Administrative Review Reform Issues Paper

The purpose of merits review tribunals has been expressed in a variety of different ways, as noted in the Issues Paper. In our view, these principles can be framed around five overarching themes:

(1) Public trust – the tribunal should engender public trust and confidence in its role. Both in its institutional structure and practical operation, the tribunal should embody the principles of accountability, independence, impartiality, transparency and fairness.

(2) Best practice in government decision-making – the tribunal should provide access to “merits”-based outcomes that offer the correct and preferable decision in individual cases. It should also support the ongoing improvement of government decision-making practices, including by modelling best practice in decision-making, and by supporting agencies to exercise power according to those same principles.

(3) Accountability – the tribunal should support the accountability of government. This includes providing meaningful and independent oversight of the implementation of law and policy, promoting transparency in government decision-making, and drawing attention to potential systemic problems in the administration of government. The tribunal itself must be accountable for the exercise of its powers through processes of appellate and judicial review.

(4) User experience – the tribunal should operate in a manner that engenders public confidence in the fairness of its procedures and outcomes. This requires attention to accessibility on a range of fronts, including practical access issues (eg the format and location of hearings), the provision of adequate support to parties, and adopting a flexible and informal approach that is responsive to a party’s particular background and circumstances.

(5) Efficiency – the tribunal should provide an efficient mechanism for review of government decisions. Case management should be directed at securing the best and preferable outcome without undue expense or delay, including by limiting technicality, adopting a flexible approach, and by offering offer practical support to parties to reduce the time and cost of proceedings. More broadly, there should be a focus on continuous improvement of the administrative law system to increase its effectiveness and efficiency. This includes continuous improvement within the merits review process (eg provision of support and training to members), and within government decision-making practices to reduce reliance on merits review (eg development of best practice guides and addressing systemic issues which give rise to requests for review).

The current AAT objectives remain appropriate, though we agree with views noted in the Issues Paper to the effect that the AAT has not necessarily functioned consistently with those objectives in practice. A revised set of objectives should expand on those contained in s 2A of the AAT Act to make explicit reference to accountability, independence, and concern for systemic issues.


(1) Accountability

Effective accountability is the foundation for the legitimacy of government. Integrity institutions, including merits review tribunals, are part of the core machinery through which accountability is delivered. Public trust in accountability mechanisms on the one hand, and public trust in government on the other, are inextricably linked:

“effective accountability mechanisms are capable of fostering citizens’ faith in a system of governance. Conversely, substandard accountability mechanisms may produce the inverse effect of eroding public confidence in the government” (see Ellen Rock, Measuring Accountability in Public Governance Regimes (Cambridge University Press, 2020) p 40)

In order to support public trust in the government, it is critical that the tribunal performs effectively in its accountability function of reviewing government decisions on the merits. However, it is equally important that the public perceives the tribunal to be effective in performing that function. A perceived accountability problem can be just as detrimental to public trust as an actual shortfall in accountability. For this reason, the tribunal must (both in fact and perception) operate in a manner that is independent, impartial and fair.


(2) Independence

Independence is widely regarded as an important criterion for public confidence in integrity institutions, of which merits review tribunals form a part (eg B Ackerman, ‘The New Separation of Powers’ (2000) 113 Harvard Law Review 633; C Field, ‘The Fourth Branch of Government’ 72 AIAL Forum 24; AJ Brown, ‘The Integrity Branch’ in Matthew Groves (ed), Modern Administrative Law in Australia (CUP, 2014) 301). The ARC considered independence to be “essential to the credibility of review tribunals and of the merits review system as a whole” (ARC, Better Decisions: Review of Commonwealth Merits Review Tribunals (Report No. 39, 14 September 1995) ix).

As with accountability, independence is both a matter of practice and perception, and failures on either front can be problematic. It is for this reason that administrative law embodies rules to protect against actual bias in decision-making as well as the apprehension of bias (see eg Matthew Groves, ‘Clarity and Complexity in the Bias Rule’ (2020) 44 Melbourne University Law Review 565). Whether or not an illegitimate influence actually overtakes the exercise of decision-making power, the perceived risk of that possibility is a threat to the integrity of the decision-making process. Making explicit reference to independence and public trust within the objects clause would be welcome.


(3) Systemic issues

Merits review is by nature focused on the resolution of individual grievances. That ought to remain the core focus of the tribunal’s review task. However, the broader system of administrative review should also reflect concern for identifying systemic issues, such as the implementation of problematic policies or practices which may lead to poor decision-making or “un-preferable” outcomes in cases on a wider scale.

Addressing systemic issues is clearly important from an accountability perspective. However, tackling problems at the “source” is also an important way in which the administrative law system can reduce the burden on merits review mechanisms. To borrow from a description regarding the role of the Ombudsman:

“our role in the early days was to resolve complaints (or swat the flies), rather than tackle the systemic issues within an agency which led to the complaints in the first place (hunt the lions). As time has passed, so has our focus. We now set out to hunt more lions, with the belief that if we remove the things that attract flies (the complaints), the flies will disappear of their own accord.” (R Glenn, ‘Keynote Address’ (Speech delivered at the Tax Institute 2014 Tasmanian State Convention, Launceston, 16–17 October 2014)

That analogy is equally apt to the merits review task. Addressing systemic issues before they translate into undesired outcomes not only reduces the negative impacts of problematic government decision-making on individuals, but also increases the efficiency of the merits review system as a whole. This objective should form part of the aims of the new legislative regime.

The ARC played an important and valuable role during the time of its operation. We agree that it, or a similar body, should be established to perform an oversight role in the administrative law context.

While we welcome the establishment of a new ARC body, we note that its ability to perform this important oversight role is largely dependent on government commitment to resourcing it. We recommend giving consideration to alternate funding models to reduce the likelihood of a future government de-funding the ARC as occurred in 2015.

The functions of the new ARC body should include broad responsibility for monitoring the Commonwealth administrative law system and making recommendations for improvements to that system, consistent with s 51 of the AAT Act.

The new ARC body should be specifically empowered to consider systemic issues arising in the context of government decision-making. As noted in the Issues Paper, failures in the implementation and subsequent detection of flaws in the Robodebt Scheme are a prime example of a systemic administrative law problem. The ARC was live to the potential issues associated with automated decision-making, publishing a better practice guide on that topic almost two decades ago (ARC, ‘Automated Assistance in Administrative Decision Making’ (Report No 46, November 2004). Had the ARC been active after 2015, it is possible that the issues with the Robodebt Scheme may have received more specific attention at an earlier stage.

We expand on our views about the potential role of the ARC in responding to systemic issues in response to the related question on “Decisions and Appeals” below.

The new legislation should adopt a two-pronged approach to promoting the continuous improvement of administrative decision-making:

(1) Modelling best practice –
As noted above, the tribunal may play an important role in the continuous improvement of administrative decision-making by undertaking its merits review functions, including by determining the correct and preferable decision in individual cases or providing recommendations on remittal. More generally, the tribunal can influence continuous improvement in agency practices by modelling best practice in administrative decision-making in its own case management, hearings and decisions.

(2) Developing best practice –
If re-established, the ARC may play a more active role in fostering improvement in government decision-making. The functions specified in s 51(1)(ab) of the AAT Act include inquiring into current decision-making practices, and consulting and advising on those practices. A similar function may be given to the new ARC body, allowing it to directly work with government agencies to improve decision-making, and to reduce the number of decisions which must later be corrected on merits or judicial review.

We note that the CATs in both NSW and SA have presidents with backgrounds in legal public sector leadership – specifically as State Solicitors or Solicitors General. By contrast, most presidents of the AAT (and other state CATs) have come from the bar or judicial backgrounds. This may have contributed to the excessive legalisation of which the AAT was often accused. Experience and leadership of the kind that Justices Armstrong and Hughes have seems particularly well-suited to leading a tribunal and each is regarded as an excellent President. Whomever is responsible for appointing the president of the new federal tribunal should broaden their consideration of suitable candidates beyond barristers and judges to other legally qualified people with significant organisational leadership experience.

We strongly support a merit-based appointment process being included in the legislation. Professors Gabrielle Appleby and Andrew Lynch have done extensive work looking at different merit-based appointment models, which we endorse.

In designing the process and selection criteria, public trust should be a core consideration. As noted in response to an earlier question, public trust in integrity bodies is the end product of both actual accountability and perceived accountability; actual independence and perceived independence; actual fairness and perceived fairness. In order to satisfy those public perceptions, it is critical that:

(1) the relevant selection criteria meet public expectations (eg a focus on merit and diversity rather than open-ended discretion);

(2) the structure of the appointments process is clear and transparent (ie the criteria and framework are visible to the public); and

(3) the appointments process is implemented properly and transparently (ie the actual process of appointing members is open to public view and is faithful to the established process).

We would support a fully merit-based appointment model, without political involvement.

We see no good reason why the Minister should be responsible for reappointments. If an experienced member wishes to be reappointed and the President supports reappointment, the member should be automatically reappointed. This will retain experienced, high quality members and avoid some of the problems with inexperienced members that has contributed to the current backlog in the AAT.

Knowing that their reappointment does not depend on the Minister’s satisfaction with a member's decisions would also contribute to the independence of members and the new tribunal. That point is captured neatly in the comments made by Acting President Logan in Singh (Migration) [2017] AATA 850:

“The very existence of the Tribunal and the independent, quasi‐judicial model adopted for it means that, inevitably, there will be tension from time to time [with] Ministers and others whose decisions are under review … [However] any member who allowed himself or herself to be persuaded as to an outcome by partisan or political rhetoric by a Minister … would be unworthy of the trust and confidence placed in him … For those members who do not enjoy the same security of tenure as judges, that may call at times for singular moral courage and depth of character.” (at [17]–[18])

Removing political oversight of the reappointment process would reduce the need for members to rely on their own “depth of character” when those inevitable tensions arise. That is clearly important for supporting the actual independence of tribunal members. However, it is equally important as a matter of public perception, as applicants and the general public can trust that tribunal members are not basing decisions on concern for their own job security. That in turn translates to public trust and confidence in the tribunal as an accountability institution.

Our comments are limited to matters raising systemic issues. We have recommended that one of the objectives of the new legislative regime should be to address systemic issues in the administrative law system. Below, we comment specifically on issues in detection and resolution of systemic issues.

(1) Detecting systemic issues
The tribunal is in a unique position to identify potential systemic issues, having the benefit of oversight over the implementation of government policy and evidence of its application to individual cases. We recommend that one of the tribunal’s formal functions be to identify potential systemic issues. This could be implemented through adopting practices which allow individual members to share details of potential systemic issues through internal reporting structures.

It is important to recognise that not all systemic issues will be apparent within the context of reviewing an individual case (or cases). Individual instances of decision-making may not appear to be problematic or unreasonable when they are taken in isolation. However, those individual instances may constitute problematic decision-making when taken together and viewed in their totality. Identifying underlying systemic issues must therefore be part of a wider oversight project.

An oversight body (we recommend the ARC or its replacement) would be able to perform a valuable role in monitoring issues raised by individual members, as well as identifying potential trends and concerns arising more broadly in the context of tribunal decision-making. In order for that oversight function to be effective, the collection and publication of tribunal decision-making data will be critical. We comment on this at the conclusion of our submission.

(2) Addressing systemic issues
As a general rule, individual disputes are not the most useful vehicle to consider or resolve broader underlying issues. A systemic issue (eg the implementation of a problematic policy) may manifest differently in different cases, and resolving those issues on the merits might demand a different response in different cases. The best way to address a systemic concern may not necessarily be the best way to resolve an individual dispute, and expanding the scope of the proceedings to consider those broader issues is also likely to reduce the efficiency of tribunal dispute resolution.

Where a potential systemic issue is identified within the tribunal, it is not necessary for the tribunal itself to investigate and resolve that issue. We recommend that the tribunal be empowered to refer potential systemic issues for investigation by other bodies, such as:

(a) a new ARC body, which will be well-suited to investigate and report on systemic issues in administrative decision-making practices;

(b) the Ombudsman, which is already empowered to address systemic issues in maladministration; or

(c) the National Anti-Corruption Commission, which has responsibility for investigating potential systemic corruption issues.

Many of the questions raised in the Issues Paper cannot be answered due to lack of data collection by the AAT. For example, there is no data on the effect of self-represented litigants on the tribunal’s processes, the time matters take, and outcomes. Evidence on SRLs is anecdotal and conflicting (see L Richardson, G Grant and J Boughey, ‘The Impacts of Self-Represented Litigants on Civil and Administrative Justice’ (Australasian Institute of Judicial Administration, 2018). There is a particular lack of data on settlement outcomes, as highlighted in the context of the inquiry into the NDIS (see eg PIAC, Submission 33 to Joint Standing Committee on the National Disability Insurance Scheme, Parliament of Australia, Inquiry into General issues around the implementation and performance of the NDIS (13 July 2020) [2.1]).

We have suggested that (1) accountability and (2) continuous improvement should be key design principles that inform the structure and operations of the new tribunal. Both of those objectives are largely dependent on access to detailed and accurate data, which would inform evidence-based decisions about the tribunal’s operation and use of resources, and identification of areas for improvement within the broader system. The new tribunal should be required, and funded, to collect data on its work.