First-tier Tribunal (Tax)

[2015] UKFTT 517 (TC)

 

TC04675

Appeal number:  TC/ 2010/00925 & Others

FIRST-TIER TRIBUNAL

TAX CHAMBER

 

METROPOLITAN INTERNATIONAL SCHOOLS

Appellant

 

-and-

 

 

THE COMMISSIONERS FOR HER MAJESTY’S REVENUE AND CUSTOMS

Respondents

 

Tribunal:  JUDGE HOWARD M. NOWLAN

JULIAN STAFFORD

 

 

Sitting in public at the Royal Courts of Justice in London on 13 to 16 January 2015 and 3 to 5 June 2015

Roger Thomas QC on behalf of the Appellant

Eleni Mitrophanous, counsel, on behalf of the Respondents

 

 

©CROWN COPYRIGHT 2015

 

DECISION

Introduction

1. These Appeals raised a considerable number of points.

2. The first, and the main, point was that the Appellant provided distance learning courses such that there was a dispute as to whether its supplies were single or multiple supplies and, more particularly, assuming that the provision was of one single supply, whether that supply was a supply of zero-rated books or standard-rated education. The Appellant was not the type of provider whose provision of education was exempt, so that on the single supply analysis, the dispute was between zero-rated and standard-rated treatment. The courses in question included various “trade” courses, such as electrical and plumbing courses, and also animation and “games” courses, enabling people to become proficient in designing and animating computer and iPhone type games.

3. Ignoring the history of the dispute at this stage, both parties advanced, as their principal contention, that the Appellant was rendering one single service. The Appellant claimed that the service was the zero-rated provision of manuals, i.e. books for VAT purposes, and that any other elements of service were ancillary to the manuals, such that they should take their VAT nature from the zero-rated supply of the manuals. In the alternative, if it was not appropriate in accordance with the relevant case law to proceed on the basis that the supply of the manuals was the principal supply and the other aspects ancillary elements, a secondary test for identifying the nature of the single supply was to consider which element was predominant, and then it was contended that the manuals plainly constituted the predominant element of the supply.

4. The Respondents contended that the single service was the supply of non-exempt education. The supply comprised various elements. From the perspective of the customers, the Respondents contended that the customers’ aim was to secure education and to obtain a qualification. The “overarching” description of the supply was properly to be described therefore as a supply of education, and this was so even though it was almost certainly inappropriate to describe the supply of the manuals as ancillary supplies. Indeed, the categorisation of the single supply as education would remain correct even if the Respondents were to concede (not that they did) that the predominant element of the components in the single supply was indeed the provision of the manuals.

5. The Appellant contended in a very secondary manner that if its primary contention was not accepted, then there should be analysed to be two services, with the supply of books comprising by far the major supply.

6. The above issues are the most important matters in these Appeals. They involve careful consideration of the legal tests properly to be applied in resolving the dispute between the parties and, of equal significance, an analysis of the facts. We might indicate now, though it will certainly become evident, that we have found it difficult to rationalise all of the relevant authorities and to arrive, with confidence, at the correct tests to apply in identifying the nature of the single supply. While we regret this observation, we consider that this case may well be one where there will be appeals to a higher court, and quite possibly a referral to the ECJ for guidance. 

 

7. Most of the remaining issues result from the history of the dispute itself.

8. The Appellant has been trading for a considerable time and in the past there had been much consideration and discussion between the Appellant and HMRC as to how the Appellant’s supplies should be treated. This had extended to suggestions that the right analysis was that there was just one composite supply, with the Appellant of course contending that the single composite supply was of books and the Respondents contending that it was of standard-rated education. For a long period however, up until August 2009, it had been agreed that there were two separate supplies, and the discussions between the parties thus focused on the appropriate method to be applied in dividing the supplies and the split of consideration received for them.

9. The parties periodically reviewed the issue of whether the basic methodology for dividing the zero-rated supplies from the standard-rated supplies remained appropriate, and then annually the percentage figures were adjusted in accordance with the agreed method, according to how turnover, costs and other factors taken into account under the relevant method had all evolved during the year. Very roughly the method treated 75% of the supplies as zero-rated and 25% as standard-rated. In the fourth quarter of the year, the liability to VAT for all four quarters of the year was adjusted in accordance with the retrospective consideration of the figures for the whole year.

10. In 1999, there had been discussions between HMRC and the Appellant that culminated in both parties agreeing on an appropriate method for splitting the standard-rated and zero-rated supplies. An HMRC letter of 14 January 2000 confirmed the method that the parties had agreed. The second paragraph of the letter made the following points:

“The method is to be used from 1 October 1998, and is based on the costs involved in making both the standard- and zero-rated supplies. Should there be any changes in your business which prevents this method giving a fair apportionment of the course fees you must notify us immediately. This method may be reviewed, amended or withdrawn by Customs and Excise at any time. You may apply to this office for a change of method if this one no longer produces a fair and reasonable result.”

11. Further meetings in 2002, 2006 and 2007 broadly confirmed the method, though at the 2006 meeting the officer who undertook the visit to the Appellant indicated that he would be giving consideration to whether the House of Lords’ decision in College of Estate Management v. HM C & E [2005] UKHL 62 (“CEM”) might have some bearing on the situation. The relevant officer confirmed in the hearing before us that, having been moved to different duties within HMRC, he in fact failed to consider this matter.  The different officer who made the visit in 2007 knew what his predecessor had undertaken to do, though it appears that the issue was not touched on in the 2007 visit, and it was simply assumed that the then agreed method would continue.

12. In 2009, an enquiry was instituted into the repayment claim made by the Appellant, in which a substantial amount of VAT was being reclaimed. The claim resulted from the final adjustment calculations made for the year 2006, coupled with a large bad debt claim. The then officer, Ms Abida Rashid (“Officer Rashid”), looked into the claim and decided that it was not justified, not because it failed to produce the correct result in accordance with the agreed method, but because the right analysis, in her view, was that the result of the CEM decision was that all of the Appellant’s supplies for the relevant period should have been standard-rated as supplies of education. On that analysis, there was an additional liability to VAT, and not an entitlement to a repayment at all. The contention in relation to the 2006 period, and the further indications from HMRC indicated that the 2000 agreed method was to be withdrawn retrospectively for all periods for which revised assessments could be made.

13. Complaints on the part of the Appellant to HMRC’s Complaints Team that it was altogether unacceptable for HMRC to withdraw the 2000 agreed method retrospectively resulted in a decision by Officer Harris that back assessments would indeed be improper in the light of the 2000 agreement. Shortly thereafter however this was reversed by a decision by another officer in the same team, Officer Winder, who informed the Appellant that the revised view was that HMRC did not have power to give concessionary treatment so as not to collect the tax that they now considered was properly due for all the back periods, and therefore Officer Winder said that Officer Rashid would be instructed to make the relevant back assessments.

14. As a result, in 2010, Officer Rashid made assessments for all back periods that remained in the period for which revised assessments could be made. This led to the Appellant instituting judicial review proceedings.

15. Shortly before any judicial review hearing, HMRC changed their stance again, and conceded that they should not pursue the assessments for the back periods. Implicitly for the future it was of course HMRC’s contention that the services would be single services of education, and thus standard-rated, though the Appellant was appealing this issue in the First-tier Tribunal. The judicial review application has not however been withdrawn because two matters remain in contention in connection with it. The Appellant’s hope is that we may be able to deal with the two relevant matters in this Tribunal hearing, though HMRC contend not only that we have no jurisdiction, but that the Appellant’s case is flawed in any event. In the meantime the judicial review issue, which has been remitted by Mr. Justice Warren to the Upper Tribunal, is stood behind this present Appeal.

16. The first outstanding matter is that, while the VAT treatment of supplies by the Appellant has generally involved VAT being paid on the whole of the consideration for the provision of manuals throughout, say, a three-year course, at the point when the customer signed the contract for the services, it has always been the case that VAT in respect of the consideration paid periodically over the life of a course, when financed by one particular financier, was properly payable only as and when each instalment of consideration was due. It was therefore due evenly over the assumed period of the course, generally 36 months. That basis for spreading the liability for VAT over the entire course period in the case of the supplies made through this one financier was not itself in dispute. What was in dispute was the feature that, as the Appellant had fixed the consideration for services provided via this particular financier at the outset of these contracts, it had no way of increasing or adjusting the charge on the relevant customers in the light of HMRC’s claim that 100% rather than merely 25% of the services were standard-rated, and that this analysis should apply to all supplies, even those rendered under these contracts after August 2009. Accordingly the Appellant contended that HMRC’s new analysis of the correct method to charge VAT in general should not be applied in the case of these contracts in their run-off period. VAT of approximately £5 million was at stake in relation to this point.

17. The Appellant advanced three different points in support of its claim that we had jurisdiction to hear the above issue, in other words essentially a judicial review issue. One, namely that the decision of the Upper Tribunal in HMRC v. Abdul Noor [2013] STC 998 (“Abdul Noor”) to the general effect that we had no such jurisdiction was wrong, was not advanced before us but was reserved in case this Appeal should proceed to the Upper Tribunal or beyond the Upper Tribunal. We will describe the other two contentions to the effect that we had jurisdiction to deal with the above “run-off” point in due course.

18. The other issue arises as follows.

19. The result of the decision made shortly before the judicial review hearing was to commence, i.e. the decision that we recorded in the first sentence of paragraph 15 above, resulted in HMRC repaying the VAT refund claimed in respect of the 2006 period that had led to the enquiry that we mentioned in paragraph 12 above. HMRC was, however, late in making the repayment and the Appellant claimed interest supplement in respect of the late repayment.

20. HMRC refused to make any payment of interest supplement on the basis that such supplement was only to be added when actual VAT was repaid late by HMRC to a claimant that had a “VAT credit”. In the present case, on the basis of HMRC’s principal contention that for all periods for which assessments could be made, the strict VAT position was that all the supplies were standard-rated, no repayment of actual VAT was due at all for the 2006 period. Accordingly what HMRC had paid in respect of the repayment claim in respect of the 2006 VAT period was not VAT but a payment made to reflect the Appellant’s legitimate expectations, and no interest supplement fell to be made in respect of such a payment.

21. It naturally follows that if we decide the principal issue in favour of the Appellant, i.e. that the supplies are and have always been wholly zero-rated supplies since 2006, then the grandfathering point dealt with in paragraph 16 above will drop away, and the interest supplement point referable to the 2006 period will also drop away, because the repayment will then emerge to have been properly due, and to constitute actual VAT. In that case it would appear that HMRC would then concede that interest supplement should be paid. It is only if we decide the principal issue in favour of the Respondents that the jurisdiction issues and the answers to the two points just described will become of significance.

The principal matter in dispute, and how we will deal with it

22. The clearest way to deal with the principal issue of whether the Appellant’s trade involved single or multiple supplies, and if a single supply whether that supply was of zero-rated books or standard-rated education will be:

·         first to give a general description of the Appellant’s trade;

·         then to amplify the description of the two types of example course that were considered extensively during the hearing, and to indicate the elements of non-book supplies that were offered in relation to each;

·         then to summarise the marketing of the courses, and the way in which specialist “interviewers” were involved in selling the courses;

·         and then to describe the relevant features in relation to exams and qualifications that customers might wish to pass and to obtain.

23. Following that general description, we will summarise the parties’ respective contentions, and we will then make formal findings of fact.

24. We will then give our understanding of the relevant law, and we will describe the particular features of the law that we find difficult to reconcile.

25. Finally we will give our decision on the principal issue.

26. Although considerable evidence was given and there was extensive cross-examination, we will not refer to the evidence beyond incorporating all the relevant points into the paragraphs that follow. The main evidence for the Appellant was given by Mr. Jaroslav Bradik, a director of the Appellant. Mr. Malcolm Tuckett gave evidence into the strategy adopted by the people who marketed the courses, and Mr. Deepak Shamdas gave accounting evidence, principally relevant to the issue concerning the “run-off” contracts that we referred to in paragraph 16 above. We were satisfied that all the evidence was given honestly and fairly.  Mr. Bradik’s evidence covered virtually all the points that we consider in relation to the Appellant’s trading pattern. Mr. Tuckett’s evidence related only to the point just indicated, and Mr. Shamdas’ evidence was of only marginal significance once it became clear that VAT was charged over the life of the “run-off” contracts, and that the pricing under those contracts did not permit adjustments if the charge to VAT was greatly increased.

The general description of the trade

27. The Appellant’s trade involved the provision, according to the Appellant’s own description, of “blended distance-learning courses”.

28. The essence of the feature that the courses were “distance-learning” courses was that customers were provided with seemingly highly professionally produced manuals that described the particular subject matter on a step-by-step basis in a manner that should be clear enough for the customer to learn everything relevant to the topic in question, simply by reading the manuals. We were given no detail about the following matter but we were told that the manuals were produced by the Appellant’s non-UK parent company, and that the Appellant paid royalties to its parent company for the right to use and provide the manuals to customers.

29. Whilst there were other features to the overall provision by the Appellant to its customers that we will mention shortly, there was little doubt that the prime way in which customers were intended to learn their relevant chosen subjects was by receiving the manuals on a one-by-one basis and then spending, as the Appellant suggested, between 10 and 15 hours per week in reading and seeking to understand the manuals.

30. One of the obvious advantages of a distance-learning course, in contrast to a more traditional course at a university or technical college, was that provided a customer had the aptitude to spend 10 to 15 hours a week reading the manuals, the customer would still be able to pursue his ordinary job and thus earn money. We were told that in the case of the trade courses such as the electrical and plumbing courses, it was commonly the case that customers would be working on a full-time basis in the relevant trades, often in small family businesses, such that they would be gaining practical experience when undertaking their day job, so that the purpose of enlisting on the courses would be to gain a better understanding of the technical information relevant to their trade.

31. Another feature, and potential benefit, of the Appellant’s courses was that while the courses were far from cheap, costing between £5,000 and £7,000 for courses that were generally expected to entail 36 months of study for the customer to gain a full understanding of the subject, the courses were still very much cheaper than other university or available full-time courses in relation to the same subjects.

32. The feature that the courses can reasonably aptly be described as “blended” learning courses, and the essence therefore of what was meant by the claim that the courses were blended courses, was that there were various ways in which customers could look for support and guidance from so-called “tutors”. The marketing material that we will describe shortly sought to illustrate, or at least to stress the representation, that customers would not just be burying themselves in isolation, reading their manuals in a rather dry and perhaps dull manner, since at least the customers could always discuss matters “with their tutors” and gain support, encouragement and clarification from the tutors. Whilst we will have to assess to what degree this feature was perhaps over-emphasised in the marketing material and thus the degree to which the support and liaison fell short of the representations, there was clearly an aim of making the course sound to be rather more collegiate and to be more involved with others, rather than simply to suggest that customers would be left to read absolutely vast quantities of written material in isolation.

The two categories of course explained

33. The actual elements of support are best described by considering the two courses that were extensively described to us during the hearing, and indicating then the points at which there would be some interaction with tutors, or for instance an opportunity to gain practical experience, as distinct from simply reading the manuals.

Features common to both the electrical and other trade courses and to the animation and computer games courses

34. We were told that the ideal way in which a customer would undertake a course would be first to read and assimilate the content of the first manual that would be sent to him. We should add that manuals were only sent one at a time so that, as manuals were constantly being updated, customers would always receive the latest manual with any amendments that might have been made to it by the time it was provided. We can also imagine that were the entirety of the manuals to be supplied at the outset, customers might find this somewhat daunting.

35. When the customer had finished reading and understanding the first manual, he was expected to turn to the back of it where there were a number of multiple choice questions designed to ascertain whether he had understood the content of the manual.  Each question indicated four possible answers, described as answers (a), (b), (c) and (d) in the familiar multiple choice manner.

36. The original practice had been that the customer would answer these questions on paper; send the completed answers to the Appellant and then they would be marked by the so-called tutors. In those earlier days, as we understood it, the customer would simply be told whether he had achieved whatever percentage of correct answers constituted the pass mark, so that he had either passed or failed. These tests were referred to as TMAs or “Tutor Marked Assignments”. 

37. By the period relevant to this Appeal, the practice had changed and all the TMAs were dealt with via the internet. Thus, the customer was provided with a DVD that enabled him to record all his answers onto the DVD and then download them onto the Appellant’s website, whereupon they would be marked entirely by pre-selected answers included on a computer programme. This then gave the customer more information. If the question had been answered correctly, this would be indicated in relation to each question. If a question was answered incorrectly, say the answer given was (a), whereas the correct answer had been (c), the computer’s response would be to refer to the part of the manual that indicated why (a) could not have been the right answer, and then refer to the part of the manual that indicated why (c) had in fact been the correct answer. We were not shown specimen answers but it would have been consistent with other website responses if the answers had been provided in a friendly-sounding form, suggesting that they had been provided by a human being, possibly indicating that if anything remained unclear, then “Please contact me if you are still confused.”

38. Once a TMA had been successfully completed, the Appellant would send the customer the next manual. It was stated that TMAs had to be passed before the next manual would be provided but we were certainly told that this rule was regularly ignored and that if a customer was up-to-date with his payments, assuming that he was paying on an instalment basis, and he requested the next manual, this would be supplied. We were also told that some customers never bothered with the TMAs and implicitly this did not stop them receiving the later manuals.

39. Aside from the contact with computerised “tutors” in relation to TMAs, customers were able to contact “their tutors”, by phone or email, at any time, should they be having difficulty with anything in the manuals. Again in a rather friendly manner, on signing up new customers, the customers were invited to ring up and introduce themselves to their tutors.

40. The Appellant said that the function of those receiving the phone calls or the emails was almost always just to refer the person who was unclear about some issue to the relevant passage in the manual that should provide the answer. There was, in other words, no intention to supplement the information in the manuals. The aim was simply to direct the person raising the question to where the answer was to be found.

41. The Respondents gave great attention in cross-examination to the Appellant’s claim that very little use was made of the facility to phone or email the tutors in the way just described.

42. Before indicating our summary of what we consider to be a realistic indication of the importance attached to this “tutor support” feature, we should indicate that at any time, the Appellant might have 60,000 customers undertaking courses. It was possible that some would have ceased to read the manuals and so would effectively have “dropped out”, so that the realistic figure of active customers might have been somewhere between 40,000 and 60,000. The Appellant had 14 employees, of whom six dealt with administrative matters and had nothing to do with “tutor support”. Three provided tutor support in relation to all the trade courses, of which the electrical one that was described to us was simply one example. Another three dealt with the animation and computer games courses and one was available to deal with either.

43. The statistics in relation to how often customers resorted to the tutor support facility were slightly confusing and they led to considerable dispute between the respective counsel.

44. We will ignore all the detailed figures that were disputed because we consider that the relevant points are as follows. Undoubtedly some use was made of the facility to request tutor support, and furthermore that facility was available to all who contracted to pursue any course. Ignoring the nature of the support and the question that we will have to address as to whether we consider that the function was ancillary to the reading of the manuals, it cannot be said that the tutor support was a pure fiction.

45. Relatively little use was made of it however. Probably the majority of customers, and quite possibly a significant majority, never phoned up or sent in an email. Since it was easier to ascertain who had used the tutor support on more occasions when email questions had been raised, we were able to see that only a tiny minority had emailed for clarification on more than a trivial number of occasions. It was of course then unclear whether a number of emails would have related to further questions in relation to one topic or whether each related to some different topic. A considerable number of those customers who had made at least some phone or email contact had made that contact only on an exceptionally limited basis. Viewed in contrast to the 500 to 750 hours that the Appellant suggested that customers needed to spend reading the manuals annually, the phone and email contact can only be classed as relatively trivial, albeit genuinely available. We might indicate at this point that in terms of any enquiry as to whether the Appellant’s functions of marking and returning the TMAs and providing the tutor support were “ancillary” to the provision of the manuals, we consider it far more appropriate to address that question by considering the nature of the TMA marking and the tutor support, rather than by counting up relatively insignificant numbers of phone calls and emails.

46. While the TMA marking function and the tutor support functions were available in relation to both the trade courses and the computer games courses, we will now deal with the further add-on functions that were available in relation to the two categories of course. They differed as between the trade courses and the animation type courses.

Other “add-on” provision made in relation to the trade courses

47. In the case of the trade courses, there are three additional add-on functions provided by the Appellant.

48. The first was the provision of a DVD that reproduced some of the information from the manuals, presumably as an aid to revision.

49. The second was a function on the Appellant’s website that enabled customers studying the electrical course to enter a virtual room, and click on items such as the light switch, whereupon an expanded view of the wiring would become available, and the customer could consider whether there were errors in the wiring.

50. Thirdly there was the rather more significant feature that if customers had passed all the TMAs in relation to the technical aspects covered in the manuals they could then join a one-week or two-week practical course where they could practice the skills that they had learnt. We were told that these courses were provided in several locations in the UK, generally provided by sub-contractors, and that the prime purpose of the sessions was to enable those attending to practice their skills, rather than for anyone to seek to provide any new education. It was said that there would always be a supervisor in attendance but that his function was principally to ensure that all health and safety requirements were met. It was said that if one of the customers attending found some difficulty with some particular task the supervisor might in practice assist or explain something, but it was still said that this was not the intended role of the supervisor.

51. The take up of practical courses was said to be very low. It was suggested that this was probably because most of those who would have completed the courses and thus been eligible to attend the practical sessions would have been working in the business such that they would have regarded supervised practice in undertaking wiring work, for instance, as being entirely superfluous.

Add-on functions in relation to the animation and computer games courses

52. The facts in relation to reading the manuals, hopefully submitting TMAs and requesting tutor support, were broadly the same in relation to the courses that we are now concerned with, as those functions available in the trade courses.

53. The first additional supply in relation to the computer-based courses was the provision to customers of two readily available computer programmes that could be downloaded onto customers’ computers, one being Photoshop. We were told that both programmes could be obtained by anybody for roughly £75, and that the programmes were needed to enable customers to create, modify and animate the images and the animated actions that the manuals would describe to them, and enable them to accomplish.

54. We were told that customers would be expected to produce their own design or animation, using the skills that they would progressively be learning and that if the customers chose to take one of the examinations to obtain some diploma (see below), the quality of the design or animation produced by the customer would generally contribute to up to 75% of the marks in any relevant exam. We were told that one of the Appellant’s tutors might give some useful tips as to how designs or animations might be improved but essentially this exercise was meant to show the particular customer’s own work, hence its significance in later examinations.

55. The Appellant also provided a web forum that enabled customers, and indeed non-customers, on a nil payment basis, to demonstrate their designs and concepts and to discuss them with others on the website. The Appellant provided a moderator to administer the sessions but the purpose of these discussion forums was to enable customers and others to discuss designs and animations amongst themselves, and it was not for the Appellant to provide any instruction.

56. There was no opportunity for customers on the animation and computer games courses to attend practical sessions in the manner that we described in relation to the trade courses.

Marketing of courses

57. We were told that the available courses were publicised on local television channels, by advertisements, and on the Appellant’s website. The Appellant made payments to Google to gain prominence, in relation to various obvious enquiry descriptions relevant to their courses. 

58. Once a potential customer showed an interest in a course, the customer would be visited by a sales specialist. These individuals were now working directly for the Appellant, though they had previously been engaged by a separate marketing firm that continued to train them in their present roles.

59. We had supposed and suggested that the key element of the marketing would have been to down-play the significance of the manuals and the required reading and instead to emphasise the blended nature of the course and the endeavours to make the course sound relatively collegiate, with references to “your tutors” etc. Evidence from the impressive head of the training firm, Mr. Tuckett, made it clear that the essential aim of the marketing was to get the potential customer to concentrate on why he was dissatisfied with his present job or prospects, and then to get him to realise that it was only by training that he could improve his prospects. Having achieved this, the practical benefits of the Appellant’s courses would be emphasised, namely the features that the courses were cheaper than most other options and that the “in your own time” feature of the courses enabled customers to continue with their daytime jobs. Admittedly some of the explanatory booklets did emphasise the tutor assistance and the fact that a tutor was always available, a phone call away, and efforts were made to diminish the notion that the course consisted essentially of wall-to-wall reading in isolation. We will comment below on the degree, and the possible significance, of any element of mis-representation in the marketing brochures, and indeed on the significance of what was actually claimed and described.

Examinations

60. In contrast to the position in the CEM case, the Appellant did not provide any examinations or provide any degrees, qualifications or diplomas. Its courses were generally designed to prepare customers to take third party examinations, provided for instance by the City & Guilds, and if a customer passed the requisite TMAs and in the case of the trade courses, undertook the practical experience session, the Appellant was then committed to pay the examination fee for some suitable course. We understood that if the customer failed the examination but had otherwise undertaken the course reasonably diligently, the Appellant was then committed to provide a further series of manuals. We were unclear whether on the second occasion the Appellant remained liable to pay examination fees again, though this is a relatively unimportant detail.

The parties’ respective contentions

61. We will amplify the parties’ contentions when dealing with the law and when giving our decision. In purely general terms the contentions were as follows:

The Appellant’s contentions in general terms

62. The Appellant’s general contentions were as follows:

·         It was first said that in a qualitative sense, it was difficult to find a case where the main element of the supply was so clearly the sale of the manuals, and where the “add-on” elements of supply were equally clearly ancillary.

·         On the basis that the marking of the TMAs was simply producing computer-based answers to questions set in the manuals, and the answers and comments all referred to the parts of the manuals that should be re-read when answers had been wrong, the whole exercise in relation to the TMAs was ancillary. In terms of one of the basic definitions of what was ancillary, the marking of the TMAs was not an end in itself from anyone’s perspective, but was rather a way of ensuring the better enjoyment or utilisation of the principal supply. In reality, even though it involved using a computer and logging on to the Appellant’s website, it was no different in nature from the familiar case where the answers were provided at the back of the book, or by turning the book upside down. By amplifying the answers, and referring to passages in the manuals, the computer-based system was simply easier to implement, in that it saved tutors having to mark papers, and it gave more information, thereby enhancing the utilisation of the manuals.

·         Exactly the same applied to the tutor support. It was perfectly easy to appreciate that the manuals could be used, as an end in themselves, without any resort by customers to any of the add-on services, and without any resort in particular to the tutor support. The tutor support could obviously not be seen as an independent aim, and plainly by invariably referring back people who raised questions to the relevant paragraphs in the manuals, the tutor support again simply enabled some customers (anyway a minority, since only a minority ever utilised this facility) to use and understand the manuals better.

·         The same applied to the other “add-on” elements, because they all essentially allowed the customers to practice or revise the knowledge gained from the manuals. They provided no further or separate education.

·         The Appellant claimed that the Respondents were wrong to say that the customers’ aim was to be educated and to pass examinations and obtain a qualification. It was not only the case that the Appellant’s services did not extend on any basis to examining customers and providing any sort of diploma or recognised qualification, but there was also little evidence that the customers’ aim was indeed to pass the third party examinations for which the Appellant was indeed liable to pay the fee. The Appellant was only liable, in the case of the trade courses, to pay for entry into examinations if and when the customer had achieved satisfactory answers in relation to the TMAs and then attended the practical courses. In the trade courses only a small minority satisfied these conditions, and so we assume that examination fees were only paid for that small minority. We were given no evidence as to whether those customers who did not meet the conditions still paid themselves for examinations, assuming that they had fully completed the course, or whether the situation may often have been that those working in family businesses would simply have regarded the “end result” as having been to learn a lot of technical information to assist and increase their general competence, perhaps in small family firms.

·         The Appellant’s characterisation of the fair description of the customers’ central aim was “to learn, essentially by self-study in reading and studying the manuals”, possibly then either enhancing that prime aim by limited utilisation of ancillary services, or alternatively by ignoring all the ancillary services altogether.

The Respondents’ contentions in general terms

63. The Respondents’ basic contentions were that there was a single supply and that the nature of that supply should be identified by paying regard to the only description that could be applied if one looked to the “overarching” characteristics of the supply. The only apt description of the entire supply was therefore education. This approach, essentially based on the particular case law test that the Respondents contended that we should apply, was then allied to the Respondents’ claim as to the nature of the “student’s economic purpose”. That was said in the closing submissions to be “to receive an education and if successful a qualification and it would be a mischaracterisation to say that his/her aim is to receive manuals.”

Our findings of fact

64. We make the following findings of fact:

1.      The Appellant’s aim was that the manuals should be entirely comprehensive, and that the information contained in them would be all that was required to enable customers to master the particular subjects.

2.      In contrast to the position in the CEM case there was no additional provision of classroom tuition.

3.      The “tutor support”, provided via phone calls or emails virtually always referred customers making enquiries back to the relevant passages in the manuals, reflecting the fact that the manuals had been drafted to include all the required information and explanations.

4.      The TMAs were simply a more efficient, and much more informative, means of dealing with multiple-choice questions that were set out at the end of each of the manuals. The answers invariably directed customers back to the appropriate part of the manuals when further study was required.

5.      In relation to the trade courses, the provision of DVDs designed to repeat the content of the manuals, the web-based virtual room to practice and test the information derived from the manuals, and the rarely requested practical sessions did not provide any additional technical information but each simply repeated the information and enabled customers to practice what they had learnt. The suggestion that the person overseeing a practical session might assist if one of the participants was doing something wrong did not undermine the unchallenged evidence that the purpose of the presence of that person was to ensure that health and safety requirements were duly met.

6.      In the case of the animation and computer games courses the provision of the down-loads of Photoshop etc did not provide education, but constituted the equivalent of the provision of paper and paints to an artist, namely required tools to enable customers to practice what they had learnt.

7.      In the case of these same courses, it was acknowledged that periodic tips might be offered to customers in their preparation of their own designs and animations, but the essential purpose of the exercise, on the part of the customers, of providing such designs and animation was to illustrate “their own work”.

8.      Albeit that an employee of the Appellant might act as moderator in the on-line discussion sessions, these sessions were designed for students to discuss amongst themselves what they had learnt and whether their various designs and animations met with approval from others, and the sessions barely involved any supply of any relevance from the Appellant at all.

9.      In testing whether the manuals were the principal supply and whether the “add-on” functions were ancillary, and then applying the case law test that something that constitutes an end in itself is the principal element and something that does not constitute an end in itself cannot be a separate principal function, we consider that the manuals are the Appellant’s principal provision and all other items are ancillary. This results not only from the fact that the manuals in isolation are intended to be, and we consider them to be, sufficient to achieve the customers’ desired end in itself, and the “add on” items are irrelevant in this context, but the majority of customers continued with the courses and continued to receive and, presumably, to study the manuals, yet many (indeed we were told a considerable majority) made no use of the “add-on” functions whatsoever. This is not to say that they were not available and part of the contracted supply. Of course they were. But if the end could be, and often presumably was, attained without resort to any of the “add on” functions, this demonstrates that the add-on functions must have ranked as ancillary.

10.  In the event that customers, or some customers, aimed to finish their courses in order to take third-party examinations and to obtain third-party qualifications, these ends were not part of the supply offered by the Appellant. More relevantly, when we were told that in the case of the trade courses, it was a requirement before the Appellant would pay for a customer to sit third-party examinations, that the practical sessions had been attended, and we were told that only a very small minority of trade course customers took the practical sessions, we find it difficult to accept that the end sought by the objective and average customer was “an education, culminating in an examination and a qualification”. There was no evidence as to the aims of customers beyond the confines of the actual supplies made by the Appellant. Some might have been entirely content simply to have gained considerable additional knowledge for career purposes, and in the case of the animation courses, some might simply have aimed to foster hobby activities.

11.  Some of the marketing information may have involved an element of hard selling, and may have over-stressed the supposedly blended nature of the courses. We will deal in the decision itself with whether this will have been decisive. For present purposes, we simply say that representations along the lines of “Your tutor, teacher, friend and mentor” will assist you at all stages, and that it is important to contact the tutor to discuss course work are all something of an exaggeration. They were not strictly untrue because the phone and email facility to seek help from tutors was real, but it was somewhat ramped up in the marketing information. Whether we should apply VAT by reference to the supplies that were made or to a ramped up description that will either have been believed or disregarded, we will deal with in our decision.

The law

65. This is a case in which we consider that the law is not entirely clear, and where the result will very much be governed by identifying the correct legal principle.

66. In any case, where there is a dispute as to whether a transaction constitutes the provision of two separate supplies or one supply with two elements, the rules for determining this first issue are relatively straightforward. The issue is first to be addressed by considering all the circumstances from an economic point of view, and from the perspective of the typical consumer. There is then said to be one supply where one element is the principal supply and the other element or elements are ancillary to that principal supply. Where that is not so, there will also be cases where from an economic point of view it would be artificial to split two elements into two separate supplies, so that again there will be treated as being one single supply in that situation. The issue of whether pricing is based on a composite price for the two elements, or separate prices for each element, can be of relevance but the pricing choices are not determinative. In connection with the rule mentioned above where one element is the principal element and the other or others are ancillary, an element that is not an end in itself ranks as an ancillary element. Ancillary elements generally contribute to the better enjoyment of the principal element.

67. Where there are rightly treated as being two separate supplies, and not one composite supply, the VAT nature of each may well differ and will obviously be geared to the nature of each individual supply. Where the consideration was paid in one amount for the two supplies, the consideration has to be apportioned between the two supplies.

68. There is more difficulty in identifying the nature of the one single service when the conclusion is that the two or more separate elements constitute one single supply for VAT purposes. Indeed there appear to be four distinct tests for identifying the correct nature of the single supply, or possibly a fifth if the simple application of common sense can be regarded as the fifth test. 

69. The first test (Test 1) is that where two or more distinct elements have been held to constitute a single supply because one element is the principal element and the other or others are ancillary, then the single supply takes its nature from the principal element, and this subsumes the ancillary elements. A good example of this situation is illustrated by this case. Assuming firstly that it is right to treat the provision of books and manuals as the principal supply, then the response to phone and email questions as to where to find relevant bits of information in the manuals would obviously be ancillary. It would entirely support the principal supply; it would not be an end in itself; and it would ensure the better use and enjoyment of the principal supply.

70. Before indicating the other three tests for identifying the nature of the single supply, it is worth observing that these tests have been applied in quite different factual circumstances, and the test to apply may or may not be influenced by these circumstances. For instance there will be situations where neither element can be considered as the principal element and another as ancillary, but nevertheless one element may be a component in an overall supply, with the description of the overall supply encompassing the relevant component. Examples of this situation are of course the restaurant meals provided on the ferry in the case of Faaborg-Gelting Linien A/S v. Finanzamt Flensburg [1996] STC 686 and the vaccination of patients by a doctor in Dr Benyon and Partners v. Customs and Excise Commissioners [2004] UKHL 53. The point with which we are presently concerned is not to identify the test that the respective courts actually applied in analysing the nature of the single service, but is simply to observe that in these situations, one element (the sale of food, and the actual supply of the vaccine as opposed to its administration) could be regarded as being encompassed in the wider descriptions of “the provision of restaurant services”, and “the supply of medical services.”

71. In contrast there are other quite different situations in which there can rightly be treated as being one single supply, but where the two components are altogether different. In other words neither component can be said to be encompassed in the other in the way that one can say that the provision of food or vaccines are integral components in the supply of restaurant services and medical services. An example of this form of single service is provided by the case of Mesto Zamberk v. Financni reditelstvi v. Hradci Kralove [2014] STC 1703 where one single ticket gave admission to a park in which there was provision for many sporting activities (swimming pools etc), but also provision for other non-sporting amusements and rest areas. The decision is slightly unsatisfactory in that the decision recorded all the sporting facilities, but virtually ignored the amusement and rest facilities. Nevertheless the decision was that there was one single service, and that if the national court was then to conclude that the sporting facilities were predominant, the conclusion would be that all the activities should be so treated. This was not because rest and amusement were comprised within “sporting activities”, but because the two components were sold together and in an economic sense comprised one supply for VAT purposes, and the nature of that supply should be decided by the nature of the predominant supply.

72. Having now addressed the different situations where composite activities might or might not be regarded as constituting a single service (and there may very well be other such situations as well) we now address the three further tests that the courts have adopted for defining the nature of the single supply.

73. The three additional tests are as follows. The first additional test (Test 2) accepts that the elements do not constitute a single supply by virtue of one element being the principal element and the others being ancillary, but arrives at the conclusion that there is a single supply because from an economic point of view the typical customer wants one composite service. Thus the restaurant customer wants a meal in a restaurant, with service and all the elements that he obviously expects, and he treats that as one economic supply. In the Faaborg-Gelting Linien case, the ECJ then applied what we will describe as Test 2, and decided that the supply was of restaurant services. This was not because food was ancillary to restaurant services because it was not. But the predominant supply was nevertheless the provision of the services, and therefore the single supply constituted a supply of restaurant services. The essence of Test 2, therefore, is rather akin to Test 1, where there is a principal supply with ancillary elements. The decisive factor is governed by which element of the supply is predominant.

74. Another ECJ example of the application of the same test is given in the case of Levob Verzekeringen BV and another v. Staatssecretaris van Financien [2006] STC 766 where there was a supply to an insurance company of a pre-existing software program, coupled with the supply of considerable services in tailoring the software specifically to meet the customer’s particular requirements. The designation of the composite single service as a supply of services, notwithstanding that the provision of the software in isolation would have been a supply of goods was based on the observation that the customisation services were the predominant element of the supply.

75. The next test (Test 3) is exemplified by the case of Finanzamt Frankfurt am Main V-Hochst v. Deutsche Bank AG [2012] STC 1951. In this case Deutsche Bank was rendering portfolio management services to clients where essentially it both conducted investment research in order to arrive at appropriate decisions for clients’ securities to be bought, sold or retained, and it then performed the bank’s back-room service of implementing sales and purchases (where appropriate) in accordance with the decisions. The back-room services appeared, in isolation, to qualify for a VAT exemption whilst the investment research function did not. It was held that there was one single service. The Advocate General had indicated that he thought that the single service should be treated as standard-rated because the predominant element of the service was the research, so that when the single service took the VAT nature of the research, the single service would be standard-rated. The Court took a different approach. It held that neither component could be said to be the principal or even the predominant element, but that the two elements were of equal importance, such that they had to be put on an equal footing. When one then described the “two component service” as “the provision of investment research, coupled with back-room implementation”, the composite service did not qualify within the relevant limb of the finance exemptions that had to be construed strictly. Accordingly the composite service was not exempt. So Test 3, applicable where two components are placed on an equal footing, is to describe the single service as comprising the two elements and if that results in the single composite service not being covered by any zero-rated or exempting provision, the service is standard-rated.

76. Test 4 could aptly apply in cases such as Faaborg-Gelting Linien, Dr. Benyon and CEM, though we are going to have to be careful in detecting whether in those cases the test that we describe as Test 4 was actually applied in them. Leaving that aside temporarily, the essence of Test 4 is to say that, although there are various constituent elements to a single service, one description aptly covers the composite service, and the single service should thus be taken to be the service, as described in that overarching classification of all the components. Thus, in Faaborg-Gelting Linien, the application of Test 4 would be to say that the sale of food was merely an integral and non-ancillary component of the overall supply, but that the supply could aptly be described by addressing the overarching supply of “the provision of restaurant services”. Similarly in Dr. Benyon, the application of Test 4 would say that the supply of the vaccine was just one component in a composite service, but once one treated the three elements that we will refer to below as one supply, the overarching description would be to say that there was a supply of medical services. The “supply of medical services” could of course encompass the supply of the vaccine, and would be an apt description of the entire single service. Again, in CEM, once there was analysed to be a single service, the only description of the service that could encompass all aspects was the overarching service of “education”. The single service could not be the provision of books, both because the Tribunal had held that the provision of the books was not an end in itself for the average customer, and because “education” was the only description, in other words the overarching description, that aptly encompassed the entirety of the single service. We might add that there was no, or very little, attention given in the House of Lords’ judgments to the particular test that their Lordships had applied to identify the nature of the service, though since Test 4 and the notion of the overarching description of the single supply was the only test that would have justified the outcome of the decision, it is fair to conclude that implicitly this was the test that had been adopted.

77. Test 4 is therefore the one that dictates the VAT nature of the single service by reference to the “overarching” description of the composite service that aptly describes the entire service. 

78. Before considering which cases have been covered by this fourth approach, it is pertinent to note that both Tests 2 and 4 could aptly be applied in the situation that we considered in paragraph 70 above. Thus the court might have decided the case of Faaborg-Gelting Linien by saying that the overarching description of the composite supply was the provision of restaurant services and that, under Test 4, that should govern the nature of the single supply. The important point to note, however, is that this is not what the ECJ said at all. No reference was made to any “overarching” supply, and the explanation of the decision that the single service consisted of the provision of services was because that component of the overall supply was the predominant element. In other words it was Test 2 that the ECJ in fact applied.

79. Turning now to the Dr. Benyon case (and we will see shortly that this point also emerges in the CEM case), it is not particularly clear that Lord Hoffman, with whom all the Law Lords concurred, particularly identified the test that he was applying. Both cases (Dr. Benyon and CEM) were also very heavily influenced by the fact that the House of Lords in both cases was principally, if not exclusively, considering the issue of whether there were multiple supplies or just a single supply, it simply being assumed that if the right analysis was that there was one single supply, then in the Dr. Benyon case that supply would be of medical services, without much or any attention to which test had occasioned that designation. Similarly in CEM the entire focus (at least by the time the case came to the House of Lords) was on the issue of whether there were two separate services or one single service and as, we will explore below, it was then rather assumed that the single service would be of education and there was no particular attention given to the test that occasioned that conclusion.

80. We will now consider the decisions in Dr. Benyon and in CEM more carefully to ascertain the basis on which the House of Lords concluded in each case how to analyse the nature of the single service.

81. The following two paragraphs in the only judgment given in the Dr. Benyon case, that by Lord Hoffmann, first describe the approach taken by judges in the Court of Appeal which had decided that there were two or three distinct services, and then explain why Lord Hoffmann considered that there had just been a supply of medical services.

“30.  Aldous LJ acknowledged that “at a particular level of generality” it could be said that there was one transaction. But he said that when a doctor administered a drug to a patient he was “in reality dispensing the drug to the patient and then administering it.” Chadwick LJ likewise divided the transaction into three elements:  first, the consultation and diagnosis, secondly the supply of the drug for the purposes of treatment and thirdly its administration. The first stage, he said, was “dissociable” from the second and third and constituted a separate supply. Although there might be some medical skill involved at the third stage, the dominant element was the supply of the drug and it was therefore to be classified as a supply of goods.

31.  … … … this approach seems to me to involve the kind of artificial dissection of the transaction which the Court of Justice warned against in its judgment in the Card Protection case.  In my opinion the level of generality which corresponds with social and economic reality is to regard the transaction as the patient’s visit to the doctor for treatment and not to split it into smaller units. If one takes this view, then in my opinion the correct classification is that which the NHS has always taken of the personal administration of drugs to non-reg 20 patients, namely that there is a single supply of services”.

82. It seems to us that the two critical points to derive from the paragraphs just quoted are first that Lord Hoffmann was principally concerned with the issue of whether there were two or three supplies, or just one, and secondly in deciding that there was just one, and describing that one supply as a “supply of services”, no mention was made of the particular test, and application of principle, that occasioned that decision. In other words, Lord Hoffmann certainly did not say either that he chose the services, in other words “medical services” because the medical services predominated, and he did not say either that that was the overarching description of the overall service. It is this latter observation that led to the remark that we made in paragraph 68 above, to the effect that there might almost be a fifth test, namely the simple application of common sense. In this regard, it is obvious that the individual who had received the vaccination would have said, on leaving the surgery, that “Dr. Benyon has just given me a vaccination”, and would certainly not have said that he had “bought a vaccination”, and he certainly would not have said that he had “received a supply of medical services, followed by a purchase of goods, and followed again by another supply of medical services.”

83. Sometimes, in other words, the correct description of the single supply may just be an application of common sense, and then understandably there may be no attention to predominant elements or overarching descriptions.

84. Some similar observations may be made in relation to the House of Lords’ decision in CEM. Again the whole focus of the debate was as to whether there were two separate services or just a supply of education. No attention was ever given to the other possibility, namely that the dominant supply might have been a provision of books, with other elements being ancillary, though this inevitably followed from the tribunal’s finding of fact that the provision of the books was not an “end in itself”, and that the “end in itself” was to pass an examination and obtain a qualification. Both Law Lords who gave decisions challenged the further finding of fact that the provision of the books was an ancillary supply, saying that that was both unrealistic, and an unnecessary finding once the Tribunal had identified the end result sought by the customers to be the obtaining of a qualification. Accordingly the provision of books, not regarded as an end in itself, was almost bound to be disregarded in analysing the description of the one singe supply.

85. The above observations are well illustrated by the final paragraph of Lord Rodger’s judgment, to the following effect:

“13. In the present case the tribunal, having taken into account all the factors, concluded that the College made one supply, the provision of education. In my view, the tribunal were entitled to reach that conclusion on the basis of the findings which they made – essentially their finding that the students took the courses in order to obtain the relevant qualification offered by the College. The transaction was therefore one which gave the students the opportunity, by successfully studying the printed materials and completing the other necessary steps, to obtain a valuable qualification. That was what the students were purchasing. For the reasons given by Lord Walker, I am accordingly satisfied that the Court of Appeal erred in disturbing the tribunal’s conclusion. On that basis the College made no zero-rated supply of books in terms of s.30(1) of the Act.”

86. Lord Walker’s approach was similar to that of Lord Rodger, exemplified by the passage just quoted, and the essence of his judgment was simply to arbitrate between two separate services and education, and then to respect the tribunal’s decision, once the tribunal had made their crucial finding that the provision of the books was not the students’ desired end result.

87. It was said in the hearing before us that Lord Rodger had adopted the overarching test, but we need to qualify this assertion. The single reference to the overarching notion appeared at a point where Lord Rodger was considering whether there were two services or one, and it seems to us that Lord Rodger was not applying that test to reach the conclusion that a single supply should itself be treated as being of education, rather than of books. Lord Rodger’s self-evident reason for that conclusion, not that the choice between education and books as the description of the single supply was ever mentioned, was clearly given in the passage that we have just quoted in paragraph 85 above.

88. Lord Rodger’s reference to the overarching supply of education was given in paragraph 12, as follows:

“12. But the mere fact that the supply of the printed materials cannot be described as ancillary does not mean that it is to be regarded as a separate supply for tax purposes. One has still to decide whether, as a matter of statutory interpretation, the College should properly be regarded as making a separate supply of the printed materials or, rather, a single supply of education, of which the provision of the printed materials is merely one element. Only in the latter event is there a single exempt supply, to which s. 31(1) of the Act applies and s. 30(1) does not apply. The answer to that question is not to be found simply by looking at what the taxable person actually did since ex hypothesi, in any case where this kind of question arises, on the physical plane the taxable person will have made a number of supplies. The question is whether, for tax purposes, these are to be treated as separate supplies or merely as elements in some overarching single supply.  According to the Court of Justice in Card Protection, for the purposes of the directive the criterion to be applied is whether there is a single supply “from an economic point of view”. If so, that supply should not be artificially split, so as not to distort (altérer) the functioning of the VAT system. The answer will accordingly be found by ascertaining the essential features of the transaction under which the taxable person is operating when supplying the consumer, regarded as a typical consumer.” 

89. The sentence that we have just highlighted indicates that the reference to there being an overarching supply is entirely directed to justifying the conclusion that there was one single composite supply, as distinct from two supplies. It was not expressed to be the basis on which that supply was decided to be the supply of education, not least because that separate issue was never considered by the House of Lords at all.   

90. While we reiterate that the House of Lords’ decision in CEM to describe the single service in the CEM case as a supply of education and not as a supply of books clearly resulted from the approach reflected in paragraph 13 of Lord Rodger’s decision that we quoted in paragraph 85 above, we now need to consider other applications of Test 4, the two best examples being judgments in both cases of Mr. Justice Warren, the cases being Byrom and others (trading as Salon 24) v. HMRC [2006] EWHC 111 (Ch), and Finnamore (trading as Hanbidge Storage Services) v. HMRC [2014] UKUT 336 (TCC).

91. In the first, the appellant supplied a lease or licence of a room in a building euphemistically described as a massage parlour to ladies, and the appellant also provided various massage parlour services to the ladies. Mr. Justice Warren concluded that there was a single supply and that the supply was of massage parlour services. This was essentially on the basis that “massage parlour services” was the overarching description that aptly described the totality of the supplies, once he had decided that there was a single supply.  He then made the important observation that he could still reach this conclusion even if he simultaneously considered that the most important element of the single supply was the licence over the room which would, in isolation, be an exempt supply. In this regard, therefore, and assuming that the licence to use the room was regarded as the predominant element of the supply, Tests 2 and 4 could lead to different results. Test 2 would tend to treat the dominant grant of the licence as the predominant element from which the remaining services should derive their nature, whereas Test 4 could reach the conclusion that the supply was an overarching supply of services even if the most important element was the licence over the room.

92. Exactly the same point arose in the Finnamore case where the appellant granted customers a lease or licence over a small area of an open storage depot to customers, whilst also hiring out a shipping container, resting on the relevant small area of land, so that the customer could keep goods securely, and protected from the weather, in such containers. Again the conclusion was that the overarching description of the supply was of the provision of storage services, and that was so regardless of whether the charge for the lease of the land, and the importance of that element, actually exceeded the charge for the container. The pricing in that case was relatively robust since the container could be hired at the relevant much lower price if it was to be located elsewhere than on the appellant’s premises. Whether of course the ECJ might have applied Test 2 and reached the same conclusion in that case on the basis that the supply of the container (rather than the licence over the land) was the predominant element, since clearly customers wanted storage containers in one location or another, we obviously do not know.

93. Turning now to the parties’ contentions in relation to the law, the Respondents naturally claimed that the right way to address the second part of the enquiry in this case (i.e. the decision as to whether one single supply was of education or books) was essentially to determine the overarching character of the supply, observing that that should be judged in the context of the contention that we mentioned in paragraph 63 above, namely that the overriding object was “to receive an education and obtain a qualification.” By contrast the Appellant contended that a divergence had emerged between the European authorities, all of which resolved the nature of the single supply by resorting to Tests 1 to 3 above, and not to Test 4, whilst Test 4 was one that had been evolved by the domestic courts without guidance from the ECJ. While the Appellant’s counsel invited us to reject the domestic authorities on the basis that the European ones should be treated as of greater binding force, we simply make two observations. Firstly there is little doubt that the case law authorities, which give guidance in relation to the issues of single or multiple supplies, the appropriate description of a single supply, are becoming difficult to apply with confidence. We consider it possible that further guidance will be required from the ECJ. Secondly, however, on our findings of fact, and the crucial distinction between the findings of fact that we have reached and those reached by the tribunal in the CEM case, we believe that this case may be resolved on a rather simpler basis than by reference to some of the points that we have been considering above. In other words, we consider that this case is one where Test 1, applicable to the situation where one supply is the principal supply and other elements are ancillary, is almost certainly the right approach to the designation of the single supply, assuming that we conclude that there has been one supply as distinct from multiple supplies.

94. We will now proceed to give our decision, and to amplify the points just made.

Our decision on the principal issue

95. The first step must be to ascertain the end results expected to be obtained by average customers from the supply or supplies made by the Appellant. That is crucial because if we conclude that there was one dominant end result sought by customers, with other elements being ancillary, those conclusions will inevitably lead to the conclusion that this is a case where there has been one single composite supply and not two or more distinct supplies. Furthermore a conclusion that there was one principal aim, with other elements of service being ancillary to the better enjoyment of that principal aim, is also likely or bound to govern the outcome of the second test for the identification of the correct description of the single service.

Overview

96. We consider that in this case, the end result sought by customers from the supply made by the Appellant was to learn, and to accomplish that aim essentially by reading the vast amount of printed material. 

97. The Appellant’s essential supply was the sale of manuals. We understood that the manuals had been drafted by the Appellant’s parent company, and that the whole aim of the manuals was that by reading and studying them one-by-one, the information in the manuals alone was confidently expected to be sufficiently clear and comprehensive to enable the customers to achieve the level of understanding of the topics required and expected.

98. We accept the Appellant’s contention to the effect that all of the other features of supply were appropriately regarded as “add-on” ancillary functions that ensured, where customers had a need or wish to request receipt of the various functions, that customers should simply go back to the manuals to the page or paragraph indicated, and re-read the manuals.

99. The Respondents’ case before us has been heavily influenced by the proposition, and HMRC’s whole stance itself almost entirely dictated by, the belief, that this Appeal is covered by the principles laid down in the House of Lords’ judgment in the CEM case.

100. We disagree.

101. We agree first with the Appellant’s counsel observation that the House of Lords’ decision was hardly a “ringing endorsement of the tribunal’s decision”, but rather a decision, in the light of those findings of fact that were not overturned by the House of Lords, that the tribunal was entitled to reach the decision that it reached, particularly in the light of the undisturbed finding of fact that the passing of the College’s examinations and the obtaining of the College’s qualifications was the principal aim sought by customers, the study of the books not being an aim in itself. After all “the tribunal had heard all the evidence”, so that the tribunal’s findings should not lightly be over-turned. There was no suggestion, however, that the facts and the law were such that no other decision would have been tenable.

102. The one finding of fact that the House of Lords did undermine was the finding that the provision of the printed material was ancillary to the provision of education. In the CEM case the provision of the printed material was clearly the principal method by which education was delivered. Admittedly in the CEM case there did appear to be other educational supplies in that the College, based at the University of Reading, employed 100 full-time staff, the academic staff were numerous and they undertook the assessment of students with written assignments providing them also with face-to-face teaching and preparing written materials and research.

103. Ignoring at this point the important, if not decisive, issue of examinations and qualifications, the contrasting situation in the present case is that all the preparation of the manuals is dealt with by the Appellant’s parent company, such that essentially the Appellant sells books. The few staff (7 as we have indicated) who have anything to do with TMAs and responding to customer’s rare phone calls and emails, are essentially intended to refer customers back to the manuals, and certainly not to provide further or different education. There could be no function more ancillary to the provision of the manuals, than the role of the 7 employees who were employed to field any enquiries from somewhere between 40,000 and 60,000 customers.

104. The fundamental fact underlying the House of Lords decision in CEM was that the end result sought by customers was the passing of an examination and the obtaining of a qualification, and not to learn from the manuals. The entire final paragraph of Lord Rodger’s decision, quoted in paragraph 85 above, was based on this very point, and more generally the same approach was adopted by Lord Wright.

105. In the present case, whether customers intended to sit examinations set by third parties and to obtain qualifications, such motives were beyond the boundary of the supplies made by the Appellant. Customers might thus have had various motives for entering into their contracts with the Appellant. They may have been later to obtain such qualifications. They may simply have been to enhance their knowledge in a field in which they were already working, such that they may have considered qualifications to be irrelevant. They may have enlisted in the animation type courses in order to foster hobby-type activities. But whatever their motives, and even if in all cases their motives were to obtain qualifications, that was not a conceivable end result sought from the courses provided by, and the supplies actually made by, the Appellant.

106. More practically, we do not actually know how many customers met the conditions in the trade courses for completing all the TMAs and undertaking the practicals, in order to be able to require the Appellant to pay for entrance into third party examinations. It appeared that only a small minority met the relevant conditions, and there was no evidence in relation to whether at the end of their courses customers did indeed take the examinations and receive the qualifications

107. In the 2002 VAT and Duties Tribunal case of International Correspondence Schools Ltd, Decision no 17662, the facts were startlingly similar to those in the present case. The predominant supply was plainly the printed material, and there was the further provision of little-used tutor support. In fact, the tutor support sounded to be rather more extensive than that in the present case. 

108. The short Decision in Edinburgh by the Chairman, T Gordon Coutts, commenced with the following two paragraphs:

“The Tribunal has no hesitation in concluding that what was supplied to the customer by the Appellants was a composite supply and that it was predominantly a supply of a substantial quantity of printed written matter. The material contained in the written matter could stand alone; there was no need for the completion of any questionnaires, any assessments or any involvement of any tutor.

Nothing depended upon that “tutor service”. The manuals themselves were self-contained. No doubt, there was a supply, or a promise of a supply, if required, of additional facilities such as the answering of students questions and, usually the marking of multiple choice examinations, but these are obviously ancillary, as was, perhaps, recognised in the percentage of the taxable element which had been agreed as appropriate before Card Protection Plan was decided finally.

109. It seems to us that that decision realistically identifies the principal supply, and that in the present case the facts are virtually identical. While the above decision pre-dated the decision in CEM, it is worth noting that the tribunal that dealt with the CEM appeal considered that the College in the CEM case was “in a completely different field [from the Appellant in the Edinburgh case] and there is no realistic similarity. It is providing education at a high level as an entity.” In the immediately following paragraph the Tribunal in CEM recorded the crucial fact that “In evidence, Mr. Batho recognised that each course qualification was the end that students sought.”

110. Our decision in this case is essentially identical to that in the Edinburgh tribunal decision. The customer’s desired end result was to educate himself entirely by studying the self-contained manuals.

The appropriate test for assessing whether “add-on” provisions were ancillary to the principal supply

111. Extensive attention was given in the hearing to the issue of how many customers availed themselves in any way or any significant way of the tutor support, and in the case of the trade courses, the practical sessions. As we have already indicated, the answer to that enquiry was broadly that there was relatively little take-up of the tutor support, and even less take-up in the trade courses of the practical sessions.

112. We have ignored much of the anyway inconclusive detail in relation to the quantitive significance of the tutor support and the practical sessions in the trade courses, because we regard the essence of what ranks as an ancillary provision to a principal supply to be dependent  principally on the nature of the add-on provision. Where it is simply supportive of the principal supply, facilitating the better use or enjoyment of the principal supply, and when the add-on functions are manifestly not an end in themselves, we consider that they are plainly ancillary, and would indeed remain ancillary even if, contrary to the facts, very considerable resort was made by customers to these add-on functions.

The degree to which resort was made by customers to the add-on functions

113. In case we should consider the degree to which resort was made by customers to the add-on functions of tutor support, marking of TMAs, the availability of the virtual room and the practical sessions, we are still going to ignore the detail of the minor percentages of utilisation, calculated on various different bases.  Instead, we will make what we believe are sensible general assumptions, but we concede that they may not be entirely reliable. As a generality, we assume that it is possible that the majority of customers may have completed the initial TMAs in responding to the multiple choice questions, though even that is not known for certain. Equally we assume that completion of the TMAs will have declined over the course both because some customers might have abandoned the course altogether, but also because those who continued to request the next manuals, so continuing with the learning, might have regarded the TMAs as irrelevant.

114. Whilst thus the take-up of the TMAs may have been significant, the take-up of the tutor support must be regarded as having been low or minimal, and the attendance at the practical sessions in the case of the trade courses also very low.

115. Another obvious fact in relation to the add-on functions is that many customers appeared to acquire all the manuals, and implicitly to read and study them, and doubtless to retain them for future reference, without actually resorting to any of the offered add-on facilities. While similar information was not collected in the periods relevant to the Appeal, we were told that in 1999 there was a question annexed to the contracts that people signed up at the outset, the question asking whether customers would or would not utilise the tutor support. We were told that between 85 and 90% said that they would not.  It of course remained available to them, and some might have changed their minds, but it was obviously not considered vital by the vast majority.

116. The conclusion that we draw from the fact that the majority, possibly a substantial majority, of customers appeared not to attach any importance to the add-on functions is of course that this amply demonstrates that the course materials were expected to be, and appear to have been proved to be, sufficient in themselves for people to achieve the desired end result. And plainly none of the add-on functions could possibly be regarded as an end in themselves, when qualitively they were no such thing, and secondly when most people ignored them.

The significance of the marketing material

117. We accept the Respondents’ point that the marketing materials sought to make the courses sound more user friendly, more supported and more collegiate, than was in fact the case. That of course is the essence of marketing, and it was obvious that this would be the case.

118. It cannot, however, be said that there was any actual misrepresentation because the elements such as tutor support, the marking of TMAs and the practical courses were all genuine, albeit that they may have been over-emphasised. The question for us is nevertheless the nature of the supplies, albeit as perceived by the customer, and particularly when there was evidence back in 1999, when the courses were little different, that the vast majority of customers were not for instance interested in the tutor support, it seems difficult to conclude that the marketing materials led customers to expect something quite different from the service in fact provided to them. There was no evidence that any customer complained that the course had been miss-sold. More relevantly the Respondents quoted various (possibly or probably genuine) testimonials from customers indicating how useful they had found the tutors. Whilst this was advanced in support of the untenable point that the tutor support was an end in itself or a separate supply encompassed in an overarching supply, it does actually support the Appellant’s implicit assertion (not that it was made) that for a small minority of customers the tutor support did indeed match or even exceed expectations.

119. We attach little importance to the inevitable marketing techniques that were bound to seek to make the courses sound less wholly dependent on the reading materials, and it is certainly possible that many of the customers would have viewed much of the marketing material in a similar slightly sceptical manner.

The two fundamental questions

120. We consider that it is clearly the case that there was only one single supply in the present case. This in our view was because the printed material was the principal supply, entirely sufficient in itself, and the other factors were ancillary in the strict qualitative sense of the word. Were we to have to answer the “single or multiple supply” question by reference to the other test of whether it would be inappropriate, in economic terms, to split the supply into different supplies, we would reach the same conclusion. One price was charged for the courses. Whether people wanted some or all of the add-on functions, they were available. There was no opportunity to enlist in the course by taking the manuals and disregarding everything else. On every basis, there was one single supply.

121. Addressing the second question, once we conclude that there was a principal supply and ancillary add-ons, the single supply takes its nature from that of the principal supply, namely the zero-rated provision of books. Insofar as many of the European authorities have concluded that the nature of the single supply should be based on the nature of the predominant supply, the answer remains of course the same. The European authority, exemplified by the Deutsche Bank case to the effect that a different rule applies where two components of a single supply were of equal importance such that they should be put on an equal footing, is plainly irrelevant in this case. Were we to seek to apply the test that has been adopted in some of the UK domestic decisions, namely the identification of the overarching supply, we would first say that that test was irrelevant in the present case, or that if it was relevant, then the overarching supply was still one of the provision of the manuals. It was absolutely clear in both of Mr. Justice Warren’s decision to which we have referred that where a supply is inherently ancillary to some principal element, then that ancillary element does not have to be treated as distinct for the purposes of arriving at the overarching nature of the composite supply. Thus, Mr. Justice Warren expressly said that he regarded cleaning and lighting the room that was licensed to the ladies to be ancillary functions in relation to the letting. It was the other services, not those ancillary to the licence that resulted in the composite whole having to be analysed from the perspective of the overarching supply of massage parlour services in that case. 

122. On every test, therefore, we conclude that there was one single supply of zero-rated books in the present case. We actually regard only one test as the appropriate one, though the same conclusion results from the application of the other tests.

123. This concludes our decision on the Appeal on the most important point.

124. We will now address the other points, generally of course on the basis that the conclusion just given is wrong. 

Our decision on various Applications

125. As the cover page to this Decision indicates, this case proceeded in a rather unsatisfactory way in that there was an unexpected level of cross-examination in relation to detailed facts during the four days set aside for the hearing in mid-January so that the case had to be adjourned. Counsel’s availability resulted in the remainder of the hearing being deferred until June. One result of this delay was that when we had deferred giving decisions on various Applications to amend pleadings and to raise new points until we had a better understanding of what was involved, by the time the hearing re-commenced in June, the parties had indeed had ample time in which to consider any new points raised, and indeed skeleton arguments had been provided on all the relevant points. Purely for the sake of form, and recognising that by the time of giving this decision on the various Applications, they had all been extensively canvassed by both parties, we accept that the parties were entitled to raise the points to which we now turn. We will slightly qualify that point in one respect when dealing with one specific point below.

The facts material to the additional questions

126. As we mentioned in the Introduction:

·         there had been numerous discussions in the late 1990s between the Appellant and HMRC in relation to whether the Appellant’s supplies should be treated as one single supply or as multiple supplies;

·         in 1999 both parties agreed that there were multiple supplies and more relevantly they agreed on the method for splitting the supplies and the consideration;

·         HMRC’s letter of 14 January 2000 confirmed that method and included the paragraph that we quoted in paragraph 10 above  in relation to the liberty of HMRC to withdraw the method;

·         on 27 August 2009 Officer Rashid challenged not the method but the analysis that there were two separate supplies and decided that for all periods that could be reviewed, the correct VAT position was that all the supplies were of standard-rated education, largely in accordance with the CEM decision, and

·         the Appellants then complained to the division of HMRC that dealt with complaints asserting that the Appellant had a legitimate expectation that the method settled on 14 January 2000 should not be withdrawn on a retrospective basis, and that indeed the method should remain in force until six months after the final resolution of the courts in relation to the correct method.

 

127. Office Harris in HMRC’s complaints team wrote to Grant Thornton, the accountants acting for the Appellant on 22 January 2010, the letter containing the following paragraph:

 

“I have carefully considered your representations, the reports specifically commissioned from officers/managers involved in visiting your client company, the information held in you client’s electronic folder under the above VAT registration and the relevant Departmental Guidance and Notices. On the basis of my investigation I agree that our handling of your client company’s VAT liability issues has been unsatisfactory and that they have a legitimate expectation that they could rely on the “agreed method of apportionment” of 14 January 2000 until the method was reviewed, amended or withdrawn. In accordance with that agreement it could be so reviewed, amended or withdrawn at any time and this was done on 27 August 2009 by Ms. Abida Rashid. Accordingly I will recommend that we take no retrospective action prior to 27 August 2009 over the VAT liabilities in question. I formally apologise to your client company for our failures in handling this aspect of their VAT affairs and would ask you to convey these to Messrs Bradik and Butler”.

 

128. On 22 March 2010, another officer in the Complaints Team, Mr. Winder, wrote to Grant Thornton effectively reversing Mr. Harris’ decision. The relevant paragraphs were as follows:

 

“HMRC has decided that it does not have power to refrain from collecting the tax which it believes is legally due in this case. The visiting case officer Abida Rashid has therefore been instructed to issue any appropriate assessment. Your client will be notified formally shortly.

 

I sincerely apologise for the uncertainty and any confusion caused since Mr. Harris advised you that he was recommending that concessionary treatment be applied. He made it clear that this was subject to approval and I did likewise in my letter of 19 February 2010. After full consideration of the facts by the relevant policy areas, remission has been deemed inappropriate.

 

………

 

I understand Abida Rashid’s assessment will be issued on a “best judgment” basis on the information already available to her. She will be prepared to reconsider the quantum if you feel the amounts assessed are incorrect.” 

 

129. Officer Rashid then made the assessments and by way of adding insult to injury a letter of 13 April 2010 from Officer Rashid even contained the suggestion that as the Appellant’s relevant earlier returns, made entirely in accordance with the agreement of HMRC, contained “an inaccuracy, that meant your claim for a repayment was incorrect. Because of this you may have to pay an inaccuracy penalty.”

 

130. The Appellant then commenced judicial review proceedings, claiming in particular that as had been implicit in the Appellant’s contentions at all times, the basis of its claim that back tax assessments should not be made was based on the legitimate expectation that  the agreement in the letter of 14 January 2000 would apply until withdrawn (as indeed it appears that Mr. Harris had understood the contention), and that the Appellant’s request was not for concessionary treatment, as the letter from Mr. Winder seemed to assume. The amount in contention, up to the 27 August 2009 date, was £16,293,339 in tax and £1,417,824.01 in interest.

 

131. We understand that prior to the hearing of the judicial review application, HMRC changed tack once again and conceded that the assessments that had by then been made for the earlier periods, prior to 27 August 2009, should not be enforced. HMRC’s letter dated 27 May 2010 contained the following two paragraphs:

 

“Further to your letter of 13 May 2010 and our subsequent telephone conversation my Complaint Officer, Mr MJ Harris, has been in fresh contact with our specialist advisers at Headquarters with the outcome that they have reconsidered their original advice. You will recall I summarised HMRC’s then position in my letter to you dated 22 March 2010.

 

As a result of the reconsideration I can confirm today the original position that MIS could indeed rely on the terms of the 2000 agreement until withdrawn on 27 August 2009. It follows that Mr. Harris will now restart his recommendation process and in this respect I would appreciate you letting him know whether or not MIS have or would suffer any detriment if HMRC failed to exercise discretion in MIS’s favour by not taking any action to pursue the assessment formally notified by letter dated 26 March 2010. If so, please advise him of the extent of that detriment.

 

132. As we indicated in the Introduction, that resulted in the back tax for which assessments had been made not being collected, but in due course the Appellant still complained about two issues.

 

The refusal to repay interest supplement in respect of the delayed repayment, consequent upon the “last period” review of the VAT calculations under the “method” for the year 2006

 

133. The first further complaint that we now address relates to the fact that once HMRC conceded that they had to honour the January 2000 agreed method of calculating VAT up until 27 August 2009, it followed that HMRC would indeed have to make the substantial repayment to the Appellant for the final quarter of the 2006 period that we mentioned in paragraph 19 above. HMRC was sufficiently late in actually repaying the amount in question that the Appellant demanded “repayment supplement” under section 79 VAT Act 1994.

 

134. It appears that HMRC conceded that repayment supplement would be due to the Appellant, save for the one fact that repayment supplement was only due where the claimant had a VAT credit, namely an excess of input deductions over output liability. HMRC’s claim was that since Officer Rashid had decided that the correct treatment for the period in which the repayment was sought was that all supplies should have been standard-rated, that resulted in the fact that instead of the Appellant being entitled to a VAT credit for the period in question, the strict position was that there was a significant net liability to VAT. Accordingly the Appellant had no “VAT credit”. Naturally it was accepted that Officer Rashid’s decision, and the assessments that appear not technically to have been withdrawn, were being ignored as a result of HMRC’s acceptance that the Appellant did have a legitimate expectation that it could rely on the January 2000 ruling, but nevertheless the strict position was still that the Appellant had no excess of input tax over output liabilities, and so no “VAT credit”. The amount that HMRC had paid was compensation due in respect of HMRC’s obligation to respect the agreed method, and the Appellant’s legitimate expectation. It was not a repayment of actual VAT.

 

135. The Appellant’s contention was that when a taxpayer had a legitimate expectation that for instance  VAT would be calculated in a particular manner, this gave the taxpayer a legally enforceable right to such treatment, and that there was no notion that it was only by concession that the strict position was not pursued. The Appellant relied on the decision of Mr. Justice Blake in R (oao Lower Mill Estate Ltd and Conservation Builders) v. HMRC CO/4721/2008 where he had held that when the builders making the claim in that case had agreed with HMRC the appropriate method of calculating VAT on its supply, the legitimate expectation that that method would be respected gave the builders the legal right that that method should be available unless and until withdrawn. Mr. Justice Blake considered that this was particularly important in relation to VAT since if the taxpayer relied on the method, thereby paying less VAT than was later claimed to have been strictly due, and also of course charging its customers lesser amounts of total VAT-inclusive consideration for its services, if the method was later withdrawn retrospectively the taxpayer would face great hardship because it would then be highly impractical, if not impossible, for extra amounts to be charged to customers in respect of the additional VAT liability.

 

136. Our decision in relation to the “repayment supplement” issue is relatively simple.

 

137. It first obviously follows that if our decision on the principal point is not over-turned on appeal, it will then follow that repayment supplement will be due, and we imagine that HMRC will cease disputing the issue. For the position will then be that the repayment was plainly due as a matter of the application of the strict statute law, and not just because of the agreed method and the Appellant’s legitimate expectation claim. All the supplies made by the Appellant in the period would have been zero-rated, the actual excess of actual input tax over actual output liability would have been very much greater than the amount claimed, and repayment supplement would thus plainly be due when HMRC was late in repaying the claim.

 

138. On the alternative basis that our decision on the principal issue is overturned on appeal, we consider that the Respondents’ contention is correct and that, notwithstanding the legally binding right on the part of the Appellant that it should be able to rely on the 2000 agreed method, it would not strictly have an excess of input tax over output liability, and thus would have no “VAT credit”, and no strict entitlement to repayment supplement. We do not remotely dispute that a successful claim to be able to rely on a contract with, or a representation from, HMRC that occasions a legitimate expectation that some conduct can be relied upon leads to an enforceable right. The enforceable right however is simply that the strict statutory position cannot be relied upon by HMRC where to do so would conflict with the claimant’s established legitimate expectation. It is not in other words the case that the legally enforceable legitimate expectation changes the statutory position. It merely precludes HMRC from enforcing the strict position.

 

139. We therefore agree with the Respondents and accept that if our decision in relation to the principal issue is overturned on appeal, then it will follow that the Appellant’s strict entitlement to repayment supplement will be undermined on account of the fact that the Appellant will not have had “a VAT credit”.

 

140. This is not to say that we find it acceptable that HMRC has refused to concede the equivalent financial recognition of the fact that it was late in repaying the claim in respect of the adjustment to the 2006 figures. We both consider that when HMRC has accepted that on account of the January 2000 letter and the agreement that that letter recorded, HMRC should treat the Appellant as if the agreed method had remained entirely valid until August 2009, then that is precisely what HMRC should have done. And the notion of treating the Appellant as if the method had remained valid would have plainly entitled the Appellant to be treated in an identical manner to the treatment that would have been appropriate had the method in fact reflected the correct legal positon. That treatment should have involved HMRC in making a payment equivalent to interest supplement under section 79.

 

141. The Appellant appears to have made no claim in this regard, as opposed to the claim that we have rejected, namely that it had a strict entitlement to actual repayment supplement. Were it to do so now by way of seeking judicial review, the Appellant would plainly be some years late in instituting proceedings, and in the tribunal appeal before us, we consider that we would have no jurisdiction to deal with the Appellant’s complaint even if it had been advanced. We did not understand the Appellant to be advancing any such claim before us in any event. This, however, does not change the statement made in the previous paragraph to the effect that we have found the refusal by HMRC to make some equivalent payment disappointing.

 

The Appellant’s complaint that when the January 2000 “method” was withdrawn in August 2009, HMRC should not only have refrained (as eventually they did) from withdrawing the method retrospectively for all past periods for which they could make assessments, but in addition HMRC should have allowed the 2000 “method” to continue in force for the remaining future payments under contracts entered into prior to 27 August 2009, where the contracts rendered it impossible for the Appellant to charge appropriately increased prices to customers.

 

142. As we indicated in the Introduction, the contracts sold to customers through one finance company were dealt with for VAT purposes on the basis that as the payments were paid in instalments, so too the supplies should be treated as being made for VAT purposes over the life, generally 36 months, of the contracts. There was considerable confusion as to the nature of finance, if any, provided by the relevant so called financier, and there was some discussion as to why VAT was being spread over the life of the contracts, rather than all paid at the point the contracts were entered into. There was no relevant dispute about either of these points, however, and it was common ground that in August 2009, there were future payments to be made under these contracts;  the liability for VAT was equally spread over the life of the contracts;  the Appellant had no means of changing the agreed consideration payable by the customers, and the additional VAT that would become payable were HMRC’s “single supply of education” analysis to be applied immediately to the run-off of these contracts, was approximately £5 million.

 

143. This issue involved three different arguments advanced on behalf of the Appellant to the effect that the First-tier Tribunal did have jurisdiction to deal with the issue of whether HMRC should not have sought to assess the contracts during their “run-off” periods on the “single supply of education” basis, albeit that one of the three arguments was only to be advanced if the case went to, or beyond, the Upper Tribunal. 

 

144. The Respondents contended that we had no jurisdiction on any of the three suggested bases, but that in any event the Appellant’s claim was unsound. The January 2000 letter had indicated that the agreed method could be withdrawn at any time; nobody had objected to that proposition when the letter was received, and the particular contention about the contracts in “run-off” had only been raised a very long time after the 2009 correspondence. The Appellant’s initial claim had been that it should have been given a six-month period in which to reorganise its affairs and its billing practices, and that claim was modified to request a  six-month period following the final resolution of the substantive appeal as to how the ongoing supplies should be dealt with. The particular argument in relation to endeavouring to grandfather the pre-August 2009 contracts in their “run-off” periods had only been raised not just months, but years after, the August 2009 decision.

 

The Upper Tribunal’s decision in HMRC v. Abdul Noor

 

145. The first contention in relation to jurisdiction, the one that the Appellant conceded that we could not accept, was that the Upper Tribunal decision in relation to the lack of judicial review jurisdiction at the level of the First-tier Tribunal in the case of Abdul Noor was wrong. Obviously we are bound by this decision, and simply note that the Appellant reserves the liberty to raise this issue if the case goes to or beyond the Upper Tribunal.

 

The First-tier Tribunal decision in Hollinger Print Ltd v. HMRC

 

146. The second contention was based on the decision by Judge Hellier in Hollinger Print Ltd v. HMRC TC 03117 where HMRC had made additional assessments for past periods because certain printing activities emerged to have been properly standard-rated and not zero-rated. The appellant had advanced various objections to such retrospective assessments, one of which was that it was unfair for the appellant to be burdened with the cost of extra VAT. Had the appellant appreciated that the relevant supplies had been standard-rated when the supplies were made, it would have been simple to charge the customers additional amounts, reflecting the VAT, and the vast bulk of the customers would have been indifferent to the extra charges since most were fully taxable businesses to which the additional consideration, and input tax, would not represent a cost. The liability at the later date for additional VAT would be one that in practice the Appellant would be unable to recover from past customers, and it would result in the VAT becoming a cost, when in the first place it would have flowed through to customers and not represented a cost at all.

 

147. Judge Hellier noted that HMRC’s power to assess when a trader’s return had been incorrect provided that HMRC “may make” a revised assessment, in other words using wording conveying the notion that there was some discretion as to whether to assess or not. We agree that it must be entirely proper for an HMRC officer to give attention, when considering whether he should make a revised assessment, to consider such matters as the trader’s legitimate expectations. As was decided in the Lower Mill case, a legitimate expectation may not modify the statutory position that strictly additional tax may be chargeable under the statutory provisions, but if the HMRC officer rightly perceives that the trader has a legitimate expectation that the tax should not be charged, that is a legal right that the officer can and should respect. The notion of the proper administration of the tax does not, in other words, require the officer to make assessments that comply with the statutory provisions but not the trader’s legitimate expectations, only later for the decision to be over-turned in a judicial review hearing. It is obviously entirely in order, and indeed the correct course, that the officer should pay due regard to all the factors likely to found a successful application for judicial review were they to be ignored.

 

148. When there is then a right of appeal to the First-tier Tribunal in respect of the making of an assessment under section 73, or the amount of such an assessment, the crucial question is whether that implicitly gives the tribunal jurisdiction to rule on judicial review considerations that the assessing officer ought or ought not to have considered, or whether the tribunal’s jurisdiction, as usual, is confined to the application of best judgment in applying the strict statute law alone. We consider that the feature that the statutory provisions have given a limited jurisdiction over judicial review matters to the Upper Tribunal, but none to the First-tier Tribunal, and the general trend of the authorities to the effect that the tribunal’s function is to arbitrate on the strict legal position in relation to the application of the tax provisions, rather than judicial review considerations, is the decisive point. We agree with various later tribunal decisions that express wording would be required to give the First-tier Tribunal what we might refer to as an unexpected, limited and novel jurisdiction over certain judicial review considerations. There was no such wording in the statutory provisions relevant in the Hollinger Print case, or in the present case.

 

149. We accordingly conclude that the Hollinger Print case does not give us jurisdiction to consider the judicial review matters relevant to the Appellant’s final complaint.

 

Jurisdiction under section 84(10)

 

150. The Appellant’s third contention to the effect that we have jurisdiction to deal with the Appellant’s claimed legitimate expectation in relation to the run-off treatment of the pre-August 2009 contracts was based on the rather obscure provision of section 84(10) Finance Act 1994.

 

151.  Section 84(10) provides that:

 

“(10) Where an appeal is against an HMRC decision which depended upon a prior decision taken … in relation to the appellant, the fact that the prior decision is not within section 83 shall not prevent the tribunal from allowing the appeal on the ground that it would have allowed an appeal against the prior decision.”

 

152. Before embarking on the eventually unsuccessful endeavour to discern the circumstances in which this sub-section is designed to operate, we should say that the Appellant’s contention was that Officer Rashid’s assessments were the relevant later decision. While our decision cannot normally extend to the consideration of judicial review objections to assessments made by Officer Rashid, the claim in this case was that, because the Officer’s later decision depended on the earlier decision of Officer Winder that the assessments should be made (an earlier decision over which we had no power to hear an appeal), we can then adjudicate on Officer Rashid’s later decision (to assess) and in so doing we can reach the conclusion that Officer Winder’s decision was wrong because it failed to pay due regard to the Appellant’s legitimate expectations, with the consequence that we could then hold that the relevant later decision, namely Officer Rashid’s decision, was itself wrong.

153. We confess that we actually fail to understand the intended application of section 84(10) and we are not persuaded that the contentions of either counsel clearly identified the intended application of the section. Having just observed, in the paragraphs above, the general expectation that the First-tier Tribunal has no jurisdiction to deal with judicial review complaints, it seems quite extraordinary for it to be suggested that section 84(10) deliberately confers this jurisdiction in the complex case where one decision is dependent on another, and the only thing wrong with either decision is the failure to pay regard to legitimate expectations.

154. Both parties seemed to share the commonly held belief that section 84(10) was introduced in order to change the effect of the House of Lords’ decision in HM C&E v. JH Corbitt (Numismatists) Ltd [1980] STC 231. The strange feature of this suggestion is first that the House of Lords’ decision appeared to us to have been not only correct but entirely cogent in general terms, such that we fail to understand why the decision needed to be over-turned or modified in some way by the introduction of section 84(10). Furthermore the relevant decision against which the appeal had been brought before the Tribunal did not appear anyway to have been depended on some prior decision.

155. In the JH Corbett case the essential point was that if the appellant was to establish that its method of accounting was satisfactory for the purposes of the relevant margin scheme in relation to its gold dealings, it had either to show that its scheme complied with certain requirements in a publication issued by HMRC referred to as the Blue Book, or else the commissioners had to have “recognised that [the appellant’s accounting records] were sufficient for those purposes”.  There was no dispute that the conditions of the Blue Book had not been complied with and so the only question was whether the VAT and Duties Tribunal had power to hear a case where it was alleged that the commissioners should have considered the appellant’s accounting records to be sufficient, when they had in fact decided the reverse. Not least because the decision had nothing to do with the terms of the Blue Book, we fail to see which was the earlier decision that might have influenced what appears to us to have been the only decision, namely that the records were not sufficient. In relation to that the House of Lords simply concluded that the Tribunal had no jurisdiction in relation to the exercise of the discretion as to whether the records were sufficient, and that that matter was one for judicial review.

156. The two later cases involving section 84(10) have not particularly clarified the situation since in each case it was held that the sub-section did not apply but regrettably not for any reason that has illuminated the intended scope of the sub-section. 

157. In HM C&E Commissioners v. Arnold [1996] QBD 1271, Mr. Justice Hidden held that the failure of the Commissioners to allow the appellant’s case on the basis of an extra-statutory concession was one that the VAT and Duties Tribunal had no jurisdiction to entertain, and that the appellant should have brought judicial review proceedings instead. The non-application of section 84(10) was based either on the proposition that there was no “earlier decision”, or that the earlier decision had to relate to the particular appellant. Mr. Justice Hidden held that the decision not to treat the extra-statutory condition as applicable was simply the reason for refusing the appellant’s claim and not some prior decision. In the alternative, the terms of the concession related to all taxpayers and therefore there was no earlier decision just affecting the appellant.

158. Similarly in the case of HM C&E Commissioners v. National Westminster Bank plc []2003] STC 1072 while there had been two decisions, they were on different topics, and one did not influence the other. One related to whether the appellant would be unjustly enriched by receiving a repayment of VAT when it would not have to pass on the refund to earlier customers. The other related to the complaint that other leasing companies had enjoyed different treatment from the treatment afforded to National Westminster Bank’s leasing subsidiary, Lombard. Since the one decision did not influence the later decision, as they related to altogether different issues, section 84(10) was not engaged and the VAT and Duties Tribunal had no jurisdiction.

The application of section 84(10) in the present case

159. It is contended that section 84(10) does confer jurisdiction on the First-tier Tribunal in this case because Officer Rashid’s decision was dictated by the prior decision of Officer Winder.

160. Whilst in general terms we consider it extraordinary, and surely not Parliament’s intention, that the First-tier Tribunal should be given judicial review jurisdiction merely because a later decision was influenced by an earlier decision in which an officer may have failed to address some legitimate expectation issue, we actually decide that section 84(10) is irrelevant in the present case for a different reason.

161. It seems to us that the only relevant decision that Officer Winder made was that HMRC could and should not forego collecting tax for past periods that was properly due even though there was a prior agreement between HMRC and the Appellant that an agreed method of calculating VAT would apply until withdrawn. That decision may once have influenced Officer Rashid to make the back-tax assessments that she made. Those assessments for periods prior to August 2009 have, however, either been withdrawn, or are, by agreement, not being pursued and nothing in relation to them is presently in issue. Neither of the Officers in the Complaints Team, or indeed those at headquarters giving them instructions, appeared to take any decision in relation to the point that is presently in contention, namely the issue of whether future supplies, after August 2009, albeit under pre-August 2009 contracts should be taxed on the “single supply of education” analysis. That seems simply to have derived from Officer Rashid’s belief that she should assess everything, following the withdrawal of the 2000 method in accordance with the view that she took of the correct legal position.

162. Our decision is accordingly that section 84(10) is not engaged because the relevant earlier decisions simply related to the periods prior to August 2009 and had nothing to do with the assessments in relation to the later supplies under the pre-August 2009 contracts.

The substance of the Appellant’s claim in relation to the later supplies under the pre-August 2009 contracts, and the Respondents’ claim as to why, if we had jurisdiction, we should still reject the Appellant’s claim

163. Having heard all the arguments in relation to the substantive claim in relation to the treatment of the pre-August 2009 contracts, as well as the contentions about jurisdiction, we will now indicate how we would have addressed this issue had we concluded that we had jurisdiction to deal with this issue. Indeed, should we be held to have been wrong in deciding that we had no jurisdiction, then the following observations will indicate what our decision would in that event have been.

164. This question breaks down into two questions, one of which we will not address though we will indicate the nature of the relevant first question.

165.  The first question relates to the fact that although the Appellant commenced judicial review proceedings within the short required time-scale, they were initially directed to challenging the retrospective application of HMRC’s “single supply of education” analysis to periods prior to August 2009. Admittedly the appellant requested a six-month period in which the old method should remain in force and this initial request was later modified by a request that the old method should continue in operation until six months after the final decision of a court in relation to the correct treatment. We understand that it was only after an interval of some years that the Appellant advanced the claim currently before us, namely the proposition that the pre-August 2009 contracts, where later supplies are treated as being made after August 2009, should all be grandfathered. Whether the Upper Tribunal, to which any remaining judicial review issue has been referred by Mr. Justice Warren, would allow pleadings to be amended after the long period of time is a matter on which we make no comment. This is the one Application that we indicated we would not deal with in the last sentence of paragraph 125 above.

166. We turn finally to the substantive issue.

167. The Respondents’ contentions were that the Appellant’s complaint in relation to the run off of the pre-August 2009 contracts was in any event unjustified because the Appellant had been notified in the letter of 14 January 2000 that the agreed method could be withdrawn at any time. Furthermore nobody had complained about that proposition at the time, and in particular nobody had indicated that the Appellant was making some supplies under contracts where firstly the consideration could not be altered and increased by the Appellant, and secondly the supplies would be correctly treated as made for VAT purposes as the consideration was received, in other words over the life of the contracts.

168. The Appellant contended that the clear better interpretation of the letter of 14 January 2000 was that the reference to withdrawing the method at any time related really to adjusting the method for allocating the supplies and the consideration into the two categories of zero-rated and standard-rated supplies, and that HMRC were not envisaging, in the paragraph of the letter that we quoted in paragraph 10 above, that they might, with no prior notice, completely change the whole basis of assessment that was not as such the matter that was really the subject of the agreement confirmed by HMRC in the relevant letter.

169. There was also some debate between the parties as to whether HMRC had been aware of the significance of the pre-August 2009 contracts, and the inability of the Appellant to change the pricing under those contracts. We were told that on several occasions the Appellant had referred to this feature, and the Appellant claimed that HMRC had been fully aware of it.

170. Had we had jurisdiction in relation to this matter, our decision would have been as follows.

171. We certainly conclude that the indication in the 14 January 2000 letter that the method could be withdrawn at any time was quite clearly envisaging that the method might be modified or some other quite different method might be introduced, but the focus was still on the method for dividing zero-rated supplies from standard-rated supplies. This, after all, is what the discussion, preceding the letter, had been all about, and the very reference to withdrawing the method, as distinct from insisting on a totally different analysis of the supplies that rendered any method for splitting single charging between two categories of supplies irrelevant, made it clear that the focus was on a change to the method. Furthermore, while in a rather “strong-arm” manner the letter referred to withdrawing the method without any prior notice or discussion, this seems to have been rather unrealistic. Beyond the obvious fact that HMRC would almost inevitably have needed to enter into discussions in order to be confident that some new method, or even a wholly new basis of analysing the transactions, was appropriate and realistic, the practice between HMRC and the Appellant had always been that there would have been sensible discussion before any decision was reached as to how supplies should be subjected to VAT in future.

172. Much the most important question is of course the issue, once HMRC has conceded that the Appellant’s legitimate expectation contention was sound in relation to pre-August 2009 supplies, of whether that same expectation realistically extended to the treatment of the pre-August 2009 contracts in their run-off period. We consider that it should have done. Whenever a legislative change is made, then with the rarest of exceptions, the change is not made retrospectively, and whenever future treatment under pre-existing contracts may be an issue, such future treatment is also almost invariably protected. For instance in the case of changes to capital allowance legislation in the direct tax field, if a change is made to the availability of allowances, that change will only affect future expenditure, and then there will inevitably be protection from the change for future expenditure incurred under contracts entered into before the announcement of the proposed change in legislation.

173. We note that in the present case the Appellant assumed the risk in relation to the contracts that we are considering, that if the rate of VAT was increased during the life of the relevant contracts, the Appellant would be unable to recover its increased VAT cost from customers. This was not particularly discussed during the hearing, but when VAT was effectively only charged in relation to 25% of the supplies, and a change in the rate of VAT would generally be of no more than 2%, the effective extra cost to the Appellant of a rise in the rate of VAT would be minimal. Furthermore we can imagine that the prospects of inserting into contracts with customers clauses providing for extra consideration to be paid in this event would have been a complication that would have had to be explained to customers, and a somewhat unappealing feature from a sales point of view. So we can well understand why the Appellant considered that it had to take the risk, and of course in the opposite direction with a reduction in the rate of VAT, the benefit, of changes in the rate of VAT. The feature, however, that the charge to VAT might rise by a factor of four times, with 100% rather than 25% of the supplies being charged to VAT is of an altogether different order.

174. There was discussion during the hearing, as we have said, as to whether HMRC was aware of this particular issue in relation to the treatment of the supplies under these particular contracts. The Appellant asserted that HMRC’s conduct was more challengeable because HMRC had been aware of the issue in advance. We do not see that this is relevant. Even if HMRC had been ignorant of the basis on which these contracts were subjected to VAT, we still consider that when the point was subsequently drawn to their attention, the proper course would have been for HMRC to accept that these contracts should be excluded from the application of the “single supply of education” analysis during their run-off period.

 

Summary conclusions

175. In summary form, our decisions are that:

·         on the principal issue, we decide that the Appellant’s services were entirely zero-rated supplies of books, coupled with ancillary services that take their nature from the principal supply;

·         unless the decision on the principal issue is overturned on appeal, it follows that the Appellant is entitled, as a strict matter, to repayment supplement in respect of the final adjustment for the 2006 year;

·         if the decision on the principal issue is overturned on appeal, the Appellant will not strictly be entitled to repayment supplement in respect of the final adjustment for the 2006 year;

·         we have no jurisdiction, on any of the three contentions advanced, to adjudicate on whether the pre-August 2009 contracts should have been grandfathered from the “single supply of education” analysis, assuming in that regard of course that our decision on the principal issue has been overturned on appeal; though

·         as a substantive matter, and had we had jurisdiction, and ignoring any issue of amending pleadings, we would have decided that the pre-August 2009 contracts should have been protected from any change in basis.

Costs

176. We are unaware of how the Appeals were classified in terms of being Complex or Standard. Furthermore neither party applied for costs during the hearing. Should the Appeal be one where costs might be awarded, the parties are invited to make applications for their costs in writing.

Right of Appeal

 

177. 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.

 

HOWARD M. NOWLAN

TRIBUNAL JUDGE

RELEASE DATE: 2 OCTOBER 2015