The Trustees of the Panico Panayi Accumulation and Maintenance Settlements Nos. 1 to 4 v The Commissioners for HMRC [2026] EWCA Civ 744

[2026] EWCA Civ 744Case No CA-2025-000269Case No CA-2025-000303
IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE UPPER TRIBUNAL (TAX AND CHANCERY CHAMBER)
Mr Justice Rajah & Upper Tribunal Judge Nicholas Aleksander
UT-2020-000404 & UT-2023-000095
Venue Royal Courts of Justice, Strand, London, WC2A 2LLDate 12 June 2026
Before
LADY JUSTICE ASPLINLADY JUSTICE ELISABETH LAINGLORD JUSTICE MILES
Between
THE TRUSTEES OF THE PANICO PANAYI ACCUMULATION AND MAINTENANCE SETTLEMENTS NOS. 1 TO 4Appellants
Tim Johnston (instructed by Sharpe Pritchard Solicitors) for Appellants in PanayiDaniel Margolin KC (of Joseph Hage Aaronson & Bremen LLP) for Appellants in RedevcoMark Fell KC and Ben Elliott (instructed by HMRC Legal Group) for Respondent in PanayiBen Elliott and Stacey Cranmore (instructed by HMRC Legal Group) for Respondent in RedevcoHearing Hearing dates: 21, 22 and 23 April 2026
Approved JudgmentThis judgment was handed down remotely at 3pm on 12 June 2026 by circulation to the parties or their representatives by e-mail and by release to the National Archives..............................

Introduction

[1]These two appeals were heard together in the Upper Tribunal (Tax and Chancery Chamber) (‘the UT’). I will refer to the Appellants in the first appeal as ‘the Trustees’, and the Appellant in the second as ‘Redevco’ (collectively as ‘the Appellants’). For the avoidance of doubt, ‘the Trustees’ includes to the trustees as they have been from time. I will also refer to the Trustees’ appeal as ‘Panayi’ and Redevco’s appeal as ‘Redevco’. I refer to the Respondents as ‘HMRC’. The appeals concern transactions many years ago by which the Trustees and Redevco moved from the United Kingdom to another member state of the European Union (‘the EU’). This court is now concerned with the law as it was then. Later legislative amendments and/or repeals are not relevant to the issues in these appeals. The legislative provisions which applied to those transactions (among others) were sections 59B and 59D of the Taxes Management Act 1970 (‘the TMA’) (‘the relevant provisions’). The effect of section 2 of the European Communities Act 1972 (‘the ECA’) is that those transactions were also governed by EU law. It is common ground in both appeals that, as enacted, the relevant provisions interfered with the right of freedom of establishment conferred on the Trustees and on Redevco by the Treaty on the Functioning of the European Union (‘the TFEU’). I will refer to the rights conferred by the TFEU, and to the similar rights which were conferred by earlier Treaties, as ‘Treaty Rights’.[2]The First-tier Tribunal (Tax Chamber) (‘the F-tT’) adopted a ‘conforming interpretation’ of the relevant legislation by adding words to the relevant provisions in a way which, in each case, was designed to make that interference proportionate (to oversimplify somewhat). In Panayi,but not in Redevco, it is argued that that conforming interpretation also made provision for interest. The issue on the appeals to the UT, which were heard together, was whether the F-tT had erred in law in its approach. The UT held that the F-tT had erred in law in each appeal by dealing with interest, and re-made those determinations. It essentially adopted the same conforming interpretation as the F-tT in each case, but did not, in that conforming interpretation, deal with interest. The conforming interpretation in each case was to enable a taxpayer to pay the tax which was due by means of five annual instalments instead of in one sum. In outline, the Appellants challenge the UT’s decision to uphold the F-tT’s conforming interpretation of the legislation and its decision/observations about interest. That general statement is qualified by the last sentence of paragraph 4.3, below. As Mr Johnston confirmed during the hearing, the Trustees have not paid the tax at issue, on the grounds that they are not liable to pay it. Mr Margolin KC told us that Redevco paid the tax at issue on 15 November 2021.[3]On this appeal, the Trustees were represented by Mr Johnston and Redevco by Mr Margolin KC. HMRC was represented by Mr Fell KC, Mr Elliott and Ms Cranmore. I thank counsel for their helpful written and oral arguments. Their submissions have significantly influenced the approach I have taken in this judgment.[4]For the reasons given in this judgment I would dismiss these appeals. I have reached three broad conclusions. 1. There is no error of law in the conforming interpretation adopted by the F-tT in both cases. It follows that the UT did not err in law in (in effect) upholding that interpretation, subject to sub-paragraph 3, below. 2. On analysis, the adverse consequences of that conforming interpretation which were the subject of many of the Appellants’ complaints almost all flow from the retrospectivity which is an intrinsic feature of the declaratory nature of court decisions in our legal system. 3. The F-tT had no jurisdiction to make any decision about interest. If it had purported to do so, the UT would have had jurisdiction to set aside its determination. But the F-tT did not do so in either case, as I explain below. The UT therefore had no jurisdiction to set aside and to re-make the determinations of the F-tT. Redevco, but not the Trustees, appealed on that ground.

The structure of this judgment

[5]I will start by summarising the legislation which is relevant to each appeal. I will, second, summarise the facts in each by reference to the findings of the F-tT. Third, I will refer to the relevant legal framework and to some of the authorities on conforming interpretation. I will, fourth, summarise the reasoning of the UT, and, fifth, give my reasons for dismissing these appeals except to the extent that they challenged the UT’s decision on interest. In that final section I will refer, to the extent necessary, to the parties’ submissions.[1]The tax legislation[6]There is no dispute about the text of the relevant provisions or their effect so I will summarise them briefly in each case. (2). The Trustees’ case[7]At the relevant time section 2 of the Taxation of Chargeable Gains Act 1992 (‘the TCGA’) provided that a person was chargeable to tax on capital gains ‘in respect of chargeable gains accruing to him in a year of assessment during any part of which he is resident in the United Kingdom or during which he is ordinarily resident in the United Kingdom’.[8]The TCGA had detailed provisions about what gains were chargeable and how a gain was to be calculated. The Trustees accepted that shares such as those at issue in the appeal were within the scope of the TCGA , and that, on the facts, a gain on them would have been chargeable to tax. Section 69 of the TCGA provided, in relation to the liability to capital gains tax (‘CGT’) of the trustees of a settlement, that they should be treated as a single and continuing body of persons and that that body should be treated as resident in the United Kingdom unless the general administration of the trust was ordinarily done outside the United Kingdom and the trustees (or their majority) for the time being were not resident or ordinarily resident in the United Kingdom.[9]Section 80 of the TCGA, headed ‘Trustees ceasing to be resident in the United Kingdom’, applied if the trustees of a settlement became at any time (‘the relevant time’) neither resident nor ordinarily resident in the United Kingdom (section 80(1)). In such a case, section 80(2) provided that the trustees should be deemed for all the purposes of the TMA to have disposed of the ‘defined assets’ immediately before the relevant time and immediately to have re-acquired them.[10]Section 59B(1) of the TMA provided, in short, that for any year of assessment, subject to section 59B(2) (which is irrelevant), a taxpayer had to pay the amount of income tax and capital gains tax contained in his self-assessment for that year, less the deductions specified in section 59B(1)(b). Subject to section 59B(3) (which is also irrelevant), the taxpayer had to pay any tax due ‘on or before the 31st January next following the year of assessment’ (section 59B(4)). (2).

Redevco’s case

[11]The taxable profits of a company resident in the United Kingdom included chargeable gains which accrued to that company in the relevant accounting period (section 6 of the Income and Corporation Taxes Act 1988 (‘ICTA’) and section 8 of the TCGA). Section 185 of the TCGA was headed ‘Deemed disposal of assets on company ceasing to be resident in UK’. It applied if a company ceased to be resident in the United Kingdom ‘at any time (“the relevant time”)’. In that event, and with some exceptions, the company was ‘deemed for all purposes of this Act to have disposed of all its assets…immediately before the relevant time, and…immediately to have reacquired them at their market value at the time’.[12]Paragraph 10A of Schedule 9 to the Finance Act 1996 (‘FA’) had a similar effect on the treatment of the ‘loan relationships’ of a company which ceased to be resident in the United Kingdom at any time. Judge Brooks quoted it in paragraph 29 of his determination of Redevco’s appeal in the F-tT.[13]Section 59D of the TMA provided:
‘(1) Corporation tax for an accounting period is due and payable on the day following the expiry of nine months from the end of that period. (2) If the tax payable is then exceeded by the total of any relevant amounts previously paid (as stated in the relevant company tax return), the excess shall be repaid. (3) The tax payable means the amount computed in accordance with paragraph 8 of Schedule 18 to the Finance Act 1998…’
[2]The facts and their legal consequences in outline as found by the F-tT (1). The Trustees’ case[14]Judge Mosedale recorded in paragraph 1 that she had already made a decision in the current appeal in a decision issued in 2015. The F-tT had then decided to refer a question to the Court of Justice of the European Union (‘the Court of Justice’). It had given its judgment on 14 September 2017. The parties were unable to agree what should happen next, so the F-tT had listed a further hearing. The issue had been whether, if all or the majority of the trustees ceased to be resident in the United Kingdom, the exit charge imposed on the trustees by section 80 of the TCGA was compatible with EU law, and, in particular, with the fundamental freedoms of establishment, of provision of services and of movement of capital guaranteed by the TFEU.[15]The facts had been ‘largely agreed’. The F-tT had set them out in its earlier decision. Mr Panayi had been born in Cyprus. He had moved to the United Kingdom when he was a child in the 1940s. He lived and worked here until he went back to Cyprus in 2005. In 1992 he had established four trusts for the benefit of members of his family. He was not a beneficiary. He and a United Kingdom trust company were the original trustees. His wife later became a trustee. In 2004 they decided to go back to Cyprus and resigned as trustees. He appointed three people who lived in Cyprus as trustees. Although the United Kingdom company was still a trustee, the majority of trustees were then, therefore, resident in Cyprus.[16]His wife and children had moved to Cyprus in 2005. HMRC had agreed that Mr Panayi had never become domiciled in the United Kingdom. On 19 December 2005, the Trustees had sold the shares which were assets they held. The United Kingdom company and one of the Trustees had then resigned, and Mr and Mrs Panayi were appointed as trustees. They and the other two trustees who were resident in Cyprus were the Appellants. The value of the shares at the point when the majority of trustees had ceased to reside in the United Kingdom was agreed to be £30m. The tax at issue was agreed to be nearly £333,000, plus interest. The Trustees had told HMRC about the change in their residence and the sale of the shares, but did not in the relevant tax returns assess any liability under section 80 of the TCGA. HMRC opened inquiries into the returns in 2007. The closure notices in September 2010 assessed a charge under section 80. The Trustees appealed.[17]The F-tT recorded in paragraph 10 that the Trustees did not report a liability under section 80 in the relevant returns because ‘they have always taken the position that there was no such tax liability because the UK provisions imposing the liability were in breach of EU law as they failed to recognise the trustees’ EU law rights and in particular their right to freedom of establishment anywhere in the EU’.[18]The Court of Justice ruled that an entity such as the trust could rely on freedom of establishment and that that right could be relied on when tax was charged if trustees ceased to be resident in the United Kingdom. The Court of Justice had already ruled in other cases that such a charge might be lawful if there was an option to defer payment of the tax. The F-tT therefore also asked the Court of Justice whether an exit charge without an option to defer payment was unlawful if the asset had been sold before the tax liability was due for payment. The Court of Justice’s answer was that that made no difference. The legislation was disproportionate because it made ‘no provision for the taxpayer being able to defer the time when the tax payable is paid’ (see paragraphs 59 and 60 of its judgment).[19]Judge Mosedale decided that a conforming interpretation was possible for the reasons which she had given. Section 59B of the TMA ‘at a time before the legislation was actually amended…should be read in cases where the taxpayer’s right to freedom of establishment would otherwise be infringed, as including an option to defer payment of s 80 exit tax in 5 equal annual instalments, without liability to interest. (Interest would of course arise under the normal legislative provisions (s 86 TMA) to the extent that an instalment was unpaid after its due date). Early realisation would not precipitate liability nor could security be required’ (paragraph 166).[20]She considered the arguments about disapplication, nevertheless. She repeated that, contrary to the views of the parties, the provision of domestic law which was incompatible with EU law was not section 80 of the TCGA, but section 59B of the TMA. She said that it should only be disapplied ‘to the extent that it improperly deprived taxpayers of directly enforceable Community rights but no further’, quoting paragraph 49 of Fleming v HMRC [2008] UKHL 2; [2008] 1 WLR 195 (‘Fleming’). What did that mean? Was section 59B entirely disapplied, so that the tax was assessable but never due for payment? ‘Or does it mean that section 59B is only disapplied for [a] time, with the tax becoming due at a later date?’[21]She quoted paragraph 106 of the judgment of Henderson J (as he then was) in Prudential Assurance Co Limited v HMRC [2013] EWHC 3249 (Ch); [2014] STC 1236. He held that the effect of the principle of minimal disapplication described in Fleming would, in that case, produce the same result as a conforming interpretation.[22]In paragraph 174, she repeated the principle of minimal disapplication. The Court of Justice had not required the United Kingdom to forego the exit tax charged. ‘S 59B should only be disapplied such that the tax should not become payable until a date which respects the taxpayer’s EU law right to defer; in this case it is clear that that would be no earlier than 31 January 2007’. Redevco’s case[23]The decision of the F-tT in Redevco’s case is reported ([2023] SFTD 1028). Judge Brooks described the evidence in paragraphs 5-8. He heard evidence from two witnesses for Redevco, Mr Drury and Mr Faber. Judge Brooks found the facts in paragraphs 9-26.[24]Redevco was incorporated in the United Kingdom on 16 November 2004. It was resident in the United Kingdom until 15 January 2008. It is a member of a corporate group, owned by COFRA Holding AG, which is resident in Switzerland. Redevco’s immediate parent company, which Judge Brooks referred to as ‘Holding’ was based in the Netherlands. Redevco owned a portfolio of retail estate investments as a consequence of its connection with the C&A retail chain, and the closure of C&A’s 109 stores in the United Kingdom.[25]Holding began to consider ‘migrating’ Redevco to the Netherlands in 2006 and got advice from leading counsel about whether the United Kingdom would be able to impose an exit charge if that happened. Judge Brooks considered the evidence about that, to the extent that material had been disclosed, having already commented (in paragraph 6) that Mr Drury could not remember why the instructions to counsel did not say that the decision to migrate depended on counsel’s advice, a point to which Judge Brooks returned in paragraph 16. He referred to a further gap in the evidence in paragraph 18. Counsel’s strong view, as disclosed, was that corporate exit charges were contrary to EU law. Counsel had added that the migration plan was not ‘aggressive tax planning’. It was not artificial and had a commercial motive. Mr Faber’s evidence was that on the basis of that advice, there was a good chance that Redevco could argue, successfully, that no tax was due (paragraph 20).[26]In paragraph 21 Judge Brooks quoted the agreed statement of facts and issues, which gave more details about the migration. HMRC had opened an enquiry in 2010. The parties had ultimately agreed that a partial closure notice issued in 2019 was ineffective. HMRC re-opened the enquiry in 2020 and the parties agreed the value of properties. HMRC issued a final closure notice on 2 August 2021, which the F-tT quoted. Redevco appealed. The F-tT listed the issues in paragraph 21(14). By the time of the hearing, there was only one (see the last sentence of paragraph 29, below). In paragraph 23, he quoted two versions of a conforming interpretation which had been produced by HMRC on the second day of the hearing.[27]In paragraphs 24-26, he made further findings of fact. These were necessary because Redevco had raised a legitimate expectation argument. The issues were whether the migration was ‘solely for commercial reasons and whether it would have happened at all if the [legal] advice…had been different or less positive’. He found that the migration was ‘primarily for commercial reasons’, although tax considerations ‘certainly played a part in the decision’. The contemporaneous documents did not support the assertions of the witnesses about the role which counsel’s advice had played in the decision to migrate. ‘As such, I find it more likely than not that the migration would have proceeded in any event irrespective of the advice of leading counsel’.[28]In the introduction to his judgment, Judge Brooks recorded that Redevco ‘ceased to be UK-resident for corporation tax purposes and became resident in the Netherlands’. That had two effects. First, it was deemed, under section 185 of the TCGA, to have disposed of its assets and re-acquired them at market value, giving rise to a chargeable gain of £139,700,000. Second, paragraph 10A of Schedule 9 to the FA treated Redevco as having assigned the assets and liabilities represented by its loan relationships for a consideration equal to their fair value and immediately to have re-acquired them for the same consideration, giving rise to profits, for corporation tax purposes, of £2,700,114.[29]It was agreed that when Redevco left the United Kingdom for the Netherlands, section 185 of the TCGA and paragraph 10A of Schedule 9 to the FA (‘the exit charge provisions’) when read with section 59D of the TMA were incompatible with EU law. Judge Brooks recorded in paragraph 2 that this was because domestic tax legislation did not give a taxpayer the choice to defer payment of the relevant tax. It was also agreed that Redevco could rely on ‘directly effective EU law, in particular the Treaty right of freedom of establishment’. The parties disagreed about whether the remedy was to disapply the incompatible provisions, or to give them a conforming interpretation.[30]For the reasons he gave in paragraphs 48-64, Judge Brooks decided that he would follow the decision of Judge Mosedale in Panayi and the approach of this court in Vodafone 2 v Revenue and Customs Commissioners [2009] EWCA Civ 446; [2010] Ch 77 (‘V 2’) (see paragraphs 55-67, below). The legitimate expectation argument, which had failed on the facts, was not a reason for distinguishing Panayi.[31]The conforming interpretation he gave to section 59D of the TMA was:
‘Corporation tax for an accounting period is due and payable on the day following the expiry of nine months from the end of that period or, in cases where the taxpayer’s right of freedom of establishment would otherwise be infringed, in five equal annual instalments following the end of that period’
. Judge Brooks said nothing about interest.[3]Conforming interpretation: the legal framework (1). The European Communities Act 1972[32]At the relevant time, section 2 of the ECA provided, so far as relevant: 2. — General implementation of Treaties.(1) All such rights, powers, liabilities, obligations and restrictions from time to time created or arising by or under the Treaties, and all such remedies and procedures from time to time provided for by or under the Treaties, as in accordance with the Treaties are without further enactment to be given legal effect or used in the United Kingdom shall be recognised and available in law, and be enforced, allowed and followed accordingly; and the expression “enforceable EU right” and similar expressions shall be read as referring to one to which this subsection applies.(2) Subject to Schedule 2 to this Act, at any time after its passing Her Majesty may by Order in Council, and any designated Minister or department may by order, rules, regulations or scheme, make provision— (a) for the purpose of implementing any EU obligation of the United Kingdom, or enabling any such obligation to be implemented, or of enabling any rights enjoyed or to be enjoyed by the United Kingdom under or by virtue of the Treaties to be exercised; or (b) for the purpose of dealing with matters arising out of or related to any such obligation or rights or the coming into force, or the operation from time to time, of subsection (1) above; and in the exercise of any statutory power or duty, including any power to give directions or to legislate by means of orders, rules, regulations or other subordinate instrument, the person entrusted with the power or duty may have regard to the objects of the EU and to any such obligation or rights as aforesaid. In this subsection “designated Minister or department” means such Minister of the Crown or government department as may from time to time be designated by Order in Council in relation to any matter or for any purpose, but subject to such restrictions or conditions (if any) as may be specified by the Order in Council… (4) The provision that may be made under subsection (2) above includes, subject to Schedule 2 to this Act, any such provision (of any such extent) as might be made by Act of Parliament, and any enactment passed or to be passed, other than one contained in this part of this Act, shall be construed and have effect subject to the foregoing provisions of this section; but, except as may be provided by any Act passed after this Act, Schedule 2 shall have effect in connection with the powers conferred by this and the following sections of this Act to make Orders in Council or orders, rules, regulations or schemes. (2). The Human Rights Act 1998[33]Section 1(1) of the HRA defines ‘the Convention rights’ as ‘the rights and fundamental freedoms set out in(a) Articles 2-12 and 14 of the Convention,(b) Articles 1 to 3 of the First Protocol, and(c) Article 1 of the Thirteenth Protocol’. That definition does not include article 13, the right to an effective remedy. Section 2 deals with the interpretation of Convention rights.[34]Section 3(1) provides:
‘So far as it is possible to do so, primary legislation and subordinate legislation must be read and given effect in a way which is compatible with the Convention rights’
. Section 3(1) applies to primary and subordinate legislation whenever enacted, but does not ‘affect the validity, continuing operation or enforcement of any incompatible primary legislation, or affect the validity, continuing operation or enforcement of any incompatible subordinate legislation if (disregarding any possibility of revocation) primary legislation prevents the removal of the incompatibility’.[35]In any proceedings in which a court decides whether a provision of primary legislation is compatible with a Convention right, ‘[i]f the court is satisfied that the provision is incompatible with a Convention right, it may make a declaration of that incompatibility’ (section 4(1) and (2)). Section 4(3) and (5) makes provision for proceedings in which the issue is whether a provision of subordinate legislation is compatible with a Convention right. Section 5(5) defines ‘court’ for the purposes of section 5. The definition includes, in England and Wales, this court and the High Court. Section 4(6) provides that a declaration of incompatibility ‘(a) does not affect the validity, continuing operation or enforcement of the provision in respect of which it is given; and (b) is not binding on the parties to the proceedings in which it is made’.[36]The Crown is entitled to notice of any proceedings in which a court is considering whether to make a declaration of incompatibility (section 5(1)). In any such case, a Minister of the Crown, on giving the appropriate notice, is entitled to be joined as a party to the proceedings (section 5(2)).[37]Section 6 is headed ‘Acts of public authorities’. Section 6(1) makes it unlawful to act in a way which is incompatible with a Convention right. Section 6(1) does not apply if ‘as a result of one or more provisions of primary legislation, the authority could not have acted differently’ (section 6(2)). A ‘public authority’ includes a court or tribunal, ‘any person certain of whose functions are of a public nature’, ‘but does not include either House of Parliament or a person exercising functions in connection with proceedings in Parliament’ (section 6(3)). By section 6(6), an ‘act’ includes ‘a failure to act’, but does not include a ‘failure to introduce in, or lay before Parliament a proposal for legislation’, or to ‘make any primary legislation or remedial order’.[38]A person who claims that a person has acted (or proposes to act) in a way which is made unlawful by section 6(1) may bring proceedings in the appropriate court or tribunal, or rely on the Convention right or rights in any legal proceedings, but only if he is (or would be) a victim of the unlawful act (section 7(1)). ‘Legal proceedings’ include ‘an appeal brought against the decision of a court or tribunal’ (section 6(6)). By section 6(7), a person is the victim of an unlawful act only if he would be a victim for the purposes of article 34 of the Convention if the proceedings were brought in the European Court of Human Rights.[39]Section 8 is headed ‘Judicial remedies’. If a ‘court’ finds any act or proposed act of a public authority is or would be unlawful, ‘it may grant such relief or remedy, or make such order, within its powers as it considers just and appropriate’ (section 8(1)). ‘Court’ in section 8 includes a tribunal (section 8(6)). Damages may only be awarded by a court which has power to award damages or to order the payment of compensation in civil proceedings (section 8(2)).[40]Section 10 is headed ‘Power to take remedial action’. It applies in two circumstances. One is if a provision of legislation has been declared under section 4 to be incompatible with a Convention right, and there is no scope for an appeal against that declaration (section 10(1)). The other is ‘If a Minister of the Crown considers that there are compelling reasons for proceeding under this section, he may by order make such amendments to the legislation as he considers necessary to remove the incompatibility’.[41]Section 10(7) introduces Schedule 2 which makes further provision about remedial orders. A remedial order may contain ‘such incidental, supplemental, consequential or transitional provision as the person making it considers appropriate’ (paragraph 1(1)(a)). It may ‘be made so as to have effect from a date earlier than that on which it is made’ (paragraph 1(1)(b)) and may ‘make different provision for different cases’ (paragraph 1(1)(d)). The power conferred by paragraph 1(1)(a) includes a power to amend primary legislation (including primary legislation other than that which contains the incompatible provision (paragraph 1(2)). Paragraph 2 makes provision for the procedure for making a remedial order. There is provision for any representations made about a draft order to be considered before the order is made (paragraph 3(2)).[42]Section 19(1) imposes a duty on the Minister of the Crown who is in charge of a Bill in either House, before the Second Reading of the Bill, either to make a statement to the effect that in his view the provisions of the Bill are compatible with Convention rights (as statement of compatibility) or a statement that although he cannot make a statement of compatibility, the government nevertheless wishes the House to proceed with the Bill. (3). The authorities on conforming interpretation

Pickstone v Freemans Plc

[43]In Pickstone v Freemans Plc [1989] 1 AC 66, the employer employed men and women as warehouse operatives and as checker warehouse operatives. The applicants were women warehouse operatives. One man was employed in that job. The applicants claimed that they were entitled to the same pay as men who were checker warehouse operatives because they were doing work the value of which was ‘equal’ to that of the checker warehouse operatives, within the meaning of section 1(2)(c) of the Equal Pay Act 1970 (‘the EPA’).[44]When the EPA was brought into force in 1975, the European Commission took proceedings against the United Kingdom government on the ground that it had failed to introduce measures which gave effect to Council Directive (75/117/EEC) (‘Directive 1’) which implemented article 119 of the EC Treaty, because the EPA did not enable women to get equal pay for equal work unless there had been a job evaluation study. The Court of Justice agreed with the Commission (Commission of the European Communities v United Kingdom (Case 61/81) [1982] ICR 578 at 599). The United Kingdom had then taken steps to comply with that judgment (under section 2(2) of the ECA) by making regulations which inserted a new paragraph (c) in section 1(2) of the EPA. On a literal reading, the effect of that provision was to prevent a woman from claiming equal pay for work of equal value if one man in the same establishment was employed doing work which was like the work which she did.[45]The industrial tribunal (‘the IT’) dismissed the claims on the grounds that the applicants were doing work which, for the purposes of section 1(2)(a), was like the work of men who were employed in the same establishment, and that they could not rely on article 119 of the EC Treaty. The Employment Appeal Tribunal (‘the EAT’) dismissed their appeal against the decision of the IT. This court allowed their further appeal, and the employer appealed to the House of Lords. Lord Keith gave a speech, and agreed with the speeches of Lords Templeman and Oliver. Lords Brandon and Jauncey agreed with those three speeches.[46]Lord Keith described the issue of construction at page 111 G-H. His view was that a literal construction of section 1(2)(c) of the EPA would leave ‘a large gap in the equal work provision, enabling an employer to evade it by employing one token man on the same work’ as the potential applicants, meaning that the United Kingdom had ‘yet again failed fully to implement its obligations under article 119 of the Treaty and’ Directive 1. It was obvious that that was not what Parliament had intended. It could be that, ‘in order to confine the words in question to that situation, some necessary implication falls to be made into their literal meaning. The precise terms of that implication’ did not matter. It was enough to say that ‘the words must be construed purposively in order to give effect to the manifest broad intention of the maker of the Regulations and of Parliament’ (page 112 D).[47]Lord Templeman described the background and the issues. At page 120 G he said that ‘the words in paragraph (c) on which the employers rely were not intended to create a new form of permitted discrimination…In my opinion there must be implied in paragraph (c) after the word “applies” the words “as between the woman and the man with whom she claims equality”. This construction is consistent with Community law. The employer’s construction is inconsistent with Community law and creates a permitted form of discrimination without rhyme or reason’. That construction was consistent with the intentions of the Government when it introduced the Regulations (page 122 H). (See also page 123 B-D).[48]If the employer was right, ‘Parliament had simply failed in its purpose’, said Lord Oliver, ‘and that is a conclusion the court must strive to avoid - particularly having regard to … section 2(4) of [the ECA] - unless it is compulsively driven to it’ (page 125 B) (my emphasis). ‘If…the section is to be read literally, and in accordance with its terms, paragraph (c) cannot apply to …work so long as paragraph (a) applies to it’. There were only two ways in which it could be made to apply, as he explained at page 125 F-G. ‘Either way’, he continued, ‘a construction which permits the section to operate as a proper fulfilment of the United Kingdom’s obligation under the Treaty involves not so much doing violence to the language as filling a gap by an implication which arises, not from the words used, but from the manifest purpose of the [EPA] and the mischief it was intended to remedy. The question is whether that can be justified by the necessity – indeed the obligation – to apply a purposive construction which will implement the United Kingdom’s obligations under the Treaty’.[49]For the reasons given by Lord Templeman, he was persuaded that paragraph (c) should be read as modified by him, or by the insertion of ‘(in respect of the man hereinafter mentioned)’ between ‘work in relation to which’ and ‘paragraph (a)’. To read a provision which was, on its face, unambiguous, in this way, involved ‘a departure from a number of well-established rules of construction’. This sort of legislation, however, was, as a result of section 2(1) of the ECA, in a ‘special category’ (page 126 A - E). It was not necessary to rely on the Minister’s intention in introducing the Regulations. His initial doubts had been resolved by the analysis of Lord Templeman, and by the new section 2A. The words of section 1(2)(c), though ‘unequivocal’ on their face were ‘reasonably capable of bearing a meaning’ which would not put the United Kingdom in breach of its Treaty obligations. That conclusion was justified by ‘the manifest purpose of the legislation, by its history and by the compulsive provision of section 2(4) of [the ECA]’ (page 128 C-D).

Litster v Forth Dry Dock Company Limited

[50]The Transfer of Undertakings (Protection of Employment) Regulations 1981 (‘TUPE’) were introduced to comply with Council Directive (77/187/EEC) (‘Directive 2’). The 12 applicants in Litster v Forth Dry Dock Company Limited [1990] 1 AC 546 worked for a company which went into receivership. They were dismissed one hour before the receiver transferred the business of the company to the transferee. When they heard that the transferee was recruiting workers, they applied for jobs but were not successful. They made a complaint of unfair dismissal to the IT against the transferor and the transferee. The IT upheld their claims, as did the EAT. The Court of Session allowed the employer’s appeal, and they appealed to the House of Lords.[51]Lord Keith agreed with the speeches of Lords Oliver and Templeman, and gave a speech of his own. Lords Brandon and Jauncey agreed with all three speeches. On the face of TUPE, the effect of regulation 5(3) was that the protections which it conferred on employees only applied to those who were employed by the transferor ‘immediately before the transfer’. The issue was whether regulation 5(3) should be read literally, or not.[52]Lord Keith said that if regulation 5(3) was read literally, a transferor could readily evade the effect of TUPE by dismissing employees a short time before the transfer (page 554 D). The effect of decisions of the Court of Justice was that if employees of the transferor were dismissed before the transfer for a reason connected with the transfer, they were treated for the purposes of article 3(1) of Directive 2 as if they had been employed immediately before the transfer (page 554 F). It was the court’s ‘duty to give to regulation 5 a construction which accords with the decisions of [the Court of Justice] upon the corresponding provisions of [Directive 2] to which the regulation was intended by Parliament to give effect. The precedent established by Pickstone v Freemans Plc indicates that this is to be done by implying the words necessary to achieve that result’ (page 554 G-H).[53]After describing the background, Lord Templeman said that the transferee’s argument was inconsistent with Directive 2, in particular article 4(1), which provided that a transfer must not, in itself, constitute grounds for the dismissal of a worker. He referred to a decision of the Court of Justice to that effect (pages 557 G- 578 B). He considered that regulation 5(3) was not intended to be and should not be construed so as to limit the ‘operation of regulation 5 to persons employed immediately before the transfer in point of time’. It must be construed as applying to ‘a person employed immediately before the transfer or who would have been so employed if he had not been unfairly dismissed before the transfer for a reason connected with the transfer’ (page 558 G-H).[54]Lord Oliver said that ‘The approach to the construction of primary and subordinate legislation enacted to give effect to the United Kingdom’s obligations under the EC Treaty’ was not in doubt. He referred to Pickstone v Freemans Plc. If the legislation could ‘reasonably be construed to conform with those obligations’ which were to be gathered ‘not only from the wording of the relevant Directive’ but from its interpretation by the Court of Justice, ‘such a purposive interpretation will be applied even though…it may involve some departure from the strict and literal application of the words which the legislature has elected to use’ (page 559 D-E).

Vodafone 2 v Revenue and Customs Commissioners

[55]The issue in V 2 was whether the taxpayer was liable to tax on the profits of a wholly-owned subsidiary in Luxembourg under the controlled foreign companies (‘CFC’) provisions in the ICTA. The normal rule was that a parent company established in the United Kingdom was not taxed on the profits of subsidiaries as they arose. Section 747(3) of ICTA created a different rule about foreign subsidiaries. Section 748 was headed ‘Cases where section 747(3) does not apply’. Section 747(1) listed five exceptions to section 747(3). A further exception, in section 748(3), depended on the motive of the taxpayer. HMRC had started an inquiry into one of the taxpayer’s tax returns to see whether it was liable under the CFC provisions.[56]On 12 June 2006 the Court of Justice had decided in an earlier case that the CFC provisions were a restriction on the right of freedom of establishment conferred by article 43 of the EC Treaty (Cadbury Schweppes plc v Inland Revenue Commissioners (Case C-196/04) [2007] Ch 30). It held that that restriction could only be justified if its objective was to prevent wholly artificial arrangements designed to escape the tax which would otherwise be due. If, on the other hand, it was shown, by reference to objective factors, that, despite the existence of tax motives, the foreign company was actually established in a member state and carried on genuine economic activities there, that restriction was contrary to article 43.[57]The taxpayer applied to the Commissioners for an immediate closure notice in relation to the inquiry. The Commissioners held that the motive test could be interpreted in such a way that the CFC provisions would not apply even if there was an intention to get a tax advantage, if it could be shown objectively that the subsidiary was properly established in the member state. Evans-Lombe J held that the Commissioners’ interpretation of section 747(3) was wrong. He therefore disapplied the CFC legislation because it was ‘contrary to Community law’. He allowed the taxpayer’s appeal. HMRC appealed to this court.[58]The Chancellor of the High Court, giving a judgment with which the other members of this court agreed, said that the issue was whether ‘the CFC legislation could be interpreted, as required by section 2(4) of [the ECA], in a manner which did not unlawfully restrict V2’s freedom of establishment, as declared by the Court of Justice in the Cadbury Schweppes case’ (paragraph 4). Like the Chancellor, I will refer to the taxpayer in that case as ‘V2’. It was common ground that the right of freedom of establishment had direct effect, and that, ‘pursuant to section 2 of [the ECA] the CFC legislation must be either construed, or, if that is not possible, “disapplied” so as not unlawfully to restrict that freedom’. There were therefore two issues: was it ‘possible’ so to construe the legislation, and, if not, ‘to what extent (and in what manner) should it be “disapplied?”’ (paragraph 5). In paragraph 8 the Chancellor said it was clear that the exceptions in section 478(1) were not mutually exclusive and (in paragraph 9) that the application of the exceptions in section 478(1) depended on ‘the objective existence of specific circumstances’.[59]He quoted paragraph 55 of the judgment of the Court of Justice in paragraph 19. The Court of Justice had decided that a restriction on freedom of establishment could be justified on the ground that it prevented abusive practices, but its specific objective must be to ‘prevent conduct involving the creation of wholly artificial arrangements which do not reflect economic reality, with a view to escaping the tax normally due on the profits generated by activities carried out on national territory’.[60]It was not disputed that the conflict between V2’s right of freedom of establishment and ‘the prima facie entitlement’ of the Commissioners to tax V2 on the profits of its Luxembourg subsidiary ‘must be resolved in accordance with the provisions of section 2 of’ the ECA. It was not disputed that V2’s freedom of establishment was ‘an enforceable Community right within subsection (1) to be given legal effect without further enactment’. Nor was it disputed that ‘in accordance with subsection (4), the CFC legislation “shall be construed and have effect to subject to” that right’ (paragraph 25). The first issue related to ‘such construction’ and the second to ‘such effect’. He added that ‘If such a construction is impossible, the process by which effect is given to the enforceable community right is described as “disapplication”’. Disapplication involves treating the relevant provision in the CFC legislation as if it were expressed to be “without prejudice to” V2’s freedom of establishment…’ (paragraph 26). HMRC’s case was that on construction, or on disapplication, it could maintain its inquiry (paragraph 27).[61]The Commissioners and the Judge had only considered the construction of section 478(3). The Chancellor rejected the submission that the reasoning of the Court of Justice in Cadbury Schweppes meant that this court could only consider whether section 478(3) was amenable to a conforming interpretation. The Court of Justice did not have jurisdiction to do more than to interpret the relevant Treaty ‘and the other matters referred to in article 234EC. They do not include the interpretation of the legislation of a member state…Further, as those citations show, the obligation of the national court is to examine the whole of national law to consider how far it may be applied so as to conform with enforceable Community rights’ (paragraphs 34-36).[62]This court listed the authorities on the principles which apply to a conforming interpretation and to which it had been referred in paragraph 37. It also quoted HMRC’s agreed summary of those principles. One of those authorities was Ghaidan v Godin-Mendoza [2004] UKHL 30; [2004] 2 AC 557 (‘Ghaidan’). Unlike the other cases which were cited in V2, Ghaidan concerns the obligation imposed by section 3(1) of the HRA. In paragraph 38, the Chancellor quoted HMRC’s submissions about the constraints on ‘the broad and far-reaching nature of the interpretative obligation’. They were agreed by the taxpayer.[63]HMRC submitted that the ‘grain’ or ‘thrust’ of the legislation’ was to ‘cast the initial net wide’ and then to ‘narrow it by the overlapping exceptions’ in section 478(1) and in section 478(3). A conforming interpretation would be to add another exception to the list in section 748(1). Its exact words did not matter. V2’s third objection to that construction was that it would operate retrospectively and involve legislative or policy decisions, and offend the principle of legal certainty (paragraph 40). The Chancellor remarked in paragraph 41 that the terms of section 3(1) of the HRA ‘are in substance the same as the terms of section 2(4) of’ the ECA. He added, however, that ‘the consequence of the inability to find a conforming interpretation is a declaration of incompatibility not disapplication in the manner I have explained’.[64]The Chancellor considered that the extension of the exceptions to the CFC legislation proposed by HMRC was ‘as permissible as either of those which found favour in’ Ghaidan or in R (IDT Card Services Ireland Limited) v Customs and Excise Commissioners [2006] EWCA Civ 29; [2006] STC 1252.It added to the exceptions, but did not change their effect. It did not go against ‘the grain’ or the ‘thrust’ of the legislation. The existing exceptions were not mutually exclusive and could be changed (see section 748(1)(e), which gave HMRC power to specify territories in regulations subject to such conditions as HMRC might also specify). Section 278(1)(e) also showed that a geographical distinction went with ‘the grain’ of the legislation (paragraph 54).[65]In paragraph 56, the Chancellor said that several points were ‘wrapped up’ in V2’s third objection to HMRC’s conforming interpretation (see paragraph 63, above). It was ‘inevitable’, first, that a conforming interpretation would be retrospective. There was no occasion to consider a conforming interpretation unless and until it was argued that a provision did not comply with EU law. The retrospectivity of such an interpretation was no more an objection in this area than in the case of domestic statutory construction. Second, it is not a condition of a conforming interpretation that ‘it should be capable of precise formulation’. It was ‘inevitable’ that such an interpretation would ‘lack the crispness to be expected of properly considered legislation’. That was an insufficient objection.[66]There could be no objection, third, to a conforming interpretation which followed ‘precisely that element’ of the legislation which ‘the Court of Justice held to constitute the hindrance to freedom of establishment’, and excepted it from the CFC legislation. That was, by definition, sufficiently certain for a conforming interpretation. Nor, for a similar reason, could there be any objection to the exclusion. ‘It follows precisely the formulation of the justification for the hindrance which the Court of Justice found to be acceptable’ (paragraph 58). If there were other ways to achieve the same objective, such a choice might well involve policy decisions. But that could not, of itself, make a conforming interpretation illegitimate (paragraph 59).[67]The Chancellor considered the question of disapplication in paragraphs 61-66. That reasoning was not necessary to the decision of the court. The Chancellor described the parties’ rival contentions in paragraphs 62-64. The essential difference between the four versions (two versions put forward by each) was whether ‘the limitation on the freedom of establishment constituted by the justification found by the Court of Justice to be permissible is to be read into it so as to restrict the class of person who… “are so circumstanced that the offending provisions must not be invoked against them, either in particular cases or at all”’ (paragraph 65). He did not reach a concluded view about that, but said he would need ‘a good deal of persuading’ that ‘it was appropriate simply to disapply the CFC legislation’ in the relevant period as that would be to ignore the justification for the restriction found in by the Court of Justice in the Cadbury Schweppes case, ‘and to disapply the legislation more extensively than the Court of Justice considered to be necessary to preserve the relevant freedom of establishment’ (paragraph 66).

British Gas Trading Limited v Lock

[68]British Gas Trading Limited v Lock [2016] EWCA Civ 983; [2017] ICR 1 (‘Lock’) concerned the calculation of holiday pay under the Working Time Regulations (1998 SI No 1833) (‘the WTR’). The WTR were made to implement Directive 2003/88/EC (‘Directive 3’). The claimant brought a claim in the Employment Tribunal (‘the ET’). The ET adopted a conforming interpretation of the WTR. It held that it was necessary to add a new sub-paragraph to regulation 16 of the WTR. The EAT upheld the ET’s decision. The employer appealed to this court. Sir Colin Rimer gave a judgment dismissing the appeal. The other members of this court agreed with it.[69]Sir Colin started by considering some authorities about conforming interpretation, in paragraphs 31-40. The first was Marleasing S.A. v La Comercial Internacional de Alimentación S.A. (Case C-106/89) [1992] 1 CMLR 305 (‘Marleasing’) (see paragraphs 87-89, below). The second was Ghaidan. He recorded in paragraph 32 that the parties agreed that the principles governing the obligation imposed by section 3 of the HRA were the same as those which apply under Marleasing. The third was V 2. In paragraphs 39 and 40 he referred to two decisions of the Supreme Court about the obligation to interpret domestic legislation ‘so far as possible in the light of the wording and purpose of the Directive which it seeks to implement’.[70]In paragraphs 100-104 Sir Colin discussed the contra legem principle, partly by reference to the judgment of the Court of Justice and the opinion of the Advocate General in Dansk Industri v Estate of Rasmussen (Case C-441/14 [2016] ICR D9 (‘Dansk’). He said that the question for this court was whether it was possible to interpret the WTR ‘in a way that conforms to’ the requirements of the relevant Directive as revealed in decisions of the Court of Justice which post-dated the WTR and in a domestic authority which pre-dated those decisions. It was agreed that the earlier domestic authority did not force this court to ‘answer that question in the negative. One element of the contra legem principle referred to by the EU decisions requires us to leave [the earlier domestic case] out of account when approaching the task of finding, if possible, a conforming interpretation: see the judgment in Dansk, para 33’ (paragraph 100).[71]Another aspect of that principle ‘reflects the acceptance by the CJEU that, when engaging in a conforming interpretation…a national court may find it impossible to adopt a conforming interpretation when “the clear unequivocal wording of a provision of national law appears to be irreconcilable with the wording of a directive”…’ (paragraph 101). Sir Colin considered that if that aspect of that principle ‘as explained in the EU cases is applied at anything approaching face value, it would be likely to frustrate the possibility of a conforming interpretation in many cases’. It was difficult to see how the House of Lords could have decided Pickstone as it had done ‘had it applied the …principle as so explained’. What emerged from the later domestic cases (including two decisions of the Supreme Court) was that the United Kingdom had dealt with the contra legem principle ‘in a manner that is manifestly more in line with the EU objective of conforming interpretation at member state level’ than anything in the Opinion of the Advocate General in Dansk. The courts of the United Kingdom do not confine themselves to the literal meaning of ‘language which may appear to stand in their way’. Rather, they asked a broader question, whether a conforming interpretation would be in line with the grain of the underlying legislation. ‘That is an approach which ought, I would think, to attract nothing but commendation by the CJEU’ (paragraph 103). He repeated in paragraph 104 that ‘the critical question’ is ‘whether [this] conforming interpretation of the WTR …is or is not within the grain or underlying thrust of that legislation. If it is, I consider it ought to follow that the interpretation favoured by the courts below is one this court should uphold. If it is not, a conforming interpretation is not possible’.[72]The court had objectively to assess whether ‘a legislative choice has been made that is directly at odds with the requirements of the Directive’. The provisions of the WTR were more consistent with the view that ‘the legislature simply’ had not ‘foreseen the particular problem that was in due course to arise with the subsequent decisions of the CJEU…’ (paragraph 110). He was not prepared to decide that the two anomalies which the litigation had brought to light ‘reflect a positive legislative choice deliberately directed at discriminating’ against the workers concerned. ‘As a matter of objective inference, I regard it as more likely that the differential treatment inherent in the scheme of the WTR was simply not foreseen at the time they were enacted’ (paragraph 111).[73]The ‘grain or thrust’ of the WTR was to provide for holiday pay in the circumstances set out in Directive 3, as explained by the Court of Justice. This court should therefore interpret the WTR as providing for holiday pay in the circumstances of that case. ‘So to interpret the WTR does of course require the implication…of words necessary to make that meaning clear. But so to imply words is not a judicial exercise amounting to the repeal or amendment of the legislation. It is rather an example of the court performing its duty to provide a conforming interpretation to legislation introduced for the purpose of implementing a Directive’ (paragraph 112).

Revenue and Customs Commissioners v Applicants in the Post Prudential Closure Notice Applications Group Litigation

[74]Revenue and Customs Commissioners v Applicants in the Post Prudential Closure Notice Applications Group Litigation [2025] EWCA Civ 166; [2025] 1 WLR 4463 (‘Prudential’) concerned the tax treatment of dividends from ‘portfolio’ holdings of shares in companies which were not resident in the EU, and the availability of double taxation relief (‘DTR’). Decisions of the Court of Justice before 2009 established that the taxation of such holdings pursuant to section 790 of ICTA was contrary to EU law, but that in the case of portfolio holdings, that breach could be cured by a conforming interpretation of section 790. The taxpayers then relied on section 790, so construed, as entitling them to tax refunds or reliefs. HMRC refused those applications. Some of them appealed to the F-tT against closure notices refusing such applications. Others applied to the F-tT for it to direct HMRC requiring them to issue closure notices in relation to (other) open inquiries. On appeal, the UT decided most of the issues in favour of HMRC. Both the applicants and HMRC appealed to this court. In a judgment with which the other members of this court agreed, Falk LJ dismissed the taxpayers’ appeals and allowed the appeals of HMRC.[75]Falk LJ considered, in detail, a range of arguments about the ramifications of a conforming interpretation which is applied many years after the end of the accounting years for which, as that interpretation later shows, reliefs and credits were available, albeit the taxpayer did not then know that they were available. As she put it in paragraph 6 of her judgment, ‘In broad terms the Taxpayers…seek to rely on their statutory rights under tax legislation, as interpreted to conform with EU law, to generate tax refunds or reliefs as an alternative to common law claims in restitution’. Most of the disputes were about procedural mechanics. Two ‘overarching points’ ran through the apparently disparate points in the appeal. The first was ‘the extent to which’ the procedural regime should be interpreted so as to give all the benefits of a tax regime which complied with EU law to the taxpayers, when they did not know what their rights were before the relevant time limits expired, and so made (as HMRC submitted) ineffective claims. The second was the extent to which the conforming interpretation benefitted the taxpayers (paragraph 15).[76]Falk LJ summarised the authorities about the relevant conforming interpretation in paragraph 21. In essence, it was that in addition to being entitled to relief from withholding tax, an underlying tax credit at (in practice) the foreign nominal rate (‘FNR’) was available for portfolio dividends. The problem for the taxpayers was that a claim for DTR was necessary to vindicate their EU rights and that correct legal analysis had been gradually established over a long period. For a long time it had not been clear how the breach of EU law was to be addressed, or what the extent of any claim might be. DTR has to be claimed. A further difficulty was that the legislation did not make that as clear as it might have done (paragraph 24).[77]The taxpayers argued that they had made every conceivable claim to vindicate their EU rights. They complained that the UT had ‘retrospectively invalidated’ those claims. The statutory procedures had been applied as if section 790 had always provided for a credit at the FNR, although the taxpayers could not have anticipated that when they made their claims. That was said to be contrary to EU law and to their Convention rights. The taxpayers argued that the UT’s approach breached the principles of equivalence, effectiveness, legal certainty and effective judicial protection.[78]HMRC argued that there was no EU regime for the restitution of improperly levied charges, with the result that that was a matter for the domestic legal system, so long as any remedial regime did not breach the principles of equivalence or of effectiveness.[79]This court dismissed the appeal on ground 1. Falk LJ said that it was based on the incorrect premise that ‘the domestic procedural rules would otherwise be treated as retrospectively re-written to deprive the Taxpayers of an effective remedy. The true position’ was that the domestic procedural rules complied with the relevant requirements of EU law. They had not been re-written and there was no reason to change them now so as to treat otherwise ineffective claims as effective (paragraph 39).[80]The ‘starting point’ was that the relevant decisions were ‘in their nature retrospective in effect, in the sense that they declared what the law had been at all relevant times. That accords with the normal declaratory theory of judicial decision-making’. The long period of uncertainty was the product of the developing case-law of the Court of Justice and not of any aspect of national procedural law (paragraph 40). The courts in the United Kingdom were obliged to give effect to the decisions of the Court of Justice, and they were applied ‘in accordance with the normal declaratory approach’. The courts gave effect to EU law by a conforming interpretation. As the Chancellor had pointed out in paragraph 56 of V 2 (see paragraph 65, above), in the context of the taxpayer’s argument about legal certainty, ‘conforming interpretation is inevitably retrospective in its operation’ (paragraph 41).[81]In paragraph 42, Falk LJ quoted, with approval, the statement by Lord Sumption JSC in paragraph 176 of Test Claimants in the FII Group Litigation v Revenue and Customs Commissioners [2012] UKSC 19; [2012] 2 AC 337: ‘It is no doubt correct that, however strained a conforming construction may be, and however unlikely it is to have occurred to a reasonable person reading the statute at the time, a later judicial decision to adopt a conforming construction will be deemed to declare the law retrospectively in the same way as any other judicial decision’.[82]The same point was relevant to the principle of effectiveness. Falk LJ said, in paragraph 44, that there was nothing in the EU law principles relied on by the taxpayers ‘(whether equivalence, effectiveness, legal certainty or effective judicial protection) that mandates a different approach’. There was no basis in principle for a distinction between a taxpayer who does not know that a principle of EU law has been infringed and a taxpayer who has ‘some, imperfect knowledge’ which leads him to make the wrong claim. As long as any relevant time limit in national law is ‘reasonable and fixed in advance’ it will be valid, ‘whatever the state of knowledge of the claimant and however reasonably (or otherwise)’ he acted. Nor was there any obligation to interpret domestic rules as requiring a claim for ‘x’ to be treated, instead, as a claim for ‘y’. For the reasons given in paragraph 46, ‘The Taxpayers’ complaint is in truth about uncertainty in the development of the substantive law and not about domestic procedural rules’.

The authorities on section 3 of the HRA

[83]We were referred to many authorities about the obligation imposed on courts by section 3 of the HRA. It has been assumed or agreed in the cases to which we referred in argument that the approaches under section 2 of the ECA and under section 3 of the HRA are the same. I will also make that assumption because we heard no argument to the contrary. It seems to me, nevertheless, that the obligations imposed by section 2 of the ECA may be stronger than those imposed by section 3 of the HRA; that much is strongly suggested by the reasoning in Pickstone and Litster (see paragraphs 43-49, and 50-54, above) which refer, for example, to the compulsive effect of section 2(4), in particular. Domestic courts were bound to disapply legislation which was incompatible with EU law, whereas a court which holds that it is not ‘possible’ to interpret legislation so as to comply with Convention rights can only make a declaration of incompatibility. A further potentially material point, which was not really addressed in argument, is that this (unlike many of the relevant authorities) is a case in which domestic legislation restricts Treaty rights. It is not a case about national legislation the purpose of which is to implement EU legislation.[84]In, for example, V 2 and Lock, this court has used phrases and metaphors which are similar to those in the cases about section 3. It is not necessary to make this judgment any longer by fully summarising the authorities about section 3. Those authorities, show, broadly, first, that an interpretation is ‘possible’ for the purposes of section 3 of the HRA if it ‘goes with the grain’ or with the ‘thrust’ of the relevant legislation. The first metaphor is apt to be misunderstood. It does not refer to granular compliance with the detail of the relevant legislation. What it evokes, instead, is a craftsman planing a plank of wood. He will always work the plane smoothly in the direction of the grain of the wood, and not in the opposite direction, against the grain. That is why the alternative metaphor is that a conforming interpretation must ‘go with the thrust’ of the legislation.[85]The second broad principle (in the domestic sphere) is that the court has no power to make legislation, which is Parliament’s job. The court’s duty to ‘interpret’ legislation is not also a power to make legislation. The court cannot, for example, make the types of policy choices which it is for Parliament to make. It ineluctably follows, therefore, that the ‘interpretation’ can and must go no further than is necessary to discharge the interpretative obligation in conformity with EU law (or with Convention rights, as the case may be). We were referred to many elaborations of this principle in the cases, but none of those is a statute, and none establishes binding rules about how a court is to interpret legislation.

The relevant decisions of the Court of Justice

[86]There were 20 decisions of the Court of Justice in the authorities bundles for this hearing. I will only summarise the judgments which are, in my view, essential to a decision on the issues in these appeals.

Marleasing

[87]Marleasing concerned the interpretation of article 11 of Directive 68/151 (‘the Directive’). The Kingdom of Spain had been obliged to implement the Directive from the date of its accession to the EU, but had not complied with that obligation. The Spanish Civil Code (‘the SCC’) contained provisions which were inconsistent with article 11 of the Directive. The Court of Justice held that the SCC must be interpreted so as to give effect to article 11 as if it had been implemented.[88]The Advocate General said that ‘The obligation to interpret a provision of national law in conformity with a directive arises whenever the provision is to any extent open to interpretation. In those circumstances the national court must, having regard to the usual methods of interpretation in its legal system, give precedence to the method which enables it to construe the national provision concerned in a manner consistent with the directive’ (paragraph 8). In a footnote he referred to Litster as an example of that approach. That obligation was limited by Community law and by the principles of legal certainty and non-retroactivity.[89]In paragraph 8 of its judgment the Court of Justice said that, for the reasons given in that paragraph, ‘…in applying national law, whether the provisions were adopted before or after [a] directive, the national court called upon to interpret it is required to do so, so far as possible, in the light of the wording and purpose of the directive in order to achieve the result’ pursued by that directive. The answer to the question referred was that the national court hearing a case falling within the scope of the directive ‘is required to interpret its national law in the light of wording and purpose of [the directive] in order to preclude a declaration of nullity of a public limited company on a ground other than those listed in’ article 11 of the directive.

Teckal Srl v Commune di Viano

[90]The Fifth Chamber of the Court of Justice considered the scope of article 177 of the EC Treaty in Teckal Srl v Commune di Viano (Case C-107/98, 18 November 1999).[91]In paragraph 29 of its judgment the Court of Justice said that article 177 was based on ‘a clear separation of functions between the national courts and the Court of Justice, which means that, when ruling on the interpretation or validity of Community provisions, the Court of Justice is empowered to do so only on the basis of the facts which the national court puts before it….’. It was therefore for the national court to ‘ascertain the facts which have given rise to the dispute and to establish the consequences which they have for the judgment which it is required to deliver’ (paragraph 30). The division of functions in article 177 meant that it was ‘for the national court to apply the rules of Community law to a specific case. No such application is possible without a comprehensive appraisal of the facts of the case’ (paragraph 31).[92]In paragraph 33, it considered the powers of the Court of Justice in relation to the interpretation of national legislation. It said that in the context of article 177, ‘the Court has no jurisdiction to rule either on the interpretation of provisions of national law or regulations or on their conformity with Community law. It may, however, supply the national court with an interpretation of Community law that will enable that court to resolve the legal problem before it…’.

Trustees of the P Panayi Accumulation and Maintenance Settlements v Revenue and Customs Commissioners

[93]The Court of Justice has considered in several cases whether, and if so, to what extent, exit charges are unjustified restrictions of Treaty rights. Trustees of the P Panayi Accumulation and Maintenance Settlements v Revenue and Customs Commissioners (Case C-646/15) [2017] 4 WLR 210 (‘Panayi CJ’) isthe outcome of the reference which the F-tT made in Panayi. The Opinion of the Advocate General and the judgment in Panayi CJ conveniently summarise most of the relevant principles. They concern whether a requirement for immediate payment is contrary to EU law, and whether an exit charge with option to defer payment over a period is nevertheless compatible with EU law.

The requirement for the immediate payment of exit charges

[94]The two main issues in Panayi CJ were whether a trust could benefit from Treaty rights and whether the domestic legislation was disproportionate. The Court of Justice decided the first issue in favour of the Trustees. I will now summarise its reasoning on the second issue.[95]The Court of Justice noted that the shares had been sold on 19 December 2005, over a year after the majority of the Trustees had ceased to be resident in the United Kingdom. If the Trustees had still been resident in the United Kingdom, then that sale would have triggered a charge to tax on any gains in the value of those shares, which would have become payable on 31 January 2007 (paragraph 19). The Court of Justice quoted the three questions which the F-tT had referred to it in paragraph 21. The Court of Justice considered that the F-tT had in substance asked whether the provisions of the Treaty about freedom of establishment ‘preclude’ national legislation which provides for the taxation of unrealised gains in the value of assets held by trustees when their majority moved to another member state and does not permit deferred payment of the relevant tax (paragraph 22) (my emphasis).[96]The Court of Justice considered, first, whether trusts could benefit from freedom of establishment (paragraphs 24-40).[97]In paragraphs 41-61, it considered whether there was a restriction on freedom of establishment. The caselaw of the Court of Justice showed that article 49 of the Treaty ‘requires the elimination of restrictions on freedom of establishment’ (paragraph 41). Measures which ‘prohibit, impede or render [it] less attractive’ to move to another member state amount to such restrictions (paragraph 43). Had the newly appointed trustees been resident in the United Kingdom, the unrealised capital gains would not have been liable to taxation (paragraph 45). The legislation treated a trust, the trustees of which were resident in the United Kingdom, differently from a trust the trustees of which were resident in a different member state (paragraph 46). That difference amounted to a relevant restriction, for the reasons given in paragraph 47.[98]The Court of Justice repeated a point it had made in earlier decisions: that ‘the preservation of a balanced allocation of powers of taxation between member states is a legitimate objective recognised by the court’. In the absence of any harmonising measures in the EU, member states ‘retain the power to define, by treaty or unilaterally, the criteria for allocating their powers of taxation, with a view to eliminating double taxation’ (paragraph 51). The fact that the trustees were no longer resident in the United Kingdom ‘cannot mean that [the United Kingdom] has to abandon its right to tax a capital gain which arose within the ambit of its powers of taxation before that transfer’. The Court had held that the member state was entitled to charge a tax on those gains when the taxpayer leaves the country. Such a measure can be justified (paragraph 52).[99]But it can only be justified where the member state is ‘actually prevented from’ taxing such income (paragraph 53). The Court of Justice rejected the Trustees’ argument that the potential liability of the beneficiaries for gains attributed to them ‘could be taxed as gains accruing to them’ (paragraphs 54 and 55). That meant that legislation which provided for the taxation of unrealised gains in the value of trust assets when the place where the trust was managed was moved to another member state, even if the first member state retained some relevant powers of taxation, ‘is a suitable means of ensuring the preservation of the allocation of powers of taxation between member states…’ (paragraph 56).[100]The Court of Justice then considered the proportionality of the legislation. It followed from the court’s case law that ‘the fact that’ the member state ‘determines the amount of tax due on the unrealised capital gains that have arisen on its territory at the time when its power of taxation in respect of those capital gains ceases to exist, in this case when the place of management of the trust is transferred to another member state, is compatible with the principle of proportionality…Further legislation…which provides that a trust which transfers its place of management to another member state may choose between immediate payment of the tax due or deferred payment of that tax, together with, if appropriate, interest in accordance with the applicable national legislation, would constitute a measure less harmful to freedom of establishment than the immediate payment of the tax due…’ (paragraph 57) (my emphasis).[101]Deferred payment cannot mean that the member state is obliged to take into account losses which occur after the transfer of the place of management (paragraph 58). The legislation at issue did not provide for deferred payment. It ‘goes beyond what is necessary to achieve the objective’ in question, and was therefore ‘an unjustified restriction on freedom of establishment’ (paragraph 59). That conclusion was not affected by the fact that the gains were made ‘after the establishment of the amount of the tax, but before that tax became payable, given that the disproportionality of the legislation…is due to the fact that that legislation makes no provision for the taxpayer...to defer the time when the tax payable is paid’ (paragraph 60).

The period for which payment must be deferred

[102]In her Opinion in Panayi, Advocate General Kokott observed, in paragraph 2, that by 21 December 2016, there was ‘an extensive body of case-law on exit taxation’. The cases had ‘formulated certain requirements to which exit taxation is subject’.[103]She also explained that the United Kingdom government had argued that the charge was not disproportionate on the facts, because the Trustees had actually realised the value of the shares by selling them before the date when the tax was due to be paid. It argued that ‘since the profit from that sale was sufficient to pay the tax debt previously incurred on account of the deemed disposal, the exit tax is proportionate in this particular case’ (paragraph 54).[104]In language echoed by the terms of paragraph 57 of the judgment (see paragraph 100, above), she said, in paragraph 55, that the Court of Justice had ‘already held on a number of occasions that the taxable person must have the choice between immediate taxation and deferred payment, together with, if appropriate, interest in accordance with the applicable national legislation…’ (my emphasis). She referred to the judgments in DMC v Beteiligungsgesellschaft mbH v Finanzamt Hamburg-Mitte (C-164/12) [2014] STC 1345 (paragraph 64) and Verder LabTec GmbH & Co KG v Finanzamt Hilden (C-657/13) [2015] 3 CMLR 39. In the first case, the Court of Justice considered legislation which provided for the recovery of tax ‘over a period of five years rather than immediately’, and held that that was proportionate. In the second case, the legislation provided for instalments to be paid over ten years. Unsurprisingly, the Court of Justice held that that, too, was proportionate. The tax in this case was disproportionate, even though the shares were sold before the date when the tax was due. Whether or not the tax was disproportionate could not depend on the conduct of the particular taxpayer (paragraph 57). The tax was disproportionate despite the sale because ‘there was no option to defer payment at the time of assessment to tax. Such an option to defer does not necessarily have to amount to the five years referred to in…the case-law’ (my emphasis; footnote omitted). The legislation ‘could instead have linked the option to defer to realisation of hidden reserves within that five-year period’ (my emphasis) (paragraph 58).[105]Paragraph 60 of the Opinion is somewhat controversial. Judge Mosedale found it difficult to understand (see paragraph 76 of her determination). We have been provided with the German and French versions of paragraph 60 and with Google translations of those. I agree with Judge Mosedale that the official English translation is difficult to understand. The French version, if accurate, is at odds with the English text, as it suggests that a solution may be found in a conforming interpretation of domestic procedural tax legislation which provides for staged payments. What is puzzling, given their current stance, is that it was the Trustees who seem to have suggested in the proceedings in the Court of Justice that there was the potential for a relevant discretion to be exercised in the United Kingdom. It is not necessary to speculate further about what the Advocate General meant for two reasons. First, this cryptic paragraph is not reflected in the judgment. Second, the interpretation of domestic legislation is for the national courts; not for the Advocate General, and not for the Court of Justice.[4]The decision of the UT[106]The UT described the parties’ arguments in impressive detail (paragraphs 65-127 of determination 3). Its discussion of the issues starts at paragraph 128 of determination 3. I will summarise its principal conclusions. The UT held that the first question was to identify the nature of the incompatibility of domestic legislation with EU law. The court should then consider whether that incompatibility could be remedied by a conforming interpretation; and, if so, in what terms. If it cannot do so, the court should then decide whether to disapply the legislation, and the effect of such disapplication (paragraphs 128-129).[107]In paragraphs 130-134, it held that the imposition of an exit tax was not unlawful. The breach of EU law, rather, was that there was no option to defer liability to that tax. It relied on paragraphs 52 and 57 of the decision of the Court of Justice in Panayi CJ. It was not, therefore, section 80 of the TCGA which was unlawful, but section 59B of the TMA, as the F-tT had held in paragraph 92 of Panayi.The same analysis applied to the corporate tax provisions in Redevco (paragraph 134).[108]The next question, a question of domestic law, was whether the UT could find a conforming interpretation of the relevant legislation in each case (paragraph 135). The Court of Justice in Panayi had not considered whether a conforming interpretation was available, as that question was not referred to it. The Court of Justice had held, in other cases, that the payment of instalments over five, or even ten, years was compatible with EU law (paragraph 136).[109]The legal principles governing conforming interpretation were not in dispute. The argument was about their application to the facts (paragraph 137). The UT had to consider ‘whether a muscular approach to interpretation is able to remedy the breach of EU law. In adopting a conforming interpretation, we cannot go beyond what is necessary to remedy the breach (see Re S). The precise form of words does not matter, and we do not need to engage in legislative drafting’ (paragraph 140). The reference to ‘Re S’ is to Re S (Care[110]The UT described its conforming interpretation in paragraph 141 of determination 3. Sections 59B and 59D of the TMA should be read, ‘in cases where the taxpayer’s right of freedom of establishment would otherwise be infringed…to provide that, at a time before the legislation was actually amended to comply with EU law… as including an option to defer payment of the exit tax in five equal instalments, with the first instalment payable on the normal due date for payment, and at yearly intervals thereafter’. It gave effect to the taxpayers’ EU law rights, and gave the ‘best fit’ with the requirements of EU law (paragraph 142). The UT’s ‘choices’ were ‘resolved by the application of the principles which apply to a conforming construction’ (paragraph 142). In paragraphs 143-148, the UT listed some factors which in its view supported that approach.[111]In paragraph 149, the UT rejected various arguments that the fact or content of the later amending legislation showed that the UT’s interpretation either went against the grain of the legislation or showed that it was contra legem. The complexity of the scheme of the amending legislation was irrelevant, as a conforming interpretation did not require (or permit) such complexity (paragraph 150). In paragraphs 151-156, the UT distinguished several of the authorities on which the taxpayers relied.[112]The UT considered arguments about legitimate expectation and legal certainty in paragraph 157, essentially on the grounds that a conforming interpretation is by its nature retrospective. Any legitimate expectation would have to take into account the potential availability of such an interpretation and of its effect. The false premise of the legitimate aim argument was that it was certain that the courts of the United Kingdom would disapply the exit provisions in full. The F-tT had found that Redevco would have left the United Kingdom in any event. Redevco’s uncertainty about its legal position was no different from that of the taxpayer in V 2. The UT returned to this finding of fact in paragraph 161, noting that it was not challenged on Redevco’s appeal and, in any event, holding that it was correct. It also held that the advice which Redevco had been given ‘cannot improve their remedy for any breach of EU law, or give rise to any justiciable legitimate expectation’.[113]In paragraph 158, the UT answered the submission that a conforming interpretation did not vindicate the Trustees’ EU rights as it was now too late for them to make the choice which the conforming interpretation would have given them at the relevant time. The conforming interpretation did not lead to the retrospective imposition of any liability. The Trustees’ argument started ‘at the wrong place’. The retrospective nature of the conforming interpretation meant that the Trustees had always had the right to elect to defer payment. The UT was vindicating that right with its conforming interpretation of the legislation. The fact that the Trustees did not know that they had the right at the relevant time was ‘irrelevant’. That was clear from many authorities about the effect of the Principle VAT Directive.[114]In paragraphs 159-161, the UT summarised some of Redevco’s arguments. It had not pursued many of its challenges to the F-tT’s reasoning at the hearing. Mr Margolin had confirmed that it did not challenge the F-tT’s finding that it would have moved to the Netherlands irrespective of counsel’s advice.[115]Paragraphs 165-169 of determination 3 are headed ‘Disposition’. The UT said that the conforming interpretation by the F-tT in each case provided that there would be no liability to interest (otherwise than on late payments). That aspect of the determinations went ‘beyond what is required to give effect to’ the EU law rights of the Trustees and Redevco. The F-tT had erred in law in doing so. There are no statutory rights of appeal to the F-tT in respect of interest. Liability to interest, therefore, was not before the F-tT. The UT therefore set aside the F-tT’s determinations. The UT decided to remake the determinations. There was an available conforming interpretation of the relevant provisions. The UT read into them words ‘to provide that, at a time before the legislation was actually amended to comply with EU law, those provisions should be read in cases where a taxpayer’s right to freedom of establishment would otherwise be infringed as including an option to defer payment of the exit tax due in five equal annual instalments, with the first instalment payable on the normal due date for payment, and at yearly intervals thereafter’. The UT acknowledged that ‘interest is strictly outside the scope of our decision’ but made some observations about it, nevertheless, in paragraph 169.[5]Discussion (1).

(i) Panayi

[116]The Trustees have four grounds of appeal. 1. The UT’s conforming interpretation infringes the relevant principles that a conforming interpretation must nota. insert a novel statutory scheme in legislation,b. involve the court in policy decisions which are for Parliament,c. offend against the principle of legal certainty, ord. amend legislation in a way which has unforeseeable consequences. 2. The UT erred in law in holding that its conforming interpretation vindicated Panayi’s EU rights. Those rights required Panayi to have a choice to pay a lump sum or by instalments when it left the United Kingdom. That conforming interpretation irreparably deprives them of that choice. 3. The UT chose the wrong conforming interpretation. The right conforming interpretation required payment to be deferred until the realisation of the asset. 4. The UT was wrong to make a policy decision and to overturn the decision of the F-tT about interest. (ii). Redevco[117]Redevco’s grounds of appeal ascribe five broad errors of law to the UT. 1. Its conforming interpretation failed to give effect to the principles of legitimate expectation and legal certainty. 2. Its reasons for deciding that Redevco could not rely on the principle of legal certainty were wrong because it should not have asked what Redevco would have done if the legislation had been different and was wrong to conclude that Redevco’s decision could not have been affected by any change in the legislation. 3. The UT identified the wrong breach of EU law. 4. The UT was wrong to engage in judicial legislation. 5. The UT was wrong to set aside the F-tT’s decision about interest. (2). The issues (i). What is not in dispute?[118]Three points are common ground. 1. The F-tT was entitled to find, in Redevco, that Redevco would have moved to the Netherlands regardless of the actual legal position. 2. (As I have already said) the relevant provisions interfered with the right of freedom of establishment conferred on the Appellants by the TFEU (see paragraph 51 of determination 3). 3. Subject (perhaps) to the issue whether an interpretation is ‘contra legem’, the question whether or not a conforming interpretation is available is a question for the domestic courts. (ii). What is in dispute?[119]There are several sub-issues, which I will consider in the appropriate place. There are five main issues or potential issues. 1. Did the UT identify the wrong breach of EU law? 2. What relevant principles apply to the process of conforming interpretation? 3. Did the UT infringe any of those principles? 4. Are there distinct principles of EU law which(a) entitled the Trustees, when they moved to Cyprus, to have a choice of when to pay the tax, or(b) required the charge to tax to be deferred until the sale of the shares? 5. If the F-tT erred in law in its conforming interpretation of the relevant provisions, can HMRC achieve the same result by a partial disapplication of the relevant provisions? This issue is raised by HMRC’s Respondent’s Notice (‘RN’). 6. What should this court do about interest? I will consider issues 2 and 3 together.[1]Did the UT identify the wrong breach of EU law?[120]Redevco argues that the relevant breach of EU law was the charge to tax and that the possibility of a deferred payment regime is only relevant to justification. I reject that submission. It is based on an incomplete description of the relevant breach of EU law. The decisions of the Court of Justice which are summarised in Panayi CJ show that an exit tax is an interference with freedom of establishment, but that it is capable of being justified by the public interest in the allocation of taxation powers between member states. That interference is not contrary to EU law if it goes no further than is necessary to meet that public interest. If there is adequate provision for deferred payment, any such restriction will be proportionate. In order to conform with EU law, an exit charge must therefore be accompanied by provisions which enable payment to be deferred. The relevant breach of EU law is the exit charge and the absence of provisions for deferred payment. The F-tT and the UT therefore identified the relevant breach of EU law correctly.[2]and 3. What principles apply to the process of conforming interpretation? Did the UT’s conforming interpretation breach those principles?[121]The taxpayers argued, variously, that the conforming interpretation of the UT infringed a legitimate expectation and/or the principle of legal certainty and/or the contra legem principle, and amounted to judicial legislation.[122]The first two objections, when analysed, are complaints about the effect of the declaratory theory. The reasoning of Falk LJ in Prudential (see, in particular, paragraphs 74-82, above) is a complete answer to these complaints. The effect of the declaratory theory is that (if the UT’s conforming interpretation is otherwise appropriate), it declares the relevant law about exit charges as it has always been. It does not matter what the taxpayers thought at the time, although it is clear that both (rightly) considered that the exit charge was contrary to EU law. We do not have a full picture about the legal advice which the Trustees or Redevco had been given when they decided to cease to be resident in the United Kingdom. I am nevertheless prepared to accept that they might not have anticipated that, many years later, a conforming interpretation would defeat their hope that they would not have to pay any exit charges. Criticisms based on legitimate expectation or legal certainty are inapt. They do not make the UT’s conforming interpretation wrong. Any such interpretation is inevitably retrospective. For that reason, I do not consider that the F-tT’s or the UT’s conforming interpretation is somehow invalidated by Parliament’s decision not to make its later amendments to the legislation retrospective.[123]Mr Johnston relied on the last two sentences of paragraph 106 of the judgment of Lewison LJ (with whom the other members of this court agreed) in Revenue and Customs Commissioners v Ampleaward Ltd [2021] EWCA Civ 1459; [2021] STC 2260. He ‘note[d]’ that when Parliament eventually amended the legislation, that legislation was not retrospective. He added:
‘The suggested conforming interpretation would have retrospective effect, although Parliament had expressly decided that it should not’
. The ground on which this court dismissed HMRC’s appeal in that case was that HMRC’s conforming interpretation directly contradicted the words of the relevant legislation (see paragraph 106) and that that interpretation did not remedy an error in the domestic legislation (paragraph 109). The passage on which Mr Johnston relies is an obiter comment and does not bind this court. I am not, in any event persuaded by it, for three reasons. First, legislation and construction are different acts done by different bodies. Second, the legislation which the tribunals were construing pre-dates the amending legislation. The meaning and effect of that earlier legislation cannot logically be affected by Parliament’s later legislative choice. Third, there is a recognised presumption against retrospective legislation, but the effect of decisions of the courts is necessarily retrospective. That effect is not in itself a reason for not adopting a conforming interpretation.[124]For the reasons given by Sir Colin Rimer in Lock (see paragraphs 71 and 72, above), the contra legem principle does not prevent a domestic court from adopting a conforming interpretation which adds to domestic provisions the language of which does not positively point against such an insertion. There is nothing in the relevant provisions which is antagonistic to the addition of words which permit the taxpayer to discharge his liability in five annual instalments. For that reason, this case is nothing like Fleming. A further difference between this case and Fleming is that this case does not concern an accrued right, or anything resembling an accrued right.[125]The conforming interpretation does not amount to judicial legislation for four broad reasons.[126]First, while it was in force, section 2(1) of the ECA was a clear statement of Parliament’s general intention that Treaty rights should be ‘recognised and available in law, and be enforced, allowed and followed accordingly’. Section 2(4), while it was in force, was a clear statement of Parliament’s general intention about the construction of all statutes, whether passed before or after the commencement of section 2(4). Its intention was that all legislation should be construed and have effect subject to section 2(1).[127]Second, the authorities on EU law and on section 3 both show that in an appropriate case, a conforming interpretation may require a court to add words, sentences, or whole provisions to the legislation in question. I accept Mr Elliott’s submission that the authorities define the boundaries of ‘possible’ interpretation. The authorities show that the duty to give legislation a conforming interpretation may in an appropriate case enable a court to insert missing words in a provision. Pickstone and Litster are early and good examples of such cases. In Litster the domestic regulations implementing a directive, as it had been interpreted by the Court of Justice, did not give it its full effect in EU law. In Pickstone the House of Lords inserted words in primary legislation. In Litster the House of Lords had no difficulty in making extensive additions to the words of the relevant regulations in order to make those regulations conform with EU law. In V 2 this court inserted a further exception in a list of exceptions in primary legislation.[128]Third, the breach of EU law here is a gap in the legislation: the legislation did not provide for payment by instalments. The starting point for considering whether the construction ‘goes with the grain’ of the legislation is a proper understanding of that metaphor in this particular context; the court is not concerned with tiny concordances or inconsistencies (see paragraph 84, above). There are two further points. First, the relevant legislation for this purpose includes the coercive effects of sections 2(1) and 2(4) of the ECA. Second, the conforming interpretation does go with the grain of the legislation in two senses. The general thrust of the legislation is to impose a charge to tax. Giving the taxpayer more time to pay tax which is due does not contradict, frustrate, or go against, that thrust. There is no express prohibition of such a payment method in the legislation, so the insertion of such a provision is not inconsistent with a fundamental feature of the legislation; and section 281 of the TCGA already permitted the payment of CGT on gifts in ten annual instalments.[129]Fourth, the Court of Justice does not establish the minimum which is required to comply with EU law. It can only decide the questions of EU law which happen to be referred to it by national courts. The minimum required for such compliance is a question for a national court. The UT’s conforming interpretation went no further than is necessary to comply with what, the relevant decisions of the Court of Justice show, are the relevant requirements of EU law. Both Mr Margolin and Mr Johnston relied on the legislation which Parliament later enacted to contend that the UT had engaged in legislation. Mr Johnston even went as far as to submit that the UT had decided that ‘the detailed statutory scheme in Schedule 7 (first established in 2019) could be read back into the legislation as it applied in 2004’ (skeleton argument, paragraph 6). The actual legislation shows the opposite. It is much more detailed than the UT’s conforming interpretation and in it, Parliament has made choices which it was not open to the UT to make, and which the UT did not make. The UT, by contrast, did not make any choices. It did not ‘read back’ Schedule 7 into the legislation as it stood in 2004. It merely did the bare minimum which was required to comply with the requirements as they have emerged from decisions of the Court of Justice; that is to provide for payment to be deferred in five annual instalments.[4]a. Does EU law require that the taxpayers have a choice when they decide to move to another member state?[130]Mr Johnston relied on various passages in the EU authorities to argue that EU law requires that the taxpayer is able to make a conscious choice at the point when he decides to move to another member state whether to discharge his liability at once or in instalments. I have italicised the relevant words (‘choose’, ‘option’, ‘choice’) in the Opinion of the Advocate General and in the judgment in Panayi CJ (see paragraphs 93-105, above) as examples. The tribunals’ conforming interpretations (both of which used the word ‘option’) were contrary to EU law, he contended, because they were imposed ex post facto, with the result that the Trustees never had the choice which was their right. I disagree. The retrospective effect of the declaratory theory means that we now know that the Trustees did, in law, have the option of paying by instalments when they decided to move to Cyprus. That they could not exercise that option, in fact, is a consequence of the time which has elapsed since then as this dispute has clarified the relevant law (cf Prudential). It is not contrary to any principle of EU law.[4]b. Does EU law require the taxpayer to have an option to pay the tax due when the assets are realised and/or did the UT make a policy choice by rejecting that option?[131]The premise of the reasoning in Panayi CJ is that it is inherent in an exit charge that the assets in question had not been realised when it was imposed. The cases show that an exit charge incurred before any assets have been realised is compatible with EU law if provision is made for payment of the charge to be deferred and made by instalments. Mr Johnston’s argument was that deferring payment until the realisation of the assets involved ‘the least violence’ to the domestic legislation: section 80 could simply be amended. For this purpose, the actual facts of Panayi are to be ignored (see footnote 5 of his skeleton argument). Mr Johnston checked during the hearing, and he told us that there is no decision of the Court of Justice which suggests that a failure to provide for payment of an exit charge to be deferred until the realisation of the assets is a breach of EU law. That may be a simple consequence of the legislative schemes which have been considered by the Court of Justice in this area. But the fact remains that there is no steer to this effect from the Court of Justice. The signs from Luxembourg all point in the same direction. I reject the argument that the tribunals made a choice of any kind, still less an impermissible policy choice, by following those signs, and by not adopting this suggested solution instead. Indeed had the tribunals chosen it, they would have been vulnerable to Mr Johnston’s criticism that it had made a policy choice of the kind which only Parliament can make.[5]Partial disapplication?[132]I have described the relevant reasoning of Judge Mosedale in paragraphs 20-22, above. She and Henderson J described a temporary, rather than a static, disapplication of the incompatible provision. They did not therefore disapply the provision to the extent that it was incompatible by excising the whole provision, or of any of its words but held, rather, that it should be held in abeyance (or disapplied) until a date which, on the facts, respected the taxpayer’s EU law right to defer payment of the tax due. I do not have to decide this issue, and do not do so. While I can see the attraction of this approach, it is, nevertheless, contrary to the obiter view of this court in V 2 (see paragraph 67, above).[6]Interest

Judge Mosedale’s conforming interpretation (paragraph 161:

[133]see paragraph 19, above) referred to interest. I do not read the reference to interest as being part of her conforming interpretation: she was simply making clear, for the avoidance of doubt, that the conforming interpretation did not deal with interest because that was already catered for elsewhere in the legislation. The phrase ‘without liability to interest’ is best understood in that light. The conforming interpretation adopted by Judge Brooks did not deal with interest at all. I consider that may be because, as I have done, he understood that Judge Mosedale’s conforming interpretation, with which he agreed, did not relate to interest.[134]I therefore consider that the UT was wrong to say, in paragraph 165, that the conforming interpretation of the F-tT in both appeals ‘provided that there would be no liability to interest (other than interest on late payments)’, for two reasons. First, only Judge Mosedale referred to interest, and, second, properly understood, her reference to interest was not part of her conforming interpretation but was an aside, confirming that liability to interest was dealt with elsewhere in the legislation.[135]The consequence of this is that the UT had no power to remake either decision or to substitute its conforming interpretation for that of the F-tT in either case. Redevco relies on this point in its skeleton argument. The Trustees do not. Nor, finally, did the UT have any power to make any binding decision about interest.

Conclusion

[136]I would dismiss both appeals except to the extent that they challenged the UT’s decision on interest. I would allow Redevco’s appeal to the extent that I would restore the conforming interpretation of the F-tT in Redevco. The UT’s conforming interpretation in Panayi stands. It is not necessary for me to decide the cross-appeal and I do not do so.[137]I agree.[138]I also agree.

Order

Implementation of Care Plan) [2002] UKHL 10; [2002] 2 AC 291.