West v Commissioner of Taxation (NSW) [1937] HCA 26; (1937) 56 CLR 657 (3 June 1937)

HIGH COURT OF AUSTRALIA

West Appellant; and The Commissioner of Taxation (New South Wales) Respondent.

H C of A

On removal from the Supreme Court of New South Wales.

3 June 1937

Latham C.J., Rich, Starke, Dixon, Evatt and McTiernan JJ.

Roper, for the appellant.

Weston K.C. (with him Hooton), for the respondent.

Roper, in reply.

The following written judgments were delivered:—

1937, June 3

Latham C.J.

The question which arises upon this case stated is whether moneys received by a retired Federal public servant by way of pension under the Superannuation Act 1922-1934 are subject to taxation under the Special Income and Wages Tax (Management) Act 1933 of New South Wales.

In the case of salaries there is such an indication in the Commonwealth Salaries Act 1907. That Act necessarily involves the propostion that Commonwealth salaries are not to be taxable otherwise than in accordance with and subject to the limits imposed by the Act. But the Act does not deal with pensions paid to persons who are no longer officers of the Commonwealth.

As already stated, the Superannuation Act, in sec. 5 and sec. 80, deals expressly with the applicability of State laws to matters connected with Federal superannuation. It is a fair conclusion that in those provisions the Commonwealth Parliament dealt completely with the subject of excluding the application of State laws to pensions and pensioners. The provisions of the Financial Emergency Acts dealing with State taxation of salaries and pensions assume that they are already taxable by the States. The Act provides only for regulations prescribing a maximum limit beyond which they shall not be so taxed. The absence of regulations dealing with pensions merely brings about the result that, if Commonwealth pensions are taxable by the States, no maximum limit has been prescribed beyond which such taxation cannot be imposed. The absence of a regulation imposing a maximum limit cannot be regarded as amounting to a positive provision that they shall not be taxed at all.

Thus, in my opinion, Commonwealth pensions may validly be taxed under general State taxation laws unless the Commonwealth Parliament prohibits such taxation in whole or in part. There is no law of the Commonwealth Parliament which contains any such prohibition and therefore, in my opinion, such pensions are subject to the laws of the States relating to income tax to the same extent as other income received by pensioners. The questions asked in the case should be answered accordingly.

(Since this judgment was written the case of The Judges v. Attorney-General of Saskatchewan[51] has been decided by the Judicial Committee of the Privy Council. The decision in the present case is in accord with the principles there stated.)

Rich J.

In Amalgamated Society of Engineers v. Adelaide Steamship Co. Ltd.[52] the following passage occurs:—"Deakin v. Webb and Lyne v. Webb3(1904) [1904] HCA 57; 1 C.L.R. 585. were cases in which it was held that the State Income Tax Act of Victoria did not validly extend to tax moneys which had been received as Commonwealth salary. The decision was rested on two grounds, both found in the American case of Dobbins v. Commissioners of Erie County4(1842) [1842] USSC 6; 41 U.S. 435.. The first ground is that taxation of a person who is a Federal officer necessarily, per se, so far as it reaches money he received as salary, and although it so reaches that money by reason of provisions which apply generally to the whole community without discrimination, is an interference with the means employed by the Commonwealth for the performance of its constitutional functions. The second ground is that the State Income Tax Act was in conflict with the Commonwealth law fixing the officer's salary. The law, as laid down in those cases, was dissented from by the Privy Council in Webb v. Outrim5(1907) A.C. 81; 4 C.L.R. 356., and was disapproved by two justices as against three in the subsequent case of Baxter v. Commissioners of Taxation (N.S.W.)6(1907) [1907] HCA 76; 4 C.L.R. 1087.. Having regard to the principles we have stated, the first ground is erroneous. An Act of the State legislature discriminating against Commonwealth officers might well be held to have the necessary effect of conflicting with the provision made by the Commonwealth law for its officers relatively to the rest of the community. The second ground depends on the construction of the Commonwealth Act with which the State Act is alleged to conflict. If, on a proper construction of both Acts, they conflict, the State Act is, to that extent, invalid. But that is so by force of the express words of sec. 109, and not by reason of any implied prohibition. The final result is to be reached, not by a Commonwealth Act permitting the State legislature to exercise a power it does not possess—except where the Constitution itself so provides, as in sec. 91 and sec. 114—but by valid Commonwealth legislation expressly or impliedly by marking limits conflicting with State legislation which is valid except for the operation of sec. 109. It is on this ground that the actual decision in Chaplin v. Commissioner of Taxes (S.A.)1(1911) [1911] HCA 13; 12 C.L.R. 375. is to be upheld as correct. Baxter's Case2(1907) [1907] HCA 76; 4 C.L.R. 1087., of course, is in the same position as Deakin v. Webb3(1904) [1904] HCA 57; 1 C.L.R. 585.." It appears to me quite unnecessary in the present case to do more than apply this passage to the legislation governing the appellant's position in relation to his pension. The decisive provisions in that legislation consist in secs. 19 and 26 of the Financial Emergency Acts 1931, i.e., the Act No. 10 of that year as amended by the Act No. 47 of the same year. Under those provisions it was open to the Governor-General in Council to make regulations prescribing the extent to which the pension should be subject to State taxation of the description now in question. Unfortunately no such regulation was made. It may be that this was accidental and that it was believed that the regulation actually made under sec. 19 in relation to the remuneration of officers would be extended by sec. 26 to the pensions of former officers. Much as I should like to give this construction to sec. 26 I have found it impossible to do so. But whether the omission to make the regulation was intentional or due to accident or misapprehension, the fact remains that the legislation specifically provides a method for determining what shall be the liability of the pension to State tax and that the Federal authority has left the pension fully exposed to State tax. The decisive statement in the passage I have quoted from the Engineers' Case[60] is contained in the sentence, "If, on a proper construction of both Acts, they conflict, the State Act is, to that extent, invalid." Now in the present case sec. 19 as introduced by the amendment shows clearly that, except to the extent that the Governor-General may otherwise prescribe, the remuneration with which it deals shall be subject to State tax. Sec. 26 makes this provision applicable to pensions. This legislation plainly declares the degree of conflict. The conclusion is inevitable that in the absence of a regulation by the Governor-General in Council relating to pensions there is no conflict between the Federal statutory provisions and the State taxing Act. I am therefore of opinion that the State legislative provision under which the tax is levied is neither ultra vires nor inoperative in so far as it affects the pension. I wish to emphasize the fact that my judgment is based altogether on the effect of sec. 19 and sec. 26 (as amended) of the Financial Emergency Acts 1931. My dissent in Pirrie v. McFarlane[61] sufficiently indicates my view that the Commonwealth and its servants and agents are not liable to an interference for which no justification, authorization or allowance can be discovered in the Constitution or in legislation validly enacted thereunder. In the course of our consideration of this case it has come to our notice that the Court of Appeal of Manitoba has adopted the view that under the British North America Act it is not open to the Provinces to require a commissioned officer of the military forces of the Dominion to take out a licence or permit to drive a Dominion motor car although the provincial law was expressed in general terms and did not discriminate against the Dominion or its forces (R. v. Anderson[62]).

The first question in the case stated should be answered in the affirmative.

Starke J.

The appellant is entitled to and receives a pension under the provisions of the Commonwealth Superannuation Act 1922-1934. He has been assessed to income tax in respect of the pension received in the income year which ended on 30th June 1933, pursuant to the Special Income and Wages Tax (Management) Act 1933 N.S.W.. It was conceded—rightly, I think—that this pension receipt is caught by, and is assessable to income tax under, the provisions of that Act (See Act, secs. 2, 6). But it is contended that the provisions of the State Act contravene the Constitution, or are inconsistent with Federal legislation, and to that extent invalid. The general authority, however, of the State of New South Wales to impose such a tax cannot be questioned (Webb v. Outrim[63]). In that case the Judicial Committee rejected the elusive doctrine of the immunity of instrumentalities, an effect of which was that the Constitution by implication rendered immune from State taxation moneys received by a Federal officer as salary. The doctrine had been adopted in this court, but in the Engineers' Case[64] the court accepted the view of the Judicial Committee, and abandoned the doctrine. In so far as the decisions of this court, such as Deakin v. Webb; Lyne v. Webb[65]; Baxter v. Commissioners of Taxation (N.S.W.)[66], rest upon that doctrine, they can no longer be regarded as law. No distinction can be drawn between pensions under the Superannuation Act and salaries. Consequently, there is nothing in the Special Income and Wages Tax (Management) Act of New South Wales, in so far as the tax reaches money received by a person under the Federal law in the form of a pension, that is in contravention of the Constitution. The decision in Webb v. Outrim[67] is in itself decisive, and since the Engineers' Case[68] there is no valid reason for refusing to act upon it.

It was next argued that the tax imposed by the New South Wales Act was inconsistent with Commonwealth law, and that, by force of sec. 109 of the Constitution, the Act was invalid to the extent of the inconsistency. It is established, I think, by authority, and I assume that the Commonwealth might competently pass legislation protecting its officers, and exempting from State taxation salaries paid and pensions granted by it (Chaplin v. Commissioner of Taxes (S.A.)[69]; R. v. Brisbane Licensing Court[70]; The Commonwealth v. Queensland; Ex parte Daniell[71]; Smith v. Oldham[72]). The Superannuation Act 1922-1934 merely provides that a person contributing under the Act shall be entitled to a pension on his retirement on or after attaining the maximum age for retirement. It protects pensions from assignment, garnishee orders, and so forth (sec. 80). But it nowhere provides that the recipient shall not be liable to taxes which fall on all citizens alike. The cases are clear that the New South Wales Act is in nowise inconsistent with Federal legislation in the form of the Superannuation Act (Webb v. Outrim[73]; Abbott v. City of St. John[74]; Caron v. The King[75]; Pirrie v. McFarlane[76]; Cooper v. Commissioner of Income Tax (Q.)[77]).

The Commonwealth Salaries Act 1907 was referred to, and also the case of Chaplin v. Commissioner of Taxes (S.A.)[78], decided under it. But that Act does not extend to pensions granted under the Superannuation Act 1922-1934. In Chaplin's Case[79], Griffith C.J. said that the Commonwealth Salaries Act in substance enacted that "the grant of a salary to an officer of the Commonwealth is made on condition that he is subject to a law of the State as to taxation to the same extent as any other citizen, and is a solemn declaration that such taxation is not an interference with the exercise of the powers of the Commonwealth." I venture to suggest that it would be more correct to say that the Act, upon its true construction, protected Federal officers, and exempted them from State taxation in respect of the salaries received by them except so far as sanctioned by the Act.

The Financial Emergency Acts 1931, secs. 19 and 26, were also referred to. But it was conceded that the provisions of sec. 26 have not been implemented by any statutory regulation (See Financial Emergency (State Taxation) Regulations 1931).

The questions stated should be answered:—1. Yes. 2. No.

Dixon J.

An examination of the provisions of the Special Income and Wages Tax (Management) Act 1933 of New South Wales leaves no doubt of its meaning in relation to pensions payable under the Superannuation Act 1922-1934 of the Commonwealth. Its intention is that they shall be included in the assessable income upon the net amount of which the special income tax is to be levied.

The question for decision is whether the State Act can operate in accordance with that intention.

Two important considerations affect the determination of the question. The first is that the special income tax is imposed without differentiation upon the net assessable income of all persons. It does not discriminate in any way against the income received from the Commonwealth or against pensions. The second consideration is that the legislation of the Commonwealth, upon which the nature and characteristics of the pension necessarily depend, treats it as potentially subject to taxation by a State for the purposes of general revenue or to meet expenditure for a special purpose. For the Governor-General is authorized to prescribe the maximum extent of taxation to which pensions payable under the Superannuation Act 1922-1934 may be subject under the laws of a State imposing taxes on income both for the purpose of the general revenue of the State and where the law of the State expressly provides that the revenue received from the tax is to be applied to meet expenditure incurred by the State for any special purpose, or the regulations of the Governor-General prescribe that the tax shall be deemed to be of that nature. The Governor-General has not exercised his authority of setting a limit upon the State taxation of Federal pensions. See sec. 26 of Act No. 10 of 1931; sec. 3 of Act No. 47 of 1931; S.R. 1931, No. 138, and No. 154, and 1935, No. 28.

In my opinion these two considerations combine to destroy the foundation of the claim that the State legislation cannot operate to include pensions payable under the Commonwealth Superannuation Act 1922-1934 among the items of assessable income upon the net amount of which special income tax is imposed. The one consideration shows that the Commonwealth pensioner is affected by the State tax, not because his pension comes from the Commonwealth, but because as an ordinary member of the community he is in receipt of income. The other consideration shows that the administrative purposes of the Commonwealth are fulfilled by the payment of a pension possessing no special quality segregating it from the general resources of the recipient or distinguishing it in point of enjoyment from any other form of income. Thus the persons upon whom the tax is levied are chosen by the State without regard to any past or present connection with the Commonwealth and the inclusion of the Federal pension in the subject of the tax involves no inconsistency with the full effectuation of the purposes of the Commonwealth. Much caution should, I think, be used in applying decisions upon the British North America Act to questions arising upon the Commonwealth Constitution. The instruments are very different and few or none of the difficulties to which secs. 91 and 92 of the Canadian Constitution continue to give rise have any real counterpart in the Australian Constitution. But, having regard to the manner in which the Commonwealth legislation deals with the liability of the pension to State taxation, there is not much danger in applying to the present case the decisions upon the question whether there is an opposition between, on the one hand, the provision by the Dominion for its servants of a defined amount of salary, and, on the other hand, the inclusion of that salary in the income taxable under a Provincial law making no discrimination between Dominion servants and other taxpayers. In Abbott v. City of St. John[80] the Supreme Court of Canada held that no such opposition exists. The decision was cited with apparent approval by the Privy Council in Caron v. The King[81], a case depending upon somewhat different considerations. In the recent case of Forbes v. Attorney-General for Manitoba[82] the Privy Council reached the same conclusion and[83] adopted the decision in Abbott v. City of St. John[84]. But in relying upon these cases, it is necessary to notice the exact point at which they become applicable to the question before us. To the Dominion Parliament the British North America Act assigns the exclusive power of "fixing and providing for the salaries and allowances of civil and other officers of the Government of Canada." Upon the Provinces it confers the power of "direct taxation within the Province in order to the raising of a revenue for Provincial purpose." The first power was not considered to involve upon the part of the Dominion the provision of a salary or allowances protected against general Provincial taxation. Duff C.J., as he now is, said in Abbott's Case[85]:—"I am quite unable to perceive that the power thus conferred in any way restricts the operation of the power of taxation committed to the Province. The fixing and providing for salaries seems to be, as a subject of legislation, quite distinct from the power to levy taxes in respect of income. The principle upon which the burden of the fiscal contributions exacted by a municipality or a Province shall be distributed among those persons subject to its fiscal jurisdiction seems to be a subject as far removed as possible from that dealt with in sub-section 8 of sec. 91." The situation is thus analogous to that arising in the present case, but arising not as a result of any limitation upon the powers of the Federal Parliament or of any general doctrine as to the qualities attaching to Federal salaries or pensions, but as a result of the provisions of the Federal legislation which assume to protect the pension from State taxation only to the extent prescribed by the Executive and thus imply that otherwise its payment is to operate like any other contribution to the general resources of the pensioner.

It will be seen that the grounds I have given for deciding the present case concede that if a State tax discriminated against pensions, salaries, or other payments made by the Commonwealth, it could not be supported. If the right to the payment were conferred by a law of the Commonwealth, prima facie the scope and intention of that law would be enough to make inoperative any attempt by the States to impose upon the payee a special burden because of his occupying that character. The State tax would be inconsistent with the law of the Commonwealth in making enjoyment of the right or benefit conferred by the latter the special occasion of a burden. The invalidity of the State law would then be a result of sec. 109 of the Constitution. But I do not think that this is the only ground upon which a discriminatory State tax on salaries, pensions or other payments received from the Commonwealth would be invalid. In the ordinary course of administration many obligations not created or defined by statute are contracted by the Commonwealth and discharged out of moneys lawfully available for the purpose. This is done in the exercise of the powers conferred by the Constitution, as for instance, under the operation of secs. 61, 67, 69, 70, 81, 82 and 83, or one or more of them. Surely it is implicit in the power given to the Executive Government of the Commonwealth that the incidents and consequences of its exercise shall not be made the subject of special liabilities or burdens under State law. The principles which have been adopted for determining for the purposes of sec. 109 whether a State law is consistent with a Federal statute are no less applicable when the question is whether the State law is consistent with the Federal Constitution. Since the Engineers' Case[86] a notion seems to have gained currency that in interpreting the Constitution no implications can be made. Such a method of construction would defeat the intention of any instrument, but of all instruments a written constitution seems the last to which it could be applied. I do not think that the judgment of the majority of the court in the Engineers' Case[87] meant to propound such a doctrine. It is inconsistent with many of the reasons afterwards advanced by Isaacs J. himself for his dissent in Pirrie v. McFarlane[88]. Indeed, he there refers to "the natural and fundamental principle that, where by the one Constitution separate and exclusive governmental powers have been allotted to two distinct organisms, neither is intended, in the absence of distinct provision to the contrary, to destroy or weaken the capacity or functions expressly conferred on the other." He adds: "Such attempted destruction or weakening is prima facie outside the respective grants of power." There is little justification for seeking to find in the Engineers' Case[89] authority for more than was decided. The importance alike of the principle there applied and of the application given to it is sufficiently great and far reaching. It is a principle adopted for the interpretation of the legislative powers of the Parliament. The principle is that whenever the Constitution confers a power to make laws in respect of a specific subject matter, prima facie it is to be understood as enabling the Parliament to make laws affecting the operations of the States and their agencies. The prima facie meaning may be displaced by considerations based on the nature or the subject matter of the power or the language in which it is conferred or on some other provision in the Constitution. But, unless the contrary thus appears, then, subject to two reservations, the power must be construed as extending to the States. The first reservation is that in the Engineers' Case[90] the question was left open whether the principle would warrant legislation affecting the exercise of a prerogative of the Crown in right of the States. The second is that the decision does not appear to deal with or affect the question whether the Parliament is authorized to enact legislation discriminating against the States or their agencies.

The decision applied the principle of interpretation thus adopted to sec. 51 (xxxv.), which was accordingly construed as extending to operations notwithstanding that they are carried on by or on behalf of the States.

In so describing the effect of the decision, I have done little but repeat what I said on a former occasion (Australian Railways Union v. Victorian Railways Commissioners[91]). But, in spite of its compendiousness, I believe it to be an accurate enough statement of the law laid down by which we are to be guided. It is, perhaps, desirable to add that, in applying the general principle to a legislative power of the Commonwealth, the words at the head of secs. 51 and 52, "Subject to this Constitution," must not be overlooked and that these words together with sec. 106 and perhaps sec. 107 may be of great importance in a question how far a law of the Commonwealth may affect the States; and this is well illustrated by the views expressed by Starke J. in Australian Railways Union v. Victorian Railways Commissioners[92] and New South Wales v. The Commonwealth [No. 1][93].

The majority judgment in the Engineers' Case[94] contains an examination of Deakin v. Webb[95] and Baxter v. Commissioners of Taxation (N.S.W.)[96] which leaves those cases depending on the interpretation of "the Commonwealth law fixing the officer's salary." In every case where an opposition is said to exist between the law which secures remuneration or other payments and a law which taxes them, the scope and purpose of the law which secures the payments must be ascertained. The differing opinions delivered in Evans v. Gore[97] supply an example of a case in which the conceptions of the purpose and extent of the protection given by the law securing the salary determined the answers given to the question (Cf. Cooper v. Commissioner of Income Tax (Q.)[98]).

The Commonwealth legislation in the present case clearly means to leave the pension without any protection unless the Executive Government thought fit to confer it. This circumstance completely distinguishes the case from Deakin v. Webb[99] as explained in the Engineers' Case[100] and also from Evans v. Gore[101].

In my opinion the question in the case stated should be answered in favour of the Commissioner of Taxation for New South Wales.

Evatt J.

The New South Wales Special Income and Wages Tax (Management) Act 1933 made the appellant (a resident of New South Wales) liable to pay special income tax in respect of the sum of £258, which, as a superannuated officer of the Commonwealth, he had received during the income year ending June 30th, 1933, pursuant to the Commonwealth Superannuation Act 1922. The notice of assessment was issued on October 18th, 1934. The New South Wales Act in question did not authorize or require deduction of the tax at the source from which the appellant's pension was received, and it applied indifferently to all pensions, salaries and incomes received.

The question raised by the stated case is whether, by virtue of the Commonwealth Constitution or of Commonwealth legislation, the New South Wales Act cannot validly apply to income wherever it has comprised payments under Commonwealth statutes by way of pension or superannuation allowance to a retired officer of the Commonwealth.

At first glance this claim of immunity from the operation of legislation of a State which imposes taxes upon all incomes and pensions indifferently, would appear to be extremely novel, not to say daring. But a number of prior decisions of this court negative such appearance, and some reference must be made to

Old, unhappy, far-off things And battles long ago.

Before this is done, let us first turn to the terms of the Constitution. By sec. 114, a State is unable to impose any tax on Commonwealth property, but so is the Commonwealth unable to impose any tax upon a State's property. Sec. 114 can have no application to legislation taxing beneficiaries in respect of payments of superannuation which they have previously received, even although the fund from which the payments were made belong to the Commonwealth or to a State.

The Constitution seems to make it reasonably plain that no power is given to the Commonwealth Parliament to add to any grant of salaries and pensions to its officers an immunity from the general or non-discriminatory taxation legislation which a State imposes upon its residents or upon those deriving income from the State. The power of the Commonwealth to legislate, contained in sec. 52 (ii.) and sec. 51 (xxxix.), enables it to provide for salaries for its present officers and for pensions to officers who have retired from its service; but it seems equally clear that the Commonwealth Parliament cannot attach a condition or quality which will attach to its grant of salary or pension and operate after it has reached the hands of the recipient. It cannot directly provide that the recipient is immune from State taxation in respect of such receipt; nor can the same thing be done indirectly by supposing that the Commonwealth Parliament can create a new species of property (here it would be the old species—cash) with the attractive characteristic that one owner of it (the Commonwealth salaried officer or pensioner) and one alone, need not subsequently pay State taxation in respect of its receipt by him from Commonwealth sources. The legislation in relation to Commonwealth salaries and pensions has not been framed in the indirect method of assigning such special character to the income received. But the question may be important, because in one case (The Commonwealth v. Queensland[102]) such method was suggested, so that the "tax-free" securities would carry with them from hand to hand the same immunity from payment of State taxation. In the case of Commonwealth salaries and pensions the indirect method of conferring immunity from State taxation is not feasible, because it would require that a special quality should be attached to cash received, solely because it represented a payment of Commonwealth salary or pension, yet such quality would immediately vanish after the cash had been paid over to the recipient's grocer or milk supplier, who could never set up the same claim of immunity. In the case of salaries and pensions immunity must be supposed to arise more directly.

There is nothing in the Constitution itself which suggests that, in respect of granting immunity to its officers or ex-officers from payment of taxation, the Commonwealth is any more powerful than a tate would be to grant its officers or ex-officers immunity in respect of taxation by the Commonwealth.

Later, I will analyse the theory that sec. 109 of the Constitution, which has established a general rule for resolving actual conflicts between valid Commonwealth and valid State legislation, gives the Commonwealth some general "supremacy" over the States. It will then be seen that, for present purposes, it is not possible for the Commonwealth to force its salary or pensions legislation into direct conflict with the income taxation laws of a State without destroying the validity of its own legislation.

Nor can the power of the Commonwealth Parliament under sec. 51 (ii.) to make laws with respect to "taxation" assist, because that power relates only to taxation by the Commonwealth for the purpose of its raising a Commonwealth revenue. The only instance in our constitutional history where this fact was sought to be concealed was in the Commonwealth Act No. 11 of 1932—the Financial Emergency (State Legislation) Act 1932, which labelled itself an Act with respect to taxation, as well as with respect to other subject matters specified in sec. 51 of the Constitution. Of the Act as an exercise of the "taxation" power of the Commonwealth, Sir Robert Garran, in the official "Case" published on behalf of the Commonwealth in answer to the Case of Western Australia in favour of its secession from the Commonwealth, said:—

The draughtsman was evidently scratching round for a peg on which to hang the Bill; but taxation was certainly a forlorn hope. The power of the Commonwealth Parliament as to taxation is a power to make laws "for the peace, order and good government of the Commonwealth" with respect to "Taxation, but so as not to discriminate between States or parts of States." It can hardly be questioned that these words refer only to Commonwealth taxation, uniform throughout the Commonwealth, for Commonwealth purposes and do not cover control of State taxation. Nothing in any decision of the High Court suggests a doubt of this; and indeed the principles of interpretation laid down by the court make doubt impossible (The Case for Union, p. 26).

Even in The Commonwealth v. Queensland[103], where, in a quiet and unobtrusive way, the Commonwealth legislative power seemed to have been stretched at least up to the limits of constitutional elasticity, the judgments which affirmed the immunity of the holder of certain Commonwealth securities from the constitutional taxing power of Queensland, all based themselves on sec. 51 (iv.)—the power of "borrowing money on the public credit of the Commonwealth." As sec. 51 (ii.) was never invoked on behalf of the Commonwealth, we can surely take it as axiomatic that that sub-section gives the Commonwealth no authority to grant exemptions or immunities from taxation by the States.

Therefore, it would seem probable that the Commonwealth cannot lawfully add to its grant of salaries or pensions a further grant, viz., an immunity from State taxation. It is quite consistent with this that a law of a State which discriminates against the salaries or pensions of Commonwealth officials or ex-officials should be deemed invalid. Such a law would probably reveal itself as a poorly disguised attempt to interfere with the normal working of the Commonwealth services, for the additional State revenue obtainable would be almost negligible, so that the real purpose of the law could hardly be to obtain revenue. Moreover, if the Commonwealth sought to pass a so-called taxation law discriminating against State officials, a similar result should follow. The importance of the fact of discrimination is suggested in several of the judgments delivered in this court in Stuart-Robertson v. Lloyd[104], and it is fully recognized under the constitutional system of Canada (Kennedy, Law of the Taxing Power).

In one view, a law of the Commonwealth discriminating against State officials would not be a "taxation" law within the meaning of sec. 51 (ii.) of the Constitution. As stated in my own judgment in New South Wales v. The Commonwealth [No. 1][105]:

The Constitution makes no general prohibition against the passing of discriminatory enactments, but their presence in this, as in other Commonwealth legislation may reveal its real nature and character.

A different angle of approach to the question of discriminatory legislation is this, that it must at least be implied in the Constitution, as an instrument of Federal Government, that neither the Commonwealth nor a State legislature is at liberty to direct its legislation toward the destruction of the normal activities of the Commonwealth or States. Such a principle is not inconsistent with the rejection by the Engineers' Case[106] of the earlier doctrine of "immunity of instrumentalities," though it is inconsistent with the unqualified dogma that the Constitution leaves no room whatever for implications arising from the co-existence side by side of seven legislatures each of which is, despite the suggestions in The Commonwealth v. New South Wales[107], sovereign within the limits fixed by the distribution of constitutional functions (James v. The Commonwealth[108]; New South Wales v. The Commonwealth [No. 1][109]; see authorities cited in The King and His Dominion Governors, p. 206).

The above analysis suggests the following provisional conclusions:—

But it is necessary to examine the cases with some care in order to see whether such conclusions are sound. If they are, it will be unnecessary to examine the precise terms of the Commonwealth legislation which are said either to establish or disestablish the immunity which alone can invalidate the State's assessment against the present appellant.

In D'Emden v. Pedder[110] it appeared that a Commonwealth officer gave to a paying officer of the Commonwealth a written receipt for salaries received, but failed to stamp the receipt. A Tasmanian statute, general in its terms, required payment of a duty of 2d. on all such receipts, a stamp having to be affixed by every recipient of any amount between £5 and £50. The Commonwealth Audit Act made it the duty of its salaried officers to give a receipt to the paying officer. The court (Griffith C.J., Barton and O'Connor JJ.) held:—

This analysis of D'Emden v. Pedder[115] shows how little historical warrant there is for the statement in the Engineers' Case[116] that the former decision "rests on the supremacy created by sec. 109." No part of the Commonwealth's case against Tasmania was based upon sec. 109 of the Constitution. The question whether there was, in fact, inconsistency between the two enactments is not so easy to determine. In the first place, it was quite possible for the Commonwealth officer to obey both Commonwealth and State laws, so that there was no inconsistency between the laws in the sense of direct repugnancy, conflict or collision. Nor is it so clear that the Commonwealth Audit Act "covered the field," if I may use that hackneyed expression containing the dangers and ambiguities which I have elsewhere discussed (Stock Motor Ploughs Ltd. v. Forsyth[117]). The Audit Act concerned itself only with requiring the authentication of payments of money to officers by the giving of receipts. It is not easy to see how any "plan or scheme" of the Commonwealth would be at once "hindered and obstructed" so soon as the subject was "touched upon ... by State authority"[118].

Even if the Audit Act had expressly provided that no State law requiring a duty stamp to be affixed to receipts should apply to Commonwealth receipts, another constitutional question would have arisen. There would be "inconsistency" between the legislation, but could the Commonwealth Parliament validly pass such an enactment? On the whole it would seem that a Commonwealth Act in such a form would be warranted, because of possible or probable delay, inconvenience and embarrassment to the Commonwealth service occasioned by requiring the affixing of a duty stamp at the time of the signing of the receipt. But, clearly, a Commonwealth law dispensing with the necessity of observing the requirement of such a State law might have been countered by the State Parliament's passing supplementary legislation as part of its stamp duties law, requiring all persons in Tasmania who had given receipts without affixing a State stamp to pay to the State direct a sum equivalent to the stamp duty which would have been payable but for the Commonwealth legislation. Such State legislation would (a) remove the sole ground of inconsistency between it and the suggested Commonwealth legislation, (b) be a valid law with respect to taxation, in no way discriminating against Commonwealth officers, and (c) not be capable of being rendered void by Commonwealth legislation aiming at an exemption of Commonwealth servants from the payment of State duty in respect of their actual receipt of income because, prima facie, such Commonwealth legislation would not be a valid law in relation to its departments or officers.

It is to be noted that nowhere in Deakin v. Webb[125] is it suggested that sec. 109 of the Constitution has anything to do with the issue. The diminution of the income of the Federal officer so often referred to is relied upon, not as proving "inconsistency" between the Constitution or Commonwealth legislation and the Victorian Act, but merely as an additional reason for applying the doctrine of non-interference, i.e., the rule in D'Emden v. Pedder[126] (See[127]). Even in argument, only an unimportant reference was made to sec. 109.

That the decision in Deakin v. Webb[128] was not based in any degree upon sec. 109 of the Constitution is conclusively proved by the subsequent judgment of the court refusing an application for a certificate under sec. 74 of the Constitution so as to permit an appeal to the Privy Council. In dealing with such application, the court held that the question they had to decide was "one as to the limits inter se of the constitutional powers of the Commonwealth and a State"[129]. This was elaborated at p. 623, where "the only questions in the case" are set out. None of them concern sec. 109 of the Constitution, which necessarily assumes the general validity of the State legislation in question, and only invalidates such provisions of it as are deemed inconsistent with the Commonwealth legislation.

Deakin v. Webb[130], like D'Emden v. Pedder[131], rested in no way upon sec. 109 of the Constitution. In each case, State legislation was deemed invalid because of its attempt to "control the free exercise" of Commonwealth authority, contrary to the rule of immunity asserted in D'Emden v. Pedder[132].

Webb v. Outrim[133] was decided by the Privy Council in December 1906. It is now an historical fact to say that some of the reasons for the decision were quite unsatisfactory, and immediately regarded as unsound. But the decision itself may be regarded as a rejection of the theory that income tax legislation of a State cannot validly apply to a Commonwealth salaried official because it "might interfere with the free exercise of the legislative or executive power of the Commonwealth"[134], the theory being that such interference is "impliedly forbidden by the Constitution."

Judgment in the Railway Servants' Case[135] was pronounced on December 17th, 1906. The court held that the question was the converse of that decided in D'Emden v. Pedder[136] but that "the doctrine is equally applicable"[137]; i.e., the State authority may be protected from attempted invasions on the part of the Commonwealth, although "whether the alleged invasion is really one or not is an entirely different question"[138]. The conclusion was that "the doctrine of mutual freedom from interference as between the Commonwealth and State Governments would be sufficient to exclude any implication that sec. 51 (xxxv.) was intended to extend to State railways"[139].

Several months after Webb v. Outrim[140] the question of the application to Commonwealth officers of a general income tax law of a State again arose before the High Court in Baxter v. Commissioners of Taxation (N.S.W.)[141]. In Baxter's Case[142], Isaacs and Higgins JJ. were members of the Bench. The majority of the court (Griffith C.J., Barton and O'Connor JJ.) adhered to their views as expressed in D'Emden v. Pedder[143] and Deakin v. Webb[144]. The essential basis of the majority view in Baxter's Case[145] was that, the question dealt with in Deakin v. Webb being an inter se question, the court was "in no way bound" by the decision of the Judicial Committee (at p. 1118). Further, the possible bearing of sec. 109 upon the questions in D'Emden v. Pedder[146] and Deakin v. Webb was expressly excluded (See p. 1119). The reasoning of Deakin v. Webb was restated at pp. 1120-1122. None of it is concerned with sec. 109, and, finally, at p. 1129, the majority of the court said that the provisions of sec. 109 "had no application to the present controversy."

In view of later developments, the judgment of Isaacs J. is of great significance. At p. 1154 he argued that a question of repugnancy or inconsistency under sec. 109 was involved. But he added that the question of "obstruction" of the Federal Government's operations raised a broader issue, and that issue had to be determined (p. 1155).

With regard to the attack on the Income Tax Act based on sec. 109 Isaacs J. said:

The first ground of attack is outside sec. 74, and, if that were the only one, I should, in accordance with what I have already said, feel compelled to follow without discussion the decision in Webb v. Outrim[147], by which it has been held that such Acts are not repugnant to the Commonwealth legislation (at p. 1156).

Upon the broader question of the doctrine of immunity,

In relation to the Commonwealth officer, he concluded that the State Act

merely requires of him his just share of the ordinary burden of his fellow citizens in return for the protection and benefits the State affords him (at p. 1161).

Higgins J. said:

I concur with my brother Isaacs in the view that it is not an improper interference with a Federal agent for a State to collect from him a tax upon his income, on the same scale as from other citizens of the State, even though his salary as a Federal agent has to be included in his return (p. 1165).

Upon the subsequent application for a certificate under sec. 74, Griffith C.J. repeated the suggestion made by him in Baxter v. Commissioners of Taxation (N.S.W.3(1907) [1907] HCA 76; 4 C.L.R. 1087.), viz., that

the question whether a State tax upon the emoluments of Federal officers is within the prohibition is a minor question, for the Federal Parliament can make its grants subject to such a tax Quilibet potest renunciare juri pro se introducto[150].

In Flint v. Webb[151] Higgins J. said of Griffith C.J.'s suggestion "that the Federal Parliament may make its grants of salary subject to the rights of the State to tax them" that

at present I cannot see how, if an income tax upon the salary of a Federal servant is made invalid by the Constitution, the Federal Parliament cannot alter the Constitution by making the income tax payable. However, I do not wish to make any final pronouncement on the suggestion, which, as far as my memory serves me, has not been mentioned before in this court (p. 1194).

But the suggestion of Griffith C.J. was adopted in the Commonwealth Salaries Act 1907, with the result that the Privy Council in Commissioners of Taxation (N.S.W.) v. Baxter[152] was content to regard the controversy as a closed one. Thus it was not until May 1911 that the interpretation of the Act of 1907 came before the High Court in Chaplin's Case[153]. The Act merely declared, in accordance with the above suggestion of Griffith C.J., that "the taxation by a State, in common with other-salaries earned within the State" of the salaries earned in the State by Commonwealth officers should not, if the quantum of tax was not higher than that imposed on salaries of the same amount earned in the State, "be deemed ... an interference with the exercise of any power of the Commonwealth."

It will be observed that the legislation did not purport to authorize a State to tax Commonwealth officers' salaries. Accordingly the argument that, under sec. 109 of the Constitution, a Commonwealth law providing a salary for a Commonwealth officer was inconsistent with a State law taxing that salary (in common with others) still remained open to the taxpayer. It was not employed, nor was any reference made to sec. 109 in the judgment. This again shows that the only ground upon which Griffith C.J., Barton and O'Connor JJ. (Isaacs and Higgins JJ. did not sit in Chaplin's Case) thought invalidity could attach to the operation of State income tax legislation against salaries of Commonwealth officers was that of the "implied immunity of Commonwealth (and State) instrumentalities," and that immunity was not a personal right of the servant, but a right of the Government, which could waive it by appropriately worded legislation[154].

In Chaplin's Case[155] the court said, in relation to the grant of salaries:

The grant, if no more is said, is free from taxation by the State, but in making the grant the Commonwealth may say that the grant to the individual is subject to State taxation

This passage also is concerned only with the doctrine of immunity of instrumentalities, and cannot be used as any justification for the theory that, unless grants of Commonwealth salaries or pensions are expressly subjected to general income tax legislation of a State, sec. 109 should be deemed to invalidate the State legislation so far as applicable to such salaries or pensions.

The general doctrine of immunity of instrumentalities had numerous and startling applications before it was overthrown in the Engineers' Case[156]. As Higgins J. said in 1925,

it was held also on the same principle that a Federal officer is not liable to State income tax, although he gets the same benefit from the State's activities as others (Deakin v. Webb[157]; Baxter v. Commissioners of Taxation (N.S.W.)[158]. It was held that Federal properties are exempt from municipal rates although they enjoy the benefit of the municipal services (Sydney Municipal Council v. Commonwealth[159])... It was held that a Commonwealth Act which applied the provisions of the Conciliation Act to employer and employees in State undertakings was invalid (Railway Servants' Case[160]). It was held that a night-soil contractor is entitled, if his contract be with the Commonwealth, to remove night-soil from Commonwealth premises without taking out the licence required by the municipality (Roberts v. Ahern[161]). Without prejudging these and other similar cases should the same question arise again, I may say that the questions will have to be considered now in the light of the Engineers' Case[162][163].

The outstanding feature of the Engineers' Case is that the only question before the court was this—Did the Parliament of the Commonwealth have power "to make laws binding on the States" under sec. 51 (xxxv.)—the legislative power in relation to arbitration for the settlement of disputes extending beyond the limits of one State. The question was answered: Yes (p. 177). But, ten years later, in the case of Australian Railways Union v. Victorian Railways Commissioners[164], the old answer was added to by the qualification that "binding on the States" must not be interpreted as a synonym for enforceable against the revenues of the State.

The actual decision in the Engineers' Case was inevitable. Otherwise, in some industries, the Commonwealth arbitration power under sec. 51 (xxxv.) would have been entirely nullified, because the leading employers of labour in the industry were State governmental authorities, and, although they became parties in fact to an industrial dispute of the kind mentioned in sec. 51 (xxxv.), they could never be governed by an arbitrator's award, made in settlement of one industrial dispute. Under the contrary view expressed in the Railway Servants' Case, acute disputes in some industries where State Governments were the sole or main employers could not be settled by arbitration at all; yet the sole purpose of the power in sec. 51 (xxxv.) was to prevent or settle all such disputes either by conciliation or arbitration.

Even if the rule in D'Emden v. Pedder[165] had been retained in favourof both States and Commonwealth, there was little reason for believing that the normal exercise by the Commonwealth of its legislative power under sec. 51 (xxxv.) would amount to an unjustified "invasion" of State authority. However, in the Engineers' Case[166] the court rejected and denounced the rule in D'Emden v. Pedder[167] altogether. The method of denunciation appears at pp. 144 et seq. The method is perhaps open to the objection that, despite the criticism levelled at Griffith C.J., it is fairly plain that all Griffith C.J. meant in his remarks as quoted at p. 145 was that the bare words of the document did not cover the whole field of permissible action by Commonwealth and State, and the court had the ultimate duty of protecting both Commonwealth and State from hostile attacks on the part of State or Commonwealth. The method also left itself open to the pungent criticism made upon it by Mr. T. C. Brennan in his recent work Interpreting the Constitution (1935) (at pp. 152, 153). It was also misleading to suggest that, after Webb v. Outrim[168], only three of the original justices of the court accepted the rule of immunity of instrumentalities laid down in D'Emden v. Pedder, for, in Baxter's Case[169], Isaacs J. accepted the rule, and he also suggested that the rule enured for the benefit of State as well as Commonwealth.

But, however the result was reached, it must now be accepted that the "rule" of immunity from mutual interference was too broadly stated in the earlier cases, and as so stated it is now an exploded doctrine. But it does not follow that discriminatory legislation passed by the Commonwealth or a State and directed against the other can be properly passed upon by the Court without recourse to principles of the Constitution which are not immediately discoverable in its words. Something of what I mean is indicated by Gavan Duffy J. in his dissenting judgment in the Engineers' Case[170] where he declares that "the existence of the State as a polity is as essential to the Constitution as the existence of the Commonwealth," the converse proposition, of course, applying equally. The reminder is admittedly a vague one, because the dangers to be guarded against cannot be precisely defined, and will seldom, if ever, occur. But I am not prepared to accept all the obiter dicta in the Engineers' Case[171] as having achieved the impossible task of anticipating every future difficulty in the working of our Federal constitutional scheme.

Although, therefore, the attempt of Griffith C.J. to enunciate the rule of non-interference ended in disaster, this was due, perhaps, to over-anxiety as to the dangers which confronted the newly established Commonwealth. These dangers were greatly, and, as we would now suppose, even absurdly, exaggerated. For instance, how the Commonwealth itself could possibly be injured because its officers had to bear the same taxation burdens as their fellow citizens, it is almost impossible to appreciate.

For present purposes, the most important aspect of the Engineers' Case is its treatment of prior decisions. D'Emden v. Pedder[172], as a decision only, was regarded as sound as it "rests on the supremacy created by sec. 109" (p. 156). We have seen that the High Court, which decided D'Emden v. Pedder, never said that this was so. "Supremacy" was the new euphemism for the less ambiguous words employed in sec. 109 itself in order to resolve actual conflicts between valid Commonwealth and valid State legislation. The basis on which D'Emden v. Pedder was decided has been sufficiently described. One may conceive that the Commonwealth could, by differently framed legislation, have validly prevented the Tasmanian Stamp Duties Act from operating so as to incommode the actual performance of duties by a Commonwealth officer. But, whether actual "inconsistency" was established between the two Acts, as actually framed, is a very different question.

Of the decision in Deakin v. Webb; Lyne v. Webb[173], the Engineers' Case judgment said it rested upon (1) the doctrine of immunity of instrumentalities, and (2) sec. 109 of the Constitution[174]. Ground 1 was declared to be erroneous. The second ground was discussed in somewhat vague language. As I interpret the passage at p. 157 in the leading judgment, it emphasizes that the law as laid down in Deakin v. Webb and Lyne v. Webb was dissented from by the Privy Council in Webb v. Outrim,[175] and disapproved by two justices as against three in the subsequent case of Baxter[176]; and it deliberately refrains from giving approval to the actual decisions in Deakin v. Webb[177] and Lyne v. Webb[178].

Further, the Engineers' Case[179] declares that the "actual decision" in Chaplin's Case[180] is to be upheld as correct. But the "actual decision" was merely that the tax assessment upon the Commonwealth, officer was lawfully made by the State. And such decision might have been reached by three different methods, viz., by holding (1) that the doctrine of immunity of instrumentalities could be waived to any extent deemed fit by the Parliament concerned, i.e., in the particular case, the Commonwealth Parliament, or (2) that there was not and never had been any inconsistency between the Federal legislation granting Commonwealth salaries and the State income tax legislation, the 1907 Commonwealth Salaries Act being quite unnecessary and, indeed invalid, or (3) that whether or not there existed any such inconsistency, the 1907 Act could be interpreted as a declaration of the Commonwealth Parliament, which, by yielding to the State legislation, could be regarded as negativing any such "inconsistency."

The learned judges who decided Chaplin's Case decided it upon the first of the three grounds, but the Engineers' Case makes such ground no longer tenable. I should have supposed that Chaplin's Case was correctly decided upon the second ground specified above, a ground which would cover the present case. Upon that footing, (1) the Commonwealth Act of 1907 was void, because the Commonwealth Parliament cannot add to the grant of its salaries an immunity from general State income taxation, nor is it of any avail to make such grant subject to the taxing power of the State, which rests upon the Federal and State Constitutions, and not upon Commonwealth legislation: but (2) there was no "inconsistency" between the Commonwealth Act granting salaries and the State Act imposing taxation, so that the State assessments were good.

However, in the case of The Commonwealth v. Queensland[181], subsequently discussed, it was suggested, in one judgment at least, that the ground upon which the decision in Chaplin's Case is to be supported is the third of those set out above.

Whether the decision in Chaplin's Case[182] is to be rested upon ground 2 or 3 should here be determined, and, in my view, the correctness of Chaplin's Case depends upon the correctness of ground 2 stated above.

The passage in the Engineers' Case[183] commencing with "the final result" is somewhat obscure, because "the final result" referred to must be the invalidity of State legislation. Yet the next sentence commences "It is on this ground" etc., suggesting that a similar result—invalidity of State legislation—was reached in Chaplin's Case, whereas Chaplin's Case actually determined its validity.

Desirous of overthrowing the troublesome rule in D'Emden v. Pedder[184] the court deciding the Engineers' Case interpreted D'Emden v. Pedder as declaring Commonwealth "supremacy." But that was not intended by the first three judges who enunciated the rule. Hence the criticism in the Engineers' Case —"Mutual supremacy is a contradiction of [sic] terms"[185] was hardly valid, considering that the word "supremacy" had never been used by the judges who propounded the rule, and that the very point in issue in the Engineers' Case was the correctness of the Railway Servants' Case[186], where the rule in D'Emden v. Pedder was regarded as one of mutual non-interference, and not one of mutual "supremacy." Similarily, the supposed "hopeless opposition"[187] between the Railway Servants' Case and the Steel Rails Case[188] is based upon the assumption that the insistence by the Commonwealth upon subjecting a State Government, which had imported steel rails, to the same customs law as bound every other importer, amounted in fact to a "drastic interference" with the Government of New South Wales.

No doubt, the decision to reject the general rule of "mutual immunity of instrumentalities" was a wise one, and it must be followed by us. But it is quite erroneous to regard the Engineers' Case as having established a new and valid constitutional principle, under which, either by direct declaration as to the termination of specified State legislation, or as to the State's legislative power, or by indirectly creating conditions or qualities under Commonwealth legislation which will achieve the same objectives, the Commonwealth Parliament is enabled, by the exercise of its own legislative power, to rid itself of any State legislative "interference" or "impediment." This constitutional principle or doctrine is a dangerous feature of the Engineers' Case[189], and any proposed application of it should be most carefully watched. Already there have been startling consequences, to which reference will be made. I do not think that it was ever intended by the decision of the majority in the Engineers' Case to set in operation any such principle as I have stated, because it was stated, though without emphasis, that there could be no case of State legislation being annulled except by the operation of "valid Commonwealth legislation expressly or impliedly by marking limits conflicting with State legislation which is valid except for the operation of sec. 109" (p. 157—italics are mine),

In one very important aspect, the principles discussed again arose for consideration in the same year as the Engineers' Case was decided. The Court upheld the validity of Commonwealth legislation which provided that the interest from certain Commonwealth securities should not be liable to income tax under any law of the Commonwealth or a State, unless the interest was declared to be so liable by the Commonwealth loan prospectus (The Commonwealth v. Queensland[190]). There was, of course, actual collision between the Commonwealth legislation and any State legislation directly taxing a bondholder as a State citizen in respect of the annual interest received by him from the Commonwealth securities. In the interpretation given to sec. 52B of the Commonwealth Inscribed Stock Act 1911, the Court extended the area of actual collision between Commonwealth and State legislation, so as to cover a further form of State taxation. That part of the decision is not relevant for present purposes, for its vital feature was that the Commonwealth, purporting to exercise its legislative power with respect to "borrowing money on the public credit of the Commonwealth" (sec. 51 (iv.)), not only provided for the issue of securities, the payment of interest and the repayment of the loan, not only gave a promise as to future taxation by the Commonwealth itself, which, though it could not fetter the Commonwealth's legislative power, concerned only the Commonwealth and not the States, but actually purported to forbid the States from exercising their ordinary constitutional powers of taxation wherever such exercise would impose a tax upon the interest received from Commonwealth securities.

In this vital case, the States as a group were not represented. Further, the only argument of principle advanced by the State of Queensland was to the effect that sec. 51 (iv.) of the Constitution "only authorizes legislation as to the terms of the contracts between the Commonwealth and the lender, and confers no powers to impose conditions upon third persons" (p. 6). But the outstanding feature of the case was not that the Commonwealth was seeking to bind third persons generally, but that it was making an extremely audacious attempt to set at nought the great constitutional powers of each State to impose taxation upon all its citizens for the maintenance of its internal peace, order and good government. In considering so grave a question, it could make no difference whether the Commonwealth Parliament said directly: "The States shall not tax you, our bondholder, in respect of any money received by you from us as interest"—or whether it preferred to say: "Our securities are to be given a special quality or attribute, viz., interest payable under them shall be free of any taxation imposed by a State."

The great point as to the constitutional powers of the States was not dealt with by Knox C.J., who apparently considered that any condition might be included in the contract between Commonwealth and bondholder, so that, by parity of reasoning, the Commonwealth could have granted its bondholder, in respect of his interest, an immunity from execution at the suit of the bondholder's creditors. Isaacs and Rich JJ. rightly stated that the Commonwealth may "guarantee that the lender shall have, and may retain to the full ... the remuneration promised him by the Commonwealth" (p. 21) (Italics are mine).

The only question which might arise would be as to whether the phrase "retain to the full" implied that the lender would not only obtain the interest in full, but might also be put in the position, quoad State taxation, of never having obtained it at all.

In the same judgment Isaacs and Rich JJ. state (p. 22) that Chaplin's Case[191] "established the immunity of Federal salaries from State income tax. This court has recently, in Amalgamated Society of Engineers v. Adelaide Steamship Co.2(1920) [1920] HCA 54; 28 C.L.R. 129., expressed its opinion that the decision was correct, by reason of the controlling force of Federal legislation" (The Commonwealth v. Queensland[193]).

It is difficult to apply this reference. As already stressed, the actual decision in Chaplin's Case merely established the liability of Federal salaries to undiscriminatory State income taxation. If Chaplin's Case is to be regarded as having been based upon the 1907 Commonwealth Salaries Act, it is hard to understand how Deakin v. Webb[194], Lyne v. Webb[195] and Baxter's Case[196] were not also right in their actual result, because they were decided prior to the 1907 Act. Yet the fair inference from the judgment in the Engineers' Case is that those three decisions were all erroneous, and that in Baxter's Case, the actual dissent of Isaacs and Higgins JJ. was sound. Further, in the Engineers' Case[197] the court attacked the device of "a Commonwealth Act permitting the State legislature to exercise a power it does not possess," and this attack was meaningless unless levelled at the form of the 1907 legislation. Further, the 1907 legislation never purported to confer power on the States.

In The Commonwealth v. Queensland[198] Higgins J., referring to the Commonwealth Parliament, said:

Just as it can, under the defence power, isolate specified ground from the intrusion of the public, so it can under the borrowing power, isolate its stock from State taxation.

With respect, it is not easy to see any analogy whatever between the protection of defence areas from actual trespass—a necessary and well-recognized military precaution—and the direct attempt of the Commonwealth to grant an immunity to its lender from obedience to the general taxation laws of the State in which the lender obtains the protection of the laws. It would seem that an immunity from taxation by a State had no connection with the constitutional power of legislation with respect to the Commonwealth's borrowing money, except perhaps this, that every loan can be made more attractive if the borrower is able to offer monetary concessions over and above the repayment of his principal and interest. But the question was whether such concessions can lawfully be made at the sacrifice of the powers of the States. Of course, the Commonwealth could pay its lender an amount of interest which, after deduction of all taxation thereon, including State taxation, would yield a specified net return. But such a payment would have been expensive to the Commonwealth, which preferred to be generous to its lenders, not at its own expense, but at that of the States.

The above analysis of The Commonwealth v. Queensland[199] is of importance because, although the actual decision is not being called into question, there is an analogy between the attempt by the Commonwealth Parliament to attach to the grant of its salaries to its own officers a guarantee of freedom from income taxation of the States in respect of the past receipt of such salaries, and its attempt to prevent the State from exercising its ordinary constitutional authority to enact an undiscriminatory tax upon the income of all its residents (not excluding income from Commonwealth securities). If both of such attempts proved successful, the Commonwealth could confer upon favoured classes or individuals an unexampled series of special privileges, including a privilege to ignore the constitutional powers of the States. In order to deny the possibility of such success, no one need rely upon any doctrine of the "reserve powers" of a State, and it would be sufficient to invoke the principle which may be expressed in the words of Duff J. (of the Supreme Court of Canada) in the well-known case of Abbott v. City of St. John[200]. There, Duff J. dealt with the exclusive power of the Dominion to legislate in reference to the provision of salaries for Dominion officers, and said:—

I am quite unable to perceive that the power thus conferred in any way restricts the operation of the power of taxation committed to the province. The fixing and providing for salaries seems to be, as a subject of legislation, quite distinct from the power to levy taxes in respect of income. The principle upon which the burden of the fiscal contributions exacted by a municipality or a province shall be distributed among those persons subject to its fiscal jurisdiction seems to be a subject as far removed as possible from that dealt with in sub-sec. 8 of sec. 91. If one were to speculate upon the intentions of the framers of the Act, I should suppose nothing further from their intentions than the exemption of Federal office holders as a class from the fiscal burdens incident to provincial or municipal citizenship[201].

The decision in Abbott v. City of St. John[202] was approved by the Privy Council in Caron v. The King[203] and Forbes v. Attorney-General for Manitoba[204].

It is indisputable that the Engineers' Case[205] finally establishes two important points in relation to sec. 51 of the Constitution. First, a rule of construction according to which the theory that all State powers or activities must be treated as having been previously reserved from the scope of sec. 51 is rejected, the true rule being that each subject matter in sec. 51 must be regarded as a separate subject matter, so that if, upon its fair construction, any State activity is included within the subject, it is for the Commonwealth Parliament alone to decide upon the desirability of applying its legislation to such activity. Second, that, under sec. 51 (xxxv.), the activities of a State as employer in industries are within the scope of the subject matter of that sub-section. But the Engineers' Case is no warrant for the extreme view that the result reached in relation to sec. 51 (xxxv.) necessarily applies indifferently to all the placita in sec. 51, still less is it a warrant for the view that the Commonwealth is at liberty to nullify the ordinary constitutional powers of a State by purporting to exercise one or other of the powers of legislation set out in sec. 51. For instance, even the Commonwealth's power to restrict or restrain by legislation under sec. 51 (xxxv.) the proceedings and determinations of State industrial tribunals is qualified, not absolute (Australian Timber Workers' Union v. Sydney and Suburban Timber Merchants' Association[206]). Further, it must not be supposed that, in dealing with "bankruptcy and insolvency," the Commonwealth could make a sequestration order against a State, though it could bind a State in relation to its position as creditor in a bankruptcy. Outstanding factors in interpreting the thirty-nine placita in sec. 51 are the separable nature of each of the placita and the principle that the legislative powers of the States are concurrent with those of the Commonwealth (James v. The Commonwealth[207]).

Further, as I endeavoured to illustrate in the case of Stock Motor Ploughs Ltd. v. Forsyth[208], attempts by the Commonwealth Parliament to manufacture "inconsistency" between its own legislation and that of the States will often be essayed only at the price of making the Commonwealth legislation ultra vires. Of course, lawful attempts by the Commonwealth may occur, as in R. v. Brisbane Licensing Court; Ex parte Daniell[209], where State referenda and general elections were forbidden to be held on the same day as Commonwealth elections. No doubt, the State's legislation, actual and prospective, was avoided pro tanto. But the Commonwealth's legislative power over its own electoral system was deemed sufficient to enable it to prevent the awkwardness and confusion which might well result from a simultaneous Commonwealth and State election. In fact, the State was not impeded in its constitutional functions, for 364 other days in the year were left for it to choose from. On the other hand, a Commonwealth electoral law which forbade the holding of State elections for six months prior to a Commonwealth election would obviously be invalid.

In some respects Pirrie v. McFarlane[210] represents the culminating point of the theory that any inconvenient State legislation may be invalidated by the Commonwealth's manufacturing "inconsistency" within the meaning of sec. 109 of the Constitution. There, the Motor Car Act 1915 of the State of Victoria required that all persons driving a motor car upon any public highway should have a licence for that purpose. In the interpretation of the State Act, the majority of the High Court held that persons whether in the service of the Commonwealth or of a State were subjected to the licensing requirements of sec. 6. Then the question arose as to whether, under the State Act, there could be a valid conviction of a person who was admittedly driving a car upon a public highway without a licence, but who (i.) was an enlisted member of the Commonwealth Air Force, and (ii.) acted in accordance with instructions to drive the car through the streets of Melbourne in order to pick up a passenger, the latter being an officer of the same force.

The majority of the court held that the defendant was rightly convicted, and they rejected the argument that, under the Constitution or the Defence Act in its existing form, members of the Defence Force when on duty in time of profound peace were to be regarded as exempt from obedience to the requirements of motor traffic legislation. The conclusion, however, was accompanied by certain obiter dicta (Knox C.J., p. 184; Higgins J., p. 218; Starke J., p. 229) to the effect that the Commonwealth Parliament might, by passing amended defence legislation, exempt its officers from the obligation of being licensed under State law. This suggestion perhaps overlooks the fact, stressed by Higgins J. on the question of interpretation, that for traffic regulation "to be effective, all the traffic must be bound" (p. 218). The same fact is very important in dealing with the question of legislative power. The Commonwealth Parliament has no general power to regulate traffic so as to secure the safety of drivers and pedestrians; with that subject the State alone can deal. It requires a licence from a driver in order to ensure safety, and the charge imposed is small, practically nominal. Higgins J. rightly referred to the Canadian case of Abbott v. City of St. John[211] already discussed above, which affirmed the power of a Province to tax Dominion officials "in like manner as all other residents." He regarded the motor traffic legislation in question in Pirrie v. McFarlane[212] as illustrative of an analogous principle. A general exemption in favour of members of the defence force might be destructive of any effective regulation of motor traffic.

Perhaps it might be said that the broad scheme of the Australian Constitution is that of mutual subjection to law of both Commonwealth and the States so that, generally speaking, all the laws of the Commonwealth may bind all the people of the Commonwealth (including, of course, State servants), and that all the laws of a State shall bind all the people of the State (including, of course, Commonwealth servants in that State). But the point left outstanding in Pirrie v. McFarlane is whether the Commonwealth Parliament can validly confer upon its servants a special immunity from the operation of State traffic legislation and regulations. It certainly could not give a general exemption from obedience to the directions of State officers as to traffic and speed observances. In time of war or emergency the position might be different, but any such Commonwealth legislation would have to be conditioned upon the existence of war or emergency. The spectacle conjured up by Isaacs J. (see p. 211) is not very convincing. In the particular case under investigation, there was no evidence as to what the officer to be "picked up" was doing. He might not even have been engaged on the business of the department. According to the reasoning, as expressed at p. 205, the messenger of a Federal judge, sent to obtain books of reference, may disobey traffic regulations, because "soldiers perform their official duties necessarily on the highways of the country as well as in barracks," and "judges and members of Parliament are by the very nature of their duties to be left to their free exercise, unfettered by State legislation." Some of the reasoning used by Isaacs J. bears a resemblance to the doctrines propounded in D'Emden v. Pedder[213]. I do not see how the grant of general exemptions from obedience to the requirements of State traffic legislation could possibly be regarded as a valid law with respect to defence; although, if the legislation were limited to occasions of emergency or military necessity, the result might be different. Further, State traffic legislation, though incidentally affecting Commonwealth (including military) officers, cannot possibly be regarded as a law with respect to defence, and as intruding upon the exclusive domain of Commonwealth legislative or executive authority over defence.

The conclusions to which I have come in the present case are:—

The stated case should be answered:—1. Yes. 2. No.

McTiernan J.

The appellant upon his retirement from the public service of the Commonwealth became entitled to a pension under the Commonwealth Superannuation Act 1922-1934. It was paid to him out of the fund established under that Act to provide superannuation benefits for persons in the Commonwealth Public Service and other benefits for their families. The fund is maintained by the contributions which the Act obliges such employees to make out of their salaries and by the statutory payments made by the Commonwealth out of revenue. The benefits provided from this fund to employees, although paid upon the termination of their services, are part of the emoluments attached to their employment. The appellant is resident in New South Wales and the moneys which he has received from the Commonwealth are required by the Special Income and Wages Tax (Management) Act 1933 of New South Wales to be included in his income for purposes of assessment for tax under that Act. The Commonwealth Superannuation Act was passed under the exclusive powers of the Commonwealth to make laws with respect to matters relating to the public service of the Commonwealth (Constitution. sec. 52). The State Act is general and non-discriminatory and was passed under the powers which are reserved to the States by the Commonwealth Constitution. The question for decision is whether the State Act validly extends to tax the appellant's Commonwealth pension.

The analogous and equally delicate question, whether a Commonwealth officer is liable to State taxation in respect of his salary, was settled by the Commonwealth Salaries Act 1907, the legislature thereby embracing a solution which had judicial recommendation (See Baxter v. Commissioners of Taxation (N.S.W.)[215]; Flint v. Webb[216]; Chaplin's Case[217]). Before this Act was passed the Privy Council had decided that a Commonwealth officer was not exempt from State taxation in respect of his salary (Webb v. Outrim[218]), whereas the High Court reached the opposite conclusion both before and after the Privy Council had given its decision (Deakin v. Webb[219]; Baxter's Case[220]). The application of the Commonwealth Salaries Act 1907 is limited to the salaries of Commonwealth employees. It does not assume to remove any obstacle which may exist to a State levying taxation on the superannuation benefits paid to former employees of the Commonwealth living within the territorial jurisdiction of a State.

The High Court had, in the early cases above referred to, declared that the taxation by a State of moneys received as salary by a Commonwealth officer was an illegal interference with the means employed by the Commonwealth to perform its constitutional functions. But it decided in the Engineers' Case[221] that the immunity of the salary of a Commonwealth officer from State taxation could not, consistently with the Constitution, be defended on that ground. That case, in which four of the justices so decided, was more directly concerned with settling the difficulty which arose from the conflict of laws where the Commonwealth by an Act passed under sec. 51 of the Constitution, enters a field already in the lawful legislative occupancy of the States. The court there decided that this difficulty was resolved by the express rule in sec. 109 of the Constitution which gives legal supremacy to Commonwealth law, and that the doctrine of "implied prohibition" could not be invoked in aid of the States. Knox C.J., Isaacs, Rich and Starke JJ., in a joint judgment, said:—"The doctrine of implied prohibition finds no place where the ordinary principles of construction are applied so as to discover in the actual terms of the instrument their expressed or necessarily implied meaning. The principle we apply to the Commonwealth we apply also to the States, leaving their respective acts of legislation full operation within their respective areas and subject matters, but, in case of conflict, giving to valid Commonwealth legislation the supremacy expressly declared by the Constitution, measuring that supremacy according to the very words of sec. 109. That section which says When a law of a State is inconsistent with a law of the Commonwealth, the latter shall prevail, and the former shall, to the extent of the inconsistency, be invalid, gives supremacy, not to any particular class of Commonwealth Acts but to every Commonwealth Act, over not merely State Acts passed under concurrent powers but all State Acts, though passed under an exclusive power, if any provisions of the two conflict; as they may—if they do not, then cadit quæstio" (Engineers' Case[222]). In expressing the court's opinion that the doctrine of implied prohibition was not a sound ground upon which to base the immunity of Commonwealth salaries from State taxation, the justices referred to the character of the State legislation imposing taxation with which the previous decisions were concerned. Attention was directed to its non-discriminatory character. It imposed on the Commonwealth officer, whom it assumed to tax, no different burden from that borne by other persons in the territory of the State who were affected by it. The justices also dealt with the question whether a State Act discriminating against persons who were Commonwealth officers would be invalid because of its special character. They said: "An Act of the State legislature discriminating against Commonwealth officers might well be held to have the necessary effect of conflicting with the provisions made by the Commonwealth law for its officers relatively to the rest of the community" (Engineers' Case[223]).

The court had in the previous cases relied on another ground as well, for its conclusion that Commonwealth salaries were exempt from the operation of State income tax laws. Referring to that ground, the justices in the Engineers' Case[224] said:—"The second ground depends on the construction of the Commonwealth Act with which the State Act is alleged to conflict. If, on a proper construction of both Acts, they conflict, the State Act is, to that extent, invalid. But that is so by force of the express words of sec. 109, and not by reason of any implied prohibition." The present problem depends for its solution on the same considerations. The authority of the Engineers' Case was not impugned in argument. It follows that the appellant's liability to State taxation turns on the question whether it would be consistent with the Commonwealth fixing the measure of the benefit which that law intends that the appellant should receive by way of pension, for the State Act to include his pension within the income upon which he is liable to pay income tax to the Government of New South Wales. In Chaplin's Case[225], Griffith C.J., with whom Barton and O'Connor JJ. agreed, said that the grant of a salary by the Commonwealth to an employee for the performance of his duties is the same in principle as the grant of a franchise to an individual or a company. He added: "The grant, if no more is said, is free from taxation by the State, but in making the grant the Commonwealth may say that the grant to the individual is subject to State taxation." (Compare The Commonwealth v. Queensland[226]; Fairbairn v. Comptroller of Stamps (Vict.)[227].) The reason given by the Chief Justice for this view is not unrelated to the condemned doctrine of implied prohibition. Nevertheless the conclusion reached in Chaplin's Case[228] that the salary was taxable within the limits marked out by the Commonwealth Salaries Act 1907 was approved in the Engineers' Case, on the ground that after the passing of the Commonwealth Salaries Act there was no inconsistency between Commonwealth law with respect to the remuneration of Commonwealth employees and State law which included that remuneration within his taxable income, provided that the taxation was within the limits marked out by the Commonwealth Salaries Act.

The decision of the present question is not embarrassed by the want of judicial unanimity on the question whether an Act taxing the recipient of a salary in respect of that income is consistent with a law fixing the amount of the salary (See Deakin v. Webb[229]; Baxter's Case[230]; Dobbins v. Commissioners of Erie County[231]; Evans v. Gore[232]; Caron v. The King[233]). The terms of the relevant Commonwealth legislation show that in the present case there is no inconsistency between Commonwealth and State law.

The Superannuation Act, which is the Act fixing the amount of the pension, must, for the present purpose, be read with the provisions of the Financial Emergency Acts of the Commonwealth, which contain provisions with respect to the same subject matter. These provisions are fully referred to in the judgment of the Chief Justice and it is unnecessary to quote them again. They show that after the passing of the Financial Emergency Acts the Commonwealth did not intend to grant superannuation benefits which would be exempt from all taxation which a State might under its own proper authority levy in respect of that income. The fact that the Governor-General has not prescribed the maximum which may be imposed by State law does not negative the intention apparent on the face of the relevant Commonwealth legislation that the measure of the grant to be enjoyed by Commonwealth pensioners should be liable to be affected by valid State income tax legislation. The consequence is that to include the moneys received by the appellant as pension within the income in respect of which he is liable to taxation under the relevant State Act is not to derogate from the rights which he derives from Commonwealth law.

In my opinion the questions for decision should be answered in favour of the Commissioner of Taxation.

Questions in the case stated answered:—1. Yes. 2. No.

Solicitor for the appellant, Percy R. Watts.

Solicitor for the respondent, J. E. Clark, Crown Solicitor for New South Wales.