Mandarin Consulting Limited v The Commissioners for HM Revenue and Customs [2021] UKUT 0292 (TCC) [2021] UKUT 0292 (TCC)

UPPER TRIBUNAL

TAX AND CHANCERY CHAMBER

[2021] UKUT 0292 (TCC)

Case No UT/2020/000389

Between

MANDARIN CONSULTING LIMITEDAppellant

THE COMMISSIONERS FOR HER MAJESTY’S REVENUE & CUSTOMSRespondent

Before

MR JUSTICE MEADE

JUDGE JONATHAN RICHARDS

Natasha Barnes (instructed by the General Counsel and Solicitor for Her Majesty’s Revenue & Customs) for Respondents

Date 23 November 2021

Category: Tax

[1]The appellant company (“Mandarin”) supplies career coaching and support services to students of Chinese origin. Those services would be outside the scope of VAT if supplied to persons whose usual residence was outside the EU. Mandarin did not take the steps required by Council Implementing Regulation 282/2011/EU (the “Implementing Regulation”) to establish the usual residence of its customers before the time it made supplies to them. In a decision reported as Mandarin Consulting v HMRC

[2020] UKFTT 228 (TC)

(the “Decision”), the First-tier Tribunal (Tax Chamber) (the “FTT”) held that these failings precluded Mandarin from establishing that its supplies were outside the scope of VAT. With the permission of the FTT, Mandarin appeals against that decision. The decision under appeal

[2]No appeal is made against the FTT’s findings of fact. Moreover, the issues between the parties have narrowed since the Decision was released since a number of the FTT’s conclusions on those facts are not challenged. Accordingly, in this section, we will not summarise the entire Decision, but rather will set out the principal facts, and the FTT’s conclusions on them, which are sufficient to put the appeal into context. References to numbers in square brackets are to paragraphs of the Decision unless we say otherwise.

[3]The FTT found (at [114] to [119]) that the services Mandarin was supplying constituted the services of consultants, falling within Article 59 of the Principal VAT Directive (the “PVD”) rather than educational services falling within Article 54. That conclusion, which is not appealed, is important because the rules applicable to the place of supply of educational services are different from those applicable to consultancy services. Because Mandarin was supplying consultancy services, its services were treated, by Article 59 as supplied where the recipients of those supplies, who were non-taxable individuals, had their permanent address, or usually resided.

[4]Accordingly, establishing the place of supply of Mandarin’s consultancy services involved determining(i) who the recipients of those services were and

(ii) where those recipients had their permanent address or usually resided.

[5]The answer to issue (i) changed over periods relevant to this dispute largely because, from July 2016 onwards, Mandarin contracted with students’ parents, rather than with students themselves. The FTT’s conclusion, which is not appealed, was that until July 2016, Mandarin supplied its services to the students themselves, but from July 2016 onwards, it supplied its services to students’ parents.

[6]This conclusion had a knock-on effect on issue (ii). It was common ground that students’ parents had their usual residence in China ([6]). Therefore, from July 2016 onwards, the FTT held that Mandarin’s supplies were outside the scope of VAT (because Mandarin was supplying consultancy services to students’ parents who had their usual residence in China). That left the question of how issue (ii) should be determined prior to July 2016, when Mandarin’s services were supplied to students. Since Mandarin did not typically gather information as to students’ “permanent 3 addresses” in China (although they did typically obtain information on where they were living when at university in the UK), issue (ii) reduced to an analysis of where students had their “usual residence” prior to July 2016.

[7]As will be seen, Mandarin obtained relatively patchy information relevant to the usual residence of individual students. Therefore, at least in part, its case before the FTT relied on propositions about those students generally. The FTT summarised the following propositions made by various of Mandarin’s witnesses and since the evidence of those witnesses was not challenged to any significant extent, we will proceed on the basis that the following facts were established:(1) The students came almost exclusively mostly from mainland China, as did their parents. Their parents were almost exclusively resident and usually resident in China ([23] to [24]).

(2) Typically Mandarin identified students who might be interested in its services by going to the UK universities at which they were studying ([25]).

(3) The parents of students typically paid Mandarin’s fees, rather than the students themselves ([25]).

(4) Mandarin’s coaching was intended to help students from mainland China obtain jobs with large Western international business (including investment banks and accounting firms). That would be achieved by helping students to navigate the different cultural norms that operated in such businesses, understand the process of applying for jobs and focus on “soft skills” that would not always have been developed as part of their education in China. Parents and grandparents were prepared to pay for Mandarin’s services because they wanted to give the students the best possible advantages ([26], [29], [30] and [33]).

(5) Once Mandarin had received payment, the first step of its programme would involve Mandarin giving a student information on application requirements of the sorts of companies to which the student wished to apply. Some of this coaching involved students participating by video conference from “home in China” (which we assumed to be a reference to their parents’ homes where the students would stay during vacations). As part of this process, students were invited to produce a Western-style CV and would receive feedback on that CV from a coach at Mandarin ([36], [46], [47]).

(6) Mandarin maintained electronic folders for each student receiving its services. There was no express finding to the effect that the CVs described at [(5)] were invariably stored on those folders. However, that seems to be implicit from [59] of the Decision which refers to email correspondence between Mandarin and students being stored electronically in the folders. We are prepared to infer that the process of students preparing CVs and Mandarin commenting on them would have taken place by email and we were taken to some examples of electronic folders for particular students during the hearing before us which included copies of CVs. 4

(7) Mandarin put into evidence a sample of 12 electronic work folders selected at random and their contents to which the FTT referred as the “Sample”. A further 10 electronic work folders, which had not been selected randomly, were also referred to. The FTT referred to these 22 work folders as the “Enlarged Sample” ([60] and [61]).

[8]It was common ground before us, as before the FTT, that the time at which Mandarin made supplies to students for VAT purposes was the time when it received payment (typically from students’ parents or grandparents). It follows that Mandarin obtained CVs from students, and the specific information contained in those CVs, after the point at which the supply to those students had been crystallised.

[9]At [62] to [71], the FTT made some findings on the contents of the work folders. It looked at all 22 work folders in the Enlarged Sample but it paid particular attention to the 12 students in the Sample. The FTT concluded:(1) It was not satisfied that the 12 folders contained in the Sample were a representative random sample ([62]).

(2) With one exception, none of the students referenced in the Enlarged Sample had any family ties in the UK ([67]).

(3) Mandarin had no system in place to check where students were usually resident and it made no systematic attempt to establish the place of usual residence of its students ([70]). At [148], the FTT said:

[148][Counsel for Mandarin] submitted that the vast majority of candidates had their permanent address or usual residence in China and not in the UK. However, in our opinion the Appellant’s records as shown in the Sample did not sufficiently demonstrate this.

[11]At [170] to [174], the FTT rejected HMRC’s arguments to the effect that students’ entitlement, under Tier 4 visas, to reside in the UK while attending their university course, or their residence in the UK while actually attending that course, established a “usual residence” in the UK. That conclusion is not challenged.

[12]At [175] to [180], the FTT considered Mandarin’s compliance or otherwise with the provisions of the Implementing Regulation. After quoting from Article 23 of the Implementing Regulation at [175], the FTT said:

[176]It seems to us that this is the key provision. Article 23, worded in mandatory terms, places the burden on the supplier (i.e. Mandarin) to establish the place where the recipient of the services usually resides. That must be done based on factual information provided by the customer. The supplier must then verify that information. Article 23 is directly applicable (Article 65).

[177]In our view, Mandarin did none of this prior to July 2016. It had no system for checking or verifying the usual place of residence of the candidates prior to July 2016. Indeed, Mandarin seemed to be unaware of the importance of asking for that information in the first place. The 5 work folders appeared to us to be fragmentary and incomplete. In five out of twenty-two cases (all pre- July 2016), there was no contract on file. In two cases, the work folders contained both a passport and a visa for the candidate. In two cases, the work folders contained a copy of the candidate’s passport and in three other cases a copy of the candidate’s visa. However, in all of the cases where there were passport or visa details in the relevant work folders, these concerned contracts entered into by Mandarin with the parents after July 2016. Before that time none of the folders contained passport or visa details. Before July 2016 there was no other personal information about where the candidate had come from or, for example, who the candidate’s next of kin might be. Before July 2016, at the time when a candidate signed their contracts with Mandarin and paid Mandarin’s charges, Mandarin did not appear to request any kind of CV from the candidate which might have indicated where the candidate’s family ties lay and which might be subject to verification. The CVs on the file seemed to have been supplied subsequently as part of the coaching process. In short, Mandarin simply did not carry out the verification exercise required by Article 23. It was therefore unable to prove that the usual residence of the candidates (before July 2016) was outside the UK.

[13]The FTT went on to express the following conclusions at [180] and [183], which are not challenged in this appeal:

[180]…[At] the time of payment, Mandarin would have had little or minimal information about the personal ties of the candidates with whom they contracted. It simply did not verify the candidates’ usual place of residence as Article 23 required it to do. ...

[183]Accordingly, we have decided that, prior to July 2016, Mandarin has failed to establish the usual residence of the candidates in the manner required by Article 23 of the Implementing Regulation.

[14]HMRC had made two separate assessments on Mandarin of under-declared output tax. One assessment, for £628,198 was in respect of Mandarin’s VAT periods 12/13 to 12/15. The second, for £799,407.19, was in respect of the VAT periods 03/16 to 06/17. In consequence of the conclusions that we have summarised, the FTT allowed Mandarin’s appeal in respect of periods from July 2016 onwards but dismissed it in relation to periods prior to July 2016.

[15]Neither counsel who appeared before this tribunal appeared before the FTT. Indeed Mr Lall, who appeared for Mandarin below, has sadly since passed away. However, it seems clear both from the Decision and the skeleton arguments before the FTT that it did not have the benefit of the arguments that we have heard as to whether Mandarin was entitled to demonstrate that students had their usual residence outside the EU in circumstances where it had not complied with its obligations under Article 23. That situation may well have arisen because the parties’ respective cases before the FTT were somewhat different from their cases now. Paragraph 66 of HMRC’s skeleton argument before the FTT demonstrates that their core argument on the “usual residence” of students was that such residence was in the UK because students were 6 living in the UK under Tier 4 visas at the time Mandarin provided the consultancy services. If correct, that analysis would have rendered any examination of Article 23 unnecessary and accordingly it seems to us that the parties did not focus on Article 23 before the FTT to the extent they might otherwise have done. Statutory and EU law

[16]Article 59 of the PVD sets out an exception to the general rule applicable (in Articles 44 and 45 of the PVD) to supplies of “business to customer” services. Article 59 provides, so far as material:

[59]The place of supply of the following services to a non-taxable person who is established or has his permanent address or usually resides outside the Community, shall be the place where that person is established, has his permanent address or usually resides: … (c) the services of consultants…

[17]Thus, since Mandarin was providing the service of a consultant to individual students who are “non-taxable persons”, the place of supply is determined by reference to where the students have their permanent address or usually reside. We have already explained that Mandarin places little emphasis on students’ “permanent addresses” and it was not argued to us as a basis for overturning the Decision. Therefore, in the remainder of this decision, we will focus our analysis of Article 59 on the concept of “usual residence”.

[18]It is common ground that the Implementing Regulation was directly applicable and therefore, in periods material to these proceedings, had the force of a UK statutory provision without the need for any implementing legislation. The Implementing Regulation contains the following recitals: (4) The objective of this Regulation is to ensure uniform application of the current VAT system by laying down rules implementing Directive 2006/112/EC, in particular in respect of taxable persons, the supply of goods and services, and the place of taxable transactions. In accordance with the principle of proportionality as set out in Article 5(4) of the Treaty on European Union, this Regulation does not go beyond what is necessary in order to achieve this objective. Since it is binding and directly applicable in all Member States, uniformity of application will be best ensured by a Regulation… (7) For certain services, it is sufficient for the supplier to demonstrate that the customer for these services, whether or not a taxable person, is located outside the Community for the supply of those services to fall outside the scope of VAT… (18) The correct application of the rules governing the place of supply of services relies mainly on the status of the customer as a taxable or non-taxable person, and on the capacity in which he is acting. In order to determine the customer's status as a taxable person, it is necessary to 7 establish what the supplier should be required to obtain as evidence from his customer…. (20) In order to determine the customer's place of establishment precisely, the supplier of the service is required to verify the information provided by the customer… (22) The time at which the supplier of the service must determine the status, the capacity and the location of the customer, whether a taxable person or not, should also be specified.

[19]We have added the emphasis in the above quotations because HMRC rely on the recitals as demonstrating an intention that a taxpayer should be required to obtain evidence as to a customer’s place of residence (and suffer sanctions if the requisite evidence is not obtained and verified) and, moreover, that the necessary evidence should be obtained before the time of supply.

[20]Article 3 provides, so far as material, as follows: Article 3 Without prejudice to point (b) of the first paragraph of Article 59a [of the PVD – which, it is common ground, is not relevant in the circumstances of this appeal], the supply of the following services is not subject to VAT if the supplier demonstrates that the place of supply determined in accordance with Subsections 3 and 4 of Section V of this Regulation is outside the community: … (c) the services listed in Article 59 of [the PVD]

[21]Article 13 of the Implementing Regulation provides as follows: Article 13 The place where a natural person ‘usually resides’, whether or not a taxable person, as referred to in [the PVD] shall be the place where the natural person usually lives as a result of personal and occupational ties. Where the occupational ties are in a country different from that of the personal ties, or where no occupational ties exist, the place of usual residence shall be determined by personal ties which show close links between the natural person and a place where he is living.

[22]Neither party was able to refer us to any authority on the application of the concept of “usual residence” for the purposes of Article 13 of the Implementing Regulation. However, it is common ground that decisions of the CJEU (by which expression we include both the Court of Justice of the European Union and the predecessor European Court of Justice) on the concepts of “residence” or “normal residence” provided a guide. We were referred to, among other cases, Case 284/87 Schäflein v Commission and

Case C-297/89

Ryborg. These authorities did not deal with the situation of a supplier verifying the residence of an individual customer but instead concerned the question of where individual taxpayers were resident for the purposes of particular legal provisions having some application to their personal circumstances. In Schäflein, the question was where an official of the European Communities was “resident” for the 8 purposes of deciding whether he was entitled to have the weighting for Switzerland applied to his termination of service allowance. In Ryborg the question was where a person had his “normal residence” for the purpose of deciding whether he was obliged to register a motor vehicle in Denmark.

[23]Both authorities approached the question as involving a multi-factorial assessment of all facts relevant to the location of a person’s permanent centre of interests. In Ryborg, the CJEU declined to issue guidance as to all factors to be taken into account, since this was a question of evaluation for the national courts. However, the CJEU’s judgment demonstrates that in principle, where a person works and lives are important. A person who has a home and job in one jurisdiction, but who makes frequent and lengthy visits to a romantic partner in a different jurisdiction could in principle be resident in either jurisdiction with a relevant question in such a case being whether there was a settled intention to live permanently with that partner.

[24]Article 23 of the Implementing Regulation provides, so far as material as follows: Article 23 …

[2]Where, in accordance with Articles 58 and 59 of [the PVD], a supply of services is taxable at the place where the customer is established, or, in the absence of an establishment, where he has his permanent address or usually resides, the supplier shall establish that place based on factual information provided by the customer, and verify that information by normal commercial security measures such as those relating to identity or payment checks.

[25]Article 25 of the Implementing Regulation provides: Article 25 For the application of the rules governing the place of supply of services, only the circumstances existing at the time of the chargeable event shall be taken into account. Any subsequent changes to the use of the service received shall not affect the determination of the place of supply, provided there is no abusive practice.

[26]In the paragraphs above, we have referred only to the EU legislation. We have not quoted UK legislation implementing the provisions of the PVD (no legislation being necessary to give effect to the Implementing Regulation which was directly applicable at material times) since it was common ground that there was no difference between the provisions of the PVD and UK statute law so that this dispute can be resolved entirely by reference to the EU legislation. Grounds of appeal

[27]Mandarin sought, and obtained, permission to appeal against the Decision from the FTT. In its application for permission, Mandarin summarised its grounds of appeal as follows: 9(1) The FTT erred in failing to establish the meaning or scope of the verification requirement in Article 23 of the Implementing Regulation, taking into account the wording of that Article and the principle of proportionality.

(2) Alternatively, to the extent that the FTT attributed a meaning to the verification requirement in Article 23 of the Implementing Regulation, the FTT construed that requirement unduly narrowly, which was disproportionate taking into account the wording of Article 23, the factual matters that were not in dispute and the FTT’s other findings.

[28]HMRC said that Mandarin’s arguments set out in Mr Hill’s skeleton argument went beyond the scope of the permission it was granted but pragmatically decided that they would make no formal objection on this ground. It was common ground that this appeal can be resolved by reference to the following issues:(1) Issue 1 – In deciding whether Mandarin had satisfied the requirements of Article 23 of the Implementing Regulation, (a) should the FTT have had regard to “informal evidence” (the term used by Mr Hill in his skeleton argument to connote descriptive general evidence from Mandarin’s personnel), as well as the documentary evidence contained in the work folders? (b) What evidence can Mandarin rely upon to establish that students had a “permanent address” or “usual residence” outside the Community? Is Mandarin limited to such documentary evidence as it had in its possession prior to the time of supply, or can Mandarin in principle rely on all evidence available to it, whether obtained before or after the time of supply, including witness evidence given in connection with the FTT proceedings?

(2) Issue 2 – Taking into account the answers to Issues 1(a) and 1(b), had the evidence that Mandarin put forward established a prima facie case that its supplies were to persons with a “permanent address” or “usual residence” outside the Community? If so, was there an evidential burden on HMRC to rebut that prima facie case which HMRC had failed to discharge?

(3) Issue 3 – To the extent that Mandarin had failed to satisfy the requirements of Article 23 of the Implementing Regulation, was that fatal to its claim to treat supplies made to students prior to July 2016 as outside the scope of VAT? Issues 1 and 3

[29]It became clear during the hearing before us that there was much overlap between Issues 1 and 3 because they both involved consideration of the nature of the shortcomings of Mandarin’s evidence and of the typical situation of traders in Mandarin’s sort of position.

[30]We will, therefore, deal with Issues 1 and 3 together by addressing the following points:(1) Whether, as a matter of construction of the provisions of the Directive and the Implementing Regulation, independent of the facts of this particular 10 case, a trader who fails to satisfy the requirements of Article 23 in relation to a particular customer is necessarily precluded from establishing that that customer’s usual residence is outside the EU.

(2) Whether, again as a matter of pure construction, a trader is permitted only to rely on material gathered before the time of supply, in pursuance of the obligations imposed by Article 23, to support a conclusion that a customer’s usual residence is outside the EU.

(3) Which of the evidence that Mandarin offered in this case could be relied upon to support its case that its customers were usually resident outside the EU. In particular, we will consider whether Mandarin could rely only upon documentary evidence contained in the work folders or whether it was also entitled to rely upon what Mr Hill referred to as “informal evidence” consisting of its knowledge and experience of students generally. Whether failure to comply Article 23 necessarily means that Mandarin’s supplies were subject to VAT

[31]Mandarin has not sought to challenge the FTT’s findings of fact to the effect that it failed to comply with Article 23, at least insofar as it failed to “verify” information. In our judgment, the FTT also determined that Mandarin failed to “establish” the position based on factual information from customers (see [176]-[177] and [183] of the Decision). In those circumstances, the question of construction is whether Mandarin is entitled nevertheless to seek to prove that its customers had their usual residence outside the EU.

[32]HMRC rely strongly on Article 3 of the Implementing Regulation. They emphasise that Article 3 provides for supplies not to be subject to VAT if the supplier demonstrates that the place of supply determined in accordance with Subsections 3 and 4 of Section 4 of Chapter V is outside the EU. Article 23 falls within Subsection 3 of Section 4 of Chapter V. It requires the supplier to “establish” the place of a customer’s usual residence, based on factual information provided by the customer. It requires the supplier to “verify” the information provided. Accordingly, argue HMRC, the effect of Articles 3 and 23 read together is that if a trader cannot “demonstrate” that a customer is usually resident outside the EU by meeting the requirements of Article 23, that customer must be taken to have a usual residence within the EU.

[33]Our first objection to that analysis is textual. Article 3 provides that if Article 23 is satisfied then the relevant supply is not subject to VAT. That does not necessarily compel the conclusion that, if the requirements of Article 23 are failed, the supply is taxable. In other words, Article 3 does not expressly provide for non-chargeability to VAT if and only if the requirements of Article 23 are satisfied.

[34]We acknowledge, however, that the Implementing Regulation is to be interpreted purposively and a purposive interpretation could, in principle, displace the textual objection we have highlighted. HMRC argue that considerations of purpose serve to reinforce their interpretation. They submit that Article 23 of the Implementing Regulation and the recitals that we have quoted at [18] above emphasise the need for 11 traders to gather information systematically and to verify it. That requirement is imposed for an important purpose: to ensure that the VAT status of a supply is ascertained at the time it is made so that the correct amount of VAT, if any, can be accounted for. It would be inconsistent with the principle of legal certainty for traders who fail to meet the requirements of Article 23 nevertheless to be able to treat their supplies as not subject to VAT. Moreover, such an interpretation would render Article 23 effectively redundant.

[35]We consider that this explanation of purpose omits some important considerations. As Mr Hill identified in his written and oral submissions, a trader making supplies of consultancy services to customers who are non-taxable individuals faces a difficult task when seeking to ascertain the place of supply of those services. That place of supply can depend on the place of “usual residence” of the customers. However, as we have noted at [23], ascertaining that place of usual residence could involve a multifactorial assessment of all relevant aspects of the customer’s personal and professional life much of which will be outside the trader’s knowledge. Therefore, an important aspect of the purpose of the provisions, which HMRC’s explanation overlooks, is the need for traders’ tasks to be manageable: they cannot be required to act as a detective agency, as Mr Hill put it, by delving into all aspects of their customers’ personal and professional lives but, at the same time, must make sufficient attempts to ascertain their customers’ place of usual residence.

[36]In our judgment Articles 3 and 23 of the Implementing Regulation are intended to strike that balance. Article 3 does not alter the rule on place of supply in Article 59 of the PVD. Accordingly, the rule remains that the place of supply can be determined by the place of a customer’s usual residence with all the difficulties in ascertaining the place of residence that we have identified. However, Article 3 overlays the place of supply rules by providing that, irrespective of the true place of supply, determined in accordance with Article 59 of the PVD, a supply is “not subject to VAT” if the supplier can demonstrate compliance with, among other provisions, Article 23 of the Implementing Regulation.

[37]Article 23, however, is not a pure “safe harbour” provision that a trader is free to rely on or not. The requirements of Article 23 to obtain and verify information by means of “normal security measures” are mandatory, reflecting the importance of traders gathering information with a view to ensuring that the correct amount of VAT is accounted for.

[38]It follows, in our judgment, that a trader who cannot demonstrate that a supply is “not subject to VAT”, by virtue of Article 3 of the Implementing Regulation (because, for example, the trader fails to satisfy the requirements of Article 23) retains the right to argue that the place of supply is outside the EU by operation of Article 59. The position could not be otherwise because the Implementing Regulation imposes no qualification on the ordinary place of supply rules in Article 59. We do not, however, accept HMRC’s submission that this interpretation deprives Article 23 of any effect. First, as we have noted, Article 23 imposes mandatory requirements and a trader who fails to meet those requirements may have to answer for such consequences as member states choose to impose in domestic law. Second, a trader who fails to comply with 12 Article 23 loses the ability to demonstrate, by taking relatively well-defined and limited steps, that a particular transaction is not subject to VAT. If that trader chooses to fall back on the ordinary place of supply rules in Article 59 of the PVD, the task of demonstrating the usual place of residence of a customer may be much more difficult given the multi-factorial nature of that test.

[39]We are reinforced in our conclusion by conclusions to be drawn from the jurisprudence of the CJEU on the distinction between “formal” and “substantive requirements”. In paragraphs 41 and 42 of its judgment in

Case C-590/13

Idexx Laboratories Italia Srl v Agenzia delle Entrate the CJEU, in the course of a judgment dealing with the right to deduct input tax, characterised substantive requirements as those which govern the “actual substance and scope of that right” and formal requirements as those governing the “exercise and monitoring thereof” and the “smooth functioning of the VAT system, such as the obligations relating to accounts, invoices and filing returns”.

[40]In this case, Mandarin failed to comply with the requirements imposed by Article 23. Given our explanation of the purpose behind the various articles set out at [36] to [38] above, and the careful balance they seek to strike, this requirement is not as obvious an example of a formal requirement as is, for example, the requirement to have a valid VAT invoice as evidence of a claim to deduct input tax. HMRC argued that, because Article 3 and Article 23 of the Implementing Regulation place emphasis on the trader “demonstrating” that conditions are met, that elevates the requirement to obtain and verify information into a substantive requirement. We disagree. Recognising that the boundary between formal and substantive requirements is not as clear in this case as it might be in others, we nevertheless consider that the requirement to obtain and verify information is more naturally described as a formal requirement since it is designed, at least in part, to enable HMRC to monitor the basis on which traders take decisions to treat services as supplied outside the EU and to facilitate the smooth operation of the VAT system.

[41]In

Case C-24/15

Josef Plöckl v Finanzamt Schrobenhausen the CJEU held that it was not permissible for a tax authority in a member state to refuse to grant exemption from VAT in respect of an intra-Community transfer on the sole ground that the supplier had failed to comply with a formal requirement to provide a particular VAT identification number. Plöckl was a case dealing with the power of member states to deny rights under EU law where a formal requirement, imposed by domestic law, was not met. This case is different. To the extent that there is any “formal requirement”, it comes from a directly applicable Regulation and so no question arises as to the power of member states to restrict EU law rights. Moreover the question we are addressing in this section is one of construction of an EU Regulation, whereas in Plöckl the CJEU was considering the permissible scope of domestic legislation.

[42]Nevertheless, we consider that the principle emerging from Plöckl is of assistance in construing the relevant provisions of EU law. Paragraph 80 of the Advocate General’s opinion in Plöckl speaks of the CJEU’s case law as: 13 …characterised by a rejection of formalism. That rejection of formalism is expressed, in practice, by an obligation on the tax authorities of the Member States to grant a right where all of the substantive requirements are satisfied, even where some formal requirements have not been met.

[43]In paragraphs 43 to 46 of its judgment, the CJEU endorsed the Advocate General’s articulation of the principle, and the two exceptions to it that the Advocate General identified. HMRC argue that the principle should be limited to cases concerning the application of a fundamental EU law right (such as the right to deduct input tax), but we reject that narrow interpretation of the principle. In paragraph 37 of its judgment the CJEU made it clear that the reason why it was in principle wrong for the failure of a formal requirement to result in a denial of an exemption without any examination of the substantive requirements was because such a treatment would result in transactions being taxed otherwise than by reference to their objective characteristics.

[44]Those principles of EU jurisprudence should, in our judgment, inform the construction of the Implementing Regulation. The requirement to obtain and verify information from a customer pursuant to Article 23 is more naturally characterised as a formal requirement than a substantive requirement. As such, the “rejection of formalism” in EU law makes it correspondingly unlikely that a failure to comply with that requirement should result in a trader losing any right to demonstrate that, in fact, a particular customer did have a usual residence outside the EU since that would run the risk of transactions being taxed otherwise than in accordance with their objective characteristics.

[45]As we have noted, EU law recognises two exceptions to the “rejection of formalism” that we have highlighted. There is no suggestion that Mandarin is party to any tax evasion and we therefore need not examine the first exception. The second exception applies where non-compliance with the formal requirement would effectively prevent the production of conclusive evidence that the substantive requirement has been met. Both parties struggled to identify a real-world example in which this exception would operate. Perhaps they came closest in giving the situation of a trader who chooses to destroy all books and records. However, in such a case a tax authority would presumably make some form of estimated assessment and it would not be a “formal requirement” that prevented the taxpayer from challenging such an assessment but rather a practical inability to put forward an alternative calculation of the tax due. We need not dwell on this difficulty, however. A failure to collect and verify information to the standard required by Article 23 would not prevent the production of evidence to the effect that particular customers have their usual residence outside the EU. Whether permissible evidence is limited to that gathered before the time of supply

[46]HMRC argue that the combination of Article 3 and Article 25 of the Implementing Regulation necessarily limit relevant evidence to that gathered before the time of supply.

[47]We consider, however, that HMRC’s reliance on Article 25 is misplaced. As Mandarin points out, Article 25 is simply saying that the place of supply of services is to be determined by reference to circumstances existing at the time of supply. 14 Therefore, as the second sentence of Article 25 makes express, if there is a change in those circumstances after the service is supplied, that cannot alter the place of supply of services that has already taken place. The flaw in HMRC’s argument is to assume that “circumstances existing” at the time of supply and a taxpayer’s knowledge of such “circumstances” are one and the same. However, they are not. It is entirely possible that a taxpayer may only realise after making a supply that facts and circumstances in existence at the time of supply caused that supply to be treated as made outside the EU. We see no indication on the face of Article 25 that a taxpayer is to be precluded from relying on particular facts and circumstances that were in existence at the time of supply but were discovered too late.

[48]In urging us to a different conclusion, HMRC rely on the importance of a trader knowing, before the supply is made, whether it is chargeable to VAT or not, as otherwise, there would be the inevitable risk that tax would not be properly charged when the supply is made. We acknowledge this as a practical point. However, for reasons similar to those we have set out in the previous section, this practical point does not mean that the legislation should be construed as meaning that a trader who failed to obtain the requisite information in time should necessarily be precluded from relying on information gathered later to establish the true tax liability associated with the transaction.

[49]If, as HMRC argue, a curfew comes into effect at the time of supply, with evidence obtained after the curfew counting for nothing, there would be scope for disproportionate and unfair outcomes in marginal cases. In his oral submissions, Mr Hill gave the example of a student contacting Mandarin in a bit of a hurry because of a job interview the next day. In the rush, Mandarin might form the mistaken view that the student is usually resident in the UK and charge VAT. Shortly after making payment, the student comes to Mandarin’s office and, during the ensuing discussion, it becomes clear that the student in fact has a usual residence in China. On HMRC’s construction of the Implementing Regulation, because Mandarin discovered the true position a few minutes after receiving payment, rather than a few minutes before, the supply would immutably be characterised as subject to VAT. Considerations of proportionality inform a construction of EU legislation and we see no reason why the Implementing Regulation should be construed as giving rise to such disproportionate outcomes given that there is a perfectly tenable construction of the provisions that enables them to operate more proportionately. The nature of the evidence that can be relied upon

[50]This issue can be addressed briefly given the conclusions that we have set out above.

[51]Article 23 does impose some limitation on evidence that can be relied upon under that Article. The evidence in question must be factual and it must be provided by the customer. Therefore, oral expressions of opinion might not be sufficient for the purposes of Article 23, even if given by the customer, since such expressions might not be factual. Factual information provided by someone other than the customer might not be good enough either. However, these distinctions aside, Article 23 does not make any 15 distinction between “formal” and “informal” evidence; nor does it distinguish between documentary and oral evidence.

[52]Moreover, as we have concluded, Mandarin is entitled to seek to establish that its customers had a usual residence outside the EU even though it has not complied with the requirements of Article 23. We do not consider that there is any limitation in principle on the nature of the evidence that it can rely upon in support of such a contention. As we have noted, determining an individual’s place of usual residence involves a multi-factorial assessment. It extends beyond matters that can be demonstrated by purely documentary evidence, such as where a person lives or works, and extends to questions of intention, such as, in the case of Ryborg, whether an individual had a subjective intention to settle permanently with a romantic partner in a different jurisdiction. In those circumstances, we do not consider that there is any limitation of principle on the kind of evidence on which Mandarin is entitled to rely. Conclusion on Issues 1 and 3

[53]We therefore conclude:(1) Even though Mandarin did not satisfy the requirements of Article 23, it is entitled nevertheless to seek to establish that its customers had their usual residence outside the EU.

(2) When doing so, Mandarin is not limited to evidence that it gathered at, or before, the time of supply.

(3) Nor is there any limitation in principle on the nature of evidence that Mandarin is entitled to deploy in support of its assertion.

[54]Mandarin is correct to say that, at [176] to [183] of the Decision, the FTT approached matters on the basis that the only means available to Mandarin to demonstrate that its customers had a usual residence outside the EU involved compliance with Article 23 of the Implementing Regulation. As we have noted it seems to us that the FTT did not have the benefit of the full submissions on this issue that we have had. However, our conclusion at [53(1)] leads us to conclude that the Decision contains an error of law.

[55]In those circumstances, s12 of the Tribunals, Courts and Enforcement Act 2007 (“TCEA”) gives us the power, though not the obligation, to set aside the Decision. If we do set aside the Decision, we must either re-make it ourselves or remit the case back to the FTT with directions for its reconsideration. We will consider how to exercise our powers under s12 of TCEA in the light of our conclusions on Issue 2 which follow. Issue 2

[56]We will apply the following principles in our consideration of Issue 2:(1) Mandarin bore the burden of proving that the supplies it made were not subject to VAT. If it had sought to comply with Article 23 of the Implementing Regulation, that burden would have been imposed by Article 16 3 and Article 23 both of which impose obligations on the supplier to “establish” relevant matters. In fact, Mandarin did not comply with Article 23 and is, in our judgment permissibly, seeking to establish that its customers were nevertheless usually resident outside the EU. It bears the burden of proof on that issue because of UK domestic law to the effect that a taxpayer seeking to displace an HMRC assessment in tribunal proceedings bears the burden of establishing the correct amount of tax payable.

(2) As a matter of civil litigation procedure it was open to Mandarin to discharge its burden of proof by adducing sufficient evidence to establish a prima facie case. If it did so, an evidential burden was placed on HMRC either to rebut or challenge Mandarin’s evidence. If it did not do so, Mandarin’s burden of proof would be discharged.

[57]Mandarin’s position on Issue 2 is that the FTT was wrong to decide the appeal purely by reference to the question of whether it met the requirements of Article 23. The FTT should, therefore, have considered the additional question of whether Mandarin’s evidence was sufficient to discharge its burden of proving that its customers were usually resident outside the EU. In Mandarin’s submission, it put forward sufficient evidence to establish a prima facie case and so to place an evidential burden on HMRC which they did not meet because they produced no evidence of their own and made no significant challenge to Mandarin’s evidence. Accordingly, Mandarin essentially asks this tribunal to set aside the Decision and remake it by deciding that supplies made prior to July 2016 were to persons usually resident outside the EU.

[58]By contrast, HMRC point to findings that the FTT made at, for example [70] and [148] of the Decision, as to the inadequacy of the specific information contained in the work folders which, they submit, demonstrate that the FTT was entitled to conclude that Mandarin had not discharged its burden of proof.

[59]Therefore, the parties approach Issue 2 from different perspectives. Mandarin asks us to make a decision which it considers the FTT should have made but did not. HMRC, by contrast, ask us to uphold findings of fact which the FTT did make. The matter is made more complicated by the fact that, as we have noted at [55], we have the power to set aside the Decision and remake it with the result that we are not necessarily bound by any determinations of fact that the FTT made in the Decision.

[60]We will, therefore, approach Issue 2 by considering first what relevant conclusions the FTT did reach and then considering how we should exercise our power to set aside and remake the Decision. The conclusions of the FTT

[61]In our judgment, Mandarin was putting its case before the FTT on an “all or nothing” basis. It was not inviting the FTT to make separate findings as to the usual residence of each of its customers individually. That was why it put into evidence a sample only of the work folders and why the evidence on which it relied consisted to a significant extent of generic evidence as to Mandarin’s business and customer base as a whole, rather than an examination of the situations of individual students. Mandarin’s 17 “all or nothing” case involved it making some observations on the situations of individual students: for example those students whose work folders were included in the Sample or the Enlarged Sample, and the student identified in Mr Peckham’s evidence summarised at [29] of the Decision whose family had settled in the UK. However, overall Mandarin’s case was that(i) the Sample was representative so that to the extent students within the Sample were usually resident in China, so were students generally and

(ii) Mandarin’s generic evidence supported the proposition that its student customers as a whole were usually resident in China. By putting its case in this way, Mandarin was accepting the risk that any deficiencies in the evidence contained within the work folders could result in the FTT declining to accept that any of its students were usually resident in China. Similarly, the FTT might conclude that the generic evidence was not enough to support conclusions as to the usual residence of any of Mandarin’s students.

[62]It is clear that the FTT conducted a detailed examination of the evidence contained in work folders comprising both the Sample and the Enlarged Sample. Since those work folders included CVs that students prepared shortly after engaging Mandarin to provide coaching services (see [7] above) the FTT would have been engaging with the detail of information on 22 students, said to be representative of Mandarin’s customer base as a whole.

[63]Having performed its review, the FTT made two important findings, neither of which is challenged. First, it was not satisfied that the Sample was representative (see [62] of the Decision). That finding alone dealt a significant blow to Mandarin’s “all or nothing” case since, if the Sample was not representative, there would be no sufficient basis for the FTT to make findings on the usual residence of Mandarin’s students generally based on the Sample. Second, the FTT concluded at [148] of the Decision that the records contained within the Sample were not sufficient to demonstrate the residence of the students with whom they were concerned.

[64]However, the FTT’s findings in this respect were limited to the Sample and the Enlarged Sample. It made no findings as to what, if any, conclusions it could draw as to the usual place of residence of Mandarin’s customers from Mandarin’s “generic” or “informal” evidence as to its business and customer base. The approach we should take in the light of the FTT’s findings

[65]The FTT had the advantage of seeing all of the evidence first-hand. The findings of fact that it made in relation to that evidence have not been challenged. While we have the power to set aside the Decision and remake it given the error of law that we have identified, in our judgment we should not exercise that power to interfere with findings of fact that are not challenged or are not affected by the errors of law.

[66]In our judgment, the findings we have identified in paragraph [63] above fall into this category. Those findings have not been challenged and were reached following a detailed review of the work folders. We do not consider that those findings are vitiated by the FTT’s flawed conclusions on the effect of Article 23 of the Implementing Regulation. There is no suggestion, for example, that the FTT performed a cursory 18 review only of the work folders comprised within the Sample or the Extended Sample thinking that a more detailed review would be unnecessary because Mandarin’s failure to comply with Article 23 was in any event fatal to its appeal. We will, therefore, adopt those findings of the FTT.

[67]In our judgment, those findings mean that, even if the FTT had correctly realised that a failure to comply with Article 23 was not fatal to Mandarin’s case, Mandarin would still not have succeeded in demonstrating, by reference to the Sample and Extended Sample alone, that its customers were usually resident outside the EU.

[68]It follows that, if the FTT had followed the correct approach, Mandarin could only have succeeded in its appeal if it had persuaded the FTT that its generic evidence as to its business and customers generally could fill the gaps in the evidence contained in the work folders. For the reasons that follow, we do not ourselves consider that the generic evidence does fill those gaps.

[69]We have read the witness statements of all of Mandarin’s witnesses who gave evidence. The statements are summarised in detail in the Decision and Mr Hill referred to them extensively in his submissions. We bear in mind that only Mr Peckham was cross-examined, and only briefly, but the lack of cross-examination does not prevent a critical assessment of what the evidence demonstrates in relation to Mandarin’s customers as a whole.

[70]The first obvious point that strikes us from those witness statements is that the “generic” evidence was just that: generic evidence about Mandarin’s business and its customer base. Certainly some of that evidence could, in a general sense, be regarded as having some bearing on the place of “usual residence” of Mandarin’s students. For example, Mr Peckham’s evidence that Mandarin’s students “came from” mainland China, that their parents lived in mainland China and that the students were so economically dependent on their parents that they relied on their parents to pay Mandarin’s fees was plainly of some relevance to the question of residence. So was his, and Mr Waley’s, evidence that Mandarin’s students were sufficiently unfamiliar with cultural norms in Western businesses as to need coaching to help them to secure jobs with such businesses. Mr Waley’s evidence that some students would attend coaching systems online from their “homes” in China was not irrelevant. Mr Latham’s evidence that students were living in the UK under Tier 4 visas which would expire shortly after their courses ended obviously called into question whether students could be usually resident in the UK, given that their right to reside in the UK was time limited. However, ultimately all of this evidence was given at a high level of generality. It was not capable of confirming the settled intentions of all of Mandarin’s students which, as noted from the case of Ryborg, are an aspect of any examination of usual residence.

[71]Moreover, the generic evidence of Mandarin’s witnesses was inherently based, in large measure if not entirely, on information that Mandarin obtained from its students. The FTT had already engaged with the raw information obtained from certain students as captured in the work folders and concluded that the raw information was insufficient to demonstrate usual residence outside the EU. If the underlying information in the Sample was not enough to demonstrate the usual residence even of those students which 19 the Sample covered, it is difficult to see a basis on which the generic evidence of Mandarin’s witnesses could be reliable as to whether all of Mandarin’s students were usually resident outside the EU.

[72]We have nevertheless taken a step back to ensure that the conclusions we have expressed do not overlook key matters of fact. It is clear from the FTT’s findings that(i) Mandarin’s students were all in their early 20s, were still full-time students and so unlikely to have much settled employment history and were still economically dependent on their parents;

(ii) the students’ parents were “almost exclusively resident and usually resident in China “ (see [24] of the Decision) and

(iii) with one exception, none of the students included within the Sample or Enlarged Sample had any family ties in the UK. We have asked ourselves whether these factors established a prima facie case that all of Mandarin’s students were necessarily usually resident in the same jurisdiction as their parents, namely in China.

[73]While we of course acknowledge that students in this position will often be usually resident in the same jurisdiction as their parents, we do not consider that this is the same as a prima facie case that all of Mandarin’s students were so resident. That is simply because the multi-factorial nature of the test of residence, which extends to matters such as subjective intention and commitment to romantic partners, makes it impossible to draw firm conclusions as to the usual residence of a large number of disparate individuals from even the above facts that they have in common.

[74]In a sense, the analysis we have just performed exemplifies the role for Article 23 and the importance of traders complying with it. Had Mandarin sought to comply with Article 23 by, for example, collecting evidence in the form of a high school diploma, or a written confirmation of a student’s address in China (and we reject Mandarin’s submission that that would have been unduly onerous or impractical) it would have found it much easier to demonstrate that its supplies to those students were not subject to VAT. By failing to comply with Article 23 Mandarin has had to fall back on a general multi-factorial determination of residence which was precisely the task that Article 23 sought to render avoidable. Given the multi-factorial nature of the test, it is not surprising that Mandarin has found it difficult to establish the usual residence of a large number of individuals. Conclusion on Issue 2

[75]For the reasons that we have given, we are not satisfied that, Mandarin could demonstrate, on the evidence that was put before the FTT, that supplies to all of its students were made outside the EU pursuant to Article 59 of the PVD. Disposition

[76]For reasons that we have given, the Decision contains an error of law. We regard that error of law as material to the Decision and we will, therefore, set the Decision aside. However, given our conclusions on Issue 2, we will remake the Decision so as to leave the result unchanged in relation to periods prior to July 2016. 20 Signed on Original MR JUSTICE MEADE JUDGE JONATHAN RICHARDS RELEASE DATE: 23 November 2021