Farnborough Airport Properties Company & Anor v Revenue And Customs

In the normal way, group relief from UK corporation tax may be claimed and surrendered between sister companies in a group if they are both “75% subsidiaries” (as defined) of a third company, and if the numerous other requirements for the grant of such relief are satisfied. Group relief is also available between two companies if one is the 75% subsidiary of the other, but in this case we are concerned only with “horizontal” claims between two fellow subsidiaries. The relevant legislation was most recently restated with minor changes, pursuant to the Tax Rewrite Project, in Part 5 of the Corporation Tax Act 2010 (“CTA 2010”), which extends from sections 97 to 188.

According to the basic definition of “75% subsidiary” in section 1154 of CTA 2010, which is applied for the purposes of group relief by section 151(1), a body corporate

Section 151(4) contains further provisions which, for the purposes of group relief, reinforce the requirement of beneficial ownership by A of B’s ordinary share capital. In short, the parent company (A) must also be beneficially entitled to receive at least 75% of any profits available for distribution to equity holders of the subsidiary (B), and at least 75% of any assets of B available for distribution to such equity holders on a winding up, in each case in accordance with detailed provisions contained in Chapter 6 of CTA 2010 (sections 157 to 182). In other words, the beneficial interest must reflect a corresponding economic interest in the subsidiary, both when its profits are distributed and when it is wound up.

In addition to the requirements of section 151(4), companies A and B are also disqualified from membership of the same group if arrangements of a specified type are in place during the accounting period for which one of them has amounts available for surrender: see section 154. For present purposes, the relevant parts of section 154 read as follows:

“154.

Arrangements for transfer of member of group of companies etc

(1)

This section applies if, apart from this section, one company (“the first company”) and another company (“the second company”) would be members of the same group of companies.

(2)

For the purposes of this Part the companies are not members of the same group of companies if –

(a)

one of the companies has surrenderable amounts for an accounting period (“the current period”), and (b) arrangements within subsection (3) are in place.

(3)

Arrangements are within this subsection if they have any of the following effects…

…

Effect 2

At some time during or after the current period a person (other than the first or second company) has or could obtain, or persons together (other than those companies) have or could obtain, control of the first company but not of the second company.

…”

It can be seen, therefore, that “arrangements” will disqualify companies A and B from being members of the same group if (a) the arrangements are “in place” during the surrendering company’s accounting period (“the current period”), and (b) the arrangements have the “effect” that, at some time during or after the current period, a person other than A or B has or could obtain control of one, but not of both, of those companies (or, alternatively, persons together other than A or B have or could obtain such control). Thus, for example, this condition would clearly be satisfied if the effect of the arrangements is to transfer control of either A or B to a third party, but it would also be satisfied if their effect is merely to leave the 75% parent company of A and B with control of one, but not the other, of them.

To discover the meaning of “control” in this context, it is necessary to turn to section 1124(2) of CTA 2010, which contains a general definition applicable to any provisions of the Corporation Tax Acts which apply it, or to which it is applied: see subsection (1). In the case of section 154, the definition is applied by virtue of sections 1118(2), (5) and 1119. The definition in subsection (2) states that:

“In relation to a body corporate (“company A”), “control” means the power of a person (“P”) to secure –

(a)

by means of the holding of shares or the possession of voting power in relation to that or any other body corporate, or

(b)

as a result of any powers conferred by the articles of association or other document regulating that or any other body corporate,

that the affairs of company A are conducted in accordance with

P’s wishes.”

The definition is thus concerned with the power of a person to secure that the affairs of the subject company are conducted in accordance with that person’s wishes. Such power must be exercisable in one or other of the specified ways, the first of which depends on the holding of shares or the possession of voting power, and the second of which depends on the exercise of any powers conferred by the articles of association or by any “other document” which regulates that or any other company. In very broad terms, it would seem that the former limb of the definition is looking at the ability to exert control at shareholder level, for example by taking steps to appoint or remove directors, while the second limb focuses on the exercise of powers conferred by the articles or any other document which regulates the company’s affairs, typically at board level. In either case, however, the person exerting the control must thereby be able to achieve the result that the company’s affairs are conducted in accordance with his wishes.

As to the meaning of “arrangements”, section 156(2) provides that:

““Arrangements” –

(a)

means arrangements of any kind (whether or not in writing), but

(b)

does not include a power of a Minister of the Crown, the Scottish Ministers or a Northern Ireland department to give directions to a statutory body as to the disposal of assets belonging to the body or to a subsidiary of the body.”

In the light of this legislative background, I can now identify the key issue which arises on this second appeal by the taxpayer companies, Farnborough Airport Properties Company and Farnborough Properties Company (which I will call “FA” and “FP” respectively, and together “Farnborough”) against the refusal by HMRC (in closure notices issued on 24 December 2014, and upheld on review on 6 March 2015) of their claims to group relief surrendered by their fellow 75% subsidiary, Piccadilly Hotels 2 Limited (“PH2L”), in respect of PH2L’s accounting periods ending on 31 January 2012 and 2013. The relevant losses of PH2L for which relief was claimed amounted to £7,114,177 in its accounting period to 31 January 2012, and £3,513,532 in the following period, totalling £10,627,709. Both amounts were claimed by FA and FP in relation to their accounting periods ended 31 May 2012.

It is common ground that, at all material times before 27 June 2011, FA, FP and

The problem arises because, on 27 June 2011, PH2L was placed into receivership by Bank of Scotland PLC, in exercise of its rights as Security Trustee under a Deed of

HMRC’s argument, reduced to its simplest terms, is that the appointment of the receivers, and the conduct of the receivership in accordance with the terms of the Debenture, constituted arrangements within the broad definition in section 156(2), that those arrangements were in place during the current periods of PH2L for which the relief was claimed, and that the effect of those arrangements was to deprive Kelucia of its previous control of PH2L because it could no longer secure that the latter company’s affairs were conducted in accordance with its wishes. That is so, submit HMRC, whether or not the right analysis is that after the appointment of the receivers no person had control (in the statutory sense) of PH2L, or the receivers had such control, or control was vested in Kelucia and the receivers jointly. The crucial point is that Kelucia alone no longer had control of PH2L, whereas it is common ground that it continued to have control of FA and FP.

This analysis was contested by the taxpayer companies, which argued that the receivership did not constitute arrangements within the meaning of section 156(2), and that in any event the receivership did not trigger Effect 2, because Kelucia continued to control PH2L at shareholder level. They also argued that control in the statutory sense could not in any event have passed to the receivers, because they held no shares or voting power in relation to PH2L, nor did they have any powers conferred by the articles of association of PH2L, or by any other document regulating PH2L or any other company, which empowered them to conduct the affairs of PH2L as they wished. Their powers were derived solely from the Debenture, which was not an “other document” of the kind contemplated by section 1124(2)(b).

Farnborough’s appeal to the FTT was heard by Judge Christopher McNall in January 2016. The parties were represented, as they have been throughout, by Mr Philip Ridgway of counsel for the appellants, and Mr Jonathan Bremner QC for HMRC. In his decision released on 17 June 2016 (“the FTT Decision”), Judge McNall dismissed the appeals. The neutral citation of the FTT Decision is [2016] UKFTT 0431 (TC), and it is reported at [2016] SFTD 826.

Farnborough then appealed to the Upper Tribunal (Judge Sinfield and Judge Poole), which heard the appeals in June 2017. By their decision released on 4 October 2017 (“the UT Decision”), the appeals were dismissed: see [2017] UKUT 0394 (TCC), [2017] STC 2293.

Farnborough’s second appeal to this court is brought with permission granted by Floyd LJ. In his short written reasons for granting permission, Floyd LJ noted that the appeal “raises fairly short points of construction”. Although the legislation which we

I have already summarised the basic group structure (which does not need to be understood in any more detail for the purposes of this appeal) and the procedural history, as recorded in the brief agreed statement of facts. No written or oral evidence was adduced before the FTT, but there was a bundle of documents which included the Debenture and the notice of appointment of the receivers in form LQ01 filed at Companies House on 27 June 2011.

On the basis of this scanty material, the FTT made the following unchallenged further findings of fact, at [5] to [12] of the FTT Decision:

The statutory Notice of Appointment (Form LQ01) records

The Bank of Scotland’s rights flowed ultimately from its appointment as Security Trustee under a Deed of Debenture dated 10 October 2006. By clause 3.1 of that Deed of Debenture, PH2L, as one of the Original Chargors, granted security in favour of the Security Trustee by way of (i) a first legal mortgage against all its Property (meaning the Real Property from time to time owned by the Chargor or in which the Chargor has an interest together with all proceeds of sale deriving from any such Real Property, the benefit of all covenants given in respect of such Real Property and any monies paid or payable in respect of such covenants), (ii) a first fixed charge on all the land and buildings, (iii) a first fixed charge on plant and machinery, other chattels, investments, insurances, book debts, bank balances, intellectual property, authorisations, goodwill and uncalled capital.

By way of Clause 3.5 of that Deed of Debenture, PH2L granted a first floating charge of “all their assets and undertakings whatsoever and wheresoever both present and future not effectively charged by way of legal mortgage or fixed charge pursuant to the provisions of clause 3.1 (fixed charges) or effectively assigned by way of security pursuant to clause 3.2 (assignment by way of security), but extending over all its property, assets, rights and revenue as are situated in Scotland or governed by Scottish law”.

Clause 3.7.2 of the Deed provides for the automatic crystallisation of the rights under the floating charge if a receiver is appointed in respect of PH2L, thereby converting the rights under it into a fixed charge.

Clauses 9.6 and 9.7 of that Deed of Debenture provide for Enforcement of Security as follows:

The “powers” which are referred to in Clause 9.6.2 are to be found in Schedule 1 of the Insolvency Act 1986 (“Powers of Administrator or Administrative Receiver”) and include (except insofar as they are inconsistent with any of the provisions of the Deed of Debenture) the following powers:

“12.

Power to do all such things (including the carrying out of works) as may be necessary for the realisation of the property of the company.

13.

Power to carry on the business of the company”

Those powers are not inconsistent with the Deed of Debenture.

Schedule 12 of the Deed of Debenture sets out the Receiver’s Specific Powers. The Receiver will have “full power and authority in relation to the chargor… it is appointed to act as agent for” including the following power:

“2.

CARRY ON BUSINESS

generally to manage the Secured Assets and to manage or carry on, reconstruct, amalgamate, diversify or concur in the carrying on the business of that Chargor or any part of it as he may think fit””

The Debenture was governed by English law: see clause 17.1. Although PH2L was a

Section 154 of CTA 2010 is the statutory successor of legislation which was first enacted as section 29 of the Finance Act 1973. That section (with the marginal note “Group relief: effect of arrangements for transfer of company to another group, etc”) was framed in materially similar terms, as follows:

“(1)

If, apart from this section, two companies (in this subsection referred to as “the first company” and “the second company”) would be treated as members of the same group of companies and -

(a)

in an accounting period which ends on or after 6^th March 1973, one of the two companies has trading losses or other amounts eligible for relief from corporation tax which it would, apart from this section, be entitled to surrender as mentioned in subsection (1) of section 258 of the Taxes Act, and

(b)

arrangements are in existence by virtue of which, at some time during or after the expiry of that accounting period, -

(i)

the first company or any successor of it could cease to be a member of the same group of companies as the second company and could become a member of the same group of companies as a third company, or

(ii)

any person has or could obtain, or any persons together have or could obtain, control of the first company but not of the second, or

(iii)…

then, for the purposes of the enactments relating to group relief, the first company shall be treated on and after 6^th March 1973 as not being a member of the same group of companies as the second company.

…

(5)

In subsections (1) and (2) above –

…

“control” has the meaning assigned to it by section 534 of the Taxes Act [i.e. section 534 of the Income and Corporation

Taxes Act 1970 (“ICTA 1970”), which was in materially similar terms to section 1124 of CTA 2012]”

Section 29 of the Finance Act 1973 formed part of a package of measures introduced, following consultation, to counter certain perceived abuses of the group relief legislation. The other measures, mainly contained in section 28 of the 1973 Act, included the further “economic” requirements for being a 75% or 90% subsidiary which are now contained in section 151(4) of CTA 2010. Introducing these measures in his Budget Statement on 6March 1973, the Chancellor of the Exchequer gave this explanation of them:

“The Finance Bill will provide two significant changes affecting groups of companies. These are the outcome of consultations between the Inland Revenue and representatives of industry.

I was urged last year by a number of important groups of companies, and in particular those with large overseas interests, to make more flexible the provisions under which companies may surrender advanced corporation tax to their subsidiaries. The problem was that I could not contemplate any relaxation in the treatment of groups unless I could also take action to counter certain artificial manipulations of the group relief provisions involving in effect the sale of capital and other allowances, at a discount. This abuse was spreading rapidly and there was reason to fear a possible loss of tax of the order of £100 million a year.

We have now found a solution to deal with this problem. This is another instance of the value of the kind of consultation, at professional level, which we have tried constantly to foster in the programme of taxation reform.”

The “main outlines of the Government’s proposals” were then set out in an Annex, which was said to have “no statutory force” and to be “published solely for information”. Paragraphs 8 to 10 of the Annex described the proposals for what became section 29 of the 1973 Act, but said nothing about its possible application in cases where the company concerned remained under the same ownership, or where it went into receivership. Nor was there any mention of what is now Effect 2 in section 154(3) of CTA 2010.

A good example of the kind of abuse at which section 73 was explicitly aimed is provided by the leading case on the application of the section, which went to the House of Lords: Pilkington Bros. Ltd. v Inland Revenue Commissioners [1982] 1

After explaining that group relief was first introduced by sections 258 and following of ICTA 1970, and referring to the “important new provisions… designed to limit the availability of group relief” introduced by the Finance Act 1973, Lord Bridge described the issue in the appeal in the following terms, at 144B-H:

“The matter in issue in the appeal arises out of a series of agreements concluded between Pilkingtons and Manchester Liners Ltd (“Manchester Liners”). Pilkingtons are the wellknown manufacturers of glass; Manchester Liners are shipowners. Manchester Liners wanted to acquire a new container ship which was to cost more than £11m. This would give rise to a claim for capital allowances exceeding any taxable profits of Manchester Liners, or any company in the same group as Manchester Liners, against which the claim could be set. Pilkingtons were willing to purchase the claim at a discount.

Pilkingtons had two wholly owned subsidiaries, Hello TV Ltd (“HTV”) and Villamoor Ltd (“Villamoor”). Manchester Liners had a wholly subsidiary, Golden Cross. Without examining the details it is sufficient to say that agreements were concluded between these companies which produced the following end result as regards the structure of the companies. Pilkingtons retained its holding of all the shares in HTV. Pilkingtons retained 50 per cent of the shares in Villamoor; the remaining 50 per cent was acquired by Manchester Liners. HTV and Villamoor each acquired 50 per cent of the shares in Golden Cross. It is not disputed that the effect of this distribution of shares was to constitute Golden Cross a “75 per cent subsidiary” of Pilkingtons both within the original definition in section 532 of the Taxes Act and in accordance with provisions qualifying that definition in section 28(2) of the Act of 1973. Certain necessary alterations were effected to the articles of association of HTV, Villamoor and Golden Cross. Again, it is not disputed that the effect of the provisions of the articles of association of Villamoor and Golden Cross, as applied to the equally divided shareholdings in both those companies, was to render each a fully “dead-locked company, in that Villamoor was not under the control of either Pilkingtons or Manchester Liners and Golden Cross was not under the control of either HTV or Villamoor. It follows, of course, that Golden Cross was not under the control of Pilkingtons.

Golden Cross concluded an agreement with shipbuilders for the purchase of the container ship and agreed to surrender its claim to capital allowances arising from that purchase, not exceeding £13m, to Pilkingtons, who were, in turn, to pay to Manchester Liners 87½ per cent of the corporation tax which they would, it was hoped, thus be enabled to save by way of group relief. In short, Pilkingtons purchased the claim to capital allowances from Golden Cross at a discount of 12½ per cent.”

On these facts, the majority in the House of Lords held that Pilkington and Golden Cross were not members of the same group, because arrangements were in existence by virtue of which the shareholders of Pilkington together had control of Pilkington, but not (because of the “deadlocked” holdings of the shares in Villamoor and Golden Cross) of Golden Cross. Accordingly, the conditions of what is now Effect 2 were satisfied.

The main reasoning which led Lord Bridge to this conclusion is set out in a passage which I need to quote in full, from 146H to 148A:

“First, the definition of “arrangements” as meaning arrangements of any kind predisposes me against imposing any limitation on the ordinary meaning of the word unless forcibly driven to do so by the context.

Secondly, I turn to consider in detail the language of the critical sentence which has to be construed, sc. “… arrangements are in existence by virtue of which… any person has… or any persons together have… control of the first company but not of the second.” “Arrangements” is in the plural, not the singular, and I can see no justification for applying to the plural the concept of a combination for a singular purpose derived from the dictionary definition of a singular “arrangement.” The arrangements to be considered are such as “are in existence”; it matters not when they came into existence. When the draftsman of the Act of 1973 wishes to distinguish arrangements which “come into existence” after a given date from those which “are in existence” on that date, he does so: see Schedule 12, paragraph 10(2). The next critical phrase is “by virtue of which.” This directs attention to the effect of the arrangements not to their purpose. But, to my mind, the consideration of overriding significance is that the whole sentence is concerned with those arrangements which determine the control of both the companies whose entitlement to be treated as members of the same group is in issue. To construe “arrangements” as excluding, notwithstanding the definition of “control” in section 534 of the Taxes Act, those arrangements which regulate the conduct of the affairs of either of the companies in accordance with the wishes of its controlling shareholders seems to me simply to negate the plain meaning of the statutory language. I should add, under this head, that I can attach no significance, in applying the language in section 29(1)(b)(ii) to the circumstances of the case, to the fact that the shareholders in Pilkingtons were not “parties” to the scheme devised to enable Pilkingtons to acquire a claim to group relief, in the sense that they, as shareholders, as distinct from the board of directors carrying on the management of the company, participated in the planning and negotiation of the scheme. There is nothing in the statutory language to indicate a requirement that they should be.

Thirdly, if one seeks to discern a legislative purpose underlying section 29(1)(b)(ii) one is driven, I think, to conclude that this provision was intended to introduce a requirement, as a qualification for entitlement to group relief, in addition to those introduced by section 28, that the two companies, (in the original terminology of the Taxes Act “the surrendering company” and “the claimant company”) claiming membership of the same group of companies should be under the same control. This requirement is, I would assume, introduced in section 29(1) rather than in section 28 because the draftsman found it convenient to include in a single provision both the original requirement of unified control and a requirement that there should be no arrangements in existence during the relevant accounting period making provision for a future severance of control and it is in section 29(1) that he deals with other cases where the benefit of group relief will be lost by reason of existing arrangements providing, in one way or another, for future severance of the group. The narrow construction of “arrangements” adopted by the learned judge would have what to my mind would be the startling consequence that the only kind of scheme setting up a group of companies, where none existed before, for the purpose of obtaining group relief, which would be liable to disqualification under section 29(1)(b)(ii), would be a scheme specifically designed to embody the very disqualifying features at which the provision is directed. Such a construction must, it seems to me, effectively deprive the provision of any practical operation as limiting the circumstances in which group relief is to be available. It was presumably intended to have such a practical operation and I can see no room here for applying any restrictive interpretation so as to cut down the plain meaning of the statutory language to make it accord with some supposedly legislative intent.”

At this point, it is worth noting a significant difference between the group arrangements which were brought into existence in Pilkington and the group structure in the present case. Because Pilkington was itself the claimant company for group relief, it followed that the test of severance of control could only be satisfied by looking at those who controlled (or ceased to control) both Pilkington and Golden Cross. It was not sufficient to look at Pilkington alone, because a company obviously cannot control itself in the relevant sense. Thus the reasoning of the majority depended on the proposition that the shareholders of Pilkington, which was a listed public company with a large and fluctuating body of individual shareholders, together had “control” of Pilkington in the statutory sense. Lord Bridge clearly had no difficulty in accepting that underlying shareholder control of this nature would suffice, even though the shareholders played no part in the relevant arrangements. Lord Wilberforce, however, found it “at first sight strange that the critical persons to consider should be the shareholders of Pilkingtons at all” (see 140C), and added at 140G:

“The shareholders of Pilkingtons had no part in these arrangements, they were not consulted, they did not agree to them. Can we then add in, as arrangements, those made (we do not know when, but probably they were spread over years as each shareholder acquired his shares) by which these shareholders were able, ultimately, to control Pilkingtons, arrangements which had nothing to do with the creation of the group structure? I cannot think so.”

Two points of potential importance emerge. The first is that no similar problem arises in the present case, because both the surrendering company (PH2L) and the claimant companies (FA and FP) are agreed to have been 75% subsidiaries of a third company, Kelucia, so there is no need to look beyond Kelucia to its own shareholders. Secondly, all of the members of the Appellate Committee in Pilkington clearly accepted that ultimate control at shareholder level was sufficient to satisfy the statutory definition, even if the shareholders were a large and fluctuating body. The ground upon which Lord Wilberforce and Lord Russell dissented was, rather, that, even giving “arrangements” the widest possible meaning, they could not properly be regarded as extending to the internal structure of Pilkington itself: see 140H-141C and 143C-F.

The only other case to reach the High Court on the relevant provisions in section 29 of the 1973 Act again involved an artificial scheme whereby a profitable trading group (in this case Tesco) entered into complex arrangements designed to acquire by way of group relief capital allowances obtained by a previously unconnected company (Rebron) on expenditure incurred on the acquisition of a ship, in return for payment of

Having reached this conclusion, Walton J observed, at 910c, that so far as he was concerned it was “strictly unnecessary to decide whether the “control” spoken of in s 534 is control at board or company meeting level: whichever it is, I think that the shareholders of Holdings did not have it so far as Rebron is concerned.” Nevertheless, Walton J proceeded to deal with the point, albeit obiter, in case he was wrong about the position at board level, and in deference to the careful arguments which had been addressed to him on the point. After referring to the definition of “control” in section 534 of ICTA 1970, the judge set out article 80 of Table A, in its then current form, which applied to Rebron and provided that:

“The business of the company shall be managed by the directors, who may … exercise all such powers of the company as are not, by the [Companies Act 1948] or by these regulations, required to be exercised by the company in general meeting…”

Walton J continued, at 910j: “The word used there is “business”: the word used in s 534 is “affairs”. Is there any difference?

Those observations would not bind us, even if they were not obiter, but in general terms I respectfully agree with them. As I have already indicated, it seems to me that the alternative tests of control in what is now section 1124(2) of CTA 2010 do distinguish between underlying control of a basic and permanent nature at shareholder level, on the one hand, and control which may be of a more temporary nature, particularly at board level, as a result of powers conferred by the articles of association or other similar documents, on the other hand.

The FTT Decision

Judge McNall began his discussion of the substantive issues by considering the purpose of section 154 of CTA 2010. He correctly pointed out that the legislative purpose of the predecessor provision in section 29(1)(b)(ii) of the Finance Act 1973 was authoritatively stated by Lord Bridge in Pilkington as the introduction of a new requirement that the two companies claiming membership of the same group “should be under the same control”: see [1982] 1 WLR 136, 147 F-G. Given that clear statement, the judge found no real assistance in either the marginal notes to sections 29 and 154 or the Explanatory Notes to CTA 2010. Furthermore, the judge was unable to discern any limitation on the scope of section 154 so that it applied only to cases of deliberate tax avoidance, or where the arrangements in question were artificially contrived to secure group relief. Referring to the guidance given by Lewison LJ in Pollen Estate Trustee Co Ltd v Revenue & Customs Commissioners [2013] EWCA Civ 753, [2013] 1 WLR 3785, at [24], the judge said at [47] of the FTT Decision:

“Taking the above into account, it seems to me that the clear purpose of section 154, read purposively, is simply to make group relief unavailable between companies which are not under the same control. Applying the guidance in Pollen, then the nature of the transaction to which section 154(3) and Effect 2 was intended to apply was simply a transaction whereby the control of two companies came to be in separate hands, irrespective of whether that motive or purpose was (put neutrally) a salutary one or not.”

I would endorse that conclusion, although I agree with the Upper Tribunal (see below) that it would be preferable to substitute the word “arrangements” for “transaction”.

The judge then turned to whether the circumstances of the present case amounted to “arrangements”. Relying on the reasoning of Lord Bridge in Pilkington, and a number of other matters discussed at [50] to [55], he concluded at [56] “that the appointment of Receivers pursuant to the Deed of Debenture constituted “arrangements” for the purposes of s154.”

On the issue of “control”, the judge reminded himself that the question had to be answered by reference to the statutory definition in section 1124(2) of CTA 2010. In the light of Pilkington, the correct approach was simply to ask whether PH2L, FA and FP “are under the same control”: see [58]. Since it was common ground that the receivers did not control FA and FP, the judge said at [59] that:

“the question of whether there is still common control of the Appellants and PH2L is most intelligibly addressed by assessing whether someone else (that is, the Receivers) are in control of PH2L. If they are in control, then the question of whether Effect 2 is engaged is answered.”

The judge then referred to certain dicta concerning the effect of the appointment by debenture-holders of a receiver and manager in Moss Steamship Company Ltd vWhinney [1912] AC 254. In that case, by an order made in a debenture-holders’ action, the court appointed Mr Whinney receiver and manager of a brewery company. The circumstances were far removed from those of the present case, and gave rise to differences of judicial opinion in both the Court of Appeal and the House of Lords. However, the three members of the majority in the House of Lords made observations about the effect of Mr Whinney’s appointment which Judge McNall set out at [61] and [62], before saying in [63] that he regarded them “as useful and instructive analyses”. It is enough for me to quote what the Earl of Halsbury said at 260:

“A great many joint stock companies obtain their capital, or a considerable part of it, by the issue of debentures, and one form of securing debenture-holders in their rights is a well-known form of application to the Court, which practically removes the conduct and guidance of the undertaking from the directors appointed by the company and places it in the hands of a manager and receiver, who thereupon absolutely supersedes the company itself, which becomes incapable of making any contract on its own behalf or exercising any control over any part of its property or assets.”

See too the speeches of Lord Loreburn L.C. at 257, and Lord Atkinson at 263 (who said that the appointment “entirely supersedes the company in the conduct of its business” and that the company’s powers in relation to its business and assets “are entirely in abeyance”).

The FTT also found assistance in Commissioners of Inland Revenue v Lithgows, Ltd (1960) 39 TC 270, where the First Division of the Court of Session considered the meaning of materially identical language in the definition of “control” in section 333(1) of the Income Tax Act 1952. In that case, Lord Guthrie said at 278:

“In order that a person may have “control” he must be in a position to secure that the affairs of the company are conducted according to his wishes. That phrase means that ability to achieve an isolated result, the power to carry a particular resolution, is insufficient to establish control in the statutory sense; and that what is required is power to secure the continuing conduct of the company’s affairs in accordance with the will of that person.”

Drawing on this material, Judge McNall concluded that, once the receivers had been appointed, they were in control of the company within the meaning of section 1124(2). He expressed his reasons in this way:

“67.

In my view, the receivers of PH2L, on the facts of these appeals, control it within the meaning of CTA 2010 s1124(2). Both on the Agreed Facts, and on the terms of their appointment as set out above:

(1)

The whole of the property of the company was put into the hands of the Receivers;

(2)

The Receivers had very extensive powers, including the power to do or omit to do anything which they considered appropriate in relation to the Secured Assets;

(3)

The Receivers had the power to do all such things (including the carrying out of works) as may be necessary for the realisation of the property of the company;

(4)

The Receivers had the power to carry on the business of the company.

68.

It seems to me that the entire affairs of PH2L, read practically, were put into the hands of the Receivers.

69.

As a cross-check, it does not seem to me as if there were any relevant powers outside the scope of the receivership, nor any suggestion that the receivers had disavowed any of their powers, so as to permit any realistic argument, on the facts, that some sufficient control of PH2L’s affairs, for the purposes of CTA s1124(2) remained vested in the directors or shareholders of PH2L. I have already remarked that there is no evidence as to what the Receivers have actually been doing. Whilst the directors did remain in office, their powers of management were rendered incapable of being exercised. The receivers replaced the boards as the person having the authority to exercise the companies’ powers.

70.

In my view, once the receivers had been appointed, the affairs of PH2L were no longer being conducted in accordance with the wishes of the Appellants’ shareholders. That was sufficient to degroup PH2L. Whilst the Appellants’ shareholders continued to have control over the Appellants, they did not have control over PH2L, in the sense that they were no longer able to secure that PH2L’s affairs were conducted in accordance with their wishes.”

After referring to the facts, the legislation and the FTT Decision, the Upper Tribunal prefaced their consideration of the issues with a “preliminary point” which they discussed at [20] to [29] of the UT Decision. They said there appeared to have been

“at the very least a degree of ambiguity”, both in the parties’ arguments and in the FTT Decision, as to where “control” of both Farnborough (i.e. FA and FP) and PH2L actually resided, leaving aside the effect of the appointment of the receivers. If one looked merely at the immediate parent of FA and FP (i.e. Gatevalley), as the FTT appeared to do when it referred to “the Appellants’ shareholders” in the FTT Decision at [70], there could be no question of that company also controlling PH2L. It was only “several tiers upwards in the group structure that the link between Farnborough and PH2L is established” through Kelucia, their indirect common parent company. This, if I may say so, is clearly correct, and in this judgment I have referred to Kelucia as the company which admittedly had common control of FA, FP and PH2L before 27 June 2011.

The Upper Tribunal went on, however, at [27] to ask themselves whether “control” in a group situation ought properly to be found to reside in the parent company of the group, or in its shareholder. They pointed out, correctly, that in Pilkington the relevant control had been found to reside in the shareholders of Pilkington, acting together, from which they concluded at [29] that in “normal” group situations:

“the shareholders of the ultimate parent company are to be regarded, for the purposes of “Effect 2” as “persons together” having control of the parent company and, through it, of all companies in the group; and that this precludes any company in the group (including the parent company) from having such control over any other group company. It follows that when considering the issue of separation of control, the starting point must be that the shareholders of KL’s ultimate holding company (“the Shareholders”) must be regarded as initially having “control” of both Farnborough and PH2L. The substantive issues raised by the parties must then be argued from this starting point.”

With respect to the Upper Tribunal, this was in my view a misconception. As I have pointed out, the reason why in Pilkington the necessary control had to be located in Pilkington’s own shareholders was that Pilkington was itself the claimant company, and it could not in any meaningful sense be said to control itself (a notion which Nourse J at first instance had described as “curious, even comical”: see [1981] STC 219 at 234f). In the case of Kelucia, however, this problem does not arise, and there is no need to look any further up the group hierarchy in order to locate the common control. Fortunately, nothing turns on this point, as the Upper Tribunal acknowledged at the end of [29]; but it needs to be borne in mind when reading the UT Decision, and references in it to “the Shareholders” should where necessary be read as references to Kelucia.

In the next main section of their decision, running from [33] to [51], the Upper Tribunal dealt with the correct approach to construction of the legislation. They referred to some recent and well-known statements by the Supreme Court of the need to adopt a purposive approach to the interpretation of taxing statutes, and to the attempt (no longer pursued) by the appellants to rely on certain passages in Hansard for statements as to the intended purpose of the 1973 amendments to the group relief legislation. The Upper Tribunal also briefly considered other historical material relied on by Mr Ridgway, including the 1973 Budget press releases, and the Explanatory Notes to CTA 2010, but agreed (as do I) with the FTT’s conclusion at [47] of the FTT Decision, quoted above, with the possible substitution of “arrangements” for “transaction”.

In relation to the issue of “control”, the Upper Tribunal first considered what it called “the Receivers’ Control Issue”, that is to say whether the receivers had or could obtain, at any relevant time, control of PH2L for the purposes of section 1124(2). As they rightly pointed out at [53], this question could only be answered in the affirmative if the receivers had power to secure, as a result of any powers conferred by any “other document regulating” PH2L or any other body corporate, that the affairs of PH2L were conducted in accordance with their wishes. Although the FTT had not dealt explicitly with this question, Judge McNall must have considered that the Debenture was such an “other document regulating” PH2L, because of his conclusion (at [67] of the FTT Decision) that the receivers controlled PH2L within the meaning of section 1124(2). In arguing for a negative answer to the question, Mr

The Upper Tribunal accepted Mr Ridgway’s submission, saying at [57]:

“It is clear from the syntax that the words “regulating that or any other body corporate” refer back both to “the articles of association” and to the “other document” contemplated in section 1124(2)(b), and accordingly the type of regulation being referred to must be similar in relation to both. Therefore, the phrase “other document regulating that or any other body corporate” when read in context must, in our view, refer to a constitutional document akin to articles of association (i.e. one which sets out the governance arrangements for a body corporate which is binding upon members and officers by virtue of their status as such, without the need for them to agree separately to its terms). We infer that in referring to “other document”, the draftsman had mainly in mind the constitutional documents governing “non-standard” types of body corporate (e.g. companies incorporated overseas or bodies established by Royal Charter, where the legal terminology often does not include the phrase “articles of association”).”

It followed from this conclusion, as the Upper Tribunal said at [58], that however extensive the powers conferred on them by the Debenture, the receivers never had “control” of PH2L within the meaning of section 1124(2). Accordingly, they decided the Receivers’ Control Issue in favour of the appellants. This made it unnecessary for them to consider Mr Ridgway’s other arguments on the point: see [59].

The Upper Tribunal then turned to the question whether the relevant shareholders (i.e. Kelucia, on my analysis) ceased to have control of PH2L when the receivers were appointed. The Upper Tribunal called this the “Shareholders’ Control Issue”, and they discussed it at [60] to [77]. I will not summarise the arguments which the parties addressed to the Upper Tribunal on this issue, save to note Mr Ridgway’s unchallenged submission, recorded at [63], that the shareholders of PH2L continued to have powers under its articles of association:

The Upper Tribunal’s conclusion was that the appointment of the receivers did indeed result in the loss of shareholder control of PH2L. Their main reasoning is contained in paragraphs [71] to [74]:

“71.

The argument as put forward by Mr Bremner was effectively that the powers which the Receivers acquired over PH2L and its business and assets were so extensive that, whether or not the Receivers technically obtained “control” of PH2L, the Shareholders were deprived of it. Mr Ridgway effectively argued that section 1124 was concerned with “constitutional” control – by reference to voting rights in relation to what might be called “structural” matters rather than day to day operations. Furthermore, in relation to control, the law abhors a vacuum and therefore if the Receivers did not have control, it must have still resided in the Shareholders.

72.

Whilst Mr Ridgway’s arguments are superficially attractive, we do not consider they can be justified by reference to the wording of the statute. When section 1124 talks of constitutional matters (voting rights etc) it does so by reference to the means through which the putative controller’s power is exercised; but the fact remains that the power which must be held in order to have “control” is “the power to secure that the affairs of [PH2L] are conducted in accordance with P’s wishes”. We see no warrant for qualifying the word “affairs” by the word “constitutional” or any interpretation to a similar effect. Were it otherwise, the “common control” purpose of Effect 2 could easily be circumvented by means of some contractual arrangement which separated the day to day commercial operation of a company’s business from its constitutional structure.

73.

Once the Receivers were appointed, their powers (as summarised in the [FTT] Decision at [9] to [12]) were, in our view, so extensive that the shareholders could no longer be fairly said to have “the power to secure that the affairs” of PH2L were “conducted in accordance with” their wishes.

74.

It does not in our view matter that, on this interpretation, neither the Shareholders nor the Receivers had “control” of PH2L within the meaning of section 1124 CTA 2010. The wording of Effect 2 in section 154 CTA 2010 recognises that

“control” can be vested in more than one person “together”, and it may well have been the case that the Shareholders and the Receivers together had “control” of PH2L.”

In the final main section of the UT Decision, running from [78] to [94], the Upper

In the light of these conclusions, the Upper Tribunal dismissed Farnborough’s appeals.

In their grounds of appeal, the appellants contend that the Upper Tribunal erred in law

By a respondent’s notice, filed on 30 April 2018, HMRC maintain their primary case that the Upper Tribunal was correct to dismiss the appeals for the reasons that it gave. It is then contended that the UT Decision should be upheld for a variety of additional and/or different reasons, including:

I will begin with what seems to me to be the central issue in the case, namely whether the appointment of the receivers on 27 June 2011 had the result that Kelucia’s indirect shareholder control of PH2L came to an end. If it did, the conditions of Effect 2 were prima facie satisfied, because Kelucia admittedly continued to have shareholder control of the companies which claimed group relief, FA and FP. The situation would thus be one in which “…during or after the current period a person (other than the first or second company) has… control of the first company [FA or FP, as the case maybe] but not of the second company”. It would then be necessary to consider the further question whether the receivership constituted “arrangements” within the meaning of section 156(2). If the answer to this further question is yes, I do not understand it to be disputed that there was the necessary causal link between the arrangements and Effect 2, with the consequence that the appeals would have to be dismissed.

If, on the other hand, the question of loss of control were answered in the appellants’ favour, the appeal would then have to be allowed whether or not the receivership constituted “arrangements”. The conditions of Effect 2 would not be satisfied, and it is common ground that there is no other ground on which group relief could be denied.

Mr Ridgway put his argument in various ways, but his key submission was that Kelucia continued to have indirect shareholder control of PH2L after 26 June 2011, just as it had before. The shareholding structure remained the same, and the shareholder rights of Kelucia and the intermediate holding companies were unaltered. At a constitutional level, there was no change. The only things that did change related to the conduct of PH2L’s business. Before the appointment of the receivers, Article 90 of PH2L’s articles of association, which were in standard form, provided that “the business of the Company shall be managed by the Directors who may exercise all the powers of the Company.” In other words, the company was managed, in the usual way, by its board. The effect of the receivership was to vest these managerial powers in the receivers, but only for the proper purposes of the receivership and while it remained in force. The duties of the receivers, in general terms, were to realise the secured property and business assets to the extent necessary to discharge the secured debt and to repay the secured lenders, with any remaining funds being returned to the company. But none of this had any effect on the constitutional rights of PH2L’s shareholders.

Mr Ridgway also adopted a suggestion from the bench, to the effect that the receivership could realistically be viewed as no more than the working out of machinery put in place by the Debenture in October 2006. There is no suggestion that the grant of the Debenture to Bank of Scotland was anything other than a normal business transaction, entered into in the usual course of business. Accordingly, when steps were later taken by the Bank to enforce its security by the appointment of receivers, this was no more than a consequence of a normal business transaction, even if it had not turned out as well as the directors originally hoped.

In addition, Mr Ridgway relied on the fact that, under the Debenture, the relevant powers were conferred on the receivers as agents for the company, even though the agency of receivers admittedly has some peculiar features and is primarily a device to protect the mortgagee: see the discussion of the relevant principles by Lightman J, giving the judgment of this court, in Silven Properties Ltd v Royal Bank of Scotlandplc [2003] EWCA Civ 1409, [2004] 1 WLR 997. Furthermore, even if the effect of the receivership is to supplant the directors’ powers while it subsists, the directors remain in office throughout.

More generally, Mr Ridgway submits that the circumstances of the present case are far removed from the sort of artificial manipulation of the group relief system which led to the enactment of section 29 of the Finance Act 1973, exemplified by cases such as Pilkington and Tesco. The present case concerns an unobjectionable attempt to surrender genuine business losses to fellow subsidiaries within an established group relationship which remained wholly unchanged by the receivership.

In his submissions on behalf of HMRC, Mr Bremner places particular emphasis on the final limb of the definition of control in section 1124(2): “control” means the power of a person to secure, in either of the two specified ways, “that the affairs of company A are conducted in accordance with [his] wishes”. Thus control is not only a question of having the requisite powers at shareholder or board level, but also the ability to bring about a practical result, namely that the company’s affairs are in fact conducted in accordance with the wishes of the person exercising those powers. This critical element of the definition, submits Mr Bremner, is largely, if not entirely, overlooked by the appellants.

Mr Bremner derives support for this submission, not only from the clear language of the definition itself, but also from some of the reasoning of the Court of Session in the Lithgows case. As I have already said, the issue in that case (which concerned an early form of legislation designed to combat “transfer pricing”) turned on the materially identical definition of “control” in section 333(1) of the Income Tax Act 1952. The case concerned a sale of two ships by company A to company B, at less than market price. At the time of each sale, a majority of the shares in each company was held by trustees of a family settlement, each settlement having been made by the same settlor. The trustees of the two settlements overlapped, but were not identical. The same trustee, however, Sir Andrew Macharg, was the person first named on the register of shareholders of each company. The question was whether this circumstance alone sufficed to establish common “control” of the two companies within the meaning of the section. This argument was rejected by the First Division of the Court of Session. The leading judgment was delivered by the Lord President (Clyde), with whom Lord Carmont agreed. Lord Guthrie delivered a judgment concurring in the result, but his reasoning was not adopted by the other members of the court.

Lord Clyde began his analysis by pointing out, at 274, that Sir Andrew Macharg could secure that “both companies would act in accordance with the way he cast his votes”, but this was not enough to constitute control within the relevant definition because “[h]is power must be such that he can secure that the affairs of the two companies were conducted not merely according to his vote but according to his wishes.” Lord Clyde continued:

“If he were holding these blocks of shares as an individual, clearly he could secure that the affairs of both companies were conducted according to his wishes, and that, indeed, is the very situation with which Section 469 is primarily designed to deal. For an individual in that position is able, by his very real control over the two companies, so to arrange a transaction between them as to evade a liability to tax. But one of three or of four trustees is in a very different position. He is merely a joint holder of the trust shares along with his co-trustees, whose names in this case are on the register of the companies’ shareholders, and, even although he is first-named trustee, that does not give him power to secure that the affairs of the company are conducted in accordance with his wishes alone. A trustee is in a fiduciary position and has a duty not merely to the beneficiaries, but to his co-trustees.

…

It was argued for the Crown that it was enough that, at a meeting called to agree to the sale or purchase of the ships, Sir Andrew as first-named trustee could carry a resolution which he wished to carry although this was in defiance of the wishes of his co-trustees. But this argument is unsound. In the first place, if he did so he could not “secure” that the resolution stood, for it would obviously be reducible by his co-trustees (Wolfe v Richardson, 1927 S.L.T. 220 and 490). In the second place, under the articles of these companies…, the shareholders never would be called on to make the contract of sale, for this was done by the managers. But in the third place, Section 333(1) is not directed to the passing of particular resolutions but something much more general, namely the conduct of the affairs of the company. In my opinion, the mere fact that Sir Andrew is first-named on the register of a block of majority shares which he holds jointly with others will not give him the power to secure that the affairs of the company are conducted as he wishes. Indeed, if this simple fact of his name being first on the register is enough, then the statutory provision is of little benefit to the Crown, for all that would be necessary to avoid it is to alter the order of trustees in the register of one of the two companies, so that Sir Andrew’s name does not appear first in both registers. In my opinion, what the Sub-section is referring to is real control by one person, so that the company is really his creature. Such a situation does not apply to one of a body of trustee shareholders such as in the present case.”

Mr Bremner draws our attention to Lord Clyde’s reference, near the end of the passage I have just quoted, to “real control by one person, so that the company is really his creature”. He also relies, as did the FTT, on what Lord Guthrie said at 278:

“In my opinion, it is not sufficient, to satisfy the requirements of the definition, that a person is in a position to carry a particular resolution at a meeting of the company. In order that person may have “control” he must be in a position to secure that the affairs of the company are conducted according to his wishes. That phrase means that ability to achieve an isolated result, the power to carry a particular resolution, is insufficient to establish control in the statutory sense; and that what is required is power to secure the continuing conduct of the company’s affairs in accordance with the will of the person. Secondly, the definition does not state that control is the power of a person to secure that the affairs of the company are conducted according to his votes. The use of the word “wishes” suggests that the Statute requires that he shall be able to achieve his personal aims.”

In the light of this guidance, submits Mr Bremner, it is clear that following the appointment of the receivers the shareholders of PH2L were no longer able to secure the continuing conduct of the company’s affairs in accordance with their will, or to achieve their personal aims. It was only the receivers who had power to carry on the company’s business, and the wishes of the shareholders were necessarily subordinated to the decisions of the receivers. In no realistic sense could it be said that the shareholders continued to control PH2L.

In so far as the question is one of fact, Mr Bremner submits that the two Tribunals were amply justified in concluding that, once the receivers were appointed, their powers were so extensive that the shareholders could no longer be fairly said to have the requisite ability to secure that the company’s affairs were conducted in accordance with their wishes: see the FTT Decision at [67] to [70], and the UT Decision at [73]. Mr Bremner further submits that these conclusions reflect the usual position when a receiver is appointed. As the editors of Palmer’s Company Law say at para 14.112, with reference to a receiver appointed by the court, “in relation to the particular assets to which the security attaches, his appointment entirely supersedes the powers of the company and the authority of its directors in the conduct of its business which remain in abeyance during his appointment.” The authority cited for this proposition is the decision of the House of Lords in the Moss Steamship case. The position is no different, says Mr Bremner, where receivers are appointed out of court by a bank: see Gomba Holdings UK Ltd v Homan [1986] 1 WLR 1301 at 1306, per Hoffmann J.

In answer to the suggestion that the receivership is no more than the working out of a transaction undertaken in the ordinary course of the company’s business, Mr Bremner submits that this does not provide an answer to the relevant question. That question (whether the affairs of the company are conducted in accordance with the shareholders’ wishes) has to be answered on a continuing basis, from accounting period to accounting period, in the light of circumstances as they then are. The grant of the Debenture in 2006 did not trigger Effect 2, because the directors remained in full control of the company’s affairs unless and until receivers were appointed following an event of default. Once that happened, however, the situation was entirely different. The directors no longer had any managerial control of the business, and although the shareholders could still remove and replace them at will, this would achieve nothing of any practical significance while the receivership continued. Full power to run the business was now vested in the receivers alone, and the shareholders could do nothing to impose their own wishes in priority.

Furthermore, submits Mr Bremner, if it is necessary to find a rationale for the denial of group relief when receivers have been appointed in cases of the present type, it may be found in the fact that the two companies forming each group for group relief purposes no longer function as part of a single economic unit. The receivers owe their primary duties to the bank which appointed them, and their job is to get in the assets to repay the secured lender. The receivers’ powers of management are ancillary to that duty: see Silven Properties at [27] and Gomba Holdings, loc. cit., at 1305C.

I have not found this an easy question, and I see considerable attraction in some of the appellants’ submissions. In the end, however, I am satisfied that HMRC’s submissions should prevail. A telling initial point, as it seems to me, is that section 154 of CTA 2010 only applies in the first place if the two companies concerned would otherwise be members of the same group: see subsection (1). In other words, one starts from a situation where one company is the 75% subsidiary of the other, or (as here) both are 75% subsidiaries of a third company. The definition of “75% subsidiary” in section 1154 depends on ownership of at least 75% of the subsidiary’s ordinary share capital, and section 151(4) introduces the further requirement of a corresponding beneficial entitlement to dividends and distributions on a winding up. In those circumstances, the parent company will already have underlying shareholder control of the subsidiary, which to my mind strongly suggests that the requirement of control in Effect 2 in section 154(3) must have been intended by Parliament to go further. If HMRC’s submissions are correct, that is indeed the case, because of the need to establish that the affairs of the company are conducted in accordance with the shareholders’ wishes. On the appellants’ case, by contrast, that further requirement is either ignored or treated as an automatic consequence of the possession of a majority shareholding or voting power.

In agreement with Mr Bremner, I consider that the final limb of section 1124(2) is of crucial importance, both as a matter of ordinary language and because the test is formulated in terms of an ability to bring about an end result by specified means. By choosing to incorporate this definition for the purposes of section 154, Parliament must be taken to have intended this separate requirement to be replicated for the purposes of Effect 2.

At first sight, the appellants’ approach to the definition of “control” may be thought to gain some support from Pilkington, where all members of the Appellate Committee seem to have proceeded on the footing that the shareholders of Pilkington together had control of the company in the statutory sense merely by virtue of their constitutional position as shareholders. Indeed, it is commonplace to talk of control of a company residing ultimately with its shareholders, because they have the power in general meeting to decide how and by whom it should be run in accordance with the articles of association (which in turn they have the power to alter by the requisite majority). It was clearly not seen as an obstacle to this approach that the wishes of individual shareholders, in a body which probably numbered several thousand, could

Furthermore, there are no grounds for supposing that the receivership was likely to be limited in duration, or that there was any realistic prospect of the company resuming operation as a going concern. As I have pointed out, no evidence at all was led before the FTT about the nature of the company’s business, the circumstances of the receivership, or its actual outcome. In the absence of evidence, which it would have been incumbent on the appellants to lead if they wished to rely on it, the FTT was clearly entitled to assume that the receivership was effectively the end of the road for the company, subject only to realisation of its assets for the benefit of the secured creditors.

Strong support for HMRC’s case on control is also provided by the reasoning of the Court of Session in Lithgows. Although not strictly binding on us, we should normally follow a decision of the Court of Session in a tax case “because it is undesirable that there should be conflicting decisions on revenue matters in Scotland and England”: see Abbott v Philbin [1961] AC 352 at 373, per Lord Reid. The passages which I have quoted from the judgment of Lord Clyde form part of the ratio of the decision, and the observations of Lord Guthrie which I have quoted are to similar effect, although in reading both judgments it needs to be remembered that the focus of the court was on a factual situation where the wishes of only one individual shareholder were in issue. Clearly, the question of “wishes” cannot be analysed in quite the same way when one is dealing with corporate shareholders (as in the present case) or a multitude of individual shareholders (as in Pilkington).

I was for a time attracted by the notion that the receivership should be regarded as the working out of machinery which the board of PH2L deliberately put in place when it entered into the Debenture, and thus as something which happened in accordance with the shareholders’ wishes. Probably wisely, however, this contention does not appear to have been advanced by Mr Ridgway below, and he only adopted it in this court when I suggested it. In any event, I now consider that Mr Bremner was able to provide a satisfactory answer to it: see [62] above. I would also accept Mr Bremner’s submissions about the effect in law of the receivership, and the limited nature of the contractual relationship of agency between PH2L and the receivers provided for in the Debenture. The agency was of a purely ancillary nature, designed to facilitate realisation of the security for the benefit of the secured creditors. It did not in any sense enable the directors, and still less the shareholders, to give directions to the receivers about how the business should be carried on.

For these reasons, therefore, I conclude that the effect of the appointment of the receivers was indeed to deprive Kelucia (and the intermediate companies in the corporate chain) of their shareholder control of PH2L. It follows that the requirements of Effect 2 were satisfied.

It remains to consider whether there were “arrangements…in place”, within the meaning of section 154(2)(b), which had Effect 2. It will be remembered that “arrangements” are very broadly defined, in section 156(2)(a), as meaning “arrangements of any kind (whether or not in writing)”; and that in Pilkington, Lord Bridge gave the following guidance at 147B:

“ “Arrangements” is in the plural, not the singular, and I can see no justification for applying to the plural the concept of a combination for a singular purpose derived from the dictionary definition of a singular “arrangement”. The arrangements to be considered are such as “are in existence”; it matters not when they came into existence.”

The issue is thus whether the appointment of the receivers, and their subsequent conduct of the receivership in accordance with the Debenture, constituted

“arrangements” for the purposes of section 154. I do not propose to spend long on this question, because in my view each Tribunal was clearly right to find that the necessary arrangements were in place.

Mr Ridgway submits that the word “arrangements” should be interpreted purposively. I agree. But he then goes on to identify the purpose of section 154 as being to ensure that the benefits of group relief should not be available when the effect, in reality, would be to pass those benefits to an unrelated company outside the group. He then says that the appointment of the receivers did not disturb the economic group, nor is there any question in the present case of the benefit of losses being removed from the group. Arrangements which do not have such an effect, he says, are not the kind of arrangements which Parliament intended to render ineffective for group relief purposes when section 29 of the 1973 Act was enacted, so the word must be construed as excluding them.

In my judgment, this submission has to be rejected. It confuses the motives which led to the original enactment of what is now section 154 of CTA 2010 with the types of transaction to which the remedial measures, properly construed, actually apply. Neither in 1973, nor in 2010, was the legislation framed in explicitly anti-avoidance terms; and even the marginal note, to the extent that regard may be had to it at all, by the use of the word “etc” shows that the section is not exclusively concerned with the transfer of a group member outside its original group. The situations to which section 154 applies can only be identified by reading the section as a whole, together with the relevant definitions. I can find nothing in the wording of the section, so read, which suggests that “arrangements” should be given a restricted interpretation. On the contrary, everything seems to me to point to a broad meaning for the term, as Lord

Furthermore, the breadth of the definition is reinforced by the exception in section 156(2)(b). The power of a Minister of the Crown to give directions to a statutory body as to the disposal of assets belonging to the body would not normally be thought of as itself constituting an arrangement, although the actual giving of such directions, in an appropriate context, could perhaps more naturally be so regarded. Yet Parliament thought it necessary to provide, if only for the avoidance of doubt, that the mere existence of such a power was not to be treated as an “arrangement”. There would have been no need to spell this out, if the existence of such a power could not even arguably have been thought to fall within the definition.

More generally, it seems clear to me that the principal limitations on the scope of section 154 are to be found in the formulation of the three specified Effects which the arrangements must have if the first and second companies are to be degrouped. It is only by a process of circular wishful thinking, in my view, that the concept of “arrangements”, which is of course a staple feature of much anti-avoidance legislation, could be confined in its scope by reference to an unstated statutory purpose. Compare the remarks of Lord Bridge in Pilkington at the end of the passage quoted in [24] above (from 147H to 148A).

For these reasons, I would dismiss each of Farnborough’s grounds of appeal. This makes it unnecessary to consider the points raised in the respondent’s notice. Since the shareholders of PH2L lost control of it when the receivers were appointed, it does not matter whether the receivers themselves acquired control within the meaning of section 1124(2)(b), either alone or together with the shareholders.

I would therefore dismiss both appeals.

I agree.

I also agree.