Whose knowledge counts? Singularis v. Daiwa and Attribution
Neil Hext QC | 6 Nov 2019
Last week, the Supreme Court handed down its decision in Singularis Holdings Ltd v. Daiwa Capital Markets Europe Ltd  UKSC 50. That case got the attention that it did because of the tension with the result in Stone & Rolls Ltd v. Moore Stephens. Others have dealt with the detail of the decision in Singularis (including an excellent article by my colleague, Mark Cannon QC). I want to look more generally at the issues created by attribution in a corporate context, and how the courts in recent years have approached them.
The Company’s Agents
It has frequently been said that a company is a fiction. It doesn’t exist anywhere other than in our minds – a creation of the rules that we have all agreed underpin our commercial life. So a company can only act through its agents. Decisions are made by the board of directors. Authority is delegated to employees or other agents to sign off on transactions. Other employees put those transactions into effect.
We don’t have any difficulty in recognising these things as the acts of the company itself. Indeed we often don’t even think about it, assuming (as we are often entitled to do) that the relevant agents have the authority to bind the company, that the company will perform, and that if it doesn’t we will have an appropriate remedy through the courts.
But sometimes real questions can arise as to whether the acts of individuals in positions of authority within the company are to be treated as the acts of the company itself. And similar questions can arise in relation to the knowledge of those individuals. Recent cases have focused on the question of illegality. So, where a director has abused his position to commit a fraud, is the knowledge of the director imputed to the company for the purposes of a subsequent claim? But it can arise in a wide range of other situations as well. For example, is the knowledge of an individual within the company imputed to the company for the purposes of its insurance cover? What about for the purposes of statutes where liability is created dependent on the state of mind of the company?
Directing Mind and Will
Perhaps surprisingly, given the length of time for which companies have been deeply entrenched in the commercial world, the law has taken quite a time to get to grips with this problem. For a long time, the search was for “the directing mind and will” of the company. A company would be held to have the knowledge of those that most closely embodied the “mind” of the company – often, but not always, the board of directors. It would also normally be held to know what an agent instructed to deal with a particular transaction knew, as long as that knowledge could fairly be said to concern the transaction in question. There was an exception of a somewhat uncertain ambit – sometimes called “the fraud exception,” or the principle in Hampshire Land – by which a company was not treated as having the knowledge of a director who was acting in fraud of the company.
It was not until the Privy Council case of Meridian Global Funds Management Asia Ltd v. Securities  2 AC 500 that some discipline was introduced to the topic. That case focused primarily on whose acts were to be treated as the acts of the company. But it also dealt with the question of whose knowledge would count.
Lord Hoffmann broke down the “rules of attribution” into three categories. First, the “primary rules of attribution.” These are the fundamental rules that determine what types of decision bind the company and are generally found in the company’s constitution and the principles of company law. Examples are the rules governing the appointment of directors and how the board makes decisions.
Next are the rules of agency. Employees and other agents will be appointed to act for the company and to enter into transactions on its behalf. These rules are the same as apply to natural persons and enable the company to bind itself contractually, as well as to become vicariously liable in tort.
Lord Hoffmann said that, normally, those two categories of rules would produce an answer as to attribution. But “in exceptional cases” they would not. That might be the situation where a rule of law, whether expressly or by implication, excludes attribution on the basis of the general principles of agency or vicarious liability. And the example he gave was a situation in which a criminal statute required one to establish the state of mind of the defendant company. In such a case, he said, it is sometimes necessary to “fashion a special rule of interpretation for the particular substantive rule.” This is a question of interpretation of the rule:
“Whose act (or knowledge, or state of mind) was for this purpose intended to count as the act etc. of the company? One finds the answer to this question by applying the usual canons of interpretation, taking into account the language of the rule (if it is a statute) and its content and policy.”
Meridian Global introduced clarity into this area of the law, but not much in the way of certainty. What marked out the “exceptional cases” in which one would depart from the ordinary rules of agency and apply the “for what purposes” approach?
Delinquent Directors: Stone & Rolls and Bilta
There then followed three high profile attribution cases, culminating in Singularis, all relating to illegality. The first of these, if anything, seemed to moved away from Meridian Global. In Stone & Rolls Ltd v. Moore Stephens  1 AC 1391, it was held that a one-man company – i.e. a company in which there was a single controlling director who was also the sole shareholder – could not sue its auditors for negligently failing to identify fraud on the part of controller of the company. The essence of that case seemed to be that, Meridian Global notwithstanding, where one had a single individual with complete control and ownership, that individual had to be treated as “the directing mind and will.” Thus the knowledge of that person’s fraud was necessarily imputed to the company. The consequence was that the company’s claim against the auditors inevitably failed for illegality.
The next case was Bilta (UK) Ltd v. Nazir  AC 1 in which liquidators of a company originally owned and controlled by an individual and his associates were, in contrast to Stone & Rolls, not prevented by the illegality principle from suing those persons in fraud. The difference between Stone & Rolls and Bilta was explained on the basis that one might get a different answer to the question, should the knowledge of the director be imputed to the company, depending on why one was asking. In the case of Bilta, the reason why the question was asked was for the purposes of establishing the liability of the directors to the company itself. Once that was recognised, it became obvious that any guilty knowledge on the part of the directors should not be imputed to the company, and that was so even if the directors were together to be regarded as the “directing mind and will.”
Thus one sees in Bilta a movement back towards the language of Meridian Global. But where that latter authority had spoken of “exceptional cases” in which one might apply the more flexible “for what purposes” approach, in Bilta – at least in the context of questions of knowledge – one would always look at the purpose of the rule that gave rise to the question of attribution before answering it. And that was what explained what had hitherto been referred to as the “fraud exception” in Hampshire Land.
Singularis reinforces this approach. In that case, Lady Hale said of Bilta that:
“The court explained that the key to any question of attribution was always to be found in considerations of the context and the purpose for which the attribution was relevant. Where the purpose was to apportion responsibility between the company and its agents so as to determine their rights and liabilities to one another, the answer might not be the same as where the purpose was to apportion responsibility between the company and a third party.”
Singularis featured a claim by a company in liquidation against a bank that had failed in its Quincecare duty to identify, and thus avert, significant misappropriations by the company’s sole active director. Save that this case concerned a bank as opposed to auditors, the facts of Singularis were not a million miles away from those of Stone & Rolls. Yet the Supreme Court held, applying the “for what purposes” approach, that the director’s knowledge of his own fraud was not to be imputed to the company. So the bank’s illegality defence failed. The whole purpose of the Quincecare duty to identify suspect payments is to protect the company from misappropriation by its trusted agents. To attribute the fraud of those agents to the company would “denude the duty of any value in cases where it was most needed.”
Lady Hale distinguished Stone & Rolls on the basis that that case concerned auditors, whose job it is simply to report to shareholders. If everyone within the company, including shareholders, already know about the fraud, because they are all in on it, the auditor’s negligence does not cause the loss. Had the auditor told the company of its concerns, it would have made no difference.
But this disguises a real difference between Stone & Rolls and Singularis. In the former, a one-man company was treated by some of the law lords as if it would for all purposes know what the single director/shareholder knew. After Singularis that would appear to be incorrect. Whether or not the knowledge of that director/shareholder should be imputed to the company will depend upon the purpose of the inquiry.
Of course, that begs the question as to whether the decision on attribution in Stone & Rolls was correct. That will depend upon an analysis of the purpose of audit duties. Moreover, as Lady Hale pointed out, a similar case may still fail for other reasons such as causation.
Whose Knowledge Counts?
But where does it leave us in relation to the question of attribution more generally? The “for what purposes” test provides logical clarity. In some situations, the answer will be explicit – so, for example, in the context of non-disclosure of material information under the Insurance Act 2015, the legislation itself sets out in section 4(3) what normally counts as the knowledge of a company for these purposes.
Likewise, where one is dealing with a case in which liquidators of a company are claiming damages for losses arising out of default on the part of its directors, it would militate against the whole purpose of the duties imposed upon directors if their knowledge were imputed to the company to defeat the claim: see Bilta itself.
And it seems that the “for what purposes” test is unlikely to produce a revolution in agency law. As Lord Toulson and Lord Hodge said in Bilta, in most circumstances the acts and state of mind of its directors and agents can be attributed to a company by applying the pre-existing rules of the law of agency. So, for example, the knowledge of an agent duly authorised to enter into a contract will very often be imputed to the company.
But there will remain cases in which working out what counts for the purposes of the company’s knowledge will be more difficult. It is not always straightforward to ascertain what the purpose of any given rule is; and even when ascertained, the question of whether attribution would or would not accord with that purpose may not be clear. It seems likely that fraud cases will continue to test the boundaries of attribution. But that is not the only area of potential difficulty. Contractual or statutory requirements for knowledge or notice on the part of a company provide further examples. One thinks, again in an insurance context, of a policy requiring notice of known circumstances that may give rise to a claim. Singularis provides clarity in terms of approach. In practice, the scope for dispute may have been increased.
© Neil Hext QC of 4 New Square, November 2019. The author assumes no responsibility to any party in respect of this article. Specific legal advice tailored to specific problems should always be obtained.