Nuffield Health v London Borough of Merton

The London Borough of Merton (Merton) appeals against an order made by Mr Stuart Isaacs QC, sitting as a Deputy Judge of the High Court.

The issue before the judge, and on this appeal, is whether Nuffield Health is entitled to mandatory relief from non-domestic rates in respect of its occupation of premises at Merton Abbey, 29 Chapter Way, London SW19 2RP (the Premises), under section 43(6)(a) of the Local Government Finance Act 1988 (the 1988 Act) which applies where:

“the ratepayer is a charity or trustees of a charity and the hereditament is wholly or mainly used for charitable purposes

(whether of that charity or of that and other charities).”

It is common ground before us, as it was before the judge, that if Nuffield Health satisfies section 43(6)(a) as regards its occupation of the Premises, it was entitled to mandatory relief from the start of its occupation on 1 August 2016 and continues to be entitled to it. The relief represents 80% of the rates otherwise payable in respect of the Premises.

Nuffield Health is a company limited by guarantee and a registered charity, established in 1957. Clause 3.1 of its memorandum of association currently states as its principal object “to advance, promote and maintain health and healthcare of all descriptions and to prevent, relieve and cure sickness and ill health of any kind, all for the public benefit”. Originally, Nuffield Health operated nursing homes and later expanded to include hospitals. It now owns and operates 31 hospitals, 112 fitness and wellbeing centres and 5 medical centres, and it operates over 200 gyms and health assessment facilities in workplaces across the United Kingdom. In evidence filed in these proceedings, Mr Toby Newman, the general counsel and company secretary of Nuffield Health, stated:

“9.

From the mid-2000s, thecharity shifted its focus to prevention of illness as much as cure. It acquired companies that operated under the brand of ‘Proactive Health’ and then, in November 2007, purchased the Cannons Health & Fitness business that operated a network of gyms. In May-July 2008, the integration of the concepts of fitness, prevention and cure was recognised by merging the separate services and facilities offered by Nuffield Hospitals, Proactive Health and Cannons into a single brand, governance and management structure under the new name of “Nuffield Health”.

10.

This was followed in June 2010 by the formal amendment of the charitable objects, so that the charity now exists simply

“to advance, promote and maintain health and healthcare of all

descriptions and to prevent, relieve and cure sickness and ill health of every kind, all for the public benefit”.

11.

It aims to achieve those objects by maintaining a widelyaccessible health system which connects three key elements: first, the promotion of fitness, emotional wellbeing and health education as a means of maintaining good health; secondly, the identification, assessment and containment of health risks; and thirdly, the treatment of diagnosed health problems, including rehabilitation following treatment. To that end, Nuffield Health has now established a network of fitness and wellbeing centres, diagnostic units, hospitals, and medical clinics, complemented by digital health and wellbeing services. There are 31 hospitals, 112 fitness and wellbeing centres, 5 medical centres, and over 200 further gyms and health assessment facilities operated by Nuffield Health in workplaces across the UK.”

Mr Newman explains that Nuffield Health is a trading charity, with a group turnover of £909.1 million in the year ended 31 December 2017. While it used to raise funds by donations, it became clear by the early 1990s that this was inadequate to meet the rising costs of modern medical provision. Donations are no longer sought to fund activities and the primary source of revenue is the fees charged for its services. It is intended that the fees will more than cover the full costs of the services, including all variable costs and fixed and overhead costs, so as to maintain a margin to reinvest. Mr Newman states that this “is normally achieved by setting prices by reference to market rates”.

Mr Newman states that the primary means by which Nuffield Health seeks to deliver public benefit is through the core trading activities undertaken at its hospitals, clinics and fitness and wellbeing centres. As regards the latter, Mr Newman states:

“The fitness and wellbeing centres (such as the one at Merton Abbey) provide conventional gym equipment, personal training and exercise classes, as well as a range of other health-related benefits such as health checks and screening, weight loss programmes, physiotherapy, nutritional therapy, emotional wellbeing advice, mental health treatment, and health education events.Given the known benefits of physical activity in improving physical health and mental wellbeing, the use of the gym and pool facilities by people directly fulfils the charity’s objects.”

Mr Newman explains that, as part of the duty of the trustees (the directors) to oversee the proper delivery of public benefit, they have adopted and keep under review a policy document entitled “Delivering Public Benefit” which sets out benchmarks for the mix and type of activities that Nuffield Health will engage in to ensure its activities remain aligned to its charitable objects. The policy recognises the provision of health products and services in its fitness and wellbeing centres as one of its directly charitable activities.

Mr Newman goes on to explain that:

“The policy also recognises that the poor must not be excluded and that the charity must ensure that it makes provision for the poor to access services (which must be more than minimal or token provision). Many of the charity’s services (especially at its hospitals) are provided at what would be regarded as relatively high fees. The trustees therefore take into account the ability of people who could not otherwise afford such services to have them funded by insurance or the NHS, as well as accessing services that are free of charge or have relatively low fees.”

The Premises are one of 35 gyms acquired on 1 August 2016 when Nuffield Health purchased the business of Virgin Active, to add to its existing network of fitness and wellbeing centres. The total purchase price was £64 million, including an amount of £38.1 million in respect of goodwill. Goodwill is the difference between the total price paid for a business and the market values of its net separable assets, and it is generally taken to be the price paid for the profit-earning capacity of the business.

The general layout and core facilities at the Premises have stayed broadly the same as they were when it was owned by Virgin Active with only minor adjustments. The Premises comprise a basement car park, with spaces for 42 cars for members, a ground floor and a first floor. The pedestrian entrance is on the ground floor leading to a reception area. There is a suite of a treatment room, a waiting room and two offices to one side of a small waiting area. Beyond the reception desk are gates, which can be opened only by using a pass card and which lead to the rest of the Premises. There is a large open area which is principally used as a space for people to rest after exercise, and which until January 2019 included a café. Changing rooms and toilets, and two small rooms where members can meet staff, are off this area. At the rear of the ground floor is a 25-metre swimming pool, a smaller spa pool and a sauna/steam room. Most of the first floor is taken up by the gym, together with a dedicated room for spin cycle classes, two studios where exercise classes are held, and two consultation rooms where health MOTs are carried out. There is also an Ofstedregistered crèche available for members with childcare needs for up to two hours a day. There are also staff offices, a staff room and storerooms.

The facilities at the Premises are available to members on payment of either a single annual charge of £852 or a monthly charge of £80, making a total charge of £960 for a year. Membership gives access to all bar two of Nuffield Health’s 112 fitness and wellbeing centres across the country, a discounted rate on a range of health and wellbeing services and free health MOTs. Use of the crèche costs £10 per month for children under three and £17 per month for other children. I refer later to the services available at the Premises to non-members.

The rateable value of the Premises is £565,000.

In August 2016, Nuffield Health applied to Merton for mandatory and discretionary rate relief. Merton refused discretionary relief but initially applied mandatory relief.

The proceedings were issued as a Part 8 claim in April 2019, supported by witness statements of Mr Newman and Mr Anthony Platt, the Head of Estates at Nuffield Health. Mr David Keppler, the Head of Revenues and Benefits for Merton, made a statement in opposition. A further witness statement was made by Mr Keppler concerning a map and demographic information for the area in which the Premises is situated. No oral evidence was given and there was no challenge to the evidence of fact provided by the parties.

The claim was heard by the judge in January 2020. He made a declaration that Nuffield Health is and has at all times since 1 August 2016 been entitled pursuant to section 43(6)(a) of the 1988 Act to mandatory relief from non-domestic rates in respect of the Premises. He ordered the repayment of rates for which relief should have been given, amounting to £930,823.95, together with interest.

Permission to appeal was given by Lewison LJ on four grounds. Ground 1 is that the judge was wrong to hold that Nuffield Health was not required to show that the

I have earlier set out section 43(6)(a) of the 1988 Act which provides for mandatory rate relief for premises occupied by a charity and used for charitable purposes. The antecedents for this provision were summarised by Lord Cross in Oxfam v Birmingham City Council [1976] AC 126 (Oxfam) at 135. Before 1955, there was no statutory provision for rate relief for charities but in practice many local authorities made, as a matter of grace, “sympathetic” assessments. This benevolent but no doubt inconsistent approach appears to have changed after responsibility for valuations and assessments was transferred to the Inland Revenue by the Local Government Act 1948. Statutory provision for rate relief was first made by the Rating and Valuation

Since the enactment of the 1988 Act, there have been significant statutory developments as regards charity law. In particular, and directly relevant to the present appeal, “charity” and “charitable purposes” were for the first time given statutory meanings by the Charities Act 2006 (the 2006 Act) which was replaced by the Charities Act 2011 (the 2011 Act), a consolidating statute. Prior to the 2006 Act, the meaning of these terms derived from authorities decided over a long period. The effect of the case law was to a substantial extent preserved by the provisions now appearing in the 2011 Act.

Nothing turns on the meaning of “charity” in the present case, because it is common ground that, as a registered charity, Nuffield Health is conclusively presumed to be a charity for all relevant purposes: section 37(1) of the 2011 Act.

As regards the meaning of “charitable purposes”, section 2 so far as relevant provides:

“(1)

For the purposes of the law of England and Wales, a

charitable purpose is a purpose which-

(a)

falls within section 3(1), and

(b)

is for the public benefit (see section 4).

(2)

Any reference in any enactment or document (in whatever terms)-

(a)

to charitable purposes, or

(b)

to institutions having purposes that are charitable under the law relating to charities in England and Wales, is to be read in accordance with subsection (1).

(3)

Subsection (2) does not apply where the context otherwise requires.”

Section 2(1)(a) refers to purposes falling within section 3(1). The purpose relevant to Nuffield Health is 3(1)(d): “the advancement of health or the saving of lives”. Section 3(2)(b) provides that “the advancement of health” includes “the prevention of or relief of sickness, disease or human suffering”.

Section 4 concerns the public benefit and provides:

“(1)

In this Act “the public benefit requirement” means the requirement in section 2(1)(b) that a purpose falling within section 3(1) must be for the public benefit if it is to be a charitable purpose.

(2)

In determining whether the public benefit requirement is satisfied in relation to any purpose falling within section 3(1), it is not to be presumed that a purpose of a particular description is for the public benefit.

(3)

In this Chapter any reference to the public benefit is a

reference to the public benefit as that term is understood for the purposes of the law relating to charities in England and Wales.

(4)

Subsection (3) is subject to subsection (2).”

Section 4 preserves the common law as regards the requirement that a purpose, to be charitable, must be for the public benefit. This in turn has two aspects or senses, which Peter Jackson LJ aptly summarised in argument as the nature and the reach of the purpose. The first sense is that the purpose itself must be such as to constitute a benefit to the public. A trust for the training of spiritualist mediums lacks such public benefit: Re Hummeltenberg [1923] Ch 237. It is not disputed in the present case that the services offered by Nuffield Health are for the public benefit in this sense.

The second sense is summarised in Tudor on Charities (10^th ed. 2015) at para. 1-043 as follows: “those who may benefit from the carrying out of the purpose must be sufficiently numerous, and identified in such manner, as to constitute what is described in the authorities as “the public or a section of the public””. Thus, a trust to make grants towards the cost of education which in fact applied over 75% of its income to pay the school fees of children of employees of a particular company lacked public benefit in this sense: IRC v Educational Grants Association Ltd [1967] 1 Ch 993. This can give rise to difficult questions as to whether the trust is for the benefit of a section of the public, when the beneficiaries are defined, or are in fact chosen, by reference to some particular criterion such as the area where they live or their membership of an identified group. This again is not an issue in the present case, where there is no restriction of that sort as to who can benefit from membership of Nuffield Health’s fitness centres.

The issue if it arises at all – Nuffield Health says that it does not, and the judge agreed – is whether the restriction of membership to those able to afford the membership fees has the effect that Nuffield Health is not using the Premises for charitable purposes, as required by section 43(6)(a) of the 1988 Act, because such purpose lacks the necessary quality of public benefit.

This leads conveniently to consideration of the first ground of appeal.

Merton submitted to the judge that the effect of section 43(6)(a), requiring the premises in question to be used for “charitable purposes”, was that the charity occupying them must establish that the use of those premises was for the public benefit. The focus was on the use of the premises, not on the overall activities and purposes of the charity.

The judge rejected this submission. Having summarised at [32] the submission, he said at [33]:

“In my judgment, the defendant’s submission fails at the outset. Under section 2(1) of the 2011 Act, in order for one of the purposes falling within section 3(1) of the 2011 Act to be a charitable purpose it must be for the public benefit. Under section 2(2)(a) of the 2011 Act, any reference in any enactment to charitable purposes is to be read in accordance with subsection (1). Accordingly, the need for the public benefit requirement to be fulfilled is imported into the expression “charitable purposes” in section 43(6)(a) of the 1988 Act and section 1(1)(a) of the 2011 Act. An institution will only be a “charity” in accordance with section 1(1)(a) of the 2011 Act if it is established for charitable purposes only, which again requires the purposes in question to be for the public benefit. The public benefit requirement is thus to be applied to the purposes of the charity and not to its activities carried on at the individual hereditament. In ISC, the court stated at [195] that the inquiry is whether the activities overall of the charity are for the public benefit. Further, as already stated, under section 37(1) of the 2011 Act, an institution is, for all purposes other than rectification of the register, “conclusively presumed to be or to have been a charity at any time when it is or was on the register”. The conclusive presumption that the claimant is a charity means that the requirements for it to be a charity are conclusively presumed to have been met, namely that it is established for purposes which are within section 3(1) of the 2011 Act and that the purposes in question fulfil the public benefit requirement.”

On this appeal, as before the judge, Merton relies on the express terms of section 43(6)(a) (“the hereditament is wholly or mainly used for charitable purposes (whether of that charity or of that and other charities)”), read with sections 2 and 4 of the 2011 Act which require a charitable purpose both to fall within the list in section 3(1) and to be “for the public benefit” (as explained in section 4). It submitted that the judge was wrong to treat the public benefit as relevant only to the purpose or status of the charity, and not to the use of the hereditament. As it is common ground that Nuffield Health’s overall purpose is the advancement of health and that its overall purpose is for the public benefit, it remained for Nuffield Health to show that its use of thePremises was not only for the advancement of health but was for the public benefit.

Merton submitted that this was consistent with the nature of non-domestic rates as a tax on a specific property, with the focus being on the use and occupation of that property: see Woolway (Valuation Officer) v Mazars LLP [2015] UKSC 53, [2015] AC 1862 at [1] per Lord Sumption. It also gave effect to the policy which, it said, evidently underlay the availability of mandatory relief under section 43(6)(a), that the exemption from 80% of non-domestic rates should be allowed only if the use of the premises was in fact substantially for the public benefit.

The judge’s approach, which looked not at the actual use of the premises in question, but at the overall purposes and activities of the charity, was inconsistent with these basic factors applicable to non-domestic rates. Further, the reference in section 43(6)(a) to “charitable purposes (whether of that charity or of that and other charities)” could, on the judge’s approach, implausibly require a rating authority or the court, in assessing an individual hereditament, to look at the whole of the activities of not just one but a number of different charities. Merton’s approach involved no practical difficulty. It was common for a rating authority, or a court, to make factsensitive assessments when deciding on rating relief or exemption. It would not

Nuffield Health submitted that it was the duty of charity trustees to operate the charity so as to advance the public benefit and that, in assessing whether trustees were performing that duty, the correct approach was to look at the activities of the charity as a whole. The requirement under section 43(6)(a) was that the hereditament be used wholly or mainly in the pursuit of charitable purpose(s), and this was to be judged by reference to the activities of the charity as a whole. It was completely wrong to look only at what was happening at the premises in question. The use of the premises viewed in isolation does not have to be charitable, but it must be of material benefit to the delivery by the charity of its charitable purposes. Relying on the decision of the House of Lords in Oxfam, it was submitted that the question for the rating authority was simply to consider the nature of the user of the premises: are they used for charitable purposes or for investment or fundraising purposes. No further assessment was required.

It supported this submission with further considerations. First, Merton’s test was unworkable. It was for the Charity Commission, with its expertise, to decide whether a charity was pursuing its purposes for the public benefit, not for individual rating authorities, with the potential for inconsistent results across different authorities. It was inconsistent with the recommendation of the Pritchard Committee for a simple scheme of mandatory relief. It could lead to inefficiencies and distortions in the way that charities which operate over more than one site allocate their functions as between their properties, so as to be able to show that each can be shown in isolation to be mainly used for the public benefit.

Mr Kolinsky QC, on behalf of Nuffield Health, submitted that it was entitled to relief under section 43(6)(a) even if the public benefit requirement were only satisfied in relation to its activities as a charity as a whole and even if the public benefit requirement was not satisfied at the premises in question if viewed on their own. This, he submitted, was consistent with the approach of the House of Lords in Oxfam and in the earlier case of Glasgow Corporation v Johnstone [1965] AC 609.

In considering these submissions, it is necessary to start with the obvious point that the issue is one of the proper construction of section 43(6)(a), which is to be determined by a consideration of its express terms in their context.

The first requirement is that the premises be occupied by a charity (or the trustees of a charity). Unless it is acting in breach of its charitable purposes, it can be assumed that the charity is occupying the premises to further those purposes. The second requirement is that the premises are “wholly or mainly used for charitable purposes (whether of that charity or of that and other charities)”. It is common ground that there are some uses of premises which are legitimate for a charity, but which are not uses for charitable purposes. Examples are premises used as shops in order to raise funds (Oxfam) and premises occupied as investments in order to raise funds (Polish Historical Institution Ltd v Hove Corporation (1963) 61 LGR 438, 10 RRC 73).

Section 43(6)(a) does not, however, restrict the second requirement so as to exclude only premises used for fundraising or investment. Instead, it fixes attention more generally on whether the premises in question are used for charitable purposes. The

I am not greatly impressed by Nuffield Health’s submissions made as to the practical difficulties and scope for inconsistency which it suggests would result from Merton’s case. As Mr Ormondroyd, who argued this part of the case for Merton, submitted, rating authorities are accustomed to making fact-sensitive assessments as to the use of individual hereditaments. As the law reports show, difficult questions can arise but, insofar as inconsistency might be a problem, it can be solved by the conventional means of authoritative rulings. These considerations are certainly not, in my judgment, sufficient to justify a construction at odds with the natural reading of the terms of the section.

I find more pertinent the submission made on behalf of Merton that its reading of section 43(6)(a) is in keeping with the basic feature of non-domestic rates as a tax on individual properties, leading to a conclusion that the property itself must be used for the public benefit to justify a mandatory reduction in the tax payable in respect of that property.

Both parties sought support from authorities for their submissions but, while some establish important general principles, none of them provides any substantial assistance on the issue raised by Ground 1.

Glasgow Corporation v Johnstone was a decision under legislation applicable to Scotland for relief for charities from non-domestic rates in substantially the same terms as section 43(6)(a). It concerned a single building comprising a church and a house, the only access to the house being through the church. The house was used as the residence of the church officer, who was a full-time employee of the church (acting through its congregational board) performing a range of organisational and similar duties. He was required by his contract of employment to live in the house. Two issues arose. First, was the occupier of the house the congregational board of the church or the church officer? It was held that the board was the occupier. This led to the proposition, on which Mr Kolinsky relied, that it was the charity’s use of the house that the court must consider. Since there is no issue on the present case that the occupier is Nuffield Health, it is clearly its use of the Premises that must be considered. The second issue was whether the board’s use of the house was “for charitable purposes”. The House of Lords held that it was. Their use of the house was to ensure that the officer was on the spot to assist them in the more efficient performance of their charitable activities. Lord Reid said at p. 622 that it was too narrow a view to see whether any charitable activity was carried on in the house. He said: “If the use which the charity makes of the premises is directly to facilitate the carrying out of its main charitable purposes, that is, in my view, sufficient to satisfy the requirement that the premises are used for charitable purposes”.

It was on this basis that, in Oxfam, Lord Cross expressed the view at p.139 that the head office of Oxfam would qualify for rating relief. The issue in that case was whether a shop, occupied by Oxfam, was used for charitable purposes. It was held that it was not, essentially because it was used as a retail operation with the profits being applied to Oxfam’s charitable purposes. Lord Cross said at p.146 that the requirement that the use should be for charitable purposes excluded “from relief user for the purpose of getting in, raising or earning money for the charity, as opposed to user for purposes directly related to the achievement of the objects of the charity”. Lord Morris of Borth-y-Gest said at pp.148-49 that there was “a distinction between, on the one hand, activities which a charity may undertake, and, on the other hand, activities which consist in the actual carrying out of its charitable purposes” and that “each case must be decided by an application of the relevant statutory words to some particular facts or set of circumstances”. He considered that “user “for charitable purposes” denotes user in actually carrying out the charitable purposes: that may include doing something which is a necessary or essential or incidental part of, or which directly facilitates, or which is ancillary to, what is being done in the actual carrying out of the charitable purpose”.

It is not Nuffield Health’s case, nor could it be, that its use of the Premises is wholly ancillary to, or directly facilitates, the carrying on of its charitable purposes. Its case is that its use of the Premises is a substantive part of “the actual carrying out of the charitable purpose”. In my judgment, the terms of section 43(6)(a), and the approach of the House of Lords in Oxfam, require that in such a case the nature of the activities at the Premises and the basis on which they are made available must satisfy the statutory requirement of use for charitable purposes. I do not accept Mr Kolinsky’s submission that the requirement is satisfied even if, viewed on its own, the use of the Premises would not be for charitable purposes, provided that they form part of the overall charitable activities of Nuffield Health.

Mr Ormondroyd relied on a number of first instance decisions. I will not go through them because they are at best tangential, save to mention Public Safety Charitable Trust v Milton Keynes Council [2013] EWHC 1237 (Admin), [2013] EGLR 133 in which at [34], Sales J said:

“In the contextof this legislation and having regard to the language used, it is reasonable to infer that Parliament intended that the substantial mandatory exemption from rates for a charity in occupation of a building should depend upon the charity actually making extensive use of the premises for charitable purposes (i.e use of the building which is substantially and in real terms for the public benefit, so as to justify exemption from ordinary tax in the form of nondomestic rates), rather than leaving them mainly unused.”

While that statement supports Merton’s case, it is fair to note that the decision concerned an entirely different issue and there is nothing to suggest that the issue in this case was the subject of any submissions, nor was there any reason why it should be.

For the reasons given above, I accept Merton’s case on Ground 1 and consider that the judge was wrong in his interpretation of the meaning and effect of the requirement in section 43(6)(a) that a hereditament be used for charitable purposes.

This ground of appeal assumes that Merton has succeeded on its first ground. It seeks to draw a distinction, in terms of what the public benefit requires, between a “core” charity such as one that advances health through relief of sickness or injury, for example by providing hospital services, or that provides education, and a charity that advances health through the provision of facilities to promote fitness and wellbeing. It submits that the latter is essentially a recreational charity, and the authorities show that, to satisfy the public benefit test, such a charity must normally be for the public or for all members of the public who need the facilities which are provided.

Mr Fowles, who argued this part of the appeal for Merton, relied in particular on IRC v Baddeley [1955] AC 572. The issue in that case was whether the trusts declared by two conveyances of land were charitable. One conveyance was of land largely comprising a church, lecture hall and store, while the other was of a playing field, with a pavilion and a groundsman’s bungalow. By a majority (Lord Reid dissenting), the House of Lords held that neither conveyance created a charitable trust, on the grounds that the purposes were too broadly expressed, in particular by including “social well-being”.

The ratio of the decision has no bearing on the present appeal, but reliance was placed on the views expressed by two of the majority that at least one of the trusts was not charitable for the additional reason that the class of beneficiaries was by the terms of the conveyance restricted to Methodists, and prospective Methodists, of limited means resident in a particular area.

Viscount Simonds accepted that a gift of land for use as a recreation ground by the community at large or by the inhabitants of a particular area may well be charitable. He reserved his view as to whether such a trust confined to adherents to a particular religion would be charitable, but he was of the view that a trust confined to members or potential members of a particular church within a limited geographical area would not be charitable: see his speech at pp.589-93. His starting point was that such a trust would fall, if at all, within the fourth of Lord Macnaghten’s categories in Income Tax Special Purposes Commissioners v Pemsel [1891] AC 531 at 583, comprising trusts of “general public utility”. While in such cases, it was often difficult to draw the line between public and private purposes, he would conclude from the need to demonstrate general public utility that a trust for the benefit of a class of persons confined not only to a particular area but selected from a particular creed would not be charitable. Considerations of this kind were important in the case of all charities, including education and the other objects in Lord Macnaghten’s first three categories, but “they have even greater weight in the case of trusts which by their nominal classification depend for their validity upon general public utility”.

Lord Reid, dissenting, disagreed with Lord Simonds’ view, and Lords Porter and Hodson expressly reserved their position. Lord Somervell, expressing himself more widely than Lord Simonds, took the view at p.615 that the nature of a particular object

Mr Fowles submitted that it followed, from authority and from section 5 of the 2011 Act (and its predecessor, section 1 of the Recreational Charities Act 1958) dealing with the charitable provision of facilities for recreation, that the court should be slower in a recreational case of the present type to find that the public benefit test was satisfied than in the case of the educational charities considered in R (Independent Schools Council) v Charity Commission [2011] UKUT 421 (TCC), [2012] Ch 214 (ISC).

In part, Mr Fowles was proposing a hard-edged test. Certain charities – either recreational charities or, more broadly, charities of general public utility – are normally required to be for the benefit of the public at large or for those members of the public in need of the charity’s provision. Support for such a requirement might be found in the speech of Lord Somervell in IRC v Baddeley, but not in any of the other speeches nor, so far as I can detect, in any other authority.

It is not in dispute that the nature of the charity may affect the breadth of the class for whom benefits may be provided. The judge in the present case agreed at [37] with Merton that, when considering the public benefit requirement, “a central aspect is that those who may benefit from the carrying out of the purpose must be sufficiently numerous, and identified in such manner as, to constitute a section of the public” and that “[w]hat satisfies the public benefit requirement may differ markedly between different types of allegedly charitable purposes and so caution must be exercised in applying authorities decided in one area of charity law to another area”. I consider the judge was right in these respects but, like the judge, I do not think that a firm proposition such as that advanced on behalf of Merton can be established.

Nor do I think that such a test can be applied to recreational charities. They are governed by section 5 of the 2011 Act, which contains some specific requirements close to those advanced by Mr Fowles in section 5(3)(b). The public benefit requirement is expressly preserved by section 5(5) and is not to be defined, in effect, by the other provisions of section 5. More important still for the present case, Nuffield Health is not a recreational charity to which section 5 applies, nor is it analogous to the charities providing recreation grounds with which the authorities before the Recreational Charities Act 1958 were concerned.

The real issue in the present case is different. It is whether charging membership fees at market rates is compatible with the public benefit requirement, either on its own or when taken with other benefits provided by Nuffield Health at the Premises. This is the subject of Ground 3.

Merton submits, on the assumption that it is correct on its first ground of appeal, that it was incumbent on Nuffield Health to show that it was using the Premises for the public benefit and that it failed to do so, by reason of the level of membership fees and the absence of any sufficient public benefit in the provision of services to nonmembers.

Before looking at the way that the judge dealt with this submission, it is convenient to refer to relevant principles established by the authorities.

First, it is not an objection per se that a charity makes charges for its services. This long-established principle was reiterated by the House of Lords in Scottish Burial Reform and Cremation Society Ltd v Glasgow Corporation [1968] AC 138, a case which, like the present, concerned liability for rates. The issue was whether a crematorium, which the appellant had built and operated in Glasgow, was “wholly used for charitable purposes”. This turned on whether the promotion of cremation was capable, as a matter of law, of being charitable. The case was not connected with the fact that the appellant charged for cremations. It was expressly accepted by the respondent local authority that the fact that economic charges were made for services did not disqualify the appellant from being a charity: see p.143B. Although the matter was not in dispute, the position was restated in some of the speeches: see Lord Reid at pp.147-48, Lord Upjohn at p.149 and Lord Wilberforce at p. 156. Lord Wilberforce noted that the fees were “apparently modest”.

Second, in order to be for the public benefit, the relevant services must be provided to the public or a sufficiently large class of the public and the “poor”, the meaning of which I refer to below, must not be excluded. The leading case is the decision of the Privy Council, on appeal from New South Wales, in Re Resch’s Will Trusts [1969] 1 AC 514. The relevant issue was whether a bequest to the Sisters of Mercy, a religious order, for the general purposes of a private hospital run by them was charitable. Alongside the private hospital, the Sisters ran a larger, adjacent public hospital. The evidence showed that the reason for establishing the private hospital was to relieve the pressing demands of the public for admission to the general hospital which was quite inadequate to the demand on it (p.539A). There was also evidence that the close proximity and association between the two hospitals had advantages from a medical aspect (p.540A).

In giving the judgment of the Board, Lord Wilberforce said at p.540C-D that the purposes of the hospital “were essentially the provision of a certain type of medical and nursing care and treatment for which there is a need and which the general hospital does not give”. As for the charges made by the hospital, “[i]t has been part and parcel of these purposes to provide such care and treatment at the lowest cost at which this can practically be done”. The charges were comparable to those made by similar, charitable institutions, and they would be paid by patients either out of their own resources or under hospital benefit schemes to which they had contributed.

Lord Wilberforce noted at pp.540G-541A that, while a gift for a hospital was prima facie charitable, a hospital would not be a charitable institution if “the benefits it provides are not for the public, or a sufficiently large class of the public, to satisfy the necessary tests of public character”. It was objected that the private hospital in question did not satisfy this criterion because “it provides only for persons of means who are capable of paying the substantial fees required as a condition of admission” (p.542D). The objection was not that benefits were provided to people who were not poor, but that the poor were excluded from participation in its benefits. This objection was rejected. The general benefit to the community of the facilities provided by the hospital “results from the relief to the beds and medical staff of the general hospital, the availability of a particular type of nursing and treatment which supplements that provided by the general hospital and the benefit to the standard of medical care in the general hospital which arises from the juxtaposition of the two institutions”. The service provided by the hospital was “needed by all, not only by the well-to-do” (p.544).

As for the charges, which were at approximately cost price, Lord Wilberforce said at p.544:

“So far as its nature permits it is open to all: the charges are not low, but the evidence shows that it cannot be said that the poor are excluded: such exclusion as there is, is of some of the poor – namely, those who have (a) not contributed sufficiently to a medical benefit scheme or (b) need to stay longer than their benefit will cover or (c) cannot get a reduction of or exemption from the charges.”

The effect of charging for services on charitable status was discussed by the Upper Tribunal (Warren J and UT Judges McKenna and Ovey) in ISC, in the context of feepaying schools. As regards Re Resch, they said, correctly in my view at [164]:

“In relation to the observation that the poor were not excluded, Mr Giffin remarks that Lord Wilberforce does not say that the poor were able to benefit to any specified extent or even to a reasonable extent; and that there is no factual finding or discussion of what is meant by a reasonable extent and how the hospital meets that requirement. It seems to us clear, however, that in Lord Wilberforce’s view the benefits of the services provided by the hospital were to some degree directly available to the poor, in which class he clearly included persons who were nonetheless able to pay because they held a medical benefits insurance contract, and we do not think he can fairly be supposed to have thought that the degree of availability to the poor was negligible.”

Having considered a number of authorities, the Upper Tribunal concluded at [173] that, in applying the principle that a trust which excludes the poor is not charitable, “the poor” equates not to the destitute but to those of modest means. As they observed, even in the context of trusts for the relief of poverty, “poor” does not mean “destitute”. In the case of the schemes held to be charitable in Joseph Rowntree Memorial Trust Housing Association Ltd v Attorney-General [1983] 1 Ch 159, where the beneficiaries necessarily had some means, the “poor” were not thereby excluded and the element of public benefit was strongly present, in circumstances where the schemes “provided for a real need which was not adequately met by either local authorities or the private sector”. But, as the Upper Tribunal warned at [174], “matters must not be pushed too far”, and the Joseph Rowntree case “is not authority for the proposition that an institution is a charity if it offers at full cost a service which is

In ISC, the Upper Tribunal was not concerned with whether a particular school was a charity, still less with whether particular rateable premises were being used for charitable purposes. The proceedings comprised a claim for judicial review of the general guidance on fee-paying schools issued by the Charity Commission and a reference by the Attorney General under the Charities Act 1993 of specific and elaborate questions about the application of charity law to a hypothetical independent school.

Question A1 asked whether charity law operates so as to cause a school which “performs its objects solely by providing certain services for which it charges fees which cannot be afforded by a significant proportion of the population of England and Wales necessarily to be operating otherwise than for the public benefit”. The answer given at [237] was:

“The answer to this question is Not necessarily, provided that it does not exclude the “poor” altogether. If the basis of the question is that the institution charges full fees for all of its students without providing any identifiable public benefit other than the general benefit to the public of an educated population, the answer is No. But if the institution, although charging fees, has in its student body a not insignificant number of persons whose fees are funded from other charitable sources, the answer is Not necessarily: it will depend on whether there is a sufficient degree of public benefit.”

The very detailed questions B1-10 were designed to elucidate the extent of “public” benefits required to make the operation of a fee-paying school charitable. The Upper Tribunal declined to give any definitive rulings, correctly observing at [242] that each real case will depend on its own factual circumstances, requiring the tribunal in such a case to have a wide range of detailed information. However, it did state this general conclusion at [244]:

“It must be borne in mind that, according to our analysis earlier in the Decision, the “poor” cannot be excluded from benefit either as a matter of the school’s constitution or, other than purely temporarily, in practice. But that is not all. Provision for the “poor” going beyond a de minimis or token benefit may be present, but it is not necessarily enough; the level of provision for them (taken with benefits to the not-so-poor who would otherwise be unable to afford the fees) must be at a level which equals or exceeds the minimum which any reasonable trustee could be expected to provide. The question therefore in all cases is whether the trustees are acting consistently with their obligations, not whether they have provided a particular level of benefit for the “poor”, although some provision must, on any footing, be made.”

Although the judge in the present case had decided that the relevant question was not whether Nuffield Health were using the Premises for the public benefit, he sensibly went on to consider it, albeit as he said “briefly”. His conclusion was that the use of the Premises satisfied the public benefit requirement for the reasons given at [41][42]:

“41.

The defendant submitted that the user of the Premises could not be regarded as being wholly or mainly for charitable purposes because they are used to provide high-end services already available on the open market at full price, at a level which excludes the poor. It referred to the following matters, which to some extent overlap with the matters it relied on in the context of its argument that the main purpose of the Premises’ use is fundraising: (i) the fees charged being set at market rates so as to maintain a margin to reinvest; (ii) the level of membership fees, which would deter those of modest means from membership, including in the local area and which are higher than the cost of membership of other local facilities; (iii) the crèche fees and charges for other services; (iv) the unavailability at the Premises of some free and reduced-fee services offered by the claimant elsewhere; (v) the limited and token additional services offered free or at a reduced cost to those who otherwise could not afford to use the Premises, which are no more than would be expected of any business seeking to advertise and promote its sales on a commercial basis.

42.

In my judgment, none of those matters, either individually or collectively, leads to the conclusion that those of modest or some means are excluded from benefitting from the use of the Premises. I consider that the comparisons made by Mr Keppler in his first statement between the Premises and other facilities in the area are not of assistance in determining whether the Premises themselves exclude the poor. To the extent that the demographics data and map evidence are relied on in the context of the public benefit requirement, as the claimant pointed out, without further elucidation they provide insufficient detail so as to be relied on and their relevance is unclear. In particular, it is unclear why the existence of other facilities in the area and the absence of any specific need for the Premises (cf Joseph Rowntree Memorial Trust Housing Association Ltd v Attorney General [1983] Ch 159) should point to the public benefit requirement not being satisfied in the instant case. In the light of Mr Platt’s evidence, I do not regard the Premises as providing only token facilities for the poor.”

As the great majority of benefits provided at the Premises are available only to members, and as there is no scheme for reduced membership fees, the first question is whether the membership fees are set at a level that excludes those of modest means. The judge said at [42] that none of the matters relied on by Merton and listed in [41]

“either individually or collectively, leads to the conclusion that those of modest or some means are excluded from benefitting from the use of the Premises”. I will take this to mean that he found that the membership fees were not set at a level which would exclude those of modest means.

This finding as to the level of fees is, regrettably, unreasoned. The judge gives no basis and refers to no evidence for his assessment that monthly fees of £80, making £960 pa, are affordable to those of modest means. (I proceed on the assumption that it is unlikely that those of modest means are able to afford a single annual fee of £852.) I would have expected some evidence as to income levels in the local area, from which the great bulk of the members using the Premises are drawn, so as to arrive at some assessment of modest means in this context and evidence as to levels of necessary expenditure, so as to arrive at some assessment of the surplus income from which such monthly fees could be paid by those of modest means. There was no evidence of the sort that was before the Supreme Court, in the different context of whether fees for claims in the employment tribunals were reasonably affordable, in R(Unison) v Lord Chancellor [2017] UKSC 51, [2020] AC 869. Merton adduced some evidence as to local demographics and ranges of income levels, but the judge found, reasonably in my view, that without further elucidation the evidence did not assist. It was, however, for Nuffield Health to lead the evidence, if this was their case, that the membership fees were set at a level that did not exclude those of modest means. It did not do so, and in my judgment there was no basis before the judge on which he could reach a reliable finding on that issue.

Contrary to the judge’s view, I consider that it is relevant that other gyms in the same area were providing similar services and facilities for the promotion of health at materially lower membership charges, suggesting demand at those more affordable levels. As against a monthly fee of £80 charged by Nuffield Health, some commercial gyms charged under £40 per month and one of Merton’s own gyms charged £21.95 per month for a basic package and between £46 and £54 for other packages. It does not of course determine whether either the fees at those other gyms or the fees charged by Nuffield Health were set at a level which excluded those of modest means. It does mean that Nuffield Health cannot say that it is satisfying a public need which is not otherwise available from market providers.

In the absence of a sustainable finding that the membership fees were at a level that did not exclude those of modest means, Nuffield Health must rely on services provided to non-members at the Premises to demonstrate that they were being used for charitable purposes. These were described by Merton as limited and token, and no more than would be expected of any business seeking to advertise and promote its sales on a commercial basis: see the judgment at [41(v)]. The judge concluded at the end of [42] that the Premises were not providing “only token facilities for the poor” but it is not clear whether he was saying that the facilities provided to non-members were sufficient by themselves to amount to the use of the Premises for charitable purposes.

The evidence discloses that the following services are provided at or from the Premises. “Health MOTs” are provided to members of the public on roughly a monthly basis. “At least four times a year” there is a “Meet our Experts” event at the Premises, at which staff provide “expert advice and guidance on a range of health topics including, for example, back complaints, cancer awareness, and joint pain

A non-member can access a free 15-minute initial consultation with a physiotherapist. Further consultations are charged at £36 per half hour for members and £45 for nonmembers, and non-members are given a free one-month gym membership when they start treatment.

Mr Platt described the following additional services in his first witness statement:

“12.

Health MOTs are also offered to targeted non-member groups, namely those who are identified as being at particular risk of ill-health due to their demographic and socio-economic status. For example, Merton Abbey staff recently visited several local schools to carry out “mini health MOT” testing (blood pressure tests etc) for the parents; the parents at the school which had the highest level of problematic results, Haslemere (a local primary school), were offered free 7-day passes to Merton Abbey (with the option of having full health MOTs). Merton Abbey staff also went into the school to run classes on health and wellbeing for the pupils.

13.

Merton Abbey staff also hold “mini health MOT” days in partnership with local businesses. Recent events have been held at M&S, Sainsbury’s and Holland & Barrett, with both staff and customers offered free basic health tests (blood pressure etc) free of charge; free 7-day passes to Merton Abbey are also handed out at these events. Merton Abbey staff also run a mini Health MOT stand at the nearby Boots store on events such as

National No Smoking Day – again free of charge.”

Two local schools use the swimming pool, with the pool closed for their sole use twice a week, paying £67.50 per 1½ hour class.

Compared with the normal operations at the Premises, these services available to nonmembers are very limited and, in many cases, are typical of services provided by commercial operators which, it can be inferred, are essentially promotional. As earlier noted, the judge did not analyse the basis on which it could be said that the provision of these services constituted Nuffield Health’s use of the Premises as being for charitable purposes. In my judgment, they are so limited as to merit the description of token, and they are insufficient to satisfy the requirement for charitable use.

In supporting the admittedly brief analysis of the judge, Mr Kolinsky submitted that the judge was right to defer to the judgment of the trustees that the provision of services by Nuffield Health was for the public benefit. This was based on statements in the Upper Tribunal’s decision in ISC. At [220], they said:

“This is all a matter of judgment for the trustees. There will be no one right answer. There will be one or more minimum benefits below which no reasonable trustee would go but subject to that, the level of provision and the method of its provision is properly a matter for them and not for the Charity Commission or the court.”

The Upper Tribunal at [221]-[223] accepted that, beyond satisfying the requirement for more than token benefit for those of modest means, it is for the trustees to act in accordance with their own considered assessment in the circumstances pertaining to the charity.

Reliance on ISC does not assist Nuffield Health. First, the services do not, for the reasons given above, meet the required minimum threshold. Second, in the context of the hypothetical questions posed in ISC, the Upper Tribunal was right to say that it is a matter of judgment for the trustees “acting in the interests of the community as a whole” (see [215]). However, in a specific case under section 43(6), the judgment as to whether the premises in question are being used for charitable purposes is, in my judgment, ultimately one for the court.

It was straightforward in a case such as Oxfam, where the premises in question were being used solely as a retail shop in a way which could not be said directly to promote the organisation’s charitable purposes, to say that the premises were being used to raise funds. In the present case, by contrast, the Premises are being used to deliver services which Nuffield Health says are part of its charitable activities. Ground 4 presupposes that to be the case, but Merton submits that the Premises are nonetheless being mainly used for fundraising.

Merton reasonably pointed to the very substantial goodwill element in the price paid by Nuffield Health for the chainof gyms previously operated by Virgin Active. This might raise an issue as to whether the purchase was made primarily as an investment, with a view of generating profits from the gyms in order to subsidise other parts of Nuffield Health’s (charitable) activities.

The issue under section 43(6) is not, however, concerned with either the chain of gyms or with all the gyms operated by Nuffield Health. It is concerned specifically with the Premises. Ground 4 would succeed only if Nuffield Health’s main use of the Premises was fundraising. An answer to that question could be provided only with evidence of the budgeted and actual income and surplus generated by the operations at the Premises, and the uses to which Nuffield Health put any surplus. Even with that evidence, it does not follow from the generation of a surplus by a charity at a particular location that the premises in question are being used mainly for fundraising. That judgment would significantly depend on the scale of the surplus when set against the extent of the use of the premises for the public benefit. None of that evidence was before the judge, nor was it sought by Merton from Nuffield Health. Merton was unable to support its submission that “the Premises appear to subsidise other parts of the claimant’s operations” (judgment at [27](v)]. The judge was right, in my view, to say at [28] that it was not made out on the evidence. I do not consider that, in the circumstances of this case as opposed to a case such as Oxfam, the burden was on Nuffield Health to establish the negative proposition, that the Premises were not being used mainly for fundraising.

For the reasons given above, I would allow the appeal, on Grounds 1 and 3.

I have had the very great advantage of reading the judgments of the other members of the court in draft. Like Nugee LJ, I am grateful to David Richards LJ for his explanation of the facts and the issues.

I agree that Ground 2 does not assist Merton. However, I would uphold Ground 3, for the reasons given by David Richards LJ and summarised by him at paragraph 78. In my view, the character of Nuffield Health’s activity at the Premises was indistinguishable from those of any commercial operator in the sector and contained a minimal offering to the poor. The element of public benefit (in the sense required to show that it was acting for charitable purposes) was in my view absent. The judge’s limited reasoning at paragraph 42 does not justify his contrary conclusion. I shall say something more about the significance of this outcome at the end of this judgment.

As to Ground 4, I agree that Merton did not succeed in showing that the premises were mainly being used for fundraising, as opposed to being an expression of Nuffield Health’s overall purpose.

Turning then to the central issue of principle, Merton’s argument under Ground 1 is that in order to qualify for rate relief Nuffield Health has to show that its use of the Premises is for the public benefit, and that this question is not answered in its favour by the fact that the organisation as a whole enjoys charitable status. In the light of our conclusion on Ground 3, I have some sympathy with this argument, but I have nonetheless concluded that this is not sound and that the judge’s reasoning at paragraph 33, quoted at paragraph 28 above, is correct. I agree with the analysis of Nugee LJ on this issue, but as the point is a novel and important one, I shall explain why that is so.

It is, at least to me, surprising that such a fundamental issue of principle has not arisen before. In earlier times, when the contours of charity law were being formed, charitable foundations were no doubt predominantly unitary and local in nature, but for a considerable period, and certainly for the past century, national charities with local branches have been common. That does not mean that Merton’s argument is wrong, only that it is curiously unprecedented. The corollary of this is that the authorities to which we were referred are of limited assistance on this issue, because they were not addressing it. I have, however, found that the historical development of the statutory provisions granting rating relief to charities sheds some light on the question of statutory construction, and I have gained some assistance from the Pritchard Report in understanding the problem that the modern rating legislation set out to address.

I will consider in turn the statutory language, the legislative history, the case law, and matters of policy and practicality.

The first task is to read the provision in section 43(6)(a) as a whole in the context of the statute. If its effect is clear on its face, there is no need to deconstruct it. However, to my mind the effect of this provision is not clear on its face. The judge construed the public benefit requirement as being “applied to the purposes of the charity and not to its activities carried on at the individual hereditament”. I consider that this is a tenable, though not inevitable reading of the words, and in this respect I most respectfully disagree with David Richards LJ.

The provision forms part of a rating statute, and in consequence the unit of rating is the individual hereditament. Merton draws from this that all matters are to be assessed at the level of the hereditament itself. I do not think that conclusion follows, or that the wording of the subsection demands it. The wording tells us what the relief attaches to (the hereditament), but it does not unambiguously dictate the basis upon which it is to be granted.

The search for meaning next calls for us to look more closely at the individual words and phrases. The first thing to note is that the wording of subsection (6) reflects subsection (1), whereby non-domestic rates are levied on a daily basis. This may not be without significance. Let us suppose that every Wednesday the Premises were open to the public at a cost that was affordable to all but the destitute, or that local schools were allowed free swimming all day on Thursdays in term time. It might be said that the public benefit requirement was met on those days, but not on other days of the week. But that is not how rating works and it is unreal to suppose that the annual charge would be fragmented in this way. In reality the ratepayer will be charged for days in occupation of the hereditament, and not on the nature of the occupation on any particular day. It was not suggested otherwise by either party, but the example shows that the provision must be read for its overall effect without placing undue emphasis on particular words (such as charity or hereditament).

Then there is the requirement under this provision that the ratepayer is a charity. Another provision (section 47) allows for discretionary relief for other deserving nonprofit organisations. We will see how this distinction arose when we look at the legislative history. For present purposes, it is not sufficient under section 43 that the hereditament is used for charitable purposes by any other type of ratepayer. It cannot be construed as if it read:

“(6)

This subsection applies where on the day concerned

(a)

the hereditament is wholly or mainly used for charitable purposes…”

The next part of the subsection that may be relevant is that the charitable purposes must be “of that charity or of that and other charities”. The words “charitable purposes… of that charity” may be thought to support a reading that the hereditament must be used by the charity for its charitable purposes. This focuses attention on the

Then there are the words “wholly or mainly”, which are designed to ensure that the charitable activity does not drop below a certain level, as it did in Oxfam. At the same time, the fact that the level of charitable use is to be judged hereditament by hereditament does not, as I have said above, dictate that the same applies to the question of whether the nature of the use is charitable.

Coming then to the core words “the hereditament is… used for charitable purposes”, I consider them, in the setting of the subsection as a whole, to be neutral as between the competing outcomes. The subsection is clearly of universal application. It must apply to single-hereditament charities and to multi-hereditament charities, but there is nothing to indicate that each hereditament of a multi-hereditament charity is to be treated as if it was a single-hereditament charity.

Putting the words of the subsection back together, I prefer the broader construction that treats the charitable purpose, and hence the public benefit requirement, as being subsumed in the status of the charity ratepayer. I now test that construction by examining the legislative history.

At paragraph 17 above, David Richards LJ has summarised the development of mandatory rate relief for charities. I will describe this in somewhat more detail, assisted by the agreed note provided by the parties.

The first statutory provision for rate relief appears in the Rating and Valuation (Miscellaneous Provisions) Act 1955. This was a holding provision pending the further legislation that followed, and it preserved the pre-existing levels of discretionary rate relief for three years. It provided:

“8 Provisions as to rates payable by charitable and other organisations

(1)

This section applies to the following hereditaments, that is to say—

(a)

any hereditament occupied for the purposes of an organisation (whether corporate or unincorporate) which is not established or conducted for profit and whose main objects are charitable or are otherwise concerned with the advancement of religion, education or social welfare; …”

It can be seen that the ratepayer was not specified to be a charity, but a non-profit organisation of a generally benevolent character. Further, in the case of any hereditament, relief followed from occupation being “for the purposes of [the] organisation”.

I think it is clear that Nuffield Health’s argument would have succeeded under the 1955 Act. However, the wording changed in later legislation. The Rating and Valuation Act 1961 provided:

“11 Reduction and remission of rates payable by charitable and other organisations

(1)

If notice in writing is given to the rating authority that—

(a)

any hereditament occupied by, or by trustees for, a charity and wholly or mainly used for charitable purposes

(whether of that charity or of that and other charities); …”

As David Richards LJ says, this is in materially the same terms as section 40 (1) (a) of the General Rate Act 1967, which was in turn replacedby section 43 (6) (a) of the 1988 Act.

The question therefore arises as to whether the difference in wording as between the 1955 Act and later legislation was intended to bring about a change of approach to the question with which we are concerned. The main thing to note in this regard is that the later legislation draws a firm distinction between charities and other benevolent organisations. One qualifies for mandatory relief, the other for discretionary relief.

As to how the modern legislation on relief for charities came into being, it is instructive to consider the Pritchard Report, which led to the 1961 Act and its successors. As Nugee LJ says, the report is not to be treated as a guide to construction, but in my view it helps us to understand the parliamentary intention lying behind the legislation.

The Pritchard Report, running to just 46 pages plus a few appendices after an 18month inquiry, represents the lost art of brevity. Part II concerns section 8 of the 1955 Act, cited above. The conclusion was that the Act had worked reasonably well as a holding provision but that by freezing existing reliefs it had perpetuated anomalies under the previous discretionary system.

In Part III the Committee made recommendations for the future, beginning with these general observations:

“65.

… [The rating system] has become a patchwork system which it is easy to denigrate but for which no acceptable replacement has so far been found. There is much in it that is arbitrary, and perhaps not the least arbitrary feature has been the granting of reliefs. We think that it is impracticable to eliminate all elements of arbitrariness; in considering charities and kindred bodies our object has been to find a reasonable balance of conflicting arguments and interests, consistent with simplicity, certainty and economy in administration.”

The Committee then considered the arguments from charities and from local authorities, and questions about the form of relief and whether it should be mandatory or discretionary. It concluded that the relief should be by rate reduction, not derating, and that in the case of a charity it should be mandatory.

Coming to the issue that is most relevant for our purposes, the Committee considered what it described as the field for mandatory rate relief. It set out a number of proposals that for one reason or another it rejected, and continued:

“91.

Indeed, we think it is plain that, however the field is defined, the results will be to some extent arbitrary, and the solution we propose is the arbitrary one of giving mandatory relief to charities. These bodies have for long enjoyed special privileges under the general law and appear in practice to have been the principal beneficiaries in the past from sympathetic under-valuation by local authorities.

92.

This solution may admit to relief a number of bodies which are not particularly in need or deserving of relief—the sort of body which the man in the street would not ordinarily, in the view of some of our local government witnesses, regard as charitable. We do not think that this result can be avoided. It is doubtful whether there would be even near-unanimity among the general public as to the charities to which the privileges of charitable status should be denied but, even if there were, we can see no justification in principle for redefining the term “charity” for rating purposes only. If charity is to be given some new statutory meaning, the new definition should apply equally for all purposes. …

93.

A further reason for leaving “charity” with the same content for rating purposes as for the purposes of the general law is the proposal later in this Report to make registration for the latter purposes conclusive as to entitlement to mandatory rating relief (see paragraph 119). This proposal makes an important practical contribution to ease and simplicity of administration. Finally, as we note later in the Report. the ultimate financial effect on other ratepayers will generally be small: the effect of any change in the definition of “charity” would necessarily be even smaller.

…

95.

Relief should be given only in respect of those hereditaments which are occupied for the purposes of the charity and not, for example, in respect of hereditaments held as an investment.”

It can be seen that in this last paragraph the report uses the language of occupation

“for the purposes of the charity”, which is the formulation used in the 1955 Act. It can also be seen that it recommended giving mandatory relief to charities and making registration conclusive as to entitlement, while acknowledging at paragraph 91 the arbitrary nature of the choice. Further, at paragraph 92 it recognised that this solution might privilege “a number of bodies which are not particularly in need or deserving of relief”.

The Committee went on to consider possible exceptions to this general rule. Some local authority witnesses had argued that national charities which serve the needs of an area wider than the rating area should not benefit. It was also said that bodies that were not dependent for a substantial part of their income upon voluntary contributions should be excluded. These arguments were rejected, with the conclusion summarised in this way:

“99.

We have therefore had little difficulty in agreeing that, with regard to charities in general, there should be no exclusion from relief on the ground that the body is national or on the ground that it is in receipt of Exchequer grant or of fees, or because its voluntary income is small. We are satisfied that the practical arguments against such exclusion are strong; that the exclusion would greatly complicate the scheme for relief; and that it would have only a marginal effect on local authorities’ income.”

The point that arises here is not the same as the question we are considering, but it is another example of the Committee’s general preference for what might be described as clear, broad-brush solutions. Another example of its approach is found in this passage from paragraph 98:

“Other bodies are not substantially dependent upon voluntary contributions because they make charges for the services they provide. Some of them provide these services to people who, in the view of some of our local government witnesses, could well afford to pay the extra money which the bodies would need if they were fully rated. But it would be impracticable to distinguish between charities by reference to the means of the people whom they benefit; and a general exclusion of charities which make charges would certainly penalise many worthy bodies which provide services to those in need.”

The Committee then considered how eligibility for relief should be established. It considered and rejected the setting up of a tribunal to decide whether or not an organisation was within the class eligible for rate relief. At paragraph 119 (previously referenced at paragraph 93, see above), it referred to the imminent establishment of a register of charities, duly achieved by the Charities Act 1960, and said:

“119.

Moreover there is in prospect an alternative method of testing the status of an applicant for mandatory relief. The Government have announced their intention of requiring certain classes of charitable trust to register with the Charity Commissioners or the Minister of Education. If compulsory registration were extended to include all charities in rateable occupation of land, it could be made applicable for rating purposes, and we recommend that this extension should be made. Registration would meet the need for a cheap and expeditious test. We assume that a body which considered itself within the field of compulsory registration but was refused registration would have a right of appeal to the courts; and that rating authorities would have the right to challenge in the courts the registration of any body which they did not consider to be a charity.

120.

The compilation of the register may take some time, and possibly the changes in the law concerning the rating of charities which we recommend may become effective before the register is complete. Any disputes which may arise during the interim period should be left for settlement by the courts in the usual way.”

The concluding sentences of paragraph 119 show that the Committee thought that eligibility for rate relief could and should be disputed at the level of charitable status, either by the ratepayer or by the local authority. It might have proposed that local authorities could raise challenges to eligibility at the level of the individual hereditament, but this possibility is not mentioned. Instead, it favoured a system offering “simplicity, certainty and economy in administration” and “a cheap and expeditious test” (paragraphs 65 and 119).

As would be expected, the rating and charities legislation harmonise with each other. The 1988 Act defines a charity as an institution or other organisation established for charitable purposes only: section 67 (10). The same definition appears in section 1 of the Charities Act 2011. An institution registered by the Charity Commission is conclusively presumed to be a charity, originally by virtue of section 5 of the Charities Act 1960, now section 37 of the 2011 Act. The ability to claim or object to the registration of an institution as a charity appeared first in section 5 of the 1960 Act and is now contained in section 36 of the 2011 Act.

In summary, the Pritchard Report shows why the 1961 Act specified that the ratepayer must be a charity, while the 1955 Act did not, but it contains no indication that the wording of section 11 (1) (a) of the 1961 Act was intended to achieve a different result to section 8 (1) (a) of the 1955 Act in any other respect.

Overall, I derive no support for Merton’s argument from the legislative history. On the contrary, it contains a number of clear pointers towards the correctness of the Judge’s conclusion:

The main authorities, Glasgow and Oxfam, have been summarised by the judge and by David Richards LJ above. They make clear that the mere fact that a body is a charity is not sufficient to satisfy the statutory test and that something more is required. In Glasgow the relevant question was whether the use of the employee’s house directly facilitated the charitable purposes of the church. The Oxfam case was concerned with whether the shops were wholly or mainly used for charitable purposes or for fundraising. These issues do not arise in our case, where the premises were wholly used for the purposes of Nuffield Health, a charity. I think the Judge was right in his summary:

“21.

These cases, and the quoted passages from them, show first that it is the charity’s use of the premises that must be considered; second, that if the use which the charity makes of the premises is directly to facilitate or ancillary to the carrying out of its main charitable purpose, that is sufficient to satisfy the requirement that the premises are used for charitable purposes; and third, conversely, that if the charity’s use of the premises is not for its charitable purposes or does not directly facilitate or is not ancillary to its charitable purposes, the charity will not be entitled to relief.”

I do not read the passages from Oxfam, cited by David Richards LJ at paragraph 42, as providing support for Merton’s case. I agree with the treatment of that decision by Nugee LJ at paragraphs 146-148. Mr Ormondroyd relied particularly on the sole opinion of Lord Morris at 148D:

“… the inquiry becomes whether the whole or the main user was for Oxfam's charitable purposes. This in turn involves ascertaining Oxfam's charitable purposes and ascertaining what was the user of the premises and then deciding whether that user was “for” the charitable purposes.”

At 149A, Lord Morris described this as “user in the actual carrying out of the charitable purposes”. If anything, these passages seem to me to point towards a focus on the purposes of the charity. They do not sustain the argument that it is for Nuffield Health to show a public benefit at the premises.

The only other case that I would mention is the ISC decision, which concerned charity law and not rating. The considerations affecting private schools have some resonance in the present case. The central conclusion was that schools that did not provide the necessary element of public benefit could not qualify as a charity: paragraph 177. The Upper Tribunal considered that, when assessing whether a charity is carrying out its purposes in a way which is for the public benefit, the inquiry is whether the activities of the charity overall are for the public benefit: paragraph 195. Importantly,

A number of the policy arguments have already been touched upon. It is said by Merton that absurdities would arise if one did not have to look for public benefit at all in individual hereditaments. An occupier that was registered as a charity would still be entitled to rate relief even if nothing of any public benefit was carried out on the premises. For example, Nuffield Health could raise its fees to five times their current levels with impunity so far as rate relief was concerned: it could get the benefits of being a charity without behaving as a charity. In support of this submission, Mr Ormondroyd cited three cases concerning reliefs from income tax and rates: IRC v Educational Grants Association, Church of Jesus Christ of Latter-Day Saints v Henning [1964] AC 420, and the Public Safety Charitable Trust case. I agree with David Richards LJ that these cases are tangential, and I note that in any event each claim for relief failed, in the first two cases because the defined activity lacked a public character and in the third because the premises were not being used by the charity.

Mr Ormondroyd then submitted that a purposive reading of the statute supported Merton’s case. He took us to the Pritchard Report, but only for the proposition that nothing in it supports a conclusion that charitable purposes should exclude public benefit. I think that the Report goes further than that. He then took us to two instances in which he argued that the court had been prepared to consider the public benefit element; having done so, he realistically accepted that this was not his strongest point, and I need say no more about it.

Finally, it was argued that the burden on a charity of proving public benefit at the individual hereditament was not an onerous one, given the very substantial relief claimed, and that rating authorities are well able for the task. That may be so, but I cannot think that it strongly influences the question of construction. There are a number of public policy factors underlying the legislation and some of these point the other way.

Mr Kolinsky QC’s response can be pithily expressed as “wrong body, wrong test”. As recorded by David Richards LJ at paragraph 33, he argues that Merton’s approach would lead to inconsistencies and distortions arising from the fact that individual local authorities are not the appropriate arbiter of whether a charity is performing its responsibilities. I agree with what Nugee LJ says about this at paragraphs 149-151.

Having considered the statutory language, the legislative history, the case law, and the issues of policy, I have reached the clear conclusion that the construction of section 43(6)(a) proposed by Merton is not correct and I would therefore dismiss Ground 1, and with it the appeal.

I would only add this. Nuffield Health may have succeeded under the rating legislation, but its failure, on our unanimous view, on Ground 3 may not be without consequences in the context of charity law. Its trustees are obliged to satisfy themselves in good faith that its provision is for the public benefit. If the situation at the Premises is replicated across its several hundred fitness centres and gyms, the organisation may face scrutiny through the Charity Commission and ultimately through the courts, as occurred in the ISC case.

I am very grateful to David Richards LJ for his account of the facts and explanation of the issues that arise in this appeal. I can therefore go directly to consideration of Ground 1 which is not only a point of general application but one that I have not myself found entirely easy to resolve. I have however the misfortune to have reached a different conclusion from that of David Richards LJ on this ground and I will try to explain why.

We are concerned in this case with a well-trodden principle, namely the requirement that a charity carry out its activities for the benefit of the public, but in the particular setting of rating law. It is obviously right, as David Richards LJ says (paragraph 35 above), that the overall question is one of statutory construction of section 43(6)(a) of the 1988 Act and in particular of the words “is wholly or mainly used for charitable purposes (whether of that charity or of that and other charities)”. I will refer to the question whether a hereditament is so used as “the statutory question”.

I will start by noting certain points that are common ground or otherwise seem to me relatively straightforward. First, the statutory question has to be asked in relation to the particular hereditament in question, here the premises at Merton Abbey. Local authority rates are the oldest tax in continuous existence in England, having been originally introduced in the reign of Queen Elizabeth I by the Poor Relief Act 1601 (coincidentally in the same session as the Charitable Uses Act 1601, the preamble to which has played such a significant role in the law of charity), and the unit of assessment has always been the hereditament: Woolway (Valuation Officer) v Mazars LLP [2015] UKSC 53 at [1] per Lord Sumption JSC. That remains the case under the 1988 Act, section 43(1) of which provides:

“43 Occupied hereditaments: liability

(1)

A person (the ratepayer) shall as regards a hereditament be subject to a non-domestic rate in respect of a chargeable financial year if the following conditions are fulfilled in respect of any day in the year–

(a)

on the day the ratepayer is in occupation of all or part of the hereditament, and

(b)

the hereditament is shown for the day in a local non-domestic rating list in force for the year.”

Section 43(6)(a) is similarly expressed by reference to whether the hereditament is wholly or mainly used for charitable purposes.

Second, the statutory question requires one to look at the use of the particular hereditament. It is well established that a charity may, lawfully and without being in breach of trust, occupy and use premises for a purpose that is not one of its charitable purposes. The paradigm case is where premises are used to raise money: see the examples referred to by David Richards LJ at paragraph 36 above (premises used as shops (Oxfam) or as investments (Polish Historical Institution Ltd v Hove Corporation)). The distinctive feature of premises used in this way is that the use is not directly in fulfilment of the charity’s objects. Oxfam’s objects were to relieve poverty, distress and suffering in any part of the world, and selling goods in its shop did not (directly) relieve poverty. The Polish Historical Institution was formed to promote research into, and the study of, modern Polish history, and the running of a boarding house in Hove did not (directly) promote such research or study. What made the use of the premises lawful for each charity was that holding assets and turning them to account was a means of raising funds which supported its charitable work, but the use of the premises itself was not part of that work.

Third, an institution that is registered with the Charity Commission is by section 37 of the 2011 Act conclusively presumed (save for the purpose of rectifying the register) to be a “charity”. Nuffield Health is on the register and so benefits from this presumption. Although there is no dispute about this, it is worth spelling out what this means. By section 1(1) of the Act a charity means an institution which (a) is established for charitable purposes only and (b) is subject to the control of the High Court in the exercise of its charity jurisdiction. I can pass over (b) on which nothing turns, but section 1(1)(a) brings in section 2(1) which provides that a charitable purpose is a purpose which (a) falls within section 3(1) and (b) is for the public benefit, as explained in section 4. It follows that the effect of section 37 is that Nuffield Health’s purposes are conclusively presumed not only to fall within section 3(1) (in fact within section 3(1)(d) (“the advancement of health…”) and perhaps also section 3(1)(j) (“the relief of those in need because of … ill-health…”)), but to be for the public benefit within the meaning of section 4, and hence charitable. It is also to be noted that because an institution can only be a charity if is established for charitable purposes only, that all its purposes are presumed to be charitable and hence to be both within section 3(1) and for the public benefit.

Fourth, with that introduction, the issue which divides the parties and which we have to resolve can I think be summarised as follows. Where a registered charity (here Nuffield Health) uses a hereditament (here Merton Abbey) for a particular purpose (here providing gym and other facilities), what is the statutory question which determines whether it is entitled to mandatory rate relief as provided for by section 43(5) and 43(6)(a) of the 1988 Act? Is it:

As with any question of statutory construction, resolution of this issue depends on the use of a number of interpretative tools among which are the language used, the relevant legal context and the practical consequences of the rival interpretations, the overall aim being to ascertain Parliament’s intention in using the words that it did.

Starting with the language, words in a statute are of course to be given their natural and ordinary meaning unless there is good reason not to. For some time I was attracted by the suggestion put forward by Mr Ormondroyd on behalf of Merton that a simple reading of section 43(6)(a) compelled one to ask the question whether the use of the hereditament in question was itself for charitable purposes and hence to ask whether such use was itself for the public benefit (ie Question (B)). But on further reflection I have come to the view that the language does not inevitably lead to this conclusion and that Question (A) is itself a perfectly natural reading of the statutory language.

That can be seen by breaking Question (A) down into steps. The first step is to identify the purposes of the charity in question. Here that is found in clause 3.1 of Nuffield Health’s memorandum of association, as amended in 2010, namely:

“to advance, promote and maintain health and healthcare of all descriptions and to prevent, relieve and cure sickness and ill health of any kind, all for the public benefit”

(see paragraph 4 of David Richards LJ’s judgment above). By virtue of the registration of Nuffield Health as a charity, those purposes are, as already explained, conclusively presumed to be charitable. The second step is to ask if the relevant hereditament, Merton Abbey, is being used for those purposes, or one of those purposes. Here the answer is Yes, as the provision of the gym and other facilities at Merton Abbey is a means adopted by Nuffield Health of pursuing its objects of advancing health and of preventing sickness and ill health. I did not understand Mr Ormondroyd to contend to the contrary. His argument was not that what Nuffield Health did at Merton Abbey was not for the advancement of health, but that it was not for the advancement of health for the public benefit.

The language of section 43(6)(a) requires the charity to use the hereditament for charitable purposes. To my mind it is a perfectly natural reading of this language to answer it by saying that a charity is using a hereditament for charitable purposes if (i) it is using it for its purposes and (ii) those purposes are charitable. Indeed in argument Peter Jackson LJ said that the argument for Nuffield Health could be summarised as being that the occupier has to be a charity that is using the premises for its purposes.

I agree, and if section 43(6)(a) had read:

“the ratepayer is a charity … and the hereditament is … used for its

charitable purposes” then I think that Question (A) would be the right question. One first identifies the charity’s charitable purposes and then asks if the hereditament is used for those purposes.

Now in fact section 43(6)(a) reads as follows:

“the ratepayer is a charity … and the hereditament is … used for charitable purposes (whether of that charity or of that and other charities).”

But it is to be noted that the statutory question is not simply whether the premises are used for charitable purposes. It includes the words in brackets, which are designed to cater for two possibilities, namely one where the purposes are those of the charity in question alone, and the other where the purposes are those of that charity and other charities. If one splits out these two possibilities, the question can be restated as follows:

“the ratepayer is a charity … and the hereditament is … used for charitable purposes of that charity, or is … used for charitable purposes of that and other charities.”

In the present case of course there is no suggestion that Merton Abbey is used for the purposes of some other charity as well as Nuffield Health. So the only question that needs to be answered is this:

“the ratepayer is a charity … and the hereditament is … used for charitable purposes of that charity.”

That is to my mind indistinguishable from Peter Jackson LJ’s suggestion that the question is whether the hereditament is being used by a charity for its charitable purposes.

On analysis, therefore, not only do I think that Question (A) is a perfectly tenable reading of the statutory question, but I have come to the conclusion that it is actually the preferable reading, purely as a matter of language. In the present case it can be answered simply by Nuffield Health as follows: “I am a registered charity. My purposes, namely the advancement of health, are therefore (conclusively presumed to be) exclusively charitable. I am using Merton Abbey for those purposes. That is sufficient.”

Having reached this conclusion on the language of the 1988 Act, the remaining question is whether there is anything in the legal context, or the practicalities, or other considerations, which militate against this conclusion. I will say at once that I do not think there is.

So far as legal context is concerned, I start with the general law of charity which is by now well established. The question of public benefit in charity law has given rise to a fair amount of authority, some of which we were shown in connection with Questions

For present purposes it is not necessary to go over that material in any detail. But a few points can be noted. First it has always been the case that a purpose will only be charitable if it is for the public benefit. This is now given statutory force by section 2(1)(b) of the 2011 Act, but this has not in general changed the substance of the law: see section 4(3), and the consideration of this point in ISC. Second, there are two aspects to the requirement of public benefit, namely the nature of the purpose and the reach of the purpose, and in the present case we are concerned with the latter (see paragraph 23 of David Richards LJ’s judgment). Third, in assessing whether an institution’s purposes are for the public benefit, the inquiry is whether the activities overall are for the public benefit: ISC at [95]. That is why an institution whose purposes are charitable overall may nevertheless carry out particular activities in pursuit of those purposes that would not be sufficiently for the public benefit, and hence would not be charitable, if carried out alone.

Against that legal background, there is I think some force in the submission made by Mr Kolinsky that it cuts across the established principles of charity law to suggest that for rating purposes one not only looks at each hereditament separately to see what use is being made of it (which I accept is necessary, as I have already said) but also looks at that use separately to see if it would be charitable if it stood by itself. It is a fundamental principle of charity law (now found in section 1(1)(a) of the 2011 Act) that an institution is only a charity if it is established for charitable purposes only. Given that, it seems odd to say that for the purposes of charity law Nuffield Health’s objects or purposes are exclusively charitable, but for rating purposes what it does in pursuance of those objects on particular premises might or might not be.

David Richards LJ has at paragraph 17 above traced some of the history of the introduction of mandatory rating relief for charities which is now found in section 43(5) and section 43(6)(a) of the 1988 Act. As there explained such relief was first introduced in what was in effect its current form in section 11 of the Rating and Valuation Act 1961, following the Pritchard Report. That report refers to some of the anomalies and difficulties under the previous law and recommends the introduction of mandatory rating relief for charities. The following points can be noted:

It is not disputed that this report is admissible. Such reports are admissible not for the purpose of shedding light directly on the language of the statutory provision in question but for the purpose of identifying the mischief at which the statute was directed. They are therefore of limited assistance, and Mr Kolinsky accepted that he did not mean to place too much reliance on the Pritchard Report. Nevertheless the general point that the aim of the legislation was to produce uniformity and certainty, and be practicable to operate, does seem to me to be of some help. I will come back below to the practical consequences of each party’s interpretation.

So far as the authorities are concerned, the leading ones are Glasgow Corporation v Johnstone [1965] AC 609 (“Glasgow”) and Oxfam. The issue in Glasgow is set out in David Richards LJ’s judgment at paragraph 41 above. As he explains the House of Lords held that the use by the board of the house was for charitable purposes, Lord Reid saying at 622:

“If the use which the charity makes of the premises is directly to facilitate the carrying out of its main charitable purposes, that is, in my view, sufficient to satisfy the requirement that the premises are used for charitable purposes”

He also referred (again at 622) to the question whether use of the premises was “wholly ancillary” to an institution’s charitable purpose, giving the example of accommodation for nurses at a hospital. In the present case it is not suggested that the use by Nuffield Health is such as “directly to facilitate” the carrying out of its main charitable purposes or is “wholly ancillary” to them; but it is one of the means by which Nuffield Health carries out those purposes. To my mind it seems surprising that something which is merely ancillary to, or facilitates, the carrying out of a charitable purpose should qualify for relief, but the actual carrying out of activities falling within its main purpose should not.

Oxfam, as David Richards LJ explains at paragraph 42 above, concerned the question whether shops operated by Oxfam were entitled to mandatory rating relief. The leading judgment was given by Lord Cross, with whom Lords Simon, EdmundDavies and Fraser agreed. Lord Cross, having pointed out that the relevant section required a line to be drawn between user by a charity for charitable purposes and user for purposes that were not charitable (at 135F-G), identified where that line should be drawn at 146A-D, saying:

“I think, in agreement with counsel, that the choice is between (A) drawing the line so as to exclude from relief user for the purpose of getting in, raising or earning money for the charity, as opposed to user for purposes directly related to the achievement of the objects of the charity, and (B) only excluding from relief user for the purpose of carrying on a business to earn money for the charity….

In my judgment, the first alternative is to be preferred.”

If that is where the line is to be drawn, it seems to me to follow that Lord Cross was of the view that user for purposes directly related to the achievement of the objects of the charity would qualify for relief. But the user by Nuffield Health of Merton Abbey is for purposes directly related to the achievement of its objects of advancing health and preventing ill-health.

Lord Morris gave a concurring judgment. At 148D he said this:

“As there is no suggestion that the user was for the charitable purposes of any other charity, the inquiry becomes whether the whole or the main user was Oxfam’s charitable purposes.”

That seems to me to be the same point as I have tried to make in paragraph 137 above. He continued at 148F-G:

“There being a distinction between, on the one hand, activities which a charity may undertake, and, on the other hand, activities which consist in the actual carrying out of its charitable purposes, it is manifest that some activities are on one side of the line and some activities are on the other.”

If that is the test, it seems to me that the activities that Nuffield Health carry out at Merton Abbey are activities which consist in the actual carrying out of its charitable purposes.

I accept that in neither case was there any consideration of the present point, as in neither case was there a question raised whether the use of the particular hereditament was for the public benefit. They are not therefore directly binding on the point. But for the reasons I have given they seem to me more consistent with Nuffield Health’s position than with Merton’s. Both cases suggest that the line is drawn between the charity using the premises directly to further its charitable work (or in other words a functional use), and using the premises indirectly to support its work by getting in, raising or earning money (an investment or business use). Or, to put it another way, the relevant question is whether the charity is using the hereditament for its charitable purposes, not whether the activity carried on at the hereditament would qualify as a

Finally on Ground 1, I add a few words on the practical considerations. First, if Nuffield Health is right, the question should be relatively easy to answer, and consistent across different local rating authorities. Such authorities would not need to concern themselves with the question whether the particular use of the hereditament was sufficiently for the public benefit, something which, as the present case illustrates, might give rise to quite extensive investigation. In the present case that involved not only demographic evidence designed to address whether the membership fees excluded those of modest means (although the evidence on this was not in fact particularly helpful), but also evidence as to whether the services provided to nonmembers were token or not: see David Richards LJ’s judgment at paragraphs 70 to 78 above. If Merton is right, then, depending on the evidence, gyms offering much the same services might be held to be charitable in one local authority area but not in another. But if Nuffield Health is right, the only enquiry a local authority would need to undertake is to ascertain what the charity’s objects are, something which will usually be readily ascertainable from its written constitution, and then ask whether the hereditament in question is being directly used for those purposes. There will be no need to carry out any investigation into the affordability of the services provided, or the extent to which they benefit the public, at the particular site.

Second, I think there is something in Mr Kolinsky’s point that it is preferable for the assessment of whether an institution’s purposes satisfy the public benefit requirement to be carried out once by the Charity Commission (based on a review of the institution’s activities as a whole) rather than repeatedly by different local authorities based on a review of its activities on a site by site basis. Many charities of course only occupy one hereditament, but large charities such as Nuffield Health operate in many different local authorities, and on Merton’s argument the question of public benefit would have to be considered anew by each local authority and in relation to each site. That seems duplicative and potentially wasteful of resources; moreover one would have thought that the Charity Commission, by virtue of its role as regulator of charities, was both better placed and more experienced in making such assessments than local authorities. The latter have many varied functions to fulfil, but regulating charities is not one of them.

Third, I also accept Mr Kolinsky’s submission that Merton’s argument might lead to anomalous results in the case of an institution split across several sites. He gave the example of a charitable trust that operated five fee-paying schools on different sites, where free education was provided on Saturdays at one of the sites. On Merton’s approach, the other four sites would not qualify for rating relief. That does not seem a particularly sensible result, given that the entire trust is charitable and, if right, might lead to behaviour deliberately designed to counteract it, for example by rotating the free Saturday education from site to site. It does not seem very sensible that charities should be obliged to adopt what might be inefficient practices of this sort in order to access rating relief.

Having considered not only the language of section 43(6)(a) but also the legal context and certain practical considerations, I have come to the conclusion that the better view is that it was not Parliament’s intention that the question of public benefit should be assessed separately for each site on which a charity carries out its charitable activity. I prefer Nuffield Health’s submissions on Ground 1 and would dismiss the appeal on this ground.

If I am right on Ground 1, Grounds 2 and 3 do not arise, as Mr Ormondroyd accepted. If I am wrong on Ground 1, I agree with David Richards LJ’s analysis and conclusions on both Ground 2 and Ground 3.

Mr Ormondroyd submitted that Ground 4 was a freestanding ground that did not depend on Ground 1. That was not I think accepted by Mr Kolinsky, and I have doubts about it. Merton Abbey was not being used for two purposes, one functional and one not. It was being used for a single purpose, that of providing gym and other facilities to members in return for membership fees which were (according to Merton) set at a level designed to generate a surplus. That was part of the way in which Nuffield Health fulfilled its object of advancing health. In those circumstances I incline to the view that it was being “wholly” used for that purpose, and the fact, if it was a fact, that it was operated in such a way to generate a surplus, did not mean that it was being used for some other purpose. If therefore I am right on Ground 1, I do not think that Ground 4 can assist Merton. But in any event, if I am wrong on this, I

I would therefore dismiss the appeal, although I agree that if I am wrong on Ground 1, the appeal should be allowed on Ground 3.