Skip to content

The Hon. Justice Duncan Kerr Chev LH

Excellence in Government Decision-Making: An AGS Symposium, Canberra
21 June 2013

Little has been written about how an administrative tribunal that is 'not bound by the rules of evidence' exercising merits review best can be informed about facts and opinions relevant to its functions. Apart from Dennis Pearce's Administrative Appeals Tribunal (very recently in its 3rd edition) and JRS Forbes' Justice in Tribunals there has been limited academic interest in examining the vast underlying canvas of tribunal practice.

By contrast there is considerable academic and professional writing from the perspective of judicial review. That literature has focused upon the circumstances in which a tribunal is to be found to have fallen into error. Risking a sweeping generalisation tribunals and their members have been left to infer what is permissible and correct through a process of eliminating that which is prohibited.

An exception to the general academic disinterest in tribunal procedures is Professor Neil Rees' 'Procedure and evidence in "court substitute tribunals"' (2006) 28 Australian Bar Review 41.

Rees argues that courts have been rarely troubled by issues regarding procedural fairness because the requirements of the fair hearing rule 'are built into the procedural and evidentiary rules of law that govern [their] proceedings...' (1) If the rules of evidence do not apply that safety net is removed:

The parties, and their lawyers if they are represented, will be unaware of the way in which the claim in issue, and any evidence which supports or opposes it, should be prepared and presented. (2)

In Rees' opinion this problem would be overcome if 'court substitute' tribunals were to apply the Uniform Evidence Act and adopt relatively detailed procedural rules specifically designed for tribunals. (3)

It may be unkind to respond to Rees' substantial and thought-provoking contribution to discussion of tribunal practices to devote much of this paper to challenging his basic premises but, in my opinion, both his analysis and his conclusion are contestable.

Rees uses the term 'court substitute' tribunals as a 'catch-all' expression to describe the many and various statutory Australian decision making bodies that can exercise both administrative and judicial power. However, the choice of that term appears to assume the answer he then provides. To refer to such bodies as 'court substitute' tribunals suggests a predetermined view that their substantive function is identical to that which otherwise would be vested in the ordinary hierarchy of the judicial system.

If the work of such bodies substitutes for the work of the courts, Rees' conclusion that they should behave as courts do is hardly surprising. It would be hard to contest such a circular conclusion.

However, referring not only to state courts conferred with administrative power but also to state and federal administrative bodies conducting independent merits review in a rolled up way as 'court substitute' tribunals, in my opinion, ignores or blurs a key distinction fundamental to administrative law.

That of course is why Rees' article has almost nothing to say of Commonwealth administrative tribunals which, for constitutional reasons, are unable to exercise any share of judicial power.

Ch III of the Constitution mandates a strict separation of powers which prevents the conferral of judicial power on any tribunal other than a court and precludes the admixture of Commonwealth judicial and non-judicial functions on a federal tribunal.

Merits review - that is the function of evaluating and substituting the correct or preferable decision standing in the place of a decision maker, as opposed to enforcing the law that constrains and limits the powers of the other branches of government -is, on that analysis, beyond judicial power.

The focus of Rees' article is therefore restricted to State 'court substitute tribunals' where, although subject to the Kable doctrine, the separation of powers doctrine does not apply in the same strict form as it does to the Commonwealth.

However, in my view, the underlying distinction between judicial review and administrative decision making and merits review is equally relevant to the functions of state tribunals undertaking administrative functions even if they are also invested with judicial power.

The fundamental importance of that distinction has since been reinforced by the entrenchment of judicial review of executive action as constitutionally inherent in the nature and functions of State Supreme Courts in consequence of the High Court of Australia's decision in Kirk. (4)

Judicial review contrasted with merits review.

Merits review is exclusively a creature of statute. Once rare, merits review has become an entrenched part of the Australian legal system. (5) It is more frequently accessed than is judicial review. Courts deal with thousands, perhaps even tens of thousands of applications for judicial review each year but merits review tribunals deal with hundreds of thousands of applications in the same period. Proceedings in such tribunals are, in the ordinary course, meant to be less expensive and far less daunting for participants. The AAT, and State tribunals which share the AAT's heritage, operate without fanfare offering relatively convenient, informal and cheap processes whereby citizens can challenge adverse administrative decisions. Merits review allows those tribunals to reach the correct or preferable decision - not merely to set aside a flawed decision and send it back for reconsideration.

By contrast judicial review exists to correct legal errors. In Attorney General (NSW) v Quin (1990) 170 CLR 1, at 35-36 Brennan J stated:

The duty and jurisdiction of the court to review administrative action do not go beyond the declaration and enforcing of the law which determines the limits and governs the exercise of the repository's power.

Even conceding that in practice the distinction between correcting legal error and examining the merits of a decision may sometimes be blurred 'there is in Australian legal theory a bright line between judicial review and merits review.' (6)

Judicial review can set flawed decisions aside and can compel unperformed legal duties to be performed but it cannot substitute for a flawed decision what the court thinks to be the correct or preferable decision. That is a step beyond judicial power. Only merits review can achieve that outcome.

Administrative power and merits review: on what basis should decisions be made?

Public law administrative decisions are usually made by State or Commonwealth ministers and public servants. In many such instances parties can exercise a right of review conferred by statute to an independent tribunal. In some, but rarer, instances the original decision making function itself is directly conferred on such a tribunal. (7) Whether directly conferred, or by way of review, the decision making function of such a tribunal, remains an exercise of executive rather than judicial power.

We do not expect ministers or public servants to reach their decisions only after testing the materials they rely upon against the admissibility criteria that operate in courts. It has never been suggested that such decision makers should do so.

What is expected of such an administrative decision maker is that they be unbiased and act fairly. (8) We expect them to comply with the rules of natural justice.

Where a primary decision is made by a public servant or minister and review power is conferred upon a tribunal the tribunal's task, under what has become the common form grant of power, is to reach the correct and preferable decision. A useful short-hand statement to explain the role of a tribunal undertaking that task is to say that it stands in the shoes of the person who made the decision. What that means is that, to the extent the law confers jurisdiction on the tribunal, the tribunal can exercise the same powers and discretions that the law confers on the original decision maker.

While merits review allows the opportunity for a more intensive examination of the circumstances of a challenged decision, and as a result the tribunal will often receive information and submissions additional to those before the public servant or a minister who made the decision subject to review, it is a category error to conceive of that function as substituting for judicial power even if the power to do so is invested in a body which also is a court as is constitutionally permissible at the State level. (9)

Given that primary administrative decision makers are not burdened by the rules of evidence and may, subject to the rules of natural justice, properly rely on any probative materials relevant to their functions, there is no self-evident jurisprudential reason for a tribunal standing in those shoes, in the ordinary course and in the absence of contrary statutory direction, (10) to place itself and the parties appearing before it in any different and more constrained position.

Statutory Directions

Moreover, rather than expressing the direction that tribunals should mirror courts in their procedures and in the manner of their receiving evidence, most of the statutory provisions that empower modern tribunals emphasise the importance of their accessibility, informality and procedural and evidentiary flexibility.

For example s 2A of the Administrative Appeals Tribunal Act 1975 (AAT Act) directs that in carrying out its functions;

the Tribunal must pursue the objective of providing a mechanism of review that is fair, just, economical, informal and quick.

Section 33(1)(b) of the AAT Act then directs that the proceedings of the AAT;

shall be conducted with as little formality and technicality, and with as much expedition...as a proper consideration of the matters before the Tribunal permit.

In my view section 33(1)(c) of the AAT Act that states;

the Tribunal is not bound by the rules of evidence but may inform itself on any matter in such manner as it thinks appropriate

is to be read in the light of the AAT Act as a whole including the two preceding statements of principle. Understood in that context s 33(1)(c) is not a grant of power occasionally to depart from the strict application of the rules of evidence; rather it presupposes and establishes a scheme for the reception of evidence in which, notwithstanding that the procedure of the Tribunal always remains within the Tribunal's independent control, (11) compliance with those rules is intended to be foreign.

Similar statements of principle are mirrored in other Commonwealth and State tribunals' legislation. It would be inconsistent with those principles for such a tribunal to implement Rees' proposals by adopting practice directions mandating the application of the Uniform Evidence Act.

Moreover, putting to one side those considerations, there are strong reasons to doubt some of Rees' key assumptions. Rees bases the argument that 'court substitute tribunals' should adopt the Uniform Evidence Act on 'modern theoretical writings concerned with the choice between broad principles and specific rules as a mode of regulation'. He contends that those theoretical writings support the notion that whenever the type of action to be regulated is complex, 'binding principles backing non-binding rules tend to regulate with greater certainty than principles alone'. (12)

That theory is at best contestable. It is salutary to look at what happened in the aftermath of the enactment by the Federal Parliament of procedural codes to govern the reception of evidence in the Migration Review Tribunal and the Refugee Review Tribunal.

At paragraph 6.21 of its most recent report Federal Judicial Review in Australia (2012) the Administrative Review Council (ARC) noted:

The MRT-RRT submitted that the procedural codes enabled by s 424A of the Migration Act is the subject of significant litigation without enhancing the quality of decision making by the Tribunals.

In consequence the ARC reported:

In matters remitted to the MRT-RRT with s 424A as the principal ground of contention, courts have found that tribunal action would have satisfied common law procedural fairness, but amounted to jurisdictional error in falling short of the requirements of a procedural code... The Council is mindful that the extraordinary amount of litigation and administrative work associated with the code means the scale of this issue is large. Overall, the Council notes that endeavours to achieve compliance with the code have not necessarily enhanced the fairness of the review process, at considerable cost to efficiency and increased litigation.

Accordingly the empirical evidence after the MRT-RRT experiment of attempting to codify the rules of procedural fairness suggests there may be some reason to be sceptical that the measures Rees suggests would work as he conceived. Rather codifying the rules might generate additional grounds of appeal with consequential unnecessary litigation and higher costs.

There is also reason to doubt Rees' associated thesis that courts are untroubled by issues regarding procedural fairness because the requirements of the fair hearing rule 'are built into the procedural and evidentiary rules of law that govern proceedings in the court'. (13)

Assuming, albeit without data, it to be true that courts are troubled less by issues of procedural fairness than are tribunals, that may simply be the product of their being presided over by judges who, by reason of their long legal training, have better internalised those notions than will be the case with the often part time members of tribunals, rather than because they were required to apply the evidentiary rules of law that govern the reception of evidence in criminal and civil litigation.

Moreover, it is far from clear that the existence of procedural and evidential rules per se immunises judges in respect of the obligations of procedural fairness. Natural justice involves considerations beyond those addressed by the rules of evidence. A recent case illustrates that point. In Straits Exploration (Australia) Pty Ltd v Kokatha Uwankara Native Title Claimants [2012] SASCFR 121, (2012) 114 SASR 516 a South Australian District Court judge, exercising an administrative function, but thinking he was exercising judicial power and applying the rules of evidence, nonetheless was held to have denied procedural fairness to applicants seeking a licence to undertake mining exploration.

Yet there is great force in Rees' observation that when the ground rules that are to be applied by a tribunal are not made clear from the outset, applicants and respondents, and their lawyers if they are represented, may be unaware of the way in which a claim in issue, and any evidence which supports or opposes it, should be prepared and presented. That is clearly undesirable.

However, adopting the rules of evidence is not the only way tribunals can ensure that parties will know what to expect.

It is understandable that lawyers familiar with and expert in the way things are done in court, and arguably privileged by expertise in contrast to unrepresented litigants, might want those rules to also apply in any tribunal in which they appear. It is equally understandable that such lawyers might, as does Rees, share the view expressed obiter dicta by Evatt J in R v War Pensions Entitlement Appeal Tribunal (1933) 50 CLR 228 at 256 that every departure from the rules of evidence will necessarily advantage one side and necessarily disadvantage the other. However, at least in my view, that proposition is untenable in the case of modern tribunals given their substantial workloads and their obligations of fairness, informality and speed. Tribunals must be equally open to the unrepresented as to those who can afford counsel. Adoption of Rees' proposals would, in my opinion, unfairly tip the balance against those unfamiliar with the codified rules of evidence that apply in most Australian courts.

The critical point of difference between Rees' answer to the question of how parties are to be made aware of the way in which any claim in issue and evidence which supports it or opposes it should be prepared and presented, and mine is that Rees contends that the ground rules for reception of evidence in a tribunal should be those of the Uniform Evidence Act while I contend that provisions such as s 33 of the AAT Act require that the admissibility of 'evidence' in merits review tribunals should be determined exclusively by the 'limits of relevance' rather than 'the interstices of the rules of evidence' as was elegantly concluded by Hill J in Casey v Repatriation Commission. (14) The task of a merits review tribunal is to give such weight to whatever relevant evidential material is before it as it determines it ought to bear. (15)

I conceive of this as conferring on merits review tribunals the freedom to take into account all of the relevant testimony, materials and circumstances known to it removed from the strictures of the rules of evidence. However, that freedom is not at large. It is a freedom to be fair.

Rees rightly is concerned to ensure parties will know what to expect when they appear before a tribunal. Fairness requires parties should know in advance what materials they will be permitted to adduce.

Making it clear from the outset that it will be relevance and not the rules of evidence that will govern the test of what merits review tribunals will permit to be adduced as 'evidence' provides equal certainty to parties.

In my opinion whenever a merits review tribunal which, by reason of its governing statute, is not bound by the rules of evidence requires, or considers requiring, compliance with formalities limiting the presentation of otherwise relevant materials, procedural fairness requires that that tribunal makes those circumstances known, either by well publicised general practice directions or by notice to the parties, well in advance of any hearing where such question may arise. (16) Otherwise the parties whose interests are at stake (and particularly unrepresented parties) are entitled to assume that provisions such as s 33 of the AAT Act mean what they say - that the processes will be conducted without formality and technicality.

Applicants and respondents should not be placed in the position of turning up to a hearing assuming (reasonably) that the tribunal will not apply the rules of evidence only to discover that what they had proposed to put to the tribunal is open to contest and arguments about admissibility (save those relating to relevance) such that the merits become obscured or never reached.

Of course tribunals remain bound to respect all proper claims of legal professional privilege and the rules against compelled self-incrimination. Those are commonly referred to as exclusionary rules of evidence but the High Court has unambiguously revealed that to be erroneous. The entitlements conferred on a party in respect of such claims are not procedural; rather they are substantive legal rights. They are excluded from the considerations discussed above.

Beyond theory: how tribunals actually go about their work

Although I have set out what I regard to be the correct approach that should govern the reception of 'evidence' in merits review tribunals there is a near complete absence of empirical studies to permit any assessment of whether that is how such tribunals actually go about that task. The only study I have discovered, although somewhat dated, casts some doubt on that assumption. However, it reinforces a point Rees and I share in common: that tribunals should recognise the problem caused by uncertainty regarding their procedures and should move to overcome it.

That study is described in Terese Henning and John Blackwood 'The Rules of Evidence and the Right to Procedural Fairness in Proceedings of Four Tasmanian Quasi-Judicial Tribunals'. (17) The authors' explanation of their reasons for commencing a study of the subject is worth setting out. They stated:

Throughout the history and development of the tribunal system in Australia, one of the more contentious issues has been the extent to which tribunals follow or are influenced by the rules of evidence that apply in more formal court proceedings. Part of the problem is that courts undertaking judicial review often impose a traditional legal construct on the operations of tribunals and may too readily overturn decisions that display a more informal or inquisitorial procedure. There is a related concern that tribunals often attempt to turn themselves into courts and to develop a formal legal paradigm. Of particular concern is the strategic or tactical use of the rules of evidence to prevent cogent evidence being adduced at tribunal hearings.

It was precisely those concerns that in 1999 led the Medical Council of Tasmania to seek the advice of [one of the authors]....The council, through its Medical Complaints Tribunal investigates complaints against medical practitioners for unethical behaviour or misconduct. The council was concerned that the rules of evidence and formal court procedures had become so institutionalised that the tribunal was being prevented from discharging its protective functions.

The result was....limited to a theoretical explanation of the applicable law. It was apparent that if the matter was to progress further, it would be necessary to obtain solid empirical evidence about the matters of concern. (18)

Four tribunals, the Medical Council of Tasmania, the Guardianship and Administration Board of Tasmania, the Anti-Discrimination Tribunal of Tasmania and the Resource Management and Planning Appeals Tribunal of Tasmania participated in Henning and Blackwood's study.

The article reports their findings. It reveals that there was a considerable variation between, and within, those tribunals as to how they dealt with evidentiary issues. Some examples in the report are testimony to how deeply uncomfortable some lawyers appearing before the tribunals became when they were asked to avoid such formalities. (19)

Recourse to the rules of evidence continued in those tribunals despite what the authors refer to as 'the legislative imprimatur for their non-observance'. (20) The authors concluded that the rules of evidence were not institutionalised and entrenched in any of the tribunals studied but that an environment for inconsistencies to occur had arisen because those rules continued to be applied from time to time without any explicit rationale. The authors concluded:

What remains unclear and of interest is precisely when and on what basis those rules will be applied or disregarded. When evidentiary rules have been applied to particular evidence, the tribunal concerned has not articulated its reasons for doing so. Similarly, where the tribunals have declined to accede to a party's request to apply the rules, they have not given any precise indication of the reasoning or of the adjectival principles governing their decision...Represented and unrepresented parties as well as lay members of tribunals are, therefore, unable to predict with any degree of confidence just how particular rules of evidence are likely to operate from case to case.....Finally, it creates the creates the environment for the rules to be manipulated unfairly. These matters will...be subject to further investigation and report in the concluding year of the study. (21)

Unfortunately there was to be no further investigation and report. What Henning and Blackwood expected would be the second stage of the study did not proceed after one of the tribunals withdrew from continued participation.

Yet it seems well enough established by Henning and Blackwood's initial study that inconsistency remains a troubling issue.

Rees proposes one solution to ensure greater consistency - that tribunals should routinely apply the rules of evidence as provided for in the Uniform Evidence Act. For the reasons advanced previously I do not think that solution is consistent with what the law requires. (22)

Henning and Blackwood were correct, in my view, to conclude that there is a legislative imprimatur for their non-observance.

The better solution, in my opinion, is that tribunals should not, except where clear notice has been given in advance, apply any filter, save that of relevance and fairness, to the materials they may receive. In my opinion, subject to the obligatory exclusionary rules applying to legal professional privilege and self-incrimination and subject to the hearing rule of procedural fairness, a party is entitled to expect that a tribunal which is not bound by the rules of evidence that is undertaking merits review will not exclude from its consideration any material put before it that is relevant. The weight to be given to any such material should then be for the Tribunal to determine. (23)

Proof in particular circumstances

I want to conclude this paper with a short discussion of two particular issues which commonly come before tribunals. Those issues are often discussed as if they involve a question of whether or not to apply the rules of evidence whereas, in my opinion, properly understood, they do not involve that question.

Lies

One of the more difficult issues confronting tribunals is how to respond to lies told by an applicant. Most applicants seek to tell the truth - some of course will make mistakes in recall, some will exaggerate or be selective in their recall knowingly or by time forming false memory: few will out and out lie.

Experienced decision makers recognise how hard it can be for applicants to maintain disinterest in their own proceedings and do not place too great a weight on such minor frailties. But when we detect an out and out lie there is a natural temptation not only to allow the lie to destroy the credit of the witness but also to regard the lie as relevant to establishing the truth of whatever has been put against them.

However, ordinarily the telling of a lie merely affects the credit of the witness who tells it.

There is no difficulty with a decision maker discounting what has been told to him or her by a liar in so far as that discount goes to negativing belief, or inducing scepticism, in any positive account provided by a witness or applicant.

The difficulty I am concerned with is when a lie is additionally drawn on as the basis of making a specific finding adverse to an applicant who has lied. Because of the natural inclination to assume (wrongly) that a lie is proof of what the opposite party is asserting, judges in criminal proceedings must explain the flaw in that logic to lay jurors. The judge must explain that a lie can be drawn on to support such a conclusion only in specific circumstances.

In Edwards v The Queen, Deane, Dawson and Gaudron JJ held in that for a lie to be capable of being relied upon to prove guilt that lie 'must relate to a material issue' and the telling of the lie 'must be explicable only on the basis that the truth would implicate the accused in the offence with which he is charged. It must be for that reason that he tells the lie.' (24) It must reveal consciousness of guilt.

A merits review tribunal is not bound by the rules of evidence but the propositions affirmed in Edwards are, in my opinion, not rules of evidence. Rather they are principles of logic applying to fact finding. There is no reason not to regard those principles as having universal application in respect of the circumstances in which Australian jurisprudence will permit lies to be regarded as relevant to proof. (25)

Given my contention that the overarching principle for the admissibility of evidence in merits review tribunals should be that of relevance I see no inconsistency between the continuing applicability of the principles stated in Edwards and the proposition that the rules of evidence do not apply in a tribunal.

Briginshaw

In Briginshaw v Briginshaw (1938) 60 CLR 336 the High Court of Australia, in a complex and much analysed decision, (26) held that in cases involving the civil standard of proof, where grave and serious allegations have been made, a court cannot come to a 'reasonable satisfaction' that the allegation has been established on the balance of probabilities unless the decision maker 'feels an actual persuasion' feels 'comfortably satisfied' and is not 'oppressed by reasonable doubt'.

Briginshaw is commonly asserted to require, as a matter of law, that the more serious and grave, involving moral opprobrium, the nature and circumstances of the allegations are, the stronger and more reliable the evidence put forward to prove the facts in issue will need to be. (27)

It appears to have been accepted on that basis by a number of professional disciplinary tribunals; Re a Solicitor; ex parte the Prothonotary (28) (allegations of fraud and perjury by a legal practitioner), Hobart v Medical Board of Victoria (29) (allegations of a doctor's improper relationship with a patient and supply of drugs) and Kerin v Legal Practitioners Complaints Committee (30) (allegations of a legal practitioner's conflicts of interest, misleading the Complaints Committee and illegally importing a firearm).

It also appears to have been applied on the same understanding in all Australian anti-discrimination jurisdictions; albeit 'without examining whether it is warranted'. (31)

But there are several caveats I want raise against Briginshaw's unexamined application in merits review tribunals.

First, to the extent it is a 'rule of evidence' there has been a legislative disapplication of the requirement for it to be applied for that reason. If Briginshaw is to be applied in merits review it must have some other policy or legal justification.

Second, Briginshaw does not change the burden of proof. Changes to the burden of proof must be provided for in the relevant legislation. (32)

The third caveat follows as a matter of logic from the second.

Briginshaw assumes the existence of a (civil) standard of proof. With some exceptions  - for example the tax matters in the AAT where the taxpayer bears the burden of proof of establishing that the Commissioner's assessment was excessive  - there is no burden of proof or onus, civil or criminal on an applicant or a respondent in merits review proceedings. (33) The proceedings are inquisitorial in character (34) and the duty of the tribunal is to reach the correct and preferable decision as if it was standing in the shoes of the original administrative decision maker.

This is not to suggest, assuming that the consequence of finding particular facts would be serious for an applicant, that a merits review tribunal should ignore that consequence in concluding that such conduct had occurred. (35) Administrative law has its own mechanisms to ensure that a decision maker acts within jurisdiction: it would offend Wednesbury principles and natural justice to make findings carrying serious consequences without the proper basis first having been established. (36)

But proper regard for the seriousness of the consequences, in my opinion, is inherent in the statutory duty and does not depend on the Briginshaw principles which turn on the boundaries between egregious and non-egregious conduct. In many statutory contexts there will not be any policy reason for an administrative decision maker to apply that distinction.

Consider, for example, the AAT's function to review decisions to suspend the right of a financial advisor to engage in his or her business (on review from ASIC) or cancel a pilot's licence (on review from CASA). Assume that the allegations in the first instance relate to incompetent but not self-interested advice. Assume that the allegations in the second relate to the capacity of the pilot without any suggestion of his or her wilfully putting the public in danger.

Such allegations involve no moral opprobrium. They would not attract the Briginshaw principles.

Now assume the allegations in the first instance were to relate to forgery and fraudulent defalcations against the applicant's clients. Assume the allegations in the second instance relate to knowingly flying an un-airworthy aircraft and dangerous low-flying and stunts overhead of densely populated suburbs.

Those would clearly involve 'serious and grave' allegations - which counsel almost certainly would contend, if the proceedings were in a court and applying the civil standard of proof, ought to attract the Briginshaw principles.

Counsel put identical submissions in administrative proceedings despite their constitutionally different character. Out of caution administrative decision makers have continued to use the language of Briginshaw as prophylactic against unnecessary risk of appeal or judicial review.

However, it is far from clear what principle requires, in merits review proceedings which are inquisitorial in nature and where there is no burden of proof, that the more egregious the conduct alleged by a regulator - and the greater the potential, if the allegation is correct, for harm to be occasioned to the community - the harder it must be for the regulator to sustain its case on review.

In that regard I find some comfort in what was said by Woodward J in McDonald v Director General of Social Security (37) that a tribunal determines matters by reference to 'the statutes under which it is operating, or in considerations of natural justice or common sense, [rather] than in the technical rules relating to onus of proof developed by the courts'. Briginshaw, which has generated some of the most arcane of the technical rules relating to onus of proof, must yield to that observation.

In my opinion, if the underlying statute law governing the powers and functions of an administrative decision maker truly involves not only matters of importance to an applicant but also the protection of the public, then on review both the severity of the potential consequences for the applicant and the protective nature of the power conferred on the regulator will be relevant factors; neither of which is to be ignored by a merits review tribunal in its task of reaching the correct and preferable decision.

Footnotes

(1) At 55.

(2) At 85.

(3) At 86-87.

(4) Kirk v Industrial Relations Commission; Kirk Group Holdings Pty Ltd v WorkCover Authority of New South Wales (Inspector Childs) [2010] HCA 1, (2010) 239 CLR 531.

(5) See D Kerr, 'The intersection of merits and judicial review: looking forward' (2013) 32(1) University of Queensland Law Journal (forthcoming).

(6) Stephen Gageler, 'The Legitimate Scope of Judicial Review' (2001) 21 Australian Bar Review 279, 279.

(7) It is relatively more common for such functions to be directly conferred on a tribunal at the State level.

(8) The role of administrative decision makers, including merits review tribunals, as conduct of the executive branch of government, cannot be excluded from judicial review. Administrative tribunals are subject to the supervisory jurisdiction of either the High Court of Australia pursuant to section 75(v) of the Constitution or, at the state level, the inherent jurisdiction of the several Supreme Courts: Kirk v Industrial Court (NSW) (2010) 239 CLR 531. The supervisory function extends to ensuring that administrative decision makers do not violate the rules of procedural fairness - either the bias or the fair hearing rule.

(9) See Straits Exploration (Australia) Pty Ltd v Kokatha Uwankara Native Title Claimants [2012] SASCFR 121, (2012) 114 SASR 516 in which it was held that a District Court Judge sitting as a Judge of the Environment, Resources and Development Court of South Australia to determine whether mining could be conducted at a site at Late Torrens was exercising administrative rather than judicial power. Provided there is no Kable inconsistency the conferral of quite broad non-judicial power on a State court is constitutionally permissible. It would not be possible at the Commonwealth level.

(10) The usual position for tribunals shaped on the AAT model is that tribunal processes are inquisitorial in nature and strictly without a contradictor (Bushell v Repatriation Commission: [1992] HCA 47; (1992) 175 CLR 408; Minister for Immigration and Citizenship v Li [2013] HCA 18 (8 May 2013)). In Confidential and Commissioner of Taxation [2013] AATA 382 Senior Member Fice distinguished those cases in reaching the conclusion that that those propositions do not apply in the taxation jurisdiction of the AAT where the applicant bears the onus of proof. Whether the existence of an onus of proof means that review proceedings assume an adversarial character has yet to be judicially addressed. Whichever way that matter is ultimately resolved SM Fice did not suggest that the rules of evidence applied to such proceedings.

(11) AAT Act s33(1)(a).

(12) At 86, citing J Braithwaite 'Rules and Principles: A theory of legal certainty' (2002) 27 Australian Journal of Legal Philosophy 47 at 75.

(13) Rees, at 55.

(14) (1995) 60 FCR 510 at 514.

(15) This is not to suggest that a merits review tribunal is unconstrained in that regard. Modern tribunals are almost always obliged to give reasons. If the weight placed upon particular evidence leads to the tribunal making decision which is so unreasonable that no reasonable decision-maker could have arrived at it reviewable legal error will have occurred; Associated Provincial Picture Houses Ltd v Wednesbury Corporation [1948] 1 KB 223 a review ground reinvigorated by the recent decision of the High Court in Minister for Immigration and Citizenship v Li [2013] HCA 18 (8 May 2013). Legal error may also be shown if the reasoning processes of the tribunal are irrational or illogical; acknowledging that the scope and availability of review in respect of those grounds remains to be settled following the decision of the High Court in Minister for Immigration and Citizenship v SZMDS [2010] HCA 16; (2010) 240 CLR 611. Nor do I intend to suggest that the logic underlying the principles of common law and statutory rules of evidence, if such can be discerned, should be entirely disregarded in evaluating the 'evidence' adduced in a merits review tribunal but those considerations should be related to the specific circumstances before the tribunal and apply to evaluation of weight rather than admissibility: Re Pochi and Minister for Immigration and Ethnic Affairs (1979) 2 ALD 33 at 40-42.

(16) An example of such advance notice having been given is the AAT's Guidelines for Persons Giving Expert and Opinion Evidence (2011).

(17) (2003) 10 Australian Journal of Administrative Law 84.

(18) At pp 84-5.

(19) Including when lawyers appearing before a tribunal were asked to remain seated - see at p 104.

(20) At p 103.

(21) At p 103.

(22) A tribunal may err by 'disregarding relevant information merely because it might be inadmissible in a court' JRS Forbes, Justice in Tribunals Federation Press 2nd ed. 2006 at p 201; see also Re Pochi (1979) 2 ALD 33 per Brennan J at 40-42, affirmed in Minister for Immigration and Ethnic Affairs v Pochi (1980) 44 FLR 41 (Smithers, Evatt and Deane JJ).

(23) Hill J in Casey v Repatriation Commission 60 FCR 510 at 514 and Secretary Department of Social Security v Jordan 83 FCR 34 at 43.

(24) (1993) 178 CLR 193 at pp 208-209.

(25) See FTZK v Minister for Immigration and Citizenship [2013] FCAFC 44 at [133] per Kerr J (in dissent).

(26) Briginshaw has been described as the Rorschach inkblot of Australian case law: H Bennett and GA Broe 'The civil standard of proof and the "test" in Briginshaw: Is there a neurobiological basis to being "comfortably satisfied"?' (2012) 86 Australian Law Journal 258 at p 258.

(27) Paraphrasing Bennett and Broe (ibid) at 263.

(28) (1939) 56 WN (NSW) 53.

(29) [1966] VR 292.

(30) (1996) 67 SASR 149.

(31) L De Plevitz, 'The Briginshaw "Standard of Proof" in Anti-Discrimination Law: "Pointing with a Wavering Finger"' (2003) 27(2) Melbourne University Law Review 308.

(32) Re Phillips and Inspector-General in Bankruptcy [2012] AATA 788.

(33) Bushell v Repatriation Commission: [1992] HCA 47; (1992) 175 CLR 408; Minister for Immigration and Citizenship v Li [2013] HCA 18 (8 May 2013).

(34) Ibid.

(35) Thus, for example, in Al-Habr and Minister for Immigration in minute detail' and Multicultural Affairs [1999] AATA 150 it was properly held that the AAT should examine 'in minute detail' all of the evidence against an applicant seeking to set aside a decision that he was to be denied the protection of the refugee convention because of his commission of a prior serious non-political crime given the seriousness of the possible consequences for him. See also Pochi and Minister for Immigration and Ethnic Affairs [1979] AATA 64 (24 May 1979) per Brennan P.

(36) It is an ordinary requirement of natural justice that a person bound to act judicially base his decision 'upon material which tends logically to show the existence or non-existence of facts relevant to the issue to be determined': R v Deputy Industrial Injuries Commissioner; Ex parte Moore, (1965) 1 QB 456 at 488 (affirmed in Minister for Immigration and Ethnic Affairs v Pochi [1980] FCA 85; (1980) 44 FLR 41 at 66-67).

(37) (1984) 1 FCR 354 at 356.