Table Of Contents
When is a service charge a service charge?
It is easy, in all the hullaballoo about reasonableness and consultation, to overlook the meaning of those two short words in section 18 of the Landlord and Tenant Act 1985. That is all the more the case where the definition is not completely contained in the familiar stomping ground of sections 18 to 30 of the Act.
Why is the definition so important? Because if a charge does not fall within the technical meaning of “service charge” under the 1985 Act, the FTT does not have the power to decide any disputes about it and lessees are not entitled to what Martin Rodger QC described as “the protection afforded by sections 18 to 30 of the 1985 Act”, amongst which he included:
This case examines (exhaustingly) the meaning of “separate” and “dwelling” and their impact on the FTT’s jurisdiction to deal with disputes between landlords and tenants.
The building in question was constructed in the 19th century in Liverpool’s Hatton Garden, as the HQ of the Fire Brigade. It was converted into student accommodation in 2012.
Listed building consent was granted to create 93 units, or “pods”, the majority being a lockable single bedroom and ensuite facilities. Those without ensuite facilities were entitled to use the communal bathrooms. All students had access to communal kitchens and living areas.
Heating was gas-fired by a communal boiler in the basement, but no gas was supplied to any of the pods, nor was electricity or water consumption individually metered.
Access through the building was up the original staircase or by lift.
There was a manager’s office, and there had been a gym and a communal laundry room with coin-operated washing machines, although when the FTT inspected the building, the equipment for those latter two had been removed.
The removal of the equipment hints at the sad pass to which this student idyll was reduced. It was originally occupied by students, but the developer went into administration, and JLK Limited bought it from the administrator in February 2014. By April 2014 it had become the subject of a prohibition order under section 20, Housing Act 2004, because the boiler had stopped working and the building was without water.
A prohibition order prohibits a person from using a building. It is an offence to breach such an order.
Squatters took up residence.
As is increasingly common in the student accommodation sector, each lease – for a 250 year term – demised the pod itself, and rights to use the communal kitchen areas, lounges, showers, toilets and the basement gym. The rights were to be exercised in common with everyone else who had the same rights.
Each lessee covenanted to use their pod only as a “private dwelling in the occupation of one household only”, and was obliged to contribute annually to the proportion of the sums spent by the landlord in running the building.
Section 18(1) of the 1985 Act originally defined a service charge as “an amount payable by a tenant of a flat” for various services. “Flat” was defined as property “occupied wholly or mainly as a private dwelling”.
Section 41 of the Landlord and Tenant Act 1987 removed “flat” from section 18(1) and inserted “dwelling”, thereby bringing tenants of houses and other “not-a-flat” property under the wing of sections 18 to 30 of the 1985 Act.
Section 18(1) currently defines as service charge as: “an amount payable by a tenant of a dwelling”.
By section 38 of the 1985 Act:
“dwelling” means a building or part of a building occupied or intended to be occupied as a separate dwelling, together with any yard, garden, outhouses and appurtenances belonging to it or usually enjoyed with it.
Not quite the model of clarity…:
“As a definition of “dwelling” section 38 perhaps leaves something to be desired”, observed Martin Rodger QC, “as it includes the word it seeks to define. The purpose of the definition is not to explain what a “dwelling” is, but rather to identify those dwellings which are intended to be referred to when the word is used in the 1985 Act, and to bring ancillary premises within the scope of the legislation”.
The parties agreed that dates on which the pods were required to be separate dwellings were the dates in relation to which the section 27A application related, rather than the date of grant of the leases.
Neither landlord nor lessee drew any distinction between the time when the property was subject to the prohibition order and the time when it was not.
A decision on the point was not required, but Martin Rodger QC was of the view that a prohibition/closure order would not, without more, prevent a unit from being a “dwelling” within the meaning of section 38.
He noted that in Morgan v. Kenyon (1913) 110 L.T. 197, where dwelling-houses were the subject of a closing-order, Atkin J held that:
“these premises were undoubtedly dwellinghouses at one time, and I do not think that a building ceases to be a dwellinghouse, provided it has in fact been such, because people cease to dwell in it, or because it gets into so bad a state of repair that people cannot live in it or the law will not allow them to live in it.”
The meaning of “dwelling” has been the subject of judgments:
In Uratemp, Lord Irvine of Lairg LC held that “dwelling” is a word with an ordinary English meaning and wide import, capable of having shades of meaning, but generally connoting “a place where a person lives, regarding and treating it as home”.
It is perhaps unsurprising therefore that the landlord argued that the pods must be the “home” of the lessee in order to be a “dwelling” for the purposes of section 38.
In Uratemp, a Mr Collins had for many years lived in a single room, but no other accommodation, in a hotel. He argued that this room was a “dwelling-house” within the meaning of section 1 of the Housing Act 1988. A tenant is an assured tenant with associated security of tenure:
Neither the room nor the hotel had food preparation facilities – Mr Collins lived off takeaways and “rudimentary meals” that he prepared in his room.
In the Hatton Garden pods, the landlord relied on the following four excerpts from their Lordships’ judgments in Uratemp:
The Supreme Court was split 5-2 (Lord Neuberger and Baroness Hale dissenting) in this case, also relied upon by the landlord.
Here, the court was required to decide whether the Protection from Eviction Act 1977 applied to homeless families to whom a local authority provided temporary accommodation while it investigated the scope of its duty to them.
The Protection from Eviction Act 1977 bites on accommodation “occupied as a dwelling”, and, as Martin Rodger QC noted:
“Lord Hodge acknowledged at  that, although much may depend on the context in which the word is used, as a general rule “dwelling” suggests a greater degree of settled occupation than “residence”; at  he added that “the establishment of a home” was a component of the activity of “dwelling”.”
Here, the Court of Appeal held that the tenancy of two houses, each of which contained a number of units of accommodation, was not “a tenancy under which a dwelling-house (which may be a house or part of a house) is let as a separate dwelling”. The tenant did not therefore benefit from the rent restriction provisions of the Rent Act 1968.
Scarman LJ described the policy of the Rent Acts as being “to protect the tenant in his home”.
The headlessee was the long lessee of a block of flats, and sublet individual flats. The Court of Appeal held that under the 1985 Act, the headlessee was the tenant of “part of a building occupied or intended to be occupied as a separate dwelling”.
Jonathan Parker LJ said:
“The policy underlying the service charge provisions in the 1985 Act and earlier Acts is, however, a different policy in that its emphasis is not so much on protecting the tenant in his home as on providing him with a way of challenging unreasonable charges sought to be levied by his landlord.”
This was a Lands Tribunal decision by the then President, George Bartlett QC as to whether three bungalows on a holiday park, all let on long leases and all capable of residential occupation, were “dwellings” under section 38 of the 1985 Act.
If so, sections 18 to 30 of the 1985 Act applied and the Tribunal was empowered to decide the reasonableness of the service charges. If not, they were not and it did not.
The leases provided that the bungalows must not be used “for any purpose other than as a holiday bungalow”, and planning permission was granted on the basis that the bungalows would not be used for permanent residential accommodation.
The Lands Tribunal noted that some statutes implied that “dwelling” should include a home, whereas some did not. Each statute was to be considered in its context and in the light of the policy behind it.
Despite Jonathan Parker LJ’s observations in Oakfern, George Bartlett QC held that, for the purposes of the 1985 Act, the replacement of “flat” by “dwelling” was of no real consequence, but that:
“It is clear from Uratemp that “dwelling”, where it appears in legislation conferring protection on tenants, will convey its ordinary meaning of the occupier’s home unless there is something that suggests it should not be so limited. I can see nothing that would suggest that in relation to sections 18 to 30 the protection conferred should be extended to premises that are not a person’s home.”
Is there no service charge issue that this unhappy litigation does not cover at some stage?
The early litigation turned on whether the chalets were “dwellings” within the meaning of section 38.
In the county court, HHJ Griggs determined that they were: “dwelling” was a context sensitive word, and, as Martin Rodger QC put it:
“the Judge could see no reason why the protection against abuse associated with the levying of service charges should have been intended to be afforded only to tenants who occupy premises as their home rather than owning them as a holiday home”.
It would appear that the parties in Phillips were content to accept that decision: the later litigation focused on the separate question of section 20 consultation.
Martin Rodger QC confessed that the reasoning in King caused him some trouble, and that he did not agree that sections 18 to 30 of the 1985 Act applied only to “those occupying residential accommodation as their home”.
That was because there was no residence requirement in section 38: the description of a dwelling required only that the building or part of a building be “intended to be occupied as a separate dwelling”. The interpretation of “dwelling” in Uratemp depended on its the statutory context, which intended to give a tenant protection from eviction. Oakfern showed that the context and purpose were different in the 1985 Act.
Further, both the Rent Act 1977 and the Housing Act 1988 expressly referred to occupying property as a residence or home: “dwelling” alone appeared to be insufficient to encompass the point.
Martin Rodger QC was at pains to emphasise the different wording of the statutes.
King v Udlaw was based on dicta in Uratemp, but in the latter case the House of Lords was considering different language and a different statute:
Despite the apparent similarities, he concluded that:
“In [the Rent and Housing Acts] the synonyms “dwelling” and “dwelling-house” are used to describe both the subject matter and the purpose of the tenancy. That element of duplication is missing in section 38, which defines a dwelling as “a building or part of a building occupied or intended to be occupied as a separate dwelling”,” he said.
“The distinct question considered in Uratemp, namely whether the subject of the letting was a dwelling-house, therefore does not arise under section 38. As Lord Bingham pointed out in Uratemp at  the House of Lords was not concerned with the requirement in section 1(1) of the 1988 Act that the premises be “let as a separate dwelling” because Mr Collins occupied a single bedsit and was not required or entitled to share facilities with anyone.
“For the purpose of section 38 the equivalent question regarding the subject of the letting is simply whether the premises are a building or part of a building. To answer that question requires a consideration of the physical characteristics of the premises, not whether they are someone’s home”.
The test as to whether a building or part of a building is “occupied or intended to be occupied as a separate dwelling” requires a consideration of the objective purpose for which it is occupied or intended to be occupied. That may be revealed from the layout of the premises, or the terms of the letting. The use actually made of the premises, is not relevant.
Accordingly, said Martin Rodger QC:
“If part of a building is physically capable of being occupied as a separate dwelling there seems to me to be nothing in the language, context or purpose of sections 18 to 30 of the 1985 Act to require additionally that it must be occupied as someone’s home. I respectfully disagree with the statement to that effect in King v Udlaw.”
It was not correct as a matter of law therefore to say that a unit of accommodation must be someone’s home to be a “dwelling” for the purpose of section 38.
The landlord’s case was not lost however. It argued that the pods were not occupied, nor intended to be occupied, as dwellings. Martin Rodger QC did not agree.
The landlord’s first and second grounds of appeal therefore failed.
It was common ground that the pods at Alexander Terrace were “parts of a building” for the purposes of section 38. The issue between the parties was as to whether they were “occupied or intended to be occupied as a separate dwelling”. (my emphasis)
The same considerations as to whether a dwelling is (intended to be) occupied as a separate dwelling apply in relation to the acquisition of the Right to Manage. In that regard, the landlord pointed to its successful resistance to an application, albeit only at FTT level, to acquire the Right to Manage at St Andrew’s Place, a different, but similarly configured property, also in Liverpool.
The degree of separation required for accommodation to be “occupied as a separate dwelling” was quite energetically litigated in the post-War years. There are three key cases:
The upshot of those cases was that it could not be said that a tenant occupied premises as a “separate” dwelling if he shared any living accommodation with his landlord or another tenant. That meant that the tenant was not protected by the Act.
All three were rather superseded by the Rent Act 1949.
The tenant took a tenancy of two rooms in a house and shared other rooms with the landlord. The Court of Appeal held that where to draw the line was a matter of fact and degree, but in that case, the tenant had shared use of the house: he did not have a tenancy of two rooms let as a separate dwelling.
The Court of Appeal held that a tenancy comprising a bedroom, living room and kitchen, with shared use of a bathroom and toilet on the floor above was of a separate dwelling. Morton LJ said:
“I think that the true test, where the tenant has the exclusive use of some rooms and shares certain accommodation with others, is as follows: there is a letting of part of a house as a separate dwelling, within the meaning of the relevant Acts if, and only if, the accommodation which is shared with others does not comprise any of the rooms which may fairly be described as ‘living rooms’ or ‘dwelling rooms.’ To my mind a kitchen is fairly described as a ‘living room,’ and thus nobody who shares a kitchen can be said to be tenant of a part of a house let as a separate dwelling.”
Lord Porter reviewed the position:
“(1) A portion of a house which is let by a landlord to a tenant, even if in itself separate, ceases to be a separate dwelling or to be protected by the Acts if the terms of the letting contain a provision that the tenant shall have the right of using a living room belonging to the landlord: Neale v Del Soto  KB 144;
“(2) To take away the protection of the Acts, the room over which rights are given must be a living room: a bathroom, lavatory or cupboard will not avail, but for this purpose a kitchen is a living room: see Cole v Harris  KB 474.”
Lord Reid explained why a shared living room makes it is impossible to find anything which is let as a separate dwelling:
“It cannot be the let rooms plus the right to use the other room, because that other room is not let to him at all – he is only a licensee there. And it cannot be the let rooms alone, because his having to share another room shows that the let rooms are only a part of his dwelling place.”
The effect of Baker was that any accommodation shared, whether with a landlord or a fellow tenant, excluded it from the protection of the Rent Acts.
The Landlord and Tenant (Rent Control) Act 1949 changed the Baker rule, so that the Acts’ protection was lost only if some accommodation was shared with the landlord. As Lord Millett noted in Uratemp, there is a difference between sharing accommodation with one’s landlord and sharing with another tenant, and Parliament did not intend to give security of tenure to tenants whose rights included the invasion of their landlord’s privacy.
Parliament’s intention was reiterated in section 22 of the Rent Act 1977 and section 3 of the Housing Act 1988, both of which provide that where the accommodation is shared with another tenant in such a way that it could not, without more, be described as separate, the Act protects the separate part.
By contrast with the Rent and Housing Acts above, there is nothing in the Landlord and Tenant Act 1985 about shared accommodation. There is nothing that deems accommodation to be separate even if some of it is shared.
Reviewing the arrangements at Alexander Terrace, Martin Rodger QC held that a tenant could not be the tenant of a part of the building which is occupied or intended to be occupied as a separate dwelling in circumstances where that tenant also enjoyed the right to share a kitchen, lounge and washing facilities. Applying Lord Reid’s judgment in Baker v Turner, he noted that:
The landlord’s appeal therefore succeeded: the pods were not dwellings because they were not occupied or intended to be occupied as separate dwellings.
Mr Ezekwe’s application to the FTT was therefore dismissed. The charges raised by the landlord were not service charges within the meaning of 1985 Act. The FTT had no power in relation to them.
It must be right that the protection conferred by the 1985 Act should apply to those not occupying their dwellings as their home as much as to those not so occupying. It would create an imbalance of rights in larger blocks where many flats are bought as investments, and where the liability to pay is identical, irrespective of whether the tenant lives in their flat.
One point that is worth expanding, and which has topical ramifications, is the distinction between an assured tenancy and a, well, non-assured tenancy when it comes to ending the tenancy.
I emphasise here that what follows is not intended to be a technical guide to recovering possession of a property. It is a very general overview of the consequences of the trend for rising ground rents.
Here we go.
An assured tenancy has traditionally been a short term tenancy granted for a period such as six months or a year, or indeed, in the case of housing associations, on a weekly basis. There is generally no “purchase price” for the tenancy – the tenant pays a market rent on a weekly, monthly or other basis: that is where the value of the tenancy is to be found. When the tenant moves out, there is no “sale” of the tenancy. Generally the tenant leaves because the tenancy has come to an end through effluxion of time. In other words, to the tenant, the tenancy has no sale value.
By contrast, tenancies – aka leases – for a longer period, such as 99/125/999 years, have a significant purchase and resale value, and that is where their value resides. The rent payable under the lease has traditionally been minimal – for example, a peppercorn, if demanded, or £25 per quarter.
It is easier to work out whether a tenancy is not assured, because Schedule 1 to the Housing Act 1988 lists all the types of tenancies that cannot be assured tenancies.
For long leases, the most relevant paragraphs of Schedule 1 are paragraphs 3A and 3B.
A tenancy cannot be an assured tenancy if it is:
(3A) A tenancy—
(a) which is entered into on or after 1st April 1990 (otherwise than, where the dwelling-house had a rateable value on 31st March 1990, in pursuance of a contract made before 1st April 1990), and
(b) under which the rent payable for the time being is payable at a rate of, if the dwelling-house is in Greater London, £1,000 or less a year and, if it is elsewhere, £250 or less a year.
(3B) A tenancy—
(a) which was entered into before 1st April 1990 or, where the dwelling-house had a rateable value on the 31st March 1990, on or after 1st April 1990 in pursuance of a contract made before that date, and
(b) under which the rent for the time being payable is less than two-thirds of the rateable value of the dwelling-house on 31st March 1990.
In the current climate, where doubling or high ground rent increases are entering stage left and hitting the headlines, it is easy to see how a long lease might suddenly fall into the embrace of the assured tenancy.
Why would that be a problem?
The assured tenancy was a concept that was brought in with the Housing Act 1988. It contains a complete code for recovering possession of the property let.
In very general terms, section 8 of the Housing Act 1988 provides that a landlord may not recover possession of a flat unless s/he has served a notice on the tenant identifying one of the grounds in Schedule 2 to the Act, and the facts that support it.
The grounds in Schedule 2 are divided into two: those grounds where the court must order possession shall given up if the ground and the facts supporting it are made out, and those where the court must decide whether it is reasonable to order possession, and if so, whether to suspend the possession order, even if the ground and the facts supporting it are made out.
One of the grounds on which the court must order possession is ground 8:
Both at the date of the service of the notice under section 8 of this Act relating to the proceedings for possession and at the date of the hearing—
(a) if rent is payable weekly or fortnightly, at least eight weeks’ rent is unpaid;
(b) if rent is payable monthly, at least two months ‘ rent is unpaid;
(c) if rent is payable quarterly, at least one quarter’s rent is more than three months in arrears; and
(d) if rent is payable yearly, at least three months’ rent is more than three months in arrears; and for the purpose of this ground “rent” means rent lawfully due from the tenant.
Consequently, if a tenancy/lease for 999 years is an assured tenancy, and the tenant falls behind with rent payments, the landlord is entitled to take action to recover possession of the flat on ground 8 of Schedule 2 to the Housing Act 1988. If the ground is made out, the court must order possession.
If a lease is not an assured tenancy, the landlord is normally entitled to take action to recover possession by forfeiting the lease.
Forfeiture has long been criticised for the windfall that accrues to a landlord who sees the process through to the end. The Law Commission prepared a report on reform of forfeiture as long ago as 1985.
However – whereas in forfeiture proceedings for non-payment of rent, relief from forfeiture is automatically granted in certain circumstances, and the court retains the power to suspend the execution of the possession order in others under sections 138 and 139 of the County Courts Act 1984, there is simply nothing that the court can do to save a tenancy once it has made a possession order on ground 8 of Schedule 2 to the Housing Act 1988.
Click here for a printer-friendly version of this post.
Click here to read the decision in full on the Upper Tribunal’s website.