The Facts

In 1983 the plaintiff company appointed Mrs R as paymistress, a decision which they have no doubt since regretted. Using three different methods, she was able to defraud the company of almost $1.5 million in the space of little more than three years. Mrs R was convicted, but the funds were never recovered. The case before Palmer AJ was one where the "dupes ... litigate amongst themselves, each seeking to cast the loss onto the others." ²

Most of the money, over $1.225 million, was obtained through the operation of accounts for workers who never existed, and most of the argument in the case concerns this method. The remaining money was misappropriated by procuring the drawing of cheques with fictitious payees or by the simple expediency of altering the name of the payee on an existing cheque. Due to space limitations, this article will confine itself to a discussion of the first method. The case also raises interesting questions related to the scope of the defence provided by s95 of the Cheques and Payment Orders Act 1986. These issues will be the subject of a later note.

The opportunity for fraud originally arose through an industrial disagreement. The plaintiff had become concerned about the dangers of armed robbery and so wished to replace the cash payment of wages with some other method. After negotiations, a new payroll scheme was adopted which used the services of the United Permanent Building Society, later acquired by the second defendant NMRB. For the purposes of this note, the distinction between the Building Society and the second defendant are irrelevant.

Accounts with NMRB were established for each employee. Each week a computer printout would show the amount due to each employee, including adjustments for commissions, allowances, etc. Mrs R had the responsibility of assuring the accuracy of this printout. The total amounts to be paid to the employees was totaled and a cheque for the amount was drawn on the plaintiff's banker, the first defendant. The cheque was then delivered to NMRB who, prior to the clearing of the cheque and by way of an advance to the plaintiff, would credit to each employee's account the amount due. As a result of this arrangement, the employee was able to withdraw cash from his or her account on Thursday of each week.

At the establishment of the scheme, officers of NMRB attended a meeting of employees and explained how the system would operate. They interviewed each employee and obtained an application form for the opening of an account. These forms were signed in the presence of the NMRB officer. Following processing of the forms, NMRB issued a card and PIN to the employee which enabled operation of the account through NMRB's ATMs. The card and PIN were sent directly to the employee by mail.

A relatively high level of staff turnover soon exposed the deficiencies in the system. New employees found it "inconvenient" to go to NMRB's offices for interviews. There were delays in receiving cards and PINs, so that new employees could not obtain access to their wages. A meeting of union representatives, officers of the plaintiff company and officers of NMRB produced the "solution".

The solution called for Mrs R to be provided with a number of application forms. When a new employee was hired, she would fill out the appropriate information, obtain the employee's signature and forward the forms to NMRB. In order to reduce the delay after application, Mrs R would also pick up the card and PIN for the employee's account. It was not the award-winning idea of 1985. Mrs R soon began establishing accounts in fictitious names.

It seems that the plaintiff's supervision of Mrs R was virtually non-existent. The Plaintiff's auditors on several occasions had warned that the system of supervision was inadequate, but if notice was taken of the warnings then it was not translated into changes in procedure.

Mrs R was eventually found out and convicted. The plaintiff sued both its own bank and the NMRB in an attempt to recover the amounts. The claims against its own bank were settled, and the judgment is concerned only with the claims against NMRB.

The main issues in the case were whether NMRB could rely on Mrs R's apparent authority, whether the plaintiff could recover money paid under a mistake of fact, whether NMRB had been negligent in failing to advise the plaintiff of the opportunity for fraud and whether NMRB could make out a defence to conversion under s95 of the Cheques and Payment Orders Act 1986. This note will discuss only the first of these.

Were the payments unauthorised?

Mrs R clearly did not have actual authority to make fraudulent insertions in the payroll list. It was just as obviously within her actual authority to compile the (proper) payroll lists, to make whatever handwritten corrections were necessary and to take the appropriate steps to ensure that NMRB made the specified payments in accordance with the scheme. Relying on Armagas Ltd v Mundogas SA³, Palmer AJ concluded that NMRB could rely upon the ostensible authority unless it could be shown that the reliance was not in good faith or that NMRB should have been put on enquiry. The principal is based on estoppel.⁴

Can the defendant's reliance be "in good faith" even though there has been something unwittingly done or omitted which enables the fraudulent agent to succeed? Without citing any authority, Palmer AJ suggested that a breach of contract or of some tortious duty which permits the fraud to succeed does not disqualify the defendant from relying on the ostensible authority. The principal's proper response is to cross-claim in contract or in tort.⁵

This approach is slightly different from the recent decision of Waller J in the Midland Bank case⁶. There it was held that the representee could not rely on a representation if the representation had been induced by some lack of disclosure by the representee. The test suggested is that the representee is required to reveal facts that any reasonable person in the position of the representee would appreciate that the representor would want to know before making the representation.⁷

Was NMRB put upon enquiry? Palmer AJ identified two different situations where it may be said that the defendant has been put on enquiry. The first is where there is some known limitation on the agent's actual authority. In such a case, the third party is at once put on notice as to whether the conditions exist for the proper exercise of the authority. Examples include the authority of the captain of a ship to sign bills of lading (limited to bills for goods which have been in fact shipped) or the authority of a secretary of a company to sign documents (limited to documents which the directors have given authority).⁸

Palmer AJ raises this class of case only to ignore it.⁹ It is perhaps understandable, for this general formulation is not of much help in fraud cases. NMRB clearly knew that Mrs R's authority was limited to adding the names of actual employees. Why are they not put on enquiry to check that the names added are employees in fact? In cannot be because the limitation is irrelevant since it is precisely this excess of authority which was at the heart of the fraud. Yet if the principle is applied literally, it would reduce substantially the distinction between actual and ostensible authority. It is unfortunate that this aspect of the case appears not to have been addressed in argument.

In the second class of case, the act of the agent appears to be within that class of acts which the principal holds the agent out to perform on the principal's behalf. In order to put the defendant on enquiry in this situation, it must be shown that there are circumstances known to the defendant which would excite suspicion in the mind of a reasonable person, keeping always in mind that the reasonable person is not unduly suspicious of business transactions.¹⁰

Palmer AJ considers that in order to put a party on enquiry two conditions must be satisfied. There must be a transaction which on its face suggests that the agent may be exceeding authority. That transaction must then be viewed through the eyes and with the knowledge of the third party at the time of the transaction. Palmer AJ specifically mentions "any previous course of similar dealings between the third party and the principal effected by the agent without demur from the principal".¹¹ He finds that there were no circumstances which called into question whether or not Mrs R was acting in her own interest.

Part of the difficulty with this analysis is that NMRB in fact did question Mrs R about several of the bogus accounts, whereupon Mrs R "gave plausible explanations".¹² It does not seem surprising that NMRB staff expressed concern, for it does not require a "mind alveolated with suspicion"¹³ to query the creation of new accounts and payments of larger than average amounts into those accounts when both the creation and the handwritten directions for payments are the responsibility of a single agent. Once the propriety of the accounts was questions, the fact of asking Mrs R for an explanation hardly fulfils the test of one "...such as is calculated to evoke material fit to deal with the doubt."¹⁴

What is the relevance of a long record of transactions which occur with no objection from the principal? Quoting with approval from the Morison case¹⁵, Palmer AJ found that NMRB was entitled to assume that then employer would discover that "he was being robbed in a mode so easily capable of detection".¹⁶

Morison was concerned with establishing the equivalent of the s95 defence to conversion. In that context the Morison approach has been thoroughly discredited. Usually referred to as the "lulling to sleep" defence,¹⁷ it was soundly rejected by the Privy Council in Bank of Montreal v Dominion Gresham Guarantee and Casualty Co Ltd,¹⁸ a case which has some remarkable similarities to the present one.

It does not seem to have been argued that Mrs R was not solely the agent of the plaintiff. It is not the responsibility of the plaintiff to check the identity of account holders or to receive ATM cards on behalf of the customers of NMRB. It is the responsibility of NMRB who were content to delegate it to Mrs R as part of the business arrangements. At least for these purposes, she was the agent of NMRB.

This "mixing" of the business activities of the two parties is part of the problem with the case. Both parties adopted business practices which permitted the fraud to be successful, business practices which have been found wanting in other cases. Both parties entrusted part of their own responsibilities to an agent who abused the trust. It is probably one of the clearest cases where "justice" calls for some sharing of losses, yet it is only in negligence where our law permits this to be done in a straightforward fashion.