First-tier Tribunal (Tax)

You are here: BAILII >> Databases >> First-tier Tribunal (Tax) >> Hilden Park LLP v Revenue and Customs (VAT - APPEALS : Other) [2017] UKFTT 217 (TC) (08 March 2017)
URL: http://www.bailii.org/uk/cases/UKFTT/TC/2017/TC05707.html
Cite as: [2017] UKFTT 217 (TC)

[New search] [Contents list] [Printable PDF version] [Help]

[2017] UKFTT 217 (TC)

[image removed]

TC05707

Appeal number: TC/2012/1012, 1014, 2231, 3039

VAT – Halifax abuse alleged by HMRC – burden of proof

FIRST-TIER TRIBUNAL

TAX CHAMBER

	HILDEN PARK LLP	Appellant

	- and -

	THE COMMISSIONERS FOR HER MAJESTY’S	Respondents
	REVENUE & CUSTOMS

TRIBUNAL:

JUDGE Barbara Mosedale

Sitting in public at the Royal Courts of Justice, the Strand, London on 22 February 2017

Mr K Gordon, Counsel, instructed by Sharpe Pritchard LLP, for the Appellant

Mr M Jones, Counsel, instructed by the General Counsel and Solicitor to HM Revenue and Customs, for the Respondents

DECISION

Background

1. As this was a case management hearing, I make no findings of fact. Nevertheless, the basic background to the case was not in dispute and I set it out to provide a context for the decisions I make, although it should not be taken to comprise any finding of fact.

2. At some point in time, an ordinary partnership, which might be described as a predecessor to the appellant LLP in this appeal, entered into arrangements involving the golf course the operation of which had been the business of the partnership. Those arrangements involved setting up companies to take over the business. Rent was paid by the new companies to the partnership.

3. The companies did not pay VAT on the golf club membership subscriptions, relying on the sporting exemption on the basis that they were, they claimed, non-profit making. At some point the original partnership was succeeded by the appellant LLP.

4. That first iteration of the arrangements was the subject of assessments on the companies as HMRC’s view was that the companies were profit making and, consequently, their supplies were not exempt. This led to the liquidation of those companies as they were unable to pay the assessments.

5. HMRC then assessed the original partnership and the appellant LLP on the basis that they had put in place arrangements intended to abuse a VAT exemption in order to gain a tax advantage and that under VAT law as explained by the CJEU in Halifax [2006] EUECJ C-255/02 they could be taxed as if there were no such arrangements. In other words, the golf club owner (the original partnership and its successor, the appellant LLP) should be taxed as if they, rather than the companies, had operated the golf club and made the supplies to the club members. It was, of course, accepted that the partnerships were not entitled to the exemption as they were not non-profit making bodies.

6. Those assessments were appealed by both the original partnership and appellant partnership. The appeals were joined for hearing and a single decision issued: [2013] UKFTT 391 (TC). The FTT upheld the assessments. The appellants appealed to the Upper Tribunal which upheld the FTT decision: [2015] UKUT 405 (TCC). (And if and to the extent that the above summary of the first iteration of the arrangements differs from that given by either the FTT and UT in those published determinations of the appeals, the findings of fact are as set out in those decisions. I make no finding of facts here.)

7. The appellants sought leave to appeal to the Court of Appeal and were refused both by the Upper Tribunal and Court of Appeal.

8. Following that, the appellants have made a complaint to the European Commission that the FTT, Upper Tribunal and Court of Appeal have misapplied or misunderstood the CJEU decision in Halifax. They have asked the Commission to bring infringement proceedings against the UK but as yet have had no decision by the Commission either way. As Mr Gordon pointed out, even if the Commission agrees with the appellant that there has been an infringement, the Commission still has a discretion whether or not to bring an infringement action. His point was that even a negative decision by the Commission would not by itself necessarily indicate that the Commission considered that there was no infringement of EU law. But, in any event, so far there has been no decision at all on whether to bring infringement proceedings.

9. The liquidation of the companies back in 2007 caused the appellant to put in place a second iteration of the arrangements. While the FTT and Upper Tribunal decisions can be referred to for a much more complete account of the relevant facts to the first iteration of the arrangements, I have been told virtually nothing about this second iteration. I was told it was different; it certainly involved different operating companies. It seems that these arrangements also involved supplies of golf club membership being made by two companies without VAT being accounted for on those supplies on the basis the companies considered themselves to be non-profit making.

10. HMRC made various assessments on the appellant arising out of this second iteration and these were appealed. The Tribunal directed that the four appeals relating to the second iteration would be stayed behind those relating to the first iteration. The appeal proceedings in relation to the first iteration (I will describe it as Hilden Park 1) did not conclude until 2016 and the stayed appeals (which I will refer to as Hilden Park 2) are now live again.

11. The case management hearing concerned the appropriate directions to make in Hilden Park 2.

Consolidation

12. The four live appeals all concern assessments, decisions and penalties arising out of the same arrangement between the appellant LLP and two companies relating to the same golf course. The appellant is the same in all of the appeals. Neither party suggested that that it was or ever would be appropriate to try any of the four appeals apart. They agreed the four appeals should be consolidated and I ordered that they be consolidated into a single appeal.

13. In future, the sole reference number for the appeals will be TC/2012/1012.

Abuse of process

14. As is well-understood, there is no res judicata in the Tribunal: it is open to the appellant to raise in proceedings with HMRC concerning certain tax periods the same legal question that has already been decided against it where different tax periods were in dispute: Littlewoods [2014] EWHC 868 (Ch). It is certainly open to the appellant to defend the assessments for the later periods despite losing the appeal against the assessments for the earlier periods.

15. However, where the factual matrix of the first and second set of appeals are very similar, there is the possibility that there might be an abuse of process in the second set of proceedings. It might be, for instance, an abuse of process if the appellant were to seek to re-try issues of fact in Hilden Park 2 which were decided against it in Hilden Park 1. It is certainly HMRC’s position that the findings of fact made in Hilden Park 1 will be relevant to Hilden Park 2 and it seems to me that, if HMRC are right on this, this may only true to the extent that it would be an abuse of process for the same question to be tried twice between the same parties. In other words, the earlier findings of fact may not be relevant if it is open to the appellant to bring evidence on the same matter a second time round.

16. However, all parties were agreed, as I am, that it is too early to say whether or not there will an abuse of process in the second set of proceedings. If, having read the appellant’s evidence, HMRC consider that the appellant is abusing the tribunal process, then they should apply to the Tribunal for that part of the evidence to be disallowed and the Tribunal can determine that application at that point. But, having raised the issue, I am satisfied that it is not appropriate to make directions at this time. The appeal should proceed as normal.

Should burden of proof be a preliminary issue?

17. Both parties were also agreed, however, that I should not issue normal case management directions at this point. This is because they were not agreed who would have the burden of proof in the appeal and they were both agreed that resolution of this issue should take place before case management directions were issued. Mr Jones considered that HMRC should not even be directed to prepare their statement of case before the issue was resolved: Mr Gordon’s position was that he considered it inappropriate for there to be an exchange of evidence at this stage and was prepared to agree the statement of case was unnecessary were it not for the fact he considered a statement of case useful for the reference to the CJEU for which he was applying and which application I determine below.

18. Despite the agreement between the parties, I was reluctant to follow a course which would mean Hilden Park 2, which concerned the years 2007-2010, being stayed for a further period of years with the risk of the evidence becoming ever more stale.

Could the case be prepared for hearing or even heard without knowing which party has the burden of proof?

19. Nevertheless, I was persuaded that a hearing could not take place unless the burden of proof was resolved in advance. It was vital for HMRC to know whether they had to open the hearing and lead evidence to establish a prima facie case of Halifax-abuse. It was vital for them to know this because if they were required to but could not establish a prima facie case, the appeal should be allowed without the appellant being obliged to refute the assessments.

20. I was also easily persuaded by both counsel that it was inappropriate for exchange of evidence to take place before the issue of burden of proof was resolved. As Mr Jones explained, and Mr Gordon agreed, if HMRC had the burden of proof, their approach to the case would be quite different. HMRC would have to call witnesses, possibly even expert witnesses such as on valuation, to establish their case of Halifax-abuse. On the other hand, if HMRC did not have the burden of proof, they might elect (as they had done in Hilden Park 1) to call no evidence at all but rely on cross-examination to challenge the appellant’s evidence that there was no abuse.

21. Indeed, the appellant would need to know who had the burden of proof no later than exchange of evidence because otherwise it would be prevented from making the application which an appellant which does not have the burden of proof is entitled to make where the evidence (or lack of it) served by HMRC so indicates. And that is an application for HMRC to be barred on the grounds its case does not have a reasonable prospect of success. The right to make such an application, where circumstances indicate it is appropriate, is a valuable right as, if successful, it obviates the need for an appellant to prepare for a hearing it is virtually certain to win. It saves costs.

22. I also accepted Mr Jones’ position that HMRC would approach their statement of case differently if they had the burden of proof, but I put aside a decision on whether to require a statement of case until I considered Mr Gordon’s case for a reference to the CJEU.

23. I noted that there was a secondary assessment on the appellant raised on the basis that the rents received were not exempt but that was very much in the alternative and for a much smaller figure than the main assessment. Although the appellant would bear the burden of proof to defend that assessment, the burden of proof in the appeal against the much larger assessment made on the Halifax-basis was the determining one for the preparation and hearing of the appeal.

24. My conclusion in the hearing was that the question of who bore the burden of proof had to be resolved before the case could progress.

Isn’t the question of burden of proof already resolved?

25. The second question was whether a stay is necessary in order for burden of proof to be determined. Again, both parties were agreed that it was. Mr Gordon’s position was that the Upper Tribunal’s determination of the issue in Hilden Park 1, while probably obiter, was correct and should be followed in the FTT and Upper Tribunal but nevertheless accepted that HMRC wanted the matter addressed by the Court of Appeal. Mr Jones’ view was that what the Upper Tribunal said in Hilden Park 1 on burden of proof was obiter and wrong and the HMRC would have appealed it had it not been that they had won the appeal despite the Upper Tribunal’s view that HMRC had the burden of proof. The appellant had not been given leave to appeal the Upper Tribunal decision so there was no possibility of a cross-appeal by HMRC on burden of proof.

26. I accepted that for HMRC the question of burden of proof in cases where they allege Halifax-abuse is very important. As I have already explained, if HMRC have the burden of proof, it will affect their entire preparation for the hearing and potentially put them to greater cost (if they decide to call expert evidence such as on valuation matters) in any appeal, including this one, where Halifax-abuse is alleged by HMRC.

27. I accepted that burden of proof is an important point and one which has not yet been finally determined against HMRC: I accepted that they may well be prepared to appeal the point to the Court of Appeal and so the view of the Upper Tribunal in Hilden Park 1 may not be the final ruling on the matter.

28. In conclusion, I accepted that a further, potentially very long, stay of this appeal was the only appropriate course of action while the question of burden of proof is being finally resolved. In the meantime, it would not be appropriate to issue case management directions (bar possibly one for a statement of case). On the contrary, I needed to make a decision on the burden of proof so that an appeal on this preliminary issue could be made as soon as possible, and as both parties were prepared to proceed, I heard argument on the point at the hearing and reserved my decision.

My decision on burden of proof

Was there a binding determination on burden of proof in Hilden Park 1?

29. This Tribunal is bound by any decision of the Upper Tribunal. Nevertheless, HMRC’s position was, and Mr Gordon did not really suggest otherwise, that the Upper Tribunal’s determination on burden of proof in Hilden Park 1 was not actually a part of its decision (I will refer to this as being ‘obiter dicta’) because at §56-57 the Upper Tribunal recognised it was not actually necessary for it to decide the matter as it would not affect the outcome of the appeal. I find it was obiter.

30. However, both counsel did agree that obiter dicta from a superior court given after hearing full argument, as in Hilden Park 1, while not binding, should normally be followed. Mr Gordon considered I should follow it in any event as it was in his view right: Mr Jones wished me to depart from it because he considered it was wrong.

31. As what the Upper Tribunal said in Hilden Park 1 on burden of proof was only obiter, I must decide whether or not I should follow it. I consider that dicta from a superior court that was given after full argument should be followed unless it is obviously wrong.

Was there a binding determination on burden of proof in Lower Mill?

32. Even though the Upper Tribunal’s dicta on burden of proof in Hilden Park 1 were obiter, am I in any event bound by what the Upper Tribunal said in The Lower Mill Estate Ltd [2010] UKUT 463 (TCC)? And that depends on whether what the Upper Tribunal said in that case, as there is no suggestion it was obiter, was actually a decision on the same question of where the burden of proof lies in appeals against assessments where HMRC allege Halifax-abuse.

33. Lower Mill was a case with allegations of Halifax-abuse, but one in which there was no argument on where the burden of proof lay. The dispute was over the first limb of Halifax which requires there to be an abuse in the sense of tax advantage contrary to the purpose of the VAT Directive. This led to a discussion by the Upper Tribunal of comparators, in other words, different commercial structures which would have led to the same physical result (of a holiday home sold to the buyer). As part of this discussion, the Upper Tribunal said:

“[136] If it is necessary to resolve [whether] ...one or other model is the only possible comparator, we would adopt [the comparator put forward by the taxpayer]. We see this as being a question of law or of mixed fact and law which can properly be the subject matter of an appeal on an issue of law under s11 Tribunals, Courts and Enforcement Act 2007.

[137]…The onus is on HMRC to establish that there is an abuse and thus that the [taxpayer’s comparator] is anti-purposive in the present case. Unless we are persuaded, which we are not, that transactions taking place under the [taxpayer’s comparator] are not normal commercial operations for a developer such as [the taxpayer], abuse cannot be established. In this context, compare Halifax at para 75 where the court said in relation to the second limb that it must be ‘apparent from a number of objective factors that the essential aim of the transactions concerned is to obtain a tax advantage’. This language is not consistent with an obligation on the taxpayer to show the reverse.”

34. On one view, the reference to ‘onus’ and the reference to an obligation on the taxpayer to show the reverse, would seem to indicate that the discussion was about burden of proof. But burden of proof was not an issue in the case so it would seem unnecessary for the Upper Tribunal to have made a ruling on it.

35. Mr Jones’ view was that the reference to ‘onus’ was a reference to an onus in submissions. While it is difficult to see how there could be an onus in submissions, the Upper Tribunal is clearly not using onus in the sense of proving abuse as a matter of fact: it used ‘onus’ in the sense of establishing that the taxpayer’s comparator wasn’t a valid one, a question which the Upper Tribunal had already said in [136] was a question of law (or at least mixed law and fact).

36. Indeed, if [137] was a discussion about burden of proof it would be in conflict with [136]. At [136] the Upper Tribunal is dealing with the comparators. The Tribunal is saying that choice of comparator is a question of law, or at least mixed fact and law, and, impliedly, that Edwards v Bairstow considerations do not arise. In other words, the Upper Tribunal was explaining it had a free choice of comparator and did not have to show that the FTT’s choice of HMRC’s comparator was unreasonable in order to overturn it. That would fit uneasily with [137] if [137] was about burden of proof, which is only relevant to matters of fact. If choice of comparator was a matter of fact, burden of proof would be relevant, but the Upper Tribunal would have to have decided the FTT’s decision in choosing HMRC’s comparator was unreasonable in the Edwards v Bairstow sense in order to overturn it.

37. I think the better view is that the Upper Tribunal either intended no comment on burden of proof at all in [137] or, if they did, they were talking merely of a shift in the evidential burden, in the sense that, as the appellant had shown that there was a comparator which gave equivalent tax liability to the arrangements actually adopted, the evidential burden had shifted to HMRC to establish that it was not a valid comparator.

38. My conclusion is that the Upper Tribunal in Lower Mill did not make a ruling on burden of proof and so what they said is not binding. It is not even persuasive in the sense it was not a comment about burden of proof. But the Upper Tribunal did consider burden of proof in Hilden Park 1 and what they said must be highly persuasive. HMRC consider what they said was wrong and do not wish me to follow it. So I will move on to consider what was said.

Is burden of proof significant?

39. In the hearing I decided that burden of proof was significant in this case to the extent that the appeal had to be stayed until the matter was finally resolved. However, I did ask for the parties’ comments on §62 of the Upper Tribunal decision in Hilden Park 1, as that appeared to suggest that burden of proof might not be so significant. The Upper Tribunal said:

“[62] Even if the FTT had accepted that HMRC had the burden of proving abuse of law, it is far from clear to us that the FTT would have required HMRC to open the case or that, if it had, a submission that the Appellants had no case to answer would have been either appropriate or successful. Under rules 5 and 15 of the Tribunal Procedure (First-tier Tribunal) (Tax Chamber) Rules 2009, the FTT has wide powers to regulate the conduct of proceedings and the evidence before it. Even if the FTT had concluded that HMRC had the burden of proving that the Halifax principle was engaged, it could have required the Appellants to lead evidence about the transactions and the background to them. Further, in considering a submission of no case to answer, the FTT could take account of the documents before it, whether formally produced by witnesses or not, and require witnesses to give evidence about such documents. It follows that we reject this ground of appeal.”

40. Mr Gordon’s view was that what the Upper Tribunal said in §62 was fundamentally wrong; Mr Jones’ view was that he would not say it was wrong, but just that it was difficult to perceive many circumstances in which it would be right to require the appellants to give evidence where the appellants had applied for the appeal to be allowed on the basis HMRC had failed to make out a prima facie case.

41. Taking what the Upper Tribunal actually said in stages, they said, and were clearly right to say, that HMRC could, theoretically, in a case where they have the burden of proof, rely on the documents alone, without calling any witness evidence, to establish a prima facie case. I would perhaps add the caveat that HMRC could only do that to the extent that the authenticity of the documents was not in dispute: if authenticity was in dispute, HMRC would have to call a witness to give evidence of authenticity.

42. The Upper Tribunal was also clearly right to suggest that in any case, including one where HMRC bears the burden of proof, it might well be appropriate to make an order for disclosure by the appellant of relevant documents. I might add the caveat that it might well be too late to do so at the final hearing. The Upper Tribunal also recognised that the FTT, in common with all courts, could summons a reluctant witness to give evidence to support HMRC’s case, including a witness who might originally have submitted a witness statement in support of the appellant’s case.

43. But the Upper Tribunal appears to go further and suggest that the appellant partners could have been compelled to give evidence in their own appeal. This is not a question of how wide the Tribunal’s case management powers are: like the courts it has the power to order anyone to give evidence. The question is whether it is in accordance with natural justice to require the appellant partners to give evidence to support HMRC’s appeal. And any evidence supports HMRC’s appeal if, without it, HMRC could not establish a prima facie case. So was the Upper Tribunal right to say that the appellants themselves could be compelled to give evidence about the transactions and background to them in a case where HMRC had the burden of proof and the appellant made a submission of no case to answer?

44. In a case where the appellant does have the burden of proof and the facts are in dispute, a failure by the appellant to lead any evidence will mean the appellant loses the appeal. Mr Gordon’s view, as I understand it, is that if the appellant does not have the burden of proof in an appeal, he is entitled not to lead any evidence. Moreover, he says where HMRC have the burden of proof, they should not be entitled to win the case by relying on the evidence of the appellant. It is one thing to challenge the appellant’s evidence in cross-examination and quite another to actually rely on the appellant giving evidence in order to prove HMRC’s case. Yet here in §62 the Upper Tribunal appear to suggest that the Tribunal, exercising its wide case management powers, could compel the appellant to give evidence on behalf of HMRC even if s/he would not otherwise choose to give any evidence at all.

45. There are no allegations of criminal conduct and the right not to self-incriminate is inapplicable. But the appellants have been assessed and have the right to appeal that assessment. Being required, even summonsed, to give evidence in favour of HMRC’s case may well adversely affect their ability to pursue their appeal. So if the Upper Tribunal actually meant that, where the burden of proof was on HMRC, and the appellant made a submission of no case to answer, it would be appropriate to require the appellant partners to give evidence, then I do, respectfully, find what the Upper Tribunal said very difficult to understand.

46. In conclusion, despite §62, I remain of the view that it is critically important to both parties in Hilden Park 2 to know before they prepare for the hearing where the burden of proof lies, because I don’t think the appellant partners ought to be compelled to give evidence to support HMRC’s case.

Burden of proof where assessments founded on allegations of Halifax-abuse

The rationale for the decision in Hilden Park 1 on burden of proof

47. So who has the burden of proof is important. The explanation for the Upper Tribunal’s decision that HMRC bears the burden of proof is in §60 of their decision. Mr Jones criticises both the conclusion and the explanation; Mr Gordon considers the conclusion correct but points out that there is a tension between what the Upper Tribunal said here and what they said on the so-called ‘Atrium’ point later on in their decision:

[60] In determining who bears the burden of proof in an appeal where abuse of law is alleged, it is necessary to consider which party substantially asserts that there is or has been an abuse. As discussed above, it is the nature of an abusive arrangement that the taxpayer’s appeal would succeed on the purely formal application of the legislation. The appeal will only fail if it can be shown that there is an abuse, ie the resulting tax advantage is contrary to the VAT Directives and the essential aim of the transactions is to obtain a tax advantage. If abuse were not alleged, or, having been alleged, cannot be established then the appeal must be allowed. It follows that establishing that a tax advantage is contrary to the VAT Directives and the essential aim of the transactions is to obtain a tax advantage is an essential part of HMRC’s case in an appeal where abuse of law is alleged. Accordingly, HMRC bear the burden of proving those matters.

48. Mr Gordon’s criticism is that later, at §73, the Upper Tribunal adopted the reasoning of the Upper Tribunal in Atrium Club Limited [2010] EWHC 970 (Ch) and held that the arrangements could be re-defined under the doctrine in Halifax even if they had not worked as intended. Yet here in §60 the decision on burden of proof is justified on the basis that ‘it is the nature of an abusive arrangement that the taxpayer’s appeal would succeed on the purely formal application of the legislation’.

49. Needless to say, in so far as there was conflict between §60 and §73, Mr Gordon’s view was that what the Upper Tribunal said in §60 on burden of proof was right and what they said at §73 on the ‘Atrium’ point was wrong; Mr Jones considers §60 wrong and §73 right. If there is conflict between what the Upper Tribunal said at §60 and §73, I am bound to proceed on the assumption that what was said at §73 was correct. What was said on burden of proof was obiter; what it said at §73 was part of the decision (‘ratio’).

50. But, properly analysed, what the Upper Tribunal said at §60, that ‘it is the nature of an abusive arrangement that the taxpayer’s appeal would succeed on the purely formal application of the legislation’, was quite right in the particular circumstances of that case. The assessments the subject to the appeal were made on the golf club owners and not the golf club operators. On a purely formal application of the legislation, the owners (the appellant partnerships) would not have had any tax liability arising out of the operations as they did not make the supplies to the members. Their liability only existed if the arrangements were re-defined under Halifax. So it was true to say in Hilden Park 1 ‘it is the nature of an abusive arrangement that the taxpayer’s [ie appellant partners’] appeal would succeed on the purely formal application of the legislation’.

51. Mr Gordon’s criticism is only valid, on the particular facts of the case, if the Upper Tribunal made this comment while eliding the partners with the operating companies: on a purely formal application of the law the operating companies were liable to tax so the arrangement failed even without an application of Halifax. And it is possible that at §60 the Upper Tribunal did not have in mind the distinction between the partnerships and the operating companies.

52. Moreover, the justification used by the Tribunal in §60, even if literally valid in Hilden Park 1, would not be valid in all alleged Halifax-abuse type cases, as not all Halifax-abuse type arrangements involve the interposition of operating companies (although many do). But the point is that it is not a comment that could apply to all such cases and therefore what was said about the formal success of the impugned arrangements was not a good justification on which to found a decision on burden of proof in all Halifax-abuse cases.

53. (I note in passing that the comment in §60 could apply to all Halifax-abuse type cases, but only if what the Upper Tribunal later decided in respect to Atrium was wrong. But I am bound by their decision on Atrium so there is no point in considering that possibility.)

The ordinary rule in tax cases

54. Does this difficulty with the reasoning mean that what the Upper Tribunal said is obviously wrong? Applying basic principles, where should the burden of proof lie? The normal rule in common law is that the burden of proof is on the person who makes the allegation. It might be thought, therefore, that because HMRC raise assessments, they are making the allegation of tax liability and should bear the burden of proof, but it is well established and beyond dispute that in appeals concerning liability to tax (such as appeals against assessments or refusals to repay tax) that common law rule is displaced and the burden of proof ordinarily lies on the appellant taxpayer, who alleges that he is not liable to the assessment. There are many statements of this principle:

“….The scheme of the 1972 Act appears to me to be this, that if the taxpayer omits to include in his return something which the commissioners consider, using their proper judgement, is taxable, then the commissioners can, using the best of their judgment, assess the taxpayer at a certain figure…, and if there is no appeal, that figure is then deemed to be the tax payable. If the taxpayer wishes to have the assessment altered, he must go to the tribunal, and unless the tribunal finds the commissioners are wrong, the assessment still stands. It seems to me, in those circumstances, that any taxpayer who appeals to the tribunal takes upon himself the burden of proving the assertion he makes, namely that the assessment is wrong, because unless he proves this there is nothing on which the tribunal can find an error in the assessment. The facts and figures are known to him….”

“Now there is no principle of law which I know of – and both counsel disclaim it – that where a provision is a taxing provision the onus of proof is on the tax gatherer, while if it is a mitigating provision it is on the taxpayer, and to this extent the ratio of the Ivy Café case is, in my view, wholly wrong. But that the onus of adducing evidence and satisfying the tribunal that the assessment is wrong lies on the appellant under s 40 [now s 83 VATA] I have not any doubt at all.”

Tynewydd Labour Working Men’s Club and Institute Ltd [1979] STC 570, Forbes J page 580e-f and 581b

“Before us counsel for the taxpayer company accepted (as was accepted below) that the burden of proof rested on the taxpayer company, in the sense that the taxpayer company had to show that the assessment was wrong….”

“…At no time do the commissioners have any burden to prove anything before the tribunal. Neither its case nor any aspect of the matter, factually or evidentially, carries any burden imposed on the commissioners. It is throughout, in my judgment, up to the taxpayer company, if it can, to attack the assessment in whole or in part….”

Grunwick Processing Laboratories Ltd [1987] STC 357 at 360 (Court of Appeal)

55. The point was also shortly, if colourfully, put in an Australian case:

“There is every reason to assume that the legislature did not intend to confer upon a potential taxpayer the valuable privilege of disqualifying himself [from liability to an assessment] by the simple and relatively unskilled method of losing either his memory or his books.”

Latham CJ in Trautwein (1936) 56 CLR 63 at page 87 (Australian High Court)

56. This was approved in the Privy Council, who also said:

“The element of guesswork and the almost unavoidable inaccuracy in a properly made best of judgement assessment, as the cases have established, do not serve to displace the validity of the assessments, which are prima facie right and remain right until the taxpayer shows that they are wrong and also shows positively what corrections should be made in order to make the assessments right or more nearly right. It is also relevant, when considering the sufficiency of evidence to displace an assessment, to remember that the facts are peculiarly within the knowledge of the taxpayer.”

Lord Lowry in Biflex v Carribbean Ltd [1990] UKPC 35 at page 10

57. Yet another succinct statement on burden of proof is:

“it is the taxpayer who knows and the taxpayer who is in a position (or, if not in a position, who certainly should be in a position), to provide the right answer, and chapter and verse for the right answer.”

Nicholson v Morris [1977] STC 162, Goff LJ approving the words of Walton J

58. I have considered it clear from the above citations that, fundamentally, the reason that the burden is on the appellant to disprove the assessment is because the appellant exclusively has control of the evidence. The tax system could not function effectively if HMRC had to prove assessments as the taxpayer could simply lose the necessary evidence of liability.

59. Nevertheless, there are exceptions to the exception: the taxpayer does not have the burden of proof in all tax cases.

Exception - penalties

60. It is well established that penalties are not like assessments: the burden of proof lies on HMRC. The reason for this rule is not that the allegation involves criminal conduct. While penalties are imposed for dishonest behaviour, they are also imposed for careless behaviour, which is not criminal. Indeed, liability to a penalty may not depend on any particular state of mind at all: many tax penalties are imposed merely because a taxpayer does something late.

61. It seems to me that the reason for the rule is that, because penalties are penalties, and not (alleged) liability to tax, the normal common law on burden of proof that the person who makes the allegation must prove it, is not displaced.

62. So this is not really an exception: it is just that penalties do not involve tax liability and so the rule explained in Grunwick does not apply. The rule in Grunwick was established, in my view, for the practical reason that the taxpayer exclusively controls the evidence of his own tax liability: but that is not really true where penalties are concerned, where HMRC ought to have the information necessary to know whether there has been, for example, a late payment of tax. Therefore, there was no need so far as penalties are concerned for the Grunwick exception to the normal rule that he who makes the allegation must prove it.

63. And as with the ‘normal’ rule, the appellant has the burden to prove any potentially applicable defence, such as a reasonable excuse for the default.

Exception – fraud and sham

64. The Grunwick rule that the burden of proof is on the taxpayer is also displaced where the liability to an assessment (or refusal to repay tax) depends on fraud or something akin to dishonesty being proved. So in a case where HMRC allege that the trader is not entitled to recover input tax which would otherwise be recoverable under normal principles because its transactions were connected to fraud and the trader knew that that was the case (Kittel), HMRC has the burden of proof:

“[81] … It is plain that if HMRC wishes to assert that a trader’s state of knowledge was such that his purchase is outwith the scope of the right to deduct it must prove that assertion.”

Court of Appeal in Mobilx Ltd and others [2010] EWCA Civ 517

65. Both counsel were agreed that the same is true where a sham is alleged. If the assessment depends on certain documents being shown to be a sham, then HMRC must prove that they are a sham.

66. This exception also makes sense as fraud or other misbehaviour should not be alleged without grounds for alleging it: so unless HMRC is aware of the evidence which apparently justifies such an allegation, the allegation should not be made. So HMRC has the burden of proof and must raise a prima facie case where they allege fraud or sham.

Other exceptions

67. Even apart from the question of Halifax-abuse cases, there are other exceptions. For instance, where an assessment is raised under s 29 TMA 1970 it appears it is accepted that HMRC must prove that the conditions for raising the assessment have been met, and that includes proving the requisite state of mind in the appellant. Nevertheless, this is also a case where at least to some extent HMRC ought to possess or even control the evidence (eg of discovery): the issue does not turn on liability to tax, the evidence for which is in the exclusive control of the taxpayer.

68. I have also suggested that HMRC would also have to prove allegations of means of knowledge of connection to fraud, on the basis that they control the evidence, rather than the appellant, but even if this is right, it may simply be a case of shifting of the evidential burden.

CJEU’s view

69. Lower Mill at the end of §137 might be thought to indicate that the CJEU thought the burden of proof was on the taxpayer, but the CJEU said:

[76] It is for the national court to verify in accordance with the rules of evidence of national law, provided that the effectiveness of Community law is not undermined, whether action constituting such an abusive practice has taken place in the case before it ….

70. This seems to suggest that a national procedural rule which put the burden of proof on the appellant would only be displaced if those rules of evidence made it very difficult or virtually impossible for appellants to show that their arrangements were not abusive. So the CJEU did not require the burden of proof to be on HMRC.

Comparison to direct tax

71. Mr Jones compares Halifax-abuses case to those of Ramsay type abuse, potentially applicable to direct tax cases. It is clear that taxing provisions must be construed purposively and applied to the facts realistically. This can lead to some tax avoidance arrangements failing. The burden of proof remains, as per Grunwick, on the taxpayer to disprove the assessment but that is not surprising as questions of statutory construction are merely questions of law; matters of fact are within the control of the appellant.

72. HMRC consider Halifax-abuse cases to be one of purposive statutory construction applied to the facts realistically. It is simply a question of whether the appellant has a liability to tax or not. Mr Gordon does not agree: he sees the Halifax doctrine similar to the one in Kittel where the ‘normal’ tax position is overridden if HMRC can prove a particular set of circumstances involving fraud.

Conclusion

73. Mr Gordon considers the Upper Tribunal were right and for the reason given, but on the basis it is his view that they decided the Atrium point incorrectly. HMRC’s view is that what is at issue in a Halifax-abuse type case is an assessment to tax: there are no allegations akin to fraud and the case does not concern whether the formal conditions for making an assessment are met. The taxpayer controls the relevant evidence. Halifax-abuse cases involve statutory interpretation, say HMRC, and for that reason too should be treated like a normal tax avoidance case. For these reasons, HMRC consider that the normal Grunwick rule should apply and that the Upper Tribunal was wrong to reach the conclusion which it did.

74. I certainly have difficulties with the reasoning employed by the Upper Tribunal which only applies to some Halifax-abuse cases and in any event does not really explain why Grunwick would not apply. None of the exceptions to the rule in Grunwick are obviously applicable. The taxpayer controls the evidence. The assessment concerns a liability to tax and not a penalty. The assessment does not concern an allegation of fraud or sham. There is nothing in the CJEU’s decision which would prevent the UK determining where the burden of proof lay in accordance with its national rules.

75. Nevertheless, the Upper Tribunal’s view could be explained as a decision that allegations of Halifax-abuse should be treated the same as allegations of fraud or sham in the sense that they are allegations that the Upper Tribunal did not think HMRC should make without evidence amounting to a prima facie case of what was alleged. The word ‘abuse’ could be seen as an allegation of misbehaviour, albeit that the CJEU stressed in Halifax that it was to be assessed objectively. So while I do have my doubts about the reasoning, I think that the FTT should follow the dicta in Hilden Park 1 because it was the dicta of a superior court made after full reasoning and is not obviously wrong.

76. So my ruling is that applying what the Upper Tribunal said at §60 of Hilden Park 1 the burden of proof in Hilden Park 2 will be on HMRC.

Reference to Europe

77. I have already described what I will refer to as the Atrium point of law. It was the decision in Atrium, followed by the Upper Tribunal in Hilden Park 1, that a failure of the arrangements to operate as intended did not prevent the application of the Halifax-abuse doctrine where the practical effect of the arrangements had been to confer a tax advantage.

78. It is applied by the Upper Tribunal in Hilden Park 1 as follows:

[73]...As in Atrium, it was an inherent feature of the scheme that, because they wished to be seen as non-profit making, the Companies paid as much as they could afford to the Appellants by way of rent (covert profit) and did not accumulate any profits or surpluses thereby depriving themselves of the means to pay any liability to VAT that might (and did) arise. The fact that the scheme did not work does not, therefore, affect the tax advantage that has accrued to the Appellants.

[74] Having heard full argument on the point, we agree with the conclusion in Atrium that the fact that the scheme does not work as the parties intended does not mean that no tax advantage accrued......

79. Mr Gordon asked me to refer Hilden Park 2 to the CJEU on this, the ‘Atrium’ point. He considered the Upper Tribunal decisions in Atrium and Hilden Park 1 (and of course my decision in the FTT in Hilden Park 1) wrong. The appellant has already sought to place the matter before the EU Commission in its application for infringement proceedings (§8) but, as he said, even if EU law has been infringed, whether the Commission decides to take action is discretionary. A reference by this Tribunal in Hilden Park 2 would undoubtedly bring the matter before the CJEU.

Timing

80. It seems to me that there is one factor clearly in favour of a reference being made now and that is that the appeal is already rather old and is now likely to be stayed for at least another couple of years: having a referral to the CJEU running simultaneously with the appeal against this preliminary decision on burden of proof would avoid any need to lose yet more time with a referral later on in the appeal’s progress to determination.

81. Mr Jones pointed out that the Tribunal might even be unable to make a referral in future years assuming that the UK leaves the EU: but I do not see how I can make a decision based on what the law might one day be. I can only consider it as it is now. And my view is that so far as timing is concerned, it would be better to refer now rather than later.

European law question ‘raised’?

82. But I do not have complete discretion to refer to the CJEU any question of European law. Whether an EU point of law should be referred depends on Art 267 of the Treaty which provides:

“Where such a question is raised before any…tribunal of a Member State, that ...tribunal may, if it considers that a decision on the question is necessary to enable it to give judgment, request the Court to give a ruling thereon.”

83. I accept that the point at issue in Atrium, which Mr Gordon wants me to refer, is a point of EU law to which Art 267 could potentially apply. But is the Atrium point actually raised in Hilden Park 2?

84. The problem for the appellant here is that Mr Gordon said that he was not instructed on whether or not the appellant accepted that the arrangements failed in this case. If the appellant’s case is that the arrangements did not fail, in other words that the operating companies’ supplies were exempt on a formal application of the law, then there is a possibility that it will succeed on that case. And if it succeeds, no Atrium point of law arises. No Atrium point will arise because the arrangements will have worked exactly as intended: they will not only have conferred the (alleged) tax advantage but have conferred it as intended in reliance on the sporting exemption.

85. I cannot refer a question of EU law that only may be relevant to the decision that the tribunal hearing the substantive appeal will be required to make. The question of EU law must arise before I can refer. So at this point in time, I do not have the power to refer the question which Mr Gordon wishes me to refer.

86. If, of course, the appellant at any point were to accept that the operating companies did not make exempt supplies, then it seems to be that the Atrium point will certainly arise before the Tribunal hearing the substantive appeal in Hilden Park 2. At that point, it would be possible to make a reference to the CJEU on the basis that an EU point of law ‘is raised’ before the domestic court.

Is a ruling of the CJUE ‘necessary’ for the Tribunal’s decision?

87. But that still may not be enough for there to be a reference. There is another precondition and that is that the CJEU’s ruling is ‘necessary’. What does this mean? In the well-known case of Ex parte Else [1993] QB 534 the Court of Appeal ruled:

“if the facts have been found and the Community Law issue is critical to the court’s final decision, the appropriate course is ordinarily to refer the issue to the Court of Justice unless the national court can with complete confidence resolve the issue itself….If the national court has any real doubt, it should ordinarily refer.”

88. But in the later case of Littlewoods Organisation plc [2001] EWCA Civ 1542 the Court of Appeal said:

“…A measure of self-restraint is required on the part of the national courts, if the Court of Justice is not to become overwhelmed….

…[a] development which is unquestionably significant is the emergence in recent years of a body of case-law developed by this court to which national courts and tribunal can resort in resolving new questions of Community law. Experience has shown that, in particular in many technical fields, such as customs and value added tax, national courts and tribunals are able to extrapolate from the principles developed in this court’s case law. Experience has shown that the case-law now provides sufficient guidance to enable national courts and tribunals – and in particular specialised courts and tribunals – to decide many cases for themselves without the need for a reference…”

89. My opinion is that the CJEU’s decision on the Atrium point is not necessary for any decision in Hilden Park 2. That is because, firstly, the Upper Tribunal has twice ruled that that Halifax does apply to situations where a tax advantage is generated albeit not by the intended legal success of the arrangements: if the Upper Tribunal were in doubt in either or both Atrium or Hilden Park 1, they could have referred the point to the CJEU. But they did not. Secondly, the Court of Appeal were not persuaded that the point merited consideration by the Court of Appeal and refused the appellants’ application for leave to appeal Hilden Park 1, thereby indicating that they had no real doubt that the Upper Tribunal had reached the correct decision.

90. While there is no binding ruling on me that the matter should not to be referred, I do not consider it appropriate to diverge from the clear views of the Upper Tribunal and Court of Appeal. I consider that the Atrium point is clear and a ruling from the CJEU on it is not necessary. A reference should be refused.

91. In conclusion:

(a) the four appeals are consolidated;

(b) HMRC will have the burden of proof in the appeal;

(d) if there is an appeal from this decision, HMRC will not be directed to prepare their statement of case as how they do so might well be influenced by whether or not they have the burden of proof and there is no advantage to the appellant in having a statement of case now as I am not referring any question to the CJEU. The direction, therefore, is that HMRC should provide their statement of case 60 days after expiry of the 56 days they are given to lodge an application for permission to appeal against this decision unless such an application is lodged within those 56 days, in which event no statement of case need be delivered yet.

92. This document contains full findings of fact and reasons for the decision. Any party dissatisfied with this decision has a right to apply for permission to appeal against it pursuant to Rule 39 of the Tribunal Procedure (First-tier Tribunal) (Tax Chamber) Rules 2009. The application must be received by this Tribunal not later than 56 days after this decision is sent to that party. The parties are referred to “Guidance to accompany a Decision from the First-tier Tribunal (Tax Chamber)” which accompanies and forms part of this decision notice.