FEDERAL COURT OF AUSTRALIA

Minister for Indigenous Affairs v MJD Foundation Limited [2017] FCAFC 37

ORDERS

THE COURT ORDERS THAT:

1.    The appeal be dismissed.

2.    The Appellants pay the Respondent’s costs of the appeal as taxed or agreed.

Note:    Entry of orders is dealt with in Rule 39.32 of the Federal Court Rules 2011.

REASONS FOR JUDGMENT

PERRAM J:

1    Subsection 33(1) of the Acts Interpretation Act 1901 (Cth) (‘the Acts Interpretation Act’) provides that:

Where an Act confers a power or function or imposes a duty, then the power may be exercised and the function or duty must be performed from time to time as occasion requires.

2    The two issues in this case arise from the application of s 33(1) to s 64(4) of the Aboriginal Land Rights (Northern Territory) Act 1976 (Cth) (‘the Land Rights Act’). That provision empowers the responsible Minister to make a direction that an amount be paid ‘for the benefit of Aboriginals living in the Northern Territory’. A former Minister made a direction that the sum of $10 million be paid to the respondent (‘the Foundation’). It is not in dispute that a payment to the Foundation was a payment for the benefit of Aboriginals living in the Northern Territory within the meaning of s 64(4). Before the money was paid to the Foundation, however, there was a change in government, and the incoming Minister purported to revoke the former Minister’s earlier direction.

3    The first question in this case is whether s 33(1) empowered the incoming Minister to revoke the decision of the former Minister. The second question is whether, assuming that it could do so in principle, the scheme of the Land Rights Act was such that it evinced a ‘contrary intention’ within the meaning of s 2(2) of the Acts Interpretation Act. Subsection 2(1) provides that the Acts Interpretation Act applies to all Acts, but subs 2(2) provides that ‘… the application of this Act or a provision of this Act to an Act or a provision of an Act is subject to a contrary intention.’

4    The learned primary judge concluded that the power in s 64(4) was such that even if re-exercised it could not bring about a revocation of the earlier decision. This was because the power in s 64(4) was a power to direct the making of a payment and nothing else. It was not possible, on this view, by the re-exercise of such a power, to bring about a revocation of an earlier decision. In any event, his Honour also concluded that the scheme of the Land Rights Act indicated that the power could be exercised only once and exhibited a contrary intention for the purposes of s 2(2) of the Acts Interpretation Act. The trial judge therefore granted relief to the Foundation setting aside the incoming Minister’s decision.

5    On appeal, the incoming Minister contests the correctness of both conclusions. For the reasons which follow, the trial judge erred in his treatment of both issues. In fact, s 33(1) did authorise the incoming Minister to revoke the former Minister’s direction, and the scheme of the Land Rights Act did not evince a contrary intention for the purposes of s 2(2) of the Acts Interpretation Act. Consequently, the appeal should be allowed with costs, the orders of the trial judge set aside and in lieu thereof it should be ordered that the Foundation’s application to the Court below should be dismissed. The Minister did not seek his costs at trial and the appropriate order is that each party pay its own costs.

6    These reasons are structured as follows:

(1)    Relevant facts and statutory provisions;

(2)    Subsection 33(1) and revocation;

(3)    Whether the Land Rights Act evinces a contrary intention for the purposes of s 2(2) of the Acts Interpretation Act; and

(4)    Orders.

1.    Relevant facts and statutory provisions

7    This case concerns Indigenous persons in the Northern Territory. Provision is made by the Parliament for the conferral of Aboriginal land rights in the Northern Territory by means of the Land Rights Act. There is continued in existence by Part VI of that statute the ‘Aboriginals Benefit Account’. The Account is credited with amounts equivalent to the mining royalties received by the Commonwealth and the Northern Territory as a result of mining activities conducted on Aboriginal land held under the Land Rights Act. Section 64 specifies a number of circumstances in which payments may be made by the Commonwealth and the Account correspondingly debited. One aspect of the Account is the on-going funding of the operation of Aboriginal Land Councils under s 64(1). Another, relevant to this case, is the Minister’s power to direct that an amount be paid ‘for the benefit of Aboriginals living in the Northern Territory’. This power is conferred by s 64(4), and the legal consequence of a direction under that subsection is that the Commonwealth must pay the amount specified in the direction and the Account must be debited by the same amount. A number of provisions in s 64 are relevant for this appeal, including subs (3), (4), (5A), (5B) and (7). These are as follows:

Debits from the Account

…

(3)     There must be debited from the Account and paid by the Commonwealth, from time to time, to each Land Council in the area of which a mining interest referred to in subsection 63(1) is situated, or mining operations referred to in subsection 63(4) are being carried on, an amount equal to 30% of any amounts:

(a)    credited to the Account in accordance with subsection 63(1) in respect of that mining interest; or

(b)    credited to the Account in accordance with subsection 63(4) in respect of those mining operations;

as the case may be.

(4)     There must be debited from the Account and paid by the Commonwealth such other amounts as the Minister directs to be paid or applied to or for the benefit of Aboriginals living in the Northern Territory.

…

(5A)     The Minister may, by notice in writing, specify conditions on which a payment of an amount to a person under subsection (4) (including by way of a loan) is made. The notice is not a legislative instrument.

(5B)     If a condition on which a payment of an amount to a person under subsection (4) is made is breached, the Minister, on behalf of the Commonwealth, may:

(a)    if the payment is by way of a loan—recover so much of the loan as has not been repaid, and any accrued interest that has not been paid, as a debt in a court of competent jurisdiction; or

(b)    in any other case—recover the whole or a part of the amount as a debt in a court of competent jurisdiction.

…

(7)     Amounts that the Minister directs to be debited from the Account under subsection (4), (4A) or (6) must be paid or applied in accordance with the direction.

8    It will be observed that the Minister’s power is a power only to direct the payment or application of an amount. The Minister does not have the power to debit the Account, nor to direct the Commonwealth to pay the amount in question. The debiting of the Account and the corresponding payment by the Commonwealth are instead the legal consequences of the Minister’s direction.

9    The Foundation is a charity established in 2008 to provide a better quality of life for Indigenous persons and their families living with Machado Joseph Disease in Arnhem Land and beyond. The disease is a largely hereditary neurological disorder characterised initially by clumsiness in the limbs. In 2010, the Foundation applied for a grant of $6 million from the Account. It intended to use that money as the corpus of a fund the income derived from which would support its charitable activities. This grant was paid.

10    By 2012, the Foundation believed that its circumstances had changed and it needed more capital. There were three elements to this change. First, there had been since the time of the initial grant a substantial increase in the number of Indigenous persons presenting with the disease. There had been an upswing from 41 symptomatic individuals and 410 individuals at risk to 71 symptomatic individuals and 557 individuals at risk. Secondly, the future projected prevalence figures for the disease suggested that in the next generation, up to 5% of the population of some Indigenous communities might be affected. Thirdly, there was evidence which indicated that the strain of the disease which was prevalent in the Northern Territory amongst Indigenous people was the more aggressive ‘Joseph’ strain, which causes persons with the disease to become symptomatic younger.

11    In light of this, the Foundation resolved to increase the size of its capital sum from the $6 million granted from the Account in 2010 to some $16 million. This entailed applying for a variation to the grant of $6 million (awarded in 2010) to $16 million. An expression of interest was submitted to the Minister on 10 August 2012. This was followed by a formal application dated 15 May 2013.

12    On 22 July 2013, the former Minister signed a document entitled ‘Aboriginals Benefit Account Direction 2013 (No. 11)’. It was in the following terms:

Direction under subsection 64(4)

I, JENNY MACKLIN, Minister for Families, Community Services and Indigenous Affairs and Minister for Disability Reform, pursuant to subsection 64(4) of the Aboriginal Land Rights (Northern Territory) Act 1976, direct that the amount of $10,000,000 be debited from the Aboriginals Benefit Account and paid to MJD Foundation Incorporated in relation to the Consolidation of Grant Variation (additional funding) project.

This direction takes effect on the date of signature.

13    On 31 July 2013, the former Minister notified the Foundation by letter that its application had been successful. The letter said that it was a condition of the payment that the Foundation enter into a funding agreement with the Commonwealth under s 64(5A). Subsection 64(5A) expressly authorised the Minister to impose conditions upon the making of a payment from the Account. Nothing arises from the fact that the condition did not appear in the direction.

14    After the direction was made but before any payment there occurred, on 13 September 2013, elections for the House of Representatives and half of the Senate. As a result of the former, there was a change of government and the incoming Minister was appointed to the portfolio. By 18 December 2013, the condition that a funding agreement be entered into between the Foundation and the Commonwealth under s 64(5A) had still not been satisfied, and there continued to have been no payment by the Commonwealth to the Foundation of the $10 million under s 64(4). On that day, the incoming Minister wrote to the Foundation and informed it that its application for funding had been unsuccessful. For completeness, it should be noted that it was not suggested that the fault for the non-execution of the funding agreement lay at the feet of the Foundation.

15    There was some controversy in this Court, of a reasonably formal kind, as to whether the incoming Minister’s actions on 18 December 2013 should be seen as constituting a revocation of the first Minister’s decision, or whether they should instead be seen as a decision that the sum of $0 should now be paid. The incoming Minister’s decision was communicated by letter dated 18 December 2013. The pertinent parts of the letter were as follows:

Dear Mr Westbury

I am writing in relation to the application submitted to the Aboriginals Benefit Account from the Machado Joseph Disease Foundation Limited seeking funding under subsection 64(4) of the Aboriginal Land Rights (Northern Territory) Act 1976.

Since taking office as the Minister for Indigenous Affairs, I have reviewed all grant applications made to the Aboriginals Benefit Account as part of the Government’s broader examination of all spending.

The funding application from the Machado Joseph Disease Foundation for a $10 million grant to be invested in perpetuity to enable interest earned to fund the work of the Foundation has been unsuccessful.

This application was in addition to the previous $6 million grant from the Aboriginals Benefit Account that your organisation received in 2010 to earn interest which as stated in your application would “cover operational costs for the Machado Joseph Disease Foundation to enable it to be sustainable for many, many years to come”.

It has been a longstanding policy and practice to not provide recurrent funding from the Aboriginals Benefit Account. Notwithstanding the undoubted value of your project, I am of the strong view that providing a cash advance is an unsustainable and indefensible mechanism to work around this existing policy. This Aboriginal Benefit Account grant would set a dangerous precedent for the fund and the previous grant of $6 million should not have been approved in my opinion.

…

16    Also relevant to the ascertainment of what the incoming Minister actually did on 18 December 2013 are the reasons he produced for the decision under s 13 of the Administrative Decisions (Judicial Review) Act 1977 (Cth). These were forthcoming on 10 November 2014. The first paragraph of the substantive part of those reasons suggests that the Minister understood himself to have revoked the earlier direction:

I decided to revoke the Former Minister’s direction under s 64(4) of the Aboriginal Land Rights (Northern Territory) Act 1976, that $10,000,000 be debited from the ABA and paid to MJD Foundation in relation to “the Consolidation of Grant Variation (additional funding) project”.

17    Taken together it seems that the incoming Minister understood himself to be revoking the earlier Minister’s decision to pay $10 million to the Foundation. There is no evidentiary support in either the letter of 18 December 2013 or the reasons of 10 November 2014 for the notion that the incoming Minister had decided to direct the making of a payment of $0 to the Foundation.

18    Finally, it should be noted that the proceedings before the trial judge initially involved (at a pre-trial stage that was well in advance of the commencement of the hearing) the contention that in making the revocation decision on 18 December 2013, the Minister had failed to afford the Foundation procedural fairness. Arrangements were subsequently made for the incoming Minister to afford the Foundation procedural fairness, as a result of which he made a fresh decision on 30 June 2015 which was to the same effect as his earlier decision. It was accompanied by written reasons of the same date. In the section of those reasons under the heading ‘Decision’, the incoming Minister twice referred to himself as revoking the former Minister’s decision (‘The direction of the former Minister dated 22 July 2013 should be revoked’ and ‘The direction of the former Minister dated 22 July 2013 made under s 64(4) of the Land Rights Act is revoked.’) In an earlier part of the reasons, under the heading ‘Background’, the incoming Minister also said ‘I am satisfied that, with the $10 million directed by the former Minister not having been paid, I have the power to direct that it not be paid.’ A direction that a sum previously directed to be paid not be paid is, for all intents and purposes, the same as a revocation of the first direction. Taken as a whole, the evidence suggests that the second decision of 30 June 2015 was a decision to revoke. As with the first decision, and subject to the same qualification referred to in the preceding paragraph, the evidence does not support the notion that the incoming Minister on 30 June 2015 directed that an amount of $0 be paid to the Foundation.

19    For the purposes of the proceeding before the trial judge and also on this appeal, the parties agreed that these two decisions stood or fell together. If s 33(1) of the Acts Interpretation Act was available, both were valid; if it were not, neither was.

20    No more needs to be said about the facts. Although the appeal is only concerned with s 33(1), a number of the issues which arise also touch upon ss 33(2A) and (3), which should be set out. Subsection 33(1) is set out above at [1], but for ease of readability is now set out again. The relevant provisions are as follows:

33  Exercise of powers and performance of functions or duties

Powers, functions and duties may be exercised or must be performed as the occasion requires

(1)      Where an Act confers a power or function or imposes a duty, then the power may be exercised and the function or duty must be performed from time to time as occasion requires.

Meaning of may

(2A)      Where an Act assented to after the commencement of this subsection provides that a person, court or body may do a particular act or thing, and the word may is used, the act or thing may be done at the discretion of the person, court or body.

…

Power to make instrument includes power to vary or revoke etc. instrument

(3)      Where an Act confers a power to make, grant or issue any instrument of a legislative or administrative character (including rules, regulations or by-laws) the power shall be construed as including a power exercisable in the like manner and subject to the like conditions (if any) to repeal, rescind, revoke, amend, or vary any such instrument.

21    It is useful then to begin with the operation of s 33(1).

2.    Subsection 33(1) and revocation

22    Can s 33(1) be used by the repository of a power to revoke an earlier exercise of that power? The materials available as to what s 33(1) was intended to do are sparse. In none of the second reading speeches accompanying the introduction of s 33(1) (or provisions like it in other jurisdictions) does there appear to have been any discussion of what it was intended to do. Some historical materials are, however, available. As originally enacted in 1901, s 33(1) was largely in the same form as it is now, and provided:

Where an Act confers a power or imposes a duty, then, unless the contrary intention appears, the power may be exercised and the duty shall be performed from time to time as occasion requires.

23    This version differs from the current version of s 33(1) only in that the ‘contrary intention’ exception has been relocated to s 2(2), and the original version has no reference to functions. Neither difference is material to this discussion.

24    The original version of s 33(1) was materially the same as s 32(1) of the Interpretation Act 1889 (UK). In the general discussion of that Act in Sir Courtney Ilbert’s 1901 work, Legislative Methods and Forms, there is a suggestion that several parts of the Act were drawn from the General Clauses Act 1887 (India). It is very likely that Sir Courtenay would have known this, since he was not only the assistant parliamentary counsel to the Treasury in 1889 when the Interpretation Act 1889 (UK) was passed but had been, during his time serving in India, the person who in 1887 had introduced the General Clauses Bill into the Indian Council of the Governor-General. In the course of delivering his introductory remarks to that Council, Sir Courtenay explained that in the bill he proposed to make additions to the provisions of the previous General Clauses Act 1868 which were ‘based on my personal experience during the last few years’ and whose purposes included ‘generalizing certain definitions and rules of frequent occurrence’. Sir Courtenay’s General Clauses Act 1887 included a s 5 which provided:

Any power conferred on the Governor General in Council or on a Local Government by an Act to which this Part applies may be exercised from time to time as occasion requires.

25    There was no equivalent to s 5 in the previous General Clauses Act 1868 (India), so this was new. There may be good reason to think, therefore, that s 5 was one of the provisions Sir Courtenay had in mind in his comments to the Council, although about this one cannot be absolutely sure. By way of further conjecture, it is possible that Sir Courtenay may have obtained some inspiration for s 5 from s 10 of the Acts Shortening Act 1852 (NSW), which provided:

Whenever power shall be given to do perform or submit to any act matter or thing such power shall be capable of being exercised from time to time as occasion may require unless the nature of the thing or the words used shall indicate a contrary intention.

26    This is not identical to s 5 of the General Clauses Act 1887 (India), but there are striking resemblances. There is, however, no need to determine the ultimate headwaters of the current s 33(1). Although Sir Courtenay made no particular remarks about s 5 on the introduction of the bill, a contemporaneous commentator did. Writing in 1891 in his treatise, A Second Supplement to the Anglo-Indian Codes, the distinguished Celtic philologist Whitley Stokes observed of s 5:

The English rule that a power given to the Crown by statute having been once exercised is exhausted had been applied by Indian Courts to powers conferred by the Indian Legislature on the Governor-General in Council and on the Local Governments.

27    Sir Courtenay appears to have shared that view. In Legislative Methods and Forms, he included this explanation of s 32(1) of the Interpretation Act 1889 (UK):

The effect of this sub-section is to make the insertion of the words ‘from time to time’ usually unnecessary. Those words were formerly inserted for the purpose of removing the application of the doctrine that a statutory power is exhausted by its first exercise unless its repetition is expressly authorized. The presumption on this point is now reversed. If it is intended that a statutory power should not be exercised recurrently this should be stated.

28    This statement was then picked up in the 1913 first edition of Halsbury’s Laws of England, where it was said:

Where any statute passed since 1889 confers a power or imposes a duty, then, unless a contrary intention appears, the power may be exercised and the duty performed from time to time as occasion requires (f), and, when the power is so conferred or the duty imposed on the holder of an office as such, it may be exercised or performed by the holder for the time being of the office (g).

29    Footnote (f) was in these terms:

(f) Interpretation Act, 1889 (52 & 53 Vict. c. 63), s. 32 (1); Battersea Borough Council v. County of London Electric Supply Co., Ltd., [1913] 2 Ch. 248. There was an inconvenient common law doctrine of somewhat uncertain extent to the effect that a power conferred by statute was exhausted by its first exercise.

30    The nature of this inconvenient common law doctrine has proved elusive. The reference to Battersea does not assist, as that decision contains no reference to the common law doctrine under discussion. The editors of the third edition of Halsbury’s Laws of England suggested, in their largely identical discussion of s 32(1), a possible example of the doctrine. It was set out in footnote (r):

(r) Ibid., ss. 32(1), 42. The purpose of this provision was to overcome the inconvenience formerly caused by the doctrine that a statutory power is exhausted by its first exercise unless (as, e.g., in Battersea Borough Council v. County of London Electricity Supply Co., Ltd., [1913] 2 Ch. 248, C. A.) a contrary intention can be discovered. For an example of the operation of the doctrine, see the Union with Ireland Act, 1800 (39 & 40 Geo. 3 c. 67), art. 1, by which the Crown was empowered to determine by proclamation the royal style and titles to be assumed thereafter, and, the power having been exercised on 1st January 1801, the granting by the Royal Titles Act, 1876 (39 & 40 Vict. c. 10) (repealed), of a new power of amendment for the purpose of enabling the transfer to the Crown of the Government of India to be recognised. The availability of statutory powers is sometimes confined to a specified period. This is so particularly in the case of powers to acquire land compulsorily, and the principles then applicable are discussed in title COMPULSORY ACQUISITION, Vol. 10, pp. 20, 21.

31    The example is not, perhaps, especially mainstream. Whatever the content of the doctrine was, however, it seems likely that provisions such as s 33(1) have, since at least 1889, prevented it from having any further modern application or development. The uncertain extent of the doctrine referred to by the editors of the first edition of Halsbury’s seems unlikely, therefore, ever to be clarified. For present purposes, that probably does not matter. What does matter is that s 33(1) and its progenitors were drafted with a view to overcoming a rule whose content was thought to be that a power conferred by statute was exhausted by its first exercise.

32    On its face, that does not throw direct light on the question whether s 33(1) can be used to revoke an earlier exercise of power.

33    What then can be drawn from the structure of the provision itself? In answering that question, it is useful to observe a distinction between statutory powers which, on their face, appear only to be intended for use on a single occasion, and those which are couched in general language and which appear to be intended to be exercised in relation to very many different circumstances.

34    Examples of the former are relatively rare. Article 1 of the Union with Ireland Act, 1800 (39 & 40 Geo. 3 c. 67) referred to in footnote (r) in the third edition of Halsbury’s Laws of England (set out above) may be one such power. The power of the Queen under s 3 of the Commonwealth of Australia Constitution Act 1900 (Imp) to unite by proclamation the Australian Colonies into a Federal Commonwealth is probably another. Another example, perhaps closer to home, may be afforded by the Governor-General’s power under s 34(1) of the Federal Court of Australia Act 1976 (Cth) to cause registries of this Court to be established in each State.

35    Instances of the latter are, by contrast, common. Subsection 29(1) of the Migration Act 1958 (Cth) may serve as a convenient example. It provides:

(1)    Subject to this Act, the Minister may grant a non-citizen permission, to be known as a visa, to do either or both of the following:

(a)     travel to and enter Australia;

(b)     remain in Australia.

36    It is obvious that this power was not forever exhausted when pursuant to it the responsible Minister issued the first ever visa, so that neither he nor his successors thereafter could ever again issue another. But this is not because of s 33(1). It is because it is clear from the subject matter, scope and purpose of the Migration Act 1958 (Cth) that Parliament intended that the power should be exercisable in relation to multiple applicants. Although not the subject of direct discussion, the High Court appears to have assumed a similar operation for relevantly indistinguishable provisions in Day v Hunkin (1938) 61 CLR 65.

37    In the case of such generally expressed powers, however, s 33(1) may be seen to speak to the individual instances in which the power has been exercised. To continue with the example of the visa granting power in s 29(1) of the Migration Act 1958 (Cth) (and leaving aside, for the sake of argument, the balance of that statute whose Byzantine stipulations are likely to impact upon the correctness of the analysis), s 33(1) would provide a basis for re-exercising the power in s 29(1) in relation to a particular application by a non-citizen, notwithstanding that that application had already previously been dealt with by the Minister.

38    That would mean, on the obverse of the current facts, that s 33(1) would authorise the incoming Minister to exercise the power to direct the making of a payment under s 64(4) of the Land Rights Act, notwithstanding that the former Minister had already decided in the case of a particular application for funding that that direction should not be made.

39    But does s 33(1) authorise the Minister when so re-exercising the power to decide to revoke the earlier decision that the payment be made, in effect directing that the payment not be made? Mr Free of Counsel, for the Foundation, cogently submitted that the text of s 64(4) set out the content of the power which could be re-exercised. Subsection 33(1) did not change the nature of that power, but merely allowed it to be exercised more than once. This mattered because no matter how the power to direct that a payment be made under s 64(4) was performed or how many times it was exercised, a power to direct the making of a payment could not ever be transformed into a power to revoke an earlier direction to make a payment. A silk purse could not be made from a sow’s ear.

40    This engaging argument should be rejected for reasons relating to the text of s 33(1) itself. In relation to duties and functions, s 33(1) requires that such duties and functions must be exercised from time to time as occasion requires. But this is not what is required in the case of statutory powers. Of statutory powers what is stipulated is only that they may be exercised. The meaning of ‘may’ is then conveniently explained in s 33(2A), so that where a statute says that a person, body or court ‘may’ do a particular act or thing, then ‘the act or thing may be done at the discretion of the person, court or body’.

41    In the case of statutory powers, therefore, s 33(1) by its own use of the word ‘may’ explicitly contemplates that one outcome it authorises is that the power in question will not be exercised. So in this case, for example, s 33(1) expressly contemplates that although the former Minister had already exercised the discretion under s 64(4) to direct the making of the payment, it was open to the incoming Minister to re-exercise that power so as not to make the direction. Any other reading cuts across the definition of ‘may’.

42    Two objections may be raised against this argument. The first is that to revoke the earlier decision it would be necessary to read s 33(1) as having a retroactive operation. On this view, a decision-maker who sought to revoke an earlier decision would be utilising the same power as had been initially exercised but this time would do so by not exercising it, and that decision would have effect from the date of the original decision. Such an occurrence would bring about an avoidance of the original decision ab initio. The difficulty with such a reading of s 33(1) is that the expression ‘may be exercised … from time to time as occasion requires’ is not apt to include exercises of power with retroactive effect. ‘[A]s occasion requires’ is an inherently forward-looking expression and cannot, as a matter of ordinary language, permit the making of orders having effect from a date in the past.

43    This objection is, in part, sound. Subsection 33(1) does not authorise the making of any kind of decision with retroactive effect, and this is so regardless of whether the decision is a revocation decision or even a decision to exercise a power in a positive way. For example, s 33(1) cannot be used to decide that a decision made today should take effect from an earlier date. However, that is not sufficient to mean that s 33(1) cannot be used to revoke a decision – it merely shows that a decision cannot be revoked ab initio. It does not, in itself, mean that a decision may not be made today to revoke, with effect from today, a decision made in the past.

44    When the power to revoke is considered as an exercise of power in the present, then the fact that s 33(1) uses the word ‘may’ still therefore shows that an available outcome is that the power on its second exercise might not be exercised at all.

45    The second objection is related to the first. When the power being exercised is, as here, a power to take some positive step then there is the textual difficulty, strongly relied upon by Mr Free, that there is no mechanism by which such a power may be used as a power to reverse that positive step. In this case, the power actually conferred by s 64(4) is a power to make a direction. On this view, there is no way that that power can be read as a power to unmake a direction, for that is not the power conferred by s 64(4). Further, s 33(1) does not expand or alter the nature of the underlying power, but only permits its fresh exercise.

46    However, this second objection may be readily dealt with. Since s 33(1) explicitly contemplates through the word ‘may’ that on the re-exercise of a statutory power to which it applies the power need not be exercised, the provision must be taken to include within its grant sufficient power to give effect to that outcome. The only way that a decision under s 33(1) to re-exercise a power by not exercising it may be given effect is by the revocation of the earlier decision. Subsection 33(1) must be construed, therefore, as including such a power. This is not a surprising outcome. It is a well-known principle of statutory interpretation that where a power is conferred by a statute, there is an implied power to perform it: see Pearce D and Geddes R, Statutory Interpretation in Australia (8th ed, LexisNexis Butterworths, 2014), para 2.43. The learned authors of that work refer as authority for that proposition to the decision of Lockhart J in Re Sterling; Ex parte Esanda Ltd (1980) 44 FLR 125 at 130, but there are very many such statements in the reports. Re Sterling is a useful example nevertheless for present purposes. In that case it appeared that the Bankruptcy Act 1966 (Cth), whilst making reference to an application to set aside a bankruptcy notice, at no point actually conferred any power on a bankruptcy court to do so. His Honour concluded, under this principle, that such a power was implied.

47    So too here. Subsection 33(1) assumes by reason of the word ‘may’ that the power can be re-exercised by not exercising it. This assumption can only be given effect if the provision carries with it a power of present but not retroactive revocation. Consequently, s 33(1) authorises not only the re-exercise of a statutory power, but where the second decision is a decision not to exercise the power at all, also the revocation of the earlier decision from the date of the second decision.

48    The course of authority concerning s 33(1) and its analogues largely supports this conclusion. However, it may be accepted that the authorities contain no principled explanation of why the provision can be used in this manner. Further, it seems correct to say that all the statements which appear in relation to the ability to use the provision as a source of authority to revoke an earlier decision are obiter dicta. Were it not for the fact that, as a matter of construction, I have concluded that s 33(1) can be used to revoke an earlier decision, I would not have found the authorities in this area very persuasive.

49    The principal authority usually referred to is the House of Lords decision in Lawrie v Lees (1881) 7 App Cas 19. That case was not concerned with a provision such as s 33(1), but instead with the meaning of the expression ‘from time to time’ in a special statute dealing with the affairs of a baronet who had become insane. The question arose in relation to the ‘Horse Shoe Brewery’ (famously the site of the London Beer Flood of 1814) which had been operated by a partnership whose members included Sir Henry Meux. Sir Henry had become subject to proceedings in lunacy and his affairs were subject to special legislation, Sir Henry Meux’s Estate Act, 1863. Under s 3 of that Act, the Lord Chancellor (or his delegates) could ‘from time to time’ direct to be done etc with respect to the business of the Horse Shoe Brewery anything that Sir Henry, if of sound mind, might have done himself. Part of the Horse Shoe Brewery apparently included premises called ‘The White Bear Inn’ at Hampstead. The appellant agreed with the partnership to purchase the leasehold in the White Bear and then failed to complete the purchase. Specific performance proceedings ensued. The Lord Chancellor (or those lawfully representing him) had made an order under s 3 of the Act authorising a Committee dealing with Sir Henry’s affairs to execute all such deeds relating to the head lease as were approved by the Master in Lunacy. The appellant purchaser sought to argue that specific performance of the purchase should not be decreed because, not having been authorised by s 3, this order was invalid. It was not authorised by s 3 because the words ‘from time to time’ in s 3 did not permit a prospective authorisation of the form which the Committee had used.

50    This argument appears not to have been received enthusiastically. Although pursued in the Court of Appeal, that Court had not dealt with it, prompting Lord Penzance to muse (at 30) whether it had been pursued seriously in that Court, Lord Blackburn to ponder (at 39) that ‘I hardly know whether I should call it an argument’, and Lord Watson (at 41) to describe the appellant’s arguments as ‘of a very slender and unsubstantial character’. In any event, no member of the House accepted the argument. Lord Penzance rejected it in the following terms (at 29-30):

It is not one which I think your Lordships would lightly adopt, and I see no reason whatever for adopting it, because the words “from time to time” are words which are constantly introduced where it is intended to protect a person who is empowered to act from the risk of having completely discharged his duty when he has once acted, and therefore not being able to act again in the same direction. The meaning of the words “from time to time” is that after he has made one order he may make a fresh order to add something to it, or take something from it, or reverse it altogether; and as that meaning gives sufficient force to the words and explains the use of them here it seems to me that your Lordships ought not to go further and to narrow these words by any construction which would throw impediments in the way of carrying on the business, whereas the object of the Act was to facilitate it.

(Emphasis added.)

51    Lord Blackburn and Lord Watson agreed at 39 and 41 respectively. It will be apparent from what has been set out above that the italicized portion of Lord Penzance’s reasons dealing with whether the power could be successively used to detract from or reverse an earlier exercise of power was not in issue in Lawrie v Lees, and that therefore the statement is an obiter dictum. Indeed, it is an obiter dictum in relation to a matter which was apparently not even the subject of argument. Whilst the conclusion I would reach is that s 33(1) does indeed operate as Lord Penzance suggests, for myself, I am bound to observe that, with respect, his Lordship’s reasons involve no more than an unreasoned assertion for a proposition which is, at least to my mind, far from obvious.

52    Lawrie v Lees was applied by Glass JA in Parkes Rural Distributions Pty Ltd v Glasson (1986) 7 NSWLR 332. That case was concerned with a provision which authorised an officer to certify that an amount had been received by a person under a subsidy scheme in excess of their entitlement to receive that subsidy. The officer had issued two certificates. The first had certified the sum of $152,317.70 to be repayable and the second, and subsequent, notice certified the sum of $134,065.27. The appellant was seeking to avoid having to pay either. Its argument was that the first notice was voidable, because it had been issued without affording the appellant natural justice, and that the second notice had been issued before the first notice had, in law, been declared void by the Court. It was then said not to be possible to have two inconsistent certificates on foot at the same time. Consequently, as Sir Maurice Byers QC for the appellant argued, the second notice must have been invalid and should be set aside. Further, fortuitously the first notice was now also to be set aside for the want of procedural fairness. In consequence, the appellant owed money under neither certificate.

53    It is not difficult to see that this was not an attractive argument. The precise wording of the power in question (s 8(3) of the Petroleum Products Subsidy Act 1965 (Cth)) was as follows:

Where an authorized officer is satisfied that an amount paid to a person under this Act … was not payable to that person or exceeded the amount that was payable to that person, he may give a certificate in writing that the amount paid or the amount of the excess, as the case may be, is repayable by that person to the State.

54    Glass JA dismissed Sir Maurice’s argument in brief terms (at 335-336):

The trial judge rejected this argument and found on the evidence that the first certificate had been withdrawn and replaced by the second certificate. It was argued before us that no power was vested in authorised officers which would enable this to be done. I am unable to agree. The power to issue certificates granted by s 8(3) of the Act is elaborated by the Interpretation Act 1897, s 32, which permits its exercise from time to time. It has been held that a power exercisable from time to time may be so exercised as to add to, subtract from or reverse the result of the previous exercise: Lawrie v Lees (1881) 7 App Cas 19 at 29. In other words the donee of such a power is never functus officio. Accordingly the existence of the first certificate does not render void a second certificate issued in lieu of it.

55    Samuels and Priestley JJA agreed at 339. It will be apparent that this case was not concerned with revocation, but with a second exercise of the power in a way not wholly consistent with the first exercise. It may be that the ability to re-exercise a power in a way which is partially inconsistent with an earlier exercise of power can have extracted from it as a corollary the proposition that one may revoke entirely an earlier exercise of power. But the internal workings of the steps necessary to arrive at that outcome are not obvious. I do not think, therefore, that it would be correct to say that Parkes Rural has as its ratio decidendi that s 33(1) can be used to revoke an earlier decision.

56    Subsequent to Parkes Rural, there have been a number of decisions which, based on the above statement, have reasoned that the power in s 33 (or its analogues) can be used to vary or revoke a decision. Examples include Gorman v Health Care Complaints Commission [2000] NSWSC 1228 at [34] per O’Keefe J; Mauger v Wingecarribee Shire Council [2015] NSWSC 1022 at [91] per Davies J; and the dissenting judgment of Kourakis J in Southlink Pty Ltd v WorkCover Corporation of South Australia [2009] SASC 175 at [340].

57    However, acceptance has not been uncritical or uniform. For example, in Bond Corporation Holdings Ltd v Australian Broadcasting Tribunal (1988) 84 ALR 669 Gummow J, having affirmed the orthodox proposition that s 33(1) meant that the donee of the power was never functus officio, reserved his position on whether the statement of Glass JA was correct to the extent that it suggested that s 33(1) might be used to reverse an earlier decision in a way which was adverse to a party (at 678). On the other hand, this concern appears to have receded in his Honour’s mind by the time of the decision in Minister for Immigration and Ethnic Affairs v Kurtovic (1990) 21 FCR 193. There Gummow J (with whom Ryan J agreed on this issue) explained the matter this way (at 218-219):

In the present case, there is nothing in the Migration Act which suggests an intention contrary to the presumption embodied in s 33(1) of the Acts Interpretation Act 1901, to which I have already referred. Accordingly, the power to make a deportation order is exercisable from time to time, so as to revoke or revive a deportation order previously made, whether on the same facts as before or otherwise. Even if the facts upon which the original decision was based remain constant, it may be the policy of the donee of the power which changes and thus requires a reconsideration of decisions previously made: cf Laker Airways Ltd v Department of Trade [1977] QB 643 at 707, 708-709, 728. The significance of a change in either the facts or in ministerial policy would go merely to the merits of the decision upon which the Court is not entitled to decide. The appellant could not therefore have been functus officio, and an estoppel could not be allowed which would have the effect of stifling the future exercise of the statutory discretion: Southend-on-Sea Corp v Hodgson (Wickford) Ltd [1962] 1 QB 416; Rootkin v Kent County Council (supra) at 1195. These principles were affirmed in the application of the Migration Act in Re Chan and Minister for Immigration and Ethnic Affairs (1977) 17 ALR 432, per Smithers J (at 441-442) and in Dallikavak v Minister for Immigration and Ethnic Affairs (1985) 9 FCR 98, per Northrop and Pincus JJ (at 103-104). I would respectfully agree with what was there said, that if a deportation order is revoked, there is power to make another deportation order in reliance upon the same facts as those upon which the previous deportation order was based.

58    Strictly, this too was an obiter dictum on the issue of revocation. In Kurtovic, the question was not whether a previous exercise of a power could be revoked. Rather, it was whether a power could be re-exercised where it had previously been revoked.

59    Finally, mention should be made of the Victorian Court of Appeal’s decision in Kabourakis v Medical Practitioners Board of Victoria (2006) VAR 449. The principal reasons for judgment were given by Nettle JA, with whom Warren CJ and Chernov JA agreed. A complaint had been made about Dr Kabourakis’ treatment of a patient who had died. A panel appointed by the board which was dealing with the complaint concluded that he had not engaged in unprofessional conduct. A complaint was made to the Ombudsman who recommended that the inquiry be re-opened. The board sought to reopen the inquiry under a different provision, and Dr Kabourakis sought a declaration that the board was functus officio by reason of its first decision.

60    The case was concerned with the Victorian equivalent of s 33 (s 40 of the Interpretation of Legislation Act 1984 (Vic)). Nettle JA rejected (at [82]) an attempt to apply Parkes Rural directly, distinguishing it on the basis that it had been concerned with the question whether successive inconsistent certificates could be issued under one provision, whereas Dr Kabourakis’ case was about whether he could be subjected to a disciplinary process a second time under a different provision. Nettle JA put it this way at [82]:

As I see it, however, that is not so. Parkes was concerned with the question of whether the power to issue certificates from time-to-time under s 8(3) of the Petroleum Products Subsidy Act carried with it power to rescind and replace previous exercises of the power to issue certificates under s 8(3). In point of principle that is a different question to whether the power to act from time-to-time under one section of an act (such as s 25(7) of the Act) carries with power to rescind and replace previous exercises of another power under another provision of the Act (such as s 39). Nothing in Parkes suggests that the power to exercise a statutory power from time-to-time imports a power to rescind and replace previous exercises of a different statutory power.

61    So the primary conclusion was that Parkes Rural was silent on the issue before the Court. However, Nettle JA went on to doubt the reasoning in Parkes Rural. At [83]-[85] his Honour pointed out, with respect correctly, that Lawrie v Lees was about a rather special situation. His Honour did not observe, as I would, that the critical passage in Lawrie v Lees was an obiter dictum unrelated to any issue in the case. At [84] Nettle JA did, however, say this:

While, however, it is true that Lawrie v Lees did hold that a power exercisable from time-to-time may be so exercised as to add to, subtract from or reverse the result of the previous exercise of the power, it needs to be understood that the case was concerned with the power of the Lord Chancellor under s 3 of Sir H Meux’s Settled Estate Act 1863 to make orders “from time-to-time” for the purposes of administering the affairs of the partnership of which Sir Henry Meux was a member before he became a lunatic. With respect, it is drawing a long bow to conclude that, because the Lord Chancellor’s power to make orders from time-to-time for the purposes of the administration of a lunatic’s estate imported power to add to vary or revoke a previous order, the effect of s 32 of the Interpretation Act was to enable to the repository of the power to issue certificates under a provision like s 8(3) of the Petroleum Products Subsidy Act to revoke and replace an issued certificate. The decisions in Re 56 Denton Rd, Twickenham, Walter Construction Group Ltd v Fair Trading Administration Corporation, Export Development Grants Board v EMI (Australia) Ltd and Firearm Distributors v Carson all suggest the contrary.

62    This is an obiter dictum but it is a considered one. Were it not for the fact that I have concluded that s 33(1) necessarily carries with it as a matter of implication a power of revocation, I too would be unpersuaded by Lawrie v Lees. On the other hand, and with respect, I am not sure that I can agree that the authorities his Honour referred to at [84] are against the proposition that s 33(1) can be used to revoke an earlier decision. My reading of them is that they each held that the operation of the equivalent of s 33(1) had been excluded by the surrounding statutory context, rather than that the power itself did not extend to revocation. In Re 56 Denton Rd, Twickenham [1953] Ch. 51 concerned a decision of the War Damage Commission. The Commission could award compensation to the owners of property damaged by the enemy. Having adopted one approach to the issue of compensation, the Commission then adopted another. Vaisey J thought that where a decision was made which affected rights, and which had been communicated in a way which appeared final, it could not thereafter be withdrawn. I do not read these reasons as indicating that s 32(1) of the UK Act could not be used to revoke an earlier decision. I would read this only as having been a situation where a contrary intention was revealed. That reading of In Re 56 Denton Rd, Twickenham as being concerned with the expression of a contrary intention was expressly adopted by Gummow J in Kurtovic at 211-212.

63    Walter Construction Group Limited v Fair Trading Administration Corporation [2004] NSWSC 158 also does not appear to be authority for the proposition that s 33(1) (or an equivalent) can never be used as a power of revocation. What Grove J said in that case (at [40]) was:

I do not construe that provision as vesting a power to make and unmake decisions infinitely. If power does not stretch to infinity, there must be in the circumstances of a particular case and “as occasion requires” a terminus. In this case it was reached with the communication of decision by the letter of 24 October 2002.

64    But there are many re-exercises of power coming before ‘infinity’, and in respect of these it seems clear that Grove J was accepting that the NSW equivalent to s 33(1) could be used to revoke an earlier exercise of power. No doubt as a matter of interpretation, and in particular as in In Re 56 Denton Rd, Twickenham, where rights are being affected and communicated in an apparently final way, s 33(1) may be implicitly excluded, but that is not to say that s 33(1) does confer a power of revocation when it is not excluded.

65    In the case of Re Export Development Grants Board v EMI (Australia) Limited and Thorn EMI Electronics Pty Limited [1985] FCA 284 it was explicit that the Full Court’s conclusion concerned the presence of a contrary intention. It was not a statement that s 33(1) could not be used as a power of revocation. At [40] the Full Court (Bowen CJ, Woodward and Fisher JJ) observed:

Reference was made to the Acts Interpretation Act 1901, sub-s.33(1), which provides that where an Act confers a power or imposes a duty, then, unless a contrary intention appears, the power may be exercised and the duty shall be performed from time to time as occasion requires. The suggestion appears to be that the Board may exercise the power to make a determination of incentive grant entitlement under sub-s.11(1) as often as occasion may require. Just whether a perceived error in a past determination presents the Board with an occasion requiring another exercise of its power under sub-s.11(1) may be left on one side. It is plain that the structure of the Act is inconsistent with the existence of such a general power to re-assess. In our opinion, a contrary intention appears in the Act.

(Emphasis added.)

66    The same is also true of Firearm Distributors Pty Ltd v Carson [2000] QSC 159 at [33]. I would accept that all of these cases (and for that matter Kabourakis) establish that where the statutory scheme provides for a decision concerning rights following on some process of formal determination, this will usually exhibit a sufficient contrary intention to prevent s 33(1) from applying. However, to the extent that Nettle JA in Kabourakis suggested that these decisions show that s 33(1) is not itself a source of a power to revoke, I must respectfully disagree.

67    In those circumstances, I conclude that s 33(1) may be used to revoke an earlier exercise of power. There are opposed considered appellate dicta (Kurtovic and Kabourakis). My analysis of the text of s 33(1) satisfies me that I should follow the Kurtovic line. I would note for completeness that this conclusion appears to be in line with the Supreme Court of Canada’s decision in Comeau’s Sea Foods Ltd v Canada [1997] 1 SCR 12, where Lawrie v Lees was applied directly to a revocation situation.

Other arguments about s 33

68    There were two further arguments between the parties about the content of s 33(1). First, the Foundation advanced an argument that s 33(1) did not contain a power of revocation because it was clear that where s 33 intended a power of revocation to exist, it had used the word ‘revoke’. It pointed to s 33(3) which allows a statutory instrument, in terms, to be revoked. Secondly, the incoming Minister submitted that if s 33(1) did not include a power of revocation, then his decision should be understood as a decision to direct a payment of $0.

69    The Foundation’s additional argument should be rejected. Subsections 33(1) and (3) as they presently provide are set out above at [20]. In form this argument is an example of the principle of statutory construction that no inference should be drawn about the operation of a provision in a statute if it goes against the express words Parliament has used: expressum facit cessare tacitum. Dixon J put it perhaps more clearly in The King v Wallis; ex parte Employers Association of Wool Selling Brokers (1949) 78 CLR 529 at 550:

This accords with the general principles of interpretation embodied in the maxim expressum facit cessare tacitum and in the proposition that an enactment in affirmative words appointing a course to be followed usually may be understood as importing a negative, namely, that the same matter is not to be done according to some other course.

70    Reduced to its essentials, the Foundation’s argument is that because s 33(3) deals expressly with the topic of revocation, it should be inferred that s 33(1) does not.

71    Subsections 33(1) and (3) have undergone changes in the 115 years since they were first enacted. But, as explained above, both can be traced back to s 32 of the Interpretation Act 1889 (UK), and s 33(1) can be traced back further to the General Clauses Act 1887 (India). This is not so in the case of s 33(3), which does not appear in the Indian statute. Curiously, an equivalent to s 33(3) did exist in the Acts Shortening Act 1852 (NSW). This may suggest a closer relationship between that Act and the Interpretation Act 1889 (UK), but there is no way to resolve that issue and, for the purposes of these reasons, it does not matter.

72    It is useful to begin then with the relationship which existed in 1889 between ss 32(1) and (3) of the Interpretation Act 1889 (UK). As passed, they were as follows:

32.(1.)    Where an Act passed after the commencement of this Act confers a power or imposes a duty, then, unless the contrary intention appears, the power may be exercised and the duty shall be performed from time to time as occasion requires.

…

(3.)    Where an Act passed after the commencement of this Act confers a power to make any rules, regulations, or byelaws, the power shall, unless the contrary intention appears, be construed as including a power, exerciseable in the like manner and subject to the like consent and conditions, if any, to rescind, revoke, amend, or vary the rules, regulations, or byelaws.

73    There are two available views on how these two provisions initially interacted. One is that the topic of revocation was dealt with exclusively by s 32(3) so that whatever the content of the power in s 32(1), it could not include a power which could be used to revoke an earlier decision.

74    The other view of the two provisions is that s 32(3) was directed to the topic of by-laws, rules and regulations, whereas s 32(1) was not. On this view, the two provisions simply operated in relation to different subject matters. Subsection 32(3) could be used in relation to by-laws, rules and regulations, whilst s 32(1) could only be used in relation to exercises of statutory power which did not result in by-laws, rules or regulations.

75    Both of these arguments involve the drawing of a negative implication from the text of ss 32(1) and (3). They differ only as to the nature of the negative implication which is to be drawn. One is a negative implication about the subject matter of the powers to which each provision applies. The other is a negative implication as to the ways in which those powers may be exercised. Both views seek to avoid redundancy in the provisions by ensuring that they do not intersect in the fields of their operation. Either construction achieves that outcome.

76    Despite that, it is tolerably clear that it is much more likely that the Parliament intended s 32(3) to be an exhaustive statement about ‘by-laws, rules and regulations’, rather than an exhaustive statement about ‘revocations’. Subsection 32(3) did not solely concern revocation; in fact, the language it used was ‘rescind, revoke, amend, or vary’. If s 32(3) had operated as an exhaustive statement of the circumstances in which the exercise of a power could be revoked, then it must also have operated as an exhaustive statement of the circumstances in which such an exercise of power might be rescinded, amended or varied. This is so because there is no textual mechanism by which the various expressions might be distinguished from this perspective.

77    But if s 32(3) had operated exhaustively in that manner, then there would have been no work left for s 32(1) to do. That provision, once shorn of any ability to vary, amend or revoke a particular exercise of power, leaves open only the possibility of exercising the power in the same way as the initial exercise. There would be no purpose to s 32(1) so construed. This suggests, therefore, that s 32(3) did not operate exhaustively with respect to the functions of revoking, rescinding, amending or varying, but instead operated exhaustively with respect to the topic of regulations, by-laws and rules. It follows that revocation is not excluded from s 32(1). On the other hand, that requires one to accept that ss 32(1) and (3) did have different fields of operation which did not overlap. As originally enacted this would entail that s 32(3) applied to exercises of statutory power which resulted in rules, regulations or by-laws, and s 32(1) applied to the balance of all other statutory powers to which s 32 applied.

78    With minor immaterial differences, this form of ss 32(1) and 32(3) was carried over into ss 33(1) and 33(3) of the Acts Interpretation Act as originally enacted. The structural relationship described in the preceding paragraphs was thereby preserved.

79    It is then necessary to trace through the changes made to s 33 over time, to ascertain whether that structural relationship was maintained.

80    Subsection 33(3) was first amended in 1941 by the Acts Interpretation Act 1941 (Cth). The second reading speech explained the purpose of the amendments as follows:

Opportunity has also been taken, in drafting the bill, to amend section 33 of the Principal Act in order rectify an omission that occasionally is the cause of some inconvenience. At present, subsection 3 of section 33 provides that a power to make rules, regulations or bylaws shall be construed to include a power to rescind, revoke, amend or vary such rules, regulations or by-laws. The expression “rules, regulations or bylaws” does not, however, exhaust the instruments which may be made under an act. It will readily be appreciated by honorable members that if a power be conferred to make or issue any instrument under an act, then the power to repeal or amend that instrument at some later date is necessary, even though the instrument does not fall under the description of “rules, regulations or by-laws”. Examples of instruments not at present covered by the expression are proclamations and orders. Consequently, the bill proposes to amend the sub-section so that it will refer to any instrument made, granted or issued under a power conferred by an act.

81    The amendments, in fact, made two changes. The first was to replace ‘rules, regulations, or by-laws’ with ‘instrument (including rules, regulations or by-laws)’ as contemplated in the second reading speech. The second was to replace ‘rescind, revoke, amend or vary’ with ‘repeal, rescind, revoke, amend, or vary’. This only had the effect of adding ‘repeal’. It is unlikely that this had any substantive impact on the provision.

82    Although the purpose of the first of these changes appears to have been to ensure that the word ‘instrument’ was interpreted broadly, this did not occur. A series of decisions thereafter suggested that the power in s 33(3) was limited to instruments of a legislative character: see, e.g., Australian Capital Equity v Beale (1993) 114 ALR 50 at 63. This question was, however, an unsettled one and its unsatisfactory state was usefully summarised, with respect, by Finkelstein J in Leung v Minister for Immigration and Ethnic Affairs (1997) 150 ALR 76 at 84. The question became moot in 2011 on the passage of the Acts Interpretation Amendment Act 2011 (Cth), which changed the expression used in s 33(3) to its current form of ‘instrument of a legislative or administrative character’.

83    Two matters emerge from these events. First, the changes made in both 1941 and 2011 proceeded on an assumption that s 33(1) could not have been used to achieve the desired outcome. In that sense, both sets of changes are consistent with the structural relationship between ss 33(1) and (3) which I favour. Secondly, nothing in this legislative history provides any support for the idea that the original relationship between the provisions was intended to be disturbed by the amendments.

84    It follows that the Foundation’s argument should not be accepted. It is true that ss 33(1) and (3) do not have an overlapping operation, but this is because one deals with instruments and the other with exercises of statutory power not resulting in instruments.

85    For completeness, it might be noted that the Minister expressly eschewed any suggestion that the document entitled ‘Aboriginals Benefit Account Direction 2013 (No. 11)’ (set out at [12] above) was an ‘instrument’ for the purposes of s 33(3).

86    I would also reject the incoming Minister’s additional submission that even if s 33(1) did not include a power of revocation, he could nevertheless make a direction that a payment of $0 be made and this would bring about the same outcome. As a matter of fact, this is not what occurred. The incoming Minister has never made such a direction. The question of whether he could make such a direction therefore does not directly arise in this case.

3.    Whether the Land Rights Act evinces a contrary intention for the purposes of s 2(2) of the Acts Interpretation Act

87    The trial judge accepted the Foundation’s argument that the scheme of the Land Rights Act demonstrated a sufficient contrary intention to displace the operation of s 33(1). There were four strands to the argument, each of which was also advanced in this Court.

88    The first was that s 64(4) created an immediately existing obligation to pay. The words in s 64(4) ‘[t]here must be debited’ could be seen as having no other meaning. Further, once the direction was made, s 64(7) required that the amount debited from the Account ‘must be paid or applied in accordance with the direction’. Whichever way one examined the matter, so the argument ran, the making of the direction set in train a series of events expressed in mandatory language. It would be surprising, if once that process had been commenced, it could be interrupted by the exercise of a power of revocation in s 33(1).

89    Although the Foundation sought to depict s 64(7) as showing the inevitability of the payment, in my opinion, the provision shows the opposite. The direction made under s 64(4) may have attached to it conditions under s 64(5A), and these may include, obviously enough, timing stipulations. What s 64(7) ensures is that any additional requirements caught up in the direction are enforced. One such restriction would include a direction that the amount be debited by a particular date or not until after such a date. Further, the power to impose conditions under s 64(5A) could result in a situation where the conditions were not satisfied. For example, a direction to pay might be subject to a condition that the recipient lodge financial information of a particular kind by a particular date. If that condition were not satisfied, the Foundation’s argument would lead to the situation that even though the direction might never be given effect to (because s 64(7) would never permit the payment to be made), nevertheless the Minister would remain powerless to revoke the direction which would, on this view, forever remain in place.

90    So to reason is not to advance an argument that s 64(4) should be construed to include a power of revocation. Rather, it is an observation which makes it difficult to conclude that the statutory context of s 64 evinces an intention to the contrary.

91    Secondly, the trial judge then focussed on s 64(5B), which provides for the recovery of grant monies if a condition which was to be satisfied had not been satisfied. His Honour reasoned that there would be little point to such a provision if the revocation by the Minister of the earlier direction under s 64(4) could bring about the same result. There are two aspects of this from which I would respectfully differ. The first is that the observation is of no assistance in determining the issue involved in this proceeding, which is instead whether the direction can be revoked before any money is paid. The second is that it is by no means obvious, in any event, that even if the money had been paid pursuant to the direction, it would be recoverable if the direction were then revoked. It could not be said that the Commonwealth had paid the money under a mistake because a revocation is not, for the reasons I have given, retroactive. Nor would there be a failure of consideration, because at the time of the payment the direction was in place. The subsequent non-retroactive revocation of the direction has no effect on that situation. This may be contrasted with a situation where money is paid under a judgment which is subsequently reversed. There the revocation is retroactive: see Edelman J and Bant E, Unjust Enrichment (2nd ed, Hart Publishing, 2016) p. 265. In this case, all that has happened is that the Minister has changed his mind. If the money had been paid before he did so, it could not be said that this would result in an outcome that was unjust if the recipient did not give it back. On this alternative scenario, the fact would remain that the payment was authorised at the time it was made. In those circumstances, I do not accept that a restitutionary remedy would be available.

92    Thirdly, the trial judge drew attention to s 35A. One of the obligations imposed on the Minister by s 64(3) was a particular obligation to remit to a Land Council on whose land mining operations were conducted 30% of the royalty revenue. This royalty revenue was then required to ‘trickle down’ to Aboriginal and Torres Strait Islander corporations whose members lived in areas affected by the mining operations. However, unlike the fixed proportion of 30% of the revenue which was directed to the overarching Land Council, the Act does not fix a precise amount to be passed on to these downstream corporations. Instead, s 35(2) provides for the upstream Land Council to determine the proportions in which the various downstream corporations are to share the revenue. Section 35A is part of the calculus for determining that proportion. Subsection 35A(1) sets out the various matters which a Land Council must take into account in making a determination such as, for example, financial reports. Subsections 35A(2) and (3) then provide:

35A Making of determinations under subsection 35(2), (3) or (6)

…

(2)    A determination under subsection 35(2), (3) or (6) must be in writing and must specify the period, not exceeding 5 years, for which it is in force.

(3)     A Land Council may vary or revoke a determination under subsection 35(2) or (3). The variation or revocation must be in writing.

93    The argument is that in the context of grants under s 64, the Parliament has expressly given a power of revocation in relation to a determination under s 35A(3). Hence, it is less likely that s 33(1) applies to s 64(4). I do not accept this argument. The subject matter of s 35A(3) is an instrument setting out a proportion which can stay in place for up to 5 years. In my opinion, s 35A(3) is not a comparable subject matter to s 64(4). The content of one does not have any impact on the meaning of the other.

94    The Foundation’s final argument drew upon a line of cases culminating in Kabourakis which held that a decision effecting a determination could not be revisited. These included the decisions earlier referred to in In Re 56 Denton Rd, Twickenham, Re Export Development Grants Board v EMI (Australia) Ltd (1985) 9 FCR 269 and Firearm Distributors Pty Ltd v Carson [2000] QSC 159, [2001] 2 Qd R 26 at 30. They also included my own decision in Cassaniti v Tax Agents Board of New South Wales (2009) 179 FCR 1. In that case I accepted, in principle, that the Tax Agents Board could not repeatedly discipline a tax agent for the same conduct by means of s 33(1). But each of these cases was concerned with something approaching a formal decision-making process which resulted in a decision which would ordinarily be expected to be final. In Kabourakis, the panel in question had concluded that Dr Kabourakis had not engaged in unprofessional conduct. In In Re 56 Denton Rd, Twickenham, the conclusion was that the property which had been partially demolished had a particular character. In Re Export Development Grant, the Board had to assess the earnings increment for a particular year. And, in Firearm Distributors, the decision-maker had determined what quantum of compensation was due to a plaintiff.

95    What all of these cases have in common, as the Minister correctly submitted, is that they all involve the adjudication of an entitlement to a benefit based upon demonstration of specified criteria. I would accept that where an essentially adjudicative function is involved, there will ordinarily be an implication that the statutory power involved is not intended to be re-exercised where it appears that it has been exercised in a fashion intended to be final: cf. Firearm Distributors at [40] and, in a slightly different context, Leung v Minister for Immigration and Multicultural Affairs (1997) 79 FCR 400 at 410.

96    This is not such a case. The former Minister was not adjudicating any entitlement of the Foundation. She was merely determining a grant application. I would not discern from such a statutory framework any intention to oust the operation of s 33(1).

4. Orders

97    The appeal should be allowed with costs. The orders of the primary judge made on 23 November 2015 should be set aside and in lieu thereof it should be ordered that the originating application be dismissed.

Associate:

Dated:    3 March 2017

REASONS FOR JUDGMENT

MORTIMER J:

98    I have had the advantage of reading the reasons for judgment of Perram J. I take a different view of the application and scope of s 33(1) of the Acts Interpretation Act 1901 (Cth), which has led me to a different view on the outcome of the appeal, but I gratefully adopt Perram J’s summary of the circumstances in which the issues concerning s 33(1) arise in this proceeding, and I do not propose to repeat them.

99    I agree with his Honour’s conclusion (at [17]) that the proper characterisation of the Minister’s decision in this case is that he revoked the earlier direction made by the previous Minister, rather than that he made a decision and gave a direction that $0 be paid to the respondent. I prefer not to express a view on the matters regarding the conclusion of a funding agreement in [13] to [14] of Perram J’s reasons, which in my opinion are unnecessary to decide for the purposes of this appeal.

100    In summary, the view I take is that, on the assumption that the approach taken by Perram J to s 33(1) is correct, there is a contrary intention evinced by the scheme of the Aboriginal Land Rights (Northern Territory) Act 1976 (Cth) (Land Rights Act) as a whole (and Pt VI of the Act, in which s 64(4) is contained, in particular), so that s 33(1) is not applicable. That is a sufficient basis on which to determine the appeal. However, had I not been satisfied of the existence of a contrary intention, in my opinion, the scope of s 33(1) does not extend to a general implication of an additional power to reverse or undo an exercise of power, whether by revoking a decision made in the exercise of a power or otherwise.

A contrary intention is present in the Land Rights Act

101    I respectfully agree with the following statement made by Lander J in Minister for Immigration and Multicultural and Indigenous Affairs v Watson [2005] FCAFC 181; 145 FCR 542 at [118] as to the location of a contrary intention:

The contrary intention may appear not only in the particular legislative provision which creates the power but also in the Act as a whole: Pfeiffer v Stevens (2001) 209 CLR 57 per Gleeson CJ and Hayne J at [20] and McHugh J at [56].

102    The parties’ submissions tended to focus on the terms of s 64(4) in isolation from the other provisions in s 64, and in further isolation from the whole of the Act. In my opinion, the scheme of the Land Rights Act, and the scheme for the making of grants from the Aboriginals Benefit Account in particular, evinces an intention contrary to the implication for which s 33(1) provides, even if that implication is to be construed as Perram J sets out.

103    The Land Rights Act serves a number of discrete, and quite different, purposes. What those purposes have in common is the use of Aboriginal land (as defined) in the Northern Territory, and the control, protection, benefits and other rights and responsibilities that are to flow to Aboriginal people, and those who act on their behalf, in the Northern Territory as a result of the use and exploitation of Aboriginal land. It is under Pt III of the Land Rights Act that Aboriginal Land Councils are created, by reference to geographic regions (see ss 21 and 21C). By s 23, Land Councils have a broad range of functions, including functions of representation, advice, agency, protection, consultation, development assistance – all in relation to Aboriginal land held in the Northern Territory. The Land Councils also have some supervisory functions over Aboriginal Land Trusts established under Pt II of the Land Rights Act.

104    Part VI deals with the “Aboriginals Benefit Account”, which the note to s 62(1) states was established under s 5(3) of the Financial Management Legislation Amendment Act 1999 (Cth). Section 62(1) continues the Account in existence.

105    Funds are credited to the Account pursuant to s 63. As the facts of the current proceeding illustrate, the size of the fund, and the size of payments which may be made from it for the benefit of Aboriginal people, are considerable indeed. The source for the funds is mining royalties “received by the Commonwealth or the Northern Territory in respect of a mining interest in Aboriginal land”. Subject to determinations made by the Minister under s 63(2) and (3) (which in substance allow the Commonwealth in certain circumstances to retain for itself, and not pay into the fund, monies received by reason of an increase in mining royalties), all mining royalties are to be paid into the Account. Again, in my opinion there is no suggestion from the text, context or purpose of the determination power conferred on the Minister in s 63(2) and (3) that Parliament intends the Minister may revoke such a determination. What the Minister would be able to do (irrespective of s 33(1), in my opinion) is to make a determination which altered (upwards or downwards, including removing) the part of the royalties increase not payable to the fund but payable to the Commonwealth. The Minister would achieve this by simply making a new determination, which would operate prospectively.

106    Section 63 also contains a provision which is in similar terms to s 64(4). These provisions deal with mining operations carried out by or on behalf of, broadly, government, rather than private entities. Sections 63(4) and (5) provide:

(4)    If mining operations for minerals are carried on under the Atomic Energy Act 1953 or any other Act on Aboriginal land by, or on behalf of, the Commonwealth, the Northern Territory or an Authority, there must be credited to the Account, from time to time, payments in respect of those mining operations of such amounts as are determined in accordance with subsection (5).

(5)    Amounts payable under subsection (4) are:

(a)    such amounts as would be payable to the Crown as royalties if the mining operations were carried on in accordance with the law of the Northern Territory relating to mining for minerals and the royalties were payable at the rate fixed by that law immediately before 26 January 1977; or

(b)    if the Minister administering the Act under which the mining operations are carried on and the Minister administering this Act jointly determine that amounts higher than the amounts referred to in paragraph (a) are to be payable under subsection (4) in respect of particular mining operations—the amounts so determined.

(Emphasis added.)

107    Like s 64(4), the statute imposes an obligation to pay upon – as one alternative – Ministerial direction. The crediting of amounts pursuant to s 63(4) will be an ongoing process, as royalties become payable. Correspondingly, the power in s 63(5)(b) to determine an amount other than the amount set out in s 63(5)(a) is a power that must, on its face, be exercisable from time to time. There is no need for an implication of the kind for which s 33(1) provides, because the nature of the power itself is, in its context and purpose, ambulatory. So too, there is no need to imply a revocation power because a new Ministerial direction can be made altering the amount payable.

108    The Land Rights Act also expressly incorporates the concept of “from time to time” in s 63(7):

There must be credited to the Account amounts equal to any amounts from time to time received by the Commonwealth as interest on, or as repayment of, any loan made under subsection 64(4).

109    Section 63 deals with credits to the Account, and s 64 with debits from it. Both are conceived by the scheme to be ongoing processes of payment and disbursement. There is no need for an implication of the kind for which s 33(1) provides on its face: namely, that a power may be exercised from time to time as occasion requires. That is inherent in the nature and purposes of these powers as exercisable to allow monies from mining royalties to flow into the Account, for the benefit of Aboriginal people (whether directly or through organisations and entities representing and assisting them), and monies to flow out to fund activities of benefit to Aboriginal people and their land in the Northern Territory. There is no justification for construing s 64 on anything other than a consistent basis with the provisions which precede it in Pt VI, and to which it is expressly related.

110    Turning then to s 64, s 64(1) and (3) deal with payments to Land Councils. The payments for which s 64(3) provides are fixed by operation of the provision itself, by reference to the location of the Land Councils and the particular mining interests and operations in that area, and by reference to the proportion of 30% which the provision fixes as the proportion of royalties to be paid directly to Land Councils.

111    By s 64(1), the Minister has a determination power (“such amounts as the Minister determines”) to make payments to Land Councils by reference, principally it would seem, to the expenditure estimates Land Councils are required to prepare and submit pursuant to s 34 of the Act. These are, in large part, the funds on which the Land Councils depend for the performance of their functions under s 23.

112    There is nothing in the text, context or purpose of the determination power in s 64(1) which would suggest that it was intended to be a power that could be exercised to revoke a determination made by the Minister about how much Land Councils would receive. It is clear from its text that it is a power which may be exercised from time to time: that much is evident from the use of the phrase “such amounts as the Minister determines” and the nature of the power as one by which Land Councils are funded. However, it would be inimical to the nature of the power if the Minister could determine that a Land Council should receive, for example, $50 million for a particular period of time, but then a month later (for example, after a change in federal government) revoke that decision and decide that it should receive only $10 million. To construe the power in that way would undermine the evident purpose of Pt III of the Land Rights Act, which is that Land Councils perform a wide range of functions which are dependent on receipt of stable funding. To take this approach does not fetter or control the discretion reposed in the Minister to determine the quantum of such funding. Rather it seeks to give effect to reposing that discretion in the Minister by construing the exercise of power as perfected once a determination is made and communicated. Revocation and reversal of funding arrangements committed to and communicated has large consequences for organisations dependent on such funding. If Parliament intended the scheme to empower Ministerial reversals and revocations (including based simply on changes of mind, or changes in the occupant of the Ministerial office), in my opinion this would make the scheme operate in such a significantly different way that express words would be necessary.

113    After the impugned provision in s 64(4) appears s 64(4A). That is also a directions power conferred on the Minister in relation to payments for leases and subleases entered into, broadly, by Aboriginal and Torres Strait Islander corporations under Pts II and IIA of the Land Rights Act. It is hardly to be thought that Parliament intended the Minister could direct that a certain sum be paid in relation to a certain lease or sublease and then, months later, revoke that decision. That would have the potential to throw the operation of the Land Rights Act, including the performance by corporations dependent on public funding of their obligations under leases, into chaos.

114    Some weight was placed on s 64(5), (5A) and (6), which deal with conditions that may be imposed on payments made pursuant to s 64(4), as well as (by s 64(5)) allowing for the payments to be by way of loan. In argument, and in the reasons of Perram J, the conditions provisions are considered to be connected to the duty set out in s 64(7). I deal with these arguments at [129]-[132] below.

115    Section 64(7) should be set out again. It provides:

Amounts that the Minister directs to be debited from the Account under subsection (4), (4A) or (6) must be paid or applied in accordance with the direction.

116    This is the second of two subsections in s 64 making it clear the payment of funds is a mandatory consequence of the giving of a Ministerial direction. The scheme could not, in my opinion, be clearer that it intends that consequence occur once a direction has been given. The direction is a direction to pay monies. It is not, in terms, a direction that a person or entity receive monies. The source of the duty is the direction itself. The scheme does not contemplate that a Minister (whether the same individual, or a new occupant of the office) will have a capacity to preclude or prevent that duty imposed by Parliament from being carried out.

117    Part VI is, comparatively, a small section of the Land Rights Act. In ascertaining whether there is a contrary intention evinced in s 64, the whole of the Part, and its place in the overall legislative scheme, should be examined: Project Blue Sky Inc v Australian Broadcasting Authority [1998] HCA 28; 194 CLR 355 at [69] (McHugh, Gummow, Kirby and Hayne JJ); Independent Commission Against Corruption v Cunneen [2015] HCA 14; 256 CLR 1 at [31] (French CJ, Hayne, Kiefel and Nettle JJ). By s 64, the maintenance of the Account is not to deliver any profits to the Commonwealth by way of interest: in other words, the entirety of the Account, including any interest earned, is intended by Parliament to be applied for the purposes and in the manner set out in Pt VI. By s 62A, the Minister is to ensure there is a minimum investment amount maintained in the Account. That is to be done by way of a written determination. There is no suggestion in the text or context of the scheme that the s 62A power would be affected by an implication of the sort in s 33(1) of the Acts Interpretation Act. Indeed, any such implication would be contrary to the purpose of the provision.

118    What all the provisions of Pt VI have in common is that they are funding provisions. The underlying assumption of the Land Rights Act is that funds received as profits from mining of Aboriginal land should be returned to those whose land has been exploited for mining purposes. A detailed scheme is established through a range of corporations and entities designed to serve the interests of Aboriginal people. To serve those interests and perform the functions given to them under the Land Rights Act, those organisations and entities require funding. The Minister is empowered to determine for herself or himself some of the amounts of funding, and the Land Rights Act sets others by express provision itself. For my own part, I cannot see how this is a scheme in which there is an intention that those kinds of funding decisions can be altered, reversed or revoked on the basis of a change of mind of the Minister, or a new Minister (as opposed, for example, to a recognition of excess of power in the sense identified by the High Court in Minister for Immigration and Multicultural Affairs v Bhardwaj [2002] HCA 11; 209 CLR 597). Rather, the scheme intends these decisions will be perfected and result in payments of funds so that functions can be performed. I do not see the power in s 64(4) as operating any differently. It is expressed in common terms with the other provisions, it is located with them, and it serves the same kind of purposes. The core difference is that by s 64(4) the availability of funds to be employed for the benefit of Aboriginal people is extended beyond the entities and organisations created by the Land Rights Act itself.

119    It seems to be accepted, including by the appellants, that the operation of s 33(1) is confined to a period of time before funds are paid to a recipient pursuant to the duty set out in s 64(4). In their written submissions, the appellants contended:

The analogy here would have been if the 31 July 2013 direction of the Minister had been carried into effect and a payment made from the Account. A subsequent exercise of power under s 64(4) would not deny the legal authority for the payment at the time it was made. It would thus not provide the foundation for an action for recovery of the payment. However, this says nothing about whether, before any payment is made, the direction which would provide the legal authority for the payment can be revoked.

…

The fact that, following a direction, there “must” be debited from the Account (within the time specified or, absent specification, a reasonable time) the amount the subject of the direction says nothing about whether, before the debiting occurs, the direction can be revoked. Likewise, the fact that the amount that must be debited is the amount “as the Minister directs”.

120    To my mind, this is an unsatisfactory distinction if one accepts s 33(1) operates to confer a power of revocation. If it is based on the notion of a decision being “perfected” (see E Campbell, ‘Revocation and Variation of Administrative Decisions’ (1996) 22 Monash UL Rev 30 at pp 38-39), then as Professor Campbell explains, a decision is generally considered to be perfected upon its communication to the person affected. The payment of funds under s 64(4) is a separate statutory duty, imposed on another person or entity: namely, the Commonwealth. The appellants’ approach would make the operation of s 33(1), and the nature of the implication in s 64(4), dependent on the alacrity with which the Commonwealth bureaucracy functioned in a given case. I do not accept that is how this statute was intended to operate. The learned primary judge reached the same conclusion at [30] of his Honour’s reasons and I respectfully agree.

121    Professor Campbell’s observation at p 68 has, with respect, considerable force:

Courts have, in a number of cases, concluded that administrative decisions which they have accepted as valid and perfected decisions were irrevocable by their maker. The decisions held to be irrevocable have, in the main, been ones concerning individual entitlements or liabilities, and decisions on applications for permits or grant of cash benefits. A general rule that valid and perfected decisions cannot be altered by their authors promotes finality in decision-making. It also protects the interests of individuals who have acted in reliance on those decisions and who may suffer detriment if the decisions are rescinded or varied.

122    The terms of the Direction given by Minister Macklin should also be noted: see [12] of Perram J’s reasons. The Minister’s Direction stated that it took effect on the date of signature, which on the evidence was 22 July 2013. The evidence was that the making of the direction was communicated to the respondents by a letter from Minister Macklin dated 31 July 2013, in which the Minister wrote:

I am writing to you about MJD Foundation Limited’s large project application for funding from the Aboriginals Benefit Account under subsection 64(4) of the Aboriginal Land Rights (Northern Territory) Act 1976 to support the Consolidation of Grant Variation (additional funding) project.

I am pleased to advise you that I have approved funding of $10,000,000 for the proposal.

Please note it is a condition of this payment under subsection 64(5A) of the Act that you enter into a funding agreement with the Commonwealth and comply with the terms of that agreement. A funding agreement will be sent to you by my Department.

It is pleasing to see organisations such as yours taking the initiative in putting forward proposals like this that will benefit Aboriginal people living in the Northern Territory.

I look forward to hearing about the success of the project.

123    Although there was no evidence before the trial judge that the respondent had engaged in any specific conduct in reliance on the direction given by Minister Macklin, given the nature of the respondent, the subject matter of its activities and the urgent need which was set out in the material forming part of the application, it is reasonable to infer that, at least, on receipt of the communication MJD relied on receipt of the funds pursuant to the duty in s 64(4), and relied on having those funds available for expenditure. Statutory provisions in contexts such as the present must be construed with regard to the realities of the way organisations dependent on public funding must operate.

124    At [168] below, I have reproduced an extract from Lander J’s reasons for judgment in Watson. His Honour there focuses on the connection between the exercise of power and the purpose of the statute conferring the power. Applying that approach to the legislative scheme of the Land Rights Act, I do not consider that the scheme intended to accommodate Ministerial changes of mind once a direction had been given and communicated. The power is conferred, in my opinion, on an office holder, not on an individual. Whether or not a power to recall a decision might be implied if there had been a material mistake of fact (such as to eligibility) need not be determined. On any view, the scheme does not intend for grants to be dependent on whether a new incumbent in a Ministerial position agrees, or does not agree, with a view taken by a previous incumbent. Rather, the scheme reposes in the Minister holding office at the time a power to decide whether to give a direction for the payment of funds, to whom the funds are to be given, how much is to be paid, and on what conditions the payment is to be made. Thereafter, the duty is cast on the Commonwealth to pay over the funds, subject to any conditions imposed. The conditions power is, in my opinion, the mechanism intended by the parliament to give further control over the allocation of funding to organisations and individuals under s 64(4).

125    The current approach to statutory interpretation involves courts assuming that those drafting legislation, and parliaments which enact it, are familiar with the general principles of statutory construction, and that courts will take statutory language as they find it, read with the purpose and context in which it appears. If a particular outcome or effect is intended by a statutory provision, the language used by Parliament should make that clear: see, generally, Taylor v Owners – Strata Plan No 11564H [2014] HCA 9; 253 CLR 531 at [35]-[40]; Marshall v Watson [1972] HCA 27; 124 CLR 640 at 649 (Stephen J); and see Dallikavak v Minister for Immigration and Ethnic Affairs (1985) 9 FCR 98 at 103 (Northrop and Pincus JJ), which I discuss below. Indeed, as I explain later in my reasons, in most cases dealing with reversals of an exercise of power, or revocation of a decision made in the exercise of power, the Court has either found, or not found, an implication about the authority of the repository in the statute itself.

126    In my opinion it is by no means clear that Parliament intended to confer a power on an incumbent Minister to revoke a direction to the Commonwealth Treasury to pay monies under s 64(4) which had been made by a previous Minister, or to do so based on a change of the incumbent’s mind. It is also by no means clear Parliament intended that reversal could occur before the monies were in fact paid (a matter apparently to large degree in the non-transparent discretion of the executive), but not afterwards. The better view, in relation to this scheme, is that Parliament intended that once a direction was given, it would be complied with and the performance of the statutory power and function would be completed, not subject to recall because of a change of mind, government, or policy.

127    The appellants’ submissions contended that to find a contrary intention (as the primary judge did) presents “an immediate oddity”. It was contended that “presumably”, s 33(1) applied to s 64(4) of the Land Rights Act “at least to some extent: so far as directions concern different recipients, at different times, the Minister may exercise the power in s 64(4) as the occasion arises”. It was then contended:

How, then, is a contrary intention manifested so as to displace s 33(1) of the Acts Interpretation Act partially, when it falls to be applied in relation to a recipient already the subject of an unfulfilled direction?

(Emphasis in original.)

128    Putting the rather confrontational language of the submission to one side, I see no such difficulty. That is because s 33(1) is unnecessary to reach the conclusion that the power in s 64(4) may be exercised from time to time in relation to different recipients, the funding of whose activities will be subject to different directions. So much is evident from the scheme of the Land Rights Act itself. Section 33(1) has no role to play.

Does the presence of a power to impose conditions affect this conclusion?

129    Sections 64(5A) and (5B) provide:

(5A)    The Minister may, by notice in writing, specify conditions on which a payment of an amount to a person under subsection (4) (including by way of a loan) is made. The notice is not a legislative instrument.

(5B)    If a condition on which a payment of an amount to a person under subsection (4) is made is breached, the Minister, on behalf of the Commonwealth, may:

(a)    if the payment is by way of a loan—recover so much of the loan as has not been repaid, and any accrued interest that has not been paid, as a debt in a court of competent jurisdiction; or

(b)    in any other case—recover the whole or a part of the amount as a debt in a court of competent jurisdiction.

130    I agree with Perram J (see [89]) that the text of s 64(7) (“amounts … must be paid or applied in accordance with the direction”) indicates that any additional requirements conditioning the payment of funds are given effect.

131    Section 64(5B) deals with what is to occur if conditions attached to the direction are breached after funds have been paid. The recovery powers are discretionary rather than mandatory and obviously there is room for careful consideration of particular circumstances of breach of conditions and what might be done to remedy such breaches. The scheme is attended by some flexibility in that regard.

132    However, I do not agree with Perram J (see [89]) that there is any necessary difficulty with the conditions provisions (especially s 64(5A)) if the directions power in s 64(4) does not extend to revocation, reversal or undoing of a direction to pay funds where the direction has been given and communicated to the recipient. The example given at [89] may or may not be apt. The nature of the conditions power is a power to specify (by notice in writing) “conditions on which a payment … is made”. The provision does not say “is to be made”. Consistently with the text of s 64(4) and s 64(7), the provision operates on the assumption that once a direction is given the statutory duty to pay the amount specified will be performed. That is why s 64(5B) addresses the recovery of amounts on breach of conditions. I do not consider the conditions power in s 64(5A) necessarily extends to imposing preconditions to a payment being made. Conditions powers are often construed as subject to limitations in terms of the kinds of conditions which may be imposed: see, eg, Western Australian Planning Commission v Temwood Holdings Pty Ltd [2004] HCA 63; 221 CLR 30 at [56]-[57] (McHugh J), [93] (Gummow and Hayne JJ); Mison v Randwick Municipal Council (1991) 23 NSWLR 734 at 737 (Priestley JA); Winn v Director General of National Parks and Wildlife [2001] NSWCA 17; 130 LGERA 508 at [13]-[14] (Spigelman CJ). That matter need not be determined in this appeal, however the constructional issue demonstrates why the answer to the example given by Perram J may not be straightforward.

Conclusion on contrary intention

133    For those reasons, I have concluded that Pt VI of the Land Rights Act, read in the context of that Act as a whole, evinces a contrary intention to the implication of a power in s 64(4) to revoke, reverse or undo the giving of a direction for the payment of funds from the Account.

134    In my opinion, the appeal should be dismissed on that basis, and it is not strictly necessary to determine the scope of s 33(1). However, given the matter was fully argued, I consider it is also appropriate to explain why, contrary to the assumption I have made above, I do not accept that the scope of s 33(1) is as the appellants contend, and as Perram J has found it to be.

Section 33(1) does not have the effect contended for

135    I do not consider s 33(1) effects a general implication into all statutory powers in federal legislation of a power to reverse or undo an exercise of power or to revoke a decision made in the exercise of a statutory power. Nor do I accept (as the argument must run) that it effects an implication of the same kind in relation to the performance of functions and duties. Specifically, I do not accept that s 33(1) has the effect of implying into the power in s 64(4) of the Land Rights Act a power to revoke, reverse or undo a direction given by a Minister and communicated to the person affected, in relation to the same subject matter.

136    In my opinion, s 33(1) has a more limited operation: its purpose is to make clear that the presumptive position is that powers, functions and duties, as conferred by the enabling statute, can be exercised or performed repeatedly – “as occasion requires” – rather than only once. This removes the need for the words “from time to time” to be expressed each time a power, function or duty is conferred by statute. Usually, the repeated exercise or performance will be in respect of different persons, different subject matter, and on different occasions. In some circumstances (and some of the authorities below illustrate this) there may be a repeated exercise or performance in relation to the same person or subject matter, but that will be a less common implication.

137    As I have noted above, in the modern statutory context, the proper construction of statutory powers and functions in terms of the manner and occasion for their exercise and performance will generally be ascertainable without resort to s 33(1) at all.

138    My reasons for reaching that conclusion fall into several categories. First, there are historical and circumstantial considerations, including the history and purpose of the term “functus officio”, which assist in understanding the purpose of s 33(1). Second, there is the text of s 33(1) itself. Third, there does not appear to be any binding authority the ratio of which requires me to accept that s 33(1) has the operation for which the appellants contend. Further, I do not consider there is any particular support for the appellants’ contentions to be derived from other authorities, each of which depends on its particular circumstances and statutory context. Where those authorities contain obiter dicta supporting such an approach, I respectfully disagree with those dicta.

Historical and circumstantial matters

139    At a general level, the utility, or need, for s 33(1) is more limited than it may have been in earlier times. As the history to which the parties’ supplementary submissions adverted demonstrates, s 33(1) and its predecessors and equivalents have their origins in 19th century English law. Statute law was not the overwhelming feature of the legal landscape it is today. In the 21st century, Australian statute law is made by way of detailed and complex prescription, drafted with awareness of refined principles of interpretation, in particular the emphasis on text, context and purpose as governing concepts in statutory construction, rendering statutes more comprehensive and self-contained than may have been the case in the past.

140    The historical extrinsic material and the historical cases, to which the Court was taken, focus on avoidance of the conclusion that a repository was functus officio. In contrast to Perram J’s reasons at [32], in my respectful opinion, the history of s 33(1) (such as it is) and its equivalents does inform the question of the scope of the implication arising from such provisions. The remarks of Sir Courtenay Ilbert, also reproduced at [27] of Perram J’s reasons, provide one of the clearest indications of the mischief that predecessor provisions to s 33(1) were intended to remedy:

The effect of this sub-section is to make the insertion of the words ‘from time to time’ usually unnecessary. Those words were formerly inserted for the purpose of removing the application of the doctrine that a statutory power is exhausted by its first exercise unless its repetition is expressly authorized. The presumption on this point is now reversed. If it is intended that a statutory power should not be exercised recurrently this should be stated.

141    The full titles of the Acts in which the predecessors of s 33(1) appear – the “Act for Shortening Acts of the Legislative Council”, being the Acts Shortening Act 1852 (NSW) (see s 10); the “Act for Further Shortening the Language Used in Acts of the Governor General in Council, and for Other Purposes”, being the General Clauses Act 1887 (India) (see s 5); and the “Act for Consolidating Enactments relating to the Construction of Acts of Parliament and for Further Shortening the Language Used in Acts of Parliament”, being the Interpretation Act 1889 (UK) (see s 32) – refer to the shortening of legislation as well as to its interpretation. The same is true of the full title of the Commonwealth Acts Interpretation Act today (“An Act for the Interpretation of Acts of Parliament and for Shortening their Language”). Many historical examples can be found of legislative provisions enacted prior to the Interpretation Act 1889 (UK) conferring powers to be exercised “from time to time”: see, eg, the power to set up watchboxes in s 22 of the Metropolitan Police Act 1829 (UK), the power to set the routes of public processions conferred by s 21 of the Town Police Clauses Act 1847 (UK), and the power of a company to issue new shares in s 11 of the Companies Clauses Act 1863 (UK). A large number of further examples could be given. Some of those powers might have included an implied power of revocation, or an implied power of variation; others might not have done. The point is that the mischief to which provisions such as s 33(1) and its predecessors are directed was twofold. First, to shorten statutory language. Second, in so shortening, to avoid the inconvenient possibility that a power conferred without the words “from time to time” might be taken to be exhausted on its first exercise. If still necessary, the objective of avoiding that mischief is achieved by a general implication (subject to a contrary intention) of the words “from time to time as occasion requires” into the exercise of powers and the performance of functions. It is not achieved by a general implication of a power to reverse, revoke or undo the exercise of a power or the performance of a function or duty.

142    The point was well made in the supplementary submissions of the respondent, which I accept:

… the legislative history confirms that in all of the jurisdictions that have been canvassed the provision which became s. 33(1) of the [Acts Interpretation Act] serves a particular, quite confined, purpose. It is to avoid the operation of the inconvenient common law doctrine to the effect that a power conferred by statute was exhausted by its first exercise … The material compiled by the appellants therefore confirms that the object was accurately described by Gummow J in Kurtovic at 211. The purpose of s. 33(1) is simply to ensure that (subject to a contrary intention) any statutory power – whatever its scope, purpose or character – can be exercised repeatedly, from time to time. There is a symmetry between the mischief and the remedy. The common law doctrine did not alter the scope, purpose or character of the statutory power in question. It merely had the effect that the power was exhausted after its first operation. The statutory remedy simply removes the operation of that doctrine, and otherwise leaves the statutory power as it finds it. It was not necessary, in order to remedy the mischief created by the common law doctrine, to confer a general power to revoke.

The functus officio concept

143    A brief consideration of what the functus officio concept involved, and how and in what circumstances it operated, supports the view I take.

144    The origins of the concept are explored in M Akehurst, ‘Revocation of Administrative Decisions’ [1982] PL 613 and in DM Pretorius, ‘The Origins of the Functus Officio Doctrine, with Specific Reference to Its Application in Administrative Law’ (2005) South African LJ 832. Professor Campbell also deals with it, less directly, in her article to which I have referred at [120] above. As I have noted, Professor Campbell’s division is between perfected and unperfected decisions: the former tending to suggest, in her opinion, a lack of any implied power to alter or reverse them, absent jurisdictional error. An earlier exploration of the topic can be found in G Ganz, ‘Estoppel and Res Judicata in Administrative Law’ [1965] PL 245. In the Australian context, various aspects of the problem are also discussed in M Allars, ‘Perfected Judgments and Inherently Angelical Administrative Decisions’ (2001) 21 Aust Bar Rev 50.

145    Pretorius refers (at pp 858-859) to the decision of Barons v Luscombe (1835) 3 Ad & E 589; 111 ER 537, as does Akehurst (at pp 613, 616 and 624). In it, two magistrates approved parish accounts showing £31 owing by the outgoing parish overseer. The succeeding overseers sought and obtained from two other magistrates a warrant authorising and requiring the new overseers to recover the sum from the previous overseer. Doubt then arose about the correctness of the sum, and the magistrates who had issued the warrant purported to rescind it. Lord Denman CJ held that, having approved the accounts according to law, the first two magistrates had “no right to rescind their decision” and therefore the second two magistrates “could not afterwards entitle themselves to rescind [the warrant] by saying, ‘We now doubt as to the state of the accounts’” (at Ad & E 594; ER 539). Similarly, Littledale J held that, having approved the accounts, the original magistrates “are functi officio, and cannot open the accounts” even if they had made a mistake because “[t]heir power was at an end” (at Ad & E 594-595; ER 539). Patteson J agreed in the result, but apparently on the narrower basis that the second two magistrates did not have power to reopen the decision of the first two magistrates and therefore could not rescind the warrant in order to do so (at Ad & E 595-596; ER 540). Williams J also decided that the second two magistrates had no power to suspend the warrant pending a review of the accounts which they did not have power to undertake, holding that the uncertainty about the correctness of the accounts did not amount to a mistake of fact (at Ad & E 596; ER 540).

146    Two matters of significance arise from Barons v Luscombe. First, it concerned, in modern terms, an exercise of judicial power. It may well be – and Pretorius’s analysis by reference to Roman and Roman-Dutch law tends to support this (see pp 842-843, and see pp 837-841 in relation to early examples of the concept being applied in relation to arbitration awards) – that the concept was first applied in judicial or quasi-judicial settings, in order to produce and reinforce finality in determinations settling private disputes between parties. The concept promoted certainty and finality in those settings. No doubt it did so in part by explaining that upon pronouncement of a decision or order, the judge, magistrate or arbitrator’s powers were exhausted. How it came to be translated into administrative law might well, as Pretorius suggests, only be explicable by some transferred understanding of the limits of a repository’s powers being akin to those conferred on a judicial officer, so as (with similar policy aims in mind) to produce finality and certainty. The second matter of significance from Barons v Luscombe is that the Court was concerned with a situation of alleged, or potential, mistake: there, mistake of fact. Reversal or revocation was proposed to be necessary because of mistake or error: that is, to ensure correction of error. In modern times, that will either be addressed by appeal or review (including judicial review) or by a principle such as that enunciated in Bhardwaj.

147    Other early cases referring to “functus officio” confirm that it is not a “doctrine” with any substantive legal content, so much as a description or label applied to given situations for the purpose of explaining why a power, function or duty could not be exercised or performed again.

148    One of the earliest uses of the term in a decided case appears to be Hay v Earl of Eglington (1630) Mor 637, a decision of the Scottish Court of Session in which it was held that a decision of an arbiter (Laird Caprington) made more than a year and a day after the matter was submitted to him was not void by reason of the Laird being “functus officio suo”, or, as the report states, “when he had no power”. Instead, the Court held that agreements to submit a matter to binding arbitration of the kind made by the parties “expire not after expiring of year and day … but that the same last and endure, and the judge may make his declaration at any time, so long as the party survives”.

149    The term, and variations of it, was used in several other decisions of the Court of Session in the 17th century before being taken up by the English courts in the 18th century. In Havers v Havers (1740) Barn C 22; 27 ER 538, the Ecclesiastical Court had appointed the impecunious grandmother of an infant as administrator of the infant’s estate, on receipt of an “administration bond” of £200 given by sureties. The bond was security for the proper administration of the estate. “[A]fter this Grant of Administration was perfected” (in the words of the law report), the grandmother gave a second surety of £600 for the same purpose. The Lord Chancellor held that, given the size of the estate (£600) and the grandmother’s impecuniosity, an administration bond of £200 was negligently small. Indeed the Lord Chancellor had this to say (at Barn C 23; ER 539):

The Administration during the Minority of the Infant, which has been granted in the present Case, his Lordship said, has been granted in the most careless, slovenly and scandalous Manner, that he ever saw.

150    The Lord Chancellor continued (at Barn C 24; 539-540):

This Negligence of taking so small a Security has been endeavoured however to be supplied by the taking of another Bond in a larger Sum. But by the taking of the first Bond, and the granting the Administration under it, the Surrogate [the Ecclesiastical Court] had functus Officio; and therefore his Lordship said, he did not see what Authority he had to take this second Security.

The Lord Chancellor therefore appointed a receiver to the estate.

151    The Encyclopaedic Australian Legal Dictionary (LexisNexis, 2017) defines “functus officio” as meaning “having discharged one’s duty; having completed one’s term of office; having ceased to hold some public appointment; having performed the authorised act and being unable to go back to it a second time”. According to the Oxford Latin Dictionary (Clarendon, 2000), “functus” is the past participle of “fungor”, meaning “to perform, discharge, [or] execute”, while “officio”, from “officium”, means “office”.

152    Therefore, the term is descriptive of an exhaustion of power. That may be because of an exercise of power or performance of function or duty, but it may also be (as Hay v Earl of Eglington demonstrates) because the permissible period in which to exercise a power, or to perform a function or duty, has expired.

153    In the early authorities, the term is usually used with reference to a court or quasi-judicial body. It was also used in a private law context. For example, in Kirby v Duke of Marlborough (1813) 2 M & S 18; 105 ER 289, Lord Ellenborough said that, where A provided a guarantee in respect of a particular amount to be advanced to B by C, “when the advance was made to that amount, the guarantie became functus officio, and was not a continuing guarantie” securing further advances.

154    A private law usage, with reference to revocation, is found in Barry v Heider [1914] HCA 79; 19 CLR 197. There, Isaacs J said of the purported withdrawal of a caveat by the caveator’s solicitor (at 220-221):

Now, unless it is to be laid down as a matter of law that because a solicitor is instructed to lodge a caveat he has also implied authority to withdraw it and to make a representation that purchase money has been paid when in fact it has not, then the withdrawal is not [the caveator’s] and the representation is not his.

I know of no principle or authority for such a rule of law. The authority to lodge a caveat is complete in itself, and is exhausted when the caveat is lodged. … The person authorized to lodge the caveat is then functus officio.

155    All of these cases illustrate that the word “doctrine” is a misnomer. Rather “functus officio” describes a conclusion on the legal authority of a person: that an exercise of power, or a performance of a function or duty, is complete and the person has no power left to exercise, or no function or duty left to perform. The conclusion is reached by close examination of the particular circumstances, and the nature of the power, function or duty in question.

156    The term continues to have a specific connotation, particularly with reference to criminal courts. For example, in NH v Director of Public Prosecutions (SA) [2016] HCA 33; 90 ALJR 978, a trial judge who had formally recorded acquittals and convictions following a criminal trial was described as “functus officio” in respect of the matter: at [93] (Nettle and Gordon JJ, concurring). Nevertheless, the term is still being used as a description of a legal conclusion, rather than a legal test or “doctrine”. This is apparent from the reasons of the plurality in NH at [30] (French CJ, Kiefel and Bell JJ):

It was common ground in the Full Court that the judgments of acquittal and conviction, reflecting the trial court’s acceptance of the verdicts of the jury, were perfected, at the latest, when each of the Reports was signed. That is a matter of substance and not just of form. Five Justices of this Court cautioned in Burrell v The Queen that the use of the term “perfected” must not be seen as giving form and procedure precedence over substance and principle. The real question is “What is to mark the point at which a court concludes its consideration of a controversy?” Their Honours said:

Identifying the formal recording of the order of a superior court of record as the point at which that court’s power to reconsider the matter is at an end provides a readily ascertainable and easily applied criterion. But more than that, identifying the formal recording of the order as the watershed both marks the end of the litigation in that court, and provides conclusive certainty about what was the end result in that court.

(Footnotes omitted.)

157    In an administrative law context, as a matter of substance, the question is similar: at what point is the exercise of power (or performance of function or duty) concluded? Each statutory setting will provide the answer.

158    Finally, although not expressly incorporating the language of functus officio, I consider the case of Collector of Customs (NSW) v Brian Lawlor Automotive Pty Ltd (1979) 2 ALD 1 (Bowen CJ, Smithers and Deane JJ), which was raised in oral argument, to be of some relevance. The case dealt with the purported revocation of a warehouse licence granted by the Collector of Customs under the Customs Act 1901 (Cth), and came to the Federal Court on an appeal under s 44 of the Administrative Appeals Tribunal Act 1975 (Cth). The Collector had purported to revoke the applicant’s licence on the basis it was not a “fit and proper person” to hold such a licence. There was no express power of revocation in the Customs Act. The Administrative Appeals Tribunal had overturned the revocation, and the appeal to the Federal Court was dismissed. Whether there was power to revoke the warehouse licence was a central question. The Collector relied on an implied power, but also on s 33(3) of the Acts Interpretation Act. So far as the report of the case indicates, neither the Collector, nor any party, nor any Judge, mentioned s 33(1).

159    What is apparent is that the Court considered, unremarkably, that the existence of a power of revocation depended on the statutory scheme. For example, Bowen CJ said (at 2):

Dealing with the first matter, it is true that various types of licence, particularly those relating to land, have been held to be in their nature revocable but in these cases a paramount right exists in the licensor which he may assert by revoking the licence. Cowell v Rosehill Racecourse Co Ltd (1937) 56 CLR 605; cf Winter Garden Theatre (London) Ltd v Millennium Productions Ltd [1948] AC 173. In such cases there is inherent in the licence a power of revocation. However, the position appears to be different in the case of a statutory licence such as that under the Customs Act 1901 (Com). There is nothing inherent in the nature of such a statutory licence which must lead to the conclusion that it is in its nature revocable: see Banks v Transport Regulation Board (Vic) (1968) 119 CLR 222 at 231.

160    See Smithers J at 13-14, to similar effect. At 14-15, Smithers J continued (having held s 33(3) of the Acts Interpretation Act did not apply to the licence):

Rights and duties of the officers of the Customs Department and the licensee are specified in Div 1 of Pt V of the Customs Act. It would seem therefore that the question whether the licence is revocable or not is to be found in the express terms of the Division and such implications as properly arise therefrom or from the nature of the transaction effectuated when a licence is granted thereunder.

When the Minister is minded to grant a licence to a person in respect of premises pursuant to Div 1 of Pt V of the Customs Act no doubt he could grant the licence by a simple statement to the effect that a licence under Div 1 of Pt V of the Act was thereby granted in respect of specified premises. The conditions of the licence would have to be gathered from the terms of the Act and the Regulations and from the nature of a licence. On the other hand it would seem, that subject to certain provisions of the Customs Act the terms and conditions on which a licence may be granted are within the discretion of the Minister or his delegate. He may decline to issue a licence to any particular person or to grant it in respect of any particular premises or he may issue a licence for a particular term or a term determinable in certain events. The kind of provision he could not introduce into a licence would be one specifying licence fees in amounts or payable otherwise than in accordance with the Regulations as s 80 of the Act provides that those fees and the times for their payment are to be determined by regulation. However, there is no doubt that the Minister could grant a licence reserving to himself a right to determine it on any grounds including failure on the part of the licensee to observe certain specified standards of conduct or to comply with the statutory duties imposed by Div 1 of Pt V of the Customs Act. And in the present case the Collector did grant the licence subject to certain conditions. If in such a case the Collector considers that there has been a breach of a condition entitling him to revoke the licence and decides to do so the questions that arise are whether the alleged breach had in fact occurred and whether on the proper interpretation of the licence there was a right to revoke for that breach.

(Emphasis added.)

161    After a detailed examination of the statutory scheme, Smithers J concluded (at 18-19) that an implied power to revoke a warehouse licence might arise if “the very existence of the licence is incompatible with the purposes of the Customs Act”, but that that was not this case and therefore the revocation was beyond power.

162    The observations of Bowen CJ and Smithers J provide some support for the proposition that where a statute provides for a power to confer a benefit with conditions – as subs (4) and (5) of s 64 of the Land Rights Act do – and specifies what will happen upon breach of a condition – as s 64(5B) does – the courts will be slow (despite reference to the Acts Interpretation Act) to find an implied power of revocation and thus alter the content of the power conferred.

Textual considerations

163    Implying a power of revocation, reversal or undoing may alter the content of a statutory power considerably. The language of s 33(1) does not suggest Parliament intended such an alteration, by implication, across all federal statutes. The same observation may be made concerning similarly worded equivalents of s 33(1) in state laws. An interpretation provision would not ordinarily be construed as having the effect of altering the substantive content of a statutory provision to which it applied, thus changing the nature of the power, function or duty conferred.

164    There are (relevantly for present purposes) two limbs to s 33(1). The first concerns the exercise of a power; the second concerns the performance of a function or duty. It is unclear how the revocation/reversal implication is said to work with the second limb of s 33(1). If s 33(1) has the wider scope for which the appellants contend, then it would need to operate to imply into the conferral of a function and a duty an authority, or ability, to revoke, reverse, or undo the performance of that function or duty. Theoretically, one may be able to undo the performance of a duty but it is difficult to see how that would operate in a real statutory context (cases such as Bhardwaj aside). Indeed, it would tend against the very notion of a duty to suggest that it might be reversed or undone. One can, however, perform a function, or discharge a duty, on more than one occasion in relation to the same, or different, subject matter, “as occasion requires”. These textual issues, together with the distinct use of “may” and “must” in s 33(1), support the construction I prefer.

165    The use of “may” for powers and “must” for functions and duties reflects the drafter’s consciousness that many powers conferred by statute are discretionary in nature, while functions and duties are generally conferred in terms that require the repository to perform the task specified by the statute. That explains the differential use of “may” and “must” in s 33(1). The key point is that the operative phrase “from time to time as occasion requires” applies equally to all three kinds of authority and must be given a meaning which is capable of applying sensibly to all three.

166    I do not share Perram J’s approach (see [41]) to the use of the word “may” in s 33(1). While I accept that where, in the same provision, an imperative and a permissive verb are each used, a different meaning for each is likely to be intended, in the case of s 33(1) the meaning emerges from the fact that s 33(1) deals with functions and duties as well as powers. To take s 64(4) as an example, Perram J’s view about the role of the word “may” in s 33(1) would have s 33(1) as the source of the Minister’s authority to decide not to make a direction to pay funds to the respondent. I respectfully disagree. The source of the Minister’s authority in those circumstances remains s 64(4) itself. The words “such other amounts as the Minister directs” encompass, or contemplate, that the Minister may make no direction. If the Minister makes no direction, then the duty to pay funds in s 64(4) is not triggered. If the Minister makes a direction concerning recipients A and B, but not C, the duty is enlivened only as to payments of funds to A and B. The Minister has chosen not to exercise her or his power to make a direction in relation to C. None of this is the work of the word “may” in s 33(1), even if (contrary to my opinion) it is necessary to apply it to s 64(4) in order to establish that the Minister may give more than one direction concerning the payment of funds. The entitlement of the Minister to decide not to give a direction, or to give a direction as to some but not all persons in respect of whom she or he has considered whether funds should be paid, springs from the terms and purpose of s 64(4) itself.

167    For this reason, I take a different view to that expressed by Perram J at [45]-[46] of his Honour’s reasons. As I have explained, in my opinion, the scheme of the Act makes it clear that the Minister is authorised to give more than one direction that funds be paid to identified recipients. The nature of the power itself is one which is susceptible of multiple exercises. The implication for which s 33(1) provides is unnecessary in this particular scheme. So too, the nature of the power suggests the Minister may choose not to direct that funds be paid to a particular person. If it were necessary for me to decide, which it is not, I would not favour any distinction being drawn within the operation of s 33(1) based on concepts of retroactive effect. How that distinction could be applied to a provision such s 64(4) illustrates the difficulty. Where the making of a direction results in the imposition of a duty on another (here, the Commonwealth), a revocation must have retroactive effect to avoid the consequence that a statutory duty remained unperformed.

168    A further matter supporting the approach I prefer is that on each occasion it is employed, s 33(1) concerns an implication into a particular statutory context where a governing constructional consideration is the purpose for which the power or function was conferred. As Lander J said in Watson at [117]:

The effect of s 33(1) is that unless the Act creating the power indicates a contrary intention the power may be exercised as often as is necessary to fulfil the purpose for which the power was conferred. Section 33(1) does not widen the power given by s 501 of the Act or allow it to be exercised more often than the purpose of that section permits.

(Emphasis added.)

169    Thus, the language of s 33(1), including the key phrase “from time to time as occasion requires”, concentrates on when a power may be exercised, or a function/duty performed, and does not otherwise alter the character of the power or function. A “decision” may be the manifestation of an exercise of power, but it should be borne in mind that the implication in s 33(1) operates on the exercise of power itself. Once a decision has been made, or an action taken, in the exercise of a power, s 33(1) operates to enable the repository to exercise the power again to make another decision or to take another action. But it does not operate to confer on the repository a power to revisit and reverse the original exercise of power, or cancel the effect of the original action. To assimilate a choice regarding whether to exercise a power with the outcome of that choice – namely, a decision or action taken pursuant to the power, if the repository chooses to exercise it – is to move, in my respectful opinion, too far beyond the text and purpose of s 33(1). Section 33(1) creates a presumption that the power may be exercised again, not that the original choice about how to exercise it can be remade.

170    In contrast, s 33(3) provides expressly for the implication of a power of revocation into a category of statutory powers which involve the making of an “instrument”. Where a power falls within that category, s 33(3) requires that it be construed as including a power, inter alia, to revoke the instrument, subject to a contrary intention. The word “instrument” in s 33(3) has been given a relatively broad construction (see Flaherty v Secretary, Department of Health and Ageing [2010] FCAFC 67; 184 FCR 564 at [61]) but it is clear s 33(1) applies to a broader category of powers, functions and duties, of which the category of powers in s 33(3) is a subset. That s 33(3) provides for a specific presumption in respect of that particular category tells against reading the same presumption into s 33(1), which applies to all powers, functions and duties. It should also be noted that Senior Counsel for the appellants expressly disavowed any reliance on s 33(3) in this case, in my opinion correctly, on the basis that the Minister’s direction under s 64(4) is not an “instrument”.

171    I accept the respondent’s submission that the legislative history in the UK and Australia demonstrates that the relevant contrast in language between like provisions to s 33(1) and s 33(3) has been maintained: see the examples given by the respondent, namely ss 10 and 11 of the Acts Shortening Act 1852 (NSW) and s 32 of the Interpretation Act 1889 (UK). Despite the longevity of these provisions, and their connection, terms such as “revocation” have not been introduced into s 33(1).

172    The word “revocation” may be inapt to describe what occurs where a power is exercised from time to time. The visa regime under the Migration Act 1958 (Cth) – at issue in Watson – provides an example. A person’s visa may be cancelled in an exercise of power to decide whether to do so. However, the power cannot be exercised again to decide not to cancel the visa because there is no visa to consider as the subject matter of the power. As the Court pointed out in Watson, in substance what would occur in this situation would be the grant of a new visa, for which the Act provides in other, detailed, ways: at [7] (Dowsett J), [23]-[24] (Hely J), [141]-[142] (Lander J). If the power is exercised for the first time and no cancellation results, then there remains a visa upon which the power can be exercised again as the subject matter of the power – in my opinion, at least if there are new facts or circumstances. That is what s 33(1) means when it speaks of an exercise from time to time. It is not an implication directed at undoing what has been done before. Thus (and bearing in mind the dangers of extrapolating to hypothetical legislative schemes), if a person applies for a licence and is refused, the person may re-apply even if there are no specific re-application provisions, on the basis of an implication from s 33(1). The provision operates to make clear that the power to grant or refuse a licence is a recurring power. If the licence is granted, and assuming s 33(3) did not apply, I would not construe s 33(1) as implying an additional power to reconsider whether the person should retain the licence, or have it revoked or cancelled. That is a power of a different nature and scope to a power to grant a licence. If s 33(3) did not apply, then the legislative scheme would probably need expressly to empower reversal, undoing or revocation. Most modern statutory schemes do just that. It is conceivable that a power in a given scheme could be construed, by reference to the text, context and purpose of the legislative scheme in which it is found, to contain an implied power to rescind, reverse or modify a previous exercise of power in relation to the same person or subject matter. If so, the implication arises from a particular statutory context, not from s 33(1).

Previous authorities

173    I accept the submissions made by the respondent (with the examples given at [12] to [16] of the supplementary submissions) that:

None of the authorities cited by the appellants involve ratio in support of the appellants’ broad proposition, namely that s. 12(1) of the Interpretation Act 1978 (UK) (and therefore, by extension, s. 33(1) of the AIA) confers a general power to revoke previous exercises of power, irrespective of the character of the power. Most of the UK authorities cited by the appellants … involve nothing more than illustration of the proposition that in the case of certain types of statutory powers, a further exercise of power will have the effect of revoking or varying the consequences of a previous exercise of the same power. … In some cases, the character of the power is such that a further exercise of the power necessarily has the effect of altering the state of affairs arising from a previous exercise of power, including in such a way that the previous exercise of power may be described as being effectively varied or revoked. The point is that this is because of the nature of the power that is being exercised from time to time, not from any independent additional power conferred by s. 12(1) of the Interpretation Act 1978 (UK).

174    I also accept the respondent’s supplementary written submissions at [18] to [23] regarding the Canadian decision of Comeau’s Sea Foods Ltd v Canada [1997] 1 SCR 12, relied on by the Minister and which considers the similarly worded Canadian equivalent to s 33(1). The respondent is correct, in my opinion, to submit that the decision contains no reasoning in explaining why the equivalent to s 33(1) should be read in all cases as altering “the basic characteristics of the statutory power in respect of which it is operating, by conferring an additional power to revoke previous decisions”. I accept the respondent’s submission that the Supreme Court’s perception of the need for flexibility in responding to changing conditions in relation to lobster stocks influenced its approach. However, in my opinion more importantly, Comeau concerned a two-stage statutory power: a power to authorise the issue of licences, which the Court referred to as a “devolution of authority” by the Minister; and a power to grant licences in accordance with that authorisation, exercisable by other public officials (at [29]). The Supreme Court was concerned with a revocation of the authorisation to issue, as the statutory scheme expressly circumscribed the circumstances in which an issued licence could be revoked (as one would expect). There is no relevant analogy with the scheme under the Land Rights Act, and with the scheme in Pt VI in particular. There is but one exercise of power, which then compels the payment of money to an identified recipient. Comeau should in my opinion be confined to its facts, and the legislative scheme in issue.

175    I turn now to consider in more detail the authorities which in my opinion best expose the competing approaches to the scope of s 33(1), but also illustrate the point made in the respondent’s submissions.

176    Lawrie v Lees (1881) 7 App Cas 19 concerned the meaning of the phrase “from time to time” in an Act (s 3 of Sir H. Meux’s Settled Estate Act 1863 (UK)) dealing with the ongoing administration of the business affairs of Sir Henry Meux. Section 3 conferred a discretionary power, in the nature of a power of attorney, on the Lord Chancellor to:

from time to time, so long as the said Sir Henry Meux and his estate shall continue to be subject to the jurisdiction in lunacy, order or direct to be done, permitted and suffered with respect to the business …

and otherwise in relation to the said business and the affairs and concerns thereof, all and whatsoever Sir Henry Meux, if of sound mind, might do, permit or suffer or concur in doing, permitting or suffering with respect to the same.

177    It was in this context that Lord Penzance made the well-known comments which Perram J reproduces at [50] of his Honour’s reasons, and on which the appellants rely.

178    As Perram J notes at [51], there is no reasoning exposed that leads to the (obiter) meaning Lord Penzance proffers. I respectfully agree. To this, I add the observation that his Lordship’s comments must be read in their context. That context was a statutory power directed expressly to the administration of the (considerable) business affairs of one person. The decisions to be taken in that context would often be of a contractual character, as the very dispute in Lawrie v Lees makes clear. If not contractual, then the decisions would encompass the multiplicity of decisions to be made in the running of a business: or indeed, of many businesses. In that context it is hardly surprising Lord Penzance spoke of the phrase “from time to time” involving an ability to “add something to”, “take something from”, or “reverse” a decision altogether. His Lordship was speaking of making decisions, on behalf of an incapacitated individual, about running businesses, including entering into contracts and other kinds of transactions. His Lordship was not describing, nor purporting to describe, what the phrase “from time to time” might mean in other settings, and in particular in the exercise of public power.

179    Finally, the language used by Lord Penzance at 29-30, in the passage cited in Perram J’s reasons, should be carefully noted. His Lordship spoke of the phrase “from time to time” having a protective function in respect of the repository of the power. Protective in what sense? In the sense of protecting the repository from the risk “of having completely discharged his duty when he has once acted”. Two observations should be made about this characterisation. First, as the context in Lawrie v Lees illustrates, his Lordship was concerned to “protect” the repository of the power so that the repository could perform what was on any view a beneficial activity for the incompetent person. His Lordship’s remarks about the scope of the phrase “from time to time” were made in the context of amplifying, for a beneficial purpose, the repository’s powers. Second, his Lordship was concerned to ensure that a person given a fiduciary function, indeed, a fiduciary duty, had sufficient power to perform that fiduciary duty in circumstances of ongoing need. That is, his Lordship was concerned with the performance of a duty, and a duty with complex and ongoing content.

180    In contrast is the case of Re 56 Denton Road, Twickenham [1953] Ch 51. I consider this case to be of more relevance. It concerned an assessment of compensation by the War Damage Commission, payable to the plaintiff in relation to damage to her house during the Second World War. The Commission initially informed the plaintiff her house was classified as a total loss, then reviewed that classification to a partial loss. The compensation payable to the plaintiff for “costs of works” on a partial loss basis was almost four times the compensation payable for total loss. Having notified the plaintiff it would proceed on a partial loss basis, and the plaintiff having agreed, the Commission then subsequently informed the plaintiff it had reviewed the classification and was reverting to a “total loss” compensation basis. I pause here to note that in Professor Campbell’s description, the decision and the notification (and acceptance by the plaintiff) would, it seems, mean the partial loss classification was a “perfected” administrative decision. Whether or not that conceptual terminology is ultimately of assistance in understanding how one decides whether a power or function is subject to an implied authority to revoke or recall it, noting that the decision was “perfected” or complete demonstrates some of the other conceptual underpinnings at work: finality and certainty.

181    On the plaintiff’s challenge to the Commission’s “review” and second decision to reclassify her property as a total loss, Vaisey J held the Commission had no power to revoke or recall its first decision. His Lordship at 56 described the defendants’ argument in the following terms:

And they say that they have a perfect right to change their minds as their policy changes or as fresh evidence comes to light or further advice is given to them by their technical advisers, and that unless a determination of theirs has been acted upon in some way so as to raise a case of estoppel, they can go on altering it, as, and as often as, occasion arises.

182    There must be, in this, some resonance with the appellants’ contentions in the present case. Although here we are concerned with a direction to pay, and then a revocation of that direction, in principle the appellants’ arguments must be the same if the situation were reversed. A direction not to grant funds to a particular applicant could be changed by a new Minister into a direction to pay. And if the Minister changed again before the funds were in fact paid, on the appellants’ contentions the third Minister could make a lawful decision to refuse to pay monies. On the appellants’ argument, it could be based on a broad policy change, but it could also be based simply on the view of the Minister about who was the most meritorious recipient. If there was sufficient administrative delay in funds being paid, there could be several Ministerial changes of mind and that would, on the appellants’ case, be lawful. I do not accept Parliament intended the scheme to operate in that way.

183    Vaisey J did not accept the Commission’s arguments. Rather, his Lordship accepted, and repeated in his reasons, the propositions put on behalf of the plaintiff (at 56-57):

… where Parliament confers upon a body such as the War Damage Commission the duty of deciding or determining any question, the deciding or determining of which affects the rights of the subject, such decision or determination made and communicated in terms which are not expressly preliminary or provisional is final and conclusive, and cannot in the absence of express statutory power or the consent of the person or persons affected be altered or withdrawn by that body. I accept that proposition as well-founded, and applicable to the present case. It is, I think, supported by Livingstone v. Westminster Corporation and Robertson v. Minister of Pensions.

(Footnotes omitted.)

184    No argument based on the equivalent of s 33(1) (being s 32 of the Interpretation Act 1889 (UK)) was apparently put to the court, despite Lawrie v Lees.

185    Livingstone v Westminster Corporation [1904] 2 KB 109, upon which Vaisey J relied, concerned a claim for compensation by the plaintiff, upon the abolition of the parish office held by him (a position as surveyor) under the Local Government Act 1899 (UK). The council, which made the decision to abolish the office, told the plaintiff to make his compensation claim based on an aggregation of his salaries and emoluments for a period of five years from the date of the abolition of his office. The plaintiff did this, his claim was approved, and the compensation (an annual allowance) was paid. Subsequently the council’s auditor formed the view there were several flaws (the details of which are immaterial for present purposes) in the allowance the council had determined to pay. The council purported to rescind the resolution approving the payment to the plaintiff and replace it with a lesser amount. The plaintiff’s argument was that the council had no power to rescind the first resolution. Buckley J (at 120) agreed with the plaintiff’s argument even if, as the council submitted, on review the first compensation allowance arrived at was factually incorrect, and excessive:

So soon as the resolution of August, 1901, was validly passed, there arose an obligation, which under s. 120, sub-s. 6, of the Act of 1888 was a specialty debt, enforceable as if the council had entered into a bond to pay the 5181. 12s. 4d. I find no power in the council to go back and relieve themselves from an obligation thus rendered binding upon them.

186    Buckley J held the plaintiff was entitled to recover the amount of compensation payable under the first resolution. Again, Lawrie v Lees, and the kinds of arguments which flow from that decision, do not appear to have been put to the Court.

187    A contrasting case is R v District Council of Berri (1984) 36 SASR 404, a decision of the Full Court of the Supreme Court of South Australia. It was a planning case, concerning the grant of planning permission upon a particular condition, which condition the Council later sought to revoke. The plaintiff contended the Council had no power to modify, vary or extend the conditions of its initial consent. At first instance, White J agreed, finding that the condition “once fixed and communicated, is irrevocable” (at 412). On appeal, the Full Court disagreed, and found there was an implied power to revoke the impugned condition. Legoe J drew the implication from the scheme and purposes of the legislation itself (the Planning and Development Act 1966-1967 (SA)) and did not rely on the equivalent to s 33(1): namely s 37 of the Acts Interpretation Act 1915-1981 (SA), as it was then called. Cox J (with whom King CJ agreed) also did not need to consider the operation of that provision.

188    Cox J did not agree with the Council’s submission that it had some generalised implied power of revocation. His Honour said (at 416):

Sub-section (5) of s. 41 provides (to take a typical case) that the owner of land that is subject to interim development control may not erect a building without the consent in writing of his local council. That implies a decision by the council and its written notification to the owner. If the council has merely resolved to grant its consent but the necessary notification has not yet been given to the owner, the formal act required by sub-s. (5) will not have occurred and it will be open to the council to rescind its earlier decision. Ex Parte Renouf; Shanahan v. Strathfield Municipal Council. However, once the council has notified the owner of its consent in writing, it has fulfilled its statutory function. It may afterwards regret its decision, it may possibly by following its rescission procedures formally disavow what it has done, but it cannot effectively claim that it did not give the owner the consent in writing that s. 41 requires.

(Footnotes omitted.)

189    At 57, Cox J referred to Re 56 Denton Road as an example of this principle in a non-planning context. However, his Honour did find that a particular provision, dealing with revocation of conditions, empowered the Council to do what it had done: see 418-420. In other words, the source for the implication of power was the Act in question, not the interpretation statute.

190    Southlink Pty Ltd v WorkCover Corporation of South Australia [2009] SASC 175; 104 SASR 172 again reveals different judicial approaches to re-exercises, or further exercises, of power, not all involving the application of equivalent provisions to s 33(1). The case concerned a challenge to an increase by WorkCover in a supplementary levy payable by the applicant to WorkCover. The facts and numerous judicial review grounds were complex, but the material circumstance was that there was a decision by a delegate that a supplementary levy found to be payable by Southlink should be struck at a particular figure, and then a decision by a different delegate that the supplementary levy should be increased. Both decisions were challenged on various grounds. Relevantly, the majority (Bleby J and White J) held that the relevant empowering provision (s 67 of the Workers Rehabilitation and Compensation Act 1986 (SA)) did not authorise an increase in the levy during its term. Bleby J generally agreed on this matter with White J who found (at [219]) that s 67, read with other empowering provisions, authorised the imposition of a single supplementary levy for a specified period which could not be increased during the period of its application. White J held the second delegate had “ignore[d] this limitation” on the power under s 67.

191    Kourakis J dissented, and it was his Honour who employed the South Australian equivalent of s 33(1) (s 37 of the Acts Interpretation Act 1915 (SA)) in determining the scope of the power in s 67 of the South Australian Act. His Honour referred to Parkes Rural Distributions Pty Ltd v Glasson (1986) 7 NSWLR 332, which I deal with below, and to Day v Hunkin [1938] SASR 121, a South Australian decision to which I also refer below. His Honour concluded (at [346]) that the nature and subject matter of the power in s 67 meant Parliament did not intend for it to be exercised only once. Of the state of authorities on a provision such as s 37, and in particular the decision in Day v Hunkin, Kourakis J said (at [343]):

Section 37 therefore plainly allows a power to be exercised for the purpose of revoking an earlier determination and making a further determination which disadvantages a person whose rights are affected by those determinations.

192    With respect, I do not consider the authorities stand for a proposition as broad as that.

193    Day v Hunkin dealt with a challenge by Mr Day to a reduction of his salary after he had accepted and commenced, in 1930, a five-year term as Chairman of the South Australian Pastoral Board. It was an appointment made by the South Australian Cabinet and approved by the Governor in Council, duly notified in the South Australian Government Gazette. When notified of his appointment, the plaintiff was told he was appointed “at the salary of £1,000 per annum, first division professional section minimum £1,000, maximum £1,000 for a term of five years dating from and including the 1st February 1930” (see Day v Hunkin [1937] SASR 453 at 430). The Governor in Council approved the recommendation of the Public Service Board that the plaintiff’s salary should be £1,000 per annum, however seven months later the plaintiff’s salary was reduced to £900 per annum and later was reduced to £800. The plaintiff contended there was no power to reduce his salary while he held office as he was appointed for a fixed term. The defendant relied on s 37 of the Acts Interpretation Act 1915 (SA) as implying a power in the provision of the Public Service Act 1916 (SA) under which the plaintiff’s salary was fixed (s 27) to alter, including to reduce, the plaintiff’s salary. Section 27 relevantly provided that the salary of an officer of the public service was to be “determined by the Governor on the recommendation of the Board”.

194    The trial judge held that s 37 did not apply to empower the salary reduction. After doubting the applicability of s 37 for reasons not presently relevant, Cleland J said (at 459):

Even if sec. 37 of The Acts Interpretation Act applies in its fullest extent it only operates as an extension of a power in order to avoid the implication that when once it is exercised it is necessarily exhausted.

195    Cleland J’s decision was overturned on appeal: [1938] SASR 121. It may be worthwhile recalling this decision was given during the Great Depression. Angas Parsons J (at 125) referred to the exigencies likely to arise which may require alterations to the salaries of public servants and did not consider the situation of those appointed for a fixed term as a distinguishing feature. His Honour said (at 125):

There is always the possibility of an increase or decrease in the cost of living, and in the ordinary course of things these alterations may have to be reflected in the salaries of the public service. Those and other circumstances make it reasonable, perhaps essential, that the salaries of public servants should be altered as occasion may demand. Moreover, there are times of stress as regards the State’s finance. There is nothing in sec. 27 which excludes the power of the Board to recommend a variation of the salary, or to prevent the Governor from determining from time to time the salaries to be paid to particular officers or classes of officers. The nature of these powers requires an interpretation which is flexible and practicable. No other construction is possible if the powers are to be effectively exercised. In The Australian Boot Trade Employees’ Federation v. Whybrow & Co. and Others, (1910) 11 C.L.R. 311, Isaacs J. said, at p. 337:—“It is true that the grant of a power carries with it the grant of all proper means not expressly prohibited to effectuate the power itself. …[”]

196    Angas Parsons J did not rely on s 37 of the Acts Interpretation Act 1915 (SA).

197    Napier J did rely on s 37 for the asserted implication (at 127), but did so by reference to general considerations, and to the text and purpose of the statute concerned. Napier J said at 127-128:

In my opinion, there are cogent arguments based on reason and convenience in favour of this construction. The power to increase the salary may be absolutely necessary if the State is to retain the services of the officer, and, on the other hand, the power to effect the reduction of salaries may be necessary in order to adjust expenditure and make ends meet. It may be possible to read the proviso to sec. 27 (1) as applicable to a power which is exercisable once and for all, but I think that the proviso is much more consistent with the intention to subject these salaries to periodical revision. It is difficult to see why the salaries under sub-sec. 1 should be unchangeable when the salaries under sub-sec. 2 are clearly subject to revision. Sec. 14 of the Public Service Act 1916 shews that when Parliament intended the salary to be fixed, at or before the commencement of a term of office, it has said so in language which puts the matter beyond doubt.

198    Richards J, noting at 132 that it was “a matter of common knowledge that those general reductions were made in consequence of the financial stringency and economic depression under which the State was suffering”, also found s 37 removed any doubt that the Governor had power to increase or reduce the salary of a person holding office as Mr Day did (at 131-132).

199    The absence of a contractual relationship between Mr Day and the State appeared to weigh heavily in the Court’s assessment. However, in terms of the construction of the power, it is not insignificant that this case concerned a “determination” of an ongoing entitlement – that is, to a salary, rather than a “once off” payment. The main point to be taken from Day v Hunkin, in my opinion, is that it was not a revocation or cancellation case at all: it was an alteration case. The import of a power to determine a person’s salary being exercisable from “time to time” so as to alter that salary is far removed from the statutory power in issue in this appeal. Further, as the judgment noted, the general economic situation facing South Australia weighed on the Court’s approach to construction.

200    The High Court dismissed an appeal from the decision of the Full Court, holding that, on its proper construction, the power was one which could be exercised from time to time to vary a salary and that it was unnecessary to rely upon s 37 to reach that conclusion: Day v Hunkin [1938] HCA 47; 61 CLR 65. Latham CJ stated that, while s 37 applied to the power, it was “not necessary to call [it] in aid” to conclude that the power could be re-exercised (at 77). McTiernan J expressed the same view (at 80). Dixon J did not refer to s 37 at all, saying it was “evident” the power was not confined in the way Mr Day contended (at 77). The Court’s decision demonstrates that careful regard should be paid to the particular statutory scheme when determining whether a power may be re-exercised to vary the effect of a previous decision, and that provisions such as s 33(1) (or, there, s 37 of the South Australian legislation) may have a limited role to play in that process.

201    Parkes concerned the issuing of consecutive certificates under the Petroleum Products Subsidy Act 1965 (NSW), in relation to a scheme granting financial assistance to distributors to facilitate the sale of petroleum products to be sold in rural areas at prices comparable to those in capital cities. The certificates related to a situation where an authorised officer had certified, as he was empowered to do, that an amount paid to a distributor in excess of what the officer was satisfied should have been paid was repayable to the State of New South Wales.

202    The relevant power was expressed in the following terms:

Where an authorized officer is satisfied that an amount paid to a person under this Act … was not payable to that person or exceeded the amount that was payable to that person, he may give a certificate in writing that the amount paid or the amount of the excess, as the case may be, is repayable by that person to the State.

203    The first certificate (issued by the first defendant in February 1982) certified that $152,317.70 was repayable by the plaintiff to the State. The second document (issued in November 1984 by the second defendant) certified that $134,065.27 was repayable by the plaintiff. The first certificate was found by the trial judge to have been issued in circumstances which involved a denial of natural justice and so was held to be invalid. The consequential argument put by Sir Maurice Byers QC on behalf of the plaintiff was the first certificate remained valid and operative until found to have been issued in circumstances involving a denial of natural justice; that therefore it was legally impossible to have two certificates in relation to the same distributor and the same grant payments imposing two different and inconsistent repayment obligations, meaning the second certificate was a nullity. The trial judge rejected the argument and held that what had, in law and fact, occurred was the first certificate had been withdrawn and replaced by the second certificate. This finding relied on s 32 of the Interpretation Act 1897 (NSW). This was the provision that replaced s 10 of the Acts Shortening Act 1852 (NSW) being, it would appear, one of the earliest examples of this kind of implication provision, and predating any such provision existing in English statute law. The trial judge’s opinion was upheld on appeal. Glass JA (with whom Samuels and Priestley JJA agreed) stated at 335-336:

It has been held that a power exercisable from time to time may be so exercised as to add to, subtract from or reverse the result of the previous exercise: Lawrie v Lees (1881) 7 App Cas 19 at 29. In other words the donee of such a power is never functus officio. Accordingly the existence of the first certificate does not render void a second certificate issued in lieu of it.

204    This statement is one of the more absolute renditions of what Lawrie v Lees is understood to stand for. To say that a repository is “never” functus officio would, if applied as absolutely as it is expressed, render administrative decision-making subject to endless uncertainty and without finality, including (as in Parkes) where the legal rights and liabilities of individuals are concerned. There is no suggestion that provisions such as s 32 of the New South Wales legislation, or s 33(1) in the Commonwealth legislation, are intended to obliterate finality in administrative decision-making, unless a contrary intention is found in the legislation.

205    I respectfully agree with the doubts expressed by Nettle JA (as his Honour then was) in Kabourakis v Medical Practitioners Board of Victoria [2006] VSCA 301; 25 VAR 449 at [83]-[85] about these aspects of the Court of Appeal’s reasoning, especially in its reliance on Lawrie v Lees.

206    One way to mitigate the absoluteness of Glass JA’s statement is to emphasise the importance of determining whether a contrary intention exists. Glass JA did not refer to this aspect of s 32 of the New South Wales legislation. However, in my respectful opinion that would still be giving provisions such as s 32 (and s 33(1) in the current appeal) too much work to do. Another is to read Glass JA’s statement as subject to an implicit qualification concerning the reason that a power may be capable of re-exercise. Here, for example, it would appear the second certificate was issued because the second defendant, discharging his duty under the relevant provision, had reviewed the liability imposed by the first certificate and considered it not in accordance with the legislative scheme. The report of the case does not disclose whether the issue of these certificates was subject to any form of review or appeal, whether in the empowering legislation or elsewhere. Rather it appears from the summary given by Glass JA at 335 that the certificate created a debt due to the State. Again, it would seem that without some ameliorating power, the purpose of the scheme would not be advanced by a construction which did not accommodate review and revision of a certificate. A distributor would otherwise incur a debt that did not accord with its legal liability. On the facts there was a two-year gap between the two certificates, and the report does not reveal whether there were changed facts and circumstances leading to the second certificate.

207    Further, what occurred in this case is not properly characterised as a revocation. A revocation, in relation to a power such as the one at issue in Parkes, would have occurred if there had been a reversal of the first certificate and a finding that no amount was due to the State. Instead, the same power was exercised in the same way and for the same purpose, and what occurred was that a different debt was substituted. That is an alteration or an amendment, but it is not a revocation. It was not, to adopt the description used by Nettle JA in Kabourakis at [86], an “annihilation” of the previous exercise of power. In the present circumstances, the Minister’s second decision under s 64(4) is aptly described as an annihilation of the first direction, and on the evidence intended to be so.

208    In my respectful opinion, Glass JA’s statements about “reversal” of a decision are obiter in his Honour’s reasoning in Parkes, and should not in any event be applied, in their broad and absolute tenor, to all statutory powers (or functions) where there is no express revocation or reversal power. To do so would unseat the certainty and finality of administrative decision-making where that decision-making affects the rights, liabilities or interests of persons affected by the power.

209    The need for certainty and finality was a central theme of the Full Court’s decision in Watson, in a situation concerning, as I have said, visa cancellation under the Migration Act. Mr Watson’s visa was cancelled by the Minister in March 2001, on the basis that Mr Watson did not pass the ‘character test’ in the Migration Act by reason of his criminal record. The Minister did so exercising the personal power reposed in the Minister by s 501(2) of the Migration Act. However, after the Minister’s decision, the High Court handed down its decision in Re Patterson; Ex Parte Taylor [2001] HCA 51; 207 CLR 391, in which the Court found that the Migration Act did not authorise the Minister to deport a British citizen who had arrived in Australia prior to 1973. Mr Watson was thus one of the individuals who, temporarily, enjoyed the benefit of the decision in Patterson before it was overruled in Shaw v Minister for Immigration and Multicultural Affairs [2003] HCA 72; 218 CLR 28.

210    As a result of Patterson, the Department informed Mr Watson that it would “no longer be proceeding with your removal” after he served his sentence of imprisonment. The reprieve lasted about three years, then in March 2004, Mr Watson (who was by this time in prison for a different offence) was informed that, because of the decision in Shaw,

the decision of the Minister to cancel your visa under section 501 of the Migration Act 1958 stands and you are liable to be detained under s 189 of the Migration Act 1958 and removed from Australia.

211    The primary judge held that the Minister had revoked the cancellation decision in 2001 (that is, as a result of Patterson), and this had been communicated to Mr Watson, restoring a visa to him. Therefore, the primary judge held, the Department (and the Minister) were incorrect to advise Mr Watson in March 2004 that the cancellation decision “stands”, because it had, in fact and law, been revoked. In reaching that conclusion, the primary judge held that a revocation power should be implied into the Minister’s cancellation power by reason of s 33(1) of the Acts Interpretation Act.

212    Ironically, it was the Minister in Watson who, on appeal, contended s 33(1) did not apply to the cancellation power so as to imply a power to revoke a cancellation decision. The Full Court agreed. All three judges gave slightly different reasons for reaching that conclusion. Dowsett J appeared to assume, it would appear on the basis of the decision in Immigration and Ethnic Affairs, Minister for v Kurtovic (1990) 21 FCR 193, but without any express consideration, that s 33(1) extended to implying a power to revoke a decision. His Honour however located a contrary intention in the broader scheme of the Migration Act, and in particular the detailed regime relating to the grant of visas, which his Honour considered could be undermined by implying a power to revoke a cancellation decision, thus essentially granting a person a new visa. Hely J found (at [15]) that, unlike other cancellation provisions which referred to revocation in certain circumstances:

There is, however, nothing in the Act which implies that a visa cancellation decision under s 501, and its consequences, are subject to a later Ministerial change of mind.

213    Hely J did consider (at [18]-[23]) the scope of s 33(1). His Honour noted (at [19]) the view of Branson J in Dutton v Republic of South Africa [1999] FCA 498; 162 ALR 625 at 636 that, unlike s 33(3), s 33(1) does not refer to the withdrawal or cancellation of the exercise of a power. His Honour then noted the different approach taken by Gummow J in Kurtovic. However, Hely J stated (at [22]):

It is true that in the passage quoted, Gummow J speaks in terms of a power to make a second deportation order “so as to revoke or revive” a deportation order, and refers to a reconsideration of decisions previously made. But these observations were obiter, as in Kurtovic it was not a second exercise of the s 12 power which revoked the first. As a result of the exercise of the s 20 power to revoke, the s 12 power remained available for re-exercise at the later date.

214    At [23], his Honour continued, in my opinion effectively disagreeing with the view taken by Gummow J in Kurtovic and adopting a position closer to that of Branson J in Dutton:

In my view, s 33(1) of the Interpretation Act (assuming it applies) does not have the effect that once there is a valid exercise of the s 501(2) power to cancel a visa, the visa can be effectively restored to the former holder by a second exercise of that power. Whilst the Minister has a discretion whether or not to exercise the s 501(2) power, the power in question is simply a power to cancel a visa which, if validly exercised, results in the former holder acquiring the status of an unlawful non-citizen with the consequences prescribed by the Act. There is no occasion for the re-exercise of the power once it has been validly exercised.

(Emphasis added.)

215    Thus, his Honour did not see s 33(1) as having the generalised broader effect described by Gummow J in Kurtovic, nor for that matter attributed to the like New South Wales provision in Parkes. Nor is this description of the effect of s 33(1) consistent with the broader and more absolute proposition stated by Kourakis J in Southlink. That Hely J was speaking of the scope of s 33(1) is apparent from the next paragraph, where his Honour said:

If s 33(1) did have that effect, I would conclude that the Act discloses a contrary intention so far as the power to cancel a visa is concerned.

(Emphasis added.)

216    Lander J, like Hely J, was in my opinion more cautious about the scope of s 33(1) than Gummow J had been in Kurtovic. His Honour carefully examined the scheme of the Migration Act in relation to cancellation and like powers, and where Parliament had expressly conferred a power of revocation.

217    Referring to three previous single judge decisions (Sloane v Minister for Immigration, Local Government and Ethnic Affairs (1992) 37 FCR 429, Burgess v Minister for Immigration and Multicultural Affairs [2000] FCA 926; 101 FCR 58, and VQAR v Minister for Immigration and Multicultural and Indigenous Affairs [2003] FCA 900), his Honour concluded (at [138]) the provisions of the Act to which he had referred evinced a contrary intention to the proposition that the cancellation power in s 501(2) could be exercised from time to time as the occasion required. However, his Honour had this to say (at [117]) about the scope of s 33(1) , with which I respectfully agree:

The effect of s 33 is that unless the Act creating the power indicates a contrary intention the power may be exercised as often as is necessary to fulfil the purpose for which the power was conferred. Section 33(1) does not widen the power given by s 501 of the Act or allow it to be exercised more often than the purpose of that section permits.

(Emphasis added.)

218    In Dutton, s 33(1) was employed in argument by Mr Dutton in relation to a notice issued by the Attorney-General under s 16(1) of the Extradition Act 1988 (Cth). It was contended that, in dealing with Mr Dutton’s request that the s 16 notice be withdrawn, the Attorney denied him procedural fairness. The Attorney contended there was no power to withdraw a s 16 notice. Mr Dutton contended the implication in s 33(1) applied. Branson J agreed with the Attorney’s submissions. Noting that there was no express power to withdraw, and noting the four-stage scheme of the Act and the dependency of subsequent stages on the existence of a s 16 notice, Branson J said (at [32]) that s 33(1) does not refer to the withdrawal or cancellation of an exercise of power. It appears her Honour’s attention was not drawn to Lawrie v Lees, nor to Kurtovic, nor to cases such as Parkes. Nevertheless, her Honour’s approach is in my respectful opinion correct, and absence from the text of s 33(1) of express words relating to revocation, cancellation, withdrawal or reversal of an exercise of power tells against the extended scope given to s 33(1) in obiter statements in cases such as Kurtovic and Parkes.

219    Minister for Immigration and Multicultural Affairs v Bhardwaj [2000] FCA 789; 99 FCR 251 involved a denial of procedural fairness to an applicant who, at the last moment, sought to inform the Immigration Review Tribunal that he wished to appear on his review and present evidence and arguments. His review concerned a cancellation of his student visa. The applicant’s communication was not brought to the attention of the Tribunal member constituting the Tribunal, who went ahead and made a decision affirming the delegate’s cancellation decision, including making the following finding, accepted by all parties to be incorrect:

On 15 September there was no appearance by the Applicant, his immigration agent nor was there any indication or advice in relation to the failure to attend.

220    The applicant’s migration agent made representations which resulted in the Tribunal allocating a hearing date to the applicant, notwithstanding it had made and communicated its decision on the review. At the hearing, evidence was adduced to satisfy the Tribunal that the delegate’s decision should be set aside, and that is what the Tribunal did. Relying on the decision of French J (as his Honour then was) in Sloane, the Minister contended the Tribunal was functus officio once it had made the first decision and that the second decision should be set aside. The Minister also relied on the decision of Goldberg J in Jayasinghe v Minister for Immigration and Ethnic Affairs (1997) 76 FCR 301, in which his Honour held a tribunal was functus officio once it made its decision on a review and that it could not reconsider that decision. The Full Court held neither decision was applicable because the Tribunal was not “reconsidering” its decision: rather, it had always intended to give the applicant a chance to attend a hearing which, due to an oversight, did not occur. Beaumont and Carr JJ referred to and relied on s 33(1) and Gummow J’s reasons in Kurtovic, and at [46] said that:

within the meaning of s 33(1) of the Acts Interpretation Act, the “occasion” “requires” that both the power and the duty of the Tribunal to review the matter should be exercised and performed at the time when the Tribunal was made aware that, in purportedly making its September decision, it had proceeded, in ignorance, upon the false assumption that the respondent had elected not to ask for an oral hearing.

221    Lehane J dissented. His Honour said at [58]:

Generally, s 33(1) of the Acts Interpretation Act will apply in relation to a statutory power or duty. But the statutory context may reveal a contrary intention. In my opinion, the present statutory context does so. It is one which, plainly, places a high value on certainty. There are strict time limits, detailed provisions governing the conduct of review proceedings and precise requirements as to the way in which the Tribunal is to record its decision and the reasons for it and is to notify and publish its decisions. There is then a limited form of judicial review, for which application may be made only within a time limit of twenty-eight days which cannot be extended. It would, in my view, be incongruous with that scheme for the Tribunal to have, in relation to a particular application for review, a power from time to time, as occasion requires, to make (and revoke) decisions. The considerations referred to by French J in Sloane v Minister for Immigration, Local Government and Ethnic Affairs [1992] FCA 414; (1992) 37 FCR 429 at 442-444 and by Goldberg J in Jayasinghe v Minister for Immigration and Ethnic Affairs [1997] FCA 551; (1997) 76 FCR 301 at 311-315 are, to my mind, compelling. The recent majority decision of the Full Court in Semunigus v Minister for Immigration and Multicultural Affairs [2000] FCA 240 is consistent with the approach taken in those cases. Certainly it is true that neither French J nor Goldberg J referred to s 33(1) of the Acts Interpretation Act and the observations of Goldberg J were directed to the question whether the Refugee Review Tribunal, having made a decision, was functus officio. But, as Gummow J makes clear in Kurtovic and as is implicit in the judgment in Jayasinghe, to say that a decision-maker is functus officio is simply to say that the decision-maker, having made a decision, does not have power to revoke it and make a new decision. And considerations, arising from a statutory context, which lead to a conclusion that, having made a decision, a decision-maker is functus officio substantially correspond with those which indicate an intention that s 33(1) is not to apply.

222    To that last view might be added the proposition that, as the approach in some of the authorities demonstrates, the text, context and purpose of a particular statute indicate, without the need to resort to s 33(1), whether or not there can be repeated exercises of a power, or repeated performances of a function, and in what circumstances, as well as whether there is power to revoke, reverse or undo an exercise of power or performance of a function.

223    Bhardwaj was decided on a different basis by the High Court, turning on whether a decision-maker can recall a decision affected by jurisdictional error (as the High Court described the denial of procedural fairness in this case): see Minister for Immigration and Multicultural Affairs v Bhardwaj [2002] HCA 11; 209 CLR 597 at [43]-[44], [53] (Gaudron and Gummow JJ, expressly disclaiming reliance on s 33(1)), [63] (McHugh J agreeing), [149]-[156] (Hayne J, to similar effect).

224    This brings me to what I consider to be the two most relevant prior authorities: Kurtovic and the Victorian Court of Appeal decision in Kabourakis. In summary, I prefer the approach taken by Nettle JA in Kabourakis, although on close analysis, in my opinion it may well be the case that Gummow J in Kurtovic was not making an absolute statement about s 33(1) (even as obiter) in the way some subsequent authorities have perceived what his Honour there said.

225    Mr Kurtovic had been convicted of an offence (manslaughter) which exposed him to deportation under the Migration Act and indeed such an order was made. However, after representations, the Minister decided to revoke the deportation order and instead “warned” Mr Kurtovic that any further conviction rendering him liable to deportation would “weigh heavily” against him when the Minister reconsidered his case. This revocation occurred under s 20 of the Migration Act (as it then stood), which contained an express power of revocation. At this time, Mr Kurtovic was seeking parole in relation to his manslaughter sentence. His parole application, and appeal, were refused, however in doing so the Supreme Court of New South Wales made some observations about his offending which caused the Minister to reconsider whether Mr Kurtovic should be deported. Acting on a Departmental recommendation, the Minister decided to issue a further deportation order. It was common ground there was no material change in circumstances – in the sense that Mr Kurtovic had not committed any further offences, or engaged in any further conduct than existed at the time the first deportation order was revoked. It should be noted that the relevant deportation power (at that time, in s 12 of the Migration Act) was expressed in a way which, provided the preconditions were satisfied (as to non-citizenship and conviction for an offence within the precondition), the Minister had a broad discretionary power:

… the Minister may order the deportation of the person.

226    It should be recalled that the challenge by Mr Kurtovic involved several grounds, including an argument that the Minister was estopped from reconsidering whether to deport him. The estoppel argument had succeeded before the primary judge, but did not succeed in the Full Court. It is apparent from the Court’s reasons for judgment that the estoppel arguments are seen as connected to the resolution of the power argument that involved s 33(1). On the latter issue, and relevantly for present purposes, Mr Kurtovic contended that the power conferred by s 12 had been exhausted by the making of an initial deportation order (in July 1984) and its eventual revocation (pursuant to s 20, in November 1985), at least where there were no significant changes in circumstances relevant to the exercise of the power conferred by s 12 between the date of the revocation and the making of the second deportation order in January 1988. Thus, Mr Kurtovic contended s 12 provided no authority for the making of the second deportation order.

227    It can be seen that Kurtovic is not a case about revocation. It is a case about the authority to re-exercise the same statutory power. It is also a case about re-exercise of a power in the same way, or with the same outcome, as the first exercise. No party contested the revocation in November 1985 (because it was undertaken pursuant to s 20 of the Migration Act as it then stood, which expressly authorised revocation). The contest was over the exercise, for the second time, of the power in s 12 to make a deportation order in relation to the same individual and on the basis of, in substance, the same circumstances.

228    Neaves J found (at 195) that:

There is nothing in the language of s 12, or in the context in which it appears, to warrant the conclusion that Parliament intended that, absent any change in relevant circumstances, the power was to be regarded as spent upon its initial exercise in relation to a particular non-citizen. The power is, in my opinion, clearly one which may be exercised from time to time as occasion requires.

229    This statement by Neaves J would apply, for example, if the first exercise of power resulted in a determination that a person should not be deported. It is not a statement which expands, or alters, the nature or content of the power in s 12, in contrast to an implication that s 12 authorised revocation or reversal of a previous exercise of power. It proceeds, correctly in my respectful opinion, on the basis that the power to reverse, revoke or undo the making of a deportation order lies elsewhere in the scheme.

230    Ryan J also held the s 12 power was exercisable from time to time in relation to the same individual and the same circumstances. His Honour placed some weight on the nature of the power as a “public discretion” (at 200), and drew a link between the failure of the estoppel argument and the nature of the power as one exercisable from time to time. Ryan J agreed with the reasoning of Gummow J concerning why the Minister’s power under s 12 had not been exhausted by its previous exercise.

231    It is apparent from Gummow J’s reasons at 208 that his Honour also saw a connection between the rejection of the estoppel argument (assuming, contrary to his Honour’s opinion, that there had been a sufficiently clear representation to trigger an estoppel) and the proposition that the power in s 12 was exhausted by one exercise in relation to a particular individual:

Rather, in the present case, the respondent seeks to prevent the appellant from making a decision within the latter’s power which would have the effect of altering a previous intra vires decision. The respondent is then met with the objection that to allow an estoppel in this context would hinder or prevent further exercise of the statutory discretion.

232    And at 210:

The same limitation has been said by high authority to apply where there is no duty to act, but merely a discretionary power. This has been put on the basis that in a case of a discretion, there is a duty under the statute to exercise a free and unhindered discretion and an estoppel cannot be raised (any more than a contract might be relied upon) to prevent or hinder the exercise of the discretion; the point is that the legislature intends the discretion to be exercised on the basis of a proper understanding of what is required by the statute, and that the repository of the discretion is not to be held to a decision which mistakes or forecloses that understanding: see New South Wales Trotting Club Ltd v Municipality of Glebe (1937) 37 SR (NSW) 288; Cudgen Rutile (No 2) Pty Ltd v Chalk [1975] AC 520 at 533; Southend- on-Sea Corp v Hodgson (Wickford) Ltd [1962] 1 QB 416 at 423-425.

(Emphasis added.)

233    It is significant that, in this passage, Gummow J confines the principle to circumstances where the repository of the power has, on the first exercise of the power, “mistaken” or “foreclosed” a proper understanding of what is required of the repository in the exercise of the statutory discretion. That is, the policy in precluding an estoppel argument is to advance the lawful exercise of a discretionary power, even if through a re-exercise. This, of course, was the approach which later came to the forefront in Bhardwaj.

234    At 211, in the part of his Honour’s reasons dealing with the concept of functus officio and the role of s 33(1), Gummow J continued to articulate the connection, although acknowledging the primacy of the statute in determining whether re-exercise could occur, even if:

… in any given case, a discretionary power reposed by statute in the decision maker may, upon a proper construction, be of such a character that it is not exercisable from time to time and it will be spent by the taking of the steps or the making of the statements or representations in question, treating them as a substantive exercise of the power. The result is that when the decision maker attempts to resile from his earlier position, he is prevented from doing so not from any doctrine of estoppel but because his power to do so is spent and the proposed second decision would be ultra vires. The matter is one of interpretation of the statute conferring the particular power in issue.

235    This, Gummow J continued (at 211), explains the decisions in Livingstone and Re 56 Denton Road. His Honour sought to distinguish Livingstone and Re 56 Denton Road, by reference to the decision of Rootkin v Kent County Council [1981] 1 WLR 1186, an authority relied on by the appellants before us. In Rootkin, Eveleigh LJ said (at 1197):

Counsel has argued that the decision to pay the fare was irrevocable, even if mistaken, and he has relied upon the principle in Livingstone v. Westminster Corporation [1904] 2 K.B. 109. That principle of irrevocability may well be applicable when there is a power or a duty to decide questions affecting existing legal rights. In Livingstone v. Westminster Corporation itself the council were concerned to assess compensation for loss of office, to which compensation the plaintiff had a right under the Local Government Act 1899. Generally speaking, however, a discretionary power may be exercised from time to time unless a contrary intention appears.

236    Lawton LJ had expressed a similar view at 1195. The parallels with the concepts underlying the doctrine of functus officio (especially their judicial antecedents) are apparent. Irrevocability is linked to the performance of a function or duty which brings finality to the determination of legal rights. What is and is not for this purpose a “right” is rather more narrowly confined than modern authorities determining whether an exercise of power (including discretionary power) affects the “rights or interests” of a person: see Annetts v McCann [1990] HCA 57; 170 CLR 596. Perhaps one explanation for such a rigid approach to irrevocability of the kind expressed by Lord Eveleigh may stem from the more fluid approach to judicial power in the United Kingdom than that which exists in Australia. To say, as it was put in Rootkin, that an administrator was determining an existing legal right would be a problematic statement in Australian law: see R v Kirby; Ex Parte Boilermakers Society of Australia [1956] HCA 10; 94 CLR 254 at 281-282 (Dixon CJ, McTiernan, Fullagar and Kitto JJ); South Australia v Totani [2010] HCA 39; 242 CLR 1 at [220] (Hayne J), [444] (Kiefel J).

237    The difficulty is that there is no clear distinction between statutory duties or functions and discretionary powers of this kind in the authorities (even in the English authorities). That is unsurprising given the spectrum and complexity of statutes under which powers and functions are conferred. The deportation and removal cases are an illustration. Those exercises of power, discretionary in nature, affect legal rights (notably, the right to liberty) but yet have been found not to be irrevocable, Kurtovic being but one example. Others include Re Chan and Minister for Immigration and Ethnic Affairs (1977) 17 ALR 432 and Dallikavak v Minister for Immigration and Ethnic Affairs (1985) 9 FCR 98 (discussed below). Of course, as I have noted above, those exercises of power do not, and could not in Australian law, determine an existing legal right, but they do affect the rights and interests of a person – so much is well established. That is why the distinctions made in cases such as Rootkin are unsatisfactory in the Australian context.

238    Gummow J also referred to the Full Court decision in Dallikavak and the decision of Smithers J sitting as a presidential member of the Administrative Appeals Tribunal in Chan, both decisions taking a similar approach to the deportation power in s 12. Again, however, it is apparent that these cases were dealing with a re-exercise of the power in circumstances where a revocation was supported by an express power in s 20 of the Migration Act. See, for example, the way the majority of the Court in Dallikavak expressed the point (at 103-104 per Northrop and Pincus JJ):

We would add that if the Minister, having made a deportation order, subsequently becomes aware of circumstances which lead him to doubt the correctness of his order, or to come to the view that its correctness might need lengthy re-examination, he may revoke the order. If that happened, the person affected would cease to be a deportee under the Act but would remain a prohibited non-citizen and be subject to all the restrictions imposed by the Act on persons of that status. If on re-examination the Minister is of opinion that the order should again be made, no new grounds are necessary, in our view: see the decision of Smithers J. in Re Chan and Minister for Immigration and Ethnic Affairs (1977) 17 ALR 432 at 441-442. See also Acts Interpretation Act 1901 (Cth), s 33(1).

(Emphasis added.)

239    Thus, one of the difficulties, in my respectful opinion, with the analysis in Kurtovic and its reliance on earlier authorities is the failure to distinguish clearly between cases where what was in issue was the power of revocation itself, and cases where what was in issue was a re-exercise. In the former case, a focus on the effect of such an implication on the rights and interests of those affected is understandable. In the latter, an effect on the rights and interests of those affected is precisely what the statute contemplates by conferring the power in the first place.

240    Kurtovic was really a case concerning re-exercise rather than revocation. There is also the difficulty that, where Gummow J expresses his conclusion (at 218), aside from noting that once its preconditions are satisfied s 12 “does not purport to constrain the way in which the discretion is exercised”, there is no real analysis regarding why s 33(1) extends to implying a power to revoke or reverse a decision. His Honour is not, in my respectful opinion, in any event deciding that s 33(1) applies to expand the scope of s 12 in that way. Section 20 expressly provided for such a power.

241    There remains the decision of the Victorian Court of Appeal in Kabourakis to consider. Nettle JA gave the leading judgment, with Warren CJ and Chernov JA agreeing. The facts concerned a decision of a panel of a medical board constituted under Victorian legislation which determined that the appellant, a medical practitioner, had not engaged in unprofessional conduct. The complainant was dissatisfied and after an investigation the Victorian Ombudsman recommended the board re-open its investigation and reconsider the allegations against the appellant, which it did, by notifying the appellant there would be a further hearing. The appellant contended the medical board was functus officio and had no power to re-open its investigation, having already reached a conclusion on the allegations. The primary judge did not agree, but the Court of Appeal accepted the appellant’s contention. At [47], Nettle JA concluded:

… the effect of the Act properly understood is that a finding of the panel is final and binding except upon appeal or review.

242    His Honour continued, at [48], to deal with what he considered to be the situation in relation to the exercise of statutory administrative powers:

Self evidently, an administrative decision has only such force and effect as is given to it by the law pursuant to which it is made. As was pointed out in Bhardwaj, parliament may give an administrative decision whatever force it wishes. Consequently, as the judge made plain, the question in this case comes down to whether the statute manifests an intention to permit or prohibit reconsideration in the circumstances that have arisen. But, as was also said in Bhardwaj, as a rule a statutory tribunal cannot revisit its own decision simply because it has changed its mind or recognises that it has made an error within jurisdiction. More often than not, the requirements of good administration and the need for people affected directly or indirectly by decisions to know where they stand mean that finality is the paramount consideration, and the statutory scheme, including the conferring and limitation of rights of review on appeal, will be seen to evince an intention inconsistent with capacity for self correction of non-jurisdictional error. In the bulk of cases, logic and commonsense so much incline in favour of finality as to permit of no other conclusion.

(Footnotes omitted.)

243    Nettle JA indicated (at [49]) that Re 56 Denton Road was another example of this approach. His Honour then referred (at [50]) to a New South Wales decision in a similar vein:

More recently, in New South Wales, in Walter Construction Group v Fair Trading Administration Corporation [2005] NSWCA 65, Grove J was concerned with claims under a statutory building insurance scheme. His Honour said:

I do not construe that provision as vesting a power to make and unmake decisions infinitely. If power does not stretch to infinity, there must be in the circumstances of a particular case and “as occasion requires” a terminus. In this case it was reached with the communication of decision by the letter of 24 October 2002.

An application for leave to appeal was refused by the New South Wales Court of Appeal. Santow JA, with whom Sheller JA and Tobias JA agreed, referred with apparent approval to the passage set out above.

(Footnotes omitted.)

244    At [52]-[54], Nettle JA reviewed other authorities consistent with the approach he took. As the extracts in his Honour’s reasons demonstrate, none of those authorities depended solely on the existence of a contrary intention for the purpose of s 33(1) or like provisions. Rather, their principal focus was the nature and context of the statutory power in issue.

245    Nettle JA then turned to Kurtovic, which he described as of “little assistance”, firstly on the basis that it dealt with deportation, an area of administrative law his Honour described as “different to most others” (at [55]-[56]). With respect, I would disagree that deportation involves, at a level of administrative law principles applied in a statutory context, any substantive differences from other area of administrative law. Although, as his Honour observed, it is an area where statutory discretions are affected by changing policy considerations, there are many areas of public power where this is so. His Honour then also distinguished what he described as the “revenue cases”, again largely by reference to the subject matter and context of the statutory powers, including discretionary powers, under consideration in those cases.

246    The general approach taken by Nettle JA – that is, to focus on the statutory context and purpose in an assessment of how the question of re-exercise of powers should be approached – is, however, one with which I respectfully agree. In my opinion, as I explain elsewhere in these reasons, those considerations explain (at least, by reference to the evaluation of the judges involved) why there is such a variety of approaches and outcomes apparent in the authorities. At [64]-[80], this was the task his Honour undertook in relation to the legislative scheme before the Court in Kabourakis.

Conclusion on previous authorities

247    There is no decision binding on a single judge of this Court which is authority for the proposition that s 33(1), properly construed, has the effect of implying into all federal statutes, subject to the existence of a contrary intention, a power to reverse, undo or revoke an exercise of power or the performance of a function, where what is expressly conferred is authority to exercise a power or perform a function. The English authorities are also attended by another difficulty, given their different focus on what “irrevocability” implies about the nature of a power and the determination of pre-existing legal rights, which is ill-suited to Australian law.

248    Rather, what a review of the authorities demonstrates is that in most cases it is the text, context and purpose of the particular statutory scheme in issue which governs the construction of the scope of the authority of a repository of a statutory power or function. Few authorities, apart from Kabourakis and judgments such as those of Hely J and Lander J in Watson, grapple with the fact that extending a general implication such as that found in s 33(1) to include a power to reverse, revoke or undo an exercise of power or performance of a function changes in a substantive way the nature of the statutory power, rather than describing the manner and circumstances of its exercise. Some of the authorities where language of revocation is used are in fact cases about a re-exercise of a power, with a different outcome to the first exercise. Whether the statute authorises such a re-exercise has rarely turned on the terms of s 33(1).

Conclusion

249    It is often the case, as here, that reconciliation of the authorities seemingly dealing with similar issues, but across different statutory regimes and different periods of political and legal history, is an elusive exercise.

250    As Professor Campbell notes (at p 34), in 1994 Sir William Wade suggested (see HWR Wade and CF Forsyth, Administrative Law (Clarendon, 7th ed, 1994) p 261) that provisions such as s 33(1) give “a highly misleading view of the law where the power is a power to decide questions affecting legal rights”.

251    The power in s 64(4), when first exercised to direct that a particular sum of money be paid to a particular recipient, cannot be said to affect an existing legal right of the recipient. At the time of Minister Macklin’s decision, the respondent had no existing entitlement to be paid any further sum out of the Account for its ongoing activities in relation to Machado Joseph Disease (MJD). It had, in the past, been paid sums out of the Account for that work. On this occasion, it had applied for a new grant of funds. To that extent, Sir William’s observation is not directly applicable. However, once the direction was made and communicated by the Minister, s 64(4) created a duty in the Commonwealth to pay the amount directed to the respondent. Whether a correlative right to payment was simultaneously created in the respondent, as occurred in Livingstone and Parkes by virtue of the relevant statute creating a debt, is a nice question.

252    However, that question does not need to be determined in this appeal, nor does the related question whether the respondent could have obtained orders in the nature of mandamus enforcing the Commonwealth’s duty to pay. That is because the respondent’s interests – its capacity to perform the charitable work with Aboriginal people suffering from MJD and their families in the Northern Territory – were certainly affected by the exercise of power under s 64(4). Here, there was no question of, to use the language of Gummow J in Kurtovic, “mistaken” apprehension of what the statute required for the lawful exercise of power as justifying reversal or recalling of the direction. Certainty and finality are no less important underlying values in the context of Pt VI of the Land Rights Act than they are in other statutory schemes. The giving of a direction under s 64(4) was an exercise of power of the kind, as I have attempted to show by reference to the legislative scheme as a whole and to its component parts – which assumes finality of decision-making so that the Account can be regularly administered, and the purposes of the Land Rights Act advanced.

253    In her article Professor Campbell states (at p 34):

Certainly such provisions will seldom provide authority to revoke or vary decisions made in exercise of statutory powers to determine individual rights and liabilities.

254    The footnote to this proposition states “Cf, Kurtovic”. In my respectful opinion the “Cf” is an appropriate comment. Consideration of the history and purpose of s 33(1) suggests its operation should be confined to an implication that, subject to the presence of a contrary intention, powers may be exercised and functions performed on more than one occasion, in relation to different individuals and circumstances. Whether the implication extends to the repeated exercise of the same power or performance of the same function in relation to the same person or subject matter will ultimately depend on the particular statutory context, and the nature of the particular power or function concerned.

255    For those reasons the Minister was not authorised by the terms of s 64(4) to revoke the direction that funds in the sum of $10 million be paid to the respondent.

Associate:

Dated:    3 March 2017

REASONS FOR JUDGMENT

PERRY J:

256    I have had the benefit of reading the judgment of Mortimer J in draft. I agree with her Honour’s reasons and the orders that her Honour proposes.

Associate:

Dated: 3 March 2017