Pye Motors Limited v The Commissioners for HMRC [2022] UKFTT 471 (TC)

[2022] UKFTT 00471 (TC)

Case No TC08668

FIRST-TIER TRIBUNAL

TAX CHAMBER

Venue Sitting in Manchester

Hearing Heard on: 7 and 8 September 2022

Date Judgment date: 12 December 2022

Appeal reference: TC/2014/00233

VALUE ADDED TAX – section 78 VATA 1994 – motor dealer – overpayment of output tax on demonstrator bonuses in the period 1974 to 1986 – overpayments repaid by HM Revenue & Customs – whether appellant entitled to interest – whether error on the part of HM Customs & Excise caused the overpayment or caused a delay in the appellant claiming repayment of the overpaid tax – appeal dismissed

Before

TRIBUNAL JUDGE JONATHAN CANNAN

Between

PYE MOTORS LIMITEDAppellant

THE COMMISSIONERS FOR HIS MAJESTY’S REVENUE AND CUSTOMSRespondent

Adam Rycroft, solicitor of KPMG LLP for Appellant

James Puzey of counsel, instructed by the General Counsel and Solicitor to HM Revenue and Customs for Respondents

DECISION

Introduction

[1]This appeal concerns the appellant’s claim to interest pursuant to section 78 Value Added Tax Act 1994 (“VATA 1994”). It arises following repayment by HMRC of overpaid output tax in relation to the period 1 August 1974 to 31 March 1986 (“the Relevant Period”). The appellant is and was at all material times in business as a Ford motor dealership. It overpaid VAT on bonus payments from Ford to the appellant in relation to demonstrator vehicles and certain other types of vehicle supplied by Ford to the appellant. There is now no dispute that VAT was overpaid by the appellant. The matter for determination is whether the appellant is entitled to interest on the amount repaid.

[2]The appellant’s case, in broad terms, is that the overpayment of VAT was caused by an error on the part of HM Customs & Excise (who for the sake of simplicity I shall refer to throughout as “HMRC”). Alternatively, that an error on the part of HMRC caused the appellant to delay making its claim for repayment. The circumstances in which VAT came to be overpaid are therefore highly relevant to the issues before me. I shall look at those circumstances in detail when I come to make my findings of fact. At this stage it is helpful to quote the summary of Henderson J as he then was in FJ Chalke Ltd v HM Revenue & Customs

[2009] EWHC 952 (Ch)

which concerned the entitlement of a motor trader to compound interest on overpayments of tax. Henderson J identified at [11] the two types of payments in relation to which VAT had been overpaid in that case: a) so-called manufacturers' bonuses, typically paid by a car manufacturer to a dealer who purchased a demonstrator vehicle …, and b) onward sales of demonstrator vehicles, typically after their use by the dealer for demonstration purposes and the provision of test drives to customers for a period of between six months and one year.

[3]Henderson J went on to describe what is known as the “input tax block” on demonstrator vehicles purchased by a dealer, which meant that the dealer could not recover input tax paid to the manufacturer on the supply of such vehicles. This was imposed because there was an element of non-business use in relation to demonstrators which were available to employees of the dealership for non-business use. He continued: [17]Despite the block on input tax for demonstrator cars, the Commissioners took the view that when such a car was sold to a private purchaser output tax should still be charged, although only on the difference, or ‘margin’, between the purchase price and the sale price. This system was generally known as ‘the margin scheme’, and is described as follows by Mr Easton in para 6 of his statement:

“However, in the case of a car dealer, selling a demonstrator, the car will normally be sold on relatively quickly in the course of business to a customer at (usually) a higher price than the dealer bought it for. Since the dealer has already borne VAT on the amount of the purchase price he had to pay, the Commissioners took the view that it would not be appropriate for the sale by the dealer to carry VAT on the full amount of the sale price. Rather, United Kingdom law required the dealer only to account for VAT on the 'margin' between his purchase price and the sale price.”

Before the decision of the ECJ in the Italian Republic case (see below), the block on input tax recovery on cars and the operation of the margin scheme were both contained in the Value Added Tax (Input Tax) Order 1992, SI 1992/3222. [18]According to Mr Easton, the policy view held by the Input Tax Branch of the Commissioners was that the combination of the block on input tax and the margin scheme represented ‘a pragmatic way of implementing the principle that VAT is a tax on the final consumption of goods’ … [19] This view was, however, shown to be untenable by the decision of the ECJ on 25 June 1997 in

Case C-45/95

EC Commission v Italian Republic

[1997] ECR I-3605

[1997] STC 1062

(‘Italian Republic’). In that case the Commission brought infraction proceedings against Italy under art 169 of the EC Treaty, alleging that Italy had failed to fulfil its obligations under art 13B of the Sixth Directive. In upholding the Commission's complaint, the ECJ held (para 16) that the final part of art 13B(c) requires member states to exempt the supply of goods in respect of which, by virtue of art 17(6), VAT did not become deductible when they were previously acquired or produced by the taxable person, and (para 19) that art 13B(c) does not allow member states to treat a transaction which is to be exempted as one which falls wholly outside the scope of VAT. It clearly follows from this reasoning that the United Kingdom should at all material times have treated sales of demonstrator cars as exempt supplies in respect of which no output tax could be charged, and that the margin scheme was therefore unlawful. The implications of the decision were soon realised, and on 10 October 1997 the Commissioners published Business Brief 23/97 in which they explained that, while consideration was given to what changes to UK legislation might be necessary, businesses could choose either to continue to use the margin scheme or to rely upon the ECJ judgment and treat the sale of input tax blocked cars as being exempt. The business brief went on to say that the Commissioners would accept claims for refunds of tax that had been overpaid as a result of the UK applying a margin scheme as opposed to an exemption, but that such refunds would be subject to the three year cap which was by then in force. [20]Mr Easton goes on to explain how consideration was then given to the question of how best to tax the private use of demonstrator cars and (more generally) the private use of vehicles by employees. Following consultation with motor industry trade bodies, proposals for new legislation relating to the VAT treatment of cars were published in April 1999, and new regulations were then introduced with effect from 1 March 2000, by the Value Added Tax (Supplies of Goods where Input Tax cannot be recovered) Order 1999, SI 1999 No 2833. The general effect of these regulations was to remove the input tax block and to require private use to be accounted for by means of a notional self-supply. The margin scheme was abolished with effect from the same date. [21]I now turn to the treatment of manufacturers' bonuses. Before the judgment of the ECJ in the Elida Gibbs case (see below), the Commissioners took the view that, as a matter of law, bonuses paid by car manufacturers to dealers on demonstrator vehicles, or to dealers or other customers on bulk orders, were to be treated as payments for a supply of services by the dealer or customer to the manufacturer. The normal practice seems to have been that the dealer would invoice the manufacturer for a supply of services including VAT, or alternatively the manufacturer would raise a self-bill invoice for the supply of services including VAT. Either way, the manufacturer would be entitled to deduct input tax on the supposed supply of services, but the dealer would of course have to account to the Commissioners for the output tax. With the benefit of hindsight, it can be seen that there were at least two difficulties with this treatment. First, the nature of the supposed services supplied by the dealer to the manufacturer in return for the bonus was elusive, and in many cases appears to have been an artificial construct invented to account for the fact that money was passing between two persons in a business relationship. Secondly, if it was right to regard the bonus as a discount from the price of the supply by the manufacturer, the principle of fiscal neutrality would appear to require that the taxable consideration for the original supply should be reduced by the amount of the discount. [22]On a preliminary reference by the VAT and Duties Tribunal in London, the ECJ held in

Case C-317/94

Elida Gibbs Ltd v Customs and Excise Commissioners

[1997] QB 499

[1996] ECR I-5339

[1996] STC 1387

(‘Elida Gibbs’), that retrospective discounts given by a manufacturer of toiletries under two coupon schemes (the first of which offered consumers a price reduction at the point of sale on the production of money-off coupons circulated in magazines or newspapers, and the second of which allowed the consumer to obtain a cash refund from the company by returning cash-back coupons which were printed on the label of the products) were indeed to be treated as reducing the taxable price at which the manufacturer had sold the goods in the first place. [23] The ECJ went on to hold that this principle applied even where there was no contractual relationship between the manufacturer and the final consumer, and despite the practical difficulties involved in retrospectively adjusting the taxable consideration for the supply. Considerations of that nature had weighed heavily with the Advocate General (Fennelly), who in his opinion delivered on 25 April 1996 had held that the cash-back coupons had no effect on the taxable amount of the manufacturer's original supply to the retailer or wholesaler, and that the money-off coupons should be regarded as a sales promotion scheme financed by the company to promote its commercial reputation and turnover: see the summaries of his conclusions in paras 29 and 39. The judgment of the ECJ is indeed a striking example of conceptual purity prevailing over practical convenience, and it doubtless came as a considerable surprise to the Commissioners after the strong support for their position in the Advocate General's opinion. [24]The implications of Elida Gibbs in the context of bonuses given by car manufacturers understandably took some time to consider, but on 21 July 1997 the Commissioners issued business brief 16/97 accepting that such bonuses should normally be treated as discounts by the manufacturers which reduced the value of their supplies. Businesses which believed that they had as a result overpaid VAT in the past three years were invited to contact their local VAT business advice centre. [25]The result of the decisions of the ECJ in Italian Republic and Elida Gibbs was that dealerships with demonstrator cars were likely to have overpaid VAT both (a) in respect of manufacturers' bonus payments which they had received, whether for the purchase of demonstrator cars or the achievement of specified sales volumes, and (b) in respect of the onward sale of demonstrator cars, when the margin scheme operated. It is not disputed by the Commissioners that the overpaid VAT was unlawfully levied, at any rate with effect from 1 January 1978, and that the Claimants were entitled to have it repaid in full once the unlawfulness under Community law of the three year cap had been established in Marks & Spencer I.

[4]It is worth noting by way of chronology that the ECJ judgment in Elida Gibbs concerning discounts was given on 24 October 1996 whilst the judgment in Italian Republic concerning output tax on the supply of goods where input tax had been blocked was given on 25 June 1997.

[5]Against that background, the appellant’s case is that in the Relevant Period HMRC wrongly required or represented to motor dealers that they should treat demonstrator bonus payments as consideration for a supply of services by the dealer to the manufacturer. That led the appellant to account for output tax on bonus payments received. In fact, such payments should have been treated by manufacturers and dealers as discounts, applying the judgment in Elida Gibbs, thus reducing the price paid by the dealer for the demonstrators. I shall refer to that error as “the Alleged Bonus Error”.

[6]In the alternative, the appellant says that it overpaid output tax on bonus payments because HMRC wrongly applied UK domestic legislation which required the appellant to account for output tax on the sale of demonstrators under the margin scheme. In fact, the appellant should not have accounted for output tax on the sale of demonstrators because of the input tax block, applying the judgment in Italian Republic. I shall refer to this as “the Margin Error”.

[7]In the further alternative, the appellant says that the way in which Parliament wrongly purported to introduce a 3-year cap on claims for overpaid output tax meant that it suffered a delay in receiving repayment of the sums overpaid for which interest is payable. I shall refer to this as “the Capping Error”.

[8]Section 78 VATA 1994 in so far as relevant provides for interest on overpaid output tax as follows: 78(1) Where, due to an error on the part of the Commissioners, a person has — (a) accounted to them for an amount by way of output tax which was not output tax due from him and, as a result, they are liable under section 80(2A) to pay (or repay) an amount to him,or … (d) suffered delay in receiving payment of an amount due to him from them in connection with VAT, then, if and to the extent that they would not be liable to do so apart from this section, they shall pay interest to him on that amount for the applicable period, but subject to the following provisions of this section.

[9]In a case falling within s 78(1)(a), the applicable period is defined by s 78(4) as a period beginning with the date on which VAT payable on the return in question was received by HMRC. In a case falling within s 78(1)(d), the applicable period is defined by s 78(7) as a period beginning with the date on which, apart from the error, HMRC might reasonably have been expected to authorise repayment of the amount overpaid and ending with the date on which that repayment is in fact authorised.

[10]There is no difference between the parties as to how s 78 should be construed as a matter of law.

[11]The application of s 78 was recently considered in Ford Motor Company Limited v HM Revenue & Customs

[2017] UKFTT 0407 (TC)

in which the FtT reviewed earlier decisions as to what amounts to an error, including Chartered Institute of Bankers v Customs & Excise Commissioners (1998) (Decision 15648) and CGI Pension Trustees Customs & Excise Commissioners (Decision 15926). The following principles set out at [9] in the Ford decision were common ground in that appeal and before me: 1) that ‘error’ in s78 was to be widely interpreted, and included a positive mistake or misdirection; 2) that although VAT was a self-assessed tax, Customs enjoyed statutory powers of care and management under which they could, and frequently did, give rulings on VAT liability and the practical implications of the liability and accounting rules for individual taxpayers: and, when they did so, they risked triggering s. 78; 3) that it was no answer to CGI’s case to say that Customs acted in good faith, or that [CGI] accepted the error at face value; 4) nor was it an answer to CGI’s case to say that a particular area of the law was complex and that Customs (or CGI) were confused. In such cases the tribunal should ask: from what did the taxpayer’s confusion directly flow? 5) that if an error was established there must be a causal link between the error and the taxpayer acting in one of the ways contemplated by s. 78(1). The section was satisfied if the taxpayer “relied on” Customs in taking the course it had taken, or, alternatively, if the error was the “true cause” of the taxpayer’s action or inaction; and 6) that Customs’ error need not be the sole cause of the taxpayer’s course of action. However, if there was no causal link it would be “repugnant to common sense” to describe the taxpayer’s course of action as “due to” Customs’ error.

[12]It is also common ground in this appeal that it is necessary to look realistically at the true cause of the overpayment. I respectfully agree with the FtT in Ford at [75] and [81] where it said that the test in s 78 is a simple one and should not be over-elaborated. It is necessary to identify whether the overpayment was due to an error on the part of HMRC. The error may not be the only cause of the overpayment, but it must be the “real” cause.

[13]In this context, the appellant relied on a decision of the VAT and Duties Tribunal in Wheeler v HM Customs & Excise (LON/95/1780A). In that case, the taxpayer would have been due a repayment of tax but for an assessment of tax made by HMRC. The assessment was subsequently withdrawn and the taxpayer claimed interest on the amount of the repayment pursuant to what is now s 78. The tribunal was satisfied that the assessment had been made to best judgment based on the information made available by the taxpayer to HMRC. It was only once further information was provided that the assessment was withdrawn. The tribunal had little difficulty in finding that there was no error on the part of HMRC and the real cause of the overpayment was the taxpayer’s failure to provide information on time.

[14]The FtT in Ford also considered at [76] whether an error can only exist if it is contained in specific guidance given to a particular taxpayer in relation to specific circumstances, in line with what it described as the approach in public law to ‘legitimate expectation’: [76] In my view, therefore, there is no requirement in section 78(1) that an “error” can only exist if it is contained in specific guidance given to a particular taxpayer in relation to specific circumstances. In relation to an alleged error contained in general HMRC guidance, such as a VAT Public Notice, much will depend on the terms and nature of the general guidance given but I do not think it is possible to say in advance whether a particular infelicity in public guidance can or cannot constitute an “error”.

[15]The parties in the present appeal helpfully prepared a statement of agreed facts and a statement of agreed issues. I shall deal with the facts in a separate section below. The agreed issues for determination are as follows:(a) whether the VAT at issue was overpaid due to the alleged error of HMRC in having represented that the bonus payments received by the appellant should, for the purpose of VAT, be treated as consideration for a taxable supply of services (“Issue 1”);

(b) whether the VAT at issue was overpaid due to the error of HMRC in having applied national legislation which, having excluded the right to claim deduction in respect of certain supplies/deemed supplies of vehicles, required the appellant to account for VAT under a margin scheme upon sale of those same vehicles (“Issue 2”); or

(c) whether the appellant has suffered a delay in receiving repayment of the amounts claimed due to the error of HMRC in the manner of the introduction, with effect from 5 December 1996, of a three-year limitation period in respect of claims for VAT overpaid (“Issue 3”).

[16]Whilst these issues broadly reflect the appellant’s amended grounds of appeal, it appears to me that Issue 1 does not fully reflect the grounds of appeal. The error of HMRC stated in the grounds of appeal is not that HMRC represented that bonus payments should be treated as consideration for a taxable supply of services. Rather, the grounds of appeal state that HMRC represented that in general such payments should be treated as consideration for a taxable supply. I shall return to that distinction when I come to consider the issues.

[17]It is important to note that what I have defined as the Alleged Bonus Error, is the alleged error of HMRC in representing how demonstrator bonuses should be treated for VAT purposes. It is not the error of the appellant in treating demonstrator bonuses as a taxable supply of services for VAT purposes. It is common ground that the appellant did make an error in so doing.

[18]I was invited by the parties to deal with the issues as a matter of principle, with the amount of any interest payable, if any, to be worked out by the parties in the light of my decision. I am content to take that approach. I understand that the claim for interest by the appellant amounts to approximately £200,000. There are also claims by a small number of other taxpayers operating Ford dealerships who have similar claims which are stayed pending the determination of this appeal.

Findings of fact

[19]It is common ground that the appellant overpaid VAT in relation to demonstrator vehicles in the Relevant Period. In relation to Issue 1 and Issue 2, it is necessary for me to make findings of fact as to the circumstances in which VAT was overpaid. In relation to Issue 3, it is necessary for me to make findings of fact as to the circumstances in which the appellant came to make its claim for repayment of the VAT overpaid. I shall firstly set out my findings of background facts, and then consider my primary factual findings in relation to each issue separately. Having said that, there will inevitably be some overlap. I will come on to consider the inferences I draw from the primary facts in my subsequent consideration of the issues.

[20]I should also say something about the burden of proof. It is common ground that the burden of proof lies on the appellant and that the standard of proof is the usual civil standard of the balance of probabilities.

[21]Neither party relies on any oral evidence from witnesses as to HMRC’s views and representations on the VAT treatment of demonstrator bonuses in the Relevant Period. The parties recognise that the reliability of such evidence would be extremely limited given that the period ended more than 35 years ago – see Gestmin SGPS SA v Credit Suisse (UK) Ltd

[2013] EWHC 3560 (Comm)

. Reliance is placed on documentary evidence, albeit not all contemporary documentation. It will be necessary to look closely at the documentary evidence in determining what if any inferences can be drawn from that evidence.

[22]I was also referred to the decision of the Court of Session (Inner House) in NHS Lothian Health Board v Revenue and Customs Commissioners

[2020] CSIH 14

. At the time of the hearing that case was on appeal to the Supreme Court. It concerned the correct approach to the burden and standard of proof in relation to historical claims for repayment of VAT. The Supreme Court handed down its judgment on 19 October 2022 (

[2022] UKSC 28

) and I invited the parties to make further submissions in writing in the light of that judgment.

[23]Both parties agreed that I should determine the factual issues in this case by reference to the usual principles as to the burden and standard of proof. It is for the appellant to satisfy me on the evidence available and by reference to the balance of probabilities that it is entitled to interest under s 78 VATA 1994.

[24]The appellant accepts that there is no direct evidence to establish HMRC’s policy in relation to demonstrator bonuses in the Relevant Period. However, it contends that there is sufficient evidence from which I should draw inferences as to the terms of HMRC’s policy and guidance to the motor trade in the Relevant Period. In that context, it relies on guidance issued by HMRC in 2017 in relation to historical “Fleming claims” at [4.8]: Unless there is evidence to the contrary, we will assume that compliant taxpayers operated in accordance with HMRC’s view of the law and guidance as it was at the time to which a Fleming claim relates. Where that view of the law was subsequently overturned by a judgment of the courts, absent evidence to the contrary, we shall work on the basis that the claimant corrected the position with effect from the end of the next prescribed accounting period after the date of the judgment.

[25]The respondents did not take issue with that approach, but it remains necessary for me to make findings as to what HMRC’s view of the law and guidance was during the Relevant Period.

[26]I take these principles and observations into account in making my findings of fact and in my discussion below where I consider the inferences I can properly draw from my findings of fact.

Background facts

[27]The appellant has been in business as a Ford motor dealership franchise since before 1 April 1973, when VAT was introduced. For the purposes of its business, it has used demonstrator vehicles supplied to it by Ford. Those vehicles were either purchased as such from Ford, or allocated as demonstrators by the appellant from general stock purchased from Ford.

[28]The appellant is registered for VAT through a group registration. On 1 August 1974, Pye Motors (Leasing) Limited joined the Appellant’s VAT group. That is the date that marks the start of the Relevant Period.

[29]During the Relevant Period and since, the appellant, or companies in its VAT group, received payments from Ford under the terms of its franchise agreement which were linked to the purchase, use or sale of Ford vehicles. I shall refer to these payments as ‘bonus’ payments, but elsewhere they are also referred to as ‘incentive’ or ‘support’ payments

[30]The appellant acquired vehicles either directly from Ford or via a company known at the time as Ford Motor Credit Company Limited (‘Ford Credit’). Ford Credit was a wholly owned subsidiary of Ford and in the same VAT group as Ford. Ford Credit supplied vehicles to Ford’s franchisees on finance.

[31]There were different types of bonuses commonly paid to dealers by manufacturers, including demonstrator bonus payments linked to the purchase, use and sale of demonstrator vehicles. There were also volume bonus payments linked to meeting target sales volumes. The demonstrator bonuses incentivised dealers to use vehicles for demonstration purposes and/or compensated dealers for the reduction in the resale value of the vehicle arising from use as a demonstrator.

[32]The appellant did not claim any input tax credit on purchases of demonstrator vehicles because of the input tax block. Where the appellant allocated a vehicle from general stock as a demonstrator this was treated as a “self-supply” for VAT purposes. Article 5 of the Value Added Tax (Cars) Order 1972 applied to deem a supply as having been both made by and to a motor trader in circumstances when a vehicle, purchased as general stock, was subsequently put to a use which engaged the input block.

[33]The self-supply rules up to 31 December 1977 required the deemed supply to be valued by reference to the open market value of a vehicle at the time of self-supply. The valuation rules were changed, with effect from 1 January 1978, to value the deemed supply at the cost of the goods to the taxable person.

[34]There was no uniformity in the motor trade as to how demonstrator bonuses were to be accounted for or how they were to be treated for VAT purposes. It is common ground that the following alternative forms of accounting documents were used:(1) The dealer issued an invoice to the manufacturer for services supplied, or

(2) The manufacturer issued a self-billed invoice, with a copy sent to the dealer, or

(3) The manufacturer issued a credit note to the dealer, with or without provision for VAT.

[35]In the case of the appellant, it is accepted that in the Relevant Period the appellant invoiced Ford for supplies of services to Ford in connection with demonstrator bonuses. The appellant charged VAT on those supplies and that is the VAT which was subsequently repaid by HMRC pursuant to the appellant’s claim.

Facts relevant to Issue 1

[36]In this section, I focus on the circumstances in which the appellant came to overpay VAT in the Relevant Period. In particular the alleged error on the part of HMRC in making representations to the motor industry as to the VAT treatment of demonstrator bonuses.

[37]Some facts and matters relevant to Issue 1 have been canvassed in previous tribunal decisions, most notably Why Pay More For Cars Limited v HM Revenue & Customs

[2013] UKFTT 497 (TC)

and the Ford decision in 2017. In the event, both parties acknowledged that findings of fact in those cases were based on the evidence and submissions made to the respective tribunals. I have not relied upon those findings as evidence in the present appeal.

[38]The documentary evidence is far from complete. The earliest document in evidence is a letter dated 25 October 1977 from Miss Jackson of HMRC Valuation Division to the Society of Motor Manufacturers and Traders (“SMMT”). It refers to the value for VAT purposes of self-supplied cars, and a change brought about by Finance Act 1977 with effect from 1 January 1978. The letter states as follows: At present VAT is charged on the open market value but from 1 January 1978 the basis will be the cost of the goods to the taxable person. It is therefore necessary to review the values of motor cars which were notified to you at or about the time of the introduction of VAT in 1973. For the sake of convenience these are set out below:(a) …

(b) … (iii) In all remaining cases: the purchase price (inclusive of car tax), disregarding any discounts received subsequent to the time of the self-supply, plus delivery charges to the trader’s premises (or £20 per car in lieu) if not included in the purchase price

[39]There is no copy of the SMMT’s reply in evidence, but on 1 February 1978 HMRC wrote to SMMT advising, for the purpose of the valuation of the self-supply at cost: Further to Miss Jackson’s letter of 25 October 1977 and to your reply of 5 December 1977 … In the case of self-supplies by dealers I am afraid that we must maintain the general rule that discounts allowed subsequent to the time of self-supply cannot be taken into account, but if in a particular instance the trader were able to demonstrate to his Local VAT Officer that the additional discount represented a genuine retrospective price reduction, we would of course consider this on its own merits. The revised arrangements, which are set out in full in the annexe to this letter, are effective from 1 January 1978, and are subject to review, as necessary, by either party. Thank you for your help and that of the taxation managers at Chryslers, Fords, Leylands and Vauxhalls.

[40]The annex to the letter of 1 February 1978 stated that it would be acceptable for motor traders valuing self-supplied motor cars to do so on the basis of: [t]he purchase price, inclusive of car tax, disregarding any discounts received subsequent to the time of self-supply, plus delivery charges to the trader’s premises (or £35 per car in lieu) if not included in the purchase price.

[41]It is notable that the annexe did not refer to the proviso contained in the body of the letter relating to genuine retrospective price reductions. However, Mr Rycroft for the appellant accepted that what was said in the annex should be read together with that was said in the body of the letter.

[42]The next available documents is from September 1986, when the SMMT issued guidance to its members which included reference to demonstrator cars. The guidance noted that additional payments by manufacturers to dealers could take the form of discounts or payments for services supplied by the dealership to the manufacturer for the services of demonstrating the vehicle. It provided as follows: [40] VAT AND DEMONSTRATION CARS When a dealer registers and uses a vehicle as a demonstration car it is common for the manufacturer or importer to make an additional payment in respect of that car. The payments differ in detail and may take a number of forms but can be summarised as follows: a) a special discount is given which reduces the value of the supply of the vehicle. This payment may be made at the time the vehicle is appropriated by the dealer or at some later date, possibly after a number of miles have been carried out in the demonstrator. In these circumstances, the VAT paid by the dealers on the car should reflect its lower value. When the vehicle is subsequently sold by the dealer under the second hand scheme, his records must show the reduced cost of the car, otherwise, if the vehicle is sold at a profit, the margin will be understated by the amount of the discount. It is probable, where the discount is given sometime after the vehicle is sold by the dealer, that the records will not show the lower value of the supply. Therefore if the vehicle is sold at a profit, the margin will be understated and so will be the VAT due. In such circumstances a separate record of the discounts should be kept and VAT paid on them. b) The manufacturer or importer pays the dealer for the services of demonstrating the vehicle. This is a taxable supply from the dealer. The payment can be made in a number of ways. The dealer might invoice his supplier with the service plus VAT or the manufacturer or importer might issue a self-billing invoice. In certain cases a credit note is issued which is used as a self-billing invoice. Customs officials informed the SMMT that a VAT credit note can only be issued in cases where there is a genuine price reduction for a supply. Therefore the Commissioners can no longer accept the issue of a credit note as a self-billing invoice in circumstances when the payment is treated as a supply by the dealer of demonstration services”.

[43]It is notable that the first treatment identified under paragraph (a) is for the bonus to reduce the value of the supply. It applies to bonuses paid at the time of self-supply and bonuses paid subsequently. It is paragraph (b) which refers to the bonus as representing payment for a taxable supply of services by the dealer to the manufacturer. There is no suggestion in the guidance that it is new, or as to how long it had been available to motor traders.

[44]The appellant places particular reliance on an HMRC internal briefing note date stamped 30 September 1987 prepared by Officer GS Burnes (“the Burnes Note”). It is not known what position Officer Burnes held in HMRC but he or she was clearly in a policy unit and corresponded with the SMMT in 1988. I understand that the Burnes Note was found in the records of Ford. It is headed “Draft” and is addressed to a Mr Burgess who also appears to be a policy officer. It is concerned with the valuation of motor vehicles for self-supply purposes and the VAT treatment of incentive payments. There are references to a number of policy documents which are no longer available. It is necessary to quote from the Burnes Note at length: 1. Area of Difficulty The Department has given agreement generally to a scheme operated by Fords detailed below, which effectively replaces the agreement on value for self-supply agreed with the Society of Motor Manufacturers and Traders in 1978. This scheme is radically different from that operated by the other major car manufacturers and its approval appears to give Fords a trading advantage which could be considered as unfair if we do not extend similar facilities to their competitors. 2. Present Agreement From 1 January 1978, VAT became chargeable on self-supply on the basis of the cost of the goods to the taxable person, rather· than open market value as previously. It was agreed with the SMMT that the value for self supply for dealers (other than importers) would be; 'the purchase price, inclusive of car tax , disregarding any discounts received subsequent to the time of self-supply, plus delivery charges to the traders premises (or £35 per car in lieu) if not included in the purchase price' A covering letter to the SMMT [ie dated 1 February 1978] stated; 'In the case of self-supply by dealers I am afraid that we must maintain the general rule that discounts allowed subsequent to the time of self-supply cannot be taken into account, but if in a particular instance the trader were able to demonstrate to his local VAT officer that the additional discount represented a genuine retrospective price reduction we would of course consider this on its merits' At this time [ie February 1978] VAD1 considered some of the various incentive payments made by the major manufacturers were consideration for the supply of services by the dealer to the manufacturer, rather than contingent discounts. Hence our reluctance to accept they reduced the value for self-supply. In an attempt to avoid detailed argument over which incentive payments qualified as discounts it was decided as a general rule that no such payment would normally be accepted as a discount. The background to this policy is in TIV 136/332/01, enclosures 92-112. The method adopted for dealing with incentive payments by most car dealers has therefore been for the dealer to issue a tax invoice to the manufacturer for the incentive payment made.

[45]The Burnes Note went on to record a new scheme introduced or intended to be introduced by Ford which had come to HMRC’s attention in June 1986. The new scheme involved Ford making payment of incentives to dealers at or before the time the vehicle was supplied. The dealer could retain the bonus if the vehicle was adopted for resale or as a demonstrator as appropriate. If the vehicle was not adopted the incentive was returned to Ford. It was recoded that Ford had advised dealers that if a vehicle was adopted for resale but was subsequently the subject of a self-supply then the scheme would benefit them because they could claim input tax credit when the vehicle was first supplied. If the vehicle was then the subject of a self-supply, the incentive payment would reduce the value of the self-supply. Officer Burnes described this as appearing to be: … a deliberate manipulation to recover input tax on vehicles always intended for own use.

[46]Officer Burnes went on to consider the acceptability of the Ford Scheme: VAD1 advised (encl 11 TI-IF 371/86) that incentive payments made are in effect contingent discounts. This advice is very general and possibly the various payments made need to be looked at in more detail. As a general rule however, this Branch would therefore have to accept that these payments reduce the value for self-supply. This overturns the advice currently in S8-10 paragraph 2.9(b)(ii). We would therefore accept that where Fords sell a vehicle to a dealer who adopts it simply for resale or own use it is acceptable for Fords to issue a credit note (inclusive or exclusive of VAT) which reduces the value of the supply. The related problem over the Ford General Letter was considered in depth in Mr Hyams paper of 3 October 1986 (encl 22-28 TMF 371/86). This resulted in your advice to the officer for Fords of 15 April 1987 (encl 34-35) which accepts the incentive payments may reduce the value for self-supply… On the basis of [the local VAT Officer’s] arguments the scheme seems acceptable in full. The one point that needs to be covered further is the possible manipulation involved to recover input tax… I have discussed this situation in some depth with VAD5. Despite the deliberation manipulation to obtain input tax, they consider we could not block input tax claimed by the dealer in this circumstances. They point out that if Fords had simply sold the vehicle to the dealer and then issued a VAT inclusive credit note, we would end up with the same payment of tax. I would support this view.

[47]Officer Burnes went on to consider schemes used by other dealers: Most of the other main manufacturers are controlling incentive payments by tax invoices issued by the dealer to the manufacturer. Although it is of course up to the individual manufacturer how they organise their network it may be that they continue to operate on that basis because of the 1978 agreement. We often receive calls from LVOs raising queries from individual dealers asking if these payments reduce the value for self-supply. I would presume you would agree that as long as the payments are treated as taxable supplies of services, it would be difficult for them to be treated as discounts I consider we should only do so where each individual manufacturer has set up their own system. If they chose to continue with the present arrangements, we would expect all of their dealers to comply accordingly.

[48]Officer Burnes then proposed the following action, with a suggestion that his note be referred to other policy units before taking action: I suggest we need to write to the SMMT pointing out how the Ford scheme has affected the 1978 agreement and seeking the comments of the other major manufacturers on the future position.

[49]It subsequently transpired that when the Burnes Note was prepared, Officer Burnes was not aware of the guidance issued by the SMMT to its members in September 1986.

[50]The Ford decision suggests that the Ford scheme described in the Burnes Note was not implemented. Be that as it may, the evidence does not show any response to the Burnes Note from other policy units. However, a year later a further note dated 4 October 1988 (“the Burnes Further Note”) was circulated to policy units together with a draft letter intended to be sent to the SMMT. The Burnes Further Note stated: 1. My minute of 30 September 1987 (encl 168/171) suggests we write to the SMMT explaining how our views have changed since our 1978 ruling… 2. I have now seen a copy of SMMT's September 1986 advice to their members which shows they have already indicated to dealers that discounts can reduce the value for self-supply. It is also now clear that manufacturers such as British Leyland and Vauxhall, who make their initial supply to finance companies, are not giving discounts to the dealers. 3. I think it would still be useful to formally clarify the issue with SMMT. A suggested draft is attached. I havealready written to the Motor Agents Association in similar terms (encl 3 TIV 34/88). 4. Paragraph 2.18 of the new version of S8.10 to be published shortly states that any claim that incentive payment are genuine discounts should be reported to this Branch (encl 176). I think this requirement should be removed and the paragraph expanded to explain in more detail the circumstances in which an incentive payment can qualify as a contingent discount and reduce the value for self-supply. This would cover the points raised in the latter part of my letter to SMMT and instruct [Local VAT Offices] to report any difficulties to VAD3.

[51]Comments were provided from other policy units on or about 20 October 1988 and it appears that only minor changes were made to the draft letter. Officer Burnes’ letter to the SMMT was sent on 24 October 1988 (“the Burnes Letter”). It stated as follows: We sometimes receive queries from motor dealers about bonus payments made by manufacturers and their effects on value for self-supply. I thought it might be useful to clarify the position to assist you in dealing with similar queries. Our letter of 1 February l978 to yourselves (copy attached) advised that we did not generally consider that 'discounts' allowed subsequent to the time of self-supply reduced the value for tax. This was because we then considered that most such payments were not actually discounts giving genuine price reductions but were principally payments for supplies or services by the dealer to the manufacturer. We said then, however, that we were prepared to consider each case on its merits. One major manufacturer introduced a market support programme in 1986 under which it issues VAT exclusive credit notes to its dealers. We ruled that these credit notes were valid, asthe payments made were freely given contingent discounts on specific supplies made. I note that in a circular to your members dated September 1986 (copy attached) you draw correctly a distinction between discounts which reduce the value of the supply and payments for supplies of services by the dealer to the manufacturer. In the latter case of course, the use of a credit note is not valid. Where a rebate earned retrospectively is directly referable to the supply of a car to the dealer by the manufacturer, this can normally be accepted as a contingent discount. Examples include volume rebates retroactive to the first car or dealer campaign bonuses linked to sales of a particular make. It is vitalthat the supply of the cars involved is direct from the manufacturer to the dealer. If a third party such as a finance company is interposed between the manufacturer negotiating the rebate and the dealer, the payment cannot be accepted as a discount because it is not directly referable to the supply. Furthermore, if an allowance is paid for a service performed by the recipient to the manufacturer, this may represent a taxable supply of servicesto the manufacturer. Againthis cannot be seen as a contingent discount. The dealer must issue a tax invoice for his services or account for tax on a self-billing invoice from the manufacturer. The above information is only a statement of the general position. Traders who are uncertain of the status of particular payments and their effect on the value for self supply of vehicles should clarify the position with their local VAT Office.

[52]The fourth paragraph of the Burnes Letter appears to be a reference to what is now known as “the line of supply”. In other words, where the discount is paid by the manufacturer making a supply directly to the dealer who receives that supply. HMRC in particular relies on references in the evidence to the line of supply in support of its case as to the nature of the representations it made as to the VAT treatment of demonstrator bonuses in the Relevant Period.

[53]Another officer of HMRC had written to the SMMT on 30 October 1987 in relation to the VAT liability of incentive payments by vehicle manufacturers to high volume purchasers. This is a different category of incentives paid by manufacturers to customers such as fleet purchasers. The appellant places reliance on the treatment of such incentives. This letter also referred to the line of supply and stated: As you know, the Department in the past had indicated that VAT was not applicable to these payments (often referred to as Third Party Discounts). Whilst they are not true discounts on the goods purchased because they do not relate directly to the supply – for example the customer buys from the authorised dealer but negotiates with the manufacturer for the incentives, often without the knowledge of the dealer – there was no evidence that the payments represented consideration for any supply of services by the customer to the manufacturer. We are now aware that services are invariably supplied in such circumstances. Our solicitors are clear on the point that customers are supplying services within the broad scope of Section 3(2) (b) of the VAT Act 1983 when in return for the payment they fulfil or agree to fulfil specific conditions set by manufacturers. Typical examples of these conditions are: i) the requirement to purchase a minimumnumber of vehicles before the payment is made ii) the requirement to purchase only from authorised dealers iii) the requirement topay cash iv) the requirement that the vehicle is not to be sold by the customer within a year of purchase. In accepting such conditions the customer has forgone certain rights - for example, the right to buy from whomsoever he wishes. It is important to set the matter on auniform footing: while it is likely that any tax will be deductible, we are aware of cases where manufacturers have, of their own volition, regarded these payments as consideration for taxable supplies and have consequently self billed the VAT. This represents a possible loss to the revenue if no output tax isaccounted for by the customer. In summary, our policy is now to see taxable supplies to manufacturers where the type of situation and conditions referred toabove apply. It would be advisable if manufacturers when negotiating the payments, clearly explained to the customers thattax is due from them on the amounts paid.

[54]The next available document is dated 4 June 1997, HMRC’s VAT Policy Directorate issued what is known as a “dear colleagues letter” or “DCL” which was intended to provide VAT officers with background to the treatment of demonstrator bonuses and fleet buyers bonuses. It provided examples of where the ECJ decision in Elida Gibbs had altered the previous policy and created situations where businesses may have overpaid VAT. In particular, where discounts were not paid in the line of supply. The DCL stated as follows: Background 2. Prior to the ECJ decision in Elida Gibbs we did not accept that there could be a discount, which reduced the value of supplies, when it was given to someone other than the direct recipient of the supply. Dealer demo bonuses and the fleet buyer’s bonuses are often given to dealers and customers where they are not the direct recipient of the manufacturer’s supply. In such circumstances we maintained that the payment was not a discount. Given that the manufacturer could not treat the bonus as a discount, it was our view that the payment was the consideration for a supply of services. Supply of Services 3. The decision in Elida Gibbs found that cashback payments by a manufacturer to consumers, who were not the direct customer of the manufacturer, were discounts given by the manufacturer. Asa result of this we must now accept that discounts can be made to parties further down a chain of transactions and not just to the giver's direct customer. As the bonus payment can now potentially be treated as a discount,we have had to re-examine the previous treatment applied to them. Whether a dealer demo bonus or a fleet buyers bonus should be treated as a discount by the manufacturer or as the consideration for a supply of services will depend on what criteria are laid down for the dealer or customer to qualify for the payment. In most circumstances the criteria will not be sufficient to establish that a supply of services exists and the payment should, therefore, be treated as a discount by the manufacturer. Examples where the criteria are significant enough to argue that a supply of services exists should be relatively uncommon.

[55]The DCL went on to give examples where bonus payments were to be treated as discounts. It described the possible treatments the parties may have adopted and indicated whether the correct amount of tax would have been paid. The first set of examples referred to demonstrator bonuses paid by a manufacturer to a dealer where the line of supply also included a finance house. It was stated that the previous treatments may have led to some under or over declaration of tax. The second set of examples referred to demonstrator bonuses paid by manufacturer to dealer where the line of supply was direct and did not include a finance house. In relation to the second set of examples the DCL states: c) Possible underdeclarations or overdeclarations. It is important to be clear that this situation is not affected by the Elida Gibbs decision itself. Because there is no intermediary in the supply of the car from the manufacturer to the dealer it would always have been in line with our policy that the manufacturer be able to issue a credit note to the dealer where a discount is agreed. The complication is that some manufacturers will have accepted that they were receiving a supply from the dealer, because it was necessary in situations where there was an intermediary. However, it is now generally accepted that there is no supply of services. As the bonus is to be treated as a discount given by the manufacturer to the dealer, the treatments above may again have led to some under/overdeclarations.

[56]The third set of examples relates to the payment of fleet buyer’s bonuses stating: b) Previous treatment As far as we are aware the only common approach in the past has been to treat the fleet bonus as the consideration for a supply of services from the customer to the manufacturer.

[57]On 21 July 1997 HMRC issued Business Brief 16/97 concerning demonstrator bonuses and fleet buyer bonuses. It states that in the past it had been common practice to treat these bonuses as the payment for a supply of services by dealers and customers, so VAT was due from the dealer or customer on this supply. It goes on to say that following Elida Gibbs, these bonus payments will normally be treated as discounts by manufacturers. However, it stated that in many cases the net effect of the change would not alter the net tax due. Businesses which believed they had overpaid VAT in the previous three years were invited to contact HMRC.

[58]On 17 December 2002 HMRC met with various motor industry representatives including the SMMT with a view to reaching common agreement over the various factors that would apply to claims made pursuant to Business Briefs 22/02 and 27/02. Those Business Briefs related to historical claims for overpaid VAT following the judgment of the ECJ in Marks & Spencer v HMCE

Case C-62/00

. The ECJ had held that a retrospective three year cap for claims introduced in 1997 was unlawful because there was no transitional period.The meeting records Mr Bob Lewis (BL), of HMRC, stating as follows in relation to the impact of Elida Gibbs on dealers: BL said that these claims would be complicated by the confusion that was widespread throughout the industry until the mid-90's regarding the treatment to be applied to bonus payments. BL said that in his view only a valid claim could be made for back-end discounts (i.e. paid after the sale of cars had been made) paid for blocked vehicles. Front end discounts would have been accounted for as a normal discount and so no confusion over its treatment would have occurred. 2.3 Lines of supply BL described the various lines of supply that may have applied, to both the sale of a vehicle and the related accounting transactions (focusing on the backend payment), during the relevant period. He described three scenarios that he had identified: 1. The vehicle supplied directly by the manufacturer to the dealer/retailer, with the payment and accounting also following this route; 2. The vehicle supplied by the manufacturer to a second entity (usually a finance house, this may be a captive) and then onwards by the intermediary to the dealer/retailer. The manufacturer made the backend payment directly to the dealer, missing out the finance company. 3. The vehicle supplied by the manufacturer to a second entity (usually a finance house) and then onwards by the intermediary to the dealer/retailer. The backend payment made by the manufacturer also followed this line of supply (including the finance house). BL suggested that no claim would be applicable in the first or third scenarios because he expected the VAT would have been adjusted by way of credit notes. BL said he would like more information as to who used this type of structure and at what time. It is the second scenario where a valid claim may be applicable by the dealer.

[59]A further meeting was held on 28 January 2003 and it was at this meeting that HMRC first suggested creating a table which dealers could use to calculate a claim for repayment of tax overpaid on bonus payments.

[60]Sometime after that meeting, HMRC issued general guidance to taxpayers in connection with the making of claims arising from Elida Gibbs. The guidance stated as follows: 1. Claims can be made in respect of retrospective bonuses on new cars purchased by dealers for use in their business from 1973 onwards where UK legislation required dealers• to block the input tax on the car, or

• trigger a self supply when the vehicle was taken from new car stock. Demonstrator cars fell into this category until the UK law changed on 1 December 1999. … Where the documentation of these payments followed the chain of supply of the cars, Customs' policy was always to treat them as a discount on the price of the car. No claims should arise in such cases.

[61]The guidance also enclosed a table prepared on the basis of information available to HMRC and the motor trade bodies indicating how VAT had been accounted for on bonus payments by different manufacturers in different periods (“the Elida Table”). The Elida Table identified that in the period from 1973 to 1985, Ford had accounted for VAT on bonus payments using an invoice. From 1 April 1986 onwards, Ford had issued credit notes and the payments were treated as being ‘outside the scope’ of VAT. The Elida Table also indicated that Ford’s demonstrator bonuses followed the line of supply.

[62]An updated version of the Elida Table was issued by HMRC in October 2006. For each make of vehicle, the Table indicated the accounting document used to account for the demonstrator bonus, whether the bonus followed the line of supply, and the VAT liability shown on the accounting document. The Table had certain columns identified as Columns 1 and 2, described as “Elida case” and Columns 3 and 4 described as “Non Elida case” with reference to the following notes: Note 1. Columns 1 and 2 on the right hand side of the pages are in respect of errors arising as either output tax or input tax under the terms of the Elida Gibbs decision. Note 2. Columns 3 and 4 to the right of the table cover the circumstances where accounting errors were made which gave rise to underclaims of input tax (column 3) or overdeclarations of output tax (column 4). These errors did not arise as a result of incorrect decisions by HMRC. As such, there was no Departmental error and you should refer to note 3 below. As these errors should not have arisen, there should be no assumption that businesses made the errors. Note 3. Statutory Interest (SI) is only payable where there is HMRC error. Where a claim is payable under columns 1 and 2, SI will be payable if requested. Where it is payable under columns 3 and 4, SI will not be due for periods before 4 December 1996 (or 30 April 1997, depending on the error, see Business Brief 13/06).

[63]In other words, errors in relation to demonstrator bonuses which were not in the line of supply were accepted by HMRC as arising from their error. Errors which related to demonstrator bonuses which were in the line of supply were not accepted by HMRC as arising from their error.

[64]The updated Elida Table shows that different manufacturers and their dealerships were adopting different accounting documentation and VAT treatment in respect of demonstrator bonuses. In some cases dealers issued invoices to manufacturers for services supplied. Alternatively, manufacturers might utilise self-billed invoices or issue credit notes to dealers. In some cases that documentation followed the line of supply and in some cases it did not. In some cases the documentation identified VAT which was accounted for by the manufacturer and the dealer and in some cases no VAT was identified with any supply being treated as outside the scope of VAT. If an invoice or a self-billed invoice was used this generally meant that the dealer was treated as supplying services to the manufacturer. In the case of credit notes, the position was not clear.

[65]The Elida Table shows that 7 manufacturers used credit notes prior to 1986, with 9 manufacturers who either issues self-billed invoices or required dealers to issue invoices for supplies of services. Apart from Ford, it appears that only Peugeot and Saab (in 1979 only) used dealer invoices to account for VAT on a supply of services by the dealer where bonuses were paid in the line of supply. Rover dealers also issued invoices although that was in relation to supplies outside the line of supply.

[66]HMRC also relied on an internal review of an appeal to the VAT Tribunal by another Ford dealer dated 15 February 2006. The document is heavily redacted to remove reference to the dealers involved. It includes the following passage: …I have sought clarification from [redacted]. He has explained that the supply chain for the cars passed from Ford to Ford Motor Credit to the dealer. The two Ford companies were in the same Group registration. Thus, the situation with bonus payments was that they were always seen as discounts on the selling price and not the consideration for a separate supply of services from the dealer to the manufacturer (who had not made the supply of goods to the dealer). Thus, Ford was never in an Elida type situation, and you would not expect an Elida claim. The exception would be ‘if having received a Ford bonus payment ' invoice' [redacted] made an internal accounting error and posted output tax that wasn’t due…

Facts relevant to Issue 2

[67]In this section I focus on what was an error on the part of HMRC in applying UK domestic legislation which required motor dealers to account for output tax on the profit margin on supplies of demonstrator vehicles when they are eventually sold to customers. The facts relevant to Issue 2 can be stated quite shortly. It is not disputed that HMRC wrongly required dealers to account for output tax on supplies of former demonstrator vehicles using the margin scheme. It became apparent that this was wrong when the judgment of the ECJ in Italian Republic was released in June 1997.

[68]Clearly, when a demonstrator vehicle was sold it might have been sold at a profit or a loss, or indeed neither a profit nor a loss. It is only where it was sold for a profit that the dealer will have wrongly accounted for output tax on the margin. There is no direct evidence before me as to whether and if so to what extent the appellant’s demonstrator vehicles were sold at a profit. Mr Rycroft relied on the “Italian Tables” from which he said it is implicit that demonstrator vehicles were in the vast majority of cases sold at a profit.

[69]The “Italian Tables” were published by HMRC to assist motor dealers in making claims for VAT overpaid as a result of the Margin Error. Mr Rycroft suggested that they assumed that demonstrators would be sold at a profit, before the demonstrator bonus was taken into account. I was not taken to the Italian Tables and they were not in evidence before me. In any event, I do not consider this point determinative of Issue 2. I understand that the Italian Tables were a pragmatic approach by HMRC and dealers to quantify claims arising out of the Margin Error. I give little weight to them in terms of evidence as to the extent to which demonstrators were sold at a profit.

[70]It is likely that some of the appellant’s demonstrator vehicles would have been sold at a profit. I cannot say that they would all have been sold at a profit. There was evidence in FJ Chalke that such vehicles would “usually” be sold at a profit, but there is no such evidence before me.

[71]It is part of the appellant’s case that the existence of HMRC’s Margin Error meant that the incorrect treatment of demonstrator bonuses had no financial effect on the appellant. In other words, whether or not the demonstrator bonuses were treated as a discount on the purchase price or as consideration for a taxable supply of services by the appellant, the output tax accounted for by the appellant would have been the same. The appellant would have accounted for too much output tax, either because it wrongly paid output tax on a supply of services to the manufacturer or because it would wrongly have paid output tax on an increased margin on sale of the demonstrator. I accept the logic of that proposition where a demonstrator is sold at a profit. However, if a demonstrator was sold at a loss there would have been every reason to challenge or clarify HMRC’s view as to the VAT treatment of demonstrator bonuses.

[72]Even where a demonstrator was sold at a profit, that would not mean that there was no financial effect on the appellant and no incentive for the appellant to challenge or clarify HMRC’s view as to the VAT treatment of demonstrator bonuses. Indeed, Mr Rycroft accepted that a dealer would receive a cashflow advantage if it had treated demonstrator bonuses as a discount on the price paid for the vehicle. That is because VAT on the margin would not be due until after the demonstrator was sold to a consumer. It was common ground and I find that dealers were required to keep demonstrators for specified periods of time, usually a matter of months, pursuant to the franchise agreement with the manufacturer. There was therefore some financial incentive for the appellant to challenge or clarify HMRC’s view as to the treatment of demonstrator bonuses.

Facts relevant to Issue 3

[73]In this section I focus on the circumstances in which the appellant came to make its claim for repayment of VAT overpaid in the Relevant Period. The chronology of events is as follows:

[74]On 27th May 1994, the appellant made a claim for repayment of VAT in relation to input tax credits of £714,194 on purchases of demonstrator vehicles. That claim was made on the same basis as a claim being made in Royscot Leasing v CCE

Case C-305/97

.Namely, that the input tax block on demonstrator vehicles introduced in the UK was unlawful. The claim covered the period 1 January 1988 to 31 December 1993, but was rejected by HMRC on 8 June 1994.

[75]On 10 July 1995 Cooper Lancaster Brewers, Chartered Accountants informed the Appellant of progress in the Royscot Leasing appeal. At that stage, the VAT Tribunal had dismissed that appeal. The writer of the letter referred to the appellant’s decision to prepare and submit its own claim rather than using professional advisers as having been a sound decision. In the event, the appeal in Royscot Leasing was unsuccessful following a decision of the ECJ in 1999 and the appellant’s claim went no further.

[76]On 27 June 1996, the Advocate General’s opinion in Elida Gibbs was delivered concerning the VAT treatment of promotions, including cash-back coupons printed on the wrapping of goods which entitled the purchaser to claim an amount of cash-back directly from the manufacturer. The Advocate General recommended answering the questions referred on the basis that such payments did not operate to reduce the taxable amount of the supplies.

[77]The ECJ issued its judgment in Elida Gibbs on 24 October 1996. It departed from the Advocate General’s opinion and concluded that the taxable amount was the selling price charged by the manufacturer, less the amount indicated on the coupon and refunded. The judgment in Elida Gibbs gave rise to the possibility of claims for refunds of VAT by manufacturers who made payments of the kind considered in that case.

[78]On 18 July 1996, the Paymaster General had announced the Government’s intention to introduce a three year limitation period with immediate effect in respect of claims for overpaid output tax. Parliament introduced that three year limitation period with effect from 4 December 1996.

[79]On 10 December 1996 the Advocate General’s opinion in Italian Republic was delivered. Heproposed that the Court grant the Commission’s application and declare that in enacting and maintaining in force a provision which did not exempt supplies of goods which had been excluded from the right of deduction, the Italian Republic had failed to fulfil its obligations under Article 13B(c) of the Sixth Directive.

[80]The ECJ issued its judgment in Italian Republic on 25 June 1997, following the opinion of the Advocate General.

[81]HMRC issued business brief 16/97 on 21 July 1997. It dealt with the impact of Elida Gibbs as described above, and invited businesses which believed they had overpaid VAT in the previous three years to contact HMRC.

[82]In August 1997 KPMG sent a flyer to the Appellant outlining the opportunity to make claims for VAT overpaid on bonus payments in the light of Elida Gibbs, although the case was not mentioned by name. Another KPMG flyer is dated October 1997 and referred to the possibility of claims arising from use of the margin scheme on sales of input tax blocked vehicles such as demonstrators. Italian Republic was not mentioned by name and the flyer noted that claims would be subject to the three year cap.

[83]In September 1997, Coopers & Lybrand sent the Appellant a flyer referring to the possibility of making claims for VAT previously accounted for under the margin scheme on disposals of used vehicles, also referring to the three year cap.

[84]On 10 October 1997 HMRC issued Business Brief 23/97 in which they explain that whilst consideration was being given to what changes to UK legislation might be necessary following Italian Republic,businesses could choose either to continue to use the margin scheme or to rely upon the ECJ judgment and treat the sale of input tax blocked cars as being exempt.

[85]On 20 October 1997 and 3 November 1997, Furness Motor Co Limited, which had become part of the appellant’s group in October 1996, received a circular from Armstrong Watson & Co, Chartered Accountants about the possibility of Italian Republic claims for overpaid output tax accounted for under the margin scheme.

[86]It seems that the appellant did make a limited claim in respect of tax overpaid on bonus payments, because on 17 June 1998 HMRC authorised a repayment of VAT in respect of “dealer bonuses”. The evidence does not identify any detail about the claim other than a letter dated 17 July 1998 stating that statutory interest of £88.60 had been authorised by HMRC. There is no evidence that this claim related to demonstrator bonuses.

[87]Meanwhile the validity of the three year cap introduced in December 1996 was being challenged. On 11 July 2002, the ECJ determined in Marks & Spencer that the UK was in breach of EU law having introduced the limitation period retrospectively and without making any provision for a transitional period.

[88]On 5 August 2002 HMRC issued Business Brief 22/02 in which HMRC invited new claims to be made with a new limitation period ending on 31 March 2003, subject to establishing that the taxpayer had previously made a capped claim or otherwise discovered its error in overpaying tax prior to 31 March 1997.

[89]On 8 October 2002 HMRC issued Business Brief 27/02 in which the period for submitting claims was extended to 30 June 2003. The date before which a taxpayer must have made a capped claim or discovered the error was also extended to 30 June 1997.

[90]At the meeting between HMRC and motor industry representatives in December 2002, referred to above, Mr Lewis of HMRC made reference to the validity of claims. In that context he stated: BL said that he is satisfied that sufficient publicity had been given, so that all businesses had information about developments, and had therefore discovered [the possibility of making a claim], before 30 June 1997.

[91]Some time after January 2003, HMRC issued their general guidance to motor dealers in relation to claims arising out of Elida Gibbs, including the original Elida Table.

[92]On 10 March 2005 Grant Thornton sent a letter to HMRC on behalf of the appellant which was identified as a ‘protective claim’ for VAT overpaid on bonus payments. A quantified claim was subsequently made on 19 October 2007. The fact of these claims was agreed between the parties, but I was not taken to copies of the claims.

[93]On 24 August 2006, HMRC issued Business Brief 13/06, following the judgment of the Court of Appeal in Michael Fleming (t/a Bodycraft) v HMRC

[2006] EWCA Civ 976

. This concerned the sufficiency of the action taken by HMRC in the Business Briefs referred to above in remedying the defects in UK law identified in Marks & Spencer. The Business Brief invited claims or the resubmission of previous claims for VAT overpaid in periods prior to 4 December 1996.

[94]The guidance issued in relation to Elida Gibbs and the Elida Table was revised and reissued on 20 October 2006 as previously described.

[95]On 23 January 2008 the House of Lords issued its judgment in Fleming. The House of Lords recognised the necessity for a transitional period to be set out in legislation. It also recognised that the requirement outlined in the Business Briefs, which required taxpayers to establish that they had previously made a capped claim or otherwise were aware of the error did not give effect to a taxpayer’s accrued rights under EU law.

[96]In response to the judgment in Fleming, Parliament enacted s 121 Finance Act 2008 which disapplied the three year time limit insofar as it applied to claims made for VAT overpaid prior to 4 December 1996. It provided that such claims had to be made before 1 April 2009.

[97]The appellant made claims pursuant to that extended time limit on 31 March 2009. The claims included claims for repayment of VAT overpaid on demonstrator bonuses and on sales of demonstrator vehicles. It also made a claim for interest on the sums overpaid. At that stage compound interest was being claimed.

[98]HMRC agreed those claims in a decision dated 31 March 2011 and a review decision dated 10 December 2013. However, I understand that it only agreed to pay interest from 31 March 2009, the date on which the error was notified to HMRC.

[99]Finally, I should say that the evidence before me included a witness statement from Mr Kevin Tedstone, an accountant employed by the appellant. He adduced some of the documentation referred to above between 1994 and 1998 in relation to what the appellant knew about the possibility of a claim for repayment of VAT on demonstrator bonuses. He was not responsible for the appellant’s VAT at that time. However, his evidence was that in 2001 or 2002 he was asked by the appellant’s managing director if the appellant could submit a claim for VAT overpaid on demonstrator bonuses. He informed the managing director that it was not possible to make a claim because of the three-year cap.

[100]Mr Tedstone’s evidence was not challenged and I accept it.

[101]It is worth pointing out at this stage that the appellant had no capped claim to make in 1997 when Elida Gibbs was decided because from 1986 onwards Ford had adopted non-VAT credit notes to account for the demonstrator bonus. Hence, the bonus was treated as being outside the scope of VAT and the appellant had not accounted for VAT on any supply of services to Ford. Ford itself should have treated the bonus as a discount which reduced the value of the supply to its dealers but did not do so. Ford therefore had a claim for overpaid output tax in the period 1986 to 1995 which was eventually paid by HMRC. Ford also claimed statutory interest on that amount pursuant to s 78 VATA 1994, but that claim was rejected by the FtT in the Ford decision released on 8 May 2017. Consideration of the issues Issue 1

Consideration of the issues

[102]Issue 1 is directed to the circumstances in which the appellant came to overpay VAT in the Relevant Period, and whether the overpayment was caused by the Alleged Bonus Error of HMRC.

[103]I have noted above that the alleged error by HMRC in Issue 1 is a representation that bonus payments should be treated as consideration for a taxable supply of services. The grounds of appeal state that HMRC represented that in general such payments should be treated as consideration for a taxable supply. The appeal was put forward on the basis of the allegation in the grounds of appeal and that is the basis on which I approach Issue 1.

[104]HMRC do not accept that they made any error in their representations to the motor industry as to how demonstrator bonuses should be treated. They contend that in the case of bonuses which were paid in the line of supply, which includes the demonstrator bonuses paid by Ford to the appellant, they had always recognised such bonuses as discounts which reduced the value of the supply.

[105]For the purposes of Issue 1, I must make findings as to HMRC’s views on the treatment of demonstrator bonuses, whether representations by HMRC as to how such bonuses should be treated involved an error on the part of HMRC and whether it was due to any such error that the appellant overpaid VAT on demonstrator bonuses. The appellant must establish the following facts and matters on the balance of probabilities, namely:(1) that HMRC made representations as to the correct VAT treatment of demonstrator bonuses;

(2) that those representations were disseminated to motor traders by HMRC;

(3) that the representations were incorrect; and

(4) as a result of HMRC’s error, the appellant overpaid tax because it treated demonstrator bonuses as supplies of services to manufacturers.

[106]Mr Rycroft acknowledged that there was no specific ruling given by HMRC to either Ford or the appellant. Further, the evidence does not include any document publicly available at the time which describes HMRC’s views or guidance. However, I accept that it is not necessary to identify a specific ruling, as such, in considering whether there was an error on the part of HMRC. It is sufficient if HMRC disseminated its views to the motor industry generally in the form of representations which might reasonably be relied upon by motor traders. For present purposes I shall refer to HMRC’s representations, rather than policy, because that is the term used by the parties in defining Issue 1. The fact that representations were not published generally or given specifically to the appellant does not mean that the appellant cannot rely on those representations for the purposes of s 78.

[107]Mr Puzey did not suggest that HMRC did not make any representations to the motor trade as to how demonstrator bonuses should be treated for VAT purposes. He was right not to do so, because the correspondence in 1977 and 1978 was at least representing how demonstrator bonuses might be treated. HMRC’s representations were made to the SMMT and I am satisfied that the motor trade generally, including the appellant, was entitled to act upon them. The focus in considering Issue 1 is therefore:(1) what representations were made by HMRC and were those representations incorrect;

(2) did the appellant overpay output tax on demonstrator bonuses in the Relevant Period as a result of those representations.

[108]Mr Puzey submitted that HMRC’s representations in the 1977 and 1978 correspondence were not a full statement of its position. The representations did not identify that the general treatment of demonstrator bonuses as consideration for a taxable supply only applied in relation to payments which were not in the line of supply. For payments in the line of supply, which covered payments by Ford to the appellant, HMRC’s view was always that they should be treated as a retrospective price reduction. Payments in the line of supply would fall within the proviso to the general rule.

[109]I am satisfied from the correspondence in 1977 and 1978 that HMRC was putting forward a general rule. The general rule was that demonstrator bonuses should be treated as a taxable supply of services. However, HMRC acknowledged that in appropriate circumstances a demonstrator bonus should be treated as a price reduction. Mr Puzey submits that this proviso to the general rule would have covered demonstrator bonuses paid in the line of supply. As such, the representations were not incorrect. It was always open to the appellant to assert that the demonstrator bonuses it received from Ford were a genuine price reduction because they were paid in the line of supply.

[110]Mr Rycroft accepted that the representations made by HMRC to the SMMT was subject to a proviso that if a taxpayer was able to demonstrate to its local VAT office that the discount represented a genuine retrospective price reduction, then it would be treated as such. However, he submitted:(1) There was no reference whatsoever to what has now become known as the line of supply in the documents from this period, and

(2) Taxpayers could not be expected to challenge the general position because they would infer that they would have to have good reason to do so.

[111]The contemporary correspondence does not identify what might have been treated as a “genuine retrospective price reduction”. The question therefore is what did HMRC and indeed the motor trade generally at this time regard as a genuine price reduction? The documentary evidence from 1977 and 1978 provides no clues. The only factor which either party has ventured to suggest might have been treated as a genuine price reduction is Mr Puzey’s submission that it would include a demonstrator bonus paid in the line of supply by a manufacturer to a dealer.

[112]The first reference to the line of supply as having any significance is in the Burnes Letter dated 24 October 1988. Officer Burnes states in the fourth paragraph of that letter that treating a bonus as a discount only applies where it is paid in the line of supply, although that form of words is not used, and not where a finance company was interposed.

[113]Mr Rycroft sought to explain the Burnes Letter read as a whole as establishing that HMRC had changed their view, held until 1987, that demonstrator bonuses generally represented consideration for a taxable supply. In that regard he relied heavily on the Burnes Note. He submitted that HMRC’s error was clear from the Burnes Note and relied on the fact that Officer Burnes quoted the 1978 correspondence and recognised that Ford’s new scheme was inconsistent with that policy. When it referred under the heading “Present Agreement” to the “method adopted … by most car dealers” as issuing a tax invoice to manufacturers, he submitted that Officer Burnes was recognising a link between the 1978 representations and the VAT treatment adopted by motor traders. Officer Burnes went on to accept that the Ford scheme would reduce the value of a self-supply by the bonus, and that this “overturns” the current advice. Mr Rycroft submitted this was clear evidence that HMRC was changing its position and that motor traders had relied on the previous representations.

[114]It seems to me that the significance of the Burnes Note, the Burnes Further Note and the Burnes Letter lies principally in whether they evidence a change in HMRC’s position in 1987 and 1988, and if so in what way HMRC’s position changed. In considering those matters, I must also consider the reliability of what is said by Officer Burnes in those documents.

[115]It is difficult to reconcile the contents of the Burnes Note with the SMMT guidance which had been issued a year earlier. Officer Burnes was unaware of that guidance at the time of the Burnes Note. The first approach to demonstrator bonuses described in the SMMT guidance is to treat them as discounts, reducing the value of the supply. Only if the payment represents a payment for services is the payment treated as consideration for a supply of services by the dealer. However, Officer Burnes states that there was a general rule that “no such payment would normally be accepted as a discount”. Officer Burnes also stated that “most car dealers” were issuing tax invoices for services supplied to manufacturers in relation to the payments. However, I am satisfied from the Elida Table that prior to 1986, only Ford, Peugeot, Rover and Saab (for 1979) issued VAT invoices. There were dealers of numerous other manufacturers who did not issue invoices.

[116]There were a large number of attachments to the Burnes Note, none of which were in evidence. It is also notable that Officer Burnes does not at this stage explain what circumstances were treated as falling within the proviso to the general rule. The note does refer to calls from dealers raising queries as to whether demonstrator bonuses reduced the value of the supply but does not indicate the nature of those queries or how they were answered.

[117]In relation to the acceptability of the new Ford Scheme, Officer Burnes refers to it being necessary to look at the incentive payments in more detail. He refers to a general rule that they would have to be accepted as reducing the value of a supply which would “overturn” the current advice. If the Burnes Notes is a reliable description of HMRC’s views over time and the practices of motor dealers then that is good evidence that there was a change in HMRC’s views in 1986 or 1987. However, I cannot take the contents of the Burnes Note at face value, in light of the fact that Officer Burnes was not fully informed as to the SMMT guidance and how car dealers were in practice accounting for VAT on demonstrator bonuses.

[118]The Burnes Further Note also states in the first paragraph that “ [the Burnes Note] suggests we write to the SMMT explaining how our views have changed since our 1978 ruling”. Officer Burnes then goes on to say that in light of the SMMT guidance, which he had not seen until then, it would still be useful “to formally clarify the issue with SMMT”. I infer that he was there accepting that his previous understanding as to the treatment of demonstrator bonuses, at least in practice, had been incorrect. He considers that the circumstances in which an incentive payment can qualify as a contingent discount should be explained in more detail. It seems to me that is a reference to the proviso in the general rule. There is no indication in the Burnes Further Note what that explanation might entail. However, the Burnes Letter dated 24 October 1988 does appear to explain the circumstances where the proviso would apply.

[119]The Burnes Letter appears to adopt the approach in the 1986 SMMT guidance. In other words, prima facie a demonstrator bonus will be treated as a discount. It goes further than the SMMT guidance in that it identifies two circumstances where a payment will not be treated as a discount. Firstly, where the payment is not made in the line of supply and secondly where it represents a payment for services from the dealer to the manufacturer.

[120]It is also notable that the Burnes Letter does not inform the SMMT of any change in HMRC’s views on the treatment of demonstrator bonuses. Rather, it states that “it might be useful to clarify the position” because HMRC receives queries from motor dealers.

[121]Mr Rycroft also relied on the letter dated 30 October 1987 in relation to incentive payments to high volume purchasers. The letter records HMRC’s policy to treat those incentives as outside the scope of VAT, because they were not discounts in the line of supply and there was no evidence that they represented a supply of services by the customer to the manufacturer. However, the letter goes on to say that HMRC are now aware that services are supplied by customers to the manufacturers, and that VAT should be accounted for by customers with manufacturers being entitled to input tax credit.

[122]Mr Rycroft submitted that this was clearly a change in position by HMRC in the treatment of volume incentive payments. However, the change was inconsistent with HMRC’s case on this appeal that they always viewed payments which were not in the line of supply as taxable supplies, unlike payments which were in the line of supply. One would expect demonstrator bonuses to dealers outside the line of supply and volume bonuses to customers to have the same treatment.

[123]I do not consider this letter to be significant. It concerned a different type of bonus and provides little if any assistance in determining the issue relating to demonstrator bonuses. Indeed, the letter can be read as consistent with HMRC’s case on demonstrator bonuses. It says that volume bonuses are not true discounts “because they do not relate directly to the supply”. By implication, if they were paid in the line of supply, they would be treated as discounts unless there was evidence of a supply of services.

[124]Mr Rycroft submitted that documents up to and including the Burnes Letter in 1988 were more or less contemporaneous with the end of the Relevant Period and more weight should be given to those documents. I should be cautious about subsequent documents relied on by HMRC, dating from 1997 onwards. I take that submission into account in determining what inferences I can properly draw from the documentary evidence. Before considering the later documents, it is convenient to deal with a number of arguments relied on by Mr Puzey.

[125]First, that there was no evidence as to any HMRC representations between 1973 and 1977. I do not accept that is the case. The letter dated 25 October 1977 sets out how vehicles had been valued for the purposes of self-supply since the introduction of VAT in 1973. What is not clear is whether the proviso to the general rule was always in effect or was introduced as a result of a challenge by the SMMT in December 1977. In the circumstances, I cannot be satisfied that the general rule applied without the proviso in the period 1973 to 1977. The appellant’s case therefore stands or falls in relation to the whole of the Relevant Period.

[126]Secondly, that the treatment described in the 25 October 1977 letter referred to discounts “received” subsequent to the time of supply, whereas the 1 February 1978 letter referred to discounts being “allowed” subsequent to the time of supply. Mr Puzey submitted that this begged questions as to whether they meant the same thing, what is the time of supply and what would the position be if a discount was allowed before the self-supply but only received afterwards.

[127]I do not consider that the different wording identified by Mr Puzey indicates any uncertainty about HMRC’s position at that time, or that it indicates that motor dealers should not rely on the representations. It seems likely that the distinction was simply between discounts that were paid and discounts that were credited to the dealer.

[128]Thirdly, that HMRC “back-pedalled” in 1977 when challenged by SMMT. I am satisfied that the SMMT sought to challenge the treatment described by Miss Jackson in her letter dated 25 October 1977. However, it is not clear how they sought to challenge the treatment, or whether the treatment was understood by both parties to this correspondence to have always included the proviso described in HMRC’s letter dated 1 February 1978

[129]Fourthly, it is not known what guidance was issued by the SMMT in light of the 1978 correspondence. There is force in this submission. The evidence shows that by 1986, SMMT had a position in relation to demonstrator bonuses, described in their 1986 report. Again, it is clear from that guidance that the treatment depended on the detail and form of the demonstrator bonus. Bonuses could fall within sub-paragraph (a) or (b). Sub-paragraph (a) of that guidance involved treating demonstrator bonuses as a discount which reduced the value of a self-supply. Sub-paragraph (b) involved treating the bonuses as payment for a taxable supply of services by the dealer to the manufacturer. As with HMRC’s representations, there was no indication as to how dealers should distinguish the different types of demonstrator bonus.

[130]The SMMT guidance contains no hint or suggestion that it was based on a recent development in HMRC’s position. The description of treating a discount as reducing the value of the supply is crystal clear and it is option (a). The only matter of discussion between the SMMT and HMRC appears to concern the use of credit notes, which it is stated would not be appropriate where the demonstrator bonus is treated as a taxable supply. Otherwise, the tone of the SMMT report is consistent with describing a long established position. If it was a new treatment, then it is likely this would have been made clear.

[131]The SMMT guidance which was operative in 1986 is consistent with HMRC’s representations in 1977. The difference is simply one of emphasis. Option(a) is the proviso to the HMRC general rule, whereas option

(b) is what was described by HMRC as the general rule. If, as I consider, the difference is one of emphasis, it points towards the SMMT guidance reflecting the way in which motor dealers would have been likely to approach the VAT treatment of demonstrator bonuses. Subject always to the analysis which HMRC and the SMMT guidance both required. I cannot say when the SMMT guidance was made available to motor dealers. It is certainly possible that it was available in or about 1978 at the time of discussions between the SMMT and HMRC.

[132]I now turn to later documents relied on by Mr Puzey in support of HMRC’s case on the nature of the representations in 1977 and 1978. Those documents were produced in 1997, 2003 and 2006 and all stated or at least implied that it was always HMRC policy that payments in the line of supply would be treated as discounts.

[133]The first document is the DCL from June 1997. The second group of documents comprise the original Elida Table in 2003, together with associated correspondence and the revised version in 2006. The third document is the internal HMRC document dated 15 February 2006 in connection with various tribunal appeals.

[134]Mr Rycroft submitted that the DCL was not reliable in that it simply expressed the belief of the writer at the time it was written. It was not a reliable statement of HMRC’s position in the period up to 1986. That point is well made. I am not satisfied that the DCL description of the treatment of demonstrator bonuses goes back to 1986. Indeed, there is an indication that it does not go that far back. In relation to fleet bonuses, the DCL states that as far as the writer is aware, the only common approach has been to treat them as a supply of services. However, that is inconsistent with the letter dated 30 October 1987.

[135]Mr Rycroft submitted that the views of Mr Bob Lewis expressed at the meeting in December 2002 which in due course led to the Elida Table were not factually correct. The original guidance did not suggest any distinction involving the line of supply. He suggested that this was a distinction that only gained currency with HMRC officers after the ECJ decision Elida Gibbs. I do not accept that submission. As I have found, the distinction was first referenced in the Burnes Letter in 1988. However, I give little weight to the observations made at this meeting given that they were made some 16 years after the relevant period.

[136]I take the same view of what is said in the Elida guidance issued after January 2003, the revised Elida Table and the internal review document produced in connection with other appeals.

[137]Mr Rycroft argued that the fact Ford accounted for VAT consistently with the alleged representations was itself evidence of those representations. I accept that fact is part of the evidence to be considered in making findings as to the nature and extent of HMRC’s representations and whether the appellant acted on the basis of those representations. However, the argument assumes what it sets out to prove.

[138]Finally, Mr Rycroft submitted that the ordinary consequence of the representations made by HMRC was that traders would act in accordance with those representations. That is what the appellant did and that is what resulted in the overpayment of VAT. Traders would have thought that they needed a good reason to challenge the general rule in any particular instance and that there was no point challenging the general rule because of the margin scheme. To the extent that there was any confusion, the source of that confusion was HMRC.

[139]Standing back and looking at the evidence as a whole, I am satisfied that the question of VAT on demonstrator bonuses was a matter of importance to the motor industry going back to at least 1977. HMRC’s view of the matter was subject to scrutiny by the SMMT. I do not consider that manufacturers and dealers simply accepted that demonstrator bonuses could never give rise to a genuine discount, or that that was HMRC’s inflexible position. The issue was always regarded as fact sensitive, in particular as to whether there was a genuine discount or whether a dealer was providing services to manufacturers in making demonstrator vehicles available to potential customers.

[140]HMRC represented that demonstrator bonuses would not generally be treated as discounts, but there was a proviso to that general rule and motor dealers would have been aware that there was a proviso. Each case would be looked at on its own merits. On balance, I am satisfied that HMRC would apply the proviso to demonstrator bonuses which were paid in the line of supply. There was therefore no reason that the appellant should have been misled by what was described as the general rule. It could have applied the proviso, or at least queried the position with HMRC. There is no suggestion that the appellant did query the position with HMRC, unlike other motor dealers as recorded in the Burnes Note.

[141]In the light of all the evidence, I am not satisfied on the balance of probabilities that the overpayment of VAT in the Relevant Period was caused by an error in representations made by HMRC. It is possible that the appellant did rely on the general rule put forward by HMRC and for whatever reason did not seek to apply the proviso. It is equally possible that the appellant applied the SMMT guidance but for one reason or another, including its own mistake, considered that demonstrator bonuses fell within limb (b) of that guidance. Either way, I am not satisfied that the appellant was led into error by HMRC.

Issue 2

[142]The appellant’s case on Issue 2 is put in the alternative to Issue 1. If the Alleged Bonus Error relied on in Issue 1 did not cause the overpayment, then HMRC’s Margin Error caused the overpayment. It is said that without the Margin Error, manufacturers and/or motor dealers would have sought clarification from HMRC on whether VAT was due on the demonstrator bonuses. Mr Rycroft says that based on HMRC’s case in this appeal, the advice would have been that the bonuses should be treated as a discount because the payments were made in the line of supply.

[143]Mr Rycroft relied on the decision in Wheeler where it was said to be “repugnant to common sense” to say that the overpayment of tax was caused by an “error” of HMRC. He submitted that in the present appeal it would be repugnant to common sense to say that the tax was overpaid as a result of an error by the appellant in taxing the demonstrator bonus when HMRC’s position would have been that tax was due in any event on the margin. I do not consider that the decision in Wheeler takes the appellant’s case any further forward. It is clear that the tribunal in Wheeler did not view the assessment made by HMRC as an error. The tribunal found that it was made to best judgment in circumstances where the taxpayer had failed to provide evidence to establish that no tax was in fact due.

[144]HMRC originally argued that the appellant’s case on Issue 2 was fundamentally flawed. In the event however, Mr Puzey did not pursue that argument and accepted that the question on Issue 2 is whether it can be said that the real cause of the overpayment was the Margin Error. He submitted that the appellant’s case rests on an economic theory as to what dealers would have thought and done in the 1970’s and 1980’s. In particular that the effect of the margin scheme was that they would have thought that there was no point challenging the VAT treatment of demonstrator bonuses.

[145]I agree with Mr Puzey that there is no evidence to support the appellant’s case. Further, the appellant’s case is inconsistent with what evidence there is. Namely, that:(1) local VAT offices often received calls from dealers querying the treatment of demonstrator bonuses and whether they operated as a discount, and

(2) the SMMT did actually challenge HMRC’s guidance in the 1977 correspondence.

[146]Mr Rycroft accepted that there was an advantage in treating demonstrator bonuses as a discount if vehicles were sold at a loss. That advantage would arise irrespective of the margin scheme. Whilst I cannot say to what extent the appellant’s demonstrators were sold at a loss, I am not satisfied that they were all sold at a profit.

[147]Further, there would be cashflow advantages to motor dealers if demonstrator bonuses were treated as a discount. In that case, dealers would not account for VAT on the discount when paid or received. They would only account for VAT on the increased margin when the vehicle was sold, some months after the self supply. Mr Rycroft accepted that there were cashflow advantages, and I have found as a fact that there was at least some incentive for the appellant to challenge or clarify the treatment of demonstrator bonuses.

[148]Mr Rycroft suggested that the latter two points were exceptions to a general rule and would not have affected the thinking of dealers at the time. I should infer that the financial impact of these advantages would not be a real consideration for dealers. I cannot accept that submission. I am not in a position to say how these factors might have influenced motor dealers in their consideration of the VAT treatment of demonstrator bonuses. Nor is there any evidence that the vehicles in respect of which tax on demonstrator bonuses was overpaid by the appellant were also vehicles on which VAT was overpaid on the profit margin. The appellant has assumed that all demonstrators were sold at a profit, without any evidence to that effect.

[149]Looking at the evidence as a whole, I am not satisfied that the Margin Error was a real cause of the overpayment of VAT on demonstrator bonuses. In my view it is too remote from the overpayment and I cannot say on the balance of probability what motor dealers would have done if the Margin Error had not been made.

Issue 3

[150]The appellant’s case on Issue 3 is in the alternative to its case on Issue 1 and Issue 2. The appellant says that it suffered a delay in receiving repayment of the amounts overpaid due to errors of HMRC in the manner of its introduction of a three-year cap on claims for overpaid VAT with effect from 5 December 1996.

[151]The three year cap was introduced without a transitional period. In July 2002, the ECJ held that the limitation period was invalid because there was no transitional provision. Subsequently, in 2008, the House of Lords held that HMRC’s attempts to remedy that defect by way of Business Briefs were insufficient. The judgment of the House of Lords led to the introduction of s 121 Finance Act 2008 which disapplied the three year cap in relation to claims for VAT overpaid prior to 4 December 1996, providing that a claim was made before 1 April 2009. The appellant made its claim for repayment of VAT on demonstrator bonuses paid in the Relevant Period on 31 March 2009. It also claimed interest under s 78 VATA 1994.

[152]The appellant says that it was aware that it could claim overpaid tax on demonstrator bonuses in at least 1998. HMRC’s letter dated 17 July 1998 evidences that the appellant did make a small capped claim for repayment of such tax. This was despite the fact, evidenced by the Elida Table, that Ford dealerships ceased to account for VAT on demonstrator bonuses after 1986. It is also the case that a large number of non-Ford dealers made capped claims in 1996.

[153]The appellant’s case is that but for the erroneous introduction of the three year cap, the appellant would have made its claim for repayment of VAT overpaid in the Relevant Period if the three-year cap had been introduced with a transitional period in 1996. In particular, it was aware in or about 1996 that it had overpaid output tax on demonstrator bonuses. I am not asked to identify any particular period as the period for which interest should be calculated if the appellant is successful on Issue 3.

[154]Mr Rycroft tentatively relied on the evidence of Mr Tedstone to the effect that he had advised in 2001 or 2002 that it was not possible to make a claim for VAT overpaid on demonstrator bonuses because of the three-year cap. He submitted that the same reasoning would have applied in 1997, but accepted that such evidence had little weight when it came to inviting me to draw an inference as to the position in 1997. I agree, and for that reason I give little weight to Mr Tedstone’s evidence in that regard.

[155]The appellant relies on the fact that it made a claim in May 1994 on the same basis as the Royscot Leasing claim. It invites an inference that it kept abreast of developments in VAT and when it received advice to make a claim, it did so. It was willing to make claims for overpayment of VAT in appropriate circumstances.

[156]The appellant also relies on the fact that it was being courted by firms of accountants in 1997 with a view to making claims following Elida Gibbs and Italian Republic. As such I am invited to infer that it was aware of the possibility of making such claims, and but for the three-year cap it would have made the claim at this time.

[157]I acknowledge and find that where the appellant was aware that it had overpaid VAT, it was willing to make a claim for repayment. It did so in relation to the input tax block on the purchase of demonstrator vehicles.

[158]Overall, I am not satisfied on the evidence that the appellant was aware in 1996 or 1997 that it had overpaid output tax on demonstrator bonuses in the Relevant Period. There is no direct evidence and I am unable to draw inferences to that effect on the evidence that is available.

[159]There is no evidence that the appellant’s claim in 1998 related to demonstrator bonuses. HMRC’s letter refers only to “dealer bonuses”. At that time, it must have been a claim subject to the three year cap and there is no evidence as to why the appellant might have incorrectly accounted for output tax on demonstrator bonuses in the period 1995 to 1998.

[160]I accept that the appellant received flyers from firms of accountants in 1997. If the three year cap had been introduced with a transitional period in December 1996 then I infer that those accountants would have referred to the possibility of making uncapped claims in relation to demonstrator bonuses. Mr Puzey submitted that the accountants courting the appellant were doing so in relation to Elida claims, where the bonuses did not follow the line of supply, and not in relation to claim’s such as the present where the bonuses did follow the line of supply.

[161]It is not clear that the accountants were only referring to tax overpaid on demonstrator bonuses which did not follow the line of supply. However, at that time the appellant had been correctly accounting for demonstrator bonuses since 1986. I cannot infer that the appellant would have been aware in 1997 that it had overpaid output tax on demonstrator bonuses prior to 1986.

[162]I am satisfied from Mr Tedstone’s unchallenged evidence that some consideration was given to reclaiming VAT overpaid on “demonstrators and bonuses” in 2001 or 2002. It is not clear that he is referring to VAT on the demonstrator bonuses paid in the Relevant Period. There is no evidence that the Appellant was aware in 2001 or 2002 of an overpayment of VAT on demonstrator bonuses in the Relevant Period, or that the appellant was in a position to quantify the amount of any claim it might have had. The evidence is and I find that the appellant was not aware it had a claim for the Relevant Period at any time prior to 10 March 2005 and it was not in a position to quantify that claim until 19 October 2007.

Conclusion

[163]For all the reasons given above, I am not satisfied that the VAT repaid to the appellant was paid as a result of any error on the part of HMRC. Nor that the appellant suffered any delay in receiving repayment as a result of errors by HMRC in the manner of the introduction of the three year cap prior to 10 March 2005 at the earliest.

[164]The parties were agreed that I should determine the issues raised at the hearing in principle. In the circumstances, I leave it to the parties to consider the implications of my findings and of this decision. The parties should consider how the appeal should be determined and should inform the Tribunal of their position within 60 days of the date of release of this decision.

Right to apply for permission to appeal

[165]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. JONATHAN CANNAN TRIBUNAL JUDGE Release date: 12th DECEMBER 2022