Two recent English cases illustrate a strict and a more flexible approach by the courts when considering whether the signing of a legal document has complied with, or is rendered ineffective by non-compliance with, prescribed statutory formalities. We first look at the more flexible approach.
Execution of deeds by companies without a common (or company) seal
On 1 November 2023, Eyre J gave his judgment in Lendlease Construction (Europe) Ltd v Aecom Ltd. The signing formality at issue there was the requirement under Companies Act 1985 (and now found in the Companies Act 2006) (CA) that for a deed to be validly executed by a company, unless its common seal is affixed to the deed – very rarely done nowadays – the deed must be signed by two directors or by a director in the presence of a witness who attests her signature.
Signing of a deed by “non-directors”
All that could have gone wrong with the signing of the deed (which was, in fact, a consultancy agreement signed in the form of, i.e., intended to take effect as, a deed) in accordance with these requirements, did go wrong. Thus, the two signatures for Aecom (“A”) where placed against the space marked for the fixing of the company seal, instead for the space for the directors to sign in lieu of a company seal, and the two individuals who signed the deed were not de jure (i.e., formally appointed) directors or even de facto (i.e., acting informally as) directors of the A. The individuals were simply employees of the A who had been closely involved in the negotiation of the documents that included the deed that was in dispute.
The issue about whether the consultancy agreement was properly and effectively executed as deed, rather than simply taking effect as a binding contract – which both parties accepted would be the case if the requirements to be executed as a deed were not satisfied – was relevant because A wanted to argue that the claim against it by the counterparty, Lendlease (“L”), was out-of-time, with a six year statutory limitation applying to the claim rather than a 12-year limitation that would apply to a deed. It was clear that the two individuals who signed the deed were not statutory (i.e., de jure) directors of A at the time they signed. It was also not argued that they were de facto directors (or indeed shadow directors).
A's argument that the two individuals (even if they were directors) were not purporting to execute the agreement by signing it but instead were only purporting to witness the sealing of the agreement and, since the company’s seal was never affixed to the document, their signing had no legal effect, didn’t get very far. It was clear that they had simply signed in the wrong place and that there had never been any intention that the agreement should have A’s seal affixed to it. A’s stronger argument was that since the two individuals were not directors of A, in any sense of the term, the agreement couldn’t have been executed as a deed and so only took effect as a contract. A did not dispute that the individuals had the necessary authority to sign the agreement on behalf of A.
Freeman & Lockyer v Buckhurst Properties (1964)
The judge devoted a large part of his reasoning on this valid execution question to an analysis of the decision in the above 1964 Court of Appeal case ( 2 Q.B. 480) which held that a company was bound by the actions of one of its four directors in entering into an agreement as managing director despite never having been formally appointed to that post under its articles. The court ruled that the counterparty did not have to enquire whether the director had been duly appointed under the company's articles with authority to enter into contracts by himself on behalf of the company. It was sufficient that there was power under the articles to appoint the individual with the “contracting powers” of a managing director.
A argued that the Freeman & Lockyer line of cases only established that the agreement would bind A on the basis of the ostensible authority of the two non-director individuals who signed it – arising from them having been “held out” as having the authority to sign the agreement. Those cases were not relevant to the question whether the agreement could take effect as a deed despite non-compliance with the CA requirements.
Eyre J disagreed. He ruled that the two individuals signing the consultancy agreement were purporting to sign it as deed. They had been held out as having actual or ostensible authority and while the court in Freeman & Lockyer had said this was an “estoppel by representation” Eyre J noted that L, arguing that the agreement took effect as a deed, had not actually asserted either an estoppel or reliance.
This was potentially significant since Eyre J also said that since the rules governing the execution of deeds are laid down by statute “a real degree of care is needed before it can be appropriate to find such an estoppel”. This was to be contrasted with the Freeman & Lockyer ostensible authority rule, for which the judge said, “there must be a holding out and reliance upon it but it is clear that both will be readily inferred”. However, here Eyre J was:
“satisfied that it is appropriate to infer both representation and reliance in circumstances where [the two individuals] expressly signed as directors a document which was predicated on the execution on behalf of [A] being by directors of that company and the terms of which made it clear that it was being entered on that basis; where [A] had placed those gentlemen in positions where they were able and expected to perform in that way; and where the parties thereafter proceeded on the basis that their dealings were governed by the Consultancy Agreement.”
How much difference, in practice, is there between the court’s readiness to infer an estoppel in the case of ostensible authority and in the case of statutory execution requirements? Eyre J seems, arguably, to have equated (if not to say extended) ostensible authority to enter into contracts with the concept of an “ostensible director”.
In support of this approach it should be noted that the CA requirements themselves provide for a form of what you might call “ostensible” due execution, by providing (in section 44(5)) that “in favour of a purchaser a document is deemed to have been duly executed by a company if it purports to be signed in accordance with [the CA requirements]” (i.e., by two directors or a director whose signing is witnessed). However, there is no reference to that statutory presumption of due execution in the judgment, presumably because, being a consultancy agreement, L couldn’t properly be described as “a purchaser” – which is defined as being “a purchaser in good faith for valuable consideration and includes a lessee, mortgagee or other person who for valuable consideration acquires an interest in property”. In addition, just as in Freeman & Lockyer, Pearson LJ said there was no need for a person dealing with a single director (in a normal transaction within the ordinary scope of the company’s business) to check the company’s articles to ensure that the directors had power to delegate to a single director – “such a requirement would be an absurd example of legal pettifoggery”, Pearson LJ colourfully said – it would also not be unusual for parties to deeds involving a company not to bother carrying out a company search to see whether the register – which might, of course, be out-of-date anyway – showed the persons signing as directors were recorded as being “active” directors of the company.
This approach is also consistent with the principle applied in the so-called rule in Turquand’s case which says that third parties dealing with companies do not need to make enquiries into compliance with internal governance requirements of a company.
At the end of the day, the judge seems to have been struck by the fact that several years after the signing of the consultancy agreement, A was now taking the point that it didn’t, as it purported to on its face and by bearing two signatures claiming to be those of directors, take effect as a deed but only as a “simple” contract, thus – because of the limitation period laid down for contracts as opposed to deeds - defeating the counterparty’s claim. As Eyre J said:
“The central factor is the artificiality of [A]’s position. It accepts that [the two individuals] had authority to act as they did and that “the parties proceeded and contracted on the terms of [the Consultancy Agreement]”. It is not open to [A] to accept that it was bound by the actions of [those individuals] and then to adopt an unrealistic stance as to the nature of those actions or as to their effect. I have already explained my conclusion that there was no intention that the Consultancy Agreement should be executed by the affixing of [A]’s common seal. It follows that when [the individuals] signed the Consultancy Agreement they were intending to execute a deed on behalf of [A]. They were not acting improperly in relation to [A] in doing so but were indeed doing what they were expected to do. As I have already noted the necessary representation and reliance can readily be inferred here. In those circumstances it is not open to [A] to contend that the Consultancy Agreement was not a deed. It follows that the applicable limitation period in respect of claims asserting a breach of that agreement is one of twelve years from the date of the accrual of cause of action.”
A helpful clarification?
Ironically, this decision could be viewed as affirming a primacy of “form” – signatories signing with the title of “director” – over substance – are they directors (whether de jure or de facto or shadow)? Unless, that is, it is to be read as a dealing with a question of statutory interpretation – does the reference to directors in the CA requirements include anyone who is held out as being a director? And of course, the feature of a deed as type of contract is that it depends on the satisfaction of certain formal requirements as regards its execution.
The net result, however, is that someone who was not a director has been held to satisfy the formal requirement of being a director, against a counterparty who doesn’t seem to have fallen within the statutory exception that applies to the formal requirements.
The decision might be appealed. If it represents the law, it will be seen as a useful clarification for those relying on deeds being properly executed by companies in cases where the statutory presumption doesn’t necessarily apply.
Contracting out and ouster of limitation periods
Aecom v Lendlease also addressed another question – whether, if the consultancy agreement failed as a deed and was a mere simple contract, the six-year statutory limitation period which would then apply to it had been ousted by the parties agreeing to a 12-year period for bringing claims under it. L argued that a clause which said that “no action or proceedings under or in respect of the agreement in contract or for breach of statutory duty shall be commenced against the consultant after the expiry of 12-years after the completion date for the works” constituted an agreement between the parties to replace the six-year statutory period with a contractual 12-year period.
Eyre J agreed with A’s argument that the relevant clause merely set up a contractual long stop date for claims but did not oust the relevant statutory limitation period, which was six years according to A but 12 years if (as the judge ruled was the case) the agreement took effect as a deed. Since the agreement was expressed to be a deed, the choice of 12 years was unsurprising and could not be construed as being intended to disapply a much shorter, six-year limitation period.
Based on earlier authority (Oxford Architects Partnership v Cheltenham Ladies College (2006)), the judge ruled that “clear words” would be required to oust the relevant statutory limitation period. He admitted, on the basis of more recent Supreme Court cases dealing with contractual interpretation – e.g., Wood v Capita Insurance Services, Rainy Sky v Kookmin Bank and Arnold v Britton – that there could be circumstances where “reliance on a statutory limitation defence was precluded by a contractual provision even though the provision being construed did not use express words to achieve that effect.” Nevertheless, he thought that those cases would be very rare:
“… the court will consider it inherently less likely than otherwise that the parties intended to agree terms departing from the norm for arrangements of the relevant kind. A provision to the effect that the statutory limitation period is being disapplied and replaced with terms enabling an action for breach to be brought outside that limitation period would be a significant departure from the norm. It is, accordingly, a provision of a kind which the court will regard as less likely than otherwise to have been the parties’ intention. It follows that although express words are not required as a matter of law for a provision to have that effect it is unlikely that such an effect will be achieved other than by the use of clear language such as can only properly be interpreted as having that effect.”
Requirement for an absolute legal assignment to be made under the hand of the assignor.
Another recent case this time shows the courts adopting a strict interpretation of statutory requirements with regards to the way in which documents must be executed to be legally effective. Section 136 of the Law of Property Act 1925 (“LPA”) only allows the assignee of a debt or other “legal thing in action” itself to acquire the legal rights and remedies in relation to the debt, etc., if the assignment is made in writing “under the hand of the assignor”. In Frischmann v Vaxeal Holdings SA & Ors, Master McQuail ruled very briefly in a six-line single paragraph of a 144-paragraph judgment that the “under the hand of” requirement was not satisfied by the assignment having been signed by an attorney for the assignor.
She was cited previous case law authority in support of this view, the most notable of which was a 1880 decision (in Wilson v Wallani (1880) 5 Ex.D. 155) in which it was held that a disclaimer of property under the Bankruptcy Act 1869 was invalid because the Act required the disclaimer to be made “in writing under the hand of” the bankruptcy trustee whereas the disclaimer had been signed by the trustee’s solicitor in his own name and was said simply to be made “on their [the trustee’s] behalf”.
In Frischmann, instead of the assignor himself signing the assignment, it had been signed by means of the signature of the assignee “for and behalf of [the assignor] by way of a lasting power of attorney” (and also by the assignee for himself). The assignee argued that the effect of section 7(1) of the Powers of Attorney Act 1971 (“POA”) was that the assignee’s execution of the assignment as attorney for the assignor was as legally effective as if the assignor himself had personally signed the assignment.
The Powers of Attorney Act 1971
Section 7(1) says:
“If the donee of a power of attorney is an individual, he may, if he thinks fit – (a) execute any instrument with his own signature, and (b) do any other thing in his own name, by the authority of the donor of the power; and any instrument executed or thing done in that manner shall … be as effective as if executed by the donee in any manner which would constitute due execution of that instrument by the donor or, as the case may be, as if done by the donee in the name of the donor”.
The judge held that the previous case law meant that the execution of the assignment by the assignee as attorney for the assignor was not sufficient for it to count as being made “under the hand of the assignor”. Section 7(1) should not “be treated as rewriting [the LPA], in the absence of any express reference to the earlier statute”.
Under the hand of the assignor or its attorney?
A few observations might be made briefly about this approach and decision. First, is it really so significant that the POA does not expressly refer to every statutory requirement for the signature of the principal to be applied to a document as being “overridden” by section 7(1), considering the breath of the language in that section, including that, effectively, the attorney’s signature is to be as effective as the due execution of the document by the donor?
Secondly, it is worth noting that where the assignor is a company, the LPA expressly accepts execution on behalf of the assignor by its attorney (see section 74). The judge was not prepared to add to that “attorney execution” regime, relying on the POA, in the same way that Eyre J seems to have been prepared to add to the statutory presumption of due execution in Aecom v Lendlease, relying on the doctrine of ostensible authority and estoppel.
Thirdly, what is the policy reason for not allowing legal assignments by an individual to be signed via an attorney? Especially when, as a matter of policy, the legal equivalence of the signing or execution of documents by an attorney with execution or signing by the donor of the power of attorney seems to have been recognized in section 7(1) of the POA.
Fourthly, although signing via an attorney is, as a matter of law, signing by an agent, there was still a difference in “form” between the trustee’s solicitor writing a letter in which he said he was disclaiming an asset on behalf of his client and signing that letter in his own name as the client’s solicitor (Wilson v Wallani) and (in Frischmann v Vaxeal Holdings) the assignee signing as formally appointed attorney for the assignor (albeit also using his own name, instead of the donor’s name, which the POA expressly permits). Perhaps another point to note, even if this would likely not have changed the judge’s analysis, is that the attorney could have signed in the name of the assignor, rather than in his own name, creating a further difference in “form” from the Wilson v Wallani signing. At common law, that is how an attorney would normally sign a document and the POA merely allows an attorney to sign in her own name but does not require that.
Precedential standing of this decision
Finally, being a decision of a “master”, essentially a “junior” High Court judge, there is an interesting question of the precedential standing of this ruling. Recent cases have not been unanimous on the question of whether decisions of masters and of “full” (including deputy) High Court judges have exactly the same status, meaning that while a judge is not bound to follow an earlier decision of her fellow judge, as a matter of “judicial comity”, unless she is of the view that the earlier decision was plainly wrong as a matter of law, it will be followed.
In Coral Reef Ltd v Silverbond Enterprises Ltd Master Matthews took the view, following an earlier master’s decision in Randall v Randall, that that there was no difference in terms of precedential standing between a master’s decision and that of a High Court judge and so a master should be free to prefer an earlier master’s decision over that of a judge’s ruling. However, on appeal from the master’s decision in Coral Reef Ltd, the deputy High Court judge (Foxton QC) took the view that a decision of a High Court judge “in terms of its clear ratio is binding on a master, absent either conflicting decisions of another judge at the same level of the High Court, or obviously of superior courts”. This will only be relevant to any future developments of the law in this area if a High Court judge takes a different view about the ineffectiveness of signing legal assignments under section 136 via powers of attorney and a later master is tempted to take her own, different view from that High Court judge’s decision.
Lessons to be learnt.
However debatable this ruling is that legal assignments, at least by individuals, cannot, if they are to have the benefits that section 136 LPA provides, be signed by an attorney, clearly the sensible advice must currently be to require the assignor herself to sign the document and not rely on an attorney to do that. End of story, as they say.