Melbourne Corporation v Commonwealth

74 CLR 31
[1947] ALR 377

(Judgment by: Latham CJ.)

Briginshaw v.
Briginshaw

Court:
High Court of Australia

Judges:
Latham CJ
Rich J
Starke J
Dixon J
McTiernan J
Williams J

Judgment date: 15 and 16 July, 13 August 1947


Judgment by:
Latham CJ.

The demurrer to the statement of claim in this action raises the question of the validity of s 48 of the Banking Act 1945. The plaintiff corporation, the Lord Mayor, Councillors and Citizens of the City of Melbourne, claims against the Commonwealth and the Treasurer of the Commonwealth a declaration that the Banking Act 1945 of the Commonwealth, or alternatively s 48 thereof, is invalid. The defendants' demurrer has been supported by the State of Victoria, intervening by leave, and has been opposed by the other intervening States, South Australia and Western Australia. The plaintiff has abandoned a claim that s 48 of the Act is invalid on the ground that other provisions in the Act are a law imposing taxation within the meaning of s 55 of the Constitution.

The Commonwealth Constitution, s 51(xiii), gives power to the Commonwealth Parliament to make laws for the peace, order and good government of the Commonwealth with respect to "Banking, other than State banking; also State banking extending beyond the limits of the State concerned, the incorporation of banks, and the issue of paper money."

The Commonwealth Parliament in the Banking Act 1945, s 48, has made the following provision -- "48 -- (1) Except with the consent in writing of the Treasurer, a bank shall not conduct any banking business for a State or for any authority of a State, including a local governing authority. Penalty: One thousand pounds. (2) Any consent of the Treasurer under this section may apply to all such business conducted by any particular bank or at a particular office of a bank, or to the business of any particular State or authority conducted by any particular bank or at any particular office of a bank. (3) Until a date fixed by the Treasurer by notice published in the Gazette, this section shall apply only in relation to banking business conducted for a State or for an authority of a State, including a local governing authority, specified by the Treasurer by notice in writing, and, if an office of a bank is specified in the notice, at the office so specified." This section does not "apply with respect to State banking" -- s 5(1).

The statement of claim alleges that the Commonwealth Treasurer has given notice that he proposes on or about 1st August 1947 to specify certain authorities of a State or local governing authorities, including the City of Melbourne, to be authorities in relation to which s 48 of the Act shall apply. As explained in the communication of the Treasurer, "In effect this will mean that as from the date on which this specification was made a private bank will not be able legally to conduct business on behalf of any local governing authority specified in the notice." The term "private bank" is evidently intended to describe banks other than the Commonwealth Bank. A subsequent communication states that the Treasurer is satisfied that the Commonwealth Bank is in a position to provide full banking facilities to the Council. Accordingly, the position is that the Federal Treasurer proposes, by the exercise of the powers conferred by s 48 of the Banking Act, to prohibit the "private banks" from doing any banking business with the City of Melbourne corporation.

The statement of claim alleges that the Council is empowered to make by-laws for the good rule and government of the City of Melbourne and that the Council carries on various undertakings, including the City Abattoirs, the City Cattle Market, the Queen Victoria Market, the Fish Market, public baths, refrigerators, weighbridges electric supply and hydraulic power undertakings; that it manages and controls various properties, parks and the like, and administers the Health Acts within the City of Melbourne, the Uniform Building Regulations, and the licensing and control of hackney carriages. In the course of these activities the plaintiff "receives large sums of money amounting to upwards of £1,800,000 per annum and makes payments (including appropriations to sinking fund, reserves and payments of a like character) amounting to a similar sum." In order to deal with its receipts and payments the plaintiff has established banking accounts, and the banking business of the plaintiff has been done for the plaintiff by the National Bank of Australasia, which is a bank to which s 48 applies.

The council cannot compel the Commonwealth Bank to accept its business. The effect of the specification made by the Treasurer is therefore that the Melbourne City Council will be compelled to do its banking business with the Commonwealth Bank upon terms which are acceptable to the Commonwealth Bank, or, alternatively, that it cannot conduct general banking business with any bank at all, there being no State bank in Victoria doing general banking business.

S 48 of the Banking Act is applicable to the banking business of a State or of any authority of a State, including any local governing authority. The arguments which have been submitted to the Court have been directed to the validity of the provision as applying in the case of a State. It has not been argued that s 48 may be invalid in the case of a State but valid in the case of State authorities, including local governing authorities; that is, no question of severability has been raised on behalf of the defendants. Further, the defendants have disclaimed absence of interest in the plaintiff as a ground of the demurrer, and argument was not heard on that subject. It would, I suggest, have been difficult to raise a doubt upon the matter if s 48, instead of providing that no bank should do business for a local governing authority without a licence from the Treasurer, had provided that no local governing authority should do business with a bank which did not have such a licence. The meaning of the provision is the same in either form.

The contentions for the plaintiff are -- (1) s 48 is not legislation with respect to banking; (2) alternatively, if s 48 is legislation with respect to banking, it is legislation with respect to State banking, and for this reason is expressly excluded from the scope of the Commonwealth power conferred by s 51(xiii) of the Constitution; (3) alternatively, if s 48 is legislation with respect to banking but is not legislation with respect to State banking, it is a law which is directed against an essential State Government activity, namely the custody, control and disposition of government funds; it involves a form of what is called "discrimination," which is forbidden by the Commonwealth Constitution taken as a whole and is therefore invalid.

Before examining these arguments it is desirable to refer to some other provisions of the Banking Act and certain provisions of the Commonwealth Bank Act 1945. Assent was given to both of these Acts on the same day -- 3rd August 1945. The Banking Act contains general provisions relating to authority to carry on banking business, protection of depositors, special accounts, advances and investments, and various subjects which clearly fall within any ordinary definition of "banking." S 4 defines "bank" to mean "a body corporate authorised under Pt II of this Act to carry on banking business in Australia." S 5 provides that nothing in Pt II or V or in ss 48 to 56 of the Act shall apply with respect to State banking. Pt II (ss 6 to 28) contains provisions relating to the carrying on of banking business and Pt V deals with interest rates. S 48 has already been quoted.

Pt II of the Act provides in s 6 that "Subject to this Act, a person other than a body corporate shall not, at any time after the expiration of six months from the commencement of this Part, carry on any banking business in Australia." S 7 requires a body corporate to have an authority in writing granted by the Governor-General to carry on banking business as a condition of carrying on such business. S 8 provides that "The Governor-General shall, within seven days after the commencement of this Part, grant to each body corporate specified in the First Schedule an authority to carry on banking business in Australia." The bodies specified in the First Schedule include, inter alia, the National Bank of Australasia Ltd, with which the City Council has been doing its banking business, and other specified banks which operate under charter or under State Companies and Bank Acts. S 8 also provides for the granting of licences, either unconditionally or subject to conditions specified by the Governor-General, to other bodies corporate to carry on the business of banking. Under s 8(5) it is provided that where an authority is granted under the section subject to conditions, the Governor-General may from time to time vary or revoke any of those conditions or impose additional conditions. The penalty for breach of a condition is £1000 for each day: s 8(6). The Act contains no statement of the conditions which the Governor-General may attach to the grant of an authority; but no question is raised by this case as to either the construction or the validity of this provision.

The Commonwealth Bank Act 1945 carries on the Commonwealth Bank as established under earlier Acts, which are repealed by the Act. S 8 of the Act is as follows -- "It shall be the duty of the Commonwealth Bank, within the limits of its powers, to pursue a monetary and banking policy directed to the greatest advantage of the people of Australia, and to exercise its powers under this Act and the Banking Act 1945 in such a manner as, in the opinion of the Bank, will best contribute to (a) the stability of the currency of Australia; (b) the maintenance of full employment in Australia; and (c) the economic prosperity and welfare of the people of Australia." S 9 of the Act is in the following terms -- "(1) The Bank shall, from time to time, inform the Treasurer of its monetary and banking policy. (2) In the event of any difference of opinion between the Bank and the Government as to whether the monetary and banking policy of the Bank is directed to the greatest advantage of the people of Australia, the Treasurer and the Bank shall endeavour to reach agreement. (3) If the Treasurer and the Bank are unable to reach agreement, the Treasurer may inform the Bank that the Government accepts responsibility for the adoption by the Bank of a policy in accordance with the opinion of the Government and will take such action (if any) within its powers as the Government considers to be necessary by reason of the adoption of that policy. (4) The Bank shall then give effect to that policy." The Act contains provisions relating to central banking functions of the Commonwealth Bank, to the general banking business of the bank and to other departments of the bank. Under s 8 the opinion of the bank determines the manner in which the powers of the bank shall be used for the purpose of promoting the objects specified under the headings (a), (b) and (c), and s 9 enables the Treasurer, in the event of difference of opinion, to control the monetary and banking policy of the bank. I can see no legal ground for objecting to the provision that the powers of the bank are to be utilized for the purpose of attaining the general objectives mentioned, even though the Commonwealth Parliament has no specific power of legislation with respect to certain of those objectives. The Commonwealth Parliament has full power of legislating with respect to the currency of Australia under the Constitution, s 51(xii). But the Commonwealth Parliament has no power to legislate with respect to "the maintenance of full employment in Australia" and "the economic prosperity and welfare of the people of Australia" as subjects in themselves. The Commonwealth Parliament is a Parliament which possesses only "enumerated or selected legislative powers" -- a proposition "as to which this Court has never faltered" (Amalgamated Society of Engineers v Adelaide Steamship Co Ltd (1920) 28 CLR 129, at p 150). As in the United States of America, the State Governments "are the ordinary governments of the country; the federal government is its instrument only for particular purposes" (Woodrow Wilson in Constitutional Government in the United States (1908), pp 183, 184, quoted by Douglas J in New York v United States (1946) 326 US 572, at p 592 (90 Law Ed 326, at p 339)). The Commonwealth Parliament has no general power to make laws for the peace, order and good government of the people of Australia. In s 51, which is the principal source of the legislative powers of the Commonwealth Parliament, power is conferred in the following words -- "The Parliament shall, subject to this Constitution, have power to make laws for the peace, order, and good government of the Commonwealth with respect to -- (i) Trade and commerce with other countries and among the States," and thirty-nine other specific subjects. Other sections confer powers to legislate with respect to other specific subjects. No power is conferred upon the Commonwealth Parliament to make laws with respect to the subjects of full employment in Australia or the economic prosperity and welfare of the people of Australia.

But the Commonwealth Parliament may exercise the powers which it does possess for the purpose of assisting in carrying out a policy which may affect matters which are not directly within its legislative powers. So this Court has held in Osborne v The Commonwealth (1911) 12 CLR 321; Radio Corporation Pty Ltd v The Commonwealth (1938) 59 CLR 170; and see other cases cited in South Australia v The Commonwealth (Uniform Tax Case) (1942) 65 CLR 373, at pp 424, 425. There is no legal obstacle to the use of the Commonwealth Bank as a means of aiding Government policy with respect to employment and economic conditions.

Accordingly, in my opinion, no objection to s 48 of the Banking Act can be founded merely upon the fact that legislative directions are given to the Commonwealth Bank by ss 8 and 9 of the Commonwealth Bank Act. This conclusion, however, simply means that it is within the power of the Commonwealth Parliament to give directions to, or to provide for directions being given, the Commonwealth Bank as to the manner in which it is to exercise its functions. It leaves untouched the separate and distinct question as to the degree of control over the banking business of States and State authorities which can be exercised by the Commonwealth Parliament under the power to legislate with respect to banking.

In the first place, it is contended for the plaintiff that s 48 is not legislation with respect to banking. The argument is that s 48 of the Banking Act simply picks out States and State authorities for subjection to a Commonwealth banking monopoly without any reason which can be described as a reason grounded upon any considerations relating to banking. In other words, s 48, it is said, is really a law with respect to States and State authorities, controlling them in respect of their banking business, and is not a law with respect to banking. Various analogies were suggested. It was put that if s 48 is valid in its present form a provision would be equally valid which prohibited a bank from conducting any banking business with any person except with the consent in writing of the Treasurer. If it were sought to obtain a monopoly for the Commonwealth Bank or for any other bank favoured by the Treasurer in this manner the result would be the exclusion of particular persons or classes of persons from the utilization of banking facilities. Under such a provision all persons professing a particular religion or belonging to a particular political party or following a particular occupation could be prevented from using any banking facilities or could be compelled to deal with a particular bank selected by the Treasurer. Such legislation, it was contended, would not be legislation truly with respect to banking, but would be legislation with respect to the particular classes of persons arbitrarily selected by the Treasurer for disqualification or limitation in respect of banking business.

The defendants replied to this contention by referring to Huddart Parker Ltd v The Commonwealth (1931) 44 CLR 492 and Victorian Stevedoring and General Contracting Co Pty Ltd v Dignan (1931) 46 CLR 73. In these cases it was assumed by the Court that the power of the Commonwealth Parliament to make laws with respect to trade and commerce with other countries and among the States enabled the Parliament (by the Transport Workers Act 1928-1929) to require persons to obtain licences as a condition of engaging in certain operations in such trade and commerce. S 48 of the Banking Act was, it was submitted, a similar provision requiring banks to obtain a licence from the Treasurer as a condition of engaging in banking business for States or State authorities. But there are at least two points of distinction between these cases and the present case. Under the Transport Workers Act every person had a right to obtain a renewal of a licence, and the reasoning upon which this decision was based shows, in my opinion, that every person had a right to obtain a licence (R v Mahony; Ex parte Johnson (1931) 46 CLR 131), he could be deprived of his licence only upon specified grounds which were relevant to the work of transport workers, and there was an appeal to a court against deprivation of licence. S 48 of the Banking Act leaves the grant or refusal of consent entirely to the discretion of the Treasurer. In the second place, s 48 singles out for special treatment the banking business of States and State authorities. There was no feature of this character in the transport workers' cases

Under the section the Treasurer can give consent in the case of one State and refuse it in the case of another State. The consent (subs (2)) may apply to all the banking business for a State or State authority conducted by a particular bank or at a particular office of a bank or to the business of any particular State or authority conducted by any particular bank or at a particular office of a bank. It would appear, therefore (though this matter was not fully argued), that the Treasurer could not limit his consent to some particular kind of banking business -- the consent must be to all or none of the banking business for a State conducted by a particular bank or by a particular bank at a particular place. Thus, it is argued, the character of the banking business done cannot be a ground for giving or withholding a licence, so that the section simply gives the Treasurer an arbitrary discretion, not related to any consideration affecting banking, to prohibit banking operations by a State.

The plaintiff's argument directed to s 48 as it now actually stands is based upon the proposition that there can be no reason of a banking character for making a special provision for States or State authorities. It is contended for the defendants, on the other hand, that the establishment of banks, and in particular of central banks, for the purpose of conducting the banking business of governments and managing public finance is a well-recognized department of banking. This contention, expressed in general terms, appears to me to be well founded.

There are many instances of special relations between governments and banks discharging central banking functions and to a large extent managing governmental financial operations. The Bank of England has been a financial agent of the Government of Great Britain for many years. So also in the United States of America banks have been established as instrumentalities of governments -- per Griffith CJ in Heiner v Scott (1914) 19 CLR 381, at p 392, referring to M'Culloch v Maryland (1819) 17 US 316 (4 Law Ed 579) and Osborn v Bank of United States (1824) 22 US 738 (6 Law Ed 204). But while the contention of the defendants accurately states the position in the case of a unitary constitution where it is not necessary to consider the relation in which States stand to a Commonwealth or Provinces to a Dominion or other federal organism, it cannot be accepted and applied without limitation in the case of a constitution where the States or Provinces are not subordinate to the federal power except in respect of particular matters. It is one thing for a Government to establish a bank for the purpose of doing the banking business of that Government. It is quite a different thing for the Parliament and Government of a Commonwealth to establish a bank and to require the States to do all their business with that bank. The giving of a monopoly of governmental banking business to a particular bank selected by a Government is a not abnormal feature of legislation with respect to banking, but this statement does not cover the case of one Government seeking to select a bank to do all the banking business of another Government, both governments being subject to a federal constitution. Thus, in my opinion, though the argument that s 48 is not legislation with respect to banking should not be accepted, the rejection of this argument still leaves open for consideration the question of the validity of such a provision under a constitution establishing not only a federal Government with specified and limited powers, but also State Governments which, in respect of such powers as they possess under the Constitution, are not subordinate to the federal Parliament or Government. State constitutional powers are, subject to the Commonwealth Constitution, expressly preserved by the Commonwealth Constitution -- ss 106, 107.

The second argument for the plaintiff is that if s 48 is a law with respect to banking it is a law with respect to State banking, and that the express terms of s 51(xiii) of the Constitution prevent the Commonwealth Parliament from making any law with respect to State banking. The contention that s 48 is a law with respect to State banking challenges an assumption upon which s 48 is based. S 5 of the Act expressly provides that s 48 shall not apply with respect to State banking. S 48 expressly relates to the conduct of banking business by a bank (ie a bank authorized under Pt II -- see s 4) for a State. Thus s 48 assumes that the conduct of banking business by a bank (not being a State bank -- s 5) is not "State banking." If this assumption is wrong, s 48 is invalid as dealing with a subject expressly removed by s 51(xiii) of the Constitution from Commonwealth legislative power.

A banker conducts a business of banking with customers. The customer of a banker does his banking business with the banker. That which the banker does as a banker is the business of banking, and that which the customer does as a customer of the banker is also the business of banking. If, under the power to legislate with respect to "banking," there is power to legislate with respect to the conduct of customers of bankers in their capacity as such customers -- and this is a proposition which can hardly be disputed -- then, it is argued for the plaintiff, when it becomes necessary to interpret the words "State banking" the same meaning must be given to "banking" in that phrase. That is to say, the word "banking" in the exception of "State banking" must be construed as including the business of banking if the State establishes and conducts a bank, and also the business done by the State as a customer of any bank. It is contended that it is only upon this view that the same meaning can be given to the word "banking" where it is twice used in the phrase "banking other than State banking."

This argument is plausible, but in my opinion it should not be accepted. It is true that the power to legislate with respect to banking includes a power to legislate with respect to customers of a bank in their capacity as customers of the bank. So also I interpret the exception of "State banking" as excluding any power to make laws not only with respect to banks conducted by States, but also with respect to customers, in their capacity as customers, of banks so conducted. Upon this construction the same meaning is given to the word "banking" in each case where it appears.

"State banking" cannot, in my opinion, be construed as meaning banking operations transacted within a State, ie intra-State banking. Such a construction is, it is true, suggested by the exclusion from federal legislative power of "State banking" together with the express inclusion within that power of "State banking extending beyond the limits of the State concerned." But, if placitum (xiii) were so interpreted the federal power could be exercised only under the provision relating to "State banking extending" & c, because all other banking in Australia would be banking within a State and, upon the construction suggested, would be "State banking" and therefore excluded from federal legislative power. The result would be that the power to make laws with respect to "banking" would have no possible field of operation. The phrase "State banking extending beyond the limits of the State concerned" shows that the words "State banking" are not used to mean banking transactions conducted entirely within a State, because if that were the case the phrase quoted would be self-contradictory.

The words to be construed in the exception "other than State banking" constitute a compound phrase -- "State banking." The question is -- What is the natural signification of those words? In my opinion if the question were asked with respect to a particular country, "Is there any State banking in the country?" the question would be understood to be an inquiry whether the State conducted banks in that country. It would not be understood to be an inquiry whether some State had dealings as a customer with any banks in that country. "State banking" in my opinion refers to banks established and conducted by a State or by an authority established under State law and representing a State. The exception contained in the words "other than State banking" prevents the Commonwealth Parliament from passing any laws with respect to the establishment, management and conduct of such banks by the State, or with respect to the conduct of customers of such banks in their capacity as such customers. But the exception does not prevent the Commonwealth Parliament from making laws with respect to the conduct of customers (including States) of banks other than State banks. Accordingly, in my opinion, it should not be held that s 48 of the Banking Act is invalid as being legislation with respect to State banking.

The third argument of the plaintiff is that s 48 introduces a degree of control of State banking activities which is forbidden by the Federal Constitution. The proposition upon which the plaintiff relies is that the Commonwealth Parliament cannot, even under a legislative power expressly conferred upon it, make a "discriminatory" as distinct from a general, law, which is aimed at or directed against an essential governmental power or function of a State.

It may be difficult to determine in some cases whether a function in fact undertaken by a Government is a governmental function which, under a federal constitution, cannot be controlled by another Government established under the constitution. But there can be no doubt that not only the raising of money by taxation, but also provision for the custody, management and disposition of public revenue moneys are activities which are essential to the very existence of a Government. It is equally essential that a Government should have the power of borrowing money and of providing for the custody and expenditure of loan moneys. In M'Culloch v Maryland (1819) 17 US, at p 422 (4 Law Ed, at p 605), it was held that a bank was "a convenient, a useful, and essential instrument in the prosecution of" the "fiscal operations" of the Government of the United States. The necessity of banking facilities to the Government of a country was emphasized in Weston v City Council of Charleston (1829) 27 US 449, at p 469 (7 Law Ed 481, at p 488), where Marshall CJ referred to the Bank of the United States as "an instrument essential to the fiscal operations of the Government." It would be impossible in practice for a State Government to exist without making provision for the custody and expenditure of public moneys, and it could not do this in modern conditions without using a bank.

The various State Audit Acts contain provisions with respect to the custody and disposition of public moneys which cannot operate according to their terms if s 48 is valid and is put into operation. I mention the following by way of illustration. The Audit Act 1928 (Vict), s 22, requires receivers of public revenue to pay moneys into such bank at such place and in such manner as the Governor (ie the Governor of the State) in Council appoints. For failure to comply with this provision there is a penalty not exceeding £500. S 26 is a similar provision. The Audit Act 1921-1936 (SA), s 18, is as follows -- "The Treasurer may, from time to time, agree with any bank upon such terms and conditions as he may think fit for the receipt, custody, payment, and transmission of public moneys within or without the State, and for advances to be made and for the charges in respect of the same, and for the interest payable by or to the bank upon balances or advances respectively, and generally for the conduct of the banking business of the State." S 19 provides that -- "(1) The Public Account shall be kept in such bank and under such subdivisions (if any) as the Treasurer may, in writing, direct. (2) All moneys paid into any bank to the Public Account shall be deemed to be public moneys and the property of His Majesty, and to be money lent by His Majesty to the bank." The Audit Act 1904 (WA), s 21, is in the same terms as s 18 of the South Australian Act.

If s 48 of the Banking Act is valid, these provisions will become ineffective -- it will be the Commonwealth Treasurer, and not the State Treasurer who will determine in what bank the State moneys will be kept and how they may be withdrawn and for what purposes advances may be made. These matters are plainly subjects of great importance to the States.

It is conceded for the plaintiff that the Commonwealth Parliament may pass general legislation with respect to banking, and that it may specify conditions (relevant to the subject of banking) which must be complied with by banks and by customers of banks. But s 48 is specifically directed to and limited to States as customers of banks. It has the effect of submitting their banking operations to the control of the Commonwealth Bank, which is in turn subject to the control of the Commonwealth Treasurer (Commonwealth Bank Act, ss 8 and 9). If s 48 is valid, a State and a State authority can, in the absence of any available State bank, be compelled to do all its banking business with the Commonwealth Bank. This is stated by the Treasurer to be the object (as it is plainly the consequence) of the notification proposed to be made under s 48. As the Commonwealth Bank is under no legal obligation to accept the business of any State -- either upon any particular terms or at all -- the result is that the operations of a State in paying money into a bank, in drawing out money, and in obtaining advances from a bank, will be subject to Commonwealth control. Such operations are included within banking business as generally conducted: See Commissioners of State Savings Bank of Victoria v Permewan, Wright & Co Ltd (1914) 19 CLR 457. Thus the Commonwealth Bank, acting under direction of the Commonwealth Treasurer, could, so far as legal obligation goes, decline to accept moneys or to allow cheques to be drawn for particular purposes or at all, and could refuse to make advances for particular purposes -- even though the Parliament of the State had appropriated moneys for those purposes. It is contended that legislation which is specifically directed towards -- or, as it is put, against -- a State and State authorities in relation to the custody, control and management of public revenue and loan moneys is legislation which is forbidden by the Constitution as dealing, and especially as dealing in a discriminatory manner, with an essential State governmental power, capacity, and function.

In D'Emden v Pedder (1904) 1 CLR 91, at p 111, it was decided that "when a State attempts to give to its legislative or executive authority an operation which, if valid, would fetter, control, or interfere with, the free exercise of the legislative or executive power of the Commonwealth, the attempt, unless expressly authorized by the Constitution, is to that extent invalid and inoperative." This rule was, in that case, applied in favour of the Commonwealth. In Federated Amalgamated Government Railway and Tramway Service Association v NSW Railway Traffic Employees Association (1906) 4 CLR 488, and other cases, the rule was extended so as to apply in favour of a State as a reciprocal limitation upon Commonwealth legislative power. In the Engineers' Case (1920) 28 CLR 129, this extension of the rule was repudiated. It was held that D'Emden v Pedder (1904) 1 CLR 91, at p 111 really stated in other words the effect of s 109 of the Constitution, which gives supremacy over State legislation to laws made under powers conferred upon the Commonwealth Parliament by the Constitution. It was held, therefore, that it was a fundamental mistake to treat what had been called the "reserved powers" of the States as a basis for implying any limitation upon Commonwealth power. The "reserved powers" of the States can be ascertained only after the extent of Commonwealth power has been determined. A grant of power cannot be construed by first purporting to describe the residue left by the grant. I venture to refer to what I have said on this matter in the Uniform Taxation Case (1942) 65 CLR, at p 422. In the Engineers' Case (1920) 28 CLR 129 it was decided that "laws validly made by authority of the Constitution bind, so far as they purport to do so, the people of every State considered as individuals or as political organisms called States -- in other words, bind both Crown and subjects" (1920) 28 CLR, at p 153. Thus the validity of a Commonwealth law is to be determined by reference to the terms of the Constitution, without applying any presumption that there are certain powers reserved to the States which must not be impaired or interfered with by federal laws.

But this principle does not mean that the States are in the position of subjects of the Commonwealth. The Constitution is based upon and provides for the continued co-existence of Commonwealth and States as separate Governments, each independent of the other within its own sphere. The Engineers' Case (1920) 28 CLR 129 recognizes, in the case of State legislation, a difference between "provisions which apply generally to the whole community without discrimination" and "an act of the State legislature discriminating against Commonwealth officers." The former may be valid and the latter might be invalid (1920) 28 CLR, at pp 156, 157. In Pirrie v McFarlane (1925) 36 CLR 170, there are several references to the same distinction -- see the report (1925) 36 CLR, at pp 184, 216, 217, 229. In West v Commissioner of Taxation (NSW) (1937) 56 CLR 657, at p 682, Dixon J made the observation that the Engineers' Case (1920) 28 CLR 129 "does not appear to deal with or affect the question whether the Commonwealth Parliament is authorized to enact legislation discriminating against the States or their agencies." Dixon J repeated this comment in Essendon Corporation v Criterion Theatres Ltd (above), p 1, at p 23. In West's Case (1937) 56 CLR, at p 687 Evatt J referred to the distinction between general laws and laws "discriminating against" Commonwealth or State officials and said -- "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." See also the Uniform Taxation Case (1942) 65 CLR, at p 431, as to the distinction between general legislation and legislation limited to a particular case. This distinction has been regarded as of significance by the Privy Council in determining questions of the validity of laws: See eg Great West Saddlery Co Ltd v R (1921) 2 AC 91, at p 119.

In the United States of America, as in Australia, the doctrine of immunity of State instrumentalities from federal legislative control has had a chequered career. In M'Culloch v Maryland (1819) 17 US, at p 436 (4 Law Ed, at p 609) it was decided that a State could not impose a tax upon a bank incorporated by Congress for fiscal purposes of the Government because "the states have no power, by taxation or otherwise, to retard, impede, burden or in any manner control the operations of the constitutional laws enacted by congress to carry into execution the powers vested in the general government. This is, we think, the unavoidable consequence of that supremacy which the constitution has declared." In Weston v Charleston (1829) 27 US 449 (7 Law Ed 481) the Supreme Court declared invalid a State Act taxing stock of the United States. The principle stated in these and similar cases was not limited to the power of taxation, but applied to any State attempt to control federal instrumentalities.

In Veazie Bank v Fenno (1869) 75 US 533 (19 Law Ed 482), it was held that Congress could impose a tax upon the circulation of notes issued by a bank chartered under State law of such amount as to prevent the use of such notes. In this case the Court, however, included a saving clause in its reasons for judgment in the following terms -- "It may be admitted that the reserved rights of the States, such as the right to pass laws, to give effect to laws through executive action, to administer justice through the courts, and to employ all necessary agencies for legitimate purposes of State Government, are not proper subjects of the taxing power of Congress" (1869) 75 US, at p 547 (19 Law Ed, at p 487).

In The Collector v Day (1870) 78 US 113 (20 Law Ed 122), the Court applied the general principle of M'Culloch v Maryland (1819) 17 US 316 (4 Law Ed 579) in favour of State instrumentalities and held that the salary of a State judge could not be taxed under Acts of Congress which were general taxation laws. The Court said -- "in respect to the reserved powers, the state is as sovereign and independent as the general government. And if the means and instrumentalities employed by that government to carry into operation the powers granted to it are necessarily, and, for the sake of self-preservation, exempt from taxation by the states, why are not those of the states depending upon their reserved powers, for like reasons, equally exempt from Federal taxation? Their unimpaired existence in the one case is as essential as in the other. It is admitted that there is no express provision in the Constitution that prohibits the general government from taxing the means and instrumentalities of the states, nor is there any prohibiting the states from taxing the means and instrumentalities of that government. In both cases the exemption rests upon necessary implication, and is upheld by the great law of self-preservation; as any government, whose means employed in conducting its operations, if subject to the control of another and distinct government, can exist only at the mercy of that government" (1870) 78 US, at p 127 (20 Law Ed, at pp 126, 127). The court relied upon what I have called the saving clause in Veazie Bank v Fenno (1869) 75 US 533 (19 Law Ed 482). Thus the doctrine of reciprocal immunity of Federal and State instrumentalities was fully established.

Many difficulties arose in the application of this doctrine, and in Graves v New York (1939) 306 US 466 (83 Law Ed 927), The Collector v Day (1870) 78 US 113 (20 Law Ed 122) was overruled. It was held that State income tax could validly be imposed upon Federal officers, the non-discriminatory nature of the law being made the basis of the decision. This rule was applied in other cases, such as James v Dravo Contracting Co (1937) 302 US 134 (82 Law Ed 155) (State taxation of the gross receipts of a contractor with the federal Government); Helvering v Gerhardt (1938) 304 US 405 (82 Law Ed 1427) (federal taxation of salaries of officers of a State Port Authority); Allen v Rents of University System of Georgia (1938) 304 US 439 (82 Law Ed 1448) (federal taxation of a corporation created as a State instrumentality to manage, inter alia, athletic exhibitions); O'Malley v Woodrough (1939) 307 US 277 (83 Law Ed 1289) (State taxation of the salary of a federal judge). In all these cases the law was upheld expressly on the ground that it was non-discriminatory.

The case of New York v United States (1946) 326 US 572 (90 Law Ed 326) corresponds to the Engineers' Case (1920) 28 CLR 129 in Australia. It represents a further endeavour to enunciate a principle which will allow Federal and State Governments to exercise powers, particularly of taxation, over the same persons, without conflict. The judgments, proceeding as they do upon differing grounds, illustrate the difficulties of the problem. It is recognized that a federal system fails unless the federal and State Governments can each carry out their functions as contemplated by the Constitution. The subordination of either to the other is inconsistent with a federal system. But, as in the Engineers' Case (1920) 28 CLR 129, it was held that some federal legislation, even taxation legislation, is applicable to the States. It was held that the State of New York, which owned and operated certain mineral water springs, was not immune from a tax imposed upon mineral waters by an Act of Congress. (Similarly in Australia it has been held that the States are bound to pay federal customs and excise duties (R v Sutton (Wire Netting Case) (1908) 5 CLR 789, and Attorney-General (NSW) v Collector of Customs (Steel Rails Case) (1908) 5 CLR 818).

In New York v United States (1946) 326 US 572 (90 Law Ed 326) all the justices conceded that the powers of the Federal Government cannot be used to destroy State Governments and vice versa, but themajority was of opinion that the extension of governmental activities into many trading and similar activities had made it impracticable to uphold a general rule of immunity in the broad terms of M'Culloch v Maryland (1819) 17 US 316 (4 Law Ed 519) and The Collector v Day (1870) 78 US 113 (20 Law Ed 122). It was admitted, however, by all the justices that there are some activities which are necessarily governmental in character (cf Uniform Taxation Case (1942) 65 CLR 373, at p 423), and that federal and State legislatures are limited in their powers of legislation with respect to agencies of other governments.

In the judgment of Frankfurter and Rutledge JJ it is stated that the fact that "ours is a federal constitutional system . . . carries with it implications regarding the taxing power as in other aspects of government . . . Thus, for Congress to tax State activities while leaving untaxed the same activities pursued by private persons would do violence to the presuppositions derived from the fact that we are a Nation composed of States" (1946) 326 US, at p 575 (90 Law Ed, at p 330). (This statement may be compared with those which I have quoted from Australian cases distinguishing between general legislation which includes States and State agencies and legislation described as discriminating against States and their agencies.) The limitation upon federal legislative power in relation to the States is expressed by these learned justices in the following words -- "There are, of course, State activities and State-owned property that partake of uniqueness from the point of view of intergovernmental relations. These inherently constitute a class by themselves. Only a State can own a Statehouse; only a State can get income by taxing. These could not be included for purposes of federal taxation in any abstract category of taxpayers without taxing the State as a State. But so long as Congress generally taps a source of revenue by whomsoever earned and not uniquely capable of being earned only by a State, the Constitution of the United States does not forbid it merely because its incidence falls also on a State. If Congress desires, it may of course leave untaxed enterprises pursued by States for the public good while it taxes like enterprises organized for private ends. Cf. Springfield Gas and Electric Co v City of Springfield (1921) 257 US 66 (66 Law Ed 131); University of Illinois v United States (1933) 289 US 48, at p 57 (77 Law Ed 1025, at p 1028); Puget Sound Power and Light Co v City of Seattle (1934) 291 US 619 (78 Law Ed 1025). If Congress makes no such differentiation and, as in this case, taxes all vendors of mineral water alike, whether State vendors or private vendors, it simply says, in effect, to a State: 'You may carry out your own notions of social policy in engaging in what is called business, but you must pay your share in having a nation which enables you to pursue your policy'" (1946) 326 US, at p 582 (90 Law Ed, at pp 333, 334). The federal law was upheld because it was not specifically directed against the States, but levied a tax "exacted equally from private persons upon the same subject matter" (1946) 326 US, at p 584 (90 Law Ed, at p 335).

Rutledge J, in a separate opinion, based his decision upon the absence of discrimination against a State. He explained what was meant by discrimination in this connection by saying that he took the limitation against discrimination "to mean that state functions may not be singled out for taxation when others performing them are not taxed or for special burdens when they are," and he added, "Perhaps there are other limitations also" (1946) 326 US, at p 585 (90 Law Ed, at p 335).

Stone CJ, Reed, Murphy and Burton JJ, though agreeing in regarding "as untenable" the distinction which had been drawn in earlier cases between "governmental" and "proprietary" interests of States, said: "Concededly a federal tax discriminating against a State would be an unconstitutional exertion of power over a coexisting sovereignty within the same framework of government" (1946) 326 US, at p 586 (90 Law Ed, at p 336). They proceeded to observe that even a federal tax which was not discriminatory as to the subject matter "may nevertheless so affect the State, merely because it is a State that is being taxed, as to interfere unduly with the State's performance of its sovereign functions of government" (1946) 326 US, at p 587 (90 Law Ed, at p 336). (It may be observed, with respect, that this statement of principle renders it necessary to distinguish between the "sovereign functions of Government" performed by a State and other functions assumed and performed by it.) It is said that a tax "even though non-discriminatory, may be regarded as infringing its sovereignty" (1946) 326 US, at p 587 (90 Law Ed, at p 336) because a sovereign Government was the taxpayer. The test is stated to be whether the tax, even if non-discriminatory, "unduly interferes with the performance of the State's functions of government" (1946) 326 US, at p 588 (90 Law Ed, at p 337). A law which specifically and directly interfered with a State's functions of government would plainly be invalid under this principle -- whether it was or was not a taxation law.

I quote these American decisions, not as authorities upon the construction of the Australian Constitution, but as illustrating in an instructive manner the method by which an endeavour has been made to solve a problem which necessarily arises under a federal constitution. The relevant result which emerges is the same as that which is suggested by the more recent cases in this Court to which reference has been made -- namely that federal laws expressed in general terms may apply to the States (as was shown in the Engineers' Case (1920) 28 CLR 129) but that federal laws which "discriminate" against the States are not laws authorized by the Constitution. Laws "discriminate" against the States if they single out the States for taxation or some other form of control and they will also be invalid if they "unduly interfere" with the performance of what are clearly State functions of government.

I have some difficulty in understanding how "discrimination" in a precise sense can be shown in a law applying only to one person or class of persons in respect of a particular subject matter. Discrimination appears to me to involve differences in the treatment of two or more persons or subjects. Legislation with respect only to one or more persons or with respect only to one or more subjects is not, I suggest with respect, properly described as discriminating against other persons or other subjects simply because it leaves them alone. I refer to what I have said as to the nature of discrimination (but in reference to administrative decisions) in Riverina Transport Pty Ltd v Victoria (1937) 57 CLR 327, at pp 343, 344. In New York v United States (1946) 326 US 572 (90 Law Ed 326) and the other cases to which I have referred in which it has been held that a law may be invalid on the ground of "discrimination," the word "discrimination" is, I think, really used in the sense explained by Douglas J in New York v United States (1946) 326 US 572 (90 Law Ed 326) -- that is, singling out another government and specifically legislating about it.

But why should legislation "discriminating" in this -- or any other sense -- against States or Commonwealth (as the case may be) be held to be invalid? It is true that taxation laws made by the Commonwealth Parliament must not discriminate between States or parts of States -- Constitution, s 51(ii). But this specific provision against discrimination in the case of this class of laws emphasises the absence from the Constitution of any provision prohibiting Federal legislation "discriminating" against the States or prohibiting State legislation "discriminating" against the Commonwealth.

In my opinion the reason why such legislation is invalid is that what is called "discrimination" shows that the legislation is really legislation by the Commonwealth with respect to a State or State functions as such and not with respect to the subject in respect of which it is sought to bind the State -- or, in the case of a State law specifically dealing with and seeking to control Commonwealth functions, that the State Parliament is really endeavouring to make laws with respect to the Commonwealth or Commonwealth functions as such. The Commonwealth Parliament has no power to make laws with respect to State governmental functions as such, and the State Parliaments have no power to make laws with respect to Commonwealth governmental functions as such. It is upon this ground, in my opinion, that what is called "discriminatory" legislation may properly be held to be invalid. I refer to what I said upon this subject in West v Commissioner of Taxation (1937) 56 CLR 657, at pp 668, 669.

Similarly, federal legislation which, though referring to a subject of federal power, is really legislation about what is clearly a State governmental function, may be said to "interfere unduly" with that function and therefore to be invalid. "Undue" interference is a rather vague conception, and an attempt to apply it as a standard for determining the validity of legislation would invite and would certainly produce differences of opinion which would often be due to other than objective considerations. In my opinion the invalidity of a federal law which seeks to control a State governmental function is brought about by the fact that it is in substance a law with respect to a subject as to which the Commonwealth Parliament has no power to make laws. Though there will sometimes be difficulties in applying such a criterion, this is a more satisfactory ground of decision than an opinion that a particular federal "interference" with a State function reaches a degree which is "undue."

The application of these principles in the present case brings about the conclusion that s 48 of the Banking Act is invalid. The section requires the consent of the Treasurer to the conduct of banking business by a bank only in the case of States and State authorities, including local governing authorities. It singles out States and State agencies and creates a rule for them and for no others. It is in substance legislation about States and State authorities. It can fairly be described as being aimed at or directed against States -- and it none the less falls within this disqualifying category because it is also aimed at and directed against what are called "private banks." On this ground, in my opinion, s 48 is invalid.

No reference was made in argument to a provision of the Financial Agreement (see Financial Agreement Act 1928) which, in my opinion, is relevant to the question raised by the demurrer. S 105A(5) of the Constitution provides in relation to such agreements as the Financial Agreement -- "Every such agreement and any such variation thereof shall be binding upon the Commonwealth and the States parties thereto notwithstanding anything contained in this Constitution or the Constitution of the several States or in any law of the Parliament of the Commonwealth or of any State." Thus no legislation, whether federal or State, can be valid if it is inconsistent with the Financial Agreement. The provisions in the Agreement may be varied by another agreement duly made in accordance with s 105A, but not by either federal or State statutes independently of such other agreement.

In the Financial Agreement (see schedule to the Act), Pt I(5), last paragraph, the following provision appears -- "Notwithstanding anything contained in this Agreement, any State may use for temporary purposes any public moneys of the State which are available under the laws of the State, or may, subject to maximum limits (if any) decided upon by the Loan Council from time to time for interest, brokerage, discount, and other charges, borrow money for temporary purposes by way of overdraft, or fixed, special or other deposit, and the provisions of this Agreement other than this paragraph shall not apply to such moneys."

This provision confers upon a State a right to borrow money by way of overdraft subject to certain conditions. These conditions are that the borrowing must be for temporary purposes, and that any decisions of the Loan Council as to maximum limits for interest, brokerage, discount and other charges must be observed.

Lending money on overdraft by a bank and borrowing money by a customer on overdraft from a bank are "banking business" -- Commissioners of State Savings Bank of Victoria v Permewan, Wright & Co Ltd (1914) 19 CLR 457. Such operations are therefore included within the business for which the consent of the Federal Treasurer is required when s 48 of the Banking Act is applied. The right to borrow money on overdraft includes (if, indeed, it does not actually mean) a right to borrow from a bank. The word "overdraft" is the word most commonly used to describe advances by a bank.

An ordinary borrowing by overdraft involves the right to draw cheques up to an agreed limit, reduction of indebtedness from time to time by the payment of moneys into the bank, and the charging of interest by the bank. To borrow money in this way (subject only to the conditions specified in the Agreement) is a right conferred upon a State by the Financial Agreement and no law, federal or State, can limit or restrict it. Federal law cannot validly provide that this right shall not be exercised except with the consent of the Federal Treasurer. S 48 attempts to subject the State in the exercise of this right not even to the control of the Loan Council (which consists of both Commonwealth and State representatives), but to the control of the Federal Treasurer, who is one of the members of the Loan Council. Such a provision must, in my opinion, in view of the terms of s 105A(5) of the Constitution, be held to be invalid in so far as it applies to borrowing on overdraft by any State.

My conclusion that s 48 is invalid is not affected by the facts that s 48 does not apply and, by reason of the terms of the Constitution, s 51(xiii), could not have been made applicable to State banking, and that the Parliament of Victoria could, if it thought proper, establish a State bank with general banking functions. Such a bank would, it is true, be free from Federal control under s 48 or under any law passed by the Commonwealth Parliament under the power to make laws with respect to banking. But it would still be the case that s 48 is a provision aimed at the States in respect of what is undeniably an essential governmental function, that it is "discriminatory" in the sense above explained, and that it makes the exercise of a right assured to the States by the Financial Agreement dependent upon Federal acquiescence or permission. The exclusion of State banking from s 48 thus does not provide a reply to the objections to the validity of the section upon which my conclusion is based.

For the reasons stated, I am of opinion that s 48 is wholly invalid. The demurrer should be overruled and a declaration should be made that s 48 of the Banking Act 1945 is void. The plaintiff claimed a declaration that the whole of the Banking Act is invalid, but it is not necessary, for the purposes of this case, to express an opinion as to the validity of any provision other than s 48.