NBS Bank Ltd v Cape Produce Company Pty Ltd and Others (281/99) [2001] ZASCA 107; [2002] 2 All SA 262 (A); 2002 (1) SA 396 (SCA) (28 September 2001)

IN THE SUPREME COURT OF APPEAL OF SOUTH AFRICA

CASE NO. 281/99

In the matter between

NBS Bank Limited Appellant

and

Cape Produce Company Pty Ltd First Respondent

Cape Produce Company Port Elizabeth Pty Ltd. Second Respondent

Benjy Lapiner Children’s Trust Third Respondent

Ronelle Lapiner Children’s Trust Fourth RespondentTrevor Bradley Fifth RespondentBefore: NIENABER, MARAIS, SCHUTZ, NAVSA JJA and FRONEMAN AJA

Heard: 20 AUGUST 2001

Delivered: 28 SEPTEMBER 2001

Ostensible authority of a bank manager – issuing letters of undertaking to repay deposits – part of a fraudulent scheme – not within actual authority – bank liable on basis of estoppel.

__________________________________________________________________

J U D G M E N T

SCHUTZ JA:[1] The keeping of two sets of books is one of the stock devices employed by frauds. It was the method employed by Mr Vito Assante when he was the branch manager of the Kempton Park branch of the appellant, NBS Bank Ltd (“NBS”). The ordinances of the bank required that an official accepting money on fixed deposit should enter it on the computer, so that both the receipt of the money and the identity of the depositor would be reflected in the bank’s accounting system. At the same time a computer-generated certificate would be delivered to the depositor.[2] Mr Assante did things differently. He had a scheme to circumvent the prescribed procedures. In return for deposits he would issue a typewritten letter on a NBS letterhead, which he signed as branch manager, undertaking that the NBS would re-imburse the depositor with stated interest on a given day. This letter was not entered in the computer. In fact, once it had been typed Assante’s typist was required to delete it from her word processor. The original letter issued to the depositor would be destroyed once it was surrendered, upon the ultimate repayment of the deposit. The only copy went, not into the NBS’s record system, but into Assante’s briefcase, which he took with him when he went on leave. That was the one set of books. It recorded, correctly, the NBS’s receipt of the deposit, and, again correctly, the name of the depositor.[3] The cheques issued by Assante’s investors named the NBS as the payee and in the case before us were crossed and marked “Not Negotiable”. They were deposited to the NBS’s account with its bankers. (The NBS operated as a building society as we once knew them and did not offer cheque account facilities). Its bankers were successively First National Bank and Standard Bank. So the NBS’s set of books correctly recorded the one side of the transaction, the debit to its bank account. But in this second set of books there was no accounting record of the depositor as its creditor. This was so, because Assante and his associate, Nel, an attorney, had ordered matters so that the credit would be passed to a “corporate saver account” held at the NBS by Nel’s firm, Nel Oosthuizen & Kruger, generally referred to as “NOK”. This type of account was evolved to cater for the likes of attorneys and accountants, who frequently bank money on behalf of clients. Formerly they would open separate accounts for each client. The advantages of a combination of these accounts were that messengers would not have to be occupied in making deposits on behalf of each individual, and that a combined investment would command a higher rate of interest than would separate ones. The corporate saver account was such a combined account. There was only one bank account (sometimes called the umbrella account) and only one customer, the attorney (to treat him as the example). The bank would not keep separate financial records of the attorney’s clients. That would be done by the attorney who would open sub-accounts in his books, to which individual credits would be posted.

This was simply bookkeeping. And that is how the money of Assante’s depositors was treated. But with the vital perversion that the money was not credited in NOK’s books to the persons who had issued the cheques, but to the accounts of developers nominated by Assante. The depositors were unaware of this. They believed that they had been credited in the books of the NBS.[4] Another feature of the corporate saver account was that the attorney was handed a NBS cheque book and was authorised, up to a limit, to issue NBS cheques, which would be charged to the corporate saver account upon payment. Above the limit the cheque would have to be signed by NBS officials. In either event this allowed NOK to repay depositors with a NBS cheque, so that to all appearances as far as the depositor was concerned, when he received a deposit slip reflecting the deposit of a NBS cheque, the money that he had directed to be paid to the NBS, and which he thought had been so paid, was in due course repaid to him by the NBS.[5] In short, the NBS’s set of books did not know the depositor and the depositor knew only the NBS. Almost needless to say, a scheme of this nature included attractive interest rates and the usual panoply of brokers receiving exceptionally large commissions.[6] Once the money was credited to NOK’s corporate saver account, Assante and Nel had control over it. They used that control to make advances to developers. The record does not reveal exactly why, but clearly these investments went badly awry. Some hundreds of millions went through the account. As it was a pyramid scheme (the money of later entrants was used to keep earlier ones content) it could not go on for ever. After more than two years it came to an abrupt halt in December 1996, when a bank in Port Elizabeth raised a query with the NBS head office in Durban about one of Assante’s letters of undertaking. By then some R134 million had been lost. Some twenty actions were instituted, by a veritable Who’s Who of plaintiffs. The appeal before us lies against the judgment of Nugent J, sitting in the Witwatersrand Local division, in one of them, finding for the four plaintiffs before him.[7] I have given a broad description of the fraudulent scheme as it emerges from the record, because once it is understood the huge detail and frequent irrelevance of the 39 volume record can largely be passed over. From here on I deal more narrowly with the facts of the case before us. In setting out the scheme I have, as did the trial judge, rejected the evidence of Assante and accepted the broad version of those who contradict him on material points. The essential difference in version is this. Assante says that the plaintiffs lent directly to the developers (without even knowing who they were), that he Assante did not receive the plaintiffs’ cheques, which were handed directly to NOK for loan to the developers, that NBS was not the borrower, that the brokers who dealt with the plaintiffs’ representative, one Mr Lapiner, were fully informed of the nature of the transaction, that Lapiner was in turn informed by them, and by himself Assante telephonically, and that Lapiner, dazzled by an interest rate some two percent above the market rate, took his chance with the developers. The essence of Assante’s version is that the NBS was not involved at all. The NBS has not relied on his version as a defence, but has contended for a lesser version, that Lapiner was “either aware of the risks involved or deliberately closed his eyes to them,” to quote the NBS’s heads of argument. In other words, he is contended not to have acted reasonably in relying on Assante’s representations.[8] The other main defence, if the plaintiff’s version is once accepted, is that in any event Assante did not have authority, actual or ostensible, to issue the letters of undertaking as he did, partly because he was acting for his own benefit, not that of his employer, with the consequence, so it is contended, that the NBS is not liable to the plaintiffs in contract.[9] There are four plaintiffs, Cape Produce Company (Pty) Ltd, Cape Produce Company Port Elizabeth (Pty) Ltd, Benjy Lapiner Children’s Trust and Ronelle Lapiner Children’s Trust (collectively “Cape Produce”). Mr Benjamin Lapiner, aforesaid, who has expertise in the field of hides and skins, throughout acted on behalf of the four plaintiffs in making investments. The main claim pleaded was that the NBS was contractually bound to pay the plaintiffs the combined sum of R31.5 million, in respect of the seven fixed deposits that were not previously repaid, plus the agreed interest on them. The basis of the claims was that Assante had authority to bind the NBS, either actual or ostensibe. In the alternative, and to cover the event that Assante acted without any authority, the plaintiffs relied on the alleged enrichment of the NBS as the basis for their claim. Since neither of these claims is couched in delict the question of vicarious liability does not arise. In giving judgment Nugent J made orders for payment on the contractual claim in favour of three of the plaintiffs in the sums of R 26 240.307, R 4 961 773.97 and R 2 756 541.10, with the agreed interest for the terms of the loans and mora interest thereafter. The NBS joined six third parties. The first of these was Assante. He was declared to be liable to indemnify the NBS. He has not appealed. The same declaration was made, jointly and severally, against the second, third and fourth third parties. They did not appear at the trial and are not parties to the appeal. The fifth third party was one Trevor Bradley, one of the brokers. NBS’s claim against him for indemnification failed, absolution from the instance being ordered. The NBS has appealed against that order. Bradley neither filed heads of argument nor appeared at the appeal, stating that he lacked the funds to do so. A similar claim against another broker, one Stephenson, was postponed before the trial. The respondents in the appeal are accordingly the four plaintiffs and Bradley, leave to appeal having been granted by Nugent J in respect of the five of them.[10] In order to explain the conclusion which I have already expressed with regard to the rejection of Assante’s evidence and the acceptance of that of Lapiner, and in order to examine the issues of Assante’s authority and also Bradley’s possible liability to the NBS, it is necessary to examine certain of the evidence more closely.[11] Lapiner is a businessman of experience. It was his practice to make regular enquiries as to what rates of interest were on offer in the market, with a view to investing surplus cash from time to time to best advantage. It was his practice to invest only in what he called “Triple A” companies. One day he came to hear of the excellent return being offered by the Kempton Park branch of the NBS. Assante had informed various financial brokers what was on offer. The scheme presented was that the NBS was lending to property developers who were prepared to pay high rates of interest, which allowed the NBS to offer better than average rates to substantial investors who were willing to lend NBS the funds necessary for the purpose. One of these brokers was Bradley. Bradley spoke to another broker, Mason, who knew Lapiner. The result was a meeting between Lapiner and Mason in October 1994. The latter produced a blank letter of “guarantee” from the Kempton Park branch of the NBS. The NBS complied with Lapiner’s criterion of a Triple A company. The investment was to be for a period of some months. The interest rate offered was 15%, which Lapiner described as “slightly above the going rate at the time”. Lapiner insisted in evidence that he lent on the strength of NBS’s name. He would have been “horrified” at the thought of his money being lent not to the NBS but to developers whose identity he did not even know. At no time was he aware of any developers’ names, nor had he heard of Nel or NOK. When the improbability of an experienced businessman lending millions to an unknown developer was put to Assante, his answer was, “[I]t does seem impossible, but that is in fact what happened.” The improbability, to put it at its lowest, of this happening is compounded by the consideration that many other experienced businessmen behaved in the same inexplicable way, that is, if Assante is to be believed. At this point we have a fundamental conflict of fact between Lapiner on the one hand, and Assante on the other, and also a fundamental improbability against Assante.[12] To revert to the dealings between Lapiner and Mason, it was agreed that R 4.5 million would be lent for 151 days. A cheque dated 31 October 1994 for that amount in favour of NBS was given in exchange for either an original letter of guarantee or a copy. (If it was the copy then the original arrived within a few days). The cheque was taken away by Mason, to be couriered to the NBS, Lapiner believed. In fact the cheque was couriered to Bradley. According to him (but not Assante) the first few cheques were handed to Assante. The later ones, on Assante’s instructions according to Bradley, were delivered directly to NOK. On 31 October 1994 an “Asset Dealing Advice” was issued by Mason’s firm to Cape Produce. It reflected the asset type as “Direct Bond (Fixed Deposit)”, stated that it was issued by NBS Bank Ltd and stated under the settlement details: “We collect your cheque in favour of NBS Bank.” This document is consistent with Lapiner’s version and inconsistent with Assante’s. [13] The numerous later transactions followed the same pattern as the first one. Except in the case of the last seven investments, repayment was made by means of a NBS cheque on due date of the capital plus accrued interest. The second transaction was initiated by a note dated 8 November 1994 from Mason to Cape Produce stating: “NBS have offered the following investment: R 2.5 million @ 15.0% ......... This is the same as the one we did recently.” This note also is inconsistent with Assante’s version.[14] Because of their central importance it is necessary to set out the terms of the letters issued by Assante, on the face of them on behalf of the NBS. There were two series, the terms of all within a series being identical. The first series used the word “guarantee”, the second “undertake”. The second series was brought into use from 28 August 1995 and the seven investments with which this appeal is concerned, commencing on 19 June 1996, were all made against a second series letter. The reason why there was a change in style was that Lapiner’s attorney, one Loon, had expressed the view that the original letter did not accurately reflect the agreement which Lapiner had described to him, because the word “guarantee” suggested that some unnamed third party, other than NBS, was primarily liable. When Lapiner conveyed this view to Assante telephonically, the latter was quite ready to change the format in future to one drafted by Loon and also to replace still current series one letters with series two versions.[15] Both forms of the letter were on the NBS’s Kempton Park letterhead and signed by Assante as branch manager. The original one, dated 28 October 1994, read:

“Dear Sir

FINANCE – R 4 500 000

We hereby confirm that NBS guarantees to repay the sum of R 4 779 246 . . . on 31 March, 1995 to Cape Produce Company upon presentation of this letter.”

The difference between the two figures was the agreed interest.The first of the second series letters, dated 28 August 1995, read: “Dear Sir FINANCE R 1 500 000.00

This letter serves to confirm that you have deposited with us the sum of R 1 500 000 . . ., which we NBS Bank Limited undertake to pay with interest at prime upon presentation of this letter on 7 March, 1996, this will amount to R 1 644 452.06 . . . We also confirm that the interest rate payable in terms of the above financing is linked to prime and in the event of an increase in this rate, the amount payable will be adjusted proportionally. The additional sum will be payable, together with the amount referred to above on 7 March, 1996.” (Emphasis supplied.)

[16] As might have been expected, Assante did not fare well when cross-examined on these letters. He admitted that he had no actual authority to issue the letters of undertaking. (This was also the gist of the evidence of the NBS’s witnesses Norton and Munro). Assante sought to explain the existence of the letters as “merely giving some sort of comfort to the parties that were providing the funding in the knowledge that there was a bank employee involved in exercising some sort of administrative or monitoring control over the funds.” The letters were not meant to be enforced against the NBS. What they reflected was an agreement between the broker and the developer. This simply could not and did not wash. The “you have deposited . . . with us” he attempted to explain as being a reference to the corporate saver account, which had a “domicile” in the NBS’s bank account. The “we NBS . . . undertake to pay” he tried to explain as also being a reference to the corporate saver account, but he was driven to concede that on his version of the agreement that part of the letter was false. Later he conceded that the letter as a whole was false and dishonest. [17] It is unnecessary to examine Asssante’s credibility further, other than to quote what Nugent J had to say about him (which is consistent with the record): “[N]ot only was he thoroughly dishonest in the manner in which he conducted the affairs of the NBS, but the evidence that he gave was in my view patently dishonest in material respects. I would not rely upon a word that he said without proper corroboration. Nevertheless, I should add that he is well spoken, urbane and articulate, and he has an agile mind. He is also able to tell even the most obvious untruths without a flicker of emotion or unease. It is important to bear all that in mind when assessing the reactions of those who dealt with him.” Was a contract concluded with Lapiner and Assante as the actors? [18] The NBS disputes the plaintiffs’ contention that tacit contracts to lend and repay money with interest are proved even if Lapiner’s evidence is accepted and Assante’s rejected. To my mind a series of such contracts has clearly been established by the conduct of Lapiner and Assante. Mason presented Assante’s scheme to Lapiner and he accepted it. By the time that the latter had given his cheque and retained the letter of undertaking in each case, all the necessary terms had been agreed. Nor does it matter that Assante, who had signed the letter, had the unexpressed intention that the money would not be paid to the NBS, as Lapiner had been told it would be: Pieters and Company v Salomon 1911 AD 121 at 130. In deciding whether a tacit contract has been concluded, the law objectively considers the conduct of both parties and the circumstances of the case generally: Joel Melamed and Hurwitz v Cleveland Estates (Pty) Ltd : Joel Melamed and Hurwitz v Vorner Investments (Pty) Ltd 1984(3) SA 155 (A) at 165 H and the discussion and references contained in Christie The Law of Contract 4 ed 91 et seq . To all outward appearances there was a contract. Had Cape Produce and Assante been the apparent parties, there would have been a contract between them. But the second apparent party was the NBS, so that what remains to be decided is whether Assante had ostensible authority to bind the NBS, as he claimed to do. Before setting out the law on that subject it is convenient to deal with a further largely factual question, whether Lapiner acted unreasonably in concluding that Assante did have authority. This question will be relevant to the existence of ostensible authority, as will appear later in this judgment. Did Lapiner act unreasonably? [19] In support of its contention that Lapiner deliberately closed his eyes to the risks involved or was aware of them the NBS relies upon the cumulative effect of various factors. Thus the “direct bond (fixed deposit)” on Mason’s “Asset Dealing Advice” should have told Lapiner that he was lending not to the NBS, but some borrower giving a mortgage bond as security. Again, the word “guarantee” in the first series letters should have conveyed a like message. This ignores that Lapiner had Mason sitting before him explaining to him exactly what form the investment took. Among other things Mason said that the investment was “like a fixed deposit”. Lapiner had no reason to distrust him. He had dealt with him before. With reference to the “direct bond” he said that Mason could have called it a direct bond or “He could have called it anything.” And in words reminiscent of Code 4.22 (“ plus valere quod agitur . . .”) he added: “No. Because it is not what he calls it. It is how it is done.” [20] A more weighty point made by the NBS is the high interest rate offered - about 2% above the going rate for fixed deposits, and later the prime rate itself. Some time after the first investment had been made Lapiner telephoned Assante, introduced himself, and asked him how he could offer such a high rate. Assante confirmed that the money was invested with the NBS and explained that the NBS would be approached by property developers in need of funds. Having satisfied itself as to the standard of the development, the NBS would lend money as work progressed, but on condition that when the development was completed, the NBS would be entitled to grant bonds to individual purchasers. Thus, according to Assante, the NBS scored twice and was able to offer these favourable rates. Lapiner’s comment was that he was not a property developer, but that the explanation sounded feasible. At a later stage, when interest rates were rising, and Lapiner wished to know whether the investments would still be available, Assante responded to Lapiner’s query as to what rate he would get in future, by granting him the prime rate as it was to be from time to time. Much was made of this in argument and understandably so. Prime is the lending rate offered by a bank to its best customers , so that it is difficult to see how a bank can make a profit if it customarily takes deposits at the same rate. Lapiner had an answer of sorts, when he pointed out that the interest received by Cape Produce was calculated at the end of the period, whereas interest charged on overdraft is calculated daily and debited monthly, which means that the effective rate received by Cape Produce was less than prime. This is true, but because of the short periods involved (some six or seven months) it would not make much difference. His real answer was that he had accepted Assante’s earlier explanation as to how the NBS could pay such high rates. In retrospect Lapiner’s explanations about Assante’s answers raise questions, but if people did not often accept such explanations the frauds would all be out of work. [21] Then there was reference to the high rates of commission rates paid to agents. Lapiner’s answer to this was brief. He was not concerned with commissions and did not know what commissions were being paid. His position may have been very different if he had known of the size of the commissions, that is, if had indeed invested after being told what they were and still claimed that he understood he was investing in a fixed deposit. The next point was, was it not extraordinary that only the Kempton Park branch was offering these rates? Lapiner responded that he had had previous experience of such a situation – when the Port Elizabeth branch of Trust Bank had offered 2% more than any other Trust Bank branch in South Africa. The fact that a computer-generated certificate of fixed deposit was not issued was the next reason advanced why Lapiner’s suspicion should have been aroused. The answer is that the man in the street is less concerned with the bureaucrat’s workings than the bureaucrat thinks he should be. Something was also sought to be made of the fact that Lapiner did not make an electronic transfer directly into the NBS’s account rather than hand a cheque to a broker. This is hardly a point. Not surprisingly the NBS had not warned the public that they should not trust its brokers with more traditional forms of payment. So, why shouldn’t they? Moreover, looking at the matter overall, consistently with the way in which many fraudulent schemes are operated, the victims seem to have been flattered into believing that they were being specially privileged by being allowed to participate in a limited number of opportunities to gain exceptional returns. [22] In retrospect one may consider that Lapiner was too trusting. But I agree with Nugent J that he was neither untruthful when he said that he accepted Assante’s assurances, nor unreasonable in doing so. What was emphasized by Lapiner was that he was dealing with a branch manager of a large branch of a reputable bank. This is a factor not to be underestimated. Add to that the description of Assante by the trial judge already quoted. He was an accomplished liar. Nor did Lapiner confine himself to speaking to an agent or broker. Before he made the later investments now in issue he had spoken to Assante on the telephone on three occasions, once about the high interest rates, once about the wording of the letters, and upon a third occasion about the reasons why Lapiner’s Port Elizabeth bank would not accept one of the letters as a pledge. On that occasion his bank manager, one Skinner, spoke to Assante on the telephone. An explanation was given. Again no suspicion was aroused. As already explained, Lapiner’s attorney Loon, was consulted about the form of the earlier letters. This led to an agreed amendment, but nothing untoward was sensed. There is yet another incident. In September 1996 Lapiner’s accountant, Mr Liston, received from Assante, by hand, an audit certificate reading: “Total amount of deposits held at 30.6.1996 R 33 000 000,00.” This certificate was on a NBS letterhead and was signed by Assante as branch manager. Liston, a chartered accountant who was well acquainted with Lapiner’s affairs, accepted it for what it was. [23] These events, involving not only Lapiner, but several professional persons, depose to how persuasive Assante’s fraud was. Add to this evidence the fact that by the time the deposits in issue were made, some R 60m had been invested by Lapiner all of which had already been repaid or was repaid before Assante’s exposure, and the point which I have stressed already with regard to Assante’s credibility – the strong improbability of the lender to Triple A companies being indifferent to whether his money was passing into the hands of unknowns. Given all those circumstances I do not agree with the submission that a reasonable man would necessarily have telephoned the NBS head office to query Assante’s authority before proceeding to invest. An ultra-cautious person may have done that, but it was the very status of Assante that might cause a reasonable man not even to consider such a step. Authority of Asssante to issue the letters [24] In the appeal Cape Produce abandoned reliance on actual authority and relied only on Assante’s having had ostensible authority to act as he did. The distinction between these concepts is explained simply by Denning MR in Hely-Hutchinson v Brayhead Ltd and Another [1968] 1 QB 549 (CA) at 583 A-G: “[A]ctual authority may be express or implied. It is express when it is given by express words, such as when a board of directors pass a resolution which authorises two of their number to sign cheques. It is implied when it is inferred from the conduct of the parties and the circumstances of the case, such as when the board of directors appoint one of their number to be managing director. They thereby impliedly authorise him to do all such things as fall within the usual scope of that office. Actual authority, express or implied, is binding as between the company and the agent, and also as between the company and others, whether they are within the company or outside it. Ostensible or apparent authority is the authority of an agent as it appears to others. It often coincides with actual authority. Thus, when the board appoint one of their number to be managing director, they invest him not only with implied authority, but also with ostensible authority to do all such things as fall within the usual scope of that office. Other people who see him acting as managing director are entitled to assume that he has the usual authority of a managing director. But sometimes ostensible authority exceeds actual authority. For instance, when the board appoint the managing director, they may expressly limit his authority by saying he is not to order goods worth more than £500 without the sanction of the board. In that case his actual authority is subject to the £500 limitation, but his ostensible authority includes all the usual authority of a managing director. The company is bound by his ostensible authority in his dealings with those who do not know of the limitation. He may himself do the ‘holding-out.’ Thus if he orders goods worth £1,000 and signs himself ‘Managing Director for and on behalf of the company,’ the company is bound to the other party who does not know of the £500 limitation . . .” [25] As Denning M R points out, ostensible authority flows from the appearances of authority created by the principal. Actual authority may be important, as it is in this case, in sketching the framework of the image presented, but the overall impression received by the viewer from the principal may be much more detailed. Our law has borrowed an expression, estoppel, to describe a situation where a representor may he held accountable when he has created an impression in another’s mind, even though he may not have intended to do so and even though the impression is in fact wrong. Where a principal is held liable because of the ostensible authority of an agent, agency by estoppel is said to arise. But the law stresses that the appearance, the representation, must have been created by the principal himself. The fact that another holds himself out as his agent cannot, of itself, impose liability on him. Thus, to take this case, the fact that Assante held himself out as authorised to act as he did is by the way. What Cape Produce must establish is that the NBS created the impression that he was entitled to do so on its behalf. This was much stressed in argument, and rightly so. And it is not enough that an impression was in fact created as a result of the representation. It is also necessary that the representee should have acted reasonably in forming that impression: Connock’s (SA) Motor Co Ltd v Sentraal Westelike Ko-operatiewe Maatskappy Bpk 1964(2) SA 47(T) at 50 A-D. Although an intention to mislead is not a requirement of estoppel, where such an intention is lacking and a course of conduct is relied on as constituting the representation, the conduct must be of such a kind as could reasonably have been expected by the person responsible for it, to mislead. Regard is had to the position in which he is placed and the knowledge he possesses. A court will not hold a person bound by consequences which he could not reasonably expect and are therefore not the natural result of his conduct: Monzali v Smith 1929 AD 382 at 386, Poort Sugar Planters (Pty) Ltd v Minister of Lands 1963(3) SA 352(A) at 364 A-B. [26] What Cape Produce therefore has to prove in order to establish Assante’s ostensible authority is: 1 A representation by words or conduct. 2 Made by the NBS and not merely by Assante, that he had the authority to act as he did. 3 A representation in a form such that the NBS should reasonably have expected that outsiders would act on the strength of it. 4 Reliance by Cape Produce on the representation. 5 The reasonableness of such reliance. 6 Consequent prejudice to Cape Produce. (This last element is clearly present and requires no further mention). [27] It is necessary to state that two defences that have been unsuccessfully advanced in the past cannot avail the NBS. They are, first, that Assante was acting in his own interests and in fraud not only of Cape Produce but also of his employer, the NBS: Chappell v Gohl 1928 CPD 47 , Bowstead and Reynolds on Agency 16 ed art 766 p 402. The second is that there existed internal restrictions on the actual authority of Assante even though they were not known to Cape Produce: Bowstead para 8-045 p 391-3, Broderick Motors Distributors (Pty) Ltd v Beyers 1968(2) SA 1 (O) 3 E-F, De Villiers and Macintosh The Law of Agency in SA 3 ed 150. Neither of these contentions was squarely raised, but there were rumblings of them in the argument. A representation by the NBS? [28] Turning to the first two requirements of an estoppel, the making of a representation, by the NBS, it was argued on its behalf that the sole peg on which Cape Produce’s case hangs is the appointment by the NBS of Assante as its branch manager at Kempton Park. This, the argument proceeded, is a wholly insufficient basis . Where an estoppel is sought to be derived from the appointment of an agent to a particular position, the principal is considered to represent no more than that the agent has the authority usually associated with this position (Bowstead para 8-018 p 368). The extent of such authority has to be proved by evidence or established by custom, and, so it was argued, proof and custom were lacking. Moreover, there were features of Assante’s actions which were highly unusual. Assante was committing a fraud and there could be no usual authority to do that. (True; but that does not end the matter). More realistically, the finger was pointed at the large amounts involved, the high rates of interest, the use of a broker to solicit business, to collect cheques and deliver the letters, which were couched in an unusual format, not the usual fixed deposit certificate – and so on. (As has been seen above, some of these points were also raised in the context of the unreasonableness of Lapiner’s reliance on the representations of Assante). [29] I have several difficulties with the argument that Cape Produce’s case rests upon the “mere appointment” of Assante. First, the importance of such a posting is not to be diminished. Members of the public may have an awareness of the existence of a head office and of specialist departments in a bank, even of a “wholesale” as opposed to a “retail” borrowing department and of a “money market”, but for them the branch manager is the bank, the one who is authorised to speak and act for it, if something should be beyond the competence of a lesser official. And for those who may know that for some acts, for instance “wholesale” borrowing, even he might need the confirmation of higher authority, they are entitled to assume that he knows his own limits and will respect them, so that when he speaks, he speaks with the full authority of the bank. [30] Moreover, the supposed unusualness of certain of the features raised amounts to little more than a complaint that Assante did not comply with internal rules, with which Cape Produce was unacquainted and which were not its concern. The fact that Assante at branch level was subject to limits as to amounts and rates of interest was an internal matter. Whether the occasional use of a broker is “unusual” is debatable in the light of some of the evidence led, but in any event this feature does not go to the core of Assante’s unquestioned actual authority, which was the taking, indeed the solicitation, and repayment of deposits. How he did it and whom he employed was of no concern to Cape Produce. The NBS’s accounting procedures and the format of its documents were, again, internal matters. [31] Nor do I agree with the argument for the NBS that Cape Produce’s case is limited to an appointment, that there is no evidence of what a branch manager’s usual powers are. In fact there was a good deal of evidence, some of it elicited by the judge. Not surprisingly, he did not suggest that the Mason/Lapiner transaction was everyday. Rather he took the more homely example of the man with a million rands to spare who has himself ushered in to the branch manager to enquire what the latter can do for him. In argument it was contended that his questioning of Mr Norton (by the time of the trial the former chief executive officer of NBS Holdings, the holding company of NBS) was over-vigorous and calculated to overawe. The questioning was often pointed, even persistent, perhaps undesirably so, but it was directed, to my mind, towards establishing, in the face of a determined rearguard action, the relative simplicity of a matter which had been presented as abstruse and complicated. What has to be decided is whether a branch manager of a bank has the authority to accept a deposit and issue a letter of undertaking to repay. When the enquiry becomes focused upon ostensible authority, evidence about the internal controls of the bank is largely irrelevant, despite the fact that the bureaucratic mind believes that things may not happen, do not happen, and finally, cannot happen, unless the regulations are complied with. The outsider does not think that way. Nor does the law. In my opinion a great deal of time and expense was wasted on evidence that took the NBS’s case nowhere. Cape Produce did not help matters by relying on actual authority up to the time that the appeal was argued. [32] What emerges from the evidence is not a nude appointment, but an appointment with all its trappings, set in a context. The context was a bank, whose business was the taking of deposits for a period at interest, and the lending of money on security at a higher rate of interest. It created branches to carry on this business and it appointed managers to manage them. Assante was appointed the local head of this business at Kempton Park. He commanded the staff, including his secretary, who typed the letters and then deleted them from her computer on his instructions, keeping her qualms to herself, whether out of fear, or loyalty, or both. The letterhead on which the letters were typed was provided by the NBS. The facility was created, and it functioned, for the NBS to take Cape Produce’s cheques into its bank account, and for its cheques to be issued in repayment. This state of affairs continued for some 18 months with numerous repayments, without the NBS’s own system of control detecting the abuse. When later Lapiner telephoned Assante as manager he found him at Kempton Park as manager. No doubt for good commercial reasons the NBS did not publish to the likes of Lapiner its internal restraints upon its managers, and it knew that people like him would be largely ignorant of these matters. It held out its branch managers as its front to the world and its local spokesmen. And even if the modern manager has disappeared behind screens of bullet-proof glass and chattering machines, it knew that the public still has a view as to what a bank manager is. [33] All in all the NBS created a façade (I use that word only because I am concentrating on outward appearances) of regularity and order that made it possible for Assante, for a time, to pursue his dishonest schemes. And it is in the totality of the appearances that the representation is to be found. That representation was that Assante was authorised to agree terms of deposit and take money deposited, even in non-routine transactions such as were concluded with Lapiner. Should the NBS have expected outsiders to act on the representation? [34] Of course, for purposes of the enquiry whether the NBS should reasonably have foreseen that outsiders might be misled by its actions, one must not impute knowledge of what Assante actually did to the NBS in pursuance of his fraud: R v Kritzinger 1971(2) SA57 (A) at 59 H – 60 D (decided in 1953, reported in 1971). What the NBS was aware of was the way that it presented itself, Assante, and their relationship, to the world, as already described. Thus it knew that Assante could and would convey terms of receipt of deposits to investors and cause money to be taken and repaid, which is what he did. It also knew, as already stated, that members of the public would be largely ignorant of its internal rules and procedures and would therefore not be protected by them should their operation be perverted by a dishonest manager. Indeed the very existence of some of these rules makes it more difficult for a bank to escape a finding of reasonable foresight. Many centuries of experience has taught banks that vanity, foolishness and greed may lead a manager off the path of strict probity. Hence at least some of the internal restrictions and procedures have been designed to prevent or limit consequent harm. A thieving bank manager is not a common figure but he is not unknown, and a bank knows that if it has had the misfortune to employ such a one, he will have the machinery and the status that it has placed at his disposal, to attempt to accomplish his ends. Therefore, even though the NBS did not foresee exactly what Assante later did, it could reasonably have foreseen, not only that Assante would take deposits, but even that he might thereafter misappropriate them. Reliance on the representation [35] Once the evidence of Lapiner is accepted and that of Assante rejected, as is the case, it is clear that Lapiner did in fact rely not only on what Assante conveyed to him, but also on the NBS’s representation as to who and what Assante was and what authority he had. The element of causation is established. Reasonableness of Lapiner’s reliance on the representation [36] Although the enquiry as to the reasonableness of Lapiner’s actions relates to his reliance on the NBS’s representation, it is nevertheless necessary to have regard to what was known to Lapiner more generally, which knowledge arose largely out of the way that Assante presented himself and his scheme to Lapiner through Mason. My conclusion is to be found earlier in this judgment. It is that Lapiner acted reasonably. Conclusion on contract [37] My further conclusion is therefore that Cape Produce has proved its claim to hold the NBS to its ostensible contracts of deposit, concluded through the ostensible agency of Assante. Performance of its obligations by Cape Produce [38] The NBS has contended that even if the contracts have been proved, Cape Produce is not entitled to relief because it has failed to perform its side of the contracts, namely to deposit moneys in the NBS bank account. The suggestion is that due to the manipulations of Assante and NOK the moneys were somehow never paid to the NBS but to someone else. I have difficulty in understanding this argument. The evidence is clear. The Cape Produce cheques were deposited into the NBS bank accounts either at First National or Standard. When these cheques were collected money was transferred from Cape Produce’s bank account to the NBS bank account. That is payment. In the terminology of s 1 of the Banks Act 94 of 1990 it is, as one might expect, a “deposit”, meaning “an amount of money paid by one person to another person subject to an agreement . . . .” Nor does it seem to me to matter that the NBS, that is apart from Assante, did not at the time know of Cape Produce. It cannot be heard to complain that it was not paid when a payment in fact was dispatched and was received by it, no doubt because of the manner in which the payee had been nominated and the cheques crossed, precautions taken by Lapiner, which were effective and worked. The fact that the moneys were later misappropriated is not a matter that concerns Cape Produce, at least in respect of contractual performance where no blame can be attributed to it. Unjustified Enrichment [39] Cape Produce contended that even if its claims in contract should fail, it should in any event succeed on the basis of unjustified enrichment. As it usually does, this claim led to much debate, but in view of my earlier conclusion it is unnecessary to say anything further about it. Conclusion on Cape Produce versus The NBS [40] In the result I consider that Nugent J was correct in upholding Cape Produce’s claims in contract. The appeal by NBS against the plaintiffs must therefore fail. The NBS versus Bradley [41] The essence of the NBS’s claim against Bradley for re-imbursement was that he had caused or allowed the cheques not to be credited to Cape Produce and that in so doing he had acted fraudulently as part of a conspiracy of which Assante was part, or negligently in that Bradley owed a duty of care to the NBS. The substance of the case sought to be established against Bradley was that he was aware that Assante, Nel and NOK were diverting Cape Produce’s moneys away from the NBS into the hands of the developers, yet continued to take money from Lapiner, all of this in return for a more than handsome commission. [42] Assante gave direct evidence of Bradley’s knowledge that the money was going directly to the developers, but his evidence is worthless. Bradley’s evidence was that after he had heard of the Kempton Park scheme, he and another broker, Kruger, visited Assante with a view to obtaining an explanation of the scheme and their possible participation in it. There he obtained the explanation which ultimately reached Lapiner, that is, on Lapiner’s version. The borrower was to be the NBS, not the developer. He also was taken in by Assante’s status as branch manager. Kruger gave evidence in support of Bradley. Thereafter Bradley acted as a courier sending Assante’s letters to Lapiner and taking the latter’s cheques to Assante, or more usually, on Assante’s instructions, to NOK. [43] Bradley’s major difficulty was caused by the transcript of an enquiry held by one van As on 8 February 1997, a few days after Assante’s fraud had been exposed. In it Bradley adopted a position intermediate between that of Lapiner and that of Assante. This intermediate position, however, contradicted vital parts of Bradley’s evidence in court, which corresponded with what Lapiner said had been conveyed to him by Mason. To recapitulate: Assante’s evidence was that NBS was never to be a party and that Bradley was told as much, which was the message that should have reached Lapiner. The borrowers from Cape Produce were to be the developers and that was known to all. Lapiner’s version was that the sole contracting party on the other side was to be the NBS, which would receive a fixed deposit. He was not told about the developers until months after the approach by Mason and in a context that in no way altered the sole liability of the NBS. Bradley’s evidence in court was to similar effect. But his statements to van As came to this. Bridging finance was needed by the developers and was to be provided directly to them by the investors, who were told as much by him, acting as go-between. After stating “I will categorically state that all my clients know that this is bridging finance” he proceeded: “There is no doubt in their mind. We even went as far, you’ve mentioned Cape Produce now, down the line when Mr Lapiner had got au fait with the whole sort of system and the way the whole investment was working. We even got Mr Lapiner to speak to Vito, just for comfort. Vito again explained the whole thing to Benji Lapiner about the whole bridging development and he is happy with it.” (Emphasis supplied.) [44] But although the developer was to be the borrower, “the NBS would be underwriting that investment and guaranteeing its repayment.” The suggestion that a fixed deposit was intended was contemptuously rejected in these words: “[I]f we had gone in and said to any of the of the clients this is a Fixed Deposit, my clients would have said where is my Fixed Deposit certificate. It was never ever approached from a Fixed Deposit point of view. The guy knew it was bridging finance and that is the way it was explained to him. It was bridging finance for the developer to be repaid back to you on a certain date and that was it. It’s just ludicrous if anyone just sits there and says this is a Fixed Deposit” [45] Bradley’s statement to van As is contradicted by that of Lapiner in essential respects. Its veracity need not be tested further as Bradley conceded in evidence that large parts of it were untrue. The lame excuse he gave for telling untruths was that he was exaggerating his own importance and that it was only towards the end of the meeting with van As that he realized that the finger was pointing not only at Assante, but at him too. The importance of his statements is that they contain an admission that he knew that the moneys were going straight to the developers, which was, of course, directly contrary to Lapiner’s instructions and the terms of Assante’s letters. [46] Nugent J took a charitable view of Bradley, saying that he did not get the impression that he was dishonest, rather that he was witless, but no more than that. The fact is that his two versions show him to be thoroughly dishonest. Nugent J also took into account in his favour as a probability, that Nel and Assante would not have been likely to increase their risk by revealing to him what they were about. I am not so sure. Frauds in established positions usually expect to be successful and win the profits of their fraud without detection. We know very little about what was happening in NOK, but the probability is that Nel and Assante believed that the developments would prosper and they would benefit handsomely. In order to find the necessary bridging finance they needed brokers who knew people with money to spare. Bradley was such a one. What surer way than offering him a tempting commission? The commissions paid seemed to vary but they were very high, some 5 or 6 per cent. Lapiner knew nothing of the commissions. Bradley, on the other hand, knew that both the rates of interest and the commission rates were high – a warning light to any money broker: see Durr v Absa Bank Ltd and Another 1997(3) SA 448(A) at 465 C-E. [47] Mindful of an appeal court’s reluctance to upset the credibility finding of a trial judge, I am of the view that Bradley’s version given in evidence should be rejected, and that it is probable that he was a party to the conspiracy, together with Assante and Nel. I have reached this conclusion without reference to the possibly contentious evidence of Ms Malan, the effect of which was that Bradley would sometimes instruct her as to which developer’s sub-account in the corporate saver account was to be credited, which would show that he knew where the money was going. We have Bradley’s own statement that he knew, however much he might now say that it was a lie. [48] Accordingly I consider that the NBS’s appeal against Bradley should succeed with costs. Whether Bradley should be made to bear the costs recovered from the NBS by Cape Produce, as the NBS requests, is less clear. On the one hand it may be argued that the NBS should have accepted Cape Produce’s contractual claim and saved the expense of a protracted trial. On the other it can be argued that in order to bring home its claim for indemnification against Assante and Bradley the NBS would have had to call some at least of the witnesses relied on by Cape Produce. Most of the witnesses that were called by the NBS would, in my opinion, not have been needed. A fair allocation seems to me to be that Bradley should bear one half of the costs recovered by Cape Produce. [49] In the result: 1. The appellant’s appeal as against the first four respondents (the four plaintiffs) is dismissed with costs, including the costs consequent upon the employment of two counsel. 2. (a) The appellant’s appeal as against Trevor Bradley is upheld with costs, including the costs consequent upon the employment of two counsel. (b) The following is substituted for paragraph 4 of the order of the court a quo :

“4(a) It is declared that the fifth third party is liable to the defendant jointly and severally with the first, second, third and fourth third parties in the amounts ordered to be paid by the defendant to the plaintiffs as set out in para 1 of this order and half the taxed costs recoverable by the said plaintiffs from the defendant in terms of para 2 of this order.

(b) The defendant is granted leave to approach this Honourable Court again, on the same papers, duly supplemented, with proof of payment by the defendant to the plaintiffs or any of them of the judgment debt set out in para 1 of this order and half of the taxed costs set out in para 2 of this order, for an order against the fifth third party for payment. (c) The fifth third party is ordered to pay the defendant’s costs, such costs to include those consequent upon the employment of two counsel.” 3. The defendant and the fifth third party (Trevor Bradley) are granted 14 days to make written submissions if they wish to object to the form of the orders made in paragraphs 2(b) and (c) above. Failing such objection those parts of the order will also become final. WP SCHUTZ

JUDGE OF APPEAL

CONCUR NIENABER JA MARAIS JA NAVSA JA FRONEMAN AJA