First National Bank of Southern Africa Ltd v Commissioner for Inland Revenue (343/2000) [2002] ZASCA 2; 2002 (3) SA 375 (SCA); 64 SATC 245 (7 March 2002)

REPORTABLE

CASE NO: 343/2000

IN THE SUPREME COURT OF APPEAL OF SOUTH AFRICA

In the matter between:

FIRST NATIONAL BANK OF SOUTHERN

AFRICA LTD APPELLANT

and

COMMISSIONER FOR INLAND REVENUE RESPONDENT

CORAM: SMALBERGER ADP, HARMS, STREICHER, FARLAM and BRAND JJA

DATE OF HEARING: 18 FEBRUARY 2002

DELIVERY DATE: 7 MARCH 2002

Summary: Income Tax - whether income accruing to the appellant was "from a source within . . . the Republic" in terms of the definition of "gross income" in sec 1 of the Income Tax Act - principles applicable.

___________________________________________________________________

JUDGMENT

SMALBERGER ADP

SMALBERGER ADP:

[1] The central issue in this appeal is whether certain income accruing to the appellant during the 1987 and 1988 tax years was received "from a source within . . . the Republic [of South Africa]" as envisaged in the definition of "gross income" in sec 1 of the Income Tax Act 58 of 1962 (before its amendment by Act 59 of 2000), and hence subject to tax. The appellant contends that it was not; the respondent disputes this.

[2] The appellant carries on business as a commercial bank. The appellant objected to the inclusion of the amounts of R17 633 032,00 and R20 379 947,00 (as finally calculated) in its gross income for the years of assessment ended September 1987 and September 1988 respectively. It claimed that the amounts constituted interest received from a source outside the Republic. It is common cause that if that were the case they would not have been subject to tax as part of the appellant's gross income. The appellant's objection was rejected by the respondent.

[3] The appellant appealed to the Income Tax Special Court ("the Special Court"). Its appeal was upheld. The Special Court set aside the assessments for the years in question and directed that the matter be referred back to the respondent to assess afresh in terms of its judgment. The respondent appealed to the Full Court of the Transvaal Provincial Division of the High Court ("the Court a quo"). The appeal succeeded, the order of the Special Court was set aside and the assessments in question confirmed in so far as they pertained to matters under consideration in the appeal. The present appeal lies with leave of this Court.

[4] As, for reasons which will appear later, the answer to the question whether the source of the interest received by the appellant was within the Republic depends ultimately upon the proper interpretation of the relevant factual matrix giving rise to the receipt of the interest, it is appropriate to commence with a review of the salient facts, which are by and large common cause. In doing so I propose to borrow liberally from the judgments of the Special Court and the Court a quo. The facts were deposed to by the only two witnesses who testified, Mr Evans, who at the relevant time was a manager within the appellant's international banking department, and Mr Howes, its group tax manager.

[5] At all material times the appellant had access to foreign currency borrowed by it, as and when required, from foreign banks interested in lending money to a South African bank. This enabled it to meet the foreign currency borrowing requirements of its clients. Its access to such funds was made possible because of, inter alia, the favourable state of its assets and liabilities, its sound business reputation, the quality of its customer base and the creditworthiness of South Africa as a country. The funds relevant to the present appeal which were on-lent to clients were all derived from foreign borrowings; the appellant made no use of any accumulated foreign funds of its own.

[6] When negotiating a loan, agreement would be reached between the appellant and the foreign bank concerned in respect of the amount and duration of the loan and the interest payable by the appellant. They would further agree, irrespective of the location of the foreign bank, that the loan would be paid to the appellant at the Chase Manhattan Bank in New York in the applicable foreign currency for the credit of the appellant's account at that bank. On maturity of the loan the appellant would repay it, plus interest, by effecting payment, in the currency borrowed, from its account at the Chase Manhattan Bank to the foreign bank's New York account. All the necessary arrangements in the above regard would be made by a dealer employed in the appellant's international banking department in South Africa. The loans would be arranged as clients required them.

[7] The starting point to any international financing transaction would be a request from a client of the appellant (usually a corporate client) for a foreign credit facility to fund either its exports, its imports or its working capital requirements in South Africa. The bulk of the funding was for the latter purpose. For import requirements the funding was probably required overseas; it is not clear where the funding for export requirements was needed. For the purpose of determining what I have identified as the central issue in the present appeal there is no difference in principle between the three situations. The advantage to the appellant's clients in acquiring a foreign facility lay in the lower interest rate payable in respect of such facility compared to that payable on a normal overdraft. From the appellant's perspective, although its profit margin on such foreign financing transactions was small compared to that on rand denominated loans, it constituted a profitable source of business as, because it required less infrastructure, it was cost effective.

[8] The client would request foreign currency in its rand equivalent. This requirement was set to quantify the appellant's maximum exposure. The client obtained the rand equivalent of the foreign currency in South Africa via the appellant's treasury account in New York and was debited locally, in rand, in the books of the branch of the appellant at which it was a customer. If the foreign currency was paid to the client overseas, or paid out overseas on its behalf, its branch account with the appellant was debited with the then rand equivalent. In addition to the capital of the loan, the client was debited in South African rand with the interest charged by the foreign bank together with an added margin on the interest (being the appellant's remuneration or profit) as well as a premium for forward exchange rate cover, if required. The latter served to ensure that the client, who bore the risk of currency fluctuations, would not be detrimentally exposed to such fluctuations.

[9] The loan, being a foreign currency loan, was pegged to the foreign currency in question and had to be repaid to the appellant in New York in that currency on the maturity date. Where the client utilised the appellant's services for this purpose, which was usually the case, payment was effected by converting the client's South African rand into the required foreign currency in the foreign exchange department at the appellant's head office in South Africa and passing the necessary credits by means of appropriate book entries. This resulted in the client's branch account being credited in rand with the amount repaid and the equivalent foreign currency being transferred to the appellant's Chase Manhattan Bank account via its treasury account in New York. If the client made alternative arrangements for repayment into the appellant's Chase Manhattan Bank account of the foreign currency amount that was due, its branch account in South Africa would ultimately be credited with the equivalent amount in rand.

[10] In 1985 a debt standstill was declared by the South African government. It effectively prohibited South African banks from repaying foreign obligations to foreign creditors. The appellant at that time had obligations to foreign banks of some $1.5 billion. This was by and large matched by indebtedness to the appellant by corporate clients in South Africa. Upon the appellant being repaid, it had the choice of either repaying the money in the blocked accounts to the Public Investment Commissioners (which would have rendered it useless to the appellant as an income earning asset) or prevailing upon its overseas creditors to permit it to continue to use such foreign currency for lending to its clients. It successfully followed the latter course, the blocked accounts providing a pool of foreign currency from which it could draw. The debt standstill has no significant bearing on the outcome of the appeal; the appellant's modus operandi remained essentially the same.

[11] The thrust of the appellant's argument (and this has been its case throughout) is that in our law the source of interest is determined by the place where the funds which attracted the interest are made available to the borrower. As this occurred in New York the source of the appellant's interest was located outside the Republic and was therefore excluded from its gross income. The appellant relies for this contention on the decision of this Court in Commissioner for Inland Revenue v Lever Bros and Another 1946 AD 441 ("the Lever Bros case"). It is on this narrow basis that the appellant claims the appeal should succeed on what it refers to as the "source issue".

[12] The legal principles that hold sway in matters involving questions of source were articulated by Corbett CJ in Essential Sterolin Products (Pty) Ltd v Commissioner for Inland Revenue 1993(4) SA 859 (A) ("the Essential Sterolin case") at 870 C to 871 B as follows:

"The legal principles to be applied in determining whether or not an amount was received from a source within the Republic have been stated in a number of decisions of this Court, more particularly in Commissioner for Inland Revenue v Lever Bros and Another 1946 AD 441; Commissioner for Inland Revenue v Epstein 1954 (3) SA 689 (A) ; Commissioner for Inland Revenue v Black 1957 (3) SA 536 (A) . These authorities point out that the Legislature, probably aware of the difficulty of doing so, has not attempted to define the phrase 'source . . . within the Republic' and has left it to Courts to decide on the particular facts of each case whether an amount was or was not received from such a source. As was stated by Watermeyer CJ in the Lever Bros case supra (at 450),

'. . . the source of receipts, received as income, is not the quarter whence they come, but the originating cause of their being received as income, and . . . this originating cause is the work which the taxpayer does to earn them, the quid pro quo which he gives in return for which he receives them. The work which he does may be a business which he carries on, or an enterprise which he undertakes, or an activity in which he engages and it may take the form of personal exertion, mental or physical, or it may take the form of employment of capital either by using it to earn income or by letting its use to someone else. Often the work is some combination of these.'

(See also Epstein's case supra at 698E; Black's case supra at 541.) In a particular case there may be a number of causal factors relevant to the ascertainment of source and, here it would seem, it is appropriate to weigh these factors in order to determine the dominant or main or substantial or real and basic cause of the receipt (Black's case supra at 543A-C). In a number of cases in our Courts reference has been made (in various forms) to the following remarks of Isaacs J delivering the judgment of the High Court in Australia in the case of Nathan v Federal Commissioner of Taxation [1918] HCA 45; (1918) 25 CLR 183 at 189-90:

'The Legislature in using the word "source" meant, not a legal concept, but something which a practical man would regard as a real source of income. . . (T)he ascertainment of the actual source of a given income is a practical, hard matter of fact.'

(See Rhodesia Metals Ltd (In Liquidation) v Commissioner of Taxes 1938 AD 282 at 300; Rhodesian Metals Ltd (in Liquidation) v Commissioner of Taxes 1940 AD 432 (PC) at 436; Lever Bros case supra at 454.)

In applying these general principles, the Courts have adopted certain rules and criteria for locating the source of particular types of accrual or receipt, such as dividends, annuities, director's fees, interest, payment for services, rent, royalties, and so on. None of these would seem to have relevance to the somewhat unusual character of the inability consideration. In seeking the originating cause of this amount one must, in my view, have regard to the factual matrix underlying and giving rise to the agreement in terms of which it became payable and then apply thereto the basic principles outlined above."

No substantial or persuasive challenge was directed against the applicability of these principles in the present appeal. Nor was it suggested that there might be special cases falling beyond the principles enunciated.

[13] In my view the appellant's reliance upon the Lever Bros case is misplaced. The case does not provide authority for the narrow proposition advanced by the appellant. The facts of the Lever Bros case differ materially from the present matter. Those facts, as succinctly reflected in the headnote, were the following: A company registered in South Africa entered into an agreement abroad, the result of which was that it took over an obligation entered into abroad by an overseas company to pay to the taxpayer (Lever Bros), another overseas company, interest upon a large sum of money being the unpaid portion of the purchase price of a large holding of shares in companies registered and carrying on business abroad, the shares remaining overseas pledged to the taxpayer. The interest was paid out of dividends accruing to the South African company abroad on the shares owned by the company and pledged to the taxpayer. In authorising the agreement entered into by the South African company, the Treasury had imposed a condition that no capital or interest should be paid from any funds in South Africa and this condition had been fully observed. It was held by Watermeyer CJ (Davis AJA concurring in a separate judgment, Schreiner JA dissenting) that notwithstanding the fact that the debtor in respect of the loan by Lever Bros resided in South Africa, the interest was not received from a source within the then Union and therefore did not form part of Lever Bros's gross income. The argument that the source of interest is the location of the debt was rejected by Watermeyer CJ.

[14] In the course of his judgment Watermeyer CJ stated (at 449):

"When the question has to be decided whether or not money, received by a taxpayer, is gross income within the meaning of the definition referred to above, two problems arise which have not always been differentiated from one another in decided cases. The first problem is to determine what is the source from which it has been received and when that has been determined, the second problem is to locate it in order to decide whether it is or is not within the Union."

[15] It was when dealing with the first problem that Watermeyer CJ made the statement (at 450) referred to in the passage from the Essential Sterolin case quoted above that the source of receipts was "the originating cause of their being received as income."

[16] Watermeyer CJ went on to add (at 451) that the supply of credit (or, for that matter, money) "is the service which the lender performs for the borrower, in return for which the borrower pays him interest. Consequently, this provision of credit is the originating cause or source of the interest received by the lender . . . . the borrower pays interest . . . . as consideration for the benefits allowed to him by the lender."

[17] Turning to the problem of locating a source of income, Watermeyer CJ opined (also at 451) that "it is obvious that a taxpayer's activities, which are the originating cause of a particular receipt, need not all occur in the same place and may even occur in different countries, and, consequently, after the activities which are the source of the particular 'gross income' have been identified, the problem of locating them may present considerable difficulties. . ." Later in his judgment (at 454) he referred indirectly (and seemingly with approval) to the remarks of Isaacs J in Nathan v Federal Commissioner of Taxation quoted in the Essential Sterolin case at 870 H - I (see para [12] above), to the effect that the ascertainment of the actual source of a given income is a practical, hard matter of fact.

[18] Watermeyer CJ went on (at 455-6) to consider the facts of the case, emphasizing as he did so the absence of considerations pointing to the source of the interest concerned being in South Africa. From his treatment of the evidence it is apparent that he thought it necessary to consider the relevant factual matrix in order to determine where the source of interest was located. This would have been a totally needless exercise if he intended to convey, or for it to be understood, that the sole criterion for determining the location of the source of interest was where the credit (or money, as the case may be) was made available. What the appellant contends for was neither said explicitly by Watermeyer CJ nor does it follow as a matter of necessary implication from Watermeyer CJ's treatment of the question of source. On the contrary, the contention is inconsistent with the tenor of both his judgment and that of Davis AJA. The principles and approach laid down in the Essential Sterolin case are not in any way at variance with the judgment of Watermeyer CJ.

[19] The overall factual situation relevant to the determination of the location of the source of the interest received by the appellant may be summarised as follows. The appellant is a South African institution with an essentially South African client base. The provision of foreign currency to individual South African corporate clients had its origin in a loan facility agreed to in South Africa. The foreign currency was made available in New York and had to be repaid there. The foreign currency was sourced by way of loans from a foreign bank by a foreign exchange dealer employed by the appellant in, and operating from, South Africa. The appellant did not have a branch in New York nor did the client concerned have a separate account with the appellant there. The client was debited in South Africa with the rand equivalent of the available foreign currency. In the majority of cases the foreign currency was brought to South Africa, converted into rand, through the agency of the appellant and its various divisions, none of which operated in isolation, all forming an integral part of the appellant's overall structure. The rand equivalent of the foreign currency was made available to the client, and utilised by it, in South Africa. The add-on margin of interest, which constituted the appellant's income from the overall loan transaction, was debited in rand against the client's branch account in South Africa. While notionally the client was required to repay the foreign currency loan in New York in the currency concerned, in practice the loan was repaid to the appellant (certainly in the majority of cases) in rand in South Africa before it was converted back to the required currency, using the appellant's structures in South Africa, and eventually paid into its Chase Manhattan Bank account.

[20] Apart from the fact that contractually the foreign currency was made available to the borrowing client in New York and had to be repaid there, all the other important factors which caused the interest income to arise (and which constituted the dominant cause of the receipt of the interest) had their origin in South Africa and flowed from the appellant's business activities and operations here. The narrow view taken by the appellant focuses only on where the funds were made available and had to be repaid. It overlooks the need to have regard to the essence of the whole transaction which generated the interest with a view to determining the location of its source. It was conceded on behalf of the appellant that had it borrowed foreign currency in New York, transferred it to South Africa and lent out the rand equivalent here, the source of the interest income generated by the loan would have been South Africa. There is no logical reason why the position should be any different because of the expedient of making the foreign currency available in New York to the client before transferring it to South Africa (and later back to New York) essentially using the same modus operandi. The substance of the underlying income-generating transaction remains the same, even though the means used to achieve the same result may differ. On an overall conspectus of the relevant factual matrix, and applying the principles enunciated in the Essential Sterolin case, the source of the interest, which is the subject of the present appeal, was in my view located in South Africa, and was correctly held by the Court a quo to have been part of the appellant's gross income and subject to tax.

[21] The conclusion reached on the source issue makes it unnecessary to decide the only remaining issue, namely, whether the appellant proved the quantum of the deductions it claimed, or should be allowed a further opportunity to do so. It is very likely that the appellant would have failed on this issue as well. Speaking generally, when a party, on whom the onus rests, is specifically challenged in court to prove its case in relation to quantum, accepts the challenge and undertakes to do so but then fails in that regard, which prima facie is the situation here, the party concerned would normally not be entitled to a second bite at the cherry. However, there is no need to express a firm view on the matter.

[22] In the result the appeal is dismissed with costs, including the costs of two counsel.

_____________________

J W SMALBERGER

ACTING DEPUTY PRESIDENT

HARMS JA ) Concur

STREICHER JA )

FARLAM JA )

BRAND JA )