APVCO 19 Ltd & Ors, R (on the application of) v HM Treasury & Anor

The primary issue in this case is whether certain retrospective tax legislation should be declared to be incompatible with the appellants’ rights under the European Convention on Human Rights (the “ECHR”). The rights in question are those relating to the protection of property under article 1 of protocol 1 (“A1P1”) and to a fair trial under article 6 (“article 6”) of the ECHR. The appellants have sought to take advantage of what have been described as aggressive tax avoidance schemes designed to allow them to escape the payment of Stamp Duty Land Tax (“SDLT”) on their purchases of (mostly) residential property for their own use (the “appellants’ schemes”). The appellants’ schemes were designed and marketed by Blackfriars Tax Solutions LLP (“Blackfriars”). The question here is not whether the appellants’ schemes would have actually worked, though that is hotly contested, but simply whether the retrospective legislation targeting them violated the ECHR.

In substance, there are appeals from two decisions of Mrs Justice Andrews. In the first, she refused the appellants permission to cross examine certain representatives of the respondents. In the second, she refused the appellants permission to bring judicial review proceedings against Her Majesty’s Treasury (“HMT”) and the Commissioners for her Majesty’s Revenue and Customs (“HMRC”) (together the “respondents”). The appellants are a number of persons affected by retrospectively effective legislative changes to section 45 of the Finance Act 2003 (“section 45”) relating to SDLT made by section 194(1)(a) and 194(2) of the Finance Act 2013 (the “legislative changes”). They contend, as I have said, that the legislative changes infringe A1P1 and article 6, and reserve the right to argue at a later stage that article 14 of the ECHR was also infringed. They seek a declaration of incompatibility as to the legislative changes under section 4 of the Human Rights Act 1998.

On 2^nd September 2014, Gloster LJ considered the main appeal on paper and granted the appellants permission to apply for judicial review, directing that there were special reasons why the application should be retained by the Court of Appeal.

When the detail is stripped away, the case is really quite straightforward. It is common ground that the challenged legislative changes put beyond doubt what it is also common ground was previously not beyond doubt, namely that the appellants’ schemes did not work. The appellants’ schemes have not yet been challenged in the First-tier Tribunal (Tax and Chancery Chamber) (the “FTT”) and may never be so challenged unless this appeal is successful. It is accepted on both sides that it is not appropriate for us to usurp the FTT’s function in deciding whether the scheme was actually effective in the first place. We are to confine ourselves to deciding whether the judge was right to hold: (i) that A1P1 was not engaged at all in this case because the legislative changes had not deprived the appellant of any “possessions” as that term is properly to be construed in A1P1, (ii) that, even if A1P1 were engaged, the legislative changes were lawful because they were neither arbitrary nor unforeseeable, and were proportionate, and (iii) that the legislative changes were also not a breach of article 6 of the ECHR, which was not even arguably engaged in this case.

The detailed background of the scheme and the legislation is very well set out by the judge in paragraphs 7-41 of her clear judgment. I do not intend to repeat that exposition here. In reality, the detail is not all that important because the effectiveness of the scheme itself is not what is now in issue. Instead, Mr Jeremy Woolf, counsel for the appellants, has focused his attack on the meaning of the term “possessions” in A1P1, and on the lawfulness and proportionality of the retrospective legislative changes. He has placed primary reliance on the terms and effect of the Government’s protocol of March 2011 on unscheduled announcements in tax law entitled “Tackling Tax Avoidance” (the “Protocol”), contending that the legislative changes were outside that Protocol in that the situation was not “wholly exceptional”, there was no “significant risk to the Exchequer”, and there was no proper apprehension of “significant losses to the Exchequer”.

In these circumstances, the non-exhaustive summary of the factual and legal background which follows will concentrate on those matters that underpin the appellants’ main arguments.

The Finance Act 2003 introduced SDLT in place of the old stamp duty. Section 42 provided for SDLT to be payable on “land transactions”, which were defined by section 43. Section 44 then defined when a land transaction was to be treated as having been entered into, dealing in particular with the distinction between contract and conveyance. The Finance Act 2003 aimed to place the burden of SDLT on the person who was to acquire the use and enjoyment of the property in question, and to reduce that burden on those with only a transient interest in the property. These arrangements were set up by section 45 which applied where “(a) a contract for a land transaction (“the original contract”) is entered into under which the transaction is to be completed by a conveyance” and “(b) there is an assignment, subsale, or other transaction … as a result of which a person other than the original purchaser becomes entitled to call for a conveyance to him”, referred to as a “transfer of rights”. Section 45(2) provided that “the transferee is not regarded as entering into a land transaction by reason of the transfer of rights, but section 44 … has effect in accordance with the following provisions of this section”. Section 45(3) then provided (originally) for section 44 to apply “as if there were a contract for a land transaction (a “secondary contract”) under which” the transferee was the purchaser and the consideration was, in effect, that paid for that part of the property under the original contract and that paid for the transfer of rights. The key provision of section 45(3) then said that “[t]he substantial performance or completion of the original contract at the same time as, and in connection with, the substantial performance or completion of the secondary contract shall be disregarded”. As it seems to me, this provision thereby took the completion of the original contract in such circumstances out of the charge to tax under section 42.

After 2003, these provisions became a target for tax avoidance schemes in relation mainly to residential property transactions. HMRC sought to combat them by a variety of methods including the issue of warning bulletins and fighting tax appeals.

Since the appellants place such reliance on the Protocol on tackling tax avoidance, it is necessary to set out some of its provisions. It was said in the executive summary to provide criteria that ministers would observe “when deciding whether to announce a change to tax law that has immediate effect”, and to allow decisive action when risks to the Exchequer were identified. It included the following:-

“The Government has made clear its aim to strike the right balance between restoring the UK tax system's reputation for predictability, stability and simplicity and preserving the ability to protect the Exchequer by making changes where necessary. In particular, changes to tax legislation where the change takes effect from a date earlier than the date of announcement will be wholly exceptional.

1.

Ministers undertake to observe the following criteria when considering a change to tax law which will:

•

be announced other than at Budget; and

•

take effect before the legislation implementing the change is enacted.

2.

Such changes to tax law will normally only be announced other than at Budget where:

•

there would otherwise be a significant risk to the Exchequer;

•

significant new information has emerged to identify the risk or indicate its scale; and

•

changing the law immediately is expected to prevent significant losses to the Exchequer. …”

Up to 2012, a common attempted but unsuccessful tax avoidance scheme involved a normal contract for sale at full market value, where, on completion, the buyer executed a deed granting a connected third party a call option in 35 years’ time for a sum lower than the SDLT threshold (which was never intended to be exercised) (the “option scheme”). The simultaneous grant of the call option was said to be a qualifying “transfer of rights” within section 45(1)(b) of the Finance Act 2003. Even though the option was below the SDLT threshold, the parties agree that the scheme did not work because, to avoid SDLT, the two transactions had to be completed or substantially performed at the same time, and the option was not completed at all.

A theme of the Chancellor of the Exchequer’s 2012 budget was the reduction of aggressive tax avoidance, which he described as “morally repugnant”. It announced a consultation concerning an intended general anti-abuse rule extending to SDLT and said that the Government would “take action to close down future SDLT avoidance schemes, with effect from 21 March 2012, where appropriate”. It also promised legislation to make clear that the option scheme did not work. Such legislation was introduced by the Finance Act 2012 by adding a new section 45(1A) of the Finance Act 2003 providing that the reference in section 45(1)(b) to “an assignment, subsale or other transaction” did not include the grant or assignment of an option. The Chancellor said expressly that he would not hesitate to move swiftly without notice and retrospectively if inappropriate ways around the new SDLT rules were found. The budget announcement of the same date said that the Government would “take action to close down future SDLT avoidance schemes, with effect from 21 March 2012 where appropriate”. A further consultation resulted in a general anti-abuse rule introduced by part 5 and schedule 39 of the Finance Act 2013 which is agreed to make the appellants’ scheme ineffective from the date on which the rule was introduced in 2013.

The appellants’ scheme was adopted by them around the time of 2012 legislation (there is a dispute as to precisely when). It was similar to the option scheme, save that, after exchange of the original contracts, the buyer entered into an agreement with a connected third party to grant that person an option to purchase the property on the date the original contract completed. The agreement was expressly stated not to be specifically enforceable, but upon the original contract completing, an option exercisable long into the future at a price just over the SDLT threshold was indeed granted to the third party, thus allegedly “substantially completing” the option agreement. The judge thought that, once again, the parties never intended that this option would be exercised (though that aspect is apparently also disputed).

The 2013 budget announced specific retrospective provisions to outlaw two similar, but not identical, “deferred completion” schemes with effect from 21^st March 2012. The appellants’ scheme was only disclosed expressly and formally to HMRC under the disclosure of tax avoidance schemes rules (“DOTAS”) on 22^nd April 2013.

On 17^th May 2013, Ms Jane Ewart, an HMRC officer working in the Corporation Tax, International and Stamps Directorate (“Ms Ewart”), wrote a note to the Minister recommending closing down the scheme, which was the one used by the appellants and 4 other promoters. Ms Ewart’s note referred to the Protocol and identified factors that she thought made this an exceptional case justifying retrospective legislation including repeated abuses in SDLT over a number of years, a clear warning at budget 2012 that this was unacceptable and that, if it continued, the Government would consider retrospective legislation, and the fact that it should have been obvious to users and promoters that “it [the appellants’ scheme] pushed on or beyond the boundaries of abusiveness and that the Government was likely to take further action”. A ministerial announcement followed on 4^th June 2013 indicating the intention to add a retrospective amendment to section 45(1A) of the Finance Act 2003 to take effect from 21^st March 2012. Reliance was placed throughout the process on the Chancellor’s clear warnings in the 2012 budget statement. The legislative changes duly introduced this amendment by adding the italicised words to section 45(1A) as follows: “The reference in subsection (1)(b) to an assignment, subsale or other transaction does not include the grant or assignment of an option or an agreement for the future grant or assignment of an option”. It is accepted as I have said that, subject to A1P1 and article 6, the legislative changes are effective to render the appellants’ scheme ineffective.

On 18^th June 2013, Mr David Gauke MP, Exchequer Secretary to the Treasury, explained to Parliament why the Treasury viewed the legislative changes as wholly exceptional and consistent with the Protocol. He said that there “is a history of abuse in this area of the legislation. Indeed, we have already made that clear. The policy intent has been made clear. The Chancellor gave a clear warning at Budget 2012 that he would not hesitate to use retrospective legislation if abuse of the SDLT rules continued”. He said that HMRC estimated the tax at risk over the next 5 years to be around £160 million, a not insignificant amount, and that the litigation would take a considerable length of time.

The legislative changes duly came into force having retrospective effect.

A1P1 provides as follows:-

“Every natural or legal person is entitled to the peaceful enjoyment of his possessions. No one shall be deprived of his possessions except in the public interest and subject to the conditions provided for by law and by the general principles of international law.

The preceding provisions shall not, however, in any way impair the right of a State to enforce such laws as it deems necessary ... to secure the payment of taxes ...”

Article 6 provides as follows:-

“In the determination of his civil rights and obligations or of any criminal charge against him, everyone is entitled to a fair and public hearing within a reasonable time by an independent and impartial tribunal established by law…”

The judge ruled that A1P1 was not engaged at all in this case. She said that the legislative changes did not impose a liability on the appellants to pay SDLT. They retrospectively remove an alleged, genuine, but not established, right to tax relief asserted in the appellants’ SDLT tax returns. Such a claim to a tax relief is not a “possession” within A1P1, because Strasbourg authority requires possessions to exist or to be claims in respect of which an individual has a legitimate expectation that they will be realised. Such a legitimate expectation must be based on a legal provision or a legal act (such as a judicial decision) and cannot be based on an arguable claim or a genuine dispute (see the decision of the European Court of Human Rights (“ECtHR”) in Kopecky v. Slovakia (No. 44912/98) (2005) 41 EHRR 43 at paragraph 52, and the Court of Appeal in R (on the application of Huitson) v. Revenue and Customs Commissioners [2011] STC 1860 per Mummery LJ at paragraphs 68-9, to which I shall refer as “Huitson CA”. The judge rejected Mr Woolf’s argument that the relevant possession was the money of which the appellants would be deprived by payment of the tax. She said that the legislative changes did not impose a liability to pay tax, because that suggestion was premised on the false proposition that, without the legislative changes, the appellants were entitled to keep the money. They were not. They simply had an argument that they might be so permitted. That legal argument, whatever its merits, was not a possession within the meaning of A1P1.

The judge declined to decide the underlying tax issue, but pointed out that the agreement with the third party under the appellants’ scheme gave that third party no right to call for any conveyance, which severely undermined the argument that there had been any “transfer of rights” in consequence of the completion or substantial performance of the intermediate transaction.

The judge then considered whether A1P1 would have assisted the appellants even if it had been engaged. The judge held that any interference with the peaceful enjoyment of possessions must be both lawful and proportionate (see Lord Reid at paragraphs 116-124 in AXA General Insurance Limited v. HM Advocate [2012] 1 AC 868). The factors are separate and cumulative. Lawfulness required compatibility with the rule of law, involving legal certainty. The interference must not operate in any arbitrary manner, but it will not be arbitrary if it is founded on necessity, reason or principle, even if it involves an exercise of discretion. It was common ground that the fact that the legislative changes were retrospective did not, of itself, make them incompatible with A1P1; retrospectivity has generally been considered when evaluating the requirement of proportionality. In relation to proportionality, where payment of taxes is concerned, the national authority must strike a fair balance between the demands of the general interest of the community and the requirements of the individual’s fundamental rights, but it enjoys a “wide margin of appreciation” (see National & Provincial Building Society v. United Kingdom [1997] STC 1466 at paragraphs 80-82).

The judge roundly rejected Mr Woolf’s primary argument that the legislative changes were arbitrary, insufficiently foreseeable, capricious and, therefore, unlawful. She relied on the Chancellor’s various statements, to which I have referred, and dismissed the suggestion that there was a violation of the Protocol. The judge said that the amount of tax lost was irrelevant to the objective which was to put paid to all such schemes and to ensure that the transfer of rights rules achieved the outcome for which they were originally intended. The appellants’ scheme was just one more example falling within the generic category of attempts to manipulate the transfer of rights rules to produce the opposite of the intended effect. The fact that other non-SDLT schemes had not been similarly treated did not make the legislative changes arbitrary or capricious, because two wrongs did not make a right. The reliance on the Protocol was opportunistic in that it would not have been available to the appellants if HMRC had become aware of the appellants’ scheme earlier. The Protocol did not anyway act as a straitjacket or as a fetter upon Parliament. It would be absurd to castigate as unlawful measures taken swiftly in response to a discovered variation to a type of scheme that was already the subject of anti-avoidance legislation simply on the basis that the tax lost was too small. In any event, this was an exceptional situation within the Protocol.

The judge then also rejected the argument that the legislative changes were disproportionate saying that they came nowhere near meeting the very high hurdle set in A1P1 cases. In the light of the clear warnings, the appellants had no legitimate expectation that they would be able to buy high value property and pay only a fraction of the ordinary SDLT that was due. The legislative changes achieved an important and legitimate aim of UK public policy to ensure that everybody buying property paid their fair share of SDLT.

Finally, the judge rejected the arguments under article 6 on broadly the same basis as she had rejected the arguments on A1P1. She held that, on the basis of the Strasbourg jurisprudence, article 6 was not even arguably engaged in the present case, since tax claims were not “civil” (see R(ToTel Ltd) v. First Tier Tribunal (Tax Chamber) [2011] EWHC 652 (Admin)). She thought that, even if article 6 had been engaged, there were considerable difficulties with the argument that retrospective legislation that satisfies the requirements of A1P1 could nevertheless be struck down as incompatible with article 6 because of pending litigation and the higher threshold applicable under article 6 (see Lord Brown at paragraph 80 in AXA supra, and paragraphs 105-113 in National & Provincial Building Societysupra). Even if article 6 were engaged, there was little difficulty in reaching the conclusion that the legislation easily satisfied the higher test of compelling grounds in the public interest.

On the first day of the hearing before the judge, the appellants applied to cross examine the respondents’ witnesses. The judge referred in her judgment on that point to the cases that establish that cross examination is exceptional in judicial review proceedings (e.g. Stanley Burnton LJ at paragraph 14 in Bancoult v. Secretary of State for Foreign and Commonwealth Affairs [2012] EWHC 2115 (Admin)). She held that cross examination as to Government legislative selectivity was not necessary to determine the claim fairly and justly. The judge said that the issue she had to determine was a legal one and the witnesses spoke only to the factual background. The wish to cross examine was a fishing exercise, and the question of selectivity was different from that of arbitrariness.

In addition to appealing the judge’s refusal to allow cross examination, the appellants also applied to us to allow cross examination during the hearing in the Court of Appeal. We dealt at the beginning of the hearing with the appellants’ fresh application to the Court of Appeal to cross examine Ms Ewart, because, had we allowed it, it would have been necessary to make arrangements for the cross examination to take place during the hearing of the appeal. In the event, we dismissed the application, saying that we would give our reasons in our judgments.

Mr Woolf argued that cross examination ought to have been allowed both before the judge and before us as to why the legislative changes were the only retrospective SDLT legislation apart from those concerning the deferred completion schemes, even though at least one other SDLT scheme had been disclosed to HMRC. He said that this went to the question of whether there was anything “wholly exceptional” that justified retrospective legislation in this case under the Protocol, and to whether the legislative changes were arbitrary or proportional.

In relation to the question of whether the legislative changes deprived the appellants of a possession, the appellants’ argument is technical. They say that section 45 prevents any SDLT liability arising. It is not granting relief from tax as the judge said, because it is not dependent on any claim for relief. Even if the claim that tax is not due because of the application of section 45 is not a possession, the money needed to pay the tax certainly is a possession. It is wrong to regard the possession in question as the debateable defence to the claim by HMRC rather than the money that is involved. Moreover, the judge was wrong to say that the appellants are not challenging the charging provisions. They are challenging a combination of the charging provisions and section 45 as amended by the legislative changes. Looked at as a whole, these provisions interfere with the appellants’ possessions. The cases relied upon by the judge are all distinguishable, because none of them addressed the question of whether the obligation to pay out money in tax was properly to be regarded as a possession within A1P1.

In relation to the lawfulness of the legislative changes, the appellants argued that they breached the terms of the Protocol and were not, therefore, foreseeable, and that they were arbitrary in that the justifications relied on by the respondents applied equally to many other tax avoidance arrangements in the employment and SDLT contexts. The situation that gave rise to the legislative changes was not “wholly exceptional”. Moreover paragraph 2 of the Protocol supra applied also by analogy, because it was directed at changes that take effect immediately on the announcement, and these changes were retrospective to a period even before the announcement. Paragraph 2 of the Protocol was not satisfied here because of the small size of the tax lost, the numerous other tax avoidance schemes that were not tackled, and the lack of any significant risk or significant losses to the Exchequer.

In relation to proportionality, the appellants submitted that the judge confused the concept of legitimacy with the concept of proportionality, and that selectivity is indeed relevant to proportionality. The appellants submit that they did have a legitimate expectation that legislation would not be introduced in breach of the Protocol, so the legislative changes were disproportionate. Such an issue did not arise in the cases upon which the judge relied.

In relation to the application of article 6, the appellants submit that the case did concern their “civil rights and obligations”, because debt claims are civil claims as a matter of English law. Neither the judge nor the case of R (ToTel) Ltd supra gave any consideration to that question. The judge was wrong to consider that there were compelling grounds for the legislative changes.

The appellants wanted to cross examine to obtain more context to the government’s decision to introduce the legislative changes and in order to ask about other areas in which loopholes had not been closed despite warnings that retrospective legislation was likely. They pointed particularly to a similar warning given in 2004 in relation to tax avoidance schemes concerning employment that had remained outstanding since then without retrospective legislation being brought forward, despite such schemes having been deployed.

As has been often emphasised, cross examination in judicial review applications is exceptional, and cross examination in the Court of Appeal is even more exceptional. In my judgment, nothing in the situation presented by this case demanded cross examination either before us or before the judge. It has always been clear that the appellants’ scheme was not the only SDLT avoidance scheme that was available. Mr Woolf’s main argument was that it was arbitrary for the government to close the loophole relating to the appellants’ scheme without closing all other loopholes, and that selectivity was a significant part of arbitrariness. This argument was not likely to be advanced by cross examination; indeed, cross examination would have been as likely to elicit points that supported the government’s reasoning. But either way, the facts disclosed by the written evidence were in my judgment sufficient to form the basis of the court’s decision.

Moreover, Ms Ewart would probably not have been able to speak to all other situations faced by HMRC, let alone in employment matters. What needs to be justified here is the use of retrospective legislation in this case. If the appellants are right that one loophole cannot be closed retrospectively without closing others as well, they will win this appeal without cross examination. Evidence is not needed just to give additional context or background in a case of this kind, where the facts are clear. Whilst other schemes might in theory be relevant to the question of whether the situation here was “wholly exceptional” within the Protocol, that was not the focus of the respondents’ reasoning. The respondents argued that the situation was wholly exceptional, because there had been clear advance warnings given, there was a long history of abuse, and people needed to be made to understand that SDLT had to be paid by all those buying properties for their own use - and the government would be as good as its word. Cross examination about other schemes and the actions taken or not taken in respect of them could not affect the respondents’ argument on the Protocol.

I can deal briefly with the appeal against the judge’s decision to refuse cross examination. She made a case management decision in the context of her rolled up hearing, which seems to me to be unimpeachable. Indeed for the reasons I have given, I would have made the same decision. No error of law or principle has been identified in her reasoning. In my judgment, cross examination of the respondents’ witnesses on the measures taken in relation to other SDLT schemes is and was wholly inappropriate. I would therefore dismiss that part of the appeal.

At the time of the legislative changes, the appellants’ transactions had already completed, so that there was a crystallised argument between them and HMRC as to whether SDLT was payable on each of their original sale transactions.

The legislative changes then amended section 45 so as to make it clear that the option arrangements entered into by the appellants had not constituted a “transfer of rights” under that section, so that the carve out in sections 45(2) and (3) had not operated to prevent the original transferees being regarded as if they had entered into taxable land transactions.

The question is whether that change had the effect of depriving the appellant transferees of any possession that they had as at the date of the legislative changes. There is, as the judge explained, clear authority that to deprive a person of an argument as to whether tax is or is not payable is not to deprive him of a possession for the purposes of A1P1. But the appellants here argue that they were deprived of the tax that they will have to pay as a result of the legislative changes being made.

Before answering this question, it is necessary to look more closely at what the authorities actually decide, since the parties’ arguments diverged widely on this point. It is clear from a number of cases, as Mr Woolf submitted, that unpaid tax can properly be regarded as a possession within the meaning of A1P1 (see, for example, paragraph 59 of the decision of the ECtHR in Burden v. United Kingdom [2008] STC 1305, paragraph 26 of Lord Hope’s speech and paragraph 114 of Lord Reed’s speech in AXA supra, and paragraphs 872-873 of the ECtHR’s decision in Khodorkovskiy v. Russia (2014) 59 E.H.R.R. 7).

The parties disagreed over the proper interpretation of Huitson CA, where the Court of Appeal rejected a claim that retrospective legislation in relation to double taxation of trading profits was an infringement of A1P1. The position in that case was that, before the retrospective legislation, a UK taxpayer was arguably entitled under section 788(6) of the Income and Corporation Taxes Act 1988 (“ICTA”) to relief from UK taxation that would otherwise be levied. The taxpayer had made the claim for relief which had not been processed, but had not paid the tax, before the retrospective legislation was enacted, which put it beyond doubt that he was not entitled to the relief. In the Court of Appeal, the case was put partly on the basis that the possession for which protection was sought under A1P1 was the claim for tax relief (paragraph 6 of Mummery LJ’s judgment, with which Sullivan and Tomlinson LJJ agreed), though that was hardly the central focus of the case. It was submitted that the taxpayer had a proprietary interest in a sufficiently established claim to tax relief to give rise to a legitimate expectation that would attract the protection of the ECHR, and that the taxpayer had been deprived of his possession in the form of the alleged proprietary interest in the nature of his claim for tax relief (see paragraphs 62 and 68 of Mummery LJ’s judgment). In fact, however, counsel for the taxpayer also argued that the ECHR looked, not only at the removal of relief as a possession, but also at the “effective financial burden imposed by the tax” (see paragraph 50 of Mummery LJ’s judgment).

Mummery LJ in Huitson CA did not, however, specifically differentiate between (i) the questions of whether A1P1 applied in the first place, and if so, to which possessions, and (ii) the question of whether the judge in that case had correctly applied the principle that, in securing the payment of taxes, a national authority must strike a fair balance between the general interests of the community and the protection of the individual’s fundamental rights, including the right to possessions under A1P1 (see paragraphs 28-29 of Mummery LJ’s judgment). Indeed, at paragraphs 44-45, Mummery LJ acknowledged, in dividing the arguments into 4 heads, that it was difficult to keep the grounds of appeal within any one heading.

Under his first heading of “proportionality and policy factors”, Mummery LJ decided at paragraph 57 that the judge had correctly directed himself on the issue of fair balance and proportionality. He seems to have done so on the assumption, without deciding at that stage, that there was a possession that engaged A1P1. Under Mummery LJ’s second heading of “tax efficiency of the scheme and legitimate expectation”, he held at paragraph 69 that the claim to tax relief had not been accepted by HMRC nor had it been made out in any tribunal or court, so all that had been established was the existence of a genuine dispute about whether the scheme based on the claim for tax relief worked. The first instance judge had, therefore, been right to dismiss the argument that the claimant had a legitimate expectation that HMRC would carry out a promise to challenge the scheme before the special commissioners in the usual way, and the new approach, which relied on the claim to tax relief as the possession, was no better. This was presumably because of the holding in Kopecky v. Slovakia supra, to which Mummery LJ had referred in detail at paragraph 51, to the effect that a claim to tax relief may generate a legitimate expectation if a currently enforceable claim is sufficiently established under national law, but that claim must be more concrete than a mere hope; it must be based on a legal provision or a legal act, such as a judicial decision, and not be just an arguable claim or a genuine dispute.

Bearing in mind that the claimant in Huitson CA had addressed an argument based on the effective financial burden imposed by the tax being the possession in question (see paragraph 50), it seems to me that Mummery LJ must be taken to have considered and rejected that proposition. Mr Kieron Beal QC, counsel for the respondents, contended that the ECtHR had endorsed this holding in Huitson CA, when it rejected the taxpayer’s application to that court as inadmissible (see application number 5031/12). But paragraphs 23-24 of that decision make it clear that the rejection was assuming, without deciding, that the claim to tax relief was an asset attracting the guarantees of A1P1. I cannot, therefore, accept that the ECtHR’s decision in Huitson is of any assistance.

All that said, since Mr Woolf has put this point at the forefront of his case, I think it should now be fully addressed. It should first be recorded that Mr Woolf seeks to draw a distinction between Huitson CA and this case, because he says that in Huitson CA, the taxpayer had only made a claim to tax relief, whereas here HMRC was making a claim to recover the tax from the appellants. The appellants retained the tax in their pockets and were, therefore, in possession of the money in every sensible meaning of the word “possession”. It should be recalled in this connection that “possession” in A1P1 has an autonomous meaning including a wide range of economic interests and assets.

As the judge in this case and Mummery LJ in Huitson CA made clear, a possession must either exist or be a claim in respect of which an individual has a legitimate expectation that it will be realised, and such a legitimate expectation cannot be based on just an arguable claim. Here, Mr Woolf seeks to avoid the consequences of accepting that the appellants had just an arguable claim to avoid paying tax on the original transfer of the properties to them, by focusing on the money that would be used by those appellants to pay the tax (if due) in due course. The problem with this approach is that it falsifies the debate in all the cases that lead up to Huitson CA. If it were an answer in a tax case to say that legislation closing a tax avoidance loophole was an interference with the money that the taxpayer would in due course use to pay the tax, that would be applicable in many, if not most, cases, since taxpayers rarely pay tax first and dispute their liability later. It is true that in Burden v. United Kingdom supra the ECtHR held that taxation was in principle an interference with a right guaranteed by A1P1, because it deprived the taxpayer of the money. But that case was concerned with a challenge to the imposition of any inheritance tax on two unmarried sisters on the basis that it was discriminatory.

Looking at the matter from first principles, the taxpayer can be assumed to have the relevant tax as a possession before he enters into the relevant transaction – in this case the purchase of property. But when he purchases the property under the first contract and transfer, SDLT becomes payable under sections 42-44 of the Finance Act 2003. The taxpayer also enters into an agreement for an option which gives him an argument that the SDLT charged under sections 42-44 is not payable. It is now common ground that the argument as to the meaning of section 45 (before the legislative changes) does not constitute an existing possession for A1P1 purposes, because the claim is not sufficiently certain or established within the principles set out in Kopecky supra. But the money available to pay the SDLT must, in my judgment, be affected by the argument as to whether it is payable to HMRC. Of course, the money is a possession in one sense, but it is a possession impressed with an arguable claim by HMRC, which prevents it being properly regarded as a possession for A1P1 purposes.

The question can also be answered by reference to the precise effect of the legislative changes. The legislative changes prevent the appellants arguing against HMRC in the FTT and beyond that their option agreements were a “transfer of rights” for the purposes of section 45. The effects of the legislative changes are, therefore, twofold:-

The question is whether the second consequence identified by the appellants can, by itself, be regarded as the deprivation of a possession when it is caused entirely by the first consequence of the legislative changes. In my judgment it cannot. In every case, where there is an argument as to whether tax is payable, and legislation is changed to make clear that it is, the potential taxpayer can say that he has been deprived of the tax. Effectively, he seeks to divide the alleged deprivation into two parts when it can only properly be regarded as a single step. The appellants have been deprived, as the judge said, of an argument that they were not liable to pay the tax. That is the primary effect of the legislative changes.

It would be different if the appellants had been challenging the imposition of SDLT itself. The basis of the rule about claims is that a disputed amount is not in the taxpayer’s possession if there is an arguable claim by HMRC to it, or if the taxpayer has an arguable claim to it (see paragraphs 62-70 of the ECtHR’s decision in National Provincial Building Society v. United Kingdom [1997] STC 1466). The location of the disputed fund cannot make a difference. Both sides claim they are entitled to it and it is not clearly established whether it properly belongs to the taxpayer or to HMRC. The prior question of whether the taxpayer has a right to the money must be decided before the taxpayer can claim to have been deprived by the legislative changes of a possession under A1P1. That makes sense from a practical point of view. If the taxpayer can show his scheme works, e.g. by obtaining a declaration to that effect, then he clearly has a possession. But if he does not, his claim that he has had a possession interfered with is premature, and begs the anterior question. Nobody knows whether the taxpayer has a possession for A1P1 purposes at that stage.

Mr Woolf also suggested that this analysis had to be wrong because it was impossible to imagine that there could be no relevant possession for A1P1 purposes. Thus, he said that it was counterintuitive to imagine that neither the claims to a carve out under section 45 nor the money in the appellants’ pockets were properly to be regarded as an A1P1 possession with which the legislative changes interfered. I do not agree. Of course, the appellants’ money is for some purposes a possession, as I have said. But, the question here is whether the legislative changes interfered with any of the appellants’ possessions within the meaning of A1P1. I do not think they did, because, again as I have said, by the time the legislative changes were made, the money that the appellants might have used to pay the tax was already the subject of an unresolved argument or claim by HMRC. All the legislative changes did was to remove the appellants’ argument that HMRC was not entitled to that money. The authorities show that that is insufficient to amount to an interference with a possession under A1P1. Even if the concept of a “claim” is not entirely apposite to describe the species of property affected by the legislative changes in this case, the appellants’ attempt to keep their money by arguing that they had entered into a “transfer of rights” within section 45 has exactly the same characteristics and the authorities must therefore be applicable by analogy.

Despite the fact that I have expressed my view in my own words, I should say that I think the judge was right on this point for the reasons she succinctly gave in paragraphs 52-53 of her judgment. A1P1 was not engaged by the imposition of the legislative changes. Like the judge, however, I will now go on to consider the question of whether there was an infringement of A1P1 in the event that it is later held that A1P1 was engaged.

The ECtHR has recently said that the first and most important requirement of A1P1 is that any interference with the peaceful enjoyment of possessions should be lawful, so that the “fair balance” issue becomes relevant only once it has been established that the interference in question satisfies the requirement of lawfulness and was not arbitrary. It explained that the concept of “law” in A1P1 refers to the quality of the law in question, requiring that it be accessible to the persons concerned, and precise and foreseeable in its application (see paragraphs 49-50 of Shchokin v. Ukraine [2011] STC 401, OAO Neftyanaya Kompaniya Yukos v. Russia [2011] STC 1988 at paragraph 559, and NKM v. Hungary [2013] STC 1104 at paragraphs 46-51). There was a debate before us as to whether the “fair balance” or proportionality issue should in some measure subsume the legality question, based upon the treatment of those questions by Lord Reid in AXA supra at paragraphs 116-134, and by Lord Mance at paragraphs 44-56 in In re Recovery of Medical Costs for Asbestos Diseases (Wales) Bill [2015] 2 WLR 481. But in my judgment, it is convenient to treat the questions separately in the context of this case, notwithstanding the obvious overlap between them. Lord Reed made clear in AXA at paragraph 116 (as the ECtHR had in Shchokin at paragraph 49, and at paragraph 67 in James v. United Kingdom (1986) 8 E.H.R.R. 123) that the concept of lawfulness required both the existence of a domestic law and compatibility with the rule of law.

The ECtHR has also repeatedly made it clear that, in the sphere of tax, the well-established position is that states may be afforded some degree of additional deference and latitude in the exercise of their fiscal functions under the lawfulness test (see paragraph 50 of NKM supra, paragraphs 75-83 of National & Provincial Building Society supra, and paragraph 559 of Yukos supra). As the ECtHR also pointed out in NKM at paragraph 51: “retroactive taxation can be applicable essentially to remedy technical deficiencies of the law, in particular where the measure is ultimately justified by public-interest considerations”. In any event, it has not and could not have been suggested that retrospective legislation is automatically unlawful.

In oral argument, it appeared that Mr Woolf’s primary contention was that the legislative changes could not be justified by the very general warnings that had been given and that were open ended in scope and time; they created a state of uncertainty, which was inconsistent with the rule of law in a democratic society unless there was some overwhelming justification for the measure. No such justification could be found where the same public policy applied to a number of situations, which the legislative changes left untouched.

In my judgment, however, it cannot be automatically unlawful or inimical to the rule of law to close a tax loophole retrospectively, just because there are other tax loopholes which are left open. If that were right, then it would never be possible for the government to close one loophole without being sure it had closed them all. Faced with this conundrum, Mr Woolf confined his submission to loopholes of which the government was aware, arguing that there were unclosed SDLT and employment loopholes at the time that the legislative changes were enacted. In my judgment, the premise is unsustainable. The rule of law inhibits arbitrariness in the sense of unpredictable, discretionary or subjective laws that lack a principled basis. In 2006, Lord Bingham identified 8 sub-rules to his definition of the rule of law (later included in his book “The Rule of Law” published in 2010). The first 3 sub-rules were: (i) that the law must be accessible and so far as possible intelligible, clear and predictable, (ii) that questions of legal right and liability should ordinarily be resolved by application of the law and not the exercise of discretion, and (iii) that the laws of the land should apply equally to all, save to the extent that objective differences justify differentiation.

The legislative changes were not arbitrary in any of these senses, just because there were other existing tax avoidance schemes that the government did not immediately tackle. Objective aspects of the appellants’ scheme, namely that it had attempted to get round the 2012 legislation, justified it being singled out. The government had, in the speeches and announcements that I have summarised above, made it perfectly clear that SDLT avoidance schemes based on sub-sales and options would not be tolerated, and that retrospective legislation would be used to achieve that objective. The appellants can have been in no doubt about any of that, before they decided to take advantage of a scheme devised purely to circumvent the precise wording of section 45(1A) as it was before the legislative changes. The import of the 2012 legislation was to prevent any such schemes defeating the clear intention that those purchasing property for their own use should pay the normal rate of SDLT. The legislative changes were simply enacted to make clear that the government was fulfilling its expressed intentions.

Moreover, the fact that, as an accident of timing, some other loopholes may have remained unclosed was not evidence of either unforeseeability or arbitrariness of the government’s actions. The government was taking a strong line for the reasons that the Chancellor expressed in the 2012 budget, namely that it was unfair and morally repugnant for some to avoid paying SDLT that all others buying property had to pay.

Mr Woolf also relied on the Protocol in relation to unlawfulness, on the grounds that the existence of the Protocol rendered the legislative changes unforeseeable. It is first necessary to understand the status of the Protocol. The Protocol was an extra-statutory announcement or promise made by the government. As such, it operated: “in the realm of politics, not of the courts, and the question whether the government should be held to such a promise is a political rather than a legal matter” (see paragraph 41 of Wheeler v. Office of Prime Minister [2008] EWHC 1409 (Admin), and paragraph 14 of Mitting J’s judgment in R (on the application of UNISON) v. Secretary of State for Health [2010] EWHC 2655 (Admin)). The sovereignty of Parliament cannot be confined by extra-statutory promises like the Protocol. That said, it is still necessary to examine whether, on a fair reading of it, the legislative changes were a breach of the terms of the Protocol. Before doing so, however, I would like to repeat that I am not content to engage in a close textual construction of a document like the Protocol. It was not intended to be a legally binding promise and should, therefore, be read, not as a deed, but as a statement of government intent.

As both parties have submitted, the only part of the Protocol that specifically applies to the legislative changes is the statement that such retrospective legislation will be “wholly exceptional”. Nonetheless, on a fair reading, since the 3 bullet points in numbered paragraph 2 apply to immediately implemented changes announced otherwise than at a budget (i.e. before the legislation, but not retrospective to the announcement) one might think that they should also apply to fully retrospective changes. The two relevant points that are said “normally” to be required are that there would otherwise be “a significant risk to the Exchequer” and that an immediate change was “expected to prevent significant losses to the Exchequer”. It is hard to see how, despite the use of the word “and” between the 3 bullets, the requirements can have been intended to be cumulative, since they may apply in different situations. Moreover, whilst the requirement of “significant losses to the Exchequer” plainly refers to monetary tax losses, the words “significant risk” must have a broader connotation.

In my judgment, there cannot sensibly be said to have been any breach of paragraph 2 of the Protocol. First, the promise only applies “normally” and the need to make sure that further unwarranted and abusive SDLT schemes were not devised to get round the 2012 legislation takes the situation far away from the normal. Secondly, there were significant risks to the Exchequer in the form of taxpayers thinking that the government did not mean what it said and that they could get away with minor variations on tax avoidance schemes once the rather unwieldy process of anti-avoidance legislation left a window for them to do so. Thirdly, Mr Woolf’s main point was that the losses caused by the appellant’s schemes did not occasion significant losses to HMT, since the general anti-avoidance provisions were due to come into force within a few weeks of the legislative changes. The maximum estimate of tax losses caused by the appellants’ scheme was about £7 million. I take the view that financial losses can be significant to the Exchequer without being huge in relative terms. In my judgment, Mr Woolf puts the appellants’ case as if the word “significant” was in fact “substantial”. These losses were not very large compared to the overall tax take, but were certainly significant to the Exchequer, because they represented a number of taxpayers deliberately doing what the government had said it intended to stop. It is that aspect of the matter that, in my judgment, brings these legislative changes within the Protocol.

Finally, of course, Mr Woolf argues that the situation occasioning the legislative changes was not wholly exceptional as the Protocol demanded. Again I disagree. This situation was not unique but it was very unusual. The Chancellor had issued a stark warning to taxpayers after repeated abuses of the SDLT system. He had made it clear that he intended to close all loopholes so that purchasers of properties in the UK would pay their fair share of SDLT. He said he would act retrospectively if tax avoiders persisted in trying to find ways around the clear intention of Parliament. This was serial abuse. The fact the appellants were in each case only avoiding a relatively small amount of tax – though no doubt not small to them – is nothing to the point. The respondents wanted, exceptionally and quite reasonably, to demonstrate that the government had meant what it said.

Insofar as the Protocol is relevant at all to the foreseeability of the legislative changes, I do not think it takes the appellants anywhere in showing that the changes were not entirely foreseeable and indeed to be expected bearing in mind the history of SDLT avoidance schemes based on sub-sales and options.

In my judgment, the judge in this case was entirely right to conclude, as she did in paragraph 77 for the reasons she gave, that the legislative changes were lawful in substance and form.

In Huitson CA, the “fair balance” principle was described as follows at paragraphs 29 and 57: in securing the payment of taxes a national authority must strike a fair balance between the general interests of the community and the protection of the individual’s fundamental rights, including the right to possessions in A1P1. In that balancing exercise the national authority has a margin of appreciation under the ECHR and a discretionary area of judgment under domestic law. The area of appreciation and judgment is wide in matters of social and economic policy. The judge correctly reflected these principles at paragraph 60 of her judgment. It is common ground that the question of retrospectivity falls to be considered at the “fair balance” stage.

Mr Woolf’s main attacks are on the alleged breach of the Protocol, and the supposed failure to point to any special justification for the retrospectivity of the legislative changes (see paragraph 122 of Lord Reed’s judgment in AXA supra). I have already dealt with the appellants’ contentions on the Protocol, which are not, for the reasons I have given, sustainable.

As regards, the justification for retrospectivity, it seems to me to be particularly clear in this case that the respondents had to introduce retrospective legislation to ensure that effect was given to Parliament’s existing legislative intent and to the warnings that Ministers had repeated on past occasions when abuse had been tackled. In this case, the balance between the general interests of the community and the protection of the individual’s fundamental rights falls heavily on the side of the public interest. In reaching this conclusion, I have taken all the circumstances into account and have borne in mind the factors enunciated by Kenneth Parker J at first instance in Huitson [2011] QB 174 at paragraph 75 and approved by Huitson CA at paragraphs 30-35 and 57. The factors I have identified as making the legislative changes lawful are properly taken into account again at the proportionality stage. The judge summarised the reasons appropriately in paragraphs 80-84 of her judgment. Once again, I think she was right for the reasons she gave.

In oral argument, Mr Woolf contended that article 6 was engaged because the legislative changes deprived the appellants of a fair and public hearing before the FTT and the courts “in the determination of his civil rights and obligations”. He submitted that the question of whether a particular question is a “civil right or obligation” for the purposes of article 6 should be determined first under national law; it is only if national law does not recognise the right as “civil” that one should move to consider the autonomous meaning of the term for ECHR purposes. The appellants draw an analogy with the way in which the ECtHR determines whether there is a “criminal charge” for article 6 purposes. Primary reliance was placed on Jusilla v. Finland [2009] STC 29, where the ECtHR held that in general the question of whether a matter was “criminal” depended, as a starting point, on the classification of the offence under domestic law, even where offences concerned with tax were involved. The issue arose in that case, because under the Finnish legal system, tax surcharges belonged to administrative law (see paragraph 26). Mr Woolf drew attention to the well-established three criteria to be considered in the assessment of the applicability of the criminal aspect set out in paragraph 30, and submitted that the same approach should be adopted, by analogy, to the determination of whether proceedings were “civil”. Mr Woolf pointed next to the dictum at paragraph 39 of the ECtHR’s decision in Stran Greek Refineries v. Greece where it had been said that “[a]ccording to the [ECtHR’s] case law, the concept of “civil rights and obligations” is not to be interpreted solely by reference to the respondent State’s domestic law”. So, the argument runs, domestic law must be relevant to some extent. Finally, Mr Woolf referred us to the dicta of Lords Hoffmann and Walker at paragraphs 69 and 112 in Runa Begum v. Tower Hamlets LBC [2003] 2 AC 430 to demonstrate that they, at least, might have some sympathy for the development in the law for which he contends.

In my judgment, the judge was right to place primary reliance on the ECtHR’s decision in Ferrazzini v. Italy [2001] STC 1314, where it was made clear at paragraphs 20-29 that the concept of “civil rights and obligations” in article 6 is an autonomous one, and that the ECtHR considered “that tax disputes fall outside the scope of civil rights and obligations, despite the pecuniary effects which they necessarily produce for the taxpayer”. I do not think that there is an appropriate analogy to be drawn between the approach to the determination of whether there is a “determination of … any criminal charge against him” on the one hand and the “determination of his civil rights and obligations” on the other hand. That is because the ECtHR has specifically determined that the starting point in the former question is to determine whether the offence charged is, according to national law, criminal or disciplinary or both. The autonomous definition adopted by the ECtHR then requires other factors to be addressed. There is no equivalent approach to the question of whether a right or obligation is “civil”. It is simply established autonomously that tax disputes are not “civil” for the purposes of article 6. It is true, of course, that in some jurisdictions, tax cases are regarded as purely administrative and that, in the UK, they are regarded in most situations as civil, as opposed to criminal, claims. But that does not inform the decision as to whether the dispute in this case as to whether SDLT was payable before the legislative changes was civil for the purposes of article 6. In my judgment, on the clear authority of the ECtHR, it was not. Article 6 was not engaged.

Had article 6 been engaged, I would have been inclined to agree with the judge that the legislative changes satisfied the higher test of compelling grounds in the public interest. In these circumstances, further consideration of the inter-action between A1P1 and article 6 can, I think, await a case in which the point directly arises.

For the reasons I have given, I have concluded that neither the appellants’ claims that their option agreements were transfers of rights for the purposes of section 45 of the Finance Act 2003, nor the money representing the unpaid SDLT in the appellants’ pockets are, in the circumstances of this case, properly to be regarded as “possessions” for the purposes of A1P1. A1P1 is, therefore not engaged. If it were engaged, I would hold that the legislative changes were, although retrospective, lawful. They were neither unforeseeable nor arbitrary. Moreover, the legislative changes satisfied the proportionality test. The fair balance between the public interest and the protection of the appellants’ fundamental rights falls firmly on the side of the public interest in preventing taxpayers taking advantage of abusive tax avoidance schemes after clear warnings have been given that such schemes would not be tolerated and would be tackled with retrospective legislation. Article 6 is also not engaged, since tax proceedings do not relate to the determination of a “civil” right or obligation.

I would, therefore, dismiss the appellants’ application for judicial review and also dismiss both appeals.

I agree with Vos LJ for the reasons which he gives that we should not permit cross examination on this application for judicial review, and that the judge was right to refuse it on the application before her. I also agree for the reasons given by Vos LJ that if A1P1 is engaged, the legislative changes were, although retrospective, lawful and satisfied the test of proportionality. Finally I agree that Article 6 is not engaged because the present tax proceedings do not relate to a civil right or obligation.

I also agree in the conclusion that A1P1 is not engaged, but would express my reasons essentially as follows. A1P1 prevents a person being “deprived” of “possessions” except in accordance with the prescribed conditions. For A1P1 to be engaged one must identify two things: something which falls within the legal definition of a possession, and an act which constitutes a deprivation of that possession.

On the face of it there are two ways of presenting the argument that the appellants in the present case have been deprived of a possession. The first is to identify the possession as the “argument” that the transaction into which they have entered does not attract SDLT. The second is to identify the possession as the money in the appellants’ hands which they contend was not, before the legislative change in issue, subject to any claim by HMRC, but which is now subject to an unanswerable claim.

Looking at the first of these ways of presenting their case, the appellants have no difficulty with the contention that they have been deprived of the argument. The difficulty which they face is with the assertion that a mere argument is a possession within A1P1. As Vos LJ has explained, such an assertion runs counter to the decision of the CJEU in Kopecky v. Slovakia (No. 44912/98) (2005) 41 EHRR 43. The argument in the present case is not a possession within A1P1.

This consideration forces the appellants to retreat to the second way of presenting the case, based on the impact on their money. This money, so they submit, must be their possession, and they have been deprived of it by the retrospective legislation in issue. The flaw in this argument is that what the appellants gain by retreating to solid ground on the issue of what is a possession, they lose on the issue of deprivation. I would, for my part dismiss this alternative way of putting the case on the basis that the appellants have not established that the legislation has deprived them of their money. To put the matter shortly, they have only been deprived of their money if the scheme was effective to avoid SDLT. They have not established that proposition, however: they have instead contented themselves with the common ground that the proposition is arguable. Just as an arguable claim to something is not a sufficient possession for A1P1, so also an arguable deprivation is not a sufficient deprivation. It follows that the common ground that it is arguable that the scheme was effective to avoid the incidence of SDLT is not enough to engage A1P1 whichever way the argument is run.

I would also, therefore, dismiss the appellants’ application for judicial review and dismiss both appeals.

Except on the issue of whether A1P1 is engaged in this case, I need only say that I am in agreement with Floyd LJ and Vos LJ on all issues that arise for determination in this appeal. Neither is persuaded that the case comes within A1P1 and I agree with them. However, as each has expressed his own reasons for his conclusion about this, I think I must say a few words on the subject myself.

It is important to recognise that the appellants’ case in relation to A1P1 was not advanced on the basis that the judge should have determined whether their scheme was in fact effective; at best, therefore, they had an argument that it was. It is also important to recognise that the appellants did not seek to persuade us that the argument that SDLT was not payable was itself a possession. What was said to be a possession within A1P1 was the money which would fund the SDLT if it was in fact payable, but which would otherwise remain in the appellants’ pockets.

The routes by which Floyd LJ and Vos LJ conclude that the appellants have failed to establish an A1P1 claim are, I think, subtly different. Vos LJ focuses on whether the money can be said to be a possession. In essence, he considers that the fact that, at the relevant time (namely at the time of the legislative changes), it was impressed with an arguable claim by HMRC prevents this. Floyd LJ focuses on whether, assuming the money is properly classed as a possession, the appellants have been deprived of it, and concludes that they have not established this. If they are right in their argument about the efficacy of their scheme, they have indeed been deprived of it by the legislative changes; if not, they were going to lose it anyway by operation of the existing statute. Arguably being right is not sufficient to establish the required deprivation.

Analysed in this way, there seems to me to be nothing inconsistent in the two judgments and I express my agreement, on the A1P1 point, with both.

Accordingly, I too would dismiss the appellants’ application for judicial review and dismiss both appeals.