Commentary on the Administrative Review Council’s Report No 50 (2012) - Federal Judicial Review in Australia

Nothing like the Curate’s Egg ^[1] –

A paper given on 19 June 2013 to the Constitutional and Administrative Law Section of the New South Wales Bar Association

Justice Alan Robertson, Federal Court of Australia

 RTF version

---

1  This commentary concerns the 15 main recommendations of the Administrative Review Council’s September 2012 Report “Federal Judicial Review in Australia”, a report published by the Administrative Review Council (ARC) after conducting an inquiry of its own motion in accordance with its statutory functions under s 51 of the Administrative Appeals Tribunal Act 1975 (Cth).

2  Time does not permit a consideration of the many other recommendations in the Report concerning the Schedules to the Administrative Decisions (Judicial Review) Act 1977 (Cth) (the AD(JR) Act).

3  I have found little that has been written about this important ARC Report, although there are some references to it in the latest edition of Aronson and Groves Judicial Review of Administrative Action.^[2]

4  There are also two papers in (2013) 72 AIAL Forum. The first is by Professor John McMillan AO and entitled “Restoring the ADJR Act in Federal Judicial Review”, being a paper he presented at an AIAL seminar in Canberra on 4 December 2012.^[3] Professor McMillan, as Australian Information Commissioner, is an ex officio member of the ARC. The second is by Roger Wilkins AO and Bronwen McGee and entitled “Judicial Review: a Jurisdictional Limits Model”, being a paper presented at the same AIAL seminar.^[4] Roger Wilkins AO is Secretary of the Attorney-General’s Department and an appointed member of the ARC. Ms McGee is Senior Legal Officer in the ARC Secretariat.

5  Each paper is directed towards the “primary issue” in the federal judicial review system identified by the ARC,^[5] discussed below, being the “divergence between the two avenues of judicial review available in the Federal Court (review under the ADJR Act and review under s 39B(1) of the Judiciary Act)”. Professor McMillan’s paper canvasses the five alternative ways of addressing that divergence considered by the ARC and discusses the option adopted as Recommendation 1 by a majority of the ARC. Professor McMillan concludes that this recommendation, considered below, is “an unconventional approach that is not free of doubt”, but nonetheless “the only viable option for retaining the primacy of the ADJR Act”. ^[6]

6  As will become apparent, I am sceptical about the existence of this problem that needs fixing. No doubt there are difficulties and improvements which should be made, but do they include this one and do they include this remedy?

7  Mr Wilkins’ and Ms McGee’s paper, and indeed Mr Wilkins’ minority report, explains the alternative proposal, which is set out in Appendix A to the ARC’s report and entitled “Jurisdictional Limits Model – Directions to Decision-Makers”. They draw on the article by Stephen Gageler SC (as Gageler J then was) “Impact of migration law on the development of Australian administrative law” (2010) 17 AJ Admin L 92 in particular at 104-105:

Keeping administrative decision-makers within the express limits of the lawful authority given to them by statute is as uncontroversial as it is mechanical. Keeping administrative decision-makers within the limits that are implied into the terms by which lawful authority is given to them by statute is more problematic. What are the limits to be implied? By what standards are implications to be drawn? Is truly value-free implication possible? Is it even desirable? Can the result in Plaintiff S157/2002 itself be adequately explained except by reference to values? Under the rubric of jurisdictional error, we have perhaps already reached the point where some limitations on power will be implied by way of presumption in the absence of a tolerably clear manifestation of legislative intention to the contrary. If so, what is it precisely that gives rise to the relevant presumption? These are questions for courts and for commentators.

A question for the Parliament may well be whether what is currently left to implication ought not now be expressed. After all, it is in everyone's interest for administrative decision-makers clearly to be made aware of their jurisdictional limits before those limits have been transgressed. If so, would there not be some utility in spelling out just what those limits happen to be, perhaps in some code or charter of administrative rights and responsibilities or at least in some new part of the Acts Interpretation Act 1901 (Cth)? And if we were to do that, how different would the list look from the now almost forgotten list of grounds in the ADJR Act?

I will come back to Mr Wilkins’ minority report at the end of this commentary, because there is much of interest in it.

8  I should also say that the ARC’s Report is far more substantial than the latest “Judicial Review: proposals for reform” consultation paper published by the Ministry of Justice in England in December 2012 and the Government’s response of April 2013. The response takes the following form, in summary:

Summary of reforms

Time limits

13. We will shorten the time limit for bringing a Judicial Review from three months of the grounds giving rise to the claim to six weeks in planning cases and thirty days in procurement cases.

14. We accept that this will not provide sufficient time to fulfil the requirements of the Pre-Action Protocol. We will also invite the Master of the Rolls to revise the Pre-Action Protocol to disapply it in these cases.

15. We have decided not to seek to clarify when the time limit starts to run in Judicial Review cases where the grounds giving rise to the claim are the result of an ongoing breach, relate to a delay in making a decision or taking action, or relate to a case where there have been multiple points at which decisions have been made.

Applying for permission

16. The Government intends to remove the right to a reconsideration at a hearing of the application for permission to bring Judicial Review (an oral renewal) in any case where the application is certified as totally without merit by the Judge considering the application on the papers.

17. The Government has, however, decided not to take forward the proposal to remove the right to an oral renewal in cases where permission is refused and substantially the same matter has been considered at a prior judicial hearing.

Fees

18. We have decided to introduce a fee for an oral renewal hearing. The fee will be set at the same level as the fee to fix a substantive hearing for a Judicial Review, which is currently £215. The fee for a full hearing will be waived if permission is granted at the oral renewal hearing, so that an applicant with a properly arguable case will not pay two fees.

19. The Government consulted separately on raising fees for Judicial Review cases. The level of the fee for the oral renewal is therefore subject to the outcome of that consultation exercise. The Government intends to publish the response to that consultation shortly.

9  Perhaps in England there is seen to be little in need of reform. Or perhaps the approach reflects “the heavy cloud looming overhead at the start of 2013, with frequently ill-informed, unsubstantiated and sometimes intemperate ministerial attacks on the courts' function of supervising the legality of executive action”.^[7]

10  I turn now to the ARC’s recommendations. To some extent they overlap but I shall talk about them separately as far as possible.

11  Let me first set out the ARC’s proposal and the reasons for it. I note that it does not involve the repeal of s 39B(1) of the Judiciary Act, so much is made clear by paragraph 5.1 of the Report. Neither does the Report envisage the repeal of s 39B(1A) of the Judiciary Act which has a judicial review component.

12  First, at paragraph 4.1, the ARC states that the primary issue facing the federal judicial review system is that in practice there are two systems of judicial review, being the AD(JR) Act and s 39B of the Judiciary Act. Review of migration decisions is analysed as a separate statutory review scheme.

13  Second, at paragraph 4.3, the ARC says that in the period 2007-2011 nearly half the applications for judicial review in the Federal Court were commenced under s 39B and in effect a jurisdiction that was designed to supplement the AD(JR) Act is increasingly overtaking it in importance. The ARC’s view is that judicial review under the AD(JR) Act should be maintained as the principal avenue for federal judicial review.

14  At paragraph 4.5, the ARC says that the AD(JR) Act has played a central role in improving the quality of Australian government decision-making since 1980 and elevating respect for the rule of law in government and an important body of jurisprudence has been developed under that Act. The relative ease with which proceedings can be commenced under the AD(JR) Act meant also that judicial review was more accessible and that proceedings could be commenced without professional legal assistance.

15  Those benefits were at risk, the ARC said, if the current trend continued and constitutional judicial review under s 39B became the preferred or standard avenue for federal judicial review.

16  Accordingly, the ARC’s recommendation sought to maintain what it said was the pre-eminence of the AD(JR) Act in federal judicial review and asked how that pre-eminence could be maintained if the Federal Court also had jurisdiction to hear proceedings commenced under s 39B.

17  It said the reasons for the AD(JR) Act dwindling in importance were twofold: first it was a wise precautionary step to commence judicial review proceedings under s 39B either in addition to AD(JR) Act proceedings or as an alternative. Secondly, the large majority of federal judicial review proceedings concerned decisions made under the Migration Act.

18  The ARC’s preferred option is as follows:

• to add a new head of jurisdiction to the AD(JR) Act based on s 75(v) of the Constitution so that the AD(JR) Act be amended to provide that a person otherwise able to initiate proceedings in the High Court under s 75(v) may apply for an order of review under the AD(JR) Act;
• such an action would be commenced under s 11 of the AD(JR) Act and remedies would be granted under s 16 rather than a constitutional remedy. In essence a court could make an order of review under this extended head of jurisdiction upon being satisfied that a jurisdictional error has occurred;
• there would be no listed grounds on which relief could be granted and ss 5, 6 and 7 of the AD(JR) Act would not "directly" apply to these proceedings (paragraph 4.14);
• consequential changes would be required so as to exclude the right to obtain a statement of reasons for the new head of jurisdiction.

19  This recommendation the ARC said was the most suitable way of achieving the dual objective of retaining the Federal Court's jurisdiction under s 39B while reinforcing the primacy of the Court's jurisdiction under the AD(JR) Act. The ARC expected that reliance on s 39B would be less common because under this proposed extended jurisdiction in the AD(JR) Act a proceeding under s 39B could be commenced but in a manner that was procedurally simpler (paragraph 4.19).

20  The ARC said the main source of uncertainty and potential doctrinal untidiness in its proposal was that the Federal Court would be applying two bodies of substantive law under the AD(JR) Act but this already occurred according to whether a proceeding was commenced under the AD(JR) Act or s 39B.

21  In my opinion the ARC’s Recommendation 1 may be seen as a solution in search of a problem and to the extent that there is a problem the recommendation does not solve it.

22  The problem, if that is a proper way to describe Government policy, is that there is a separate legislative scheme for judicial review of Migration Act decisions, putting the matter broadly. The recommendation does not appear to be addressed to that question.

23  Secondly, I do not accept that the Federal Court is applying two bodies of substantive law according to whether a proceeding was commenced under the AD(JR) Act or s 39B. In my opinion the bodies of substantive law are the numerous statutes in respect of which judicial review actions are brought and which ultimately define the rights and duties of the parties, rather than the mechanism for judicial review. Similarly, although the question of justiciability seems to arise more often in relation to s 39B, because it extends to non-statutory or prerogative powers, it is not the mechanism for review which constitutes the body of substantive law but, at least at the margin, “whether the issue is a suitable one for judicial resolution”. The words in quotation marks are Professor Geoff Lindell’s.^[8]

24  Next, it is not apparent to me why the mere numbers of applications to the courts are significant by themselves, in identifying or defining a problem, absent an assessment of the use of the AD(JR) Act in everyday administration by decision-makers and by those affected by decisions.

25  Next, would the recommendation implicitly still limit such additional jurisdiction to decisions under an enactment? Although this is not made explicit in Chapter 4 I think the answer is “no”. I refer especially to the discussion under the heading Judicial Review of non-statutory decisions on page 79 and following.That discussion commences with the statement as follows:

5.7  The non-statutory decisions of government fall into three main categories: decisions under executive schemes; commercial decisions of government; and other miscellaneous exercises of executive power under s 61 of the Australian Constitution. Different considerations apply to these different types of decision in terms of the possible extension of the ADJR Act.

26  There is then discussion of these categories. (There does not appear to be any discussion of acts or actions (not being decisions) by officers of the Commonwealth which would fall within s 75(v)).

27  As to executive power, the ARC’s conclusion at paragraph 5.22 is that its model would leave the development of this area of judicial review to the courts:

5.22  By giving an applicant standing under the ADJR Act where they can show an entitlement to the constitutional remedies, arguments about the scope of constitutional judicial review can occur under the auspices of the ADJR Act. Given the range of prerogative powers, it is unnecessary specifically to extend judicial review in these areas. However, certain powers of particular national significance, most appropriately dealt with by the High Court, could be excluded from the revised ADJR Act—in particular, decisions relating to war and the defence of the nation.

28  Decisions under executive schemes are considered separately at paragraph 5.23 and following, those executive schemes encompassing schemes that allow the Government to provide discretionary compensation payments such as the Scheme for Compensation for Detriment caused by Defective Administration (CDDA Scheme) and schemes that provide government grants. It is not clear why, conceptually, executive schemes require separate consideration to “Executive power” but perhaps the point is that they are written down and involve a level of detail or sometimes involve a more immediate effect on individuals. Here the conclusions are, in relation to the CDDA Scheme:

5.42  On balance, the Council’s view is that decisions made under the CDDA Scheme would continue to be reviewable in the Federal Court under s 39B of the Judiciary Act but not under the ADJR Act. As such, if the Council’s preferred model were implemented, CDDA decisions would be excluded from the ambit of the amended ADJR Act. The Council bases this recommendation on its expectation that decision making under the CDDA scheme will be undertaken in accordance with the procedures in Finance Circular 2009/09, and that there will be active oversight of CDDA administration by the Commonwealth Ombudsman.

29  I find this confusing. An expectation that decision-making will be undertaken in accordance with written procedures would not appear to be a sound basis for limiting the grounds on which judicial review is available. If there are few claims, and fewer successful claims, that may well show no more than that claimants under the scheme would be hard put to find legal errors. Equally, that there will be active oversight by the Ombudsman of an area of public administration does not suggest to me that, for that reason, there is a basis for limiting judicial review at the threshold. The ARC says at [5.39] that it could not find any examples of review of these decisions under s 39B, but Croker v Minister for the Department of Finance and Deregulation [2013] FCA 429 now provides an example. A few earlier cases were brought under the AD(JR) Act rather than under s 39B because of the statutory footing in the terms of s 33 of the Financial Management and Accountability Act 1997 (Cth). Under s 33, if the Finance Minister considers it appropriate to do so because of special circumstances, he or she may authorise the making of certain payments to a person even though the payment or payments would not otherwise be authorised by law or required to meet a legal liability.

30  The conclusion in relation to schemes that provide government grants is:

5.48  The Council considers that judicial review remedies are unlikely to be particularly useful in relation to grants decisions. Decisions regarding emergency payments tend to be time-critical and administered in such a way that few people miss out, and commercial decisions of government are likely to be subject to other legal remedies. Specifically extending review to these decisions would not significantly increase accountability for government action. The Council’s model would mean that judicial review for these decisions would be available under the ADJR Act if the matter was within the High Court’s jurisdiction under s 75(v) of the Constitution.

5.49  Overall, the Council no longer considers the specific extension of review to non‑statutory schemes funded by an appropriation by Parliament, as previously recommended by the Council and adopted in Queensland, to be the most appropriate approach. Specifically to extend statutory review only to certain non-statutory decisions, where the application of judicial review principles is currently unclear, would place an unnecessary administrative burden on agencies administering those schemes. The discussion above illustrates that executive schemes all have particular considerations—for example urgency, broad discretion as to payments or large numbers of potentially eligible recipients—associated with their administration. The Council’s model would provide for judicial review of executive schemes under the ADJR Act where it is also available under the Constitution. The Council’s model would, however, allow for limited exclusions of ADJR Act review, for example, of decisions under the CDDA scheme.

31  This is an important recommendation, which I consider in a little more detail below, but I find this also confusing at a conceptual level. That an application for judicial review is unlikely to be successful does not strike me as an appropriate basis for distinction. Also, to reason in this way rather suggests that no ground and no remedy would ever be available because of the class, government grants, whereas as the law has been developing it has been increasingly recognised in this field that broad or class distinctions do not yield principled results: Minister for Arts, Heritage & Environment, v Peko-Wallsend Ltd (1987) 15 FCR 274. Lastly, that a decision is not to be quashed does not mean that there is no useful remedy: Ainsworth v Criminal Justice Commission (1992) 175 CLR 564; Plaintiff M61/2010E v Commonwealth; Plaintiff M69 of 2010 v Commonwealth (2010) 243 CLR 319.

32  As to commercial decisions by government the conclusion of the ARC is:

5.57 As discussed above, the scope of judicial review of the commercial decisions of government in the constitutional judicial review jurisdiction is currently unresolved. Probably some commercial decisions of government are subject to limited judicial review. The Council considers that commercial actions where the Government is acting as a private individual—such as contractual actions by government, procurement decisions and tendering decisions—are subject to accountability mechanisms other than judicial review. However, the Council considers that these issues could equally be argued under an amended ADJR Act as under s 39B of the Judiciary Act, and there is therefore no reason to specifically exclude these decisions.

33  The ARC’s overall conclusion in relation to non-statutory decisions is stated as follows:

5.58  There are a range of exercises of non-statutory powers currently not reviewable under the ADJR Act. Many exercises of non-statutory powers are, at least in theory, reviewable under s 39B of the Judiciary Act and s 75(v) of the Constitution.

5.59  When many of these powers are considered individually, there is not necessarily a compelling need for judicial review of those decisions. However, the Council still considers there to be an underlying problem with the dual systems for applying for judicial review. In relation to executive schemes and grants programs, the Council notes that s 39B review would be available—even if the grounds of review are likely to be limited—and therefore considers that many of the concerns expressed by agencies are unlikely to be realised. The Council’s proposed model of drawing the constitutional review jurisdiction into the ADJR Act would therefore bring these decisions within the ambit of the ADJR Act, to the extent that constitutional judicial review already allows for their review.

5.60  However, some non-statutory decisions could be excluded from the operation of the expanded ADJR Act. The Council considers that limited exclusions would still be appropriate to address the concerns of government agencies about the expansion of review in certain areas. Exemptions which the Council considers appropriate—on the basis that only very limited review would be available under s 39B—are:

• CDDA decisions; and
• actions relating to war and the defence of the nation.

5.61  Such exemptions should be limited to decisions where there is a justification for limiting the availability of judicial review to its constitutional minimum.

34  That is, the ARC contemplates the exemption of classes of non-statutory executive action from the new expanded AD(JR) Act, which would seem to add further complexity – leaving parts of s 39B out of the new expanded AD(JR) Act. At present I do not follow the point of having two s 39B streams. Put differently, if the effect of putting s 39B cases under the AD(JR) Act is only to have standing and remedies written down and modernised, why is it necessary to bifurcate the proposed new jurisdiction in this way?

35  Does this proposal now bring in decisions of a legislative character made by officers of the Commonwealth?It appears so, as I discuss more fully below.

36  As to the position of the present Schedules to the AD(JR) Act, it would seem that no general change is recommended or contemplated, but it appears to be implicit that Schedule 1 and Schedule 2 would apply to the proposed extended reach of the AD(JR) Act to cover s 39B(1). The relevance of this to the discussion in paragraph 4.49 of the Report, set out above, may be seen by reference to the recent amendments to Schedule 1 of the AD(JR) Act by the addition of paragraphs (he) and (hf) made by clause 1 of Schedule 1 to the Financial Framework Legislation Amendment Act (No 3) 2012 (Cth) in light of the decision of the High Court in Williams v Commonwealth (2012) 288 ALR 410 (the School Chaplains case). Section 39B, as unamended by the ARC’s proposals, may be available to challenge something that is done under Division 3B of Part 4 or s 44 of the Financial Management and Accountability Act 1997 (Cth) at least where what is done affects the rights and interests of a person. Section 39B under the proposal to bring it in to the AD(JR) Act would not be available in light of paragraphs (he) and (hf) to Schedule 1 to the AD(JR) Act. Thus s 39B would be bifurcated.

37  It appears that Australian Broadcasting Tribunal v Bond (1990) 170 CLR 321 would not be reversed so far as concerns the final or operative requirement for reviewable decisions under the AD(JR) Act. The ARC said:

5.81  The Law Council of Australia supported a broader amendment to the ADJR Act to overcome the requirement in Bond that a ‘decision to which this Act applies’ must be ‘final or ultimate or operative’. The Law Council argued that this requirement is not supported by the text of the ADJR Act and has resulted in an ‘undesirable proliferation of litigation on a technical issue of justiciability’. They noted that the same requirement does not apply in relation to review under the Judiciary Act. They argued that the Court can be given an additional basis for discretion to decline to exercise jurisdiction under s 10(2)(b) of the ADJR Act in order to combat concerns about ‘abuse’ of the Act. The Council does not support this approach on the basis that the jurisdictional test in the ADJR Act is now relatively settled, and the Council’s preferred approach to expanding the jurisdiction of the Act would allow for review of preliminary decisions to the extent this is already available via constitutional judicial review.

By this means, as I understand it, the status quo in this respect would be retained. Section 13 reasons would not be available and the s 5 grounds would not be “directly” applicable: in substance a party would proceed as now under s 39B but in form under the AD(JR) Act. It appears that the rules about standing and remedies would be modernised.

38  In addition, at the heart of the recommendation seems to be the uncertain content of the conclusory statement “jurisdictional error”. The recommendation does not cure this apart from suggesting there may be a closer alignment of statutory judicial review and constitutional judicial review if the recommendation were adopted as it is possible the courts would pay explicit regard to the criteria in s 5 in elaborating the concept of jurisdictional error (paragraph 4.21).

39  At present, combining an application under s 39B and under the AD(JR) Act does not cause a particular problem. There is even a rule about it, being rule 31.01(3) of the Federal Court Rules 2011. There may be some intellectual laziness involved if the applicant or the applicant's lawyers have not given sufficient thought to whether or not they need a combined application. Sometimes it is done to avoid an objection to competency as to whether or not what has occurred constitutes a decision within Australian Broadcasting Tribunal v Bond or whether the “under an enactment” jurisdictional requirement has been met. Sometimes there may be an element of late filing where the applicant sees an advantage in not having to explain the delay by reference to the AD(JR) Act and sometimes there appears to be an advantage in avoiding the direct operation of the express discretionary powers to refuse relief set out in s 10.

40  If all of this is right, however, the recommendation of the ARC would appear to be cosmetic in the sense that it does not deal with these issues which would, apparently, continue to apply but under a different name, the AD(JR) Act.

41  Later in these comments I put forward a modest proposal to include as a ground of judicial review under the AD(JR) Act an express ground of “jurisdictional error within the meaning of s 75(v) of the Constitution” so as to make express on the face of the legislation what one may well derive from Minister for Immigration and Multicultural Affairs v Yusuf (2001) 206 CLR 323 at [83].

42  I would also suggest an express statement at the beginning of s 5 or a note at the end of it to the effect that the grounds of review are not mutually exclusive.

43  I appreciate that my proposal would not have the same effect as Recommendation 1 in the ARC’s Report but I question whether that recommendation would hit its mark.

44  May one say that the problem of bifurcation or “fragmentation of the federal judicial review landscape” (page 11) or “divergences between the two avenues of judicial review available in the Federal Court” (paragraph 4.2), whether merely procedural in terms of forms, or more substantial in terms of the ARC’s reference to “two systems of judicial review” (paragraph 4.1) does not exist? May one say the recommendation is a disproportionate response to the identified problem, if it exists? Or is this to look at it from the too narrow perspective, referred to by the ARC at paragraph 4.10, that “the framework and principles for federal judicial review must satisfy benefits beyond the convenience of experienced practitioners”?

45  For completeness I also note that the ARC does not recommend expanding judicial review beyond the current ambits of the AD(JR) Act and constitutional review. The Council considers a model of review based on a “public power” or “public function” test to be indeterminate and likely to create uncertainty about the coverage of judicial review standards: see page 13 of the Report. The reason is given in paragraph 4.12 as follows:

The chief difficulty, as explained in Chapter 5 of this report, is that it would replace ADJR Act jurisdictional tests that have been interpreted and applied for over 30 years and that have a relatively settled operation, with new tests of uncertain meaning and operation. This change could usher in a new phase of adventurous re-thinking of the scope of judicial review, the grounds of review and the objectives of administrative law. The Council does not believe that the resulting litigation would necessarily enhance adherence to the core values of administrative law, or that Parliament would support such a change.

46  In other words, the ARC does not recommend that the principle for which R v Panel on Take-overs and Mergers; Ex parte Datafin [1987] QB 815, 824–825 stands should be included in the proposed regime. In my view, this position is the correct one, at least for federal judicial review. It is not possible to ignore the express and limiting words of s 75(v) “an officer of the Commonwealth”.^[9] This still leaves the very large question of review of administrative action outsourced by government in cases where there is no involvement of an officer of the Commonwealth.

47  First, to be recalled are the broad and beneficial terms of s 3(3) of the AD(JR) Act, as follows:

3(3) Where provision is made by an enactment for the making of a report or recommendation before a decision is made in the exercise of a power under that enactment or under another law, the making of such a report or recommendation shall itself be deemed, for the purposes of this Act, to be the making of a decision.

48  In Edelsten v Health Insurance Commission (1990) 27 FCR 56 the Full Court said (at 70):

In our opinion, s 3(3) applies where there is a provision in an enactment that a particular report or recommendation be made as a condition precedent to the making of a decision under that enactment or under another law … We agree with the view expressed by Ellicott J in Ross v Costigan (1982) 59 FLR 184 at 198 that s 3(3) `contemplates a case where there is provision in an enactment for a specific report or recommendation as a condition precedent to the making of a decision under that enactment or some other'.

49  Next to be recalled is Ainsworth v Criminal Justice Commission (1992) 175 CLR 564 granting relief by way of declaration where a report had adversely affected Ainsworth’s reputation and in circumstances where the report had no legal effect.

50  Third, as decided in Hot Holdings Pty Ltd v Creasy (1996) 185 CLR 149 at 159, relief is available where a report “has a discernable effect or apparent legal effect upon rights”, but query whether that applies where the report is not a mandatory relevant consideration.

51  Fourth, in Plaintiff M61/2010E v Commonwealth; Plaintiff M69 of 2010 v Commonwealth (2010) 243 CLR 319 at 352 (the Offshore Processing case), a unanimous court of seven justices emphasised at [99]-[105] the flexibility of remedies in public law, particularly the declaration of right. Their Honours at [100] left open “whether, as was submitted on behalf of the Commonwealth and the Minister, certiorari will not go to quash a decision or recommendation prior to the final exercise of a discretion that directly affects legal rights unless that decision or recommendation must be taken into account by the ultimate decision-maker”. Nor was it “necessary to examine whether, or how, the proposition advanced on behalf of the Commonwealth and the Minister, expressed as it is in absolute terms, might permit or require modification to accommodate cases such as the present, where the right that is affected by conducting the impugned process of decision making is a right to liberty." The claims to certiorari and mandamus were refused but the Court said, at [101], a declaration should be made in each case that the processes undertaken to arrive at the reviewer's recommendation were flawed in the respects that have been identified, the declaration being moulded in terms similar to the declaration made by the Court in Ainsworth. The Court said as follows:

[105]  In each matter there should be a declaration that, in recommending to the Minister that the plaintiff was not a person to whom Australia has protection obligations, the third-named defendant made an error of law, in that he (or in the matter of Plaintiff M69, she) did not treat the provisions of the Migration Act 1958 (Cth) and the decisions of Australian courts as binding, and further, failed to observe the requirements of procedural fairness.

52  Note also in SZQDZ v Minister for Immigration and Citizenship  (2012) 200 FCR 207, dealing with a matter the Minister was not bound to take into account, the Full Court said at [44]:

[44]  The Minister can ignore entirely a reviewer’s assessment and recommendation. However, as in Plaintiff M61 at 358–360 [99]–[104], declaratory relief and, perhaps, an injunction would be an appropriate remedy to ensure that the Minister understands that a [sic] assessment or recommendation is affected by demonstrable error if he or she were otherwise minded to act upon them. In this regard, to say that the assessment and recommendation of the review was not an administrative decision is not to say that the review process was not required to be fairly and lawfully conducted.

53  Against this background it might be thought that the recommendation of the ARC takes a position behind what appears to be the current state of the law. However what is to be remembered is that the present provision, which I have set out above, deems those reports with which it deals to be decisions in their own right. Section 3(3) does not deal with the question of when a report, which has become part of the ultimate or substantive decision, may be reviewable as part of a larger attack.

54  There are intermediate statutory reports reviewable in their own right by analogy with the intermediate statutory decisions reviewable in their own right as referred to in Australian Broadcasting Tribunal v Bond (1990) 170 CLR 321 at 337.

55  Also Australian Broadcasting Tribunal v Bond (1990) 170 CLR 321 at 338 referred to antecedent conclusions or findings which contributed to the ultimate decision and may therefore not be beyond reach on judicial review of that ultimate decision.

56  For my own part, although I do not think the ARC discusses it, it may be debatable whether the condition precedent approach is too narrow. The point may be better expressed if the report was reviewable in its own right where it was a mandatory relevant consideration.

57  I also venture to doubt the practicality of providing a schedule that can be amended by regulation to list other reports and recommendations that are decisions to which the AD(JR) Act applies.

58  I also doubt that factors which indicate that a report or recommendation should not be subject to statutory judicial review are where the report or recommendation “is of a completely factual nature; and is an assessment of a professional nature and judicial review standards, such as procedural fairness, do not apply to the decision-making process”: see paragraph 5.87 of the Report. These factors seem to be both imprecise and contestable.

59  May it not be preferable to align the position with the common law so far as concerns s 3(3)? This may leave out only reports and recommendations which were neither made under an enactment nor made by an officer of the Commonwealth, at least as long as they remained mere reports and recommendations, without action being taken on them.

60  The ARC said:

5.121 This conclusion is reflected in Report No 47’s ‘Framework of indicative principles’, which says that limits on judicial review are justified in decisions in relation to criminal, civil penalty or extradition proceedings. The Council’s view is that the ADJR Act should also not apply to decisions to commence civil penalty proceedings.

My only query is whether the recommendation is sufficiently broad. Perhaps it takes as its starting point the existing law.

61  A broader approach, again perhaps converting into a principle what would be a matter going to the court’s discretion in any event, would be to exclude decisions to commence any proceedings. The policy discussed by the ARC would seem to lead in that direction. The point is that the courts in which such proceedings are brought, whatever the division between those types of proceedings, as to which see Chief Executive Officer of Customs v Labrador Liquor Wholesale Pty Ltd (2003) 216 CLR 161, are in the best position to control them. See Strictly Stainless Pty Ltd v Deputy Commissioner of Taxation (Unreported, Federal Court of Australia, Davies J, 5 November 1993) and the cases which have followed that decision such as Golden City Car & Truck Centre Pty Ltd v Deputy Commissioner of Taxation (1999) 56 ALD 177; Ruddy v Deputy Commissioner of Taxation (1998) 82 FCR 337 and Rawson Finances Pty Ltd v Deputy Commissioner of Taxation (2010) 189 FCR 189.

62  Justice Davies said:

…one can see the point of public policy involved. It seems to me to be quite inappropriate for this Court to consider whether or not it was reasonable for the Deputy Commissioner to institute wind-up proceedings in the Supreme Court. This Court should stand right away from the question and allow the Supreme Court to deal with the matter in the ordinary way.

I do not think it would be useful for me to make any comment on the facts of this present case. Indeed, I think it would be inappropriate for me to do so. The facts are for the Supreme Court to determine. If the Supreme Court wishes to consider any aspect of the matters which counsel has raised in this present case, that will be a matter for the Supreme Court.

63  Another important issue discussed in this section of the ARC’s Report, although not the subject of a numbered recommendation, is the review of subordinate legislation. The ARC accepted, at paragraph 5.88, that its proposed model of judicial review:

… allows for direct review of subordinate legislation under the ADJR Act to the same extent as it is available under constitutional judicial review. Therefore, the Council does not consider that a specific extension of the ADJR Act to allow for judicial review of subordinate legislation is necessary.

64  I support the conclusion although it may well be that it is s 39B(1A)(b) and, more particularly, s 39B(1A)(c) of the Judiciary Act that achieves this result, rather than s 39B(1).

65  In my opinion the removal of the blanket exclusion of decisions by the Governor-General under an enactment from the purview of the AD(JR) Act must be correct: the identity of the decision-maker can no longer be seen as determinative. There may be room for debate, however, in relation to the proposed prescribed exclusions or exemptions in Schedule 1. I do not see the force of the reasoning in paragraph 5.128 that “the urgent nature of the decision and the fact that the power is unlikely to be used frequently … would mean that an application to the High Court in the first instance would be the most appropriate form of review.”

66  I do see force in the exemption of certain decisions relating to defence and to appointments, and the termination of appointments, to certain statutory positions.

67  Other examples of statutory decisions which could be considered for inclusion in Schedule 1 and therefore exclusion from the general rule would be, in my opinion, dissolving Territory assemblies and issuing Royal Commissions.

68  In addition, although the ARC’s view was against it, I would be inclined to include in the AD(JR) Act a statement that justiciability is not to be assumed, particularly in relation to decisions of the Governor-General.

69  I also think it would be worthwhile saying in the AD(JR) Act that the proper respondent party in an application for judicial review of a statutory decision of the Governor-General is the responsible Minister. This would tend to encourage attention to the nature of the power rather than who, formally, was exercising the power.

70  I support this recommendation concerning the limited scope of the AD(JR) Act in light of the provision made for “merits review” either in the AAT or in the Federal Court. On one view, the effect of the Schedule is to make a rule about where tax appeals should be heard on the merits rather than leaving it to the Court's discretion on the basis of adequate alternative remedy.

71  Of course there are still issues to be worked out as to the content of the jurisdiction under s 39B to set aside assessments for “conscious maladministration” and the like. It also appears that that issue may be raised only in proceedings under s 39B, not in proceedings under Part IVC, even in the Federal Court, and not in the AAT or in the Federal Circuit Court. See Commissioner of Taxation v Futuris Corp Ltd (2008) 237 CLR 146 and the two Full Court of the Federal Court decisions in Denlay v Commissioner of Taxation (2011) 193 FCR 412 and Gashi v Commissioner of Taxation (2013) 209 FCR 301; 296 ALR 497.

72  In my view, from a technical perspective, s 44 of the Administrative Appeals Tribunal Act 1975 is not working as well as it should. There are still arguments about the framing of the question or questions of law and the apparently stricter view of the Full Court in Comcare v Etheridge (2006) 149 FCR 522 was explained in Collins v Administrative Appeals Tribunal (2007) 163 FCR 35 per Allsop J, with whom Lindgren and Emmett JJ agreed.

73  It is also the case that it is often notoriously difficult to divide questions of law from questions which are not such questions. As long ago as 1927, Professor Dickinson in Administrative Justice and the Supremacy of the Law wrote, at 55:

Matters of law grow downward into roots of fact and matters of fact reach upward without a break, into matters of law. The knife of policy alone effects an artificial cleavage where the court chooses to draw the line.

This was quoted by Professor Whitmore in an important article, as long ago as 1967.^[10] In the same article, Professor Whitmore also quoted Professor Green, writing in 1930, discussing the division of functions between judge and jury:

No two terms of legal science have rendered better service than ‘law’ and ‘fact’. They are basic assumptions; irreducible minimums and the most comprehensive maximums at the same instant. They readily accommodate themselves to any meaning we desire to give them. In them or their kind a science of law finds its strength and durability. They are the creations of centuries. What judge has not found refuge in them? The man who could succeed in defining them would be a public enemy. They may torture the souls of language mechanicians who insist that all words and phrases must have a fixed content, but they, in their flexibility, are essential to the science which has to do with the control of men through the power to pass judgment on their conduct.^[11]

Often the answer will depend on precisely how the case is put. It is therefore difficult to agree with the proposition, noted in the ARC’s Report at paragraphs 6.24 that s 44 provides the “user benefits” of a clear appeals mechanism, and procedural and cost benefits when compared to judicial review applications.

74  It may once have been intended that there should be a narrower field of review or appeal for decisions of the Administrative Appeals Tribunal than for other Executive decision-make However as a matter of history I note that the Commonwealth Administrative Review Committee Report 1971 (Parliamentary Paper No. 144) (the Kerr Committee) recommended at [247] that the Court should be invested with supervisory jurisdiction over the proposed Administrative Review Tribunal and that that supervision would be on the basis of the clarified grounds of judicial review later proposed by that Committee, that is, what we now know as the AD(JR) Act. Be that as it may, because it now seems to be generally accepted that the AD(JR) Act is also available to challenge Tribunal decisions and because s 39B is also available it is difficult to agree with the ARC at 6.25:

The Council considers that the risk of duplication is mitigated by the demarcated functions of s 44 appeals and judicial review under the ADJR Act and s 39B of the Judiciary Act.

And at 6.32-6.33:

The Council considers that framing applications on questions of law has advantages—it draws attention to the distinction between review of the merits of a case and legal error, and discourages applicants from seeking to challenge findings of fact.
Overall, the Council concludes that the procedural advantages of retaining s 44 outweigh the arguments for its removal or revision.

75  At 6.30, the ARC cites with approval the statement of Davies J in Tuite v Administrative Appeals Tribunal (1993) 40 FCR 483 at 484:

The words ‘question of law’ in s 44 encompass matters concerning not only the interpretation of a Federal enactment or the enunciation of the principle of the common law or equity, but also the breach of any duty which the Tribunal was bound by law to perform and the failure of which to perform may lead to the setting aside of the decision. The words ‘question of law’ encompass grounds enunciated in s 5 of the ADJR Act.

76  If this is so, why should there not be a provision included in s 44 of the AAT Act to say simply “‘question of law’ encompasses grounds enunciated in s 5 of the AD(JR) Act.” or, more directly, replace s 44 with a provision stating that review under the AD(JR) Act is available? In my view that would put an end to the need to bring AD(JR) Act applications at the same time as s 44 appeals and would limit, if not remove, the need for agonising about how to frame a procedural fairness point as a question of law.

77  I note that in paragraph 2.7 of the ARC’s Report 1997 No. 41 Appeals from the Administrative Appeals Tribunal to the Federal Court there is a reference back to theKerr Committee correctly perceiving at [232] the need for the general administrative appeals tribunal to be subject to correction on all matters of law and not just jurisdictional matters. I have just set out what I regard as the more relevant recommendation of the Kerr Committee at [247]. This supports the conclusion, in light of developments over the last 45 years, thats 44 should pick up the grounds of review available under the AD(JR) Act including what I have suggested as “jurisdictional error within the meaning of s 75(v) of the Constitution”. There is in my view no purpose whatever in having different formulations which merely add unnecessary complexity.

78  This is a recommendation with which it is easy to agree given that one of its purposes is to keep review rights simple.

79  The ARC said at 6.36:

Where the Government proposes to establish a new separate statutory scheme to operate alongside or in place of review under the ADJR Act, the Council considers that there should be compelling reasons for the establishment of such a scheme—such as the need for a streamlined process involving both merits and judicial review, such as in the taxation system. New statutory schemes should not attempt to restrict the grounds of review or remedies available under the ADJR Act. The Council notes that procedural measures can be implemented in relation to particular kinds of matters without establishing an entirely separate scheme of review. The Council considers that further fragmentation of the judicial review system is undesirable, and that new separate statutory schemes will rarely be justified.

80  It also would tend against the specialist courts to which Heydon J powerfully and wittily drew attention in Kirk v Industrial Relations Commission (NSW) (2010) 239 CLR 531 at [122] (and to which the majority alluded at [64]):

Our legal system has often had to balance the advantages of creating specialisation over the disadvantages of doing so. It is commonly thought better, for example, that allegations of crimes be tried by judges expert in criminal law and procedure. The same is true, mutatis mutandis, of company work, bankruptcy, personal injury claims, planning law and many other categories of litigation. Sometimes the legislature elects to create separate courts for the particular litigation. Sometimes it creates separate divisions within a court.
…

However that may be, the appellants referred in submissions to the danger of conferring jurisdiction to hear criminal proceedings on courts the practitioners in which are unfamiliar with all the relevant rules. There is a related danger in that course in that the courts on which the jurisdiction has been conferred, while in some sense specialist, are not familiar with all the relevant rules. Thus a major difficulty in setting up a particular court, like the Industrial Court, to deal with specific categories of work, one of which is a criminal jurisdiction in relation to a very important matter like industrial safety, is that the separate court tends to lose touch with the traditions, standards and mores of the wider profession and judiciary. It thus forgets fundamental matters like the incapacity of the prosecution to call the accused as a witness even if the accused consents. Another difficulty in setting up specialist courts is that they tend to become over-enthusiastic about vindicating the purposes for which they were set up. Medical students usually detect in themselves at a particular time symptoms of the diseases they happen to be studying at that time. Academic lawyers interested in a particular doctrine can too often see it as almost universally operative. So too courts set up for the purpose of dealing with a particular mischief can tend to exalt that purpose above all other considerations, and pursue it in too absolute a way. They tend to feel that they are not fulfilling their duty unless all, or almost all, complaints that that mischief has arisen are accepted. Courts which are “preoccupied with special problems”, like tribunals or administrative bodies of that kind, are “likely to develop distorted positions”: Walker, The Rule of Law (1988), p 35.

Thus Jaffe said, discussing the factual position illustrated by R v Bradford [1908] 1 KB 365 “[R]oad-maintenance authorities sorely pressed to find gravel within the parish will not place a high value on the amenities of the gentry’s parks”: “Judicial Review: Constitutional and Jurisdictional Fact”, Harvard Law Review, vol 70 (1957) 953, at p 963.

81  The first issue the ARC addressed in its discussion was whether there should be a list of grounds at all. It considered that a list of grounds should be retained on the basis that they play an educative role and are more transparent and accessible than relying solely on principles set out in judicial decisions.

82  The next issue was whether a less comprehensive and prescriptive list of grounds might be more flexible and adaptable.

83  The third issue was whether there should be some general principles included to give direction for the particular review grounds.

84  The fourth issue was whether the two “catch all” grounds of review were performing a useful role or whether they should be repealed or amended.

85  The ARC considered three potential new grounds of review: “proportionality”; “serious administrative injustice” and “serious illogicality or irrationality”.

86  The ARC considered that there was no need to add additional grounds to the current list, in part because the existing “catch all” ground was likely to incorporate the new ground of “serious illogicality” as “otherwise contrary to law”. The ARC considered that both “proportionality” and “serious administrative injustice” were grounds that raised issues about the line between merits and judicial review, and that other avenues were open to people who had suffered from serious maladministration to make complaints or seek compensation.

87  I raise for consideration whether or not “jurisdictional error within the meaning of s 75(v) of the Constitution” should be added as a ground so as to make it clear that there is no gap between the statutory grounds and the grounds available under the constitutional writs and so as to make express on the face of the legislation what one may well derive from Minister for Immigration and Multicultural Affairs v Yusuf (2001) 206 CLR 323 at [83]. Given the importance now attributed to “jurisdictional error”, the “catch all” ground of “otherwise contrary to law” seems more appropriate to a somewhat pale afterthought. Such a ground of “jurisdictional error” would perhaps pick up, for example, whatever is new about rationality and reasonableness in Minister for Immigration and Citizenship v Xiujuan Li [2013] HCA 18; (2013) 297 ALR 225.

88  I accept that this proposal is a modest one in the absence of any radical rethink of Schedules 1 and 2 of the AD(JR) Act.

89  The ARC then considered whether some of the particular paragraphs in s 5 should be amended.

90  Although not suggested by the ARC, I see merit in a note at the end of s 5 drawing attention to the Court’s discretion to grant a remedy and saying that it does not necessarily follow that the Court will grant relief even where a ground is established, for example, in the case of immaterial error.

91  In my opinion relevant and irrelevant considerations still waste a good deal of time. As to an irrelevant consideration, perhaps the minor issue, I would support a rewording: “taking into account a consideration prohibited to be taken into account by the enactment in pursuance of which the decision was purported to be made” to adapt a suggestion of Professor Aronson.

92  Similarly, in relation to relevant considerations, I would support a rewording: “failing to take into account a consideration required to be taken into account by the enactment in pursuance of which the decision was purported to be made.”

93  As to an exercise of a discretionary power without regard to the merits of the particular case I would support a rewording: “applying a rule or policy that unlawfully purports to narrow or enlarge the breadth or content of a discretionary power (conferred by the statute)”.

94  It may be that the reference to a rule or policy is relatively minor and the heart of the error is misapplying the power by taking too narrow or too broad a view of it.

95  The ARC’s positive recommendations in relation to ss 5(3) and 6(3), set out above, seem to me to be valid and important, although the word “critical” to the final decision, derived from the decision of the Full Court in SFGB v Minister for Immigration and Multicultural and Indigenous Affairs (2003) 77 ALD 402; [2003] FCAFC 231 and from Minister for Immigration and Multicultural Affairs v Rajamanikkan (2002) 210 CLR 222 may be, once in a statute, unduly restrictive. The adjective “material” might be more apposite.

96  This recommendation is perhaps more a “note to self”. It seems to me that statutory codes of procedure may be useful both to decision-makers and to those affected if “code” is not used in the sense of an exhaustive statement to the exclusion of rules of procedural fairness.

97  In my view there is merit in the idea that, where appropriate and ideally statute by statute and so as to state expressly within what limits particular powers are to be exercised, there should be an express legislative statement of the positive limits on decision-making; a clear statement as to the consequences of any non-compliance with those limits or any of them for the ensuing decision so as to assist with a jurisdictional error/Project Blue Sky Inc v Australian Broadcasting Authority (1998) 194 CLR 355 analysis; and, lastly, an express statement that the statutory steps so articulated are not intended to exclude the rules of procedural fairness. I develop this further in the context of considering Mr Wilkins’ minority report, at the end of this paper.

98  I agree with this recommendation.

99  Section 27 of the AAT Act relevantly provides:

27(2) An organization or association of persons, whether incorporated or not, shall be taken to have interests that are affected by a decision if the decision relates to a matter included in the objects or purposes of the organization or association.

27(3) Subsection (2) does not apply in relation to a decision given before the organization or association was formed or before the objects or purposes of the organization or association included the matter concerned.

100  No doubt in a curial setting, particularly a setting which involves a costs regime, unlike the Tribunal, there will be questions as to security for costs and which entity or person, of any financial substance, would be liable for a costs order. No doubt also costs capping applications may be made under the Federal Court Rules 2011 or the Federal Circuit Court Rules 2013.

101  The provision seems to have worked without difficulty. Although in Arnold (on behalf of Australians for Animals) v Queensland [1987] FCA 148; 73 ALR 607 at 612 the Full Court said there were “some difficulties of interpretation in s 27”, it now seems to be accepted that s 27(2) is a deeming provision and deems the interests of unincorporated organisations to be affected by a decision if the relevant decision relates to a matter included in their objects or purposes. This deeming provision is for the purposes of s 27(1).

102  There may be some of you who think the proposed reform does not go far enough.

103  It is easy to be in favour of the recording of reasons. However I also note that this recommendation is at the level of endeavour rather than statutory obligation.

104  I agree with the reasons of the ARC in support of the contemporaneous recording of reasons by a decision maker: paragraph 9.40. Reasons recorded at the time of making the decision are more likely to reflect a decision-maker's actual reasons. Contemporaneous reasons can be provided at the time notice of the decision is given. Importantly, the process of writing reasons materially assists decision-makers during the process – facilitating the detailed consideration of all necessary issues.

105  Having said that, I doubt whether mass or semi-automated decision-making would ever be the subject of reasons in the same sense. I wonder whether a “one size fits all” approach is realistic. Section 25D of the Acts Interpretation Act 1901 (Cth) is unlikely to be a problem because it applies only where an Act requires a tribunal, body or person making a decision to give written reasons for the decision, whether the expression “reasons”, “grounds” or any other expression is used. Only then must the instrument giving the reasons also set out the findings on material questions of fact and refer to the evidence or other material on which those findings were based.

106  One ground of resistance to the contemporaneous recording of reasons is, perhaps, the perception that reasons must always be lengthy and detailed. If they could be flexibly introduced so that in many cases the decision-maker could merely note the essential facts as found and their essential reasoning there may be less resistance. Also, once it is appreciated that there is no requirement for “model reasons” and that Minister for Immigration and Multicultural Affairs v Yusuf (2001) 206 CLR 323 establishes that it is the decision-maker’s real reasons that are required to be written down and not some ideal set of reasons or the reasons the decision-maker should have had then the prospect of complying with the ARC’s recommendation would be less daunting.

107  I also note the discussion, albeit in the context of recommendation 12, about statements of reasons that are not prepared by the decision-maker. There seems to be more than one problem here.

108  One, which seems to be relatively straightforward, is where the original decision-maker is no longer available when a statement of reasons is being prepared, on the assumption that no contemporaneous statement of reasons was prepared. I see much sense in the suggestion made in a 2010 article by Stephen Lloyd SC and Donald Mitchell, to which the ARC refers, “Statements of the decision maker’s actual reasons” (2010) 59 Admin Review 56. That suggestion, I understand, was that the person currently responsible for making that decision should, where possible, remake the decision and give their own reasons.

109  A separate issue is in relation to decisions by Ministers. Again, Stephen Lloyd and Donald Mitchell suggest that where appropriate agencies could provide a draft statement of reasons at the same time as the submission for decision which the Minister can sign or alter as necessary. Otherwise one is left with the submission together, sometimes, with handwritten underlining or marginal notes. If the statement of reasons is not contemporaneous with the decision then, it seems to me, the relevant officers should first speak to the Minister to find out what his or her reasons were, so as to guard against the wholesale adoption of a draft statement of reasons which, in those circumstances, may not reflect the actual decision-making. This was the problem in Phosphate Resources Ltd v Minister for Environment, Heritage and the Arts (2008) 251 ALR 80 where Buchanan J referred at [162] to the reasons for decision being carefully crafted in the Department after they were requested by PRL but which were not the reasons present in the mind of the Minister, or adopted by him, when he signed the decision.

110  A further and final issue noted by Stephen Lloyd and Donald Mitchell and referred to by the ARC is legal practitioners settling draft reasons to remove legal errors. In my opinion where a statement of reasons is to be considered by a lawyer this should not be done in the absence of the decision-maker except for the purposes of raising questions for the decision-maker to address. There is, in my view, an analogy with the “settling” of opinions of experts by lawyers: every care should be taken to ensure that the opinion or statement of reasons accurately reflects the actual opinions and reasons. Intrusions by lawyers should be limited to formal aspects, including any obscurities of expression.

111  The additional consideration where a statement of reasons is involved is that the reasons should be the contemporaneous reasons of the decision-maker at the time of the decision.

112  I have considered the substance of this recommendation 12 in my discussion of recommendation 11.

113  The ARC discusses from paragraphs 9.59 to 9.74 its views on the consequences of a failure to give reasons.

114  I note the terms of s 13(7) of the AD(JR) Act, which provides that where, upon application, the relevant Court considers the statement of reasons that has been provided to be inadequate, the Court may order the person who furnished the statement to provide a better statement within a specified time.

115  In my experience there is not often litigation about statements of reasons because in many cases an applicant would prefer to have and work with the record, with its faults, if any, rather than with a settled formal statement.

116  Is there a need for a specific power in the Act in the case of no reasons rather than inadequate reasons, the subject matter of s 13(7)? Given the clear existence of a duty to provide such a statement there may be thought to be no reason to add to the armoury of the courts, the present armoury including a writ of mandamus. Nevertheless, if one of the points is that the AD(JR) Act should be self-contained and accessible and because a claim of inadequate reasons is already dealt with then, in my view, such a remedy should be expressly provided in the Act. I do not understand the ARC’s reference at paragraph 9.72 that such an additional provision implies that agencies might refuse a request for reasons arbitrarily. I would have thought that if there was any implication it would be to the opposite effect.

117  I also do not understand the ARC’s reference at paragraph 9.62 that providing for judicial review of a failure to provide reasons contradicts a central purpose of the right to reasons: providing an opportunity to determine whether there is any basis for commencing an action. The unreported decision referred to, Clanwilliam Pty Ltd v Bartlett (23 May 1984, Fitzgerald J) noted at (1984) 6 ALN N61, was dealing with an argument on the question of costs after a statement of reasons had been provided but had been provided late, apparently put by the respondent and rejected, that an order for reasons should not or would not have been made because a failure to give reasons within time was itself a reviewable decision under the Act.

118  Turning to the precise terms of Recommendation 13, I would not disagree with it. My only question is whether such a provision is necessary because the court would already take into account on the question of costs whether there had been a material failure to provide a statement or an adequate statement of reasons. The question of materiality would go to whether or not proceedings would have been commenced at all or persisted in at all or maintained on a wider basis than necessary if reasons or adequate reasons had been given.

119  I also found the discussion at 9.73 and 9.74 confusing. I would not have thought the reference to Oshlack v Richmond River Council (1998) 193 CLR 72 advanced the argument unless there is a freestanding recommendation that the public interest character of litigation should be a relevant factor in a costs order, as to which see recommendation 15. Also I would not have thought that references to specific costs provisions for proceedings in the Administrative Appeals Tribunal, for example under s 66 of the Freedom of Information Act or s 67 of the Safety, Rehabilitation and Compensation Act, had any bearing on the present question. Ultimately the question is whether having a specific provision that a failure to give adequate reasons or, I would add, a failure to give reasons at all, is or may be relevant to costs is necessary or appropriate.

120  I note that there is a difference between this recommendation in its final form and paragraph 9.5 where the ARC said it recommended a statutory presumption that costs should be awarded in any cases where a person has to take the matter to court without the agency providing reasons on request. I doubt whether such a presumption is necessary but it may operate as a useful signal to a prospective respondent.

121  The ARC does not recommend the repeal of Schedule 2 of the AD(JR) Act but goes through each of those exemptions in Appendix C to its Report. Time does not permit a consideration in this paper of those detailed recommendations. Section 13A of the AD(JR) Act operates so that certain information is not required to be included in a s 13 statement of reasons.

122  The recommendation for amending s 13A(1) is certainly neat. Nevertheless I do not follow the reference in paragraph 9.91 where the ARC says, with regard to the exemption of particular classes of information, that it considers that amending s 13A would not be without difficulty as amendments could potentially broaden exemptions from the right to request reasons.

123  My comments are as follows. I have no difficulty with the principle as long as it is not related to s 25 of the Freedom of Information Act as described in paragraph 9.93 and apparently approved in paragraph 9.94 of the ARC’s Report. Section 25 seems to me to be directed at a different end, that is, not having to give information about the existence or non-existence of certain documents. For that reason, s 25 does not pick up each of the exemptions in Part IV of the Freedom of Information Act.

124  Next, to adopt recommendation 14(a) would have the result that a reader would need to go to the Freedom of Information Act to find out what classes of documents are exempt documents under that Act. In my view it would be preferable, although adding slightly to the length of s 13A(1), to give at least some indication, even by reference to the headings, of the content of the proposed exemptions. It would be necessary to refer to the public interest factors where relevant but that could be done in the same way. To avoid doubt, the one line descriptions could pick up the Freedom of Information Act to the extent necessary by including some phrase such as “for the purposes of” or “within the meaning of” that Act. For further detail, the reader could be directed by a note to the Freedom of Information Act.

125  Last, recommendation 14(b) may not be necessary if it were to be covered by a broader equivalent to s 46 of the Freedom of Information Act concerning the infringement of the privileges of an Australian Parliament or Legislative Assembly.

126  The recommendation about s 14(1) suggests a corresponding tidying up of that provision rather than any substantive change to the certification procedure by the Attorney-General.

127  This last recommendation in the body of the Report contains a substantial policy element, on which I do not comment. However I do note that the recommendation seems to proceed on the assumption that any proceeding brought under the AD(JR) Act is public interest litigation. As that term is presently understood, that assumption is incorrect.

128  In addition, although this may be contestable, Oshlack v Richmond River Council (1998) 193 CLR 72 may stand for the proposition that the fact that proceedings may be characterised as “public interest litigation” may be a factor which contributes to a conclusion of “special circumstances” so as to justify a departure from the ordinary rule that the unsuccessful party pays the successful party’s costs. However, of itself it is not sufficient to justify a departure.

129  As things stand there is provision for costs capping orders to be made both at first instance and on appeal: see rule 40.51 of the Federal Court Rules 2011.

130  It may also be thought that the proposal would involve a change in the relationship between those affected by administrative decisions, on the one hand, and those making decisions on the other. No doubt as well there are budgetary implications. In any event it would seem necessary to consider a provision similar to s 570 of the Fair Work Act 2009 (Cth) dealing with vexatious claims or claims unreasonably brought and with claims prosecuted unreasonably such as to cause the other party to incur costs. Whether that is an appropriate policy setting outside the industrial context would be for others to decide. It is however to be borne in mind, as recognised by r 40.51(2) of the Federal Court Rules 2011 that costs orders are an important tool for the efficient and fair conduct of proceedings.

The minority proposal

131  As I earlier said, I want to end by spending a little more time on Mr Wilkins’ minority proposal which is, in summary, to repeal the AD(JR) Act and rely solely on constitutional judicial review, with Parliament setting general jurisdictional limits on decision-makers.

132  Mr Wilkins says this approach would fundamentally change the nature of judicial review in Australia: rather than a list of grounds focusing on the review powers of the courts, as in the AD(JR) Act, a list of jurisdictional limits would focus on decision-makers and the extent of their powers. Parliament could state, in general terms, the limits of the power of executive officers to make decisions under legislation, emphasising the standards that contribute to good decision-making. Setting out the jurisdictional limits on decision-makers would clarify which errors could ground an application for the constitutional writs. Particular pieces of legislation could retain more specific limits as appropriate.

133  This approach could also create a unified general system of federal judicial review, as opposed to the bifurcated system which currently exists, Mr Wilkins says. A list of jurisdictional limits could be designed to make explicit the limits on decision-making power which are currently implied by the courts into particular statutory schemes. Compared to the current list of grounds in the AD(JR) Act, a list of jurisdictional limits could be directed to decision-makers rather than courts and they could be clearly stated rather than implied from the statute. Applicants would have greater certainty about the power of decision-makers, referring to clearly listed limitations in a statute rather than arguing that particular limits are implied.

134  There is, in my opinion, much logic in the proposal for the reason that, since constitutional judicial review cannot be excluded, to achieve a fully unified general system of review it is necessary to repeal the AD(JR) Act.

135  Whether or not the AD(JR) Act should be repealed and whether there is a bifurcated system at present or whether there is a problem in that respect, I put to one side because I think it tends to obscure what is radical and significant in what Mr Wilkins proposes.

136  Although the AD(JR) Act has, since 1980, performed an educative function, the listing of grounds of review has diverted attention from the prior necessity of construing the legislation. This is despite what has been clear since at least Kioa v West (1985) 159 CLR 550 that the AD(JR) Act does not generally impose requirements which would not otherwise exist.

137  Secondly, to impose requirements on decision-makers, such as under the United States Administrative Procedure Act 1946, could redress that distinct tendency since the focus would then have to be on the particular powers exercised or purportedly exercised by the decision-maker.

138  Thirdly, the Parliament could then explicitly address what the consequences of non-performance of those requirements were to be and address what is presently almost always left to implication and a Project Blue Sky Inc v Australian Broadcasting Authority (1998) 194 CLR 355 analysis. This could be done by reference to substantive compliance, for example.

139  Even though jurisdictional error within s 75(v) probably remains in a state of change, Mr Wilkins’ proposal, if taken up, would show that Parliament was no longer content to leave marking out positively the scope of powers entirely to implication and thus entirely to the courts.

140  Primarily what is attractive about his proposal is that categories of legal error, in the abstract and by reference only to s 5 of the AD(JR) Act, should cease to be the almost automatic starting point for analysis. Similarly the conclusory statement of “jurisdictional error” should be able to be founded in an analysis of express limits on the grant of power by the Parliament to the Executive and express statements of the consequences of non-observance of those limits.

141  Although it would be necessary to have a savings provision in light of a question as to the extent to which the Parliament may describe jurisdiction conclusively,^[12] the focus would be, as it should be, on the nature, terms and purpose of the power, as a matter of construction, and the manner and effect of its exercise.

142  An added benefit, in my opinion, is that the Parliament would give up, as an experiment which failed,^[13] attempting to define jurisdiction by resort to generalised statements in legislation which the courts tend to construe as privative clauses purporting to limit remedies rather than define powers.

143  Perhaps the proposal is too theoretical. It is unlikely I should think that a Parliament would wish to have no statutory qualifications to judicial review and it will always therefore be likely that any statutory jurisdiction will not cover the same ground as s 75(v).

144  Ideally however a reform-minded Parliament would undertake the task Mr Wilkins envisages statute by statute and state expressly within what positive limits particular powers are to be exercised and state clearly the consequences for the ensuing decision of non-compliance with those limits. This would go a long way towards making predictable the outcome of a jurisdictional error/Project Blue Sky Inc v Australian Broadcasting Authority (1998) 194 CLR 355 analysis. If the result of the proposal was merely an exercise purporting to limit remedies rather than define powers it would be likely to fail.

The Curate’s Egg

"True Humility",
Bishop: "I'm afraid you've got a bad egg, Mr Jones";
Curate: "Oh, no, my Lord, I assure you that parts of it are excellent!"

145 As may be seen, the cartoon is about power relationships referable to, but no doubt not limited to, England in the 1890s. You cannot have a bad egg with good parts. You can have a report, such as this one, with good parts. The Administrative Review Council’s Report No 50 (2012) – Federal Judicial Review in Australia is nothing like the curate’s egg.