So, what are we to make of the decision in K/S Victoria Street v House of Fraser (Stores Management) Ltd & Ors [2011] EWCA Civ 904; perhaps the most important commercial landlord and tenant case of the decade so far. You will no doubt be aware that this is the case that seeks to bring a conclusion to the so-called “AGA saga” under the Landlord and Tenant (Covenants) Act 1995, by providing an answer to the question of whether an assigning tenant’s guarantor, under a post-1995 Act tenancy, can lawfully guarantee performance of tenant covenants by an incoming assignee – either by way of a direct guarantee, or indirectly by guaranteeing the outgoing tenant’s guarantee of the assignee in the AGA. The High Court decision in Good Harvest Partnership Llp v Centaur Services Ltd [2010] EWHC 330 (Ch) had held that the direct guarantee of an assignee was rendered void by section 25 of the Act in that it frustrated the guarantor’s release from liability on lawful assignment under section 24(2) of the Act. Many property lawyers (including this one) believed that, by analogy, the indirect guarantee must also be void. The Court of Appeal, however, thinks otherwise.
What exactly did the CA decide?
The ratio decidendi of the K/S Victoria case is limited to two issues: (1) whether a provision in an agreement for lease requiring an assignment of the lease by the original tenant to a specific assignee, on terms that the original tenant’s guarantor would guarantee the assignee, was valid: and (2) whether the drafting of the lease alienation clause removed, in certain circumstances, not only the need for landlord’s consent to assign, but also the need to satisfy other pre-conditions of the clause. Issue (2) does not involve “Good Harvest” points and is therefore not considered in this note.
Issue (1) relates specifically to a direct guarantee of an assignee by the original tenant’s guarantor. The Court of Appeal held that any provision (such as the provision in the agreement for lease under consideration in this case) which stipulates in advance that a tenant’s guarantor, who is to obtain a statutory release on assignment under section 24(2), must re-assume that liability as a term of the assignment, would frustrate the operation of section 24(2)(b), and is therefore rendered void by section 25(1)(a). Any contractual arrangement contained in the tenancy (or in a prior agreement), which imposes an obligation on an existing or prospective guarantor of the tenant’s liabilities, to guarantee the liabilities of a future assignee is void. This conclusion would also apply to a contractual arrangement contained in a later document, for instance where the re-assumption of liability was imposed on a guarantor of an assignee’s liabilities in a later assignment or a licence to assign. Lord Neuberger summaries this by saying that, subject to what he has to say about the AGA saga (see below), an existing or contracting guarantor of a tenant cannot validly be required to commit himself in advance to guarantee the liability of a future assignee, and that a guarantor of an assignor cannot validly guarantee the liability of the assignor’s assignee. It does not matter that the original tenant’s guarantor might be perfectly happy to guarantee the assignee. Section 25 operates on any agreement which would “have effect to… frustrate the operation” of any other provision of the Act. Motive is therefore irrelevant – it is the effect of the agreement that is important. So, the direct guarantee of the incoming tenant is unlawful.
Issue (1) did not involve a consideration of the validity of an indirect guarantee – the real AGA saga. Discussion of that issue is therefore strictly obiter, although pretty powerful obiter, to say the least. Powerful, since Lord Neuberger carries the Court of Appeal with him in a single opinion to which, he says, all members have contributed. He says it is an attempt to provide some certainty in landlord and tenant law.
What did the CA say about the indirect guarantee?
Despite the central finding, the Court of Appeal is happy for a tenant’s guarantor indirectly to guarantee the incoming assignee if it does so through the medium of an AGA. Under section 24(2), an outgoing tenant’s guarantor is released “to the same extent” as the outgoing tenant. The tenant’s release is, however, qualified by section 16 in that an outgoing tenant can be required to guarantee the incoming tenant in the form of an AGA. It is therefore not inconsistent with section 24(2) for a landlord to require the outgoing tenant’s guarantor to guarantee the liability of the tenant under the AGA – i.e. the indirect guarantee of the assignee. The release will operate “to the same extent”: “The conclusion that the assignor’s guarantor can be required to guarantee the liability of the assignor under an AGA appears to comply with commercial sense and … is also consistent with the purpose of an AGA.” [Para 48]. “[W]hile the guarantor of an assignor cannot normally validly guarantee the liability of the assignee, it can validly do so by being party to a valid AGA.” [Para 49].
The case can therefore be seen in the property world as bringing to an end the AGA saga, validating the drafting of many landlords who, since the arrival of the 1995 Act, have been drafting AGAs to include covenants that are supported by the covenant of the assigning tenant’s guarantor. Keeping the tenant’s guarantor on the hook by requiring it to guarantee the AGA does not, says the Court of Appeal, frustrate the guarantor’s release from liability under section 24(2) of the Act.
Is this the end of the road?
If you like, you can just say “yes” and not read any more. No application was made in the Court of Appeal for leave to appeal to the Supreme Court. At the time of writing, it is not known whether an application for leave will be made to the Supreme Court. Certainly, Lord Neuberger’s opinion appears to have been delivered with a view to finality. He says at the end of his opinion: “We would hope that those responsible for drafting leases are aware of these conclusions, and that, as a result, the 1995 Act should not lead to many practical difficulties of the sort discussed above.” [Para 53].
A sensible approach might be to accept, and to move on. However, we are natural analysts and critics, and so the property journals and blogs will now go into overdrive. And the problem for Lord Neuberger is that he has had to make sense out of what is undoubtedly a defective piece of legislation. One finding that would have achieved certainty was the opposite, literal interpretation of the 1995 Act: i.e. that on lawful assignment, a guarantor is released by section 24(2) and that any attempt to prolong the guarantor’s liability would run up against the formidable terms of section 25 – end of story. However, that would not have satisfied either end of the property industry. He has therefore instead attempted a commercial commonsense approach to statutory interpretation, but one that was, because of the inadequacy of the legislation, bound to leave just one or two open questions …such as:
•Was Lord Neuberger’s approach to interpretation correct?
Underpinning Lord Neuberger’s interpretation is effect of the phrase “to the same extent as the tenant” in section 24(2): when the tenant is released, the guarantor is released “to the same extent”. To what extent is a tenant released from tenant covenants under the Act? Lord Neuberger regards a tenant as being released either without a requirement for an AGA, or subject to an AGA. Hence, when the guarantor is released “to the same extent” the guarantor is either released without having to guarantee the tenant’s AGA, or subject to the tenant’s AGA.
However, this does not appear to be the intended use of the phrase “to the same extent as the tenant” in section 24(2). A close analysis of the Act is required to ascertain to what extents tenants are released by the Act from the tenant covenants of the tenancy. Under section 5(2), on an assignment of the whole of the premises demised, the tenant is released from the tenant covenants of the tenancy. Under section 5(3), where the tenant assigns part only of the premises demised, he is released from the tenant covenants “only to the extent that those covenants fall to be complied with in relation to that part”. So, where section 24(2) states that the guarantor is released from a tenant covenant “to the same extent as the tenant is released from that covenant”, section 24(2) is clearly relating back to the whole/part release provisions of section 5. (This point seems to read much more clearly in the draft bill attached to the Law Commission’s 1988 Report (No 174), although not a lot can be read into the draft bill and report given the numerous amendments made to the bill before its enactment in 1995).
•What did Lord Neuberger mean when he suggested that assignments to tenants’ guarantors could be void?
Applying the ratio of the case, that section 25(1) invalidates any agreement which involves a guarantor of the assignor guaranteeing that assignor’s assignee, Lord Neuberger said [para 37] that this interpretation “would mean that, even where it suited the assignor, the assignee and the guarantor that the assignee should have the same guarantor as the assignor (because, for instance, the assignor and the assignee had the same parent company, or shared a common bank, which was the guarantor), they could not offer that guarantor. It would also appear to mean that the lease could not be assigned to the guarantor, even where both tenant and guarantor wanted it.” Are assignments by tenants to their guarantors void? What about those that have taken place in the past? Are the registrations of those transfers now to be corrected? This comment must surely be a throw-away remark amongst obiter comments which was uttered in a process of attempting to support an alternative interpretation (called “interpretation (ii)” in the opinion) which is later rejected.
•How does Lord Neuberger manage to prohibit a direct guarantee, but allow a co-guarantee in an AGA?
Lord Neuberger says: “It may well be that the guarantor could simply act as a co-guarantor under the AGA with the assignor, as might have been the position in Good Harvest [2010] Ch 426 – see at para 5 – rather than being a guarantor of the assignor’s liability, under the AGA.” [Para 47]. It would be hard to explain to the lay person that a direct guarantee of an assignee is void, but that a co-guarantee of the assignee is fine if it is located in another document (the AGA). Is it OK simply because it is in an AGA? But if the guarantor is co-guaranteeing the assignee under the AGA, is the agreement still an AGA? Could the landlord then subsequently release the tenant from the AGA, and leave the guarantor on its own? Or would section 25 then bite?
•Should a “commercial commonsense” approach to statutory interpretation override a literal interpretation and an application of the mischief rule?
The mischief at which the 1995 Act was aimed is made quite clear by the preamble to the Act. The preamble to the Act states that the 1995 Act is: “An Act to make provision for persons bound by covenants of a tenancy to be released from such covenants on the assignment of the tenancy.”
For tenants, section 5 provides the release at lawful assignment. For others, such as guarantors, section 24(2) provides the release at the same time. To compensate landlords for the loss of original tenant liability, section 16 allows assigning tenants to enter into AGAs to guarantee the obligations of incoming tenants. If the legislators had their time again, they would surely have provided that tenants’ guarantors could also enter into AGAs to guarantee the tenant’s obligations in it – but, crucially, they didn’t. It is interesting to note how, in sections 17 to 19 of the 1995 Act, the legislators expressly referred (both in section headings, and text) to former tenants and their guarantors. But section 16 only refers to tenants. Section 16 states that nothing in the Act precludes an assigning tenant from entering into an AGA; but it does not say that nothing precludes its guarantor from joining in as well. An AGA is an agreement under which “the tenant guarantees … performance … by the assignee”; but it does not say that it is an agreement under which the tenant, along with its guarantor, guarantees performance by the assignee. The whole tenor of section 16 is that assigning tenants enter into AGAs, and that other persons do not. This is the overriding defect in the Act: the legislators forgot to refer to tenants’ guarantors in section 16. Lord Neuberger has manfully, some would say masterfully, managed a silk purse from a sow’s ear. Lord Denning once said (in Magor and St Mellons RDC v Newport Corporation [1950] 2 All ER 1226) that: “We do not sit here to pull the language of Parliament to pieces and make nonsense of it. We sit here to find out the intention of Parliament and carry it out and we do this better by filling in the gaps and making sense of the enactment than by opening it up to destructive analysis.” This approach was rebuked by the Law Lords, Lord Simonds stating that “..if a gap is disclosed the remedy lies in an amending Act”.
There is a gap in section 16 LTCA 1995. Parliament should fill it. Maybe Lord Neuberger has?