International Nickel Australia Ltd v Federal Commissioner of Taxation

[1977] HCA 49

HIGH COURT OF AUSTRALIA

Gibbs, Mason and Murphy JJ.

INTERNATIONAL NICKEL AUSTRALIA LTD. v. FEDERAL COMMISSIONER OF TAXATION

(1977) 137 CLR 347

20 September 1977

Income Tax (Cth)

Income Tax (Cth)—Assessable income—Purchase of trading stock—Price payable in pounds sterling—Devaluation of pound—Exchange gain—Whether assessable income—Income Tax Assessment Act 1936 (Cth), s. 25 (1).* * Section 25 (1) of the Income Tax Assessment Act 1936 (Cth) provides that a taxpayer's assessable income shall include "(a) where the taxpayer is a resident—the gross income derived directly or indirectly from all sources whether in or out of Australia ... which is not exempt income."

Decisions

1977, September 20.
The following written judgements were delivered: -
GIBBS J. The appellant, a company incorporated in Victoria, marketed in Australia nickel products which it purchased from two companies in England. Under the contracts pursuant to which those goods were supplied to the appellant the price was payable in sterling. By arrangement with the companies payment was made by the appellant after the goods had been resold; remittances were made not in respect of specific resales, but when substantial sums of money were available to the appellant from payments received by it from its customers. The business was a continuing one: funds were sent to London by the appellant when sufficient money was available and from time to time further supplies of the goods were ordered from London. When the appellant received an invoice in respect of goods supplied it would enter in its books the price in pounds sterling and would enter also the equivalent in Australian dollars converted at an agreed rate of exchange which approximated to the Bank of New South Wales' telegraphic transfer rate. A substantial proportion of the contracts made by the appellant for the resale of the goods to its customers in Australia were for prices written in pounds sterling, but the contracts provided that payments should be made by the customers to the appellant in Australian dollars calculated by conversion at the Bank of New South Wales' telegraphic transfer selling rate of exchange ruling on the day of settlement. At all relevant times the accounts of the appellant were kept, and its income tax returns submitted, on an accruals and not on a receipts basis. Trading stock on hand at the end of each year was valued at cost. Deductions for purchases were claimed in the income tax returns in the amount of Australian dollars which appeared in the appellant's books of account in respect of the purchases of the goods. (at p349)

2. As a result of a devaluation of the pound sterling which took place on the 18th November 1967 the amount required to be remitted by the appellant to the two English companies in respect of goods sold and delivered before that date, when expressed in Australian dollars, was considerably less than the amount at which the purchases had been recorded in the appellant's books of account. On the other hand, as a result of the devaluation the appellant received from the Australian customers to whom it had resold the goods a smaller sum in dollars than it had shown in its accounts. The devaluation also made it necessary to adjust downwards the price of some of the stock on hand. The result was a net exchange gain of $271,232. Of this total, $49,513 was applicable to the year ended 31st December 1967 and $221,719 was applicable to the year ended 31st December 1968. Since all payments in respect of which the exchange gain was made were for goods supplied before 18th November 1967, it is obvious that all such payments made during the year ended 31st December 1968 were for goods which had been taken into stock by the appellant during a previous year. It does not appear from the stated case when the goods for which payment was made during the year ended 31st December 1967 were taken into stock. (at p350)

3. The Commissioner included the amount of $49,513 in the appellant's assessable income for the year ended 31st December 1967, and the amount of $221,719 in the appellant's assessable income for the year ended 31st December 1968. It is against these assessments that the appellant has appealed, unsuccessfully, to a Board of Review. (at p350)

4. It is not suggested that any of the special provisions of the Income Tax Assessment Act 1936 (Cth), as amended ("the Act") justified the assessments. The question is simply whether it was right to include these amounts for the purpose of arriving at the gross income derived by the appellant, within s. 25 (1) (a) of the Act. (at p350)

5. It can hardly be doubted that these exchange gains were trading profits as ordinarily understood. They would be taxable as trading profits under a system in which profits are brought to tax. In Davies (H. M. Inspector of Taxes) v. Shell Co. of China Ltd. Jenkins L.J. said (1951) 32 Tax Cas 133, at p 151 :

"... where a British company in the course of its trade engages in a trading transaction such as the purchase of goods abroad, which involves, as a necessary incident of the transaction itself, the purchase of currency of the foreign country concerned, then any profit resulting from an appreciation or loss resulting from a depreciation of the foreign currency embarked in the transaction as compared with sterling will prima facie be a trading profit or a trading loss for Income Tax purposes as an integral part of the trading transaction."

Counsel for the appellant submitted that the position is quite different in Australia. The Act, they said, taxes income, not profits. They submitted that in the present case nothing has "come in" to the appellant as a result of the variation of the rate of exchange; nothing has been received; all that has happened is that there has been a saving or a reduction in the amount of an outgoing, and a benefit of that kind is not a derivation of income. (at p351)

6. It may be said immediately that even if this argument were correct it would not necessarily follow that the appeal against the assessment for the year ended 31st December 1967 should succeed. If part of the gain of $49,513 was in respect of payments for goods supplied during that income year, that part would properly be brought to account in determining the amount deductible for outgoings in that year. The amount to be deducted in respect of purchases of trading stock made in a particular year is the actual amount in Australian dollars which it will cost the appellant to pay for the purchases, when that amount is known with certainty. The appellant could not deduct an amount simply because, in good faith, it had written it into its books, when it was known before the close of the year that the amount written in was not the true amount. So much in my opinion was decided in Moreau v. Federal Commissioner of Taxation (1926) 39 CLR 65 . In that case goods were purchased from France, at a price payable in francs, and when they arrived into stock the cost price in francs was entered together with a conversion into pounds at the rate then prevailing. However before the price became payable the value of the franc fell. The Commissioner successfully contended that the number of pounds eventually and actually used to pay for the goods, and not the number of pounds originally reckoned at the former rate of conversion, was the true cost price. Isaacs J. said (1926) 39 CLR, at p 70 :

"The taxpayer's error arose in thinking the Commissioner wished to tax the foreign profit made in conversion as an independent source of income. That is not so. The Commissioner really taxes the profits of the business and ignores the conversion as an independent transaction. It certainly enabled the trader to use a less number of pounds sterling to pay for his goods, but the important and only relevant fact in this connection is the actual amount of Australian money used for the purpose."

However the matter could only be dealt with in this way if the variation in the rate of exchange had occurred in the year in which the price was claimed as a deduction. The fact that there had subsequently been a variation in the rate of exchange would not justify re-opening the transactions of a previous year, at least once an assessment had been made. The entry of the price in the books at the rate of exchange then prevailing would not amount to "an error in calculation or a mistake of fact" within s. 170 (3) of the Act simply because the rate of exchange had changed in a later year. This question need not be pursued because, as I have said, the stated case does not show when the goods paid for during 1967 were purchased. (at p352)

7. However the proposition upon which the appellant founded its argument is unsound. It is not correct to say, in the case of a trader who carries on a continuing business, that there can be no income unless there has been a receipt. On the contrary, the income derived by such a trader from his business will almost necessarily be determined upon an earnings or accruals basis. Such a method will be adopted in assessing the income derived from trading if, as will generally be the case, "it is calculated to give a substantially correct reflex of the taxpayer's true income": Commissioner of Taxes (S.A.) v. Executor Trustee and Agency Co. of South Australia Ltd. (Carden's Case) (1938) 63 CLR 108, at p 154 . The judgment of Dixon J. in Carden's Case (1938) 63 CLR, at pp 151-156 , where the relevant principles are expounded, although given in respect of a different statute, has constantly been regarded as applicable to cases arising under the Act: see Arthur Murray (N.S.W.) Pty. Ltd. v. Federal Commissioner of Taxation (1965) 114 CLR 314 ; Henderson v. Federal Commissioner of Taxation (1970) 119 CLR 612 and J. Rowe and Son Pty. Ltd. v. Federal Commissioner of Taxation (1971) 124 CLR 421 . Where the income of the taxpayer is derived from trade, there is not really a difference between the concept of income and that of profit which is significant for the purpose of answering questions arising under the Act, as Walsh J. pointed out in J. Rowe and Son Pty. Ltd. v. Federal Commissioner of Taxation (1971) 124 CLR, at p 433 . (at p352)

8. Where income is assessed on an accruals basis, a purchase of stock in trade is taken into the books at the time when the purchase is made. The purchase price is an outgoing incurred in gaining or producing the taxpayer's assessable income, and is an allowable deduction in the year of the purchase. However if in a subsequent income year, when the time for payment arrives, the taxpayer is compelled, because of a change in the rate of exchange, to expend a larger amount in Australian dollars in discharging his liability for the price than had already been allowed as a deduction, the additional amount which he is required to expend - the exchange loss - is an allowable deduction in that subsequent year. This is well settled: Texas Co. (Australasia) Ltd. v. Federal Commissioner of Taxation (1940) 63 CLR 382 ; Armco (Australia) Pty. Ltd. v. Federal Commissioner of Taxation (1948) 76 CLR 584 ; Caltex Ltd. v. Federal Commissioner of Taxation (1960) 106 CLR 205 . The reason why a deduction is allowable in these circumstances was explained by Dixon J. in Armco (Australia) Pty. Ltd. v. Federal Commissioner of Taxation. The Court there was equally divided in the result, but the difference of opinion was only in relation to the question whether the exchange loss was of a capital nature. Dixon J. said (1948) 76 CLR, at p 618 :

"But it is important to see how that result would be reached. It would arise from the fact that it is a continuing business, depending upon the purchase or manufacture of goods and their resale. The accounts are, of course, made up, not on a basis of receipts and disbursements, but upon the commercial basis of valuation and credit. Nevertheless in such a case actual expenditure in a later accounting period arising from a fortuitous increase in the amount of the liabilities taken into it from the prior period may form a proper debit in the later period, notwithstanding that the item relates to purchases or costs of manufacture included in an earlier accounting period."

After pointing out that purchases are taken into account by reference to the liability incurred, irrespective of when it is discharged, he went on to deal with the case where the amount of the liability has changed, for example by reason of an alteration in the rate of exchange. He said (1948) 76 CLR, at p 618 :

"If the change takes place in a subsequent period and actual payment is then made, is the increase or decrease, as the case may be, to be attributed to the prior period and the net profit or loss reassessed? Obviously not. It is to be taken in as an item belonging to the subsequent period; for the reason that, with continuing trading, when increases beyond the estimates by which assets and liabilities are carried out of one period into the next occur, they must be treated as incidents of the system and they must be regarded as belonging to the period in which they accrue or are realized."

It is clear from this passage that his remarks were not confined to the case where the amount of the liability has increased. It would indeed be anomalous if an exchange loss gave rise to a deduction in the later period and an exchange gain did not increase the amount of the assessable income, but according to this statement both the loss and the gain are treated according to the same principle, as "incidents of the system". (at p354)

9. In Caltex Ltd. v. Federal Commissioner of Taxation (1960) 106 CLR 205 Fullagar J. also expressly recognized that the principle is not confined to losses. He said (1960) 106 CLR, at pp 226-227 :

"If the payment is made in an accounting period later than that in which the purchase was taken into account, the exchange loss or gain will be properly brought into account in the later accounting period. This is because it is only in the later accounting period that the gain has been derived or the loss incurred. The gain derived is assessable income of the later period: the loss incurred is an allowable deduction in the later period."

The same learned justice had earlier reached a similar result in Ballarat Brewing Co. Ltd v. Federal Commissioner of Taxation (1951) 82 CLR 364 . In that case sales were made on terms which allowed the customers a discount and a rebate on the purchase price of goods which they had bought from the taxpayer; disallowance of a discount or rebate was in fact rare. The Commissioner contended that the discounts and the rebates should not be brought into account until they had actually been allowed. This contention was rejected and it was held that the discounts and rebates allowable in respect of sales made during an accounting period should be brought into account during that period. Fullagar J. accepted the submission advanced on behalf of the taxpayer that no difficulty would be created if a discount or rebate treated as allowable in the accounts of one period was subsequently disallowed; the amount treated as allowable in the first period would simply be added to the figure representing sales in the second period (1951) 82 CLR, at pp 366-368 . (at p354)

10. These observations of Dixon J. and Fullagar J., although perhaps not strictly necessary for the actual decision of the cases cited, make it clear that the principle upon which those cases were decided applies to gains as well as to losses. That principle is confined to cases where the taxpayer carries on a continuing business as trader or manufactuerer. In such a case the true income of the taxpayer in the later year cannot be determined unless the exchange loss or gain made in that year is deducted from or added to the income otherwise assessed. Of course the principle does not apply if the loss or gain is referable to capital. It is a truism that in the absence of special statutory authority any loss or gain of a capital nature cannot be taken into account in determining the assessable income. In the present case the payments in connexion with which the gains were made were for goods used by the appellant as its trading stock. Such payments were made for the purpose of discharging liabilities on revenue or income account. The exchange gains were plainly of an income and not of a capital nature. (at p355)

11. In the year ended 31st December 1968 the exchange gain was rightly added as part of the assessable income. In respect of the previous year it is possible that part of the exchange gain could properly have been regarded as a reduction of the deductible outgoings, but since the stated case does not show whether this is so or not it is convenient simply to add the gain as part of the assessable income, for it is immaterial in the result in which of these two ways the gain is dealt with. (at p355)

12. I would answer the questions as follows:
(i) In respect of the year of income ending 31st December 1967 is the sum of $49,513 treated by the respondent as part of the appellant's assessable income, income of the appellant of the year of income ending 31st December 1967 within the meaning of s. 25 (1) of the Income Tax Assessment Act, or at all?
Answer: Yes.

(ii) In respect of the year of income ending 31st December 1968 is the sum of $221,719 treated by the respondent as part of the appellant's assessable income, income of the appellant of the year of income ending 31st December 1968 within the meaning of s. 25 (1) of the Income Tax Assessment Act, or at all?
Answer: Yes.

(iii) Alternatively, should the appellant's outgoings of the year ended 31st December 1967 for purchase of trading stock subject to deduction under s. 51 of the Income Tax Assessment Act, be reduced or adjusted by the amount of such net exchange gain attributable to the reduced liability for purchase, or by any part of such net exchange gain?
Unnecessary to answer.

MASON J. This is a case which I stated pursuant to s. 18 of the Judiciary Act 1903, as amended, in an appeal by the appellant taxpayer under s. 196 (1) of the Income Tax Assessment Act 1936, as amended ("the Act") from a decision of a board of review confirming, by majority, assessments made by the respondent in respect of income derived by the appellant during the years of income ended 31st December 1967 and 31st December 1968, the appellant having adopted, for the purpose of s. 18 of the Act, with the leave of the respondent, an accounting period ending on 31st December of each year. (at p356)

2. The appellant was an Australian resident, having been incorporated in Victoria on 1st July 1965. At all relevant times it has been a public company and a wholly-owned subsidiary of International Nickel Services Ltd., a Canadian corporation, which was in turn a wholly-owned subsidiary of International Nickel Co. of Canada Ltd. ("Inco") which was also a Canadian corporation. As from 1st July 1966 the appellant took over the business of marketing in Australia the nickel products of Inco and its subsidiaries, Inco being a major producer of these products. Thereafter the appellant carried on in Australia the business of the importation and sale of primary nickel and nickel products. For the purpose of obtaining its supplies the appellant dealt with International Nickel Ltd. and Henry Wiggin &Co. Ltd. which were United Kingdom companies and to which I shall refer as "the United Kingdom companies". All deliveries of primary nickel stocks were c.i.f. the Australian port of importation, whereas all deliveries of nickel alloys stocks were f.o.b. the United Kingdom port of despatch. (at p356)

3. By supply agreements made between the appellant and the United Kingdom companies, the price for stock shipped by each company to the appellant pursuant to an order made by it was in each case the prevailing price of the company concerned at the time of despatch from works or warehouse of stocks to fulfil the order (less a discount of eight per cent from International Nickel Ltd. and eight per cent from Henry Wiggin &Co. Ltd.) and was payable in pounds sterling within seven days after payment was due to the appellant for the resale of the products. Upon receipt of an invoice for an order the appellant would enter in the purchase journal of the relevant division particulars relating to the invoice including the amount due on the invoice in pounds sterling, and alongside that actual price there was entered a conversion in Australian dollars at a rate of exchange adopted by the appellant for accounting purposes. At the end of each month the purchases for each division for the month were totalled and the totals were recorded in the relevant purchase journal. The totals of the amounts entered by the appellant in Australian dollars were then entered into the appellant's general ledger to the credit of the particular supplier in accounts which brought in credits to the supplier on various accounts and all remittances to it. The conversion into Australian dollars and the entry of the amount so converted in the purchase journal and the subsequent totalling of the converted amounts and the entry of the total in the general ledger were made by the appellant for the purpose of compliance with s. 20 of the Act. (at p357)

4. The rate of exchange adopted by the appellant for the purposes of making entries in Australian dollars in the purchase journals was calculated at 2.50 Australian dollars to each pound sterling up to 20th November 1967; thereafter it was calculated at 2.1430 Australian dollars to each pound sterling. Both these rates were standard inter-company rates adopted for accounting purposes in the accounts of the Inco group and applied respectively before and after the devaluation of sterling which took place on 18th November 1967. These rates corresponded closely with the rates of exchange quoted in Australia by the Bank of New South Wales for telegraphic transfer between the United Kingdom and Australia at the relevant times. (at p357)

5. By reason of subsequent arrangements made between the appellant and the United Kingdom companies the appellant was allowed extended time for the purpose of paying its indebtedness for stock shipped. Pursuant to these arrangements, which were entered into for reasons of convenience, remittances were to be made in round lump sums at regular intervals, instead of remitting exact sums against specific resales. Payments were made by the appellant after 20th November 1967 in respect of accrued liabilities to pay for stock which had arrived in Australia in 1967 before the devaluation of sterling. Payments were made by way of lump sum remittances pursuant to these arrangements. These remittances were made by telegraphic transfer to the appropriate United Kingdom company when substantial sums of money (in the order of $100,000 or more) were available to the appellant from payments received by it from customers. In respect of each remittance, there was debited to the supplier's ledger account in the appellant's general ledger an amount in Australian dollars calculated at the standard inter-company rate of exchange which the appellant had adopted. The appellant carried on a continuing business in which funds were sent to London when sufficient money was available and in which from time to time further supplies of goods were ordered in London. Payments were not appropriated by the appellant or either of the United Kingdom companies against specific invoices, each supplier's account being in the nature of a running or current account having throughout 1967 and 1968 substantial balances. (at p357)

6. The appellant sold the stocks which it purchased from the United Kingdom companies upon terms which gave to the purchasers extended times for payment, varying from a minimum of two months from the date of delivery to a maximum of approximately seven months from that date. A substantial proportion of the appellant's contracts with its customers were for prices written in pounds sterling, the purchases being invoiced in pounds sterling, but the contracts provided that payment should be made to the appellant in Australian dollars, conversion being calculated at the Bank of New South Wales' (telegraphic transfer) selling rate of exchange ruling on the date of settlement. The goods purchased by the appellant from the United Kingdom companies and sold by it in Australia were trading stock within the meaning of the Act. (at p358)

7. At all relevant times the appellant's accounts were kept and its income tax returns were submitted not on a basis of receipts and disbursements but on the basis of the appellant's profit and loss account compiled so as to comply with the requirements of the Companies Act 1961 (Vict.) adjusted or reconciled so as to conform to the provisions of the Act. Trading stock on hand at the end of each year was valued at cost and the appellant adopted the first in first out method when making the valuation. Deductions for purchases were claimed in income tax returns in the amount of Australian dollars which appeared in the appellant's books of account in respect of those purchases. (at p358)

8. Upon devaluation of the pound sterling on 18th November 1967 the appellant calculated that it would make an exchange gain, being the reduction in the number of Australian dollars required to extinguish its liabilities to the United Kingdom companies at that date. However, the appellant also calculated that it would make exchange losses, being the reduced numbers of Australian dollars which would be received from Australian customers with whom contracts had been made in pounds sterling and from consignment stocks on hand for sale to Australian customers. The gross exchange gain was calculated as $473,817 and the net exchange gain as $271,232. (at p358)

9. A portion of the appellant's accrued liabilities as at the date of devaluation to the United Kingdom companies was paid by the appellant to them before the expiry of the year of income ended 31st December 1967; the balance (being the major part of such liabilities) was paid by the appellant to them in the following financial year. The appellant calculated that it realized exchange gains when in 1967 and 1968 it made payments to the suppliers in settlement of its liabilities at the date of devaluation, and subsequently the appellant calculated that the net exchange gain of $271,232 was realized to the extent of $49,513 during the income year ended 31st December 1967 and to the extent of $221,719 during the income year ended 31st December 1968. (at p359)

10. In its accounts the appellant included the net exchange gain of $271,232 as income in its statement of profit and loss in which the appellant calculated its "profit from all sources before income tax" but it excluded the exchange gains from income when calculating its taxable income for income tax purposes. The respondent included the amount of $271,232 in the assessable income of the appellant for the years of income ended 31st December 1967 and 1968; $49,513 was included in the 1967 assessable income and $221,719 in the 1968 assessable income. (at p359)

11. On these facts the case states the following questions of law:

"(i) In respect of the year of income ending 31st December 1967 is the sum of $49,513 treated by the respondent as part of the appellant's assessable income, income of the appellant of the year of income ending 31st December 1967 within the meaning of s. 25 (1) of the Income Tax Assessment Act, or at all? (ii) In respect of the year of income ending 31st December 1968 is the sum of $221,719 treated by the respondent as part of the appellant's assessable income, income of the appellant of the year of income ending 31st December 1968 within the meaning of s. 25 (1) of the Income Tax Assessment Act, or at all? (iii) Alternatively, should the appellant's outgoings of the year ended 31st December 1967 for purchase of trading stock subject to deduction under s. 51 of the Income Tax Assessment Act, be reduced or adjusted by the amount of such net exchange gain attributable to the reduced liability for purchase, or by any part of such net exchange gain?" (at p359)

12. It was common ground that the two amounts in question are not included within the appellant's assessable income by s. 26 of the Act. Consequently the respondent's assessment rested on s. 25 (1) which includes in a taxpayer's assessable income -

"(a) where the taxpayer is a resident - the gross income derived directly or indirectly from all sources whether in or out of Australia; ...

which is not exempt income." (at p359)

13. The appellant's case is that the two amounts in question constitute a saving in cost, calculated in Australian dollars, of purchasing the number of pounds sterling necessary to discharge a debt incurred in pounds sterling and that such a saving in an outgoing is not income. In brief, the appellant says that a reduction in outgoings is not income, and it calls in aid such well-known statements as those to be found in Tennant v. Smith (1892) AC 150, at p 164 , and Inland Revenue Commissioners v. Blott (1921) 2 AC 171, at pp 192-196 , to support the proposition that the receipt of an advantage or benefit does not constitute, without more, the receipt of income. (at p360)

14. The respondent's answer depends rather heavily on observations made by this Court in earlier cases. They began with Moreau v. Federal Commissioner of Taxation (1926) 39 CLR 65 , which was, so far as it is presently relevant, a decision of Isaacs J. on the Income Tax Assessment Act 1915, as amended. There the taxpayer, who carried on business in Australia, purchased goods in France, the price being payable in francs. On arrival into stock the cost price in francs was entered as Australian currency at the rate of twenty-five francs to the pound. Before the price became payable the value of the franc fell with the result that the amount in pounds required to pay the price in francs was smaller than that entered in the books. Isaacs J. said (1926) 39 CLR, at p 70 :

".... the Commissioner contends that the number of pounds sterling ... used to pay for the goods is the true cost price. I agree with the Commissioner .... The Commissioner really taxes the profits of the business and ignores the conversion as an independent transaction."

In this passage his Honour seems to say that the initial cost is to be recalculated by reference to the amount actually expended in Australian currency to defray the purchase price and that the profit on the transaction or the profits of the business (which constitute the relevant income) are to be ascertained on this footing. However, the facts have not been fully stated in the report and support is not to be found in the later cases for the proposition that the cost of purchases as shown in the accounts for a particular year should be subsequently reopened and recalculated in the light of variations in the exchange rate which occur at some later date. (at p360)

15. Next was Texas Co. (Australasia) Ltd. v. Federal Commissioner of Taxation (1940) 63 CLR 382 , where the taxpayer, an oil company, purchased its petroleum supplies from its parent company in the United States of America the purchase price being payable in U.S. dollars. As the taxpayer was under-capitalized the parent company allowed the price payable to remain outstanding for long periods of time, the taxpayer paying lump sums from time to time when convenient. The taxpayer kept its accounts by entering up the purchases in U.S. dollars and also entering the equivalent in Australian currency by converting the amount in dollars to Australian pounds at the then subsisting rate of exchange. It made up its annual accounts and income tax returns on the basis of purchases and trading stock in Australian currency, conversion being effected at the rate of exchange prevailing at the date of purchase. Before some of the debts were paid there was an alteration in the exchange rate resulting in a devaluation of the Australian pound as against the U.S. dollar. Consequently the cost in Australian pounds for discharging the balance of the running account exceeded the cost as formerly recorded in the books. The taxpayer claimed the difference as a deduction in the year in which it made the payment. The Commissioner asserted, first, that the initial entries in Australian dollars should have been adjusted to bring them into line with the exchange rate as altered, and, secondly, that the loan account had been capitalized, with the result that the loss was a loss of capital. (at p361)

16. Latham C.J. said (1940) 63 CLR, at p 427 :

"It has been suggested that a claim for the deduction of increased expenditure resulting from adverse exchange movements where the payment is made in respect of goods acquired in an earlier year cannot be deducted in a later year, because such a deduction would involve a substitution of a cash basis for a credit basis in assessing the company. In my opinion, however, the suggested difficulty disappears if the increased outlay required in a subsequent year to discharge the constant (in this case, dollar) debt is regarded not necessarily as payment of the price of the goods, but as a necessary outgoing made in the normal course of the continuance and maintenance of the business as an enterprise conducted for the purpose of profit. The Income Tax Assessment Act, s. 23 (1)(a), permits the deduction of outgoings (subject to exceptions) actually incurred in gaining or producing the assessable income. Although assessments to income tax are made for separate years, it is established that an expenditure made in one year which does not produce its income-gaining effect till a subsequent year may nevertheless be deducted in the year in which it is made, and so also an outgoing which arises out of income-gaining activities of a prior year may be deducted in a subsequent year when it is actually made ... "

Later his Honour said (1940) 63 CLR, at p 428 :

"Such expenditure of Australian pounds is an ordinary business expenditure, and the taxpayer is entitled to claim as a deduction the actual outgoing which he makes in order to discharge his normal business debts for stock-in-trade and the like: Cf. Moreau v. Federal Commissioner of Taxation." (1926) 39 CLR, at p 70. (at p362)

17. Dixon J. (1940) 63 CLR, at pp 464-466 expressed a similar view, a view which his Honour was to reiterate in subsequent cases and to which I shall refer in more detail later. His Honour rejected the argument that the outgoing was wholly of a capital nature. Starke J. was against the Commissioner on both questions. (at p362)

18. The third case is Armco (Australia) Pty. Ltd. v. Federal Commissioner of Taxation (1948) 76 CLR 584 where the facts were similar to the Texas Case (1940) 63 CLR 382 , save that here the balance of the accounts was capitalized, so the Court held, and used by the taxpayer to purchase capital assets in Australia. Subsequently the exchange rate moved adversely to the Australian pound with the result that a larger number of Australian pounds was required to discharge the debt owing in U.S. dollars to the parent company. The taxpayer claimed the difference as a deduction in the year in which the liability was discharged. The Court on appeal being evenly divided, the judgment of Williams J. prevailed, his Honour concluding that the outgoing was one of a capital nature, a view with which Dixon and McTiernan JJ. subsequently agreed. Latham C.J. and Starke J. were of the contrary opinion. Dixon J. said (1948) 76 CLR, at p 618 :

"It would arise from the fact that it is a continuing business, depending upon the purchase or manufacture of goods and their resale. The accounts are, of course, made up, not on a basis of receipts and disbursements, but upon the commercial basis of valuation and credit. Nevertheless in such a case actual expenditure in a later accounting period arising from a fortuitous increase in the amount of the liabilities taken into it from the prior period may form a proper debit in the later period, notwithstanding that the item relates to purchases or costs of manufacture included in an earlier accounting period. The comparison made between the beginning and end of an accounting period means that stock in trade and purchases are taken into account not by reference to what is actually paid for them, but according to the value assigned to the one and the liability incurred in acquiring the other. That is done, of course, entirely independently of the period within which the actual disbursement of money is made to discharge the liability. If, as is commonly the case, the amount of the liability is fixed and incapable of subsequent variation, the disbursement itself whenever made could never matter for the purpose of computing profit or loss in that or any subsequent period. But if for any reason the amount is capable of changing, as is the case when the indebtedness is in a foreign currency and the rate of exchange may alter, a further question arises. If the change takes place in a subsequent period and actual payment is then made, is the increase or decrease, as the case may be, to be attributed to the prior period and the net profit or loss reassessed? Obviosly not. It is to be taken in as an item belonging to the subsequent period; for the reason that, with continuing trading, when increases beyond the estimates by which assets and liabilities are carried out of one period into the next occur, they must be treated as incidents of the system and they must be regarded as belonging to the period in which they accrue or are realized."

These remarks did not touch the actual point on which there was a difference of opinion within the Court. His Honour distinguished the Texas Case (1940) 63 CLR 382 on the ground that there the purpose of allowing the fund to accumulate was that it should be employed in trading, not taken out of the business and invested, whereas in the Armco Case (1948) 76 CLR 584 the parties agreed upon a course involving the creation of a fund to be used outside the business. (at p363)

19. The fourth case was Caltex Ltd. v. Federal Commissioner of Taxation (1960) 106 CLR 205 , where an oil company controlled by non-residents incurred debts in U.S. dollars for trading stock to the United States supplier. In this period there were variations in the exchange rate, mainly adverse to Australia, which were reflected in monthly adjustments to the supplier's account in the taxpayer's books. The balance of the current account was paid by the taxpayer in 1936 by cheque drawn on its account at a New York bank, a new United States supplier having advanced the taxpayer the requisite amount in U.S. dollars. The Australian equivalent of the U.S. dollars thus paid was substantially higher than either the cost at which its supplies had in Australian currency originally been taken into account or the amount at which by reason of the adjustments in its books that cost then stood in the books. The taxpayer claimed the difference as a loss or outgoing under s. 51 (1). By majority it was held that there had been no loss or outgoing because the taxpayer had merely substituted one creditor for another and had not suffered any loss of Australian pounds. Dixon J., after reiterating the observations in the Armco Case (1948) 76 CLR, at p 618 which I have already quoted, said (1960) 106 CLR, at pp 219-220 :

"The Australian law of income taxation does not enable traders who incur liabilities in money of other countries in purchasing stock in trade to treat a movement of exchange against this country as establishing a loss deductible as such in the year in which the movement takes place simply because a liability incurred for stock in a prior year remains undischarged. The movement of exchange may have other consequences in a current assessment ... But if there is nothing amounting to a realization or definite accrual or establishing of the loss or outgoing in the manner described in Flood's Case (1953) 88 CLR, at p 506 , as one to which the taxpayer is at least definitively committed, as an outlay ascertained and unavoidable s.51 (1) will not avail him ... But the purpose of introducing the analogy of goods is to illustrate the differences between the changes of value expressed in terms of Australian money and the actual realization of a loss or the incurring of a loss or outgoing. It is one thing to deal in values without realization for the purpose of s. 31; another for the purposes of s. 51 (1). Values are ever changeable: so is a rate of exchange." (at p364)

20. Fullagar J. expressed a substantially similar view. In one respect his Honour went further than the earlier observations which I have already quoted. His Honour considered specifically the converse case, the problem which now arises, that is whether a saving in the amount of Australian currency required to meet a liability in a foreign currency, brought about by a favourable variation in the rate of exchange, is income. His Honour said (1960) 106 CLR, at pp 226-227 :

"It is not, of course, to be denied that variations in a rate of exchange between two countries may involve, as a consequence to a trader, either an assessable profit or a deductible loss. The position has been twice fully explained by Dixon C.J. - once in the Texas Co. 's Case (1940) 63 CLR, at pp 464-466 , and again in Armco (Australia) Pty. Ltd. v. Federal Commissioner of Taxation (1948) 76 CLR at p 618.down An Australian taxpayer who imports goods for which payment must be made in dollars in New York must (whether or not he keeps his accounts also in dollars) record his purchases in his books in the year of purchase in terms of Australian pounds, and for this purpose he will apply the rate of exchange prevailing at the time of purchase. Let it be assumed that that rate is $4 to the pound (A). On the day when he comes, in the same or a later accounting period, to pay for a purchase to recorded, the rate of exchange may have moved to $3 to the pound (A) or $5 to the pound (A). The taxpayer will have to expend, in order to provide the necessary dollars in New York, either more or less in pounds (A) than the amount at which the goods were entered in his books. In the former event there is an 'exchange loss': in the latter event there is an 'exchange gain'. If the payment is made in an accounting period later than that in which the purchase was taken into account, the exchange loss or gain will be properly brought into account in the later accounting period. This is because it is only in the later accounting period that the gain has been 'derived' or the loss 'incurred'. The gain derived is assessable income of the later period: the loss incurred is an allowable deduction in the later period." (at p365)

21. Although it was suggested that Fullagar J.'s observations may have been directed to the payment of a rebate following an exchange variation which was favourable to a taxpayer, the language gives no support to this notion. Rather does it indicate that his Honour was discussing the converse situation to that which actually arose for decision in the Caltex Case (1960) 106 CLR 205 . (at p365)

22. The same view of the question now under consideration was taken by the Supreme Court of Canada in Eli Lilly and Co. (Canada) Ltd. v. Minister of National Revenue (1955) 55 DTC 1139 . There after the Canadian taxpayer had incurred a debt payable in U.S. dollars for trading stock to its American parent company the exchange rate moved favourably to Canada. The taxpayer then paid the debt, claiming that the saving in Canadian dollars was a capital profit. By majority the Court decided against the taxpayer, holding that it was an established practice "that whether there be a loss or a gain in respect to the item of foreign exchange it should be taken into account as a trading loss or profit in the computation of income tax" (1955) 55 DTC, at p 1142 . Estey J., who delivered the majority judgment, cited with approval the principles enunciated by Latham C.J. and Dixon J. in the Texas Case (1940) 63 CLR 382 . (at p365)

23. The appellant criticized on several grounds the approach taken in these cases, in particular the view expressed by Fullagar J. and by the Canadian Supreme Court. First, it was pointed out that the Moreau Case (1926) 39 CLR 65 and the Texas Case (1940) 63 CLR 382 were not decisions on the Income Tax Assessment Act 1936 which differed materially from the Income Tax Assessment Acts of 1915 and 1922. The fact that there were differences may be conceded, though the principal difference was between the 1915 Act on the one hand and its successors on the other hand, but we are left with the Armco Case (1948) 76 CLR 584 and the Caltex Case (1960) 106 CLR 205 where the Court was speaking in the context of the 1936 Act. In so doing it was enunciating a general concept of income, a notion that emerges from the judgment of the Supreme Court of Canada in Eli Lilly and Co. (Canada) Ltd. v. Minister of National Revenue (1955) 55 DTC 1139 . (at p366)

24. Secondly, the appellant placed great emphasis on the scheme of taxation provided for by the Income Tax Assessment Act 1915 and other schemes of taxation such as the United Kingdom Income Tax Acts which levy taxation on the profits of a business and the different approach taken in the Income Tax Assessment Act 1936 which, it was pointed out, calls for receipts to be identified as assessable income before any consequential figure for taxable income can be ascertained after allowable deductions have been subtracted from the assessable income, with the result that a receipt must be identified as assessable income before it can be reflected in the taxable income. The existence of this distinction may be acknowledged without, in my opinion, doing any injury to the respondent's case. The question remains one of determining whether the amount in question is income within s. 25 (1) of the present Act. (at p366)

25. On this question it seems to me that the observations in the earlier cases in this Court, in particular the views expressed by Fullagar J. in the Caltex Case and the decision of the Supreme Court of Canada in the Eli Lilly Case are persuasive. They give expression, as I have said, to a general concept of income which includes within its embrace a reduction in the amount of an outgoing on revenue account. True it is, as the appellant has contended, that the point now put forward was not argued in the earlier cases which dealt with losses not gains. At the same time, their Honours could not have failed to take into account the case of a reduction in outgoings - to treat that case discordantly would have been to throw the general concept into question. (at p366)

26. The approach taken in the Texas, Armco and Caltex Cases conforms to the principle enunciated in Commissioner of Taxes (S.A.) v. Executor Trustee and Agency Co. of South Australia Ltd. (Carden's Case) (1938) 63 CLR 108, at pp 154-155 where Dixon J. said:

"Unless in the statute itself some definite direction is discoverable, I think that the admissibility of the method which in fact has been pursued must depend upon its actual appropriateness. In other words, the inquiry should be whether in the circumstances of the case it is calculated to give a substantially correct reflex of the taxpayer's true income. ... Speaking generally, in the assessment of income the object is to discover what gains have during the period of account come home to the taxpayer in a realized or immediately realizable form."

And his Honour continued:

"The reasons which underlie the practice of estimating for taxation purposes the income from trade or manufacture by means of a commercial profit and loss account consist in the impracticability of computing income in any other way and in the adoption for fiscal purposes of recognized commercial principles." (at p367)

27. The appellant insisted that these observations must be understood in a qualified sense by reason of what was said in Arthur Murray (N.S.W.) Pty. Ltd. v. Federal Commissioner of Taxation (1965) 114 CLR 314, at p 318 . But in my opinion it is a mistake to treat what has been said in the cases which have been discussed as resting only on established book-keeping practice. The observations which have been made not only record the book-keeping practice which has been followed in many cases, in itself a relevant consideration because it tends to throw light on whether income has been earned or gained, but they also indicate a legal concept of income as reflected in the accounts of a continuing business. (at p367)

28. It was in an endeavour to overthrow this view that the appellant sought to develop an argument designed to show that proper accounting or book-keeping practice would require the establishment of a special capital account or reserve into which losses and gains on exchange variations would be carried, removing them from the sphere of revenue and stamping them with the character of windfall losses or gains. That there is something to be said for this view I would concede, but to my mind it more accords with the legal conception of income and with sound book-keeping practice to treat gains and losses arising from fluctuations in the exchange rate as revenue items, at least when the gains and losses arise from liabilities for the payment of trading stock purchases in a continueing business, exchange fluctuations being a common hazard of international trade frequently and ordinarily encountered by businesses engaged in that trade. (at p367)

29. Finally, it was submitted that the appellant's case was supported by the decisions in British Mexican Petroleum Co. Ltd. v. Inland Revenue Commissioners (1932) 16 Tax Cas 570 ; St. Lucia Usines and Estates Co. Ltd. v. Colonial Treasurer of St. Lucia (1924) AC 508 and Ballarat Brewing Co. Ltd. v. Federal Commissioner of Taxation (1951) 82 CLR 364 . The British Mexican Petroleum Case certainly resulted in a reduction or elimination of outgoings but this was the consequence of an unusual transaction whereby a liability was extinguished by a release on the part of the creditor. It therefore does not touch this case where the liability is reduced by supervening circumstances ordinarily encountered in trade. (at p368)

30. The decision in the St. Lucia Case is far removed from the question which now arises. There the question was whether interest accrued but not paid in the accounting period was income arising or accruing in that period. It was held that "There must be a coming in to satisfy the word 'income'" (1924) AC, at p 512 , but this was a comment made in the context of whether a mere entitlement without receipt constituted income arising or accruing. (at p368)

31. The third case in the trilogy, Ballarat Brewing Co. Ltd. v. Federal Commissioner of Taxation (1951) 82 CLR 364 , a decision of Fullagar J., is likewise of no assistance to the appellant. There it was held that in computing the taxpayer's true gross profit for an accounting period all discounts and rebates allowed or allowable in respect of sales made during that period should be brought to account, the discounts and rebates being so seldom disallowed that any discrepancy resulting from disallowance would be negligible, and that if a discount or rebate treated as allowable in the accounts for one period was disallowed in the succeeding period and the gross price actually received, the amount treated as allowable in the first period should be added to the figure representing sales in the second period. The decision is consistent with the principle which I would apply to the present case. (at p368)

32. In the result I would answer the questions asked - (i) Yes. (ii) Yes. (iii) Does not arise. (at p368)

MURPHY J. The question is whether savings made in an income year because debts payable by the appellant taxpayer company were reduced by movements in currency exchange are assessable income (within s. 6 of the Income Tax Assessment Act 1936, as amended). Section 26 or other special provisions of the Act do not require the savings to be brought into account. They are assessable only if they are part of the gross income within s. 25 (1). The majority of the Board of Review considered that the savings represented a gain, that the gain was profit, that the profit was income, and that the income was assessable income. The appellant's case was essentially that "a reduction of outgoings is not income" (see ss. 20 and 21). (at p368)

2. The previous Australian Income Tax Acts, Nos. 34 of 1915 and 37 of 1922, were constructed similarly to the English Income Tax Acts in which income is closely tied to notions of profit (see the definition of what is to be included in taxable income set out in s. 16 of the 1922 Act). The scheme of the Australian 1936 Act is quite different from that of the pre-1936 Acts and the English Acts (see Federal Commissioner of Taxation v. Gordon, per Dixon J. (1930) 43 CLR 456, at p 461 ; per Rich J. (1930) 43 CLR, at p 469 ; American Thread Co. v. Federal Commissioner of Taxation, per Dixon J. (1946) 73 CLR 643, at p 658 ; Federal Commissioner of Taxation v. James Flood Pty. Ltd. (1953) 88 CLR 492 ; Federal Commissioner of Taxation v. W. E. Fuller Pty. Ltd., per Dixon C.J. (1959) 101 CLR 403, at pp 407-409 ; Gibb v. Federal Commissioner of Taxation (1966) 118 CLR 628 ; Scott v. Federal Commissioner of Taxation (1966) 117 CLR 514, at pp 523-524 ; Levy v. Federal Commissioner of Taxation, per Kitto J. (1961) 106 CLR 448, at p 460 ; Arthur Murray (N.S.W.) Pty. Ltd. v. Federal Commissioner of Taxation (1965) 114 CLR 314, at pp 318-320 . (at p369)

3. The appellant contended that, even in systems which tax profits of a trade (that is, a pre-1936 Australian or an English scheme), there must be something coming in to the taxpayer capable of a valuation. It relied on Tennant v. Smith (1892) AC 150, at p 164 : "a person is chargeable for income tax under Schedule D ... not on what saves his pocket, but on what goes into his pocket" (Lord Macnaghten); see also Commissioners of Inland Revenue v. Blott (1921) 2 AC 171 . It contended that in cases apparently to the contrary the question was whether the admitted receipt was income rather than capital. Whether a saving on an outgoing was income has not been decided (see Case H13, per Mr. O'Neill (1976) 76 ATC, at p 78 ). (at p369)

4. In Texas Co. (Australasia) Ltd. v. Federal Commissioner of Taxation (1940) 63 CLR 382 (which involved the 1922 Act), the taxpayer claimed deductions for the increased number of dollars (in year of payment) over the amount estimated in previous accounts on which income tax was assessed. Latham C.J. said (1940) 63 CLR, at pp 426-427 :

"If, owing to the movement of exchange, goods which in earlier years could have been paid for in dollars by an expenditure of say 1,000 pounds Australian were in fact paid for in a subsequent year by a necessarily increased amount of 1,100 pounds Australian, it is urged that it would be difficult to regard the additional sum of 100 pounds as deductible in the later year if it is regarded as the price of the goods bought in the earlier year because, ex hypothesi, the full deduction of what was in the earlier year taken to be the price, namely 1,000 pounds, had already been made."

and (1940) 63 CLR at p 427 :

"Although assessments to income tax are made for separate years, it is established that an expenditure made in one year which does not produce its income-gaining effect till a subsequent year may nevertheless be deducted in the year in which it is made, and so also an outgoing which arises out of income-gaining activities of a prior year may be deducted in a subsequent year when it is actually made."

Dixon J. said (1940) 63 CLR, at p 465 : "...discrepancies between the liabilities carried into the period and the cost of defraying them must come into the comparison as an actual reduction or increase of the profit or loss otherwise produced by the comparison ..." Rich and McTiernan JJ. concurred with him. (at p370)

5. Armco (Australia) Pty. Ltd. v. Federal Commissioner of Taxation (1948) 76 CLR 584 was decided on the 1936 Act. Although the case was mostly directed to the point of whether the item was capital or income, Dixon J. virtually repeated his remarks in the Texas Co. Case (1948) 76 CLR, at p 618 . In Caltex Ltd. v. Federal Commissioner of Taxation Dixon C.J. stated (1960) 106 CLR 205, at pp 219-220 :

"The Australian law of income taxation does not enable traders who incur liabilities in money of other countries in purchasing stock in trade to treat a movement of exchange against this country as establishing a loss deductible as such in the year in which the movement takes place simply because a liability incurred for stock in a prior year remains undischarged. The movement of exchange may have other consequences in a current assessment; it may, for example, indirectly affect the value of trading stock for the purpose of s. 31. But if there is nothing amounting to a realization or definite accrual or establishing of the loss or outgoing in the manner described in Flood's Case (1953) 88 CLR, at p 506 as one to which the taxpayer is at least definitively committed, as an outlay ascertained and unavoidable, s. 51 (1) will not avail him."

This statement confirmed that the proper course is not to falsify previous accounts, but the adjustment is to operate when and only when the outgoing is crystallized as by payment. Similarly, Fullagar J. said (1960) 106 CLR, at pp 226-227 :

"It is not, of course, to be denied that variations in a rate of exchange between two countries may involve, as a consequence to a trader, either an assessable profit or a deductible loss. The position has been twice fully explained by Dixon C.J. - once in the Texas Co.'s Case (1940) 63 CLR, at pp 464-466 and again in Armco (Australia) Pty. Ltd. v. Federal Commissioner of Taxation (1948) 76 CLR at p 618. ... If the payment is made in an accounting period later than that in which the purchase was taken into account, the exchange loss or gain will be properly brought into account in the later accounting period. This is because it is only in the later accounting period that the gain has been derived or the loss incurred. The gain derived is assessable income of the later period: the loss incurred is an allowable deduction in the later period."

Although the sole question was whether the outgoing was crystallized, the majority reached their decision against a background of the line of cases referred to in Texas (1940) 63 CLR 382 and that case. Although not necessary for the decision, Fullagar J.'s statement is precisely in point for this case. (at p371)

6. Lord Thankerton's statement in British Mexican Petroleum Co. Ltd. v. Jackson (1932) 16 Tax Cas 570, at p 592 is contrary to the views expressed in the above cases. However, in that case there was a release which seems more like a gift than a reduction in an outgoing. This view of it seems to have been taken in the Supreme Court of Canada (see Eli Lilly and Co. (Canada) Ltd. v. Minister of National Revenue (1955) 55 DTC 1139 which referred to the Texas Case and took the same approach as this Court). (at p371)

7. The appellant contended that the reason why the reduction occurs is irrelevant, so that the case of a reduction through exchange rate variation should be dealt with in exactly the same way as a saving by a release as in the British Mexican Case. Its arguments, if accepted, would produce some very strange results. If the taxpayer has to pay more on an exchange variation, this increase would be allowable, but if it pays less, the saving is not assessable. Exchange rate variations are an everyday event of international commerce and many countries, including Australia, have covering arrangements to protect their traders from losses. If the taxpayer owed money to persons in different countries, a variation might increase one debt and reduce another. The company would claim to deduct the losses but not to have the savings assessable. In a period of continuous trading with postponed payment, when there were frequent exchange fluctuations, this could mean that even where the taxpayer had a considerable net gain with a particular customer from the currency fluctuations, it would be able to deduct the losses but not be assessable in respect of the greater amount of its savings. This would be absurd. (at p371)

8. The correct approach is this. Where assessable income or allowable expenditure is subject to an exchange variation occurring in the same year in which the income or expenditure is taken into account, the income or expenditure is adjusted to reflect the actual income or expenditure. If variation occurs in a later year then if the sum already taken into account is assessable income, any increase caused by exchange variation is assessable income and any decrease is allowable expenditure. If the sum already taken into account is allowable expenditure, any later increase is also allowable expenditure and any decrease is assessable income. (at p372)

9. As the debts payable by the appellant were allowable expenditure, the decrease in that expenditure was therefore assessable income. (at p372)

10. The answer to the first two questions should be yes. (at p372)

Orders

Order that the questions referred by the case stated be answered as follows: (i) In respect of the year of income ending 31st December 1967 is the sum of $49,513 treated by the respondent as part of the appellant's assessable income, income of the appellant of the year of income ending 31st December 1967 within the meaning of s. 25 (1) of the Income Tax Assessment Act, or at all?
Answer: Yes.
(ii) In respect of the year of income ending 31st December 1968 is the sum of $221,719 treated by the respondent as part of the appellant's assessable income, income of the appellant of the year of income ending 31st December 1968 within the meaning of s. 25 (1) of the Income Tax Assessment Act, or at all?
Answer: Yes.
(iii) Alternatively, should the appellant's outgoings of the year ended 31st December 1967 for purchase of trading stock subject to deduction under s. 51 of the Income Tax Assessment Act, be reduced or adjusted by the amount of such net exchange gain attributable to the reduced liability for purchase, or by any part of such net exchange gain?
Unnecessary to answer.

Order that the respondent pay the appellant' costs of the reference to the Full Court.