Tan & Anor v Sitkowski

The question to be decided in this appeal is whether Julian Sitkowski (“the appellant”) enjoys Rent Act protection in relation to his tenancy of 335 Battersea Bridge Road, London, SW11 (“the premises”). This simple question based as it is on straight forward facts, requires a relatively difficult investigation and analysis of the relevant legislation, which is somewhat opaque, and of the relevant case law, which is often unsatisfactory and inconsistent.

Before 1970, the appellant occupied property in York Road, London, SW11, where he used the ground floor for his electrical retail business and the first floor as a residence with his family. In 1970, the York road property was acquired for demolition in connection with road-widening purposes, and the appellant was re-housed in the premises by the Greater London Council (“GLC”).

At that time, the premises were in a somewhat dilapidated condition. They were recorded in the rating list as “shop flat and premises” and appear to have been physically adapted for use as a shop on the ground floor and living accommodation on the first floor. It seems clear from the evidence that both the appellant and the GLC anticipated that he would live in the first floor with his wife and children, and would occupy the ground floor for his business as a retailer and repairer of electrical goods. That is the effect of his evidence, and it is entirely consistent with the fact that that is how he had used the York Road property.

Those were indeed the uses to which the premises were put from 1970, after the appellant had carried out works of repair and improvement, until 1989. In that year, the appellant ceased his business use of the ground floor. Subject to having had to vacate the premises to enable substantial repair works to be carried out, he continued to occupy the first floor as his home, and he used the ground floor for storage in connection with his residential use of the first floor.

Although the appellant’s first tenancy of the premises was granted in 1970, the primarily relevant tenancy for present purposes was a written agreement (“the Agreement”) dated 19 August 1976, whereby the GLC granted the appellant an annual tenancy at a rent of £650 per annum. The premises were described in the Agreement as “all that shop and premises known as 335 Battersea Park Road…”.

The only covenant of relevance in the Agreement was contained in clause 3(b) whereby the appellant agreed:

“not to use or permit to be used the premises or any part thereof for any purpose other than that described in the second schedule nor for any illegal or immoral purpose nor in such a manner as to cause a nuisance or annoyance… and nor to allow the premises to be used for sale by auction or for the sale of beer, wine, cider or spirits…”.

The purpose described in the second schedule was use as a retailer and repairer of televisions and radios.

The only other provision in the Agreement to which I should refer is the proviso for re-entry which, as usual, entitled the landlord to forfeit the tenancy in the event of the tenant being in arrear with the rent or being in breach of any of his covenants.

In the late 1980s, the GLC was abolished, and the freehold reversion of the premises became statutorily vested in the London Residuary Body (“LRB”). In 1990, the reversion was sold by the LRB to Phaik Seang Tan and Kit Yeng Tan (“the respondents”). The respondents accordingly became the appellant’s landlords. In October 2003 they served various notices to quit on the appellant, and there seems to me no doubt that at least one of the notices to quit was effective to determine the appellant’s tenancy of the premises by November 2004. For most, or at any rate much, of the period during which the respondents were the appellant’s landlords, the rent was paid by the local housing authority direct to the respondents through Housing Benefit.

The relationship between the respondents and the appellant was not good, and there has been previous litigation between them. The instant claim, brought by the respondents against the appellant, for possession and other relief, was met by a counterclaim. The proceedings came before His Honour Judge Brian Knight QC in the Central London County Court. Most of the issues in the proceedings are irrelevant for present purposes, as they were determined by the Judge adversely to the appellant, and permission to appeal was refused by Chadwick LJ on 17 June 2005.

However, there was one issue on which Chadwick LJ had some concern, and (albeit on a rather wider basis) it is the issue which is before us. As a result of directions given by Jacob LJ and myself, we have to decide whether to give permission to appeal on this issue, and, if we do, we should determine the appeal. That issue is whether the Judge was right to decide that, following the expiry of the notice to quit to which I have referred, the appellant was not a statutory tenant under the Rent Act 1977 (“the 1977 Act”).

If, as the Judge concluded, the appellant was not the statutory tenant of the premises, then his appeal must be dismissed, and that is the end of the case. If, on the other hand, the Judge was wrong, and the appellant was the statutory tenant of the premises, the case will have to be remitted to the County Court. First, there is an outstanding claim for possession based on case 3 of Schedule 15 of the 1977 Act, which would have to be determined. Secondly, the appellant would contend that he had been wrongfully deprived of his statutory tenancy (for reasons which it is unnecessary to discuss at this stage) and, as I understand it, he would be claiming damages.

As is well known, there is a statutory scheme which protects business tenants, and other statutory schemes which protect residential tenants. The scheme protecting business tenants is contained in Part II of the Landlord and Tenant Act 1954 (“the 1954 Act”). A tenant whose tenancy falls within the 1954 Act can, unless the landlord can establish a stipulated ground of opposition (which often involves paying the tenant compensation), claim a new tenancy at a market rent, when his tenancy expires.

The effect of section 23(1) of the 1954 Act is that a tenant can claim the benefit of that Act in relation to a tenancy if and so long as he occupies at least part of the premises comprised in the tenancy for the purposes of the business. An important feature of this provision is that protection under the 1954 Act depends on whether the premises concerned are actually being occupied for business purposes. If and when business use ends, there is no protection under the 1954 Act, but protection can be, as it were, revived by restarting the business use during the currency of the term.

From 1915 until 1989, when the Housing Act 1988 came into force with prospective effect, private sector residential tenancies were (subject to varying categories of irrelevant exceptions) governed by a succession of so-called Rent Restriction Acts. The relevant statute in force when the appellant moved into the premises, and indeed when the 1976 tenancy agreement was granted, was the Rent Act 1968 (“the 1968 Act”). However, subsequently it was replaced by the 1977 Act, which was a consolidating statute. For present purposes at any rate, the provisions of the 1968 Act were to all intents and purposes identical to those of the 1977 Act. Accordingly I shall refer only to the 1977 Act, save where it is necessary to refer to the 1968 Act separately.

Section 1 of the 1977 Act provides that:

“A tenancy under which a dwelling house (which may be a house or part of a house) is let as a separate dwelling is a protected tenancy for the purposes of this Act.”

Section 2 goes on to provide that, where a protected tenancy is brought to an end (e.g. by a notice to quit) the tenant thereunder becomes a statutory tenant, “if and so long as he occupies the dwellinghouse as his residence”. Protected tenancies (i.e. contractual tenancies which are within the 1977 Act) and statutory tenancies (which arise when protected tenancies determine) are known as regulated tenancies.

A landlord cannot obtain possession against a regulated tenant unless he establishes one or more of the grounds set out in section 98 of, and Schedule 15 to, the 1977 Act. Schedule 1 to the 1977 Act gives the right to certain members of a regulated tenant’s family to succeed to the tenancy on the tenant’s death. The 1977 Act enables a regulated tenant to pay a significantly lower rent than normal market forces would justify.

Section 24(3) of the 1977 Act is in these terms:

“A tenancy shall not be a regulated tenancy if it is a tenancy to which [the 1954 Act] applies (but this provision is without prejudice to the application of any other provision of this Act to a sub-tenancy of any part of the premises comprised in such a tenancy)”.

Accordingly, for present purposes, there are two important qualifying conditions for a tenancy to be a regulated tenancy within the 1977 Act. First, it must be a tenancy of “a dwelling house” “let as a … dwelling”; secondly, it must not be a tenancy which is subject to the 1954 Act.

It is clear that, when the GLC granted the appellant a tenancy in 1970, and indeed when they granted him a second tenancy in 1976, the tenancy concerned enjoyed the protection of the 1954 Act. That is because the tenant, namely the appellant, occupied part of the premises, namely the ground floor, for business purposes. However, as the Judge found, the appellant permanently ceased using any part of the premises for business purposes in 1989. Accordingly, the 1954 Act no longer applied to the tenancy.

In those circumstances, the appellant’s primary case is that section 24(3) of the 1977 Act no longer applied, and he enjoyed the protection of the 1977 Act. In other words, he contends that, by the time the notice to quit expired in 2004, he was a protected tenant, and therefore became a statutory tenant on the expiry of the notice to quit.

(It is right to say that the appellant appeared below and in this court in person, and the arguments that I am attributing to him have been developed on his behalf partly by the Court and partly by Mr Joshua Swirsky, who appears for the respondents, and who has presented his submissions in this difficult appeal in a concise and clear way.)

On behalf of the respondents, Mr Swirsky argues that the appellant’s primary argument should be rejected because there is authority binding on this court to the effect that, if a tenancy is granted for mixed business and residential use, and is accordingly subject to the 1954 Act, the tenant cannot, simply by unilaterally ceasing the business use, arrogate to himself protection under the 1977 Act. He says that, in such a case, the tenancy in question is not to be treated as being one of “a separate dwelling” within the meaning of section 1 of the 1977 Act.

Thus, the first, and to my mind the main, question on this appeal is as follows. Where premises are let for mixed business and residential use, and the tenant subsequently ceases the business use, can he claim protection under the 1977 Act in relation to the tenancy? If the answer to that is in the negative, then the appellant raises a second argument (which the Judge considered and rejected). Even if his tenancy could not be treated as a letting of “a separate dwelling” from or before 1989, he contends that it had become such a tenancy by the time that the notice to quit expired in 2004, and accordingly he is nonetheless entitled to claim a statutory tenancy. I shall consider those two arguments in turn.

Mr Swirsky’s contention is that, where premises are let partly for a use falling within the 1954 Act and partly for residential purposes (which I shall refer to as mixed business and residential use), they do not constitute a “dwelling house… let as a separate dwelling” within the meaning of section 1 of the 1977 Act. Consequently, he says, the fact that the business use subsequently ceases cannot, subject to one point, enable the tenant to claim that his tenancy has thereby become protected under that Act. The exception I have mentioned founds the basis of the second issue in this appeal, and it arises where, at the time of, or following the cesser of, the business use, the nature of the letting has in some way been altered. I shall say nothing more about that exception in this section of my judgment.

As mentioned above, the 1977 Act was the last in the line of a number of Rent Restriction Acts, enacted over a period of more than fifty years. The first of those statutes was the Increase of Rent and Mortgage Interest (War Restrictions) Act 1915 (“the 1915 Act”) which heavily restricted the right of a landlord to obtain “possession of a dwelling house”, if it was “a house or part of a house let as a separate dwelling” – see sections 1(3) and 2(2).

In Epsom Grandstand Association Ltd –v- Clarke (1919) 35 TLR 525, the Court of Appeal had to consider whether a property, let and used as a public house on the ground floor with a flat above, fell within the ambit of the 1915 Act. Bankes LJ (with whom Scrutton and Atkin LJJ agreed) said at page 526 that the property “was a dwelling house, and nonetheless so because it was also a public house”. His reasoning was that:

“The object of the legislature was to include all houses which were occupied as dwelling houses… irrespective of whether the premises were also used for some other purpose. They came within the statute, although part of the premises might be used for other purposes.”

The 1915 Act was extended and amended by the Increase of Rent and Mortgage Interest (Restrictions) Act 1919, and both statutes were then repealed and replaced by the Increase of Rent and Mortgage Interest (Restrictions) Act 1920 (“the 1920 Act”). Sections 1(1) and 12(2) provided that, like the 1915 Act, the 1920 Act applied to a “dwelling-house”, which was “a house or part of a house let as a separate dwelling”. This was then amplified by subsequent paragraphs of section 12(2), one of which, section 12(2)(ii), provided that:

“The application of this Act to any house or part of a house shall not be excluded by reason only that part of the premises is used as a shop or an office or for business, trade, or professional purposes”.

In Hicks –v- Snook (1928) 27 LGR 175, Scrutton LJ (with whom Greer and Sankey LJJ agreed) said at page 177 that the effect of section 12(2)(ii) of the 1920 Act appeared to him “to affirm in statutory form what the Court of Appeal had decided [in the Epsom Grandstand case].” Accordingly, where premises, let under a verbal agreement without any specific stipulation as to use, had been used throughout the twenty five years of the term by the tenant, as a shop on the ground floor and a residence on the first floor, the commercial use of part did not “stop the premises in which he lives from being a dwelling house”.

Following the 1920 Act, various statutes, most notably the Rent and Mortgage Interest Restriction Act 1939 (“the 1939 Act”), from time to time amended the provisions of the 1920 Act. Then, by the Rent Act 1957, many properties were forthwith taken out of protection, and the remainder were subject to provisions for gradual decontrol.

The only case decided during that period to which it is necessary to refer is Wolfe –v- Hogan [1949] 2 K.B. 194. In that case, the tenant had taken a tenancy of a large room on the basis that she would use it “as a shop and not as a dwelling house” although there was no express prohibition of the use as a dwelling house – see at page 200 per Evershed LJ. She initially used the room as a shop, but then started to live there. The question was whether, in those circumstances, she could claim the benefit of the 1920 Act.

The Court of Appeal held that she could not, because the premises did not constitute a “dwelling house… let as a separate dwelling” within section 12(2) of the 1920 Act. The principle applied by the Court was described in the judgment of Denning LJ at 204-205:

“In determining whether a house or part of a house is ‘let as a dwelling’ within the meaning of the Rent Restriction Acts, it is necessary to look at the purpose of the letting. If the lease contains an express provision as to the purpose of the letting, it is not necessary to look further. But, if there is no express provision, it is open to the court to look at the circumstances of the letting. If the house is constructed for use as a dwelling-house, it is reasonable to infer the purpose was to let it as a dwelling. But if, on the other hand, it is constructed for the purpose of being used as a lock-up shop, the reasonable inference is that it was let for business purposes. If the position were neutral, then it would be proper to look at the actual user. It is not a question of implied terms. It is a question of the purpose for which the premises were let.”

In 1965, protection for residential tenants was re-introduced by the Rent Act 1965 (“the 1965 Act”). It involved somewhat different terminology. Tenancies protected under the previous Rent Restriction Acts were called controlled tenancies, whereas those protected under the 1965 Act (and subsequent Acts) were, as mentioned above, called regulated tenancies. (Controlled tenants had rather more protection with regard to rent; controlled tenancies have been converted into regulated tenancies as a result of legislation over the subsequent years).In general, under the 1965 Act, existing tenancies which were still within the ambit of the Rent Restriction Acts remained controlled tenancies, whereas tenancies which were not controlled became, in the same way as tenancies granted after the 1965 Act came into force, regulated tenancies.

Effectively by reference to the previous Rent Restriction Acts, the 1965 Act conferred protection of a tenancy if it was of a “dwelling house… let as a separate dwelling”. However, the position with regard to mixed business and residential use tenancies was changed, so that the law applicable to regulated tenancies was different from that applicable to controlled tenancies. Section 3(3) provided that the 1965 Act would not apply in a case where the 1954 Act applied. In other words, in the case of mixed business and residential use premises, at least so long as the tenant continued to use part of the premises for business purposes, the 1954 Act would apply, and the 1965 Act would not apply.

The previous Rent Restriction Acts were subsequently (albeit substantially prospectively) repealed and substantially codified by the 1968 Act”). Some subsequent amendments were made to that Act, principally by the Rent Act 1974, and for the last time in relation to protected tenancies, the law was re-codified in the 1977 Act. I have already set out the relevant provisions of that Act, which are, as mentioned, for present purposes, the same as the provisions of the 1968 Act.

The next case I should refer to is Maunsell –v- Olins [1975] A.C. 373, which concerned the question whether a regulated sub-tenant could claim the protection of the 1968 Act where the head tenancy, which consisted of a farm, had come to an end. The issue turned on the meaning of the word “promises” in section 18(5) of the 1968 Act (now section 137(3) of the 1977 Act, albeit that it was amended following the decision in that case). The majority of the House of Lords held that, in its statutory context, the word “premises” was to be given a relatively limited meaning, which Lord Wilberforce explained at pages 388G to 389A in the following terms:

“So what should "premises" be taken to mean? One view, the narrowest view, would be that it simply means "dwelling-houses."... A less narrow view would be to say that "premises" includes not only dwelling-houses in the normal popular sense, but premises, which, for the purposes of the Rent Act, are treated as dwelling-houses. Everybody knows, and the draftsman must be taken to have known, that protection under the Rent Acts is given not merely to single, identifiable, pure dwelling-houses or dwelling units, but also to units of a mixed character - houses let with a garden or a yard or a garage or a paddock, houses part (even a substantial part) of which is used for business purposes.”

This latter meaning is the one which he, Lord Reid and Lord Cross of Chelsea adopted in that case.

The next case which should be dealt with is Pulleng –v- Curran (1980) 44 P&CR 58. In that case, a tenancy of mixed business and residential use premises had been granted, and, following its contractual determination, the tenant had exercised his rights under the 1954 Act to acquire a new tenancy. Subsequently, he claimed to have ceased the business use, and contended that he was entitled to protection under the 1977 Act. The Court of Appeal rejected that argument. The first reason, which is not germane for present purposes, was that the trial judge was entitled to find that the business use had not entirely ceased, as a result of which the 1954 Act applied, and, in light of s.24(3) of the 1977 Act, the latter statute did not.

The Court, however, went on to conclude that there were two further reasons why the tenant’s appeal should fail. The second of those further reasons is also not germane for present purposes. It was based on the proposition that, having taken the benefit of the 1954 Act to acquire a new tenancy, the tenant could not then switch to claim the benefit of a different statutory regime. It seems to me that that conclusion can be fairly said to be based more on policy than on any specific provision in the 1954 Act or in the 1977 Act.

The first of the two further reasons for rejecting the tenant’s case is, however, very much in point. It was that the tenant could not claim the protection of the 1977 Act, because the tenancy had been for mixed business and residential uses, and it was not right in principle or appropriate in practice that the tenant should be able to switch his protection from the 1954 Act to the 1977 Act simply by ceasing to use the premises for business purposes.

It must be said at once that, with one exception, I find the reasoning of Sir George Baker P, who gave the leading judgment, on this issue at pages 68 to 70 incomprehensible. The only point I understand, which again appears to me to be founded on policy, or perhaps on what is sometimes called the statutory scheme, rather than on the meaning of any specific statutory provision, is to be found at page 69, where Sir George said this:

“[I]t certainly strikes me as a most remarkable conclusion if a tenant, by simply ceasing to carry on his business …, could then say: "I am now in a position that I have the shop and all the premises subject to the Rent Restriction Acts; we have moved under that umbrella, and you, the landlord, can whistle for possession." He might indeed, if he was so minded, leave the shop to rot and simply confine himself to his upstairs premises.

The corollary, it seems to me… is that the tenant could stop, start, stop, start, as long as he liked, juggling between the two Acts of Parliament.”

Towards the end of his judgment, Sir George Baker also relied on the Wolfe case. However, it seems to me that decision has nothing to do with the issue in the Pulleng case. As already mentioned, the letting in the earlier case was of premises for purely business use, and therefore, on no view, were the premises “let as a separate dwelling”, in the light of the test explained by Denning LJ.

Cumming-Bruce LJ agreed that the appeal in the Pulleng case should be dismissed, at page 70, “for the reasons stated by Sir George Baker”. He developed his reasons on the relevant issue at pages 72 to 73. Again, I find his reasoning very hard to understand, save insofar as it relied on the Wolfe case (which was irrelevant for the reason I have explained) and on policy.

The final judgment was given by Stephenson LJ. His reasoning appears, at least to me, to be easier to understand. At page 74, he said that the letting in that case “was not a letting of tenancy as a residential dwelling”. However, he then went on to base his reasoning in part on the Wolfe case, although he also referred to an observation of Lord Denning MR in Cheryl Investments Ltd –v- Saldanha [1978] 1 WLR 1329, which undoubtedly supported his conclusion. However, Lord Denning’s observation was plainly an obiter dictum, and was not supported (or, it is fair to say, disclaimed) by the other two members of the court. It is therefore of very limited authority, even bearing in mind the eminence of its source.

I now turn to Wagle –v- Trustees of Henry Smith Charity [1990] 1 Q.B. 42. In that case, premises had been let for mixed business and residential use, and the tenant, having ceased the business use, claimed to be protected under the 1977 Act. The Court of Appeal rejected that contention.

The leading judgment was given by Dillon LJ. He made it clear at page 45A that his conclusion was reached on the basis “that this Court is bound by previous Court of Appeal decisions”. He first relied on Lord Denning’s obiter dictum in the Cheryl case, which he quoted at page 47H. I have already dealt with that: it cannot be ignored, but it can only carry a little weight.

Dillon LJ then referred to the Epsom Grandstand case and said that, because it was poorly reported, it could not carry much authority. Unfortunately, he does not seem to have appreciated that that case was strongly endorsed by this court in the fully reported Hicks case. Dillon LJ then referred to the Wolfe case, which, as I have explained, is of no relevance. He then referred to the Pulleng case, which he held to be binding authority.

Finally, he referred to Russell –v- Booker [1982] 2 EGLR 86 which appears to bear superficial similarity with a mixed business and residential use case, because the letting there was for mixed agricultural and residential purposes. The agricultural use ceased, whereupon the tenant claimed protection under the 1977 Act. However, it seems to me that cases involving agricultural holdings provide no assistance, because, as explained in the Maunsell case, premises cannot be treated as “a dwelling house” where the letting concerned is of an agricultural holding (even if it includes a dwelling), as in the Maunsell case itself. (Although the effect of that case was reversed by legislation in relation to sub-tenancies of dwelling houses, the reversal did not apply to or, at least in my judgment, bear on grants of tenancies of agricultural holdings).

The second judgment given in the Wagle case was that of Taylor LJ. His approach was rather different. He concentrated on the expression “let as a separate dwelling” in section 1 of the 1977 Act. He accepted that, under the 1915, 1920 and 1939 Acts, premises let for mixed business and residential use were “let as a separate dwelling”. He then turned to consider whether the meaning of the expression had changed by the time of the 1977 Act, given that (as under the 1965 and 1968 Acts) premises with such a mixed business and residential use would have been within the 1954 Act, and not the Rent Acts. The effect of his reasoning at pages 52G to 53B was that, as a result of that change of statutory policy, the meaning of the statutory words “let as a separate dwelling” had changed, so that they did not apply to premises let for mixed business and residential use.

Taylor LJ did not refer to the Pulleng case, but he did rely on what Lord Denning had said in the Cheryl case. The third member of the Court of Appeal in the Wagle case was Sir John Megaw, who agreed effectively on the basis that the Court of Appeal was bound by its earlier decision in the Pulleng case.

Very shortly after, the Court of Appeal had to consider a very similar point, albeit arising under the Housing Act 1985, in Webb –v- Barnet LBC (1988) 21 HLR 228. The only reasoned judgment was given by Taylor LJ, and his reasoning at pages 234 to 236 gives a somewhat more detailed explanation of why he considered that the meaning of the words “let as a separate dwelling” in the Rent Restriction Acts had a different meaning in the legislation passed after 1965 from that which it had in the legislation passed before 1954. He considered that his conclusion was supported by the obiter dictum in the Cheryl case, and the decisions in Pulleng, Russell and Wagle cases.

The next relevant case is the decision in Wellcome Trust Ltd –v- Hamad [1998] QB 638. In that case, overruling three of its previous decisions, this court held that a sub-tenant of purely residential premises could continue to claim the protection of the 1977 Act on the expiry of a head tenancy, which was of mixed business and residential use premises. The issue was therefore similar to that in the Maunsell case (save that the premises concerned were mixed business and residential use rather than mixed agricultural and residential use).

In reaching this conclusion, the Court of Appeal in the Wellcome case effectively decided that its conclusion followed from the reasoning in the Maunsell case, and, in particular, the observations I have quoted from Lord Wilberforce as to the meaning of “premises” in what is now section 137(3) of the 1977 Act: see at pages 647B-G, 654A to 655G and 657D to 658D.

In the last of those passages, at page 658A, Leggatt LJ, giving the judgment of the court, said this:

“[W]e must test the question whether property demised by a superior tenancy constitutes "premises" by asking whether it is a dwelling house within the extended meaning indicated by this court in the Epsom Grandstand case… and thereafter perpetuated in the cases to which we have referred, albeit reinforced by statute from time to time in the form of the provisos.”

Although the Pulleng and Russell cases were referred to in argument, the only one of the recent cases to which I have referred, which was discussed by the Court in the Wellcome case was the Wagle case. The Court in the Wellcome case pointed out at page 650E-F that the Court in the Wagle case appeared to be unaware of the Hicks case, which “fully confirmed the authority of [the Epsom Grandstand case]” and that accordingly those two relatively early decisions “set the scene, so far as the meaning of the word “dwelling house” is concerned, from then onwards”.

The final case which must be mentioned is Patel –v- Pirabakaran [2006] 1 WLR 3112. In that case, the Court of Appeal had to consider whether section 2 of the Protection of Eviction Act 1977, which only applies to premises “let as a dwelling”, extended to premises used for mixed business and residential purposes. In his judgment, Wilson LJ mentioned virtually all the cases I have discussed, and clearly had some doubts about the reasoning in the Pulleng,Wagle and Webb cases, but he noted at paragraph [30] that very experienced leading counsel for the tenant in the case before him accepted that those cases were binding on the Court of Appeal. At paragraph [31], Wilson LJ said:

“In the Wellcome Trust case, by one valuable judgment, this court explained that, in accordance with [the] line of authority [beginning with the Epsom Grandstand case], the premises had been let to the tenant 'as a dwelling' and that accordingly the defendant enjoyed protection against the claimant under the Rent Act 1977.”

In the end, however, the Court reached its conclusion in the Patel case, to the effect that premises let for mixed business and residential use were “let as a dwelling”, essentially on the basis of the proper construction of the Protection of Eviction Act 1977 (see paragraphs [34] to [37]).

(In case this part of the judgment is not confusing enough, it is right to add this final comment. The 1977 Act does not apply to residential tenancies granted after the Housing Act 1988 came into force. Further, public sector tenancies were outside the ambit of the 1977 Act and the 1988 Act, and are subject to the statutory protection code in the Housing Act 1985. In the light of the fact that the GLC were the appellant’s landlord until the late 1980s, and the respondents did not become the landlords until 1990, it might be thought that it would be the Housing Act 1988 Act, rather than the 1977 Act, would be the relevant statute for present purposes. However it is accepted on behalf of the respondents that, if the appellant is entitled to statutory residential protection, it would be under the 1977 Act. As I understand it, this view is based on the involvement of the LRB and the provisions of section 38 of the Housing Act 1988. Furthermore even if the latter Act, rather than the 1977 Act, applies, it seems to me that the centrally relevant provisions probably have the same effect – see section 1 of, paragraph 4 of Schedule 12 to, and ground 13 of Schedule 2 to, the Housing Act 1988.)

After that analysis of the statutory and case law history, I should start by dealing with one argument raised by the respondents although Mr Swirsky (quite rightly in my judgment) no longer pursues it, despite the fact that it was accepted below. The Judge considered that, because of the terms in which clause 3(b) of the Agreement was expressed, only business use as identified in the second Schedule to the Agreement was permitted in the premises, and that therefore there can have been no question of the premises having been, at least originally, let as a dwelling because no residential use was permitted.

In the light of the observations of Denning LJ in the Wolfe case, the terms of the lease must be scrutinised to decide whether the premises were “let as a separate dwelling”. I also accept that the primary meaning of the words of clause 3(b), if read on their own acontextually, is in accordance with the Judge’s conclusion. However, the provisions of the Agreement, like any other commercial contract, must be read in their context.

That context includes the following facts. First, the premises, when let (both initially and in 1976), were adapted for business use downstairs and residential use upstairs. Secondly, in 1970, the premises were offered by the GLC to the appellant in 1970 as a replacement for the York Road property, where he lived with his family upstairs and carried on his business downstairs, with the obvious intention that the premises be put to the same use, as indeed they were. Thirdly, by 1976, when the Agreement was entered into, the appellant was, and had for more than five years been, living on the first floor and carrying on business on the ground floor. Fourthly, no business could have been lawfully carried on in the first floor, at least without planning permission.

In those circumstances, it seems to me that clause 3(b) of the Agreement, insofar as it purports to restrict the use to a particular type of business, is and was only directed to the use of the ground floor of the premises. It was not concerned with the use of the first floor. Thus, on the facts as they were in 1976, it cannot have been the intention of the parties that the GLC should have been able to forfeit the tenancy on the ground of the appellant’s residential use of the first floor of the premises: yet that would be the result of giving clause 3(b) its strictly literal, or primary linguistic, meaning. Accordingly, in my opinion, that meaning should be rejected. In this connection, it appears to me that the present case, in terms of both the facts and the terms of the user covenant, is impossible sensibly to distinguish from Levermore –v- Jobey [1956] 1 WLR 697.

Mr Swirsky fairly accepts this proposition. But he certainly does not accept that this means that the premises can be said to have been let “as a separate dwelling” within the meaning of section 1 of the 1977 Act. I agree. All it means is that the appellant can at least get to first base in his argument on that point. In other words, when the Agreement in this case was granted, he can fairly say that it was intended and envisaged that part of the premises, namely the first floor, would be occupied for residential purposes by the tenant.

It is now necessary to face up to the main issue raised by this appeal. In effect, Mr Swirsky contends that we are bound by the decisions and reasoning in the Pulleng and Wagle cases to hold that the letting in the present case was not for premises “as a separate dwelling”, because the letting was for the purpose of mixed business and residential use, and that, in those circumstances, the appellant’s contention that he has a regulated tenancy of the premises must be rejected.

That submission has obvious force, but, as the decision and reasoning in the Wellcome case shows, decisions, even of fairly recent origin, of the Court of Appeal in this area are not immune from being overruled by this Court if we are satisfied that the reasoning does not bear analysis. Such decisions may be treated as having been decided per incuriam.

Mr Swirsky says that that is setting too low a hurdle for refusing to follow earlier decisions of this Court in light of these relatively limited exceptions to the stare decisis principle as laid down in Young –v- Bristol Aeroplane Co Ltd [1944] 1 KB 718. However, it seems to me that the effect of the Wellcome case (and other recent decisions of this court such as Starmark Enterprises Ltd –v- CPL Distribution Ltd [2002] Ch. 306) is that the circumstances in which a decision may be treated as being per incuriam may have become somewhat more generous and flexible than might have been perceived as being fifty years ago.

As I have, I hope, made clear when discussing the Pulleng, Wagle and Webb cases, it appears to me that much of the reasoning in those cases is either flawed or incomprehensible. However, in the Pulleng case, it does seem to me that at least Stephenson LJ specifically suggested that premises let for a mixed use purpose were not “let as a dwelling” for the purposes of the 1977 Act. Furthermore, Taylor LJ reached the same conclusion, more clearly and with an explanation, in the Wagle case. In addition, that was the fully argued and explained ratio of the decision of this court in the Webb case; although that case was concerned with the Housing Act 1985 rather than the 1977 Act, the reasoning was applicable to, and based on case law relating to, the Rent Restriction Acts.

This reason (unlike most of the other reasons in the judgments in the Pulleng and Wagle cases, which seem to me to be either incomprehensible or wrong) is one which appears to me to be tenable as a principle. The reason is attributable to a change in the legislative policy, and is justified on the following basis. Under the earlier legislation, premises could be treated as “let as a dwelling” if they were let for mixed residential and business purposes, because the 1915, 1920 and 1939 Acts applied to premises let for mixed business and residential uses. However, from 1965, the Rent Acts no longer applied to premises used for such mixed purposes, and applied only to premises used for purely used for residential purposes, so premises should not be treated as “let as a dwelling” unless they were let for purely residential purposes.

Furthermore, I consider that that conclusion can fairly be claimed to be supported by the practical considerations identified by Sir George Baker in the Pulleng case. Invoking policy can be reasonably said to be particularly justifiable in this context, as the reasoning in the original case on the topic, the Epsom Grandstand case, was based, at least in part, on policy, as the passage I have quoted from the judgment of Bankes LJ in that case demonstrates.

It is worth examining the policy consequences of the appellant’s argument a little further. It does seem somewhat anomalous that a tenancy granted for mixed business and residential use, which, when entered into, falls within the 1954 Act, can unilaterally be brought within the ambit of the 1977 Act simply by the tenant ceasing business use. It could fairly also be said to be somewhat unfair on the landlord.

The most common sort of case of mixed use premises, I would have thought, would be exemplified by the premises in this case, namely a ground floor shop with a flat above. It would seem both unfair on the landlord and generally socially undesirable, that a tenant of such premises, especially when (as is frequently the case) they are part of a parade of shops, should be able to shut down the business in the shop and continue in occupation of the whole of the premises for the rest of his life at an artificially reduced rent under the 1977 Act, with the possibility of passing on that right to successors (see Schedule 1 to that Act). It is true that the landlord might well be able to obtain possession of the shop part of the premises by contending that the upper part on its own constituted suitable alternative accommodation for the tenant (see section 98(1) of the 1977 Act) but, no doubt in many cases at any rate, the attraction of the premises to potential occupiers would be as a shop with a flat available for occupation above, rather than as a shop with a regulated tenant above.

It might be said that a landlord could protect himself by including in the tenancy agreement a positive obligation on the tenant to carry on business, but that is not a satisfactory answer. Not only could such a provision have a severely depreciating effect on the rent which the landlord would obtain, but, even if the tenant ceased the business use in breach of covenant, the landlord could not be sure of obtaining possession under the 1977 Act for that reason: the Court would still have to consider that it was reasonable to make an order for possession. Furthermore, it could be said to lead to an unreasonable distortion of the market for the legislation to have the effect of encouraging landlords to take positive user covenants for what might be said to be wholly artificial reasons.

Section 24(3) is not rendered redundant if Mr Swirsky’s argument is correct. It would still apply in a case where premises had been let for purely residential purposes, and the tenant subsequently uses them wholly or partly for business purposes. The fact that he could switch protection from the 1954 Act back to the 1977 Act by ceasing the business use does not undermine the policy argument raised in the Pulleng case. The protection, in terms of both security and rent, is much greater under the 1977 Act than under the 1954 Act. Accordingly, it would not be particularly odd if a tenant, who initially had the protection of the 1977 Act, could, by his unilateral action, switch to the 1954 Act, and could then unilaterally switch back to the 1977 Act whose protection he enjoyed initially; whereas it could fairly be said to be odd if a tenant, who initially had 1954 Act protection, could unilaterally switch to 1977 Act protection.

I do not consider that the respondents’ argument is undermined, although it is undoubtedly weakened, by the reasoning in the Maunsell or Wellcome cases. It seems to me that, as is illustrated by the passages I have quoted from those two cases, they were ultimately concerned with the meaning of “premises” and “dwelling house” and not with the ambit of the expression “let as a separate dwelling”. Indeed, it can fairly be said that the Epsom Grandstand and the Hicks cases only involved the implied necessary conclusion that premises let for mixed business and residential uses were “let as a separate dwelling” under the 1915 and 1920 Acts respectively, as the judgments in both cases concentrated on the meaning of “dwelling house”.

Nonetheless, I accept that some of the reasoning in the Wellcome case (in particular the reinstatement and approval of the reasoning in the Epsom Grandstand case, which impliedly dealt with the question of “let[ting] as a separate dwelling”) is difficult to reconcile wholly satisfactorily with that in the Pulleng, Wagle and Webb cases. However, I think the three decisions can logically stand together, as is demonstrated by (a) the fact that the court in the Wellcome case did not suggest it was overruling the Wagle case and did not criticise the reasoning of Taylor LJ in that case or in the Webb case, and (b) counsel’s concession (impliedly approved by Wilson LJ) to that effect in the Patel case.

In those circumstances, while it would be idle to pretend that most of the reasoning in the Wagle case was enthusiastically viewed by the Court in the Wellcome case, three points can be made. First, the Wellcome case ultimately was concerned with the meaning of the expression “dwelling house” in the Rent Act context, whereas the present case is concerned with the meaning of the words “let as a separate dwelling”. Secondly, criticism in the Wellcome case is concentrated on the judgment of Dillon LJ in the Wagle case (in my view with considerable justification), whereas nothing was said to impugn the fundamental reasoning of Taylor LJ in the Wagle case. Thirdly, in the Patel case, this court, while unhappy about the Wagle case, considered that it represented the law, at any rate at Court of Appeal level.

Quite apart from this, it is appropriate to bear in mind that the decisions in the Pulleng and Wagle cases have stood for well over fifteen years, and have no doubt been acted on many times. Similarly, with the decision in the Webb case. If those decisions were plainly wrong, or could be shown, on analysis, to have no rational basis, as was the position in relation to the cases overruled in the Wellcome case, the fact that they had stood for some time would very probably not be enough. However, where there is both a rational and a practical justification for earlier decisions of this court, it seems to me that, particularly when they have stood for some time, it would verge on judicial hooliganism, and would drive a coach and horses through the already diluted stare decisis rule, if we were to overrule those decisions simply because some of the reasoning was flawed and because we might have reached a different conclusion.

Accordingly, while I am doubtful whether I would have reached this conclusion in the absence of the decisions in the Pulleng, Wagle and Webb cases, I would hold that, subject to the remaining point to which I now turn, the appellant cannot claim protection under the 1977 Act, because, having been let for mixed business and residential use, the premises were not “let as a separate dwelling” within section 1 of that Act.

This issue can be far more easily disposed of. If the tenancy, as originally granted, was not capable of being within the 1977 Act, because it was not a letting of premises “as a separate dwelling”, it could, in principle, nonetheless have come to be within the 1977 Act if “it can properly be inferred, as a matter of fact, that the landlord has affirmatively assented to the change of user until it has become let in whole or in part [as a separate dwelling]” to adapt Denning LJ’s observation in the Wolfe case to the appropriate issue under the 1977 Act. That analysis was approved, expanded and applied in the Russell case by Slade LJ.

In the present case, the business use ceased around 14 years before the notice to quit expired, and the landlord continued accepting rent throughout that period. Without more, it seems to me that, at least in this case, it is hard to see how that could amount to a sort of positive assent which Denning LJ had in mind in the Wolfe case. As he said, it is well established that, simply by accepting rent with knowledge of the breach, a landlord does not waive the right to forfeit the lease for a continuing breach of a user covenant. By the same token, there has to be something more than mere knowledge of the change of use coupled with acceptance of rent to operate as positive consent to the change. In this case, indeed, the respondents contend, and there is no reason to doubt this contention, that, for much of the period from 1990, they did not know of the cesser of business use and had always assumed that the tenancy was within the 1954 Act.

In these circumstances, subject to one possible point, it seems to me that the Judge was right (or, to put it at its very lowest, entitled) to conclude that the nature of the tenancy did not change, so as to bring it within the 1977 Act. Acceptance of rent coupled with knowledge over a very long period indeed might, without more, be enough to entitle a court to infer consent, but, again to put it at its lowest, fourteen years would certainly not compel such a conclusion. Not having been referred to much authority on the point, I leave open the question whether it would, in any circumstances, entitle the court to reach such a conclusion, although I would tend to doubt it.

The one point on this aspect, which was of concern to Chadwick LJ, was that, for the greater part of this period, the respondents had received the rent in the form of Housing Benefit direct from the local housing authority. He thought there might be something in the argument that a landlord could not deny that he had in effect consented to a change of use from mixed business and residential use, to purely residential use once he had accepted rent in this way for a substantial period.

On analysis, I am satisfied that there is nothing in that point. Paragraph 10(4) of the Housing Benefit (General) Regulations 1987 makes it clear that Housing Benefit can be paid in respect of the rent due under premises used for mixed business and residential purposes, and therefore its seems to me that this point take matters no further. It is only fair to mention that, until Chadwick LJ raised the point, there had been no reference to the provisions of those Regulations. It should be added that I am by no means convinced that I would have reached a different conclusion on this aspect, if the Regulations had not so provided.

It is right to mention that the applicant, who is in poor health and somewhat hard of hearing, raised a number of other points in his lengthy written representations and pithy oral submissions. However, those points were either concerned with issues on which he had been refused permission to appeal or were misconceived as a matter of law (although it is quite understandable why he raised them).

In all these circumstances, I would refuse permission to appeal on the point left open by Chadwick LJ, and, while I would grant permission to appeal on what I regard as the main point in this case, namely the question whether the premises had been “let as a separate dwelling” in 1976, I would dismiss the appeal on that point.

I agree.

I also agree.