JR) Act’) must generally be ‘final or operative’.[40] As the AAT Act defines ‘decision’ in similar terms to AD(JR) Act,[41] Bond’s ruling has been assumed to apply to the AAT.[42] While a uniform interpretation accords with the Kerr Committee’s vision of a ‘comprehensive and integrated system of administrative law’,[43] the AD(JR) Act extends judicial review’s jurisdiction to ‘conduct’,[44] which covers procedural decisions and preliminary steps in decision-making.[45] Avoiding an overlap with this proviso thereby justified the Court’s narrow reading of ‘decision’ in Bond.[46] However, the AAT Act confers no such complementary jurisdiction for ‘conduct’ on the AAT; the AAT’s jurisdictional remit is strictly limited to ‘decisions’.[47] If ‘conduct’ is amenable to judicial review, it would seem sensible that merits review of intermediate decisions should also be permissible. However, this view is not accepted universally,[48] whereby ‘the impact of Bond is ... unclear as far as AAT applications are concerned’.[49] Consequently, interim executive decisions which have not yet matured into a ‘final’ decision are unlikely caught by the AAT’s jurisdictional net and are, therefore, unreviewable.[50]
Bond has been relied on to refuse merits review in various instances,[51] including Robodebt. The decision chain for administrative review (of a decision by the DHS to issue a debt notice under section 1229 of the Social Security Act) is comprised of many ‘links’ or ‘steps’.[52] First, in protest of the debt, an applicant lodges a complaint and the original decision-maker reconsiders the decision. Terry Carney explains that Robodebt ‘effectively converted [this] first [step] into a ‘virtual’ reconsideration ... because [it] insisted on exclusive engagement with the online compliance system’.[53] If unsatisfied, the applicant may request mandatory internal review by an Authorised Review Officer.[54] If still unsatisfied, only then is the decision of the Authorised Review Officer reviewable at the AAT.[55]
Centrelink’s bureaucratic culture, however, was to characterise protests lodged by applicants as only a mere ‘complaint’, ‘concern’ or ‘request for online assistance’; indeed, ‘everything, but a decision’ as to the issuance of the debt notice.[56] Sometimes complainants were diverted to a less formal ‘reassessment’ process, in which they received a letter stating their current outstanding debt amount – not a decision letter that explained the ‘reassessment’ or change to the debt amount/period.[57] In other words, Centrelink did not consider the ‘first step’ of review to be complete. This policy of Centrelink is not explicit, but vaguely alluded to, in the Royal Commission’s Report into the Robodebt Scheme.[58] Consequently, many cases were improperly dissuaded from (and not amenable to) internal review by an Authorised Review Officer, which deprived the AAT of jurisdiction for merits review.[59]
Ultimately, the AAT’s very purpose as an avenue of review free from the technicalities of judicial review is defeated. As Justice Lockhart once cautioned, ‘no narrow or pedantic approach is called for in determining whether a decision falls within the scope of review by the AAT’.[60] It must be acknowledged that the jurisdictional barriers in merits review and judicial review serve different purposes.
B Transparency of Reasoning: Two-Tier Review
The AAT was amalgamated with the Social Security Appeals Tribunal, Migration Review Tribunal and Refugee Review Tribunal in July 2015 to create a ‘one-stop shop’ for merits review without ‘unnecessary layers of bureaucracy’.[61] The amalgamated AAT reviews Centrelink decisions in its Social Services and Child Support Division (‘the First Tier’) and then, if appealed, in its General Division (‘the Second Tier’).[62] The First Tier repeatedly heard complaints concerning Robodebt’s unlawfulness[63] and made several rulings adverse to the executive.[64] However, the First Tier lacks fundamental procedural safeguards; unlike the Second Tier, for expediency and cost-saving purposes,[65] hearings are privately held[66] and reasons are not published.[67] Such lack of transparency presents an insidious threat to the rule of law; by administering justice ‘in secret’, the Scheme was insulated from public scrutiny and Centrelink could continue issuing debts unlawfully.[68] Indeed, no decision in the Second Tier has ever been published with reasons in regards to Robodebt.[69] Further, the AAT lacks protections against ‘gaming of review’ by the executive; to avoid an open hearing in the Second Tier, Centrelink deliberately did not appeal First Tier Robodebt decisions.[70]
There are no cogent policy reasons for two-tiered review,[71] it merely exists as the legacy of the absorption of the Social Security Appeals Tribunal by the AAT.[72] Notably, the AAT’s objectives of ‘economical, informal, quick’ and ‘fair, just’ merits review are ‘internally inconsistent’.[73] Without sufficient resources to publish decisions, the First Tier cannot provide transparent, ‘fair’, and ‘just’ merits review.[74] Resources are instead diverted away from the First Tier and towards the Second Tier. However, few cases ever reach this appellate stage and, out of the 174 Robodebts that did, 92 per cent were settled by consent.[75] With no guidance on which objective takes precedence, Parliament has created a ‘conundrum’ for tribunals; the tension between the objectives means the AAT cannot sufficiently scrutinise and hold the executive accountable.[76]
C Inconsistency of Reasoning
AAT decisions have only a ‘quasi-law’ status and are not binding as precedent.[77] They are, thus, predisposed to inconsistency.[78] While AAT members are expected to follow decisions with similar factual circumstances,[79] not all members uniformly adopted the reasoning of previous Robodebt decisions.[80] As early as 2016, numerous First Tier decisions set aside debts raised by way of income averaging.[81] Most notably, the decision of Member Terry Carney on 8 March 2017 held that the methodology of income averaging was itself unlawful; the algorithm defied ‘simple mathematics’ and provided an insufficient evidentiary basis for the calculation of overpayment debts.[82] This decision, and other decisions that were adverse to the DHS, reached their conclusions using considered and detailed legal reasoning. Yet, some decisions by the AAT affirmed the appropriateness of income averaging. There were 114 of such decisions by the First Tier over time and, intriguingly, none of them applied any legal principles or relevant legislation in their reasoning.[83] A significant inconsistency consequently arises; even though earlier decisions held that the foundational algorithm of Robodebt was unlawful (in other words, considered illegality in a general sense), other AAT members nonetheless ruled that debts were lawfully issued ‘in the circumstances’. Adverse decisions were often distinguished on the basis that they ‘obviously turn[ed] on particular facts of [the] particular matter’.[84] Overall, the variation or set aside rate for debts at the First Tier was 42 per cent.[85]
The degree of inconsistency in reasoning was highlighted when those members of the AAT, who overturned many of Centrelink’s decisions, were not reappointed.[86] With five separate rulings against the executive, Member Terry Carney was not reappointed upon completion of his term – despite having sat on the AAT and its predecessor for nearly four decades.[87] When Member Michael Manetta was benched, the AAT’s Deputy President explained that she ‘was concerned at the high level of [Manetta’s] decisions against the department (around 65%) as out of kilter with most of the other members’ and ‘wanted consistency of decision-making across the division’.[88]
Ultimately, because some First Tier decisions had endorsed the practice of income averaging, the executive ‘felt free’ to reject the AAT’s reasoning in decisions that were adverse to it.[89] In the words of Justice Brennan, ‘inconsistency is not merely inelegant: it brings the process of deciding into disrepute, suggesting an arbitrariness which is incompatible with commonly accepted notions of justice’.[90]
D Inadequacy of Remedies for Individual Rulings
The primary redress offered to applicants in the First Tier was to set aside Centrelink’s decision and to remit the matter for redetermination, often accompanied with directions that prohibited the use of income averaging.[91] However, Centrelink routinely ignored the AAT’s directions and reconsidered the matter using its own interpretation of the law. The 8 March 2017 decision, for instance, directed the DHS to recalculate the welfare recipient’s debt using fortnightly salary records. Nonetheless, the DHS proceeded to use income averaging on the basis that, in the circumstances, its ‘evidence gathering power [was] exhausted’; for that reason, it claimed the ‘direction is wrong in law and can be ignored’, and it was ‘still open’ to raise a debt based on income averaging ‘regardless of what the [AAT] said’.[92] In another case, the Department of Social Services had effectively given ‘an ultimatum’ that it will ‘ignore a decision of [the] Tribunal that [it] does not agree with’, ‘refuse to implement it’ and ‘apply the law as [it] sees fit’.[93] It follows that merits review failed to procure effective redress for most applicants.
Indeed, one must look at the ‘full run of cases’ to determine whether a system is fair.[94] The focus of merits review, however, is on the immediate, individual case.[95] The AAT has no formal process for group-based merits review, such as for Robodebt claims which involve a common algorithmic underpinning.[96] The AAT’s scrutiny of the executive was, therefore, reactive, piecemeal and ad hoc in nature;[97] the executive treated adverse decisions as no more than mere ‘pin-pricks’.[98] Consequently, procuring system-wide reform and distributive justice for Robodebt was beyond the remedial capacity of tribunals,[99] and the burden fell upon systemic review institutions, notably the Ombudsman, to mitigate the shortcomings of merits review.
IV COMMONWEALTH OMBUDSMAN
The Kerr Committee’s original vision for the Ombudsman was oriented towards ‘swatting flies [rather] than hunting lions’ – that is, to resolve individual complaints, not systemic issues.[100] However, systemic review of the executive has now become its more central function by virtue of its dual roles.[101] The Ombudsman is vested with ‘reactive’ powers to receive, investigate and remedy individual complaints retrospectively, and ‘proactive’ powers to conduct systemic inquiries on its own motion.[102] As the Ombudsman does not rely on individual complaints to emerge first, its system-fixing potential has been widely praised;[103] the institution has been described as bringing ‘the lamp of scrutiny to otherwise dark places’[104] where ‘other [mechanisms] ... have proved inadequate’[105] and ‘even over the resistance of those who would draw the blinds’.[106] Greg Weeks claims the Ombudsman has become so entrenched as an ‘integrity body’[107] that the question of whether it ‘uphold[s] the rule of law [has] become otiose’, with ‘no doubt that [it] is effective in holding the [executive] to account’.[108] However, close scrutiny of the Ombudsman’s culture and practices, during its own motion inquiry into Robodebt, suggests otherwise.
A Reticence Towards Pursuing Questions of Legality
The Ombudsman is vested with broad powers yet, curiously, does not appear to exercise them in practice. The judiciary’s view is that powers conferred upon the Ombudsman by statute are not to be read down, but construed liberally and given ‘ample meaning’.[109] Parliament endowed the Ombudsman with wide-ranging powers for the very purpose of increasing executive accountability and improving public administration.[110]
Indeed, the Ombudsman had sufficient power to investigate Robodebt’s lawfulness;[111] it may report on an executive decision that appears ‘contrary to law’ or based ‘wholly or partly on a mistake of law’.[112] As early as October 2016, the Ombudsman had received complaints about the Scheme, which it identified as an ‘issue of interest’.[113] In January 2017, an own motion inquiry was launched which, by April 2017, had come to end with a report titled ‘Centrelink’s Automated Debt Raising and Recovery System’ (‘the Report’).[114] However, the Ombudsman’s Report did not engage with this fundamental question of whether the Scheme was lawful.[115] Rather, the Report adopted a ‘presumption of legality’ approach,[116] relying on assertions by the executive that it was satisfied Robodebt was lawful.[117] Legality, the Ombudsman said, was ‘just one part of the broader consideration of the Scheme as required by the Ombudsman Act’.[118] It sought to finalise the Report ‘quickly’ due to Robodebt’s immediate impact on people.[119] Clearly, as observed with the AAT, the objectives of ‘quick’ and ‘just’ review simply do not align.
The Ombudsman is also bestowed with powers akin to a Royal Commission, including compelling the executive to produce information.[120] However, the Ombudsman treats these powers as a measure of ‘last resort’,[121] thus did not use them during Robodebt investigations.[122] Further, the Ombudsman’s discretion not to pursue matters where it considers external review to be more appropriate may have limited its preparedness to investigate legality.[123] Indeed, the Ombudsman thought it was ‘preferable that questions of law be ventilated in matters proceeding through the AAT’[124] and concluded that ‘complex questions’ of legality could ‘only be resolved with certainty by a court’.[125]
Ultimately, the executive was able to rely on the Ombudsman’s Report to assert that Robodebt’s issues only went ‘to the administration of the system as opposed to the fundamentals of the system’.[126]
B Separation of Powers: A Myth of Independence
The Ombudsman institution is the ‘traditional outlier’ of the administrative system.[127] The Ombudsman has been described as an ‘idea [with] no precise demarcation’,[128] a ‘graft on the Westminster system of government’,[129] and even a ‘constitutional misfit’.[130] It is conventionally classified as part of the executive – the reason being, put bluntly, is that ‘[it] fit[s] better here than anywhere else’.[131] Yet, an emerging view is that the Ombudsman institution forms part of a fourth, ‘integrity’ branch of government: regulatory institutions performing an independent ‘oversight’ role.[132]
Such an unusual constitutional position has impaired its accountability function.[133] During early stages of investigation, the Ombudsman provided the executive with a ‘draft report outline’, inviting them to ‘comment’ and ‘edit’.[134] The executive’s involvement in the drafting process meant it had, effectively, ‘co-writ[ten]’ the Ombudsman’s report.[135] Operating within the executive itself meant the Ombudsman’s ‘independence’ was ‘constitutionally constrained’[136] and, ultimately, compromised, in a manner Dicey warned would be incompatible with the rule of law.[137]
C Recommendatory Powers: Ombudsman as a ‘Blunt’ Tool
The Ombudsman lacks determinative power; it may only issue recommendations to the executive[138] and, if ignored, refer the matter to Parliament.[139] Nor can it injunct or compel an agency to take action, or award compensation.[140] The Ombudsman has thus been denigrated as a ‘toothless tiger’ – its influence rests, ultimately, on private persuasion and adverse publicity.[141] Any recommendations made or lessons learned from investigations can be easily neglected by the executive and simply ‘fall into history’.[142] Christopher Pollitt describes this phenomenon as ‘institutional amnesia’; the executive may employ a ‘wilful and strategic amnesia’ which sees them ‘deliberately consign lessons learned from past mistakes’.[143] Indeed, the Scheme continued for years after the Ombudsman’s investigations ceased.[144]
Greg Weeks suggests, however, that the Ombudsman’s strength is because of the absence of determinative power;[145] the Ombudsman’s ‘informal’ and ‘less-threatening’ nature makes agencies willing to assist investigations and abide by recommendations. Although, Howard Gadlin, formerly Ombudsman himself, reveals an important practical reality:
“[V]ery often at meetings and in discussion I hear an ‘Ombudsman don’t have power we just help people explore options’ sensibility which is ... a denial of the fundamental importance: a commitment to fairness and procedural justice which is at the heart of the Ombudsman role. The rejection of decision making authority and investigative powers has, over time, evolved into a frighteningly passive interpretation of neutrality .... ‘We don’t criticize, we don’t advise, we don’t suggest, we don’t evaluate, we don’t investigate’ has become a kind of mantra for some of us” (emphasis added in italics).[146]
Such a culture eschews any ‘system-fixing’ influence. Ultimately, Robodebt confirms long-standing concerns that the Ombudsman is a ‘blunt tool’; with only a mere ability to recommend, the Ombudsman ultimately failed to ‘fix’ the ‘system’. Instead, it enabled the executive’s ‘wilful, strategic’ ignorance of Robodebt’s unlawfulness.
V ABOLITION OF THE AAT: A SOLUTION?
Indeed, the call to revise the style of review conducted by the AAT is timely; on 16 December 2022, the government announced the abolition of the AAT. Beleaguered by political appointees, the AAT’s independence has been irreparably compromised;[147] between 2013 to 2022, partisan appointments rose from single digits to 40% of all appointments.[148] No doubt, the AAT’s failure to find fault with Robodebt – a signature policy of the Coalition – does little to advance the claim that it is independent from the government that appoints its members. The Administrative Review Tribunal Bill 2023 (Cth) (‘ART Bill’), accompanied by the Administrative Review Tribunal (Consequential and Transitional Provisions No 1) Bill 2023 (Cth), abolishes the Administrative Appeals Tribunal and replaces it with the Administrative Review Tribunal (‘ART’). This Part examines the new legislative scheme and queries whether it addresses the concerns raised previously in this article.
A Jurisdiction: Bond’s Finality Requirement
The ART must escape Bond’s ‘brooding presence’[149] and should, instead of drawing on the jurisprudence of judicial review, ‘carve out a philosophy of [its] own existence’.[150] However, no amendment to jurisdiction in this regard is apparent. Clause 12 of the ART Bill provides that ‘a decision is a reviewable decision if an Act or a legislative instrument provides for an application to be made to the Tribunal for review of the decision’. The definition of ‘decision’ remains unchanged.[151] Nor does Part 3, headed ‘Starting a Review’, operate with reference to ‘conduct’; it only refers to ‘decisions’. It emerges, therefore, that the ART Bill may be a missed opportunity to rectify the issues stemming from Bond.
B Transparency of Reasoning: Two-Tier Review
As Robin Creyke aptly concludes, the amalgamated AAT’s ‘story is of opportunities missed, rather than taken, to the [institution’s] detriment’.[152] The demise of AAT, however, comes with a clean slate. The ART Bill removes two-tier merits review in respect of social security matters, preserving only one tier of review. The Guidance and Appeals Panel (‘GAP’), established by Part 5 of the Bill, will in turn provide a channel to appeal a Tribunal decision. Moreover, the Tribunal must publish a decision, with reasons, if the Tribunal President considers that the decision ‘involves a significant conclusion of law’ or ‘has significant implications for Commonwealth policy or administration’.[153]
Albeit a welcomed change, the proposed single-tier structure must be refined in light of the following concerns. Single-tier review has the potential to enhance efficiency and incentivise the full engagement of parties in the first instance. Yet, these benefits will only accrue if there are mechanisms in place to ensure that the ART process is capable of considering all relevant issues at first instance.[154] Single-tier review may also risk becoming overly legalistic and intimidating for vulnerable, self-represented applicants. Importantly, referral of a Tribunal decision is subject to the President’s discretion and governed by a stringent criterion. Pursuant to clause 128 of the Bill, applicants may only appeal to the GAP if ‘the President is satisfied’ that the ART’s decision ‘raises an issue of significance to administrative decision-making’ or ‘may contain an error of fact or law materially affecting the ... decision’. This criterion may exclude matters where an applicant was not afforded the opportunity to present all relevant material and arguments to the ART at first instance; or where an applicant is self-represented, and has not received comprehensive or any legal advice.[155] These are exceptional circumstances faced most prevalently by applicants in social security matters.[156] Accordingly, if an applicant is unable to satisfy the high standards of clause 128, they are effectively barred from appealing an ART decision and their review rights are fundamentally reduced. The single-tier structure will only be successful in so far as funding for legal aid in social security matters is increased, and clause 128 is amended to accommodate for exceptional circumstances.
C Inconsistency of Reasoning
The revival of the Administrative Review Council (‘ARC’) by the ART Bill will be an integral step in remedying inconsistency.[157] The ARC was established by (and formally remains part of) Part V of the AAT Act. Conceptualised by the Kerr Committee as the ‘first step’ in the ‘evolution’ of Australian administrative law, it was a high level administrative authority that supervised existing review mechanisms and their adequacy.[158] The ARC was empowered to identify and address systemic issues; indeed, it was highly successful in doing so.[159] Yet, in the 2015–16 Budget, the government defunded the ARC, effectively resulting in its abolition. While the inclusion of the ARC in the ART Bill offers no guarantee of its actual existence, the Explanatory Memorandum states that $5.3 million has been dedicated to its refunding.[160]
VI CONCLUSION
The AAT failed to hold the executive accountable in the administration of the Robodebt Scheme due to its narrow jurisdiction, ‘two-tiered’ structure of review, inconsistent reasoning, and ‘blunt’ remedies. However, accountability mechanisms are usually individually weak; indeed, the AAT was ‘never intended to stand alone’.[161] Rather, the strength of the administrative law system depends on the interrelationship between each individual mechanism; collectively, they must provide a system of mutual accountability.[162] The Ombudsman’s ‘proactive’ role could, therefore, have mitigated the shortcomings of merits review. However, the Ombudsman’s failure to exercise important investigative powers and maintain independence from the executive, compounded by lack of determinative power, undermined its system-fixing potential. Such an incoherent ‘new administrative law’ system rendered it an inadequate check on executive power; confirming that the Kerr Committee’s ‘vision splendid’ has, indeed, faded. These failures ought to compel judicial and legislative reconsideration of existing administrative review mechanisms; while the ART Bill gives renewed impetus for change, further refinements to the proposed legislation are most certainly due.
* LLB (III), BCom, University of New South Wales. The author would like to thank Farid Assaf SC and Megan Caristo at Banco Chambers, Arisha Arif at King & Wood Mallesons, and Joanne Kinslor at the University of New South Wales for their insightful comments on earlier drafts of this article.
1 Prygodicz v Commonwealth [No 2] [2021] FCA 634; (2021) 173 ALD 277, 280 (Murphy J).
[2] Terry Carney, ‘Robo-Debt Illegality: The Seven Veils of Failed Guarantees of the Rule of Law?’ (2019) 44(1) Alternative Law Journal 4, 9; Matthew Groves and Janina Boughey, ‘Administrative Law in the Australian Environment’ in Matthew Groves (ed), Modern Administrative Law in Australia: Concepts and Context (Cambridge University Press, 2014) 3.
[3] Dennis Pearce, ‘The Fading of the Vision Splendid?: Administrative Law’ (1989) 58 Canberra Bulletin of Public Administration 15, 15. See also Peter Cane, Administrative Tribunals and Adjudication (Hart Publishing, 2009) 182–3, 257; Helen Murphy, ‘Administrative Review Rights and Changes to Commonwealth Government Service Provision’ (1998) 2(2) Flinders Journal of Law Reform 235, 243.
[4] Gabrielle Appleby, Alexander Reilly and Laura Grenfell, Australian Public Law (Oxford University Press, 3rd ed, 2019) 307.
[5] Linda Reif, ‘Building Democratic Institutions: The Role of National Human Rights Institutions in Good Governance and Human Rights Protection’ (2000) 13 Harvard Human Rights Journal 1, 29.
[6] Gerard Brennan, ‘Canberra Opening Address’ (Speech, The Administrative Appeals Tribunal Twentieth Anniversary Conference, 1 July 1996).
[7] Christopher Knaus, ‘Acoss Criticises Government’s “Appalling” Jail Threats to Welfare Recipients’, The Guardian (online, 7 December 2016) <https://www.theguardian.com/australia-news/2016/dec/07/acoss-criticises-governments-appalling-jail-threats-to-welfare-recipients>.
[8] Royal Commission into the Robodebt Scheme (Report, July 2023) vol 1, xxxiii.
[9] Ibid 28.
[10] Ibid.
[11] Matthew Knott, ‘“Strong Welfare Cop”: Scott Morrison’s New Self-Proclaimed Title’, The Sydney Morning Herald (online, 22 January 2015) <https://www.smh.com.au/politics/federal/strong-welfare-cop-scott-morrisons-new-selfproclaimed-title-20150122-12vkqw.html>.
[12] Anna Huggins, ‘Addressing Disconnection: Automated Decision-Making, Administrative Law and Regulatory Reform’ [2021] UNSWLawJl 37; (2021) 44(3) University of New South Wales Law Journal 1048, 1056; see also Louise Macleod, ‘Lessons Learnt about Digital Transformation and Public Administration: Centrelink’s Online Compliance Intervention’ [2017] AIAdminLawF 21; (2017) 89 Australian Institute of Administrative Law Forum 59, 59.
[13] Richard Glenn, ‘Centrelink’s Automated Debt Raising and Recovery System: A Report about the Department of Human Services’ Online Compliance Intervention System for Debt Raising and Recovery’ (Report No 02/2017, Commonwealth Ombudsman, April 2017).
[14] Commonwealth of Australia, ‘Statement of Agreed Facts’, Statement of Agreed Facts in Amato v Commonwealth, VID611/2019, 27 November 2019.
[15] Yee-Fui Ng et al, ‘Revitalising Public Law in a Technological Era: Rights, Transparency and Administrative Justice’ [2020] UNSWLawJl 37; (2020) 43(3) University of New South Wales Law Journal 1041, 1068.
[16] Prygodicz v Commonwealth [No 2] [2021] FCA 634; (2021) 173 ALD 277, 280 (Murphy J).
[17] Royal Commission into the Robodebt Scheme (Report, July 2023) vol 1, ch 10.
[18] See, most notably, Royal Commission into the Robodebt Scheme (Report, July 2023) in addition to the literature referenced throughout this article.
[19] The rate is set out in Rate Calculators such as Benefit Rate Calculator B: Social Security Act 1991 (Cth) s 1068, as applied by s 643.
[20] See, by analogy, Briginshaw v Briginshaw [1938] HCA 34; (1938) 60 CLR 336, 362 (Dixon J).
[21] Terry Carney, ‘Robo-Debt Illegality: The Seven Veils of Failed Guarantees of the Rule of Law?’ (2019) 44(1) Alternative Law Journal 4, 5.
[22] Social Security Act 1991 (Cth) ss 1222A(a), 1223(1).
[23] McDonald v Director-General of Social Security (1984) 1 FCR 354, 356–8 (Woodward J), 366 (Northrop J), 368–9 (Jenkinson J).
[24] Ibid 358 (Woodward J); Re Russell and Secretary, Department of Families, Housing, Community Services and Indigenous Affairs [2011] AATA 52, [35] (Senior Member Creyke); Terry Carney, ‘The New Digital Future for Welfare: Debts Without Legal Proofs or Moral Authority?’ [2018] (March) University of New South Wales Law Journal Forum 1, 6.
[25] Royal Commission into the Robodebt Scheme (Report, July 2023) vol 1, 41–2.
[26] Order of Davies J in Amato v Commonwealth (Federal Court of Australia, VID611/2019, 27 November 2019). See also Prygodicz v Commonwealth [No 2] [2021] FCA 634; (2021) 173 ALD 277: in approving the class action settlement, Murphy J declared that a decision that the applicant owed a debt under s 1223 of the Social Security Act 1991 (Cth) was not validly made where the Robodebt income averaging method was utilised as the sole methodology.
[27] Luke Henriques-Gomes, ‘Robodebt: Government to Refund 470,000 Unlawful Centrelink Debts Worth $721m’, The Guardian (online, 29 May 2020) <https://www.theguardian.com/australia-news/2020/may/29/robodebt-government-to-repay-470000-unlawful-centrelink-debts-worth-721m>.
[28] Robert French, ‘Judicial Review Rights’ [2001] AIAdminLawF 4; (2001) 28 Australian Institute of Administrative Law Forum 33, 35.
[29] A-G (NSW) v Quin (1990) 170 CLR 1, 36 (Brennan J); Drake v Minister for Immigration and Ethnic Affairs (1979) 24 ALR 577, 589 (Bowen CJ and Deane J).
[30] Minister for Immigration and Ethnic Affairs v Pochi (1980) 4 ALD 139, 143 (Smithers J); Frugtniet v Australian Securities and Investments Commission [2019] HCA 16; (2019) 266 CLR 250, 271 (Kiefel CJ, Keane and Nettle JJ); Shi v Migration Agents Registration Authority [2008] HCA 31; (2008) 235 CLR 286, 327 (Kiefel J).
[31] Drake v Minister for Immigration and Ethnic Affairs (1979) 24 ALR 577, 589 (Bowen CJ and Deane J); Shi v Migration Agents Registration Authority [2008] HCA 31; (2008) 235 CLR 286, 327 (Kiefel J).
[32] Commonwealth Administrative Review Committee, Parliament of Australia, Commonwealth Administrative Review Committee Report (Parliamentary Paper No 144, August 1971); Administrative Appeals Tribunal Act 1975 (Cth) s 2A.
[33] Bushell v Repatriation Commission [1992] HCA 47; (1992) 175 CLR 408, 424–5 (Brennan J); Green v Minister for Immigration and Citizenship [2008] FCA 125; (2008) 100 ALD 346, 353 (Tamberlin J); Haberfield v Department of Veterans’ Affairs as Delegate for Comcare [2002] FCA 1579; (2002) 72 ALD 333, 345 (Sackville J); Kowalski v Repatriation Commission [2011] FCAFC 43, [33]–[34] (Logan J); Calvista Australia Pty Ltd v Administrative Appeals Tribunal [2013] FCA 860; (2013) 304 ALR 489, 511–12 (Robertson J); Sullivan v Civil Aviation Safety Authority [2014] FCAFC 93; (2014) 322 ALR 581, 616 (Flick and Perry JJ); Bienstein v A-G (Cth) [2008] AATA 330; (2008) 47 AAR 354.
[34] Shi v Migration Agents Registration Authority [2008] HCA 31; (2008) 235 CLR 286; Drake v Minister for Immigration and Ethnic Affairs (1979) 24 ALR 577.
[35] Peter Cane, ‘Merits Review and Judicial Review – The AAT as Trojan Horse’ [2000] FedLawRw 11; (2000) 28(2) Federal Law Review 213.
[36] Sir Anthony Mason, ‘The Increasing Importance of Judicial Review of Administrative Action’ (Speech, Administrative Law Section of the Law Institute of Victoria’s Annual Dinner, 6 June 1994).
[37] Darren O’Donovan, ‘Social Security Appeals and Access to Justice: Learning from the Robodebt Controversy’ (2020) 158 Precedent 34, 37.
[38] Panel Discussion with Paul Henman, Terry Carney and Yee-Fui Ng, ‘Law Matters: Administrative Justice After Robodebt’ (UQ Law School, The University of Queensland, 10 August 2023) 00:15:12–00:16:59 <https://law.uq.edu.au/event/session/19911>.
[39] Ibid.
[40] Australian Broadcasting Tribunal v Bond (1990) 170 CLR 321, 337 (Mason CJ).
[41] Administrative Appeals Tribunal Act 1975 (Cth) ss 3(3), 25; cf Administrative Decisions (Judicial Review) Act 1977 (Cth) ss 3(1)–(2).
[42] Re Gallivan Investments Ltd v Australian Securities Commission [1991] AATA 211; (1991) 24 ALD 611; Re Toll v Australian Securities Commission [1993] AATA 6; (1992) 29 ALD 412; Re Adams v Veterans’ Review Board [1992] AATA 281; (1992) 16 AAR 307; Australian Postal Corporation v Matusko (Australian Administrative Tribunal, Olney J, 29 April 1996).
[43] Commonwealth Administrative Review Committee, Parliament of Australia, Commonwealth Administrative Review Committee Report (Parliamentary Paper No 144, August 1971) 114; Ward v Nicholls (1988) 84 ALR 471, 481 (Wilcox J); Lamb v Moss [1983] FCA 254; (1983) 49 ALR 533, 557 (Bowen CJ, Sheppard and Fitzgerald JJ).
[44] Administrative Decisions (Judicial Review) Act 1977 (Cth) s 6.
[45] Australian Broadcasting Tribunal v Bond (1990) 170 CLR 321, 337 (Mason CJ).
[46] Ibid; Mark Aronson, Matthew Groves and Greg Weeks, Judicial Review of Administrative Action and Government Liability (Thomson Reuters, 7th ed, 2022) 87 [2.320]; Peter Cane, Administrative Tribunals and Adjudication (Hart Publishing, 1st ed, 2009) 222–3.
[47] Administrative Appeals Tribunal Act 1975 (Cth) ss 3(3), 25(1) do not operate with reference to ‘conduct’. See Robin Creyke and Graeme Hill, ‘A Wavy Line in the Sand: Bond and Jurisdictional Issues in Judicial and Administrative Review’ [1998] FedLawRw 2; (1998) 26(1) Federal Law Review 15, 16; Peter Cane, Administrative Tribunals and Adjudication (Hart Publishing, 1st ed, 2009) 223.
[48] Director-General of Social Services v Chaney [1980] FCA 87; (1980) 31 ALR 571; Robin Creyke, ‘Administrative Tribunals’ in Matthew Groves and H P Lee (eds), Australian Administrative Law: Fundamentals, Principles and Doctrines (Cambridge University Press, 2007) 77, 89.
[49] Dennis Pearce, Administrative Appeals Tribunal (LexisNexis Butterworths, 2nd ed, 2007) 30.
[50] Robin Creyke, ‘Administrative Tribunals’ in Matthew Groves and H P Lee (eds), Australian Administrative Law: Fundamentals, Principles and Doctrines (Cambridge University Press, 2007) 77, 86–9; Peter Cane, Administrative Tribunals and Adjudication (Hart Publishing, 1st ed, 2009) 222–3; Peter Cane, Leighton McDonald and Kristen Rundle, Principles of Administrative Law (Oxford University Press, 3rd ed, 2018) 43; Robin Creyke and Graeme Hill, ‘A Wavy Line in the Sand: Bond and Jurisdictional Issues in Judicial and Administrative Review’ [1998] FedLawRw 2; (1998) 26(1) Federal Law Review 15.
[51] Peter Cane, Administrative Tribunals and Adjudication (Hart Publishing, 1st ed, 2009) 223.
[52] Terry Carney, ‘Automation in Social Security: Implications for Merits Review?’ (Research Paper No 20/16, Faculty of Law, University of Sydney, March 2020) 14–15.
[53] Ibid 15.
[54] Social Security (Administration) Act 1999 (Cth) ss 129, 135.
[56] Panel Discussion with Paul Henman, Terry Carney and Yee-Fui Ng, ‘Law Matters: Administrative Justice After Robodebt’ (UQ Law School, The University of Queensland, 10 August 2023) 00:15:12–00:16:59 <https://law.uq.edu.au/event/session/19911>.
[57] National Legal Aid, Rethink Robo-Debt: Building a Fair and Accurate System People can Trust: Submission to the Senate Community Affairs References Committee Inquiry into Centrelink’s Compliance Program (Submission No 26, 27 September 2019) 28.
[58] See, eg, Royal Commission into the Robodebt Scheme (Report, July 2023) vol 1, 207; ibid.
[59] National Legal Aid, Rethink Robo-debt: Building a Fair and Accurate System People Can Trust: Submission to the Senate Community Affairs References Committee Inquiry into Centrelink’s Compliance Program (Submission No 26, 27 September 2019) 28–30; Terry Carney, ‘Automation in Social Security: Implications for Merits Review?’ (Research Paper No 20/16, Faculty of Law, University of Sydney, March 2020) 15.
[60] Director-General of Social Services v Hales [1983] FCA 81; (1983) 78 FLR 373, 397 (Lockhart J).
[61] Tribunals Amalgamation Act 2015 (Cth); Commonwealth, Parliamentary Debates, Senate, 3 December 2014, 10070 (Senator Fifield); Moira Coombs, ‘Amalgamation of Merits Review Tribunals: Budget Review 2014–15 Index’, Parliament of Australia (Web Page) <https://www.aph.gov.au/About_Parliament/Parliamentary_Departments/Parliamentary_Library/pubs/rp/BudgetReview201415/Tribunals>; Robin Creyke, ‘Australian Tribunals: Impact of Amalgamation’ (2020) 26 Australian Journal of Administrative Law 206, 209; Robin Creyke, ‘Tribunal Amalgamation 2015: An Opportunity Lost?’ [2016] AIAdminLawF 15; (2016) 84 Australian Institute of Administrative Law Forum 54, 61; Matthew Groves and Greg Weeks, Tribunal Justice and Politics in Australia: The Rise and Fall of the Administrative Appeals Tribunal (2023) 97(4) Australian Law Journal 278, 283.
[62] Tribunals Amalgamation Act 2015 (Cth) ss 142, 179.
[63] Terry Carney, ‘The New Digital Future for Welfare: Debts Without Legal Proofs or Moral Authority?’ [2018] (March) University of New South Wales Law Journal Forum 1, 2; John Townsend, ‘Better Decisions? Robodebt and the Failings of Merits Review’ in Janina Boughey and Katie Miller (eds), The Automated State: Implications, Challenges and Opportunities for Public Law (Federation Press, 2022) 52, 63; Monash University Faculty of Law, Submission to Royal Commission into the Robodebt Scheme, Public Administration and Automation (31 January 2023) 10.
[64] Royal Commission into the Robodebt Scheme (Report, July 2023) vol 2, 556; John Townsend, ‘Better Decisions? Robodebt and the Failings of Merits Review’ in Janina Boughey and Katie Miller (eds), The Automated State: Implications, Challenges and Opportunities for Public Law (Federation Press, 2022) 52, 66–7; Anthony O’Donnell and Christopher Arup, ‘Income Support in a Time of Contagion’ (2021) 34 Australian Journal of Labour Law 43, 57; Luke Henriques-Gomes, ‘Coalition Warned Robodebt Scheme Was Unenforceable Three Years Before It Acted’, The Guardian (online, 12 February 2020) <https://www.theguardian.com/australia-news/2020/feb/12/coalition-warnedrobodebt-scheme-was-unenforceable-three-years-before-it-acted>.
[65] Terry Carney, ‘Robo-Debt Illegality: The Seven Veils of Failed Guarantees of the Rule of Law?’ (2019) 44(1) Alternative Law Journal 4, 7.
[66] Social Security (Administration) Act 1999 (Cth) s 168.
[67] Administrative Appeals Tribunal Act 1975 (Cth) s 66B; Royal Commission into the Robodebt Scheme (Report, July 2023) vol 2, 554.
[68] JWTT v Commissioner of Taxation [2017] AATA 1612, [16]–[17] (Deputy President McCabe); Royal Commission into the Robodebt Scheme (Report, July 2023) vol 2, 563; John Townsend, ‘Better Decisions? Robodebt and the Failings of Merits Review’ in Janina Boughey and Katie Miller (eds), The Automated State: Implications, Challenges and Opportunities for Public Law (Federation Press, 2022) 52, 69; Terry Carney, ‘Robo-Debt Illegality: The Seven Veils of Failed Guarantees of the Rule of Law?’ (2019) 44(1) Alternative Law Journal 4, 7; Terry Carney, ‘Vulnerability: False Hope for Vulnerable Social Security Clients?’ [2018] UNSWLawJl 27; (2018) 41(3) University of New South Wales Law Journal 783, 810; Terry Carney, ‘Robo-Debt Illegality: A failure of rule of law protections?’, Australian Public Law (Forum Post, 30 April 2018) <https://auspublaw.org/2018/04/robo-debt-illegality/>.
[69] John Townsend, ‘Better Decisions? Robodebt and the Failings of Merits Review’ in Janina Boughey and Katie Miller (eds), The Automated State: Implications, Challenges and Opportunities for Public Law (Federation Press, 2022) 52, 67; Royal Commission into the Robodebt Scheme (Report, July 2023) vol 3, appendix 9.
[70] Prygodicz v Commonwealth [No 2] [2021] FCA 634; (2021) 173 ALD 277, 281 (Murphy J); John Townsend, ‘Better Decisions? Robodebt and the Failings of Merits Review’ in Janina Boughey and Katie Miller (eds), The Automated State: Implications, Challenges and Opportunities for Public Law (Federation Press, 2022) 52, 67–8; Terry Carney, ‘Robo-Debt Illegality: The Seven Veils of Failed Guarantees of the Rule of Law?’ (2019) 44(1) Alternative Law Journal 4, 5; Peter Whiteford, ‘Debt by Design: The Anatomy of a Social Policy Fiasco – Or Was It Something Worse?’ (2021) 80 Australian Journal of Public Administration 340, 353, 356–7; Luke Henriques-Gomes, ‘“Not Above the Law”: Warning for Australia’s Department of Social Services Over Threat to Ignore Tribunal’ The Guardian (online, 7 October 2021) <https://www.theguardian.com/australia-news/2020/feb/12/coalition-warnedrobodebt-scheme-was-unenforceable-three-years-before-it-acted>; Alex Mitchell, ‘Lawyer Ignored Tribunal’s Robodebt Finding’ The Dailey Advertiser (online, 25 January 2023) <https://www.dailyadvertiser.com.au/story/8060346/lawyer-ignored-tribunals-robodebt-finding/>.
[71] Ian Callinan, Review: Section 4 of the Tribunals Amalgamation Act 2015 (Cth) (Final Report, 19 December 2019) 160 [9.21].
[72] Ibid 13 [1.14]; Robin Creyke, ‘Administrative Justice – Towards Integrity in Government’ [2007] MelbULawRw 30; (2007) 31 Melbourne University Law Review 705, 727.
[73] Administrative Appeals Tribunal Act 1975 (Cth) s 2A; Robin Creyke, ‘Administrative Tribunals’ in Matthew Groves and H P Lee (eds), Australian Administrative Law: Fundamentals, Principles and Doctrines (Cambridge University Press, 2007) 77, 94; Minister for Immigration and Multicultural Affairs v Eshetu [1999] HCA 21; (1997) 197 CLR 611, 643 (Gummow J), 628 (Gleeson CJ and McHugh J), 668 (Callinan J); Sun Zhan Qui v Minister for Immigration and Ethnic Affairs [1997] FCA 324.
[74] Robin Creyke, ‘Australian Tribunals: Impact of Amalgamation’ (2020) 26 Australian Journal of Administrative Law 206, 208; Ian Callinan, Review: Section 4 of the Tribunals Amalgamation Act 2015 (Cth) (Final Report, 19 December 2019) 163 [10.10]; Rachel Bacon, ‘Rewriting the Social Contract? The SSAT, the AAT and the Contracting out of Employment Services’ [2002] FedLawRw 2; (2002) 30(1) Federal Law Review 39, 51; Royal Commission into the Robodebt Scheme (Report, July 2023) vol 2, 554.
[75] Darren O’Donovan, ‘Social Security Appeals and Access to Justice: Learning from the Robodebt Controversy’ (2020) 158 Precedent 34, 37.
[76] Robin Creyke, ‘Administrative Tribunals’ in Matthew Groves and H P Lee (eds), Australian Administrative Law: Fundamentals, Principles and Doctrines (Cambridge University Press, 2007) 77, 94. See also Robin Creyke, ‘Tribunal Amalgamation 2015: An Opportunity Lost?’ [2016] AIAdminLawF 15; (2016) 84 Australian Institute of Administrative Law Forum 54, 69.
[77] Robin Creyke, ‘Administrative Tribunals’ in Matthew Groves and H P Lee (eds), Australian Administrative Law: Fundamentals, Principles and Doctrines (Cambridge University Press, 2007) 77, 98; Peter Cane, ‘Merits Review and Judicial Review – The AAT as Trojan Horse’ [2000] FedLawRw 11; (2000) 28(2) Federal Law Review 213, 227; Michael Asimow and Jeffrey Lubbers, ‘The Merits of “Merits Review”: A Comparative Look at the Australian Administrative Appeals Tribunal’ [2011] AIAdminLawF 18; (2011) 67 Australian Institute of Administrative Law Forum 58, 69.
[78] Peter Cane, ‘Merits Review and Judicial Review – The AAT as Trojan Horse’ [2000] FedLawRw 11; (2000) 28(2) Federal Law Review 213, 227; John Griffith, ‘50th Anniversary of the Kerr Committee’ (2022) 28 Australian Journal of Administrative Law 224, 232–3.
[79] Re Ganchov and Comcare [1990] AATA 419; (1990) 19 ALD 541, 542 (Deputy President Todd); Minister for Immigration and Ethnic Affairs v Pochi (1980) ALD 130, 154 (Deane J); Re Niola Nominees Pty Ltd and Minister for Health (1986) 9 ALN N200; John Basten, ‘Disputes Involving the Commonwealth: Observations from the Outside’ (1999) 92 Canberra Bulletin of Judicial Administration 38, 38; Michael Asimow and Jeffrey Lubbers, ‘The Merits of “Merits Review”: A Comparative Look at the Australian Administrative Appeals Tribunal’ [2011] AIAdminLawF 18; (2011) 67 Australian Institute of Administrative Law Forum 58, 69; Garry Downes, ‘Structure, Power and Duties of the Administrative Appeals Tribunal of Australia’ (Speech, Administrative Court of Thailand and Central Administrative Court of Thailand, 21 February 2006) [48].
[80] Royal Commission into the Robodebt Scheme (Report, July 2023) vol 2, 561.
[81] Ibid vol 1, 239.
[82] 2016/S104681 (Administrative Appeals Tribunal, Member Terry Carney, 8 March 2017) 10 [54]–[56]; Royal Commission into the Robodebt Scheme (Report, July 2023) vol 1, 239.
[83] Royal Commission into the Robodebt Scheme (Report, July 2023) vol 2, 556, 561.
[84] Darren O’Donovan, ‘Social Security Appeals and Access to Justice: Learning from the Robodebt Controversy’ (2020) 158 Precedent 34, 37.
[85] Ibid.
[86] Matthew Groves and Greg Weeks, ‘Tribunal Justice and Politics in Australia: The Rise and Fall of the Administrative Appeals Tribunal’ (2023) 97 Australian Law Journal 278, 289–90.
[87] Lenore Taylor, ‘Prominent Critic of Robodebt Who Ruled Against Scheme Five Times Lost AAT Job, Inquiry Hears’, The Guardian (online, 24 January 2023) <https://www.theguardian.com/australia-news/2023/jan/24/prominent-critic-of-robodebt-who-ruled-against-scheme-five-times-lost-aat-job-inquiry-hears>.
[88] Paul Karp, ‘AAT Member Says He Was Benched After Too Many Decisions Against Government’, The Guardian (online, 16 May 2022) <https://www.theguardian.com/australia-news/2022/may/16/aat-member-says-he-was-benched-after-too-many-decisions-against-government>.
[89] Royal Commission into the Robodebt Scheme (Report, July 2023) vol 2, 561.
[90] Re Drake and Minister for Immigration and Ethnic Affairs [No 2] (1979) 2 ALD 634, 639 (Brennan J).
[91] Royal Commission into the Robodebt Scheme (Report, July 2023) vol 3, appendix 9; Administrative Appeals Tribunal Act 1975 (Cth) s 43(1)(c)(ii).
[92] Royal Commission into the Robodebt Scheme (Report, July 2023) vol 2, 562.
[93] Re Secretary, Department of Social Services and Dawson [2021] AATA 3442, [42], [48], [53] (Senior Member Evans-Bonner).
[94] Lord Chancellor v Detention Action [2015] EWCA Civ 840, [27] (Lord Dyson MR).
[95] Maria O’Sullivan, ‘Automated Decision-Making and Human Rights: The Right to an Effective Remedy’ in Janina Boughey and Katie Miller (eds), The Automated State: Implications, Challenges and Opportunities for Public Law (Federation Press, 2022) 70, 81–2; John Townsend, ‘Better Decisions? Robodebt and the Failings of Merits Review’ in Janina Boughey and Katie Miller (eds), The Automated State: Implications, Challenges and Opportunities for Public Law (Federation Press, 2022) 52, 54, 56; Lynsey Blayden, ‘Designing Administrative Law for an Administrative State: The Carefully Calibrated Approach of the Kerr Committee’ (2021) 28(4) Australian Journal of Administrative Law 205, 222; Peter Cane, ‘Understanding Administrative Adjudication’ in Linda Pearson, Carol Harlow and Michael Taggart (eds), Administrative Law in a Changing State: Essays in Honour of Mark Aronson (2008, Bloomsbury Publishing) 273, 294.
[96] Royal Commission into the Robodebt Scheme (Report, July 2023) vol 2, 553; Maria O’Sullivan, ‘Automated Decision-Making and Human Rights: The Right to an Effective Remedy’ in Janina Boughey and Katie Miller (eds), The Automated State: Implications, Challenges and Opportunities for Public Law (Federation Press, 2022) 70, 72, 80–2, 85–8.
[97] Mark Aronson, Matthew Groves and Greg Weeks, Judicial Review of Administrative Action and Government Liability (Thomson Reuters, 7th ed, 2022) 5–6; Anna Huggins, ‘Addressing Disconnection: Automated Decision-Making, Administrative Law and Regulatory Reform’ [2021] UNSWLawJl 37; (2021) 44(3) University of New South Wales Law Journal 1048, 1072; Mark Aronson, Matthew Groves and Greg Weeks, Judicial Review of Administrative Action and Government Liability (Thomson Reuters, 7th ed, 2022) 5–6.
[98] Royal Commission into the Robodebt Scheme (Report, July 2023) vol 2, 553.
[99] Mark Aronson, Matthew Groves and Greg Weeks, Judicial Review of Administrative Action and Government Liability (Thomson Reuters, 7th ed, 2022) 203 [4.160]; Terry Carney, ‘Robo-Debt Illegality: The Seven Veils of Failed Guarantees of the Rule of Law?’ (2019) 44(1) Alternative Law Journal 4.
[100] Senate Standing Committee on Finance and Public Administration, Parliament of Australia, Review of the Office of the Commonwealth Ombudsman (Report, December 1991) 14 [2.43]; Shergold v Tanner [2002] HCA 19; (2002) 209 CLR 126, 130 (Gleeson CJ, McHugh, Gummow, Kirby and Hayne JJ); Yee-Fui Ng, ‘Institutional Adaptation and the Administrative State’ [2021] MelbULawRw 4; (2021) 44(3) Melbourne University Law Review 889, 900; Anita Stuhmcke, ‘“Each for Themselves” or “One for All”? The Changing Emphasis of the Commonwealth Ombudsman’ (2010) 31(1) Federal Law Review 143, 145; Anita Stuhmcke, ‘A Snapshot in Time: The Changing Systemic Role of the Australian Commonwealth Ombudsman’ (2014) 4(1) Journal of Law and Social Sciences 99, 100–2; Gerald Caiden, ‘The Institution of Ombudsman’ in Gerald Caiden (ed), International Handbook of the Ombudsman: Evolution and Present Function (Greenwood Press, 1983) 3; Ron McLeod, ‘The Ramblings of a New Ombudsman’ (Speech, Auditor-General’s Office, 7 August 1998) 11–12.
[101] Botany Council v Ombudsman (1995) 37 NSWLR 357, 363 (Kirby P); Greg Weeks, Soft Law and Public Authorities: Remedies and Reform (Hart Publishing, 2016) 236; Anita Stuhmcke, ‘“Each for Themselves” or “One for All”? The Changing Emphasis of the Commonwealth Ombudsman’ (2010) 31(1) Federal Law Review 143, 144.
[102] Ombudsman Act 1976 (Cth) s 5(1)(b); Gabrielle Appleby, Alexander Reilly and Laura Grenfell, Australian Public Law (Oxford University Press, 3rd ed, 2019) 313; Anita Stuhmcke, ‘Undefined and Undervalued: The Systemic Role of Classical Ombudsman Institution’ (Conference Paper, Annual International Conference Proceedings, Global Science and Technology Forum, 2014) 48–54.
[103] Anita Stuhmcke, ‘Ombudsmen and Integrity Review’ in Linda Pearson, Carol Harlow and Michael Taggart (eds), Administrative Law in a Changing State: Essays in Honour of Mark Aronson (2008, Bloomsbury Publishing) 349, 355; Jeremy (Wei Peng) Soh, ‘Creating a Framework for Evaluating the “Effectiveness” of the Commonwealth Ombudsman (2019) 26 Australian Journal of Administrative Law 164, 167; Mark Aronson and Matthew Groves, Judicial Review of Administrative Action (Thomson Reuters, 5th ed, 2013) [1.90]; Anita Stuhmcke, ‘Evaluating the Effectiveness of an Ombudsman: A Riddle, Wrapped in a Mystery Inside an Enigma’ (Research Paper, International Ombudsman Institute World Conference, 15 November 2012) 3; John McMillan, ‘Future Directions 2009 – The Ombudsman’ [2010] AIAdminLawF 17; (2010) 63 Australian Institute of Administrative Law Forum 13, 14–15.
[104] Re Ombudsman Act (1970) 72 WWR 176, 190, 192 (Milvain CJ).
[105] Ainsworth v The Ombudsman (1988) 17 NSWLR 276, 283 (Enderby J).
[106] Re Ombudsman Act (1970) 72 WWR 176, 190, 192 (Milvain CJ).
[107] James Spigelman, ‘Jurisdiction and Integrity’ (Speech, National Lecture Series of the Australian Institute of Administrative Law, 5 August 2004).
[108] Greg Weeks, Soft Law and Public Authorities: Remedies and Reform (Hart Publishing, 2016) 244.
[109] Botany Council v Ombudsman (1995) 37 NSWLR 357, 367–8 (Kirby J); K v NSW Ombudsman [2000] NSWSC 771, [61] (Whealy J); Ainsworth v Ombudsman (1988) 17 NSWLR 276 (Enderby J).
[110] Botany Council v Ombudsman (1995) 37 NSWLR 357, 367–8 (Kirby J).
[111] Royal Commission into the Robodebt Scheme (Report, July 2023) vol 2, 574.
[112] Ombudsman Act 1976 (Cth) s 15.
[113] Royal Commission into the Robodebt Scheme (Report, July 2023) vol 1, 208.
[114] Richard Glenn, ‘Centrelink’s Automated Debt Raising and Recovery System: A Report about the Department of Human Services’ Online Compliance Intervention System for Debt Raising and Recovery’ (Report No 02/2017, Commonwealth Ombudsman, April 2017).
[115] See Richard Glenn, ‘Centrelink’s Automated Debt Raising and Recovery System: A Report about the Department of Human Services’ Online Compliance Intervention System for Debt Raising and Recovery’ (Report No 02/2017, Commonwealth Ombudsman, April 2017) 7–8, 41–3; Royal Commission into the Robodebt Scheme (Report, July 2023) vol 2, 593; Peter Hanks, ‘Administrative Law and Welfare Rights: A 40-Year Story From Green v Daniels to ‘Robot Debt Recovery’ – Closing the Chapter’ (2021) 103 Australian Institute of Administrative Law Forum 19, 21.
[116] Terry Carney, ‘Robo-Debt Illegality: The Seven Veils of Failed Guarantees of the Rule of Law?’ (2019) 44(1) Alternative Law Journal 4, 6; Peter Hanks, ‘Administrative Law and Welfare Rights: A 40-Year Story from Green v Daniels to ‘Robot Debt Recovery’ – Closing the Chapter’ (2021) 103 Australian Institute of Administrative Law Forum 19, 21.
[117] Royal Commission into the Robodebt Scheme (Report, July 2023) vol 2, 591.
[118] Ibid 588.
[119] Ibid.
[120] Ombudsman Act 1976 (Cth) s 9.
[121] Rick Snell, ‘Australian Ombudsman: A Continual Work in Progress’ in Matthew Groves and H P Lee (eds), Australian Administrative Law: Fundamentals, Principles and Doctrines (Cambridge University Press, 2007) 100, 105.
[122] Terry Carney, ‘Robo-Debt Illegality: The Seven Veils of Failed Guarantees of the Rule of Law?’ (2019) 44(1) Alternative Law Journal 4, 6.
[123] Ombudsman Act 1976 (Cth) s 6(2); see generally Dennis Pearce, ‘The Ombudsman and the Rule of Law’ [1994] AIAdminLawF 2; (1994) 1 Australian Institute of Administrative Law Forum 1, 1.
[124] Royal Commission into the Robodebt Scheme (Report, July 2023) vol 2, 588.
[125] Ibid 596–7.
[126] Evidence to Senate Community Affairs Reference Committee, Parliament of Australia, Adelaide, 10 April 2017, 6 (Jason McNamara, General Manager, Integrity Process Modernisation, Department of Human Services).
[127] Anita Stuhmcke, ‘Ombudsmen and Integrity Review’ in Linda Pearson, Carol Harlow and Michael Taggart (eds), Administrative Law in a Changing State: Essays in Honour of Mark Aronson (2008, Bloomsbury Publishing) 349, 349.
[128] Booth v Dillon [No 1] [1976] VicRp 23; [1976] VR 291, 295 (Lush J).
[129] Anita Stuhmcke, ‘Ombudsmen and Integrity Review’ in Linda Pearson, Carol Harlow and Michael Taggart (eds), Administrative Law in a Changing State: Essays in Honour of Mark Aronson (2008, Bloomsbury Publishing) 349, 349.
[130] Dennis Pearce, ‘The Commonwealth Ombudsman: The Right Office in the Wrong Place’ in Robin Creyke and John McMillan (eds), The Kerr Vision of Australian Administrative Law at the Twenty-Five Year Mark (Australian National University Centre for International and Public Law, 1998) 54, 72. See also Ainsworth v The Ombudsman (1988) 17 NSWLR 276, 283 (Enderby J); Anita Stuhmcke, ‘Ombudsmen and Integrity Review’ in Linda Pearson, Carol Harlow and Michael Taggart (eds), Administrative Law in a Changing State: Essays in Honour of Mark Aronson (2008, Bloomsbury Publishing) 349, 349; Katrine Del Villar, ‘Who Guards the Guardians? Recent Developments Concerning the Jurisdiction and Accountability of Ombudsman’ [2003] AIAdminLawF 3; (2003) 36 Australian Institute of Administrative Law Forum 25, 44.
[131] Cheryl Saunders, It’s Your Constitution: Governing Australia Today (Federation Press, 2nd ed, 2003) 114.
[132] Justice James Spigelman, ‘Judicial Review and the Integrity Branch of Government’ (Speech, World Jurist Association Congress, Shanghai, 8 September 2005); Allan Asher, ‘Integrity Agencies: The Fourth Arm of Government’ (Speech, Public Sector Leadership Conference: Rethinking and Improving Service Delivery, 12 May 2011); Anita Stuhmcke and Anne Tran, ‘The Commonwealth Ombudsman: An Integrity Branch of Government?’ (2007) 32(4) Alternative Law Journal 233.
[133] Anita Stuhmcke, ‘Government Watchdog Agencies and Administrative Justice’ in Joe Tomlinson, Marc Hertogh and Richard Kirkham (eds), The Oxford Handbook of Administrative Justice (Oxford University Press, 2021) 115, 120–1.
[134] Royal Commission into the Robodebt Scheme (Report, July 2023) vol 2, 583.
[135] Transcript of Proceedings, Royal Commission into the Robodebt Scheme (5 December 2022) 1044–5.
[136] Gabrielle Appleby, Alexander Reilly and Laura Grenfell, Australian Public Law (Oxford University Press, 3rd ed, 2019) 307.
[137] Albert Venn Dicey, Introduction to the Study of the Law of the Constitution (Macmillan, 8th ed, 1915). See also John McMillan, ‘Re-Thinking the Separation of Powers’ (2010) 38 Federal Law Review 423, 423.
[138] Ombudsman Act 1976 (Cth) s 15; Chairperson, Aboriginal and Torres Strait Islander Commission v Commonwealth Ombudsman [1995] FCA 1584; (1995) 134 ALR 238, 243 (Einfeld J).
[139] Ombudsman Act 1976 (Cth) ss 16, 17, 18.
[140] John McMillan, ‘The Ombudsman and The Rule of Law’ [2005] AIAdminLawF 1; (2005) 44 Australian Institute of Administrative Law Forum 1, 3.
[141] Creyke et al, Control of Government Action: Text, Cases and Commentary (LexisNexis Butterworths, 5th ed, 2019).
[142] Matthew Groves and Greg Weeks, ‘Institutions and Their Memories’ (2023) 30 Australian Journal of Administrative Law 83, 83.
[143] Christopher Pollitt, ‘Institutional Amnesia: A Paradox of the Information Age?’ (2000) 19 Prometheus: Critical Studies in Information 5; Matthew Groves and Greg Weeks, ‘Institutions and Their Memories’ (2023) 30 Australian Journal of Administrative Law 83, 84.
[144] Royal Commission into the Robodebt Scheme (Report, July 2023) vol 1, 208.
[145] Greg Weeks, Soft Law and Public Authorities: Remedies and Reform (Hart Publishing, 2016) 238.
[146] Howard Gadlin, ‘Some Thoughts on Informality’ (2012) 5(1) Journal of the International Ombudsman Association 31, 35–6.
[147] Jake Evans, “‘Politicised’ Administrative Appeals Tribunal Abolished, After Attorney-General Declares Its Reputation Ruined’, ABC News (online, 16 December 2022) <https://www.abc.net.au/news/2022-12-16/administrative-appeals-tribunal-abolished-by-attorney-general/101781300>.
[148] Debra Wilkinson and Elizabeth Morison, ‘Cronyism in Appointments to the AAT: An Empirical Analysis’ (Discussion Paper, Australia Institute, May 2022) 1.
[149] Alan Robertson, ‘Remedies, Present and Future’ (1995) 44 Administrative Law Review 85, 88.
[150] John McMillan, ‘Merit Review and the AAT: A Concept Develops’ in John McMillan (ed), The AAT, Twenty Years Forward: Papers Presented at a Conference to Mark the Twentieth Anniversary of the Administrative Appeals Tribunal (Australian Institute of Administrative Law, 1998) 33; Robin Creyke, ‘Tribunals – “Carving Out the Philosophy of Their Existence”: The Challenge for the 21st Century’ [2012] AIAdminLawF 22; (2012) 71 Australian Institute of Administrative Law Forum 19.
[151] Administrative Review Tribunal Bill 2023 (Cth) cl 4; cf Administrative Appeals Tribunal Act 1975 (Cth) ss 3(3).
[152] Robin Creyke, ‘Tribunal Amalgamation 2015: An Opportunity Lost?’ [2016] AIAdminLawF 15; (2016) 84 Australian Institute of Administrative Law Forum 54, 69.
[153] Administrative Review Tribunal Bill 2023 (Cth) cl 113(2)(a). In this regard, the Bill directly implements Recommendation 20.4 of the Royal Commission into the Robodebt Scheme (Report, July 2023).
[154] Janina Boughey and Ellen Rock, Submission No 13 to Standing Committee on Social Policy and Legal Affairs, Parliament of Australia, Inquiry into the Administrative Review Tribunal Bill 2023 and the Administrative Review Tribunal (Consequential and Transitional Provisions No 1) Bill 2023 (30 January 2024) 5.
[155] Economic Justice Australia, Submission No 7 to Standing Committee on Social Policy and Legal Affairs, Parliament of Australia, Inquiry into the Administrative Review Tribunal Bill 2023 and the Administrative Review Tribunal (Consequential and Transitional Provisions No 1) Bill 2023 (18 January 2024) 6; Monash Law Clinics, Submission No 8 to Standing Committee on Social Policy and Legal Affairs, Parliament of Australia, Inquiry into the Administrative Review Tribunal Bill 2023 and the Administrative Review Tribunal (Consequential and Transitional Provisions No 1) Bill 2023 (18 January 2024) 3.
[156] Economic Justice Australia, Submission No 7 to Standing Committee on Social Policy and Legal Affairs, Parliament of Australia, Inquiry into the Administrative Review Tribunal Bill 2023 and the Administrative Review Tribunal (Consequential and Transitional Provisions No 1) Bill 2023 (18 January 2024) 3.
[157] Administrative Review Tribunal Bill 2023 (Cth) pt 9. In this regard, the Bill directly implements Recommendations 20.5 and 23.4 of the Royal Commission into the Robodebt Scheme (Report, July 2023).
[158] Commonwealth Administrative Review Committee, Parliament of Australia, Commonwealth Administrative Review Committee Report (Parliamentary Paper No 144, August 1971) 103 [346].
[159] Narelle Bedford, ‘The Kerr Report’s Vision for the Administrative Review Council and the (Sad) Modern Reality’, Australian Public Law (Forum Post, 21 May 2021) <https://www.auspublaw.org/blog/2021/05/the-kerr-reports-vision-for-the-administrative-review-council>.
[160] Explanatory Memorandum, Administrative Review Tribunal Bill 2023 (Cth), 9.
[161] Murray Gleeson, ‘Outcome, Process and the Rule of Law’ (2006) 65(3) Australian Journal of Public Administration 5, 5; Robin Creyke, ‘Administrative Justice – Towards Integrity in Government’ [2007] MelbULawRw 30; (2007) 31 Melbourne University Law Review 705, 727; AJ Brown et al, ‘Chaos or Coherence? Strengths, Opportunities and Challenges for Australia’s Integrity Systems’ (Final Report, National Integrity Systems Assessment, December 2005) 15, 17.
[162] Ibid.
AustLII:
Copyright Policy
|
Disclaimers
|
Privacy Policy
|
Feedback
URL: http://www.austlii.edu.au/au/journals/UNSWLawJlStuS/2024/13.html