Hurstwood Properties (A) Ltd and others v Rossendale Borough Council and another

The question on these appeals is whether the appellant local authorities have reasonable grounds for claiming non-domestic rates on certain empty properties from the respondent companies. In bare outline, non-domestic rates (or business rates) are charged on the occupier of a property which is not a domestic property or, if the property is unoccupied, on the owner of the property, subject to certain exceptions. The owner is defined as the person entitled to possession of the property. One of the exceptions is where the owner (as defined) is a company in liquidation, including voluntary liquidation. The respondents are the registered owners of various unoccupied commercial properties who have sought to avoid liability for business rates by means of one or other of two closely related schemes.

Both schemes involve setting up a special purpose vehicle (“SPV”) in the form of a company without any assets or business. The registered owner grants a short lease of the unoccupied property to the SPV. The premise of the schemes is that the SPV thereupon becomes the “owner” of the property for the purpose of the liability for business rates. The SPV is immediately put into members’ voluntary liquidation or, alternatively, is dissolved. In the liquidation version of the scheme, reliance is then placed for as long as possible on the exemption which applies where the owner of the property is being wound up. The dissolution version of the scheme relies on the fact that, upon dissolution, the lease and associated liability for rates is automatically transferred by law as bona vacantia to the Crown (or, as appropriate, to the Duchy of Lancaster or Cornwall). Meanwhile the registered owner is relieved from paying business rates, either until it terminates the lease because it has a tenant or other use for the property or until the lease is disclaimed by the liquidator or by the Crown.

On the assumption that the schemes work at all, both versions rely for their effectiveness over time upon administrative inertia. In the liquidation version of the scheme, the period until the necessarily onerous leases are disclaimed by the liquidator is deliberately prolonged. In the dissolution version, there may be delay before the SPV is dissolved; thereafter the scheme relies upon the unlikelihood that the local authority will find out about the dissolution for a considerable time, and then prompt the unsuspecting Crown to disclaim the lease. In either version of the scheme, if and when the relevant lease is disclaimed so that the registered owner regains its entitlement to possession, the scheme can simply be repeated using a fresh SPV.

The liquidation version of the scheme (in the form described in this judgment) has already been judicially branded an abuse of the insolvency legislation in proceedings for the winding up in the public interest of a company which promoted and managed such a scheme: see In re PAG Management Services Ltd [2015] EWHC 2404 (Ch); [2015] BCC 720. As will appear, the dissolution version of the scheme is no less an abuse of legal process and may in certain circumstances involve the commission of a criminal offence.

It is common ground that the schemes have no business or other “real world” purpose and that their sole purpose is to avoid liability to pay business rates. But, subject to one new point, dealt with below, it is also now common ground that the leases granted to the SPVs were not shams so that, as a matter of the law of real property, they conferred an entitlement to possession upon the SPVs. An argument that the leases were shams was rejected at first instance and has not been resurrected on appeal.

The two present claims, brought by Rossendale Borough Council and Wigan Council against, respectively, companies in the Hurstwood Group and Property Alliance Group Ltd, have been selected as test cases representative of 55 similar cases where one of the schemes outlined above or a materially similar scheme has been employed. The values of the claims for unpaid rates made in these cases vary from a few thousand pounds to millions of pounds.

In their particulars of claim in these proceedings the local authorities have alleged, in relation to each property in respect of which they are claiming unpaid rates: (i) that the lease to the SPV, if not a sham, was ineffective to make the SPV the “owner” of the unoccupied property within the meaning of the applicable legislation; alternatively, (ii) that the separate legal personality of the SPV should itself be ignored for this purpose. By either technique they seek to identify the defendant as the entity liable for the unpaid business rates.

On the defendants’ application to have the particulars of claim struck out on the basis that they disclose no reasonable grounds for bringing the claims, the judge (Judge Hodge QC) ruled in the defendants’ favour on the first point but not in relation to the alternative case based on “piercing the corporate veil”: [2017] EWHC 3461 (Ch). The Court of Appeal (David Richards, Henderson and Baker LJJ) decided both issues in the defendants’ favour and struck out all the claims: [2019] EWCA Civ 364; [2019] 1 WLR 4567. The local authorities appeal to this court on both points.

The first way in which the local authorities advance their claim that the defendants are liable for the unpaid rates relies on the approach to statutory interpretation associated in the field of tax legislation with the case of WT Ramsay Ltd v Inland Revenue Comrs [1982] AC 300. What has often been referred to as the Ramsay principle or doctrine may be said now to have reached a state of well-settled maturity, not least because of its restatement at the highest level in two 21st century authorities: Barclays Mercantile Business Finance Ltd v Mawson [2004] UKHL 51; [2005] 1 AC 684 and UBS AG v Revenue and Customs Comrs [2016] UKSC 13; [2016] 1 WLR 1005. Although usually deployed in relation to tax avoidance schemes, it is not in its essentials particular to tax, being based upon the modern purposive approach to the interpretation of all legislation, one which penetrated the field of tax legislation only at a relatively late stage: see Barclays Mercantile at paras 28-29; and UBS at paras 61-63.

There are numerous authoritative statements in modern case law which emphasise the central importance in interpreting any legislation of identifying its purpose. Two examples will suffice. In R (Quintavalle) v Secretary of State for Health [2003] UKHL 13; [2003] 2 AC 687, para 8, Lord Bingham of Cornhill said:

The result of applying the purposive approach to fiscal legislation has often been to disregard transactions or elements of transactions which have no business purpose and have as their sole aim the avoidance of tax. This is not because of any principle that a transaction otherwise effective to achieve a tax advantage should be treated as ineffective to do so if it is undertaken for the purpose of tax avoidance. It is because it is not generally to be expected that Parliament intends to exempt from tax a transaction which has no purpose other than tax avoidance. As Judge Learned Hand said in Gilbert v Commissioner of Internal Revenue (1957) 248 F 2d 399, 411, in a celebrated passage cited (in part) by Lord Wilberforce in Ramsay [1982] AC 300, 326:

Another aspect of the Ramsay approach is that, where a scheme aimed at avoiding tax involves a series of steps planned in advance, it is both permissible and necessary not just to consider the particular steps individually but to consider the scheme as a whole. Again, this is no more than an application of general principle. Although a statute must be applied to a state of affairs which exists, or to a transaction which occurs, at a particular point in time, the question whether the state of affairs or the transaction was part of a preconceived plan which included further steps may well be relevant to whether the state of affairs or transaction falls within the statutory description, construed in the light of its purpose. In some of the cases following Ramsay, reference was made to a series of transactions which are “pre-ordained”: see eg Inland Revenue Comrs v Burmah Oil Co Ltd [1982] STC 30, 33 (Lord Diplock); Furniss v Dawson [1984] AC 474, 527 (Lord Brightman). As a matter of principle, however, it is not necessary in order to justify taking account of later events to show that they were bound to happen - only that they were planned to happen at the time when the first transaction in the sequence took place and that they did in fact happen: see Inland Revenue Comrs v Scottish Provident Institution [2004] UKHL 52; [2004] 1 WLR 3172, para 23, where the House of Lords held that a risk that a scheme might not work as planned did not prevent it from being viewed as a whole, as it was intended to operate.

The decision of the House of Lords in the Barclays Mercantile case made it clear beyond dispute that the approach for which the Ramsay line of cases is authority is an application of general principles of statutory interpretation. Lord Nicholls of Birkenhead, delivering the joint opinion of the appellate committee (which also comprised Lord Steyn, Lord Hoffmann, Lord Hope of Craighead and Lord Walker of Gestingthorpe), identified the “essence” of the approach (at para 32) as being:

Almost all statements of principle, however broadly framed, tend to be responsive to the particular facts under review. The above statements refer to “the transaction” and most of the leading expositions of the Ramsay doctrine do the same. This is because most of the provisions being considered taxed, or as the case may be exempted, transactions. But not all do. Some involve a tax (such as stamp duty) on instruments. Others impose charges by reference to the status of a person, or their rights in relation to specified property, such as the owner of unoccupied non-domestic property in the present case. The Ramsay doctrine is no less applicable in such cases. In MacNiven (HM Inspector of Taxes) v Westmoreland Investments Ltd [2001] UKHL 6; [2003] 1 AC 311, 320, para 8, Lord Nicholls said:

In the task of ascertaining whether a particular statutory provision imposes a charge, or grants an exemption from a charge, the Ramsay approach is generally described - as it is in the statements quoted above - as involving two components or stages. The first is to ascertain the class of facts (which may or may not be transactions) intended to be affected by the charge or exemption. This is a process of interpretation of the statutory provision in the light of its purpose. The second is to discover whether the relevant facts fall within that class, in the sense that they “answer to the statutory description” (Barclays Mercantile at para 32). This may be described as a process of application of the statutory provision to the facts. It is useful to distinguish these processes, although there is no rigid demarcation between them and an iterative approach may be required.

Both interpretation and application share the need to avoid tunnel vision. The particular charging or exempting provision must be construed in the context of the whole statutory scheme within which it is contained. The identification of its purpose may require an even wider review, extending to the history of the statutory provision or scheme and its political or social objective, to the extent that this can reliably be ascertained from admissible material.

Likewise, the facts must be also be looked at in the round. In Inland Revenue Comrs v McGuckian [1997] 1 WLR 991, 999, Lord Steyn explained that it was the formalistic insistence on examining steps in a composite scheme separately that allowed tax avoidance schemes to flourish. Sometimes looking at a composite scheme as a whole allows particular steps which have no commercial purpose to be ignored. But the requirement to look at the facts in the round is not limited to such cases. Thus, in Scottish Provident where the taxing statute granted an allowance which depended upon the taxpayer having an entitlement to a specified type of property (gilts), a view of the facts in the round enabled the House of Lords to conclude that a legal entitlement to gilts generated by one element in a larger scheme failed to qualify because the entitlement was intended and expected to be cancelled out by an equal and opposite transaction.

In the present case the operative statutory provision is section 45 of the Local Government Finance Act 1988 (the “1988 Act”), which imposes a charge to rates on unoccupied property in the following terms:

The regulations referred to in section 45(1)(d) are the Non-Domestic Rating (Unoccupied Property) (England) Regulations 2008 (SI 2008/386) (the “2008 Regulations”). They replaced, without major changes, the Non-Domestic Rating (Unoccupied Property) Regulations 1989 (SI 1989/2261). By regulation 3 of the 2008 Regulations, all relevant non-domestic hereditaments are prescribed for the purposes of section 45(1)(d) other than those described in regulation 4. Regulation 4 lists classes of non-domestic hereditament which, by way of exception, do not when unoccupied give rise to a liability to pay rates. These include a hereditament:

The levying of rates on the owners of unoccupied property was not new in 1988. Rates, both on domestic and non-domestic property, had grown out of the Poor Relief Act 1601, originally as a charge on the inhabitants and occupiers of land to “enable the poor to be set to work and to maintain the lame, impotent, halt and blind”: see Cross on Local Government Law, looseleaf, ed, para 15-01. Rates were for centuries a charge on occupation rather than the ownership of land, but their rationale changed over time from poor relief to payment for the provision of local services. The boundaries of rateable occupation were essentially judge-made, mainly in 19th century cases. The charge affected all those in actual occupation, with or without title to occupy: see R v St Pancras Assessment Committee (1877) 2 QBD 581, 588-590 (Lush J).

The question whether property is occupied and, if so, who is the occupier for rating purposes is still largely governed by the common law rules: see section 65(2) of the 1988 Act. The classic statement of those rules is that of Tucker LJ in John Laing & Son Ltd v Assessment Committee for Kingswood Assessment Area [1949] 1 KB 344, 350, recently reaffirmed by this court in Cardtronics UK Ltd v Sykes (Valuation Officers) [2020] UKSC 21; [2020] 1 WLR 2184, para 13:

Provision for levying rates on unoccupied property was first made by sections 20 to 22 of the Local Government Act 1966, and then in similar terms in the General Rate Act 1967 (the “1967 Act”). In its February 1966 Command Paper, Local Government Finance England and Wales (Cmnd 2923), which preceded the introduction of the legislation, the Government proposed to “provide for the payment of part rates on properties which remain unoccupied for more than a limited period”. This reform was said to be necessary because the notion “that properties should remain empty and available for occupation for long periods in conditions of scarcity is an affront to all right-thinking people”. Making rates payable on empty property should also “reduce the present waste of accommodation”.

In Hastings Borough Council v Tarmac Properties Ltd [1985] 1 EGLR 161 Lawton LJ said that the mischief with which the relevant statutory provisions were intended to deal “can be clearly identified. Parliament wanted to stop the owners of premises … leaving them unoccupied to suit their own convenience and to their own financial advantage”.

The statutory regime remained largely intact until further reforms made by the 1988 Act, the main underlying purpose of which was to replace the domestic rate with the (then) community charge, subsequently itself replaced by the council tax. Rates on non-domestic property survived as the rump of the old general rating system, with its central rationale unchanged. During the passage through Parliament of the Bill which became the 1988 Act the Government made it clear that it was not seeking to modify the justification underpinning the rates regime. When the Bill was considered in committee in the House of Lords, the Earl of Caithness (then Minister of State for the Environment whose department was responsible for the Bill) acknowledged that:

The aim of deterring owners from leaving property unoccupied for their own financial advantage and encouraging them to bring empty property back into use for the benefit of the community at large is further reflected in the exceptions to liability listed in regulation 4 of the 2008 Regulations, and also in the zero-rating scheme for charities and community amateur sports clubs in section 45A of the 1988 Act (added by the Rating (Empty Properties) Act 2007). The thrust of the exceptions is to exclude properties where, for varying reasons, the owner either (i) may be unable to bring the property back into occupation, or (ii) may be regarded as having a reasonable excuse for not doing so, or (iii) may be making some other valuable contribution to society by being the owner, in lieu of paying rates.

Thus, within the first of these categories, regulation 4(c) provides an exemption where occupying the property is unlawful. In the second category regulation 4(d) does so where property is kept vacant ahead of planned compulsory acquisition, and regulation 4(e) does so where the property is subject to a building preservation notice or listed as such. Regulation 4(h) to (m) exclude properties whose owner is an office holder, whether a personal representative, trustee in bankruptcy, liquidator or administrator, in each case subject to duties (eg to realise by sale) which will or may conflict with securing early occupation. More generally, regulation 4(a) and (b) exclude properties which have only been unoccupied for short periods. Within the third category is regulation 4(f), which exempts ancient monuments or archaeological sites, and the zero-rating scheme for charities and community clubs, where the property is temporarily vacant prior to occupational use for public benefit.

Some of these exceptions date back to when rates were first imposed on empty property: see paragraph 2 of Schedule 1 to the 1967 Act. Others have been added over time, but they all tend to reinforce the purposes underlying the legislation as described above.

The imposition of empty property rates on owners, and the identification of the owner as the person entitled to possession of the relevant hereditament, is also as old as the legislative scheme itself: see paragraphs 1(1) and 15 of Schedule 1 to the 1967 Act. The concept of a right to possession has a well settled meaning in the law of real property. The person entitled to possession of real property is the person who, as a matter of title, against the rest of the world (including any other person interested in the property such as a landlord) has the legal right to enjoy the property for the time being, whether by personal occupation or exercising the right to put others into possession or occupation, as tenants or licensees or otherwise. Such a person also, of course, has the right to decide to keep the property unoccupied.

In Westminster City Council v Haymarket Publishing Ltd [1981] 1 WLR 677, the question arose as to whether a mortgagor or mortgagee was the “owner” of an unoccupied property for the purpose of the liability to pay rates under the 1967 Act. Lord Denning MR said (at p 680):

A central question in the present appeals is: why in the case of unoccupied property does the legislation impose the liability to pay rates on the person entitled to possession of the property? To the extent that the charge merely raises funds to help pay for local services the answer may not be obvious. Parliament may choose to levy a form of property tax on any class of persons interested in property. But in relation to the central purpose of providing an incentive to bring unoccupied property back into use, the intention is clear. It focuses the burden of the rate precisely on the person who has the ability, in the real world, to achieve that objective.

Normally the examination of the facts in the round required by stage two of the Ramsey approach takes place by reference to facts found by a judge at trial. In the present appeals, by contrast, the examination necessarily proceeds by reference to the facts alleged by the local authorities in their statements of case which are assumed to be true for the purposes of the defendants’ applications to strike out the claims. The particulars of claim are somewhat exiguous, understandably so because disclosure has not yet taken place. But the bare bones pleaded in the particulars of claim have been fleshed out to some extent by the provision of sample leases and rate demands (and responses to such demands) and by helpful annexes to the Statement of Facts and Issues agreed by the parties for the purposes of these appeals. These annexes give relevant particulars of the SPVs and leases used to implement the schemes in the present cases. One of the schemes was also examined in detail by Norris J in PAG Management Services, with the benefit of substantially greater material than is available on a strike-out application. It is not suggested that Norris J’s analysis of the scheme in issue before him was in any respect factually inaccurate.

The two test cases before this court illustrate the two different versions of the basic scheme mentioned earlier. The common features are: (i) the setting up of an SPV without any assets or any actual or intended business; and (ii) the grant of a lease by the registered owner of the unoccupied property to the SPV which, as a matter of real property law, is framed so as to confer an entitlement to possession of the property upon the SPV, but on terms which enable the lessor promptly to terminate the lease and so re-acquire entitlement to possession, if and when it identifies an occupational use or business tenant for the property.

The features which distinguish the two versions of the scheme are those which affect the life of the SPV thereafter. Under the variant employed by the Hurstwood companies (Scheme A), the SPV was dissolved without any prior liquidation process, so that the liability for rates passed from the SPV to the Crown upon the vesting of the lease in the Crown as bona vacantia. Under the variant used by Property Alliance Group Ltd (Scheme B), the SPV was placed in members’ voluntary liquidation within a few days of the grant of the lease so as to trigger the winding-up exemption in regulation 4(k) for as long as the liquidation could be made to last without the lease being disclaimed. In both cases the SPV was set up and (to the minimal extent required) operated by the promoter of the scheme for a fee consisting of a substantial part of the rates saved by the then owner of the property (“the landlord”). The SPV was not directly owned by the landlord, and it is a triable issue whether it was even beneficially owned or controlled by it. All the landlord had to do was grant the lease and, if rates were demanded by the local authority, direct the authority to the SPV as the person entitled to possession of the empty property.

Before reviewing the schemes employed by the defendants in more detail it is convenient to begin with the leases. The sample lease of Unit 8, Three Point Business Park, Charles Lane, Haslingden, Lancs, from the registered owner Hurstwood Properties (R) Ltd to the SPV, named ACE Recruitment (Northern) Ltd, is proffered by the parties as a suitable example. It granted a two-year term from 19 December 2011. Although the agreed statement of facts and issues describes the rent reserved by the leases as typically £1 or a peppercorn, this particular lease reserved a rent of £7,000 per annum. Nonetheless, it is common ground that the rent was not intended to be demanded or paid. The lease contains a business user clause defined by reference to planning use classes and detailed provision for payment of service charges, none of which can have been intended to be implemented, as well as a provision, described as a break clause, for early termination by the landlord only, on one month’s written notice.

Under the heading “The Landlord’s Obligations” and the quaint sub-heading “Quite possession” (sic), there is a qualified covenant for quiet enjoyment on the condition (amongst others) that the tenant pays the rent. At the hearing of these appeals Mr Mathew QC for the appellant local authorities submitted orally (and for the first time in these proceedings) that, because no rent was in fact paid, a precondition for conferring an entitlement to possession of the property was never satisfied, so that the schemes failed even on their own terms regardless of the two grounds of challenge raised in the particulars of claim and maintained on these appeals. That submission is, with respect, misconceived. A lease confers a right to possession on the tenant as a proprietary interest in the demised premises by virtue of the grant of exclusive possession for a term and at a rent, irrespective of whether any condition for the enforcement of the personal covenant for quiet possession, such as the actual payment of rent, has or has not been satisfied.

The particulars of claim originally included allegations that the leases were a sham. For a transaction or document to be characterised as a sham in English law, it is necessary to show that the parties intended that the transaction or document should not actually create legal rights and obligations but should merely appear to do so, with the object of deceiving third parties: Snook v London and West Riding Investments Ltd [1967] 2 QB 786, 802. The judge held that the particulars of claim disclosed no reasonable grounds for making such an allegation because the necessary facts, in particular a dishonest common intention, were not pleaded. The judge therefore ordered the allegations of sham to be struck out from the particulars of claim. There has been no appeal against that part of the judge’s order. In any event, it seems plain that the parties to the leases did indeed intend that the leases should confer a legal entitlement to possession on the SPV, so as to seek to bring the SPV within the definition of the “owner” of the property in section 65(1) of the 1988 Act. As both Lord Steyn in McGuckian at p 1001 and Lord Reed in the UBS case at para 67 noted, tax avoidance is the spur to executing genuine documents and entering into genuine arrangements. Nonetheless, the requirement to review the facts in the round calls for an examination of the other elements of the two schemes, all of which, like the formation of the SPVs and the grant of the leases, were pre-planned and designed purely for the avoidance of rates without any business or other purpose.

Under Scheme A, employed by the Hurstwood companies, the demise of the SPVs was achieved by their dissolution without any prior winding-up or administration. The dates given in the annexes to the agreed statement of facts indicate that the dissolution took place months, and sometimes years, after the grant of the relevant lease. By the time they were dissolved, the SPVs (on the assumption that the leases made them the owners of the relevant properties for the purposes of business rates) had incurred substantial liabilities for rates, even if delays before the local authorities found out about the leases and made demands on the SPVs meant that rates did not immediately become payable. The judge found that liability for rates accrues day by day, even if it does not become payable until demanded. On the clear wording of section 45 of the 1988 Act (quoted at para 18 above), he was plainly right on this point. See also Jervis v Pillar Denton Ltd [2015] Ch 87, para 74.

Two methods of dissolving a company appear to have been used. Under section 1000 of the Companies Act 2006 a company’s name may be struck off the register, after a qualifying period of inactivity and prescribed enquiries, on the initiative of the Registrar. The company is dissolved upon publication in the Gazette of a notice that its name has been struck off. Dissolution of the company does not discharge its directors or managers from their liabilities, nor prevent the company from being wound up: see section 1000(7).

An alternative route is provided by section 1003 of the Companies Act, which enables a company’s directors to apply on its behalf for its name to be struck off the register. Again, the company is dissolved upon publication of a notice in the Gazette, with similar provisions to preserve liabilities of its directors and managers and the court’s power to wind up the company. Under section 1006, a director who applies under section 1003 for the company to be struck off must secure that every creditor (including prospective creditors) of the company is notified of the application within seven days of making it. Failure to do so is a criminal offence, and an aggravated offence if done with the intention of concealing the making of the application from the person concerned: see section 1006(4).

One effect of the dissolution of a company (if not subsequently restored to the register for the purpose of being wound up) is that, pursuant to section 1012 of the Companies Act, its property (including leasehold property) vests in the Crown as bona vacantia. On the assumption that the SPV had incurred a liability for business rates, upon its dissolution this liability accordingly passed to the Crown as the owner of the relevant property: see section 65A(2)(b) of the 1988 Act. Any such liability would remain with the Crown until it disclaimed the lease under section 1013, whereupon the lease is deemed by section 1014 not to have vested in the Crown under section 1012.

At the hearing of this appeal the court asked how Scheme A could work if the local authority could, upon becoming aware of the dissolution of the SPV, simply alert the Crown and ask for the lease to be disclaimed, so that the landlord again became entitled to possession of the empty property. The answer, given by counsel for the appellant local authorities, Mr Mathew QC, but Mr Prosser QC for the respondents did not demur, is that the scheme relied upon the local authorities not finding out about the dissolution until long after it had occurred. One of the sample documents provided to this court is a letter from Hurstwood dated 3 September 2014 in response to a business rates bill from Rossendale Borough Council. The letter stated that “[t]he premises are let to Car Wash Solutions Ltd, who are directly responsible for the payment of rates” and enclosed a copy of the lease. Hurstwood did not explain that Car Wash Solutions Ltd (despite its name) had never had any actual or intended business and had, according to the agreed statement of facts, been dissolved on 5 November 2013. If it is shown that, in cases where the dissolution was triggered by an application under section 1003 of the Companies Act, the policy or practice was adopted of deliberately not informing the local authority of the application, it will follow that the scheme in those cases entailed the commission of an aggravated criminal offence contrary to section 1006(4).

More generally, the pre-planned structure of Scheme A had built into it a substantial period, between the grant of the lease and the dissolution of the SPV, when it was intended that the SPV would be accumulating an ever-increasing liability for business rates without having, or there being any prospect of it having, the ability ever to pay the sum due when demanded. Even if an application under section 1003 for voluntary striking off was made at the earliest opportunity, this period would be a minimum of five months: see sections 1003(3) and 1004(1). If no such application was made on behalf of the company, the period would last for as long as it took before the Registrar became aware that the SPV was not carrying on business or in operation and exercised the power under section 1000 to strike its name off the register. In practice, on the information provided to this court, this usually took more than a year and sometimes as long as five years. While the directors of the SPVs could perhaps rely upon an understanding that rent and service charges would never have to be paid, the same could not be said of the rates. It therefore follows that, in the absence of some other facts yet to be revealed, the directors would have been acting unlawfully when they entered into the leases in the first place, it being contrary to their statutory and fiduciary duties voluntarily to commit the company to incur a liability for rates, with no prospect that their company could ever discharge it. Put shortly, it appears from the available facts that the leases under Scheme A could not (if the scheme works) have been entered into without unlawful conduct by those responsible (whether as directors or as shadow directors) for the management of the affairs of the SPVs.

The course taken by the SPVs under Scheme B used by Property Alliance Group Ltd finessed these difficulties. Under this version of the scheme the SPVs were placed in voluntary liquidation within days of the grant of the leases, with the intention that they would be exempted by regulation 4(k) from incurring more than a minimal liability for rates. Members’ voluntary liquidation is an exit route only available for solvent companies and requires a statutory declaration of solvency made by the directors. So it must be assumed, in the absence of any allegation that the declarations of solvency were dishonest, that some arrangement was made to cover that modest amount, if ever demanded.

But a voluntary liquidation conducted in accordance with the applicable rules, and for the purpose of the orderly winding up of the affairs of the SPV, would not achieve the scheme’s objective of obtaining a substantial financial benefit for the landlord of the property and the promoter of the scheme. Liquidators would be appointed. They would quickly realise that the leases were onerous and disclaim them. The landlord of the property would then become entitled to possession again. To delay this, the scheme was arranged so as to secure that the liquidations would continue rather than proceed to a conclusion. In PAG Management Services, following a detailed forensic examination of other examples of the operation of Scheme B, Norris J found that there was no collection, realisation and distribution of assets intended or effected in the voluntary liquidations: see PAG Management Services, para 65. The liquidators were effectively nominal liquidators who dragged their feet and were eventually removed and not immediately replaced. The liquidations were artificially prolonged solely to shelter the leases, which were of no commercial value, in order to allow the empty commercial property to continue to be owned by the SPV. This was found to constitute an abuse of the insolvency legislation and to demonstrate a lack of commercial probity. Norris J said at para 67:

In the present case the judge treated the examples of Scheme B which are the subject of the claim by Wigan Council as in substance the same as those reviewed by Norris J in PAG Management Services. It has not been suggested that he was wrong to do so, nor that there was any relevant difference in how the scheme was operated in each case.

Drawing the elements of the schemes together, on the agreed or assumed facts of the present cases, the following may be said about the scheme leases and the entitlement to possession which the leases conferred on the SPVs nominated as the tenants of the unoccupied properties:

The leases were not shams and created genuine legal rights and obligations.

The leases were entered into, and the rights under them granted, solely for the purpose of avoiding liability for business rates on the part of the defendant and for no other purpose.

The leases were not granted with the intention of allowing the SPV to make any use of the property, or giving the SPV any role in its being brought back into use. To the contrary, it was an inherent part of the schemes that the SPV would have no ability to do so. The SPVs were constituted in such a way that they could do nothing with their rights under the lease, having neither the monetary nor human resources with which to do so, nor any business of which such an activity could form a part. Those rights were accordingly of no value or benefit to the SPV. Furthermore, the SPVs were intended either to go into immediate liquidation (under Scheme B) or to be dissolved after a period of inactivity which was to begin upon the grant of the lease (under Scheme A).

By the same token, the schemes were designed so that the practical ability to decide whether to continue to leave the property unoccupied remained with the defendant landlord. If at any time the defendant found a tenant or acquired a use for the property, it could simply terminate the lease. In Lord Denning MR’s phrase (see para 29 above), it was the defendant who really had control of letting the property.

Although the leases were granted with the object of imposing the legal liability for business rates initially at least upon the SPV in exoneration of the defendant, it was not intended that any business rates would actually be paid by the SPV. Nor was there any possibility that the SPV would in fact pay those rates, once demanded. It had no assets from which to do so. Under Scheme A the plan was that the liability for business rates would just accumulate, unpaid, until the SPV was dissolved, following which the further accruing liability would be incurred by the Crown, until the Crown disclaimed the lease and the liability was extinguished. Under Scheme B the plan was for the SPV to be placed immediately in liquidation to claim the shelter of the regulation 4(k) exemption and stay there for as long as could be engineered. Both schemes were managed so as to prolong the period before the disclaimer of the leases, and therefore the hiatus in payment of rates, for as long as possible.

Each scheme pursuant to which the leases were granted involved as an integral part the misuse of a legal process: namely, the dissolution of a company and the law governing dissolution in the case of Scheme A and the liquidation process and the insolvency legislation in the case of Scheme B. Furthermore, Scheme A (if effective) involved an inevitable breach of statutory and fiduciary duty by the directors of the SPVs in the very acceptance of the leases, and the section 1003 variant is likely to have involved the commission by those directors of criminal offences.

There can be no doubt that the definition of the “owner” of a hereditament in section 65(1) of the 1988 Act as “the person entitled to possession of it” is to be interpreted as denoting in a normal case the person who as a matter of the law of real property has the immediate legal right to actual physical possession of the relevant property. Arden J so held in Brown v City of London Corpn [1996] 1 WLR 1070 as a reason for concluding that the right of a receiver under a debenture to exercise a power to displace the possession of the company which was the tenant of an unoccupied property did not make the receiver the “owner” of the property for the purposes of section 65(1) and thereby liable for business rates. It has not been suggested that that conclusion was wrong. Furthermore, such an interpretation generally accords with the legislative purpose of imposing the liability for business rates on the person who controls whether the property is left unoccupied and on whom the legislation is intended to place an incentive to bring the property back into use for the benefit of the community.

In the unusual circumstances of this case, however, identifying “the person entitled to possession” in section 65(1) of the 1988 Act as the person with the immediate legal right to possession of the property would defeat the purpose of the legislation. As we have explained, the schemes were designed in such a way as to ensure that the SPV to whom a lease was granted had no real or practical control over whether the property was occupied or not and that such control remained at all times with the landlord.

In our view, Parliament cannot sensibly be taken to have intended that “the person entitled to possession” of an unoccupied property on whom the liability for rates is imposed should encompass a company which has no real or practical ability to exercise its legal right to possession and on which that legal right has been conferred for no purpose other than the avoidance of liability for rates. Still less can Parliament rationally be taken to have intended that an entitlement created with the aim of acting unlawfully and abusing procedures provided by company and insolvency law should fall within the statutory description.

In these circumstances we have no difficulty in concluding that, on the agreed and assumed facts, the SPVs to which leases were granted as part of either of the schemes we have described did not thereby become “entitled to possession” of the demised property for the purposes of the 1988 Act. Rather, throughout the term of the lease that person remained the defendant landlord. This does not involve ignoring the leases, in the way that an intermediate element in a circular transaction might be ignored under the Ramsay doctrine. Rather it involves their close examination in their context, and a conclusion that they did not transfer to the SPVs the entitlement to possession required by the Act as the badge of ownership. If the defendants did not thereby transfer their entitlement to possession it necessarily remained, for the purposes of the Act, with them. The Act requires someone to be identified as the owner. That will be the person who, in any tenurial chain, starting with the freeholder and working downwards, has not disposed of the entitlement to possession of the property in question.

We emphasise that this conclusion is not founded on the fact that the defendant’s only motive in granting the lease was to avoid paying business rates, although that was undoubtedly so. If the leases entered into by the defendants had the effect that they were not liable for business rates, their motive for granting the leases is irrelevant. Nor does it illuminate the legal issues to use words such as “artificial” or “contrived” to describe the leases, when it is now accepted that they created genuine legal rights and obligations and were not shams. Our conclusion is based squarely and solely on a purposive interpretation of the relevant statutory provisions and an analysis of the facts in the light of the provisions so construed.

The courts below appear to have received little assistance from counsel for the local authorities as regards the purpose of the rating legislation; and the same was true in this court. It is perhaps unsurprising that in these circumstances the judge and the Court of Appeal did not adopt a purposive approach to interpreting the relevant statutory provisions and considered that the “owner” as defined in section 65(1) of the 1988 Act must invariably and even on the assumed facts of these cases be identified as the person who is entitled to possession of a hereditament as a matter of the law of real property. For the judge, it was enough to constitute the SPV as the “owner” that the scheme lease vested the immediate legal right to possession of the property in the SPV and that none of the subsequent steps in the scheme re-vested that legal right in the defendant until the lease was disclaimed: see para 110 of the judgment. For the Court of Appeal, Henderson LJ said that the very concept of entitlement to possession in section 65(1) of the 1988 Act is an intrinsically legal one because the property is, by definition, unoccupied so the question cannot be answered by reference to the position on the ground and the rights of anyone in actual physical occupation: see paras 71 and 74. As he could see no scope for giving the concept of ownership in this context, as defined, anything other than its normal legal meaning, Henderson LJ concluded that the legislation is not amenable to a wider, purposive interpretation which could allow scope for the Ramsay approach to operate: see para 73. He likened this case to cases such as MacNiven and Barclays Mercantile in which arguments based on the Ramsay principle were rejected.

The description of the concept of entitlement to possession in section 65(1) of the 1988 Act as “an intrinsically legal one” seems to hark back to a distinction drawn in the speech of Lord Hoffmann in MacNiven (at para 58) between “commercial concepts” and “purely legal concepts”. Lord Hoffmann took as an example stamp duty payable on “a conveyance or transfer on sale” and stated:

The chosen example of stamp duty in fact revealed the weakness of this reasoning. The case of Ingram v Inland Revenue Comrs [1986] Ch 585 shows that stamp duty is amenable to a Ramsay approach. In a judgment which repays study, Vinelott J said (at p 602):

Lord Templeman, writing in the Law Quarterly Review about the MacNiven case described the distinction drawn between commercial and legal concepts as reflecting “ingenuity but not principle”: see Lord Templeman “Tax and the taxpayer” (2001) 117 LQR 575, 584. The distinction was considered by the Hong Kong Court of Final Appeal in Arrowtown, another case which involved a scheme to avoid stamp duty. In a judgment which carefully traced the history of the Ramsay doctrine, Lord Millett NPJ said that the supposed dichotomy between commercial and legal concepts had caused great difficulty (para 148). He said that Lord Hoffmann’s speech in MacNiven had had “most unfortunate consequences” and had “led to arid debates in an endeavour to fit the statutory language into one or other conceptual category” (para 150). Lord Millett concluded that the dichotomy should not form part of the jurisprudence in Hong Kong (para 151). The other members of the court agreed. Ribeiro PJ said that he doubted very much that Lord Hoffmann actually intended to lay down any mechanistic test based on a commercial/legal dichotomy for pre-determining whether a particular statutory provision is or is not susceptible to a Ramsay approach; if however that was the intention, he would respectfully decline to follow it.

The opportunity to clarify the position in English law arose in the Barclays Mercantile case. In its joint opinion (at para 38) the appellate committee of the House of Lords explained that Lord Hoffmann’s distinction, while “not an unreasonable generalisation” was not intended to provide a substitute for a close analysis of what the statute means. It certainly did not justify the assumption that an answer can be obtained by classifying all concepts a priori as either commercial or legal. That would be the very negation of purposive construction, as Ribeiro PJ had pointed out in Arrowtown.

We think it important also to note that the conclusions reached in MacNiven and in Barclays Mercantile itself that the statutory provisions relied on in those cases were intended to apply to particular transactions undertaken solely for the purpose of obtaining tax relief were, in each case, the result of a close analysis of the legislation and the relevant facts, following the Ramsay approach, and not of a determination that the Ramsay approach did not apply in the first place.

On the same day as Barclays Mercantile, Lord Nicholls also gave the joint opinion of the appellate committee, similarly constituted, in Scottish Provident. The question in that case depended on what the taxing statute meant by “a contract under which … a qualifying company has any entitlement … to become party to a loan relationship”, the latter term being defined as including a government security. The concept of “entitlement” is of course the concept used in section 65(1) of the 1988 Act, which the Court of Appeal in the present case considered to be an “intrinsically legal one”. In Scottish Provident, however, the House of Lords was not deterred by the “legal” nature of the concept from giving it a practical meaning. It was held that the provision did not apply to a legal entitlement which was intended and expected to be cancelled by an equal and opposite obligation, as there was in these circumstances no entitlement to gilts “in a practical sense” (para 19). As Lord Reed observed in UBS at para 71, the statutory provision was therefore construed as being concerned with a real and practical entitlement.

In a similar way in the present case we consider that the words “entitled to possession” in section 65(1) of the 1988 Act as the badge of ownership triggering liability for business rates are properly construed as being concerned with a real and practical entitlement which carries with it in particular the ability either to occupy the property in question, or to confer a right to its occupation on someone else, and thereby to decide whether or not to bring it back into occupation. The fact that the property is by definition unoccupied means, as Henderson LJ said, that there is no scope for identifying as the owner anyone in actual occupation. But it does not preclude asking the question whether a lease granted as part of a scheme for tax avoidance having the characteristics set out in para 46 above confers an entitlement to possession in the relevant real and practical sense, so as to identify the lessee as the owner for the purposes of the liability for business rates. If it does not do so, in particular because, under the scheme, there is no question of the SPV being able to exercise any of the attributes of a person with an entitlement to possession, and in particular to bring the premises back into occupation by itself or by anyone else, then the lessee under that lease will not be the owner. The landlord, as grantor of the lease, will be the owner, because the landlord will not by the grant of the lease have transferred to the lessee a real entitlement to possession.

As well as adopting the reasoning of the Court of Appeal, Mr Prosser QC in his well-focused submissions for the defendant companies on these appeals argued that to interpret section 65(1) of the 1988 Act as capable of denoting someone other than the person with the immediate legal right to possession of the property would make the test uncertain. It would, he submitted, be inconsistent with the intention reasonably to be attributed to Parliament that ratepayers and rating authorities alike should be able to determine without difficulty who is liable under section 45. There is force in this argument if the test is formulated, as it was by the appellants in their written case, as one of “genuine and real commercial entitlement as owner”. That is an amorphous formulation. But a recognition that section 65(1) is speaking of an entitlement to possession which vests in the person concerned a real and practical ability either to occupy the property or to put someone else into occupation of it, is a purposive interpretation which achieves some coherence between the language of the statute and its purpose in identifying the “owner” of an unoccupied non-domestic property as the person who is liable for business rates.

It may be that other factual situations may demonstrate that this test needs some further adjustment. For example the letting of unoccupied business property by a parent company to a wholly owned and controlled subsidiary would not of itself cause the subsidiary to fail to satisfy the ownership test merely because the management of the affairs of the subsidiary (including whether to bring the premises back into occupation) rested with the parent’s board. We would, however, reject the criticism that the test is insufficiently certain. In any ordinary case the test will easily be satisfied by identifying the person who is entitled to possession as matter of the law of real property. The fact that the law of real property may not prove a reliable guide in an unusual case of the present kind is not in our view an objection to our preferred interpretation. The value of legal certainty does not extend to construing legislation in a way which will guarantee the effectiveness of transactions undertaken solely to avoid the liability which the legislation seeks to impose.

On what we conceive to be the proper interpretation of the definition of “owner” in section 65(1) of the 1988 Act, none of the SPVs granted rights by the leases under review in these proceedings answers to that description on the agreed and assumed facts. Rather, on those facts “the person entitled to possession” of the property remained at all material times the defendant landlord. Accordingly, there is a triable issue whether the defendants remained liable for business rates throughout the duration of the leases as claimed by the local authorities and those claims should not be struck out.

The second pleaded ground of the local authorities’ claims is their contention that the use of the SPVs to avoid the liability for business rates which the defendants would otherwise have is an abuse of the SPVs’ separate corporate personality that warrants “piercing the corporate veil”. This contention is based on what Lord Sumption in the leading case of Prest v Petrodel Resources Ltd [2013] UKSC 34; [2013] 2 AC 415 called the “evasion principle”. It is a contention for which the necessary predicate is that (contrary to our view as stated above) on the assumed facts the grant of the leases did make the SPVs the owners of the relevant premises for the purpose of liability for business rates.

Talk of “piercing the corporate veil” is a metaphor that is liable to obscure more than it illuminates. As Lord Sumption said at the start of his discussion of the topic in Prest (at para 16):

In Prest Lord Sumption proposed that two distinct principles underlie the cases apparently concerned with piercing the corporate veil. The first, which he called the “concealment principle”, involves the interposition of a company or perhaps several companies to conceal the true nature of an arrangement. In these cases, the court is merely looking behind the company to discover the facts which the corporate structure is concealing and applying the ordinary legal or equitable principles to those facts. This concealment principle “is legally banal and does not involve piercing the corporate veil at all” (para 28). The second principle, which Lord Sumption dubbed the “evasion principle”, was said to comprise “a small residual category of cases where the abuse of the corporate veil to evade or frustrate the law can be addressed only by disregarding the legal personality of the company” (para 35). This principle was said to apply:

It is on this principle which the local authorities seek to rely in these proceedings. They argue that, on the assumed facts, the defendants interposed companies under their control solely in order to avoid a liability for business rates that would in each case, but for the company’s separate legal personality, be a liability of the defendant. Counsel for the local authorities submitted that this was an abuse of the separate legal personality of the SPV which justifies the court in piercing the corporate veil to deprive the defendant of the advantage that it would otherwise have obtained by the SPV’s separate legal personality.

The first problem with this argument is that, even if the proposition of law relied on is sound, the defendants are not shareholders of the SPVs. On the agreed facts, each of the SPVs had a single shareholder, who was in each case an individual who was also the company’s director. Nonetheless it is pleaded that the SPVs were “acquired” by the defendants, and we will assume that, as ultimate beneficial owners, they might be shown at trial to have been in control of them, even though this conflicts with the finding in the PAG Management Services case that the SPVs there under review were controlled by the promoter of the scheme.

More fundamentally, it is not clear to us that the “evasion principle” could ever enable the liability of a company to be extended to its shareholder or to some other person who controls it. In her judgment in Prest, at para 92, Baroness Hale of Richmond (with whom Lord Wilson agreed) drew attention to a distinction, not discussed by Lord Sumption, between two different situations which may be said to involve “piercing the corporate veil”. One is where the separate legal personality of the company is disregarded in order to obtain a remedy against the owner or the controller of the company in respect of a liability which would otherwise be that of the company alone. The other is “the converse case, where it is sought to convert the personal liability of the owner or controller into a liability of the company”. See also Edwin C Mujih, “Piercing the corporate veil: where is the reverse gear?” (2017) 133 LQR 322; and Christian Witting, “Piercing the corporate veil” in Day and Worthington (eds), Challenging Private Law: Lord Sumption on the Supreme Court (2020), ch 17.

The two authorities identified by Lord Sumption as cases in which the evasion principle was properly applied were both cases of the latter kind. In Gilford Motor Co Ltd v Horne [1933] Ch 935 Mr Horne, after leaving his employment with the plaintiff, formed a new company to compete with his previous employer in breach of an agreement by him not to do so. The Court of Appeal held that the new company was a “mere channel” used by Mr Horne to avoid his contractual obligation and granted an injunction against both him and the company to prevent them from competing with the plaintiff. Jones v Lipman [1962] 1 WLR 832 was a case of a similar kind. The defendant, Mr Lipman, agreed to sell a property to Mr Jones. However, before completion he changed his mind and transferred the property to a company which he had formed for the sole purpose of seeking to defeat Mr Jones’ right to specific performance. The judge made an order for specific performance not only against Mr Lipman but also against his company. In the case of the company, on Lord Sumption’s analysis, this was justified by the evasion principle. In both cases there was a straightforward contractual liability on the owner of the company. The question was whether the company itself could be made liable.

Prest was also a case of this type, albeit that it was held that the evasion principle was not engaged on the facts. The argument was that Mr Prest had put properties beyond the reach of his former wife by transferring the properties to companies controlled by him and that, on the wife’s claim for ancillary relief, an order should be made against the companies requiring them to transfer the properties to the wife.

Whether the evasion principle is needed or provides the best justification of cases such as Gilford Motor and Jones is itself open to debate. In his judgment in Prest Lord Neuberger of Abbotsbury said that the decision in the Gilford Motor case that an injunction should be granted against the company was “amply justified” on the basis that the company was Horne’s agent for the purpose of carrying on the business (para 71); and that in Jones it was unnecessary to invoke the doctrine, as an order for specific performance against Lipman would have been sufficient by requiring him to procure that the company conveyed the property in question to the plaintiff (para 73). Baroness Hale expressed the view that in such cases it is usually more appropriate to rely on the concepts of agency and of the “directing mind”. Lord Walker of Gestingthorpe questioned whether “piercing the corporate veil” is a coherent principle or rule of law at all, as opposed to simply a label used to describe the disparate occasions on which some rule of law produces apparent exceptions to the principle of the separate juristic personality of a corporate body (para 106). Although this is not the occasion for reaching any final view, we are inclined to share Lord Walker’s doubts.

Even if there is an “evasion principle” which may in “a small residual category of cases” (per Lord Sumption) justify holding a company liable for breach of an obligation owed by its controlling shareholder, we are not ourselves convinced that there is any real scope for applying such a principle in the opposite direction so as hold a person who owns or controls a company liable for breach of an obligation which has only ever been undertaken by the company itself. At para 34 of his judgment in Prest, Lord Sumption said:

That analysis, with which we respectfully agree, seems to us to leave very little room for reliance on the “evasion principle” to impose upon the controller of a company a fresh liability incurred by the company as distinct from its controller. But whether that is so or not, we think it clear that there is no scope for such a principle to operate in the present case. Liability for rates accrues from day to day. If the leases were effective to transfer the ownership of the demised properties for the purpose of the 1988 Act from the defendants to the SPVs (which is the premise on which the attempt to “pierce the corporate veil” becomes relevant), then in each case, from the date of the lease, the only person liable for business rates incurred thereafter was the SPV. Furthermore the interposition of the SPV had no effect at all on the liability of the landlord for rates up to the date of grant of the lease. Applying Lord Sumption’s reasoning, it is not an abuse of the separate legal personality of the SPV to cause the liability for business rates to be incurred by the SPV by granting it a lease, nor to rely on the fact (if it is a fact) that the liability was not the defendant’s because it was the company’s. Nor can the evasion principle properly be invoked so as to create a liability on the part of the defendant that would not otherwise exist.

At the heart of the evasion principle, as explained by Lord Sumption, is the necessary averment that the interposition of the company is, on the facts, an abuse of separate corporate personality. But if the rating legislation permits a landowner to transfer the on-going liability for rates by granting a lease of the property to a wholly owned subsidiary created for the purpose, on terms which make the subsidiary the owner of it, then it is not per se an abuse of corporate personality to do so. The abuse in the present case lies in the way in which the SPV’s liability for rates is then sought to be dealt with, by the abusive processes by which the SPV is either dissolved or put into liquidation. The law provides comprehensive remedies for abusive behaviour of that kind, which do not require the piercing of any corporate veil.

In these circumstances the attempt by the local authorities in this case to invoke the “evasion principle”, or any principled development of it, is in our view wholly misplaced. The submissions made by counsel for the local authorities that, when a scheme lease was granted, the defendant was under an existing liability for business rates which it evaded by interposing a company are untenable and amounted to vain efforts to force a square peg into a round hole. The plain fact is that the lease did not avoid, nor even appear or seek to avoid, the liability for business rates which the defendant had incurred up to the date when the lease was granted: that liability on any view remained with the defendant. All that the lease did (if, contrary to our view on the assumed facts, effective for this purpose) was to transfer ownership of the property to a tenant who accordingly incurred the new liability for business rates which accrued for each day that the lease continued thereafter. It made no difference in terms of the liability for business rates whether the tenant was a company or an individual or whether, in circumstances where it was a company, the company was owned or controlled by the defendant.

It follows that the Court of Appeal was, in our opinion, correct to decide that there is no principle which could justify piercing the corporate veil on the assumed facts of this case.

For these reasons, we would allow the appeal and set aside the order of the judge in so far as he struck out the claims that, on the proper interpretation of sections 45 and 65(1) the 1988 Act, the leases were ineffective to make the SPVs the owners of the relevant properties, with the result that the defendants remained liable for business rates. We would leave undisturbed the decision of the Court of Appeal in so far as it struck out the claims based on “piercing the corporate veil”.