Claimants In the Royal Mail Group Litigation v Royal Mail Group Ltd

This action is the trial of certain issues that arise in this managed litigation in which around 340 claimants seek to establish that they are entitled to demand, or to have, VAT invoices from the defendant (“Royal Mail”) for a period spanning several decades. The actions derive from the fact that until 2009 it was, by and large, generally assumed that no VAT was chargeable on the postal services provided by Royal Mail because the services were exempt. The English VAT legislation seemed so to provide, and HMRC certainly acted on that footing. Then on 23 April 2009 the Court of Justice of the European Union (“CJEU”) ruled in a case called, for short, TNT1, that not all postal services were exempt. The universal postal service was exempt, but arrangements that were individually negotiated were not. That decision is said to have been a general surprise. (Footnote: ¹) It took almost 2 years for English legislation to be brought into line (or so it was thought) but disputes as to what services are and are not within the realm of chargeable VAT continue to be the subject of dispute.

The claimants, who include commercial and non-commercial organisations such as universities and local authorities, all mount claims that in the circumstances the amounts that they paid for their services for the period up to, and in some cases beyond, the revised legislation were VATable even though Royal Mail did not seek to charge VAT through its invoices, and claim to be entitled to VAT invoices from Royal Mail so that they can use those invoices to reclaim input tax from HMRC. Mr Cordara QC, who appeared for the claimants before me, said that an optimistic ball-park figure for the total of the claims, if successful, was £500m. Any global recoveries to which they may be entitled may be very much less than that, depending on such things as whether there is a cause of action, and if so its scope; limitation; whether certain services rendered since 2011 were VATable but not properly so treated by Royal Mail; and other matters. The ball-park figure gives an idea of the potential scope of the claims.

The litigation gives rise to a large number of issues. The view has been taken that it would not be sensible to try them all at once, even by taking individual litigants and running the whole of their claims, so attempts were made to separate out some issues to try to break up the litigation into manageable and sensible chunks and perhaps to whittle down some of the claims. The issues which I am invited to decide are principally targeted at answering some questions about limitation, which has an obvious potentially limiting effect on the scope of the claims. Unfortunately, the attempt at useful preliminary issues has suffered in part from the defects which not infrequently attend such attempts, in that the argument demonstrated that they were not really set up adequately, contrary to the expectations of the parties, as will appear.

In this action Mr Roderick Cordara QC led for the test claimants; Mr Javan Herberg QC led for the defendant Royal Mail.

In these preliminary issues no distinction is drawn between Royal Mail and its predecessors, including the Post Office and Consignia plc. They can be treated as one for the purposes of liability in this judgment. I shall use “Royal Mail” throughout.

From more or less the outset, European and UK VAT legislation provided for exemption for postal services. It is unnecessary to trace the versions of the exemption from time to time, and it is sufficient to take Directive and UK statute law in force at the time the matter took an unexpected turn in 2009.

In the UK this was given effect to by section 31 and Schedule 9 of the Value Added Tax Act 1994 (“VATA”). Section 31 provides for exempt supplies if they are as described in Schedule 9, and the Schedule describes:

“1.

The conveyance of postal packets by the Post Office.

2.

The supply by the Post Office of any services in connection with the conveyance of postal packets.”

Royal Mail operated, and operates, a number of services in addition to what the consumer would understand to be the regular mail delivery service. There were and are

Until the 2009 decision of the CJEU Royal Mail did not distinguish between those services for VAT purposes. They were all treated by all concerned as being within the statutory VAT exemption. The result was that no VAT invoices showing VAT were rendered. All concerned (Royal Mail, customers and indeed HMRC) worked on the footing that that was the correct approach - that is one of the agreed asssumptions (“the Assumptions”) for the purposes of this hearing, as will appear. Accordingly, Royal Mail did not account to HMRC for any output tax, and customers did not claim credit for any input tax. So far as invoices were concerned, when rendered they did not show, or purport to charge, VAT. Not all services necessarily generated any invoices in respect of all payments - for example, it is not apparent that invoices were generated in respect of all payments for franking services.

That state of affairs was challenged by another provider of mail and parcel services and the challenge was referred to the CJEU by the Administrative Court. The result was the decision in R (on the application of TNT Post UK Ltd) v HMRC [2009] STC 1438 - “TNT1”. In that decision the CJEU held:

“44.

It follows from the requirements referred to at [31] of this judgment that the exemption provided for in art.13A(1)(a) must be both strictly interpreted and interpreted consistently with the objectives of that provision, that the supplies of services and of goods incidental thereto must be interpreted as being those that the public postal services carry out as such, that is, by virtue of their status as public postal services.

…..

49.

Consequently, the answer to the second and third questions is that the exemption provided for in art.13A(1)(a) of the Sixth Directive applies to the supply by the public postal services acting as such—that is, in their capacity as an operator who undertakes to provide all or part of the universal postal service in a Member State—of services other than passenger transport and telecommunications services, and the supply of goods incidental thereto. It does not apply to supplies of services or of goods incidental thereto for which the terms have been individually negotiated.”

That required a reconsideration of how to implement the law, and the UK sought to bring its legislation into line with that decision by amending Schedule 9 with effect from 31 January 2011 to read:

“1.

The supply of public postal services by a universal service provider.

2.

The supply of goods by a universal service provider which is incidental to the supply of public postal services by that provider.”

Various Notes follow which I do not need to set out here.

That change has not, apparently, removed all disputes about the scope of the exemption. For example, in R (on the application of Whistl UK Ltd v HMRC [2015] STC 1077 (“TNT2”) TNT, under its new name of Whistl, challenged the application of the newly formulated exemption to certain services, and sought Francovich damages for the state’s past failure to limit the exemption appropriately. The details of that case do not matter for present purposes. In relation to the damages claim, Kenneth Parker J held that the UK’s interpretation of the legislation was “reasonably tenable” (para 103). I am told that some of the claimants in the present managed litigation raised other challenges in relation to other services.

The present litigation arises out of a complaint by the claimants that they should be entitled to have VAT invoices to enable them to claim input tax for various services which they say were in fact VATable both before and after TNT1. In this litigation they say that they have an entitlement to have such invoices in respect of various supplies rendered across 3 periods - the period leading up to TNT1, the intervening period pending amendment of the UK legislation, and the period from 31 January 2011 (after the amendment). Without such invoices they say that they are deprived of the opportunity to claim input tax which they are entitled to deduct, even for the period when no-one thought that there was any input tax to deduct.

The claimants in this litigation vary in terms of the services used and the periods of use. I am not yet privy to any particular individual issues that arise but any that exist do not arise for determination at the moment. The form and nature of the claims are all the same. They claim declarations that the claimants are entitled to VAT invoices, orders that they be provided, and damages for not providing invoices. At this stage there are three test claimants - Harrier LLC, H Tempest Ltd and Leicester City Council. They seem to have been chosen as vehicles for placing before me the various forms of contract on which I was addressed, and for no other purpose.

There is some parallel litigation that is relevant. One of the present claimants in this litigation (Zipvit Ltd) is pursuing a different route for the recovery of input tax in respect of the pre-TNT1 period. It sought to make a claim for repayment of input tax on Royal Mail invoices for pre-TNT1 services without being able to produce VAT invoices to back it up (for obvious reasons). The legislation provides that HMRC has a discretion to accept alternative evidence of the payment of input tax in place of a VAT invoice, and Zipvit invited HMRC to exercise that discretion in its favour and thus accept a claim for repayment of input tax for a pre-TNT1 period without invoices. The procedure started in 2009. The Commissioners rejected the claim and it was pursued through challenges which have so far failed at all levels up to the Court of Appeal (Zipvit v HMRC [2018] STC 1502), but an appeal to the Supreme Court is now pending. This managed litigation started as a parallel track in 2015, and Zipvit was one of the claimants. If Zipvit fails in that litigation for want of VAT invoices and the absence of the exercise of discretion in its favour, then it, and others, turn to this litigation in order to be able to get invoices to establish their claim. If that appeal succeeds so as to allow, or even require, the exercise of the discretion, then it would remove or reduce the need for this litigation (probably). Because of the length of time that the Zipvit litigation was thought likely to take it was apparently not thought right to postpone the present claims pending the final determination of that litigation (which might, on one scenario, involve taking proceedings to Europe). It may also be (I do not know) that the parties had limitation in mind.

It will be apparent that HMRC has an indirect interest in these proceedings. To the extent that they ultimately result in the issue of VAT invoices, HMRC will be faced with many claims for repayment of input tax amounting to many millions, if not tens or hundreds of millions, of pounds. I was told that they have been aware of these proceedings in the past, but made no attempt to intervene. At my request, at the start of the hearing before me a letter was sent to HMRC again pointing out these proceedings and referring to the risk to HMRC, and informing them of the hearing. As I understand it nothing significant was heard from HMRC in response to that letter.

The points made in this section are parenthetical in that they do not go to the resolution of the dispute, but they were points made to me by the parties, and it is right that I record them here.

For his part Mr Herberg said that these cases involved claims to what would be a substantial windfall. For many years (in some cases decades) all parties carried on their business on the economic assumption that VAT was not chargeable on the relevant transactions and priced accordingly. Royal Mail did not account for output tax and

Mr Cordara did not accept that analysis as to the economic merits of the claim. He sought to demonstrate, at a very generalised level, that where VAT which ought to be charged is not charged in the middle of a chain of transactions (which is where the subject transactions lie, at least in cases where the claimants are commercial concerns) then what happens is that subsequent traders charge VAT on a sum which already includes VAT, which distorts the system and leads to a situation where the price to the ultimate consumer is inflated.

These points are not relevant to the questions that I have to decide, which are technical and not merits- or economics-based, and I do not have to develop or consider them. They may or may not come back at a later stage of these proceedings if the question of the grant of a discretionary remedy comes into play, and they have not been fully developed. I will confine myself to my own irrelevant observation that it is apparent from the judgment of Henderson LJ in the Court of Appeal in the Zipvit litigation that his view of the economic merits seems to coincide more with Mr Herberg’s than with Mr Cordara’s.

Apart from some sample contracts, no evidence was adduced on the trial of these issues, not even by way of agreed witness statements. The parties agreed certain clearly stated Assumptions, but, as it turned out (and as ought to have been anticipated) the original assumptions were not sufficient without some additional background matters to provide all the evidential material required to enable all the issues to be decided. Some of the gaps were filled in as we went along by the acceptance of the parties from time to time of some obvious and agreed background facts to provide necessary context. Sometimes it transpired that facts upon which one side or the other (usually the claimants) wished to rely were not evidenced and were not agreed as such, so they could not be deployed. By and large the parties managed to work round that sort of matter, but it did mean that I did not necessarily have a proper picture of some aspects of the case and some submissions of counsel could not be entertained. At its most serious it led the parties to withdraw some important restitution points from the issues which I was to decide, which is unfortunate but which was, in my view, inevitable.

The background facts to the transactions in question (which do not figure in any agreed documents) are those which appear in the preceding ‘History and background’ section of this judgment. They should be taken as facts agreed by the parties (for the purposes of this trial). In addition to that picture the parties agreed the following Assumptions on which the hearing was to take place. It was emphasised that they were assumptions and in later stages of the litigation one or other of the parties would or might wish to operate on a different basis in individual cases; but they were necessary and useful assumptions to enable decisions to be made on legal points which could be used to redefine the litigation, and the scope of claims, for the future. It may be necessary to introduce the odd additional fact into the decision-making process in this judgment. If I do it can be taken to have been another agreed assumption for the purposes of this hearing unless the contrary appears.

The general Assumptions on which argument took place and on the basis of which I am invited to make my decisions, as modified during the hearing, were as follows (and I repeat them in Appendix 3 for ease of reference):

Certain comments fall to be made in relation to those Assumptions.

Assumption 1 is very much one whose scope is limited to this particular phase of the managed litigation only. Which services were and are chargeable to VAT is a very live dispute in relation to a number of services across the period of the claims.

Assumption 2 again skates over a very live dispute. Royal Mail does not consider itself to be an emanation of the state, which would have the effect that Directives are directly enforceable against it, and that will be an argument for another day. This Assumption is made so that the effect of the Limitation Act 1980 can be tested in these proceedings on the assumption that it is an emanation. A decision on that will serve the useful function of assessing the levels of claims which will be sustainable against the Royal Mail if and insofar as the Assumption turns out to be correct.

As to Assumption 3, no invoices were in evidence before me, but the underlying hypothesis of the whole claim is that all invoices on all disputed services said nothing about VAT. In those circumstances it is assumed that the effect of the law is that the stated consideration included VAT, and it is on the footing that that assumption is accurate that the claimants say they are entitled to an invoice which correctly breaks down the consideration into price plus VAT, which they can in due course present to HMRC to claim their input tax back.

The parties did not prepare the statement contemplated by Assumption 4. Casual descriptions, apparently not disputed, were provided in the course of submissions on the particular forms of contract in issue in this case.

Assumption 5 generated a small amount of controversy. Its original form confined the understanding to the claimants, though Mr Herberg expressed the view firmly that it ought to be taken to be shared by Royal Mail, both because that is his case up to TNT1 in 2009, and because all three test claimants (and apparently many if not all others) pleaded that it was the understanding of both parties (actually not confined to the period up to 2009). After a certain amount of hesitation Mr Cordara accepted that it could be amended so that it assumed the same understanding on the part of Royal Mail, and then that it was also the position of the Revenue which seemed to have taken the same view. (For my own part I observe that the Zipvit case seems to have involved a finding that at least until 2009 Zipvit and Royal Mail believed the services to be exempt.) The practical effect of that is that any determinations I make based on matters to which this Assumption is relevant will operate only in relation to the pre-TNT1 period (because after TNT1 the parties no longer had that assumption), but the parties are content with that situation.

The last three Assumptions are relevant only to the restitution claims which have been taken out of this phase of the proceedings.

In order to get certain issues out of the way, the parties agreed the preliminary issues in the form in which they arrived before me. As I have already indicated, they are primarily intended to end up addressing certain limitation points that affect the claim. The issues are:

I have omitted the three issues which go only to restitution and which, as I have said, have been put on one side for the moment. Issues 5, 8, 9 and 14 were the subject of concessions by the defendant which meant that they did not need to be argued at the hearing before me. The other issues were grouped differently from the above for the purposes of argument and I shall deal with them below in a sequence which should make their development easier to follow.

For the purposes of this hearing the parties relied on the provisions of the Value Added Tax Act 1994 “VATA” so far as UK primary legislation is relevant, even though for part of the period of some claims earlier legislation would technically be applicable. It was not suggested that there was any material difference between the two phases of primary legislation.

It is unnecessary to set out many specific provisions of VATA at this point in this judgment. That Act, together with the regulations made under it, implement the UK’s obligations under the applicable Directives. It suffices for the present to observe the following:

The particular statutory duties relied on, which are said to be an implementation of what the Directives require, appear in regulations made under the statute. For early parts of the claim period the VAT General Regulations 1975 SI 1975/2204, and further regulations made in 1977, 1980 and 1985 would be the relevant ones. They are the regulations identified in the text of Issue 1. However, it was common ground that they are, for these purposes, to the same effect as the most recent regulations made under

Regulation 13 requires the provision of a VAT invoice:

“13.—

(1)

Save as otherwise provided in these Regulations, where a registered person—

(a)

makes a taxable supply in the United Kingdom to a taxable person …

he shall provide such persons as are mentioned above with a VAT invoice [subject to a specified irrelevant exception].”

The time for provision of the invoice is dealt with in sub-paragraph (5):

“(5)

With the exception of the supplies referred to in paragraph (6), the documents specified in paragraphs (1), (2), (3) and (4) above shall be provided within 30 days of the time when the supply is treated as taking place under section 6 of the Act, or within such longer period as the Commissioners may allow in general or special directions.”

The closing words, giving the Commissioners power to vary, are relied on by Mr Herberg.

Regulation 25 provides for three-monthly or quarterly returns, to be made within one month after the end of each three months or quarter. That period may be varied by the Commissioners. Regulation 29 is an important provision providing for claims for input tax and what is required in that respect:

“29(1) [Subject to paragraph (1A) below] , and save as the Commissioners may otherwise allow or direct either generally or specially, a person claiming deduction of input tax under section 25(2) of the Act shall do so on a return made by him for the prescribed accounting period in which the VAT became chargeable save that, where he does not at that time hold the document or invoice required by paragraph (2) below, he shall make his claim on the return for the first prescribed accounting period in which he holds that document or invoice].”

Note the reference to invoices. Subsection (2) makes the need for the provision of an invoice even clearer, but with an important proviso:

“(2)

At the time of claiming deduction of input tax in accordance with paragraph (1) above, a person shall, if the claim is in respect of—

(a)

a supply from another taxable person, hold the document which is required to be provided under regulation 13; ….

provided that where the Commissioners so direct, either generally or in relation to particular cases or classes of cases, a claimant shall hold, [or provide], such other [...] evidence of the charge to VAT as the Commissioners may direct.”

That proviso embodies the discretion, which the Commissioners have, to accept evidence other than invoices when it comes to considering claims to deduct, or be repaid, input tax. The existence of this discretion (so far as it does exist in a practical sense, which is disputed by Mr Cordara) is very much relied on by Mr Herberg as a fair and practical alternative to the presentation of an invoice where the latter is not possible.

That is a sufficient statutory background to enable the proper exposition of the case for the time being. It will be necessary to bring in other statutory provisions in relation to some of the arguments as they are dealt with, but it will be more convenient to set out them out in their context.

This issue is separate from the question of whether Royal Mail has a direct liability under the Directives as an emanation of the state in a Francovich claim. The unstated premise of this issue (not accepted by the claimants) is that there is no such direct claim, or it is inadequate in its effect, so they wish to rely on breach of an English statutory duty. This question is likely to be relevant to all business-to-business transactions where the parties are registered for VAT, whether involving Royal Mail or not. It is therefore a significant question for the VAT regime.

The parties were agreed that there was a statutory duty to provide invoices, and agreed on most of the case law on whether or not a statutory duty gave a private law right of action to individuals or corporations, but they were not agreed on whether there was a private law duty to individuals in this case. For these purposes I make no assumption as to whether, in Royal Mail’s case, there is or is not a duty based on the Directives and

This debate also took place on the footing that the (mis)understanding of the parties up to 2009 did not have any effect on the enforceability of the duty. It was not, for example, argued that on the facts of any particular case Royal Mail was exonerated from performing any statutory duty towards its customers by reason of the actual arrangements entered into. The debate was as to whether a duty arose under the statutory regime, without going further.

The relevant legal principles emerge from the following cases.

In X (Minors) v Bedfordshire CC [1995] 2 AC 633 at 731C-G Lord Browne-Wilkinson set out the requirements for the constitution of what he called a “breach of duty simpliciter”, which would describe the alleged breach of duty in this case if there is one. He said:

“(A)

Breach of statutory duty simpliciter

This category comprises those cases where the statement of claim alleges simply (a) the statutory duty, (b) a breach of that duty, causing (c) damage to the plaintiff. The cause of action depends neither on proof of any breach of the plaintiffs' common law rights nor on any allegation of carelessness by the defendant.

The principles applicable in determining whether such statutory cause of action exists are now well established, although the application of those principles in any particular case remains difficult. The basic proposition is that in the ordinary case a breach of statutory duty does not, by itself, give rise to any private law cause of action. However a private law cause of action will arise if it can be shown, as a matter of construction of the statute, that the statutory duty was imposed for the protection of a limited class of the public and that Parliament intended to confer on members of that class a private right of action for breach of the duty. There is no general rule by reference to which it can be decided whether a statute does create such a right of action but there are a number of indicators. If the statute provides no other remedy for its breach and the Parliamentary intention to protect a limited class is shown, that indicates that there may be a private right of action since otherwise there is no method of securing the protection the statute was intended to confer. If the statute does provide some other means of enforcing the duty that will normally indicate that the statutory right was intended to be enforceable by those means and not by private right of action: Cutler v. Wandsworth Stadium Ltd. [1949] A.C. 398 ; Lonrho Ltd. v. Shell Petroleum Co. Ltd. (No. 2) [1982] A.C. 173. However, the mere existence of some other statutory remedy is not necessarily decisive. It is still possible to show that on the true construction of the statute the protected class was intended by Parliament to have a private remedy. Thus the specific duties imposed on employers in relation to factory premises are enforceable by an action for damages, notwithstanding the imposition by the statutes of criminal penalties for any breach: see Groves v. Wimborne (Lord) [1898] 2 Q.B. 402.

Although the question is one of statutory construction and therefore each case turns on the provisions in the relevant statute, it is significant that your Lordships were not referred to any case where it had been held that statutory provisions establishing a regulatory system or a scheme of social welfare for the benefit of the public at large had been held to give rise to a private right of action for damages for breach of statutory duty. Although regulatory or welfare legislation affecting a particular area of activity does in fact provide protection to those individuals particularly affected by that activity, the legislation is not to be treated as being passed for the benefit of those individuals but for the benefit of society in general. Thus legislation regulating the conduct of betting or prisons did not give rise to a statutory right of action vested in those adversely affected by the breach of the statutory provisions, i.e. bookmakers and prisoners: see Cutler's case [1949] A.C. 398; Reg. v. Deputy Governor of Parkhurst Prison, Ex parte Hague [1992] 1 A.C. 58. The cases where a private right of action for breach of statutory duty have been held to arise are all cases in which the statutory duty has been very limited and specific as opposed to general administrative functions imposed on public bodies and involving the exercise of administrative discretions.”

Thus the question of whether there is a private law cause of action depends on ascertaining the intention of Parliament in that respect, which in turn involves construing the statute. Much of the argument before me turned on the non-conclusive indicia propounded by Lord Browne-Wilkinson - whether the legislation contained alternative remedies and, if so, how effective they were. Other authorities were cited to me but since they contain either insignificantly different reformulations of Lord Browne-Wilkinson’s guidance, or examples of how his indicia have and have not resulted in a statutory duty, enforceable in a civil claim by an individual, in other cases, I do not think it necessary to deal with them save for one.

In Reg. v Deputy Governor of Parkhurst Prison, Ex parte Hague [1992] 1 A.C. 58, cited by Lord Browne-Wilkinson, Lord Jauncey rejected the proposition that the mere fact that a statutory provision was intended to benefit individuals meant that a private law duty arose. He said (at p170-171):

“My Lords, I take from these authorities that it must always be a matter for consideration whether the legislature intended that private law rights of action should be conferred upon individuals in respect of breaches of the relevant statutory provision. The fact that a particular provision was intended to protect certain individuals is not of itself sufficient to confer private law rights of action upon them, something more is required to show that the legislature intended such conferment.”

The statutory duty said to exist in this case flows from Regulation 13 of the 1995 Regulations (and its predecessors so far as relevant). That provides that in a transaction such as those which took place in this case, the supplier:

“shall provide such persons as are mentioned above with a VAT invoice”.

I have emphasised the word which was emphasised by Mr Cordara. The time by which it is to be supplied is provided by Regulation 13(5):

“within 30 days of the time when the supply is treated as taking place under section 6 of the Act”.

Mr Cordara relies on this apparently straightforward expression of the duty, and puts it in its commercial context in order to demonstrate its commercial significance in normal trading. He relied heavily on the fact that, as he puts it, a VAT invoice is required to enable a trader to recover input tax. Since this is treated as a VATable transaction (see Assumptions 1 and 3) the requirements for an invoice are met - this was a supply of normally VATable services by a registered trader to a registered trader. Regulation 29 provides that a trader “shall” claim the input tax in a return (if he wished to claim it - he does not have to), for which purpose he needs (“shall hold”) the invoice required by Regulation 13. The discretion of the Commissioners to allow alternative proof under Regulation 29(2) is not a practical alternative because, Mr Cordara submits, Zipvit in effect closes it off. The ability to claim input tax is integral to the proper operation of the VAT system, without which neutrality is not achieved and the system simply does not work as intended. It is very important that the system be administered properly, and to that end it is right and necessary that traders themselves should have the right to enforce the rendering of a VAT invoice to enable input tax to be reclaimed. Authority tells us that the purpose of an invoice, or part of its purpose, is to evidence the payment of input tax so that the tax can be claimed. The non-provision of an invoice, and the consequential non-recoverability of input tax, distorts the system and makes it operate in a way which is contrary to principle. Mr Cordara accepted that there are enforcement mechanisms in the legislation (to which I will come) but said that they are inadequate to fulfil the legitimate requirements of a customer and do not point away from a directly owed duty.

Mr Cordara supported his argument by a higher level factor. He pointed to the European legislation which required Member States to set up a system which operated in accordance with the high level principles required by the Directives.

All those matters, taken together, are said by Mr Cordara to demonstrate that the duty to provide an invoice must be taken to have been intended by Parliament to be enforceable by customers against suppliers. On analysis his case seems to turn heavily on the two central factors - first, that since Zipvit in the Court of Appeal his clients (being typical customers for these purposes) have no way of claiming input tax in the absence of an invoice; and second, that any other enforcement methods under the legislation are inadequate. That means that he does not have an adequate remedy which enables a customer to deduct input tax, unless the statutory duty can be enforced directly by the customer. It should be noted that, as I have observed, his case involves a general proposition which would apply to all suppliers in VAT transactions. There is nothing special in the assumed facts of this case which require the duty in this case but not in others.

For his part Mr Herberg accepted the test as set out in the X (Minors) case but said that it did not result in a private right of action. He pointed to the fact that even where the statute provides no express remedy (which is not this case) that still does not necessarily result in a private right of action - see R v Deputy Governor of Parkhurst Prison Ex p Hague [1992] 1 AC 58. A close study of the regulations demonstrates that HMRC has various discretions to vary the content and the timing of invoices, and such discretions are said to be a contra-indication of a private cause of action; reliance was placed on Todd v Adams and Cope [2002] 2 Lloyds Rep 293. This, he submitted, is not surprising since he said that the primary purpose of the invoice was to allow HMRC to supervise the regime. Furthermore, the European legislation and the implementing UK legislation provided for an alternative route for recovery via the discretion given to the Revenue to accept alternative evidence in appropriate cases. Zipvit did not block that off. The absence of any reference to a private law right was significant too.

Mr Cordara’s case centred very much on what he said was the centrality to the VAT system of two things - a right to deduct input tax and the need for a VAT invoice, and

According to Mr Cordara, the relevant part of the legislative story goes right back to section 2 of the European Communities Act 1972:

“2(1) All such rights, powers, liabilities, obligations and restrictions from time to time created or arising by or under the Treaties, and all such remedies and procedures from time to time provided for by or under the Treaties, as in accordance with the Treaties are without further enactment to be given legal effect or used in the United Kingdom shall be recognised and available in law, and be enforced, allowed and followed accordingly; and the expression “enforceable EU right” and similar expressions shall be read as referring to one to which this subsection applies.”

It is said that, based on that provision, I should lean towards decisions which make for a “lawful” outcome. The structure of VAT goes back to the First Council Directive of 1967 (67/227/EEC) whose recitals refer to the principles of simplicity and neutrality which are intended to be reflected and in which Article 2 says:

“The principle of the common system of value added tax involves the application to goods and services of a general tax on consumption exactly proportional to the price of the goods and services, whatever the number of transactions which take place in the production and distribution process before the stage at which tax is charged.

On each transaction, value added tax, calculated on the price of the goods or services at the rate applicable to such goods or services, shall be chargeable after deduction of the amount of value added tax borne directly by the various cost components.”

The effect of that is that it is the ultimate consumer who is intended to bear the tax.

The first VAT Directive with an effect which is significant for the purposes of these actions is the Sixth Directive (33/388/EEC) which was in force until the 2006 VAT Directive (2006/112/EEC) (“the Principal VAT Directive”) superseded it. Much reliance was placed on its provisions.

Recitals which I do not need to set out refer to the need for harmonisation and the recitals go on to say:

“Whereas the obligations of taxpayers must be harmonized as far as possible so as to ensure the necessary safeguards for the collection of taxes in a uniform manner in all the Member States; whereas taxpayers should, in particular, make a periodic aggregate return of their transactions, relating to both inputs and outputs where this appears necessary for establishing and monitoring the basis of assessment of own resources”.

Mr Cordara points out the need for returns to monitor the basis of assessment, in relation to both inputs and outputs. Article 2 contains the basic charge on a supply of goods or services (this case is concerned with services) and Article 10 deals with chargeable events giving rise to the charge together with the significance of invoices:

“Article 10

1.

(a) "Chargeable event" shall mean the occurrence by virtue of which the legal conditions necessary for tax to become chargeable are fulfilled.

(b)

The tax becomes "chargeable" when the tax authority becomes entitled under the law at a given moment to claim the tax from the person liable to pay, notwithstanding that the time of payment may be deferred.

2.

The chargeable event shall occur and the tax shall become chargeable when the goods are delivered or the services are performed …

By way of derogation from the above provisions, Member States may provide that the tax shall become chargeable, for certain transactions or for certain categories of taxable person, either:

-

no later than the issue of the invoice or of the document serving as invoice, or

-

no later than receipt of the price, or

-

where an invoice or document serving as invoice is not issued, or is issued late, within a specified period from the date of the chargeable event.”

Thus invoices are not required by the system to be the chargeable event, but Member States may make them so.

Article 11 describes the taxable amount:

“Article 11

A. Within the territory of the country

1.

The taxable amount shall be: (a) in respect of supplies of goods and services other than those referred to in (b), (c) and (d) below, everything which constitutes the consideration which has been or is to be obtained by the supplier from the purchaser, the customer or a third party for such supplies including subsidies directly linked to the price of such supplies …

2.

The taxable amount shall include: (a) taxes, duties, levies and charges, excluding the value added tax itself;…”

I have already set out Article 13 which provides for exemption for postal services. Article 17 is an important Article dealing with the right to deduct:

“Article 17

Origin and scope of the right to deduct

1.

The right to deduct shall arise at the time when the deductible tax becomes chargeable.

2.

In so far as the goods and services are used for the purposes of his taxable transactions, the taxable person shall be entitled to deduct from the tax which he is liable to pay: (a) value added tax due or paid in respect of goods or services supplied or to be supplied to him by another taxable person …”

Article 18 contains “Rules governing the exercise of the right to deduct” which are very important to this case because they require an invoice to be available before a taxpayer can deduct input tax from output tax or get a refund of input tax:

“1.

To exercise his right to deduct, the taxable person must: (a) in respect of deductions under Article 17 (2) (a), hold an invoice, drawn up in accordance with Article 22 (3);”

The occasion on which the deduction occurs is when the taxpayer holds an invoice:

“2.

The taxable person shall effect the deduction by subtracting from the total amount of value added tax due for a given tax period the total amount of the tax in respect of which, during the same period, the right to deduct has arisen and can be exercised under the provisions of paragraph 1.”

The “provisions of paragraph 1” are the provisions requiring the holding of an invoice; hence the invoice’s importance to the claims of the claimants. However, Member States are given a discretion to allow claims to input tax if proved in a different way:

“3.

Member States shall determine the conditions and procedures whereby a taxable person may be authorised to make a deduction which he has not made in accordance with the provisions of paragraphs 1 and 2.”

This is the source of the discretion which the taxpayer in Zipvit is litigating about. Any repayment resulting from an excess of input tax over output tax is provided for by paragraph 4:

“4.

Where for a given tax period the amount of authorised deductions exceeds the amount of tax due, the Member States may either make a refund or carry the excess forward to the following period according to conditions which they shall determine.”

Invoices are to be required under Article 22:

“3.

(a) Every taxable person shall issue an invoice, or other document serving as invoice in respect of all goods and services supplied by him to another taxable person, and shall keep a copy thereof.

Every taxable person shall likewise issue an invoice in respect of payments on account made to him by another taxable person before the supply of goods or services is effected or completed.

(b)

The invoice shall state clearly the price exclusive of tax and the corresponding tax at each rate as well as any exemptions. (c) The Member States shall determine the criteria for considering whether a document serves as an invoice.”

Paragraph 4 requires the submission of returns no later than 2 months after the end of a tax period, and tax periods are to be fixed by Member States.

The exposition of those provisions is sufficient to demonstrate what Mr Cordara says is the overall structure of the accounting for VAT, and in particular the need for an invoice. It sets the tone for what follows.

As has appeared, this Directive was in force until the Principal VAT Directive came into force in 2006. So far as the mechanisms provided for by the Sixth Directive as set out above are concerned, there was no material change of substance in that later Directive, so it is unnecessary to set out all the equivalent provisions in the Principal VAT Directive. It suffices to set out the following.

Article 167 provides for the right to deduct input tax, splitting out the earlier provisions into separate Articles:

“Article 167

A right of deduction shall arise at the time the deductible tax becomes chargeable.

Article 168

In so far as the goods and services are used for the purposes of the taxed transactions of a taxable person, the taxable person shall be entitled, in the Member State in which he carries out these transactions, to deduct the following from the VAT which he is liable to pay:

(a)

the VAT due or paid in that Member State in respect of supplies to him of goods or services, carried out or to be carried out by another taxable person; …”

Article 178 reproduces the need for an invoice where a deduction is to be claimed: “Article 178

Article 179 reproduces the mechanism for input tax to be deducted from output tax and the discretion to have provisions which allow for method of proof alternative to an invoice is spread over Articles 180 to 182. Article 183 reproduces the deduction mechanism for input tax.

Article 220 provides again for an invoice to be provided:

“Article 220

Every taxable person shall ensure that, in respect of the following, an invoice is issued, either by himself or by his customer or, in his name and on his behalf, by a third party:

(1)

supplies of goods or services which he has made to another taxable person or to a non-taxable legal person; …”

The “Content of Invoices” is provided for in Section 4 (Articles 226ff), but Member States are permitted to dispense with some of the specified content by Article 221(1) in respect of invoices issued in their respective territories.

Article 222 deals with time limits for invoices: “Member States may impose time limits on taxable persons for the issue of invoices when supplying goods or services in their territory.”

Those provisions provide two things for the purposes of this litigation. First, they are accepted by Mr Herberg, for the purposes of this trial, as giving rise to a direct effect against Royal Mail insofar as it is an emanation of the state (which again is accepted or assumed for the purposes of this stage of the proceedings, but not otherwise). That is the content of Assumption 2 (varied, as it was, at the hearing). Second, they are relied on by Mr Cordara as constantly emphasising the need for an invoice if one is going to

It will be apparent from those arguments that HMRC’s discretion to vary the timing and content of invoices, and its discretion to allow alternative proof of inputs, are central to the argument. It will be convenient to deal with the scope of those aspects here before turning to consider their effect in the argument.

The regulations giving HMRC discretion over the content and timing of invoices have not yet been set out. In versions of the Regulations (as referred to in Issue 1) prior to 1995 there is apparently a further wider discretion. Those provisions are as follows. I shall emphasise particular relevant words where it is helpful to do so.

Regulation 8 of the 1975 Regulations contains the obligation to provide an invoice:

“8(1) Save as otherwise provided in these Regulations, a registered taxable person making a taxable supply to a taxable person shall provide him with a tax invoice.”

That is in substance the equivalent of the wording in the 1995 Regulations set out above.

However, the intervening Regulations (1977, 1980 and 1985) contained some additional wording:

“8(1) Save as otherwise provided in these Regulations, or as the Commissioners may otherwise allow, a registered taxable person making a taxable supply to a taxable person shall provide him with a tax invoice.”

(The regulation is Regulation 12 in the 1985 Regulations.) Those italicised words disappeared from the 1995 Regulations. Mr Herberg did not seek to make much about the presence of those words where they occurred but I do not think that one can ignore them. No-one suggested that any statutory duty has varied over time, so one has to give them the same effect across the period. If one were considering just those earlier regulations with the emphasised wording in, it would be a probably more significant indication of intention than Mr Herberg’s other examples.

Those other examples relate to the timing and content of invoices. I can take the 1995 Regulations as typical so far as contents are concerned.

“14(1) Subject to paragraph (2) below and regulation 16... save as the Commissioners may otherwise allow, a registered person providing a VAT invoice in accordance with regulation 13 shall state thereon the following particulars —”

and there then follow extensive particulars as to date, amount, subject matter, and many other detailed matters. One can see from their nature that many of them will be of much more interest to HMRC than to the customer recipient. Mr Herberg points to the italicised words, which have counterparts in the earlier regulations.

Timing, which has an obvious significance for a recipient as well as for HMRC, is dealt with in Regulation 13(5), which I have already set out above but which I set out again here with an appropriate emphasis:

“(5)

With the exception of the supplies referred to in paragraph (6), the documents specified in paragraphs (1), (2), (3) and (4) above shall be provided within 30 days of the time when the supply is treated as taking place under section 6 of the Act, or within such longer period as the Commissioners may allow in general or special directions.”

Mr Herberg also sought to say that the statutory scheme contains its own penalties which are imposed by section 69 of VATA. Basically there is a daily penalty per failure, up to a maximum of 100 days:

“69.— Breaches of regulatory provisions. (1) If any person fails to comply with a regulatory requirement, that is to say, a requirement imposed under—

…

(d)

any regulations or rules made under this Act, other than rules made under paragraph 9 of Schedule 12;

…he shall be liable, subject to subsections (8) and (9) below and section 76(6), to a penalty equal to the prescribed rate multiplied by the number of days on which the failure continues (up to a maximum of 100) or, if it is greater, to a penalty of £50.

….

(3)

Subject to subsection (4) below, in relation to a failure to comply with any regulatory requirement, the prescribed rate shall be determined by reference to the number of occasions in the period of 2 years preceding the beginning of the failure in question on which the person concerned has previously failed to comply with that requirement and, subject to the following provisions of this section, the prescribed rate shall be—

(a)

if there has been no such previous occasion in that period, £5;

(b)

if there has been only one such occasion in that period, £10; and

(c)

in any other case, £15.

(4)

For the purposes of subsection (3) above—

…

(b)

a continuing failure to comply with any such requirement shall be regarded as one occasion of failure occurring on the date on which the failure began;

(c)

if the same omission gives rise to a failure to comply with more than one such requirement, it shall nevertheless be regarded as the occasion of only one failure;…”

The net effect of these provisions is that a failure to provide a single VAT invoice attracts a maximum penalty of between £500 and £1500, depending on the offender’s previous record. There is, however, a “reasonable excuse” defence under subsection (8):

“(8)

A failure by any person to comply with any regulatory requirement or the requirement referred to in subsection (2) above shall not give rise to liability to a penalty under this section if the person concerned satisfies the Commissioners or, on appeal, a tribunal that there is a reasonable excuse for the failure; and a failure in respect of which the Commissioners or tribunal have been so satisfied shall be disregarded for the purposes of subsection (3) above.”

Mr Herberg says that, collectively, these provisions provide a considerable discretion to vary in relation to invoices which points to there being no separate private remedy, and he relies on Todd v Adams [2002] EWCA Civ 509 especially at para 25. In that case Neuberger J was considering whether certain safety regulations on ships were actionable by individuals, and he concluded they were not. One of the indicia he relied on was a power of the Secretary of State to exempt some vessels:

“25.

Thirdly, the fact that section 121(2) of the 1995 Act gives the Secretary of State a power (of a very wide and flexible nature as is clear from the next sub-section) to exempt vessels from the ambit of any rules, tends to support the proposition that there was no intention to impose a civil liability for breach of any of the rules. As Mr Nolan points out on behalf of the respondents, it would seem surprising if two identical vessels had accidents caused by the same defect, but that in one case there was no cause of action under the 1975 Rules, because the vessel had been exempted from the relevant part of the rules, and in the other case there was a virtually unanswerable cause of action, because it had not been so exempted. The Secretary of State might think it right, for instance, to relax a rule in relation to fishing vessels based in harbours in a specific area, for economic or (in the wider sense of the word) political reasons, and it would be a little surprising if the civil liability of the owner or master could depend on that sort of factor.”

Those matters go to a discretion to vary the scope of the obligation. The other significant discretion relied on by Mr Herberg is the alternative route available to a customer without an invoice, namely the discretion to allow alternative evidence. Mr Cordara says that this discretion is not in practice available, and is not available in the present case, because Zipvit requires an applicant to prove that its supplier has paid the relevant VAT output tax, which he cannot do.

Whether Zipvit goes that far was the subject of serious dispute in this case, and Mr Cordara frequently stressed that that is what Zipvit holds. Whether or not he is right depends on an analysis of the judgment of Henderson LJ in that case. It may seem presumptuous for me to have to decide the point when an appeal in the case is pending in the Supreme Court, but no-one suggested that I should wait for the Supreme Court decision (and a possible reference to the CJEU), and it does not seem right to do so, so I have to decide it.

As appears above, the Zipvit action arises (in part) out of a challenge by Zipvit to the decision of HMRC not to allow it to claim input tax in the face of evidence which comprised an invoice which did not refer to, or purport to claim, VAT and some supplementary evidence that Zipvit also provided. It seems that part of the argument turned on the Revenue’s discretion to allow alternative evidence. Zipvit failed in the Court of Appeal (and below). The parts of Henderson LJ’s judgment on which Mr Cordara most heavily relied follow from those parts of the judgment in which the judge identifies the significance of invoices as evidencing the payment of output tax by the supplier. It is necessary to quote extensively from what follows:

“112 … One of the main purposes of the mandatory requirement for a VAT invoice is to enable the taxing authorities to monitor payment by the supplier of the tax for which a deduction is sought, or as the Advocate General put it at point 32 of her opinion “to enable a check on whether the person issuing the invoice has paid the tax”. Zipvit remains wholly unable to satisfy this condition, because the only invoices which it can supply show the complete opposite, namely that no tax was paid because the supplies were considered to be exempt. Nor can it be said that the position was remedied by the exiguous further information supplied with the letter of claim in September 2009. All this did was to show the VAT component of the original purchase prices, on the assumption that the supplies were taxable. It provided no evidence that a penny of that tax had been paid by Royal Mail to HMRC, and still less did it do so in the form of an invoice issued by Royal Mail.

113.

Mr Thomas argues that none of this matters, because Zipvit was entitled to exercise its right to deduct input tax referable to the supplies which it made to its own customers, on which it accounted for output tax in the usual way. To deny a deduction on the sole basis that Royal Mail cannot be shown to have paid tax on the relevant supplies which it made to Zipvit is, he submits, to rely on a wholly irrelevant consideration, because it would offend the well-established principle that the right of deduction is unaffected by the question whether VAT due at an earlier stage in the chain of supply has been paid to the public purse. In my view, however, this objection misses the point. Exercise of the right to deduct is subject to a mandatory requirement to produce a VAT invoice, which must contain the specified particulars. Zipvit is unable to produce invoices which satisfy the requirements of article 226(9) and (10), and it is also unable to produce any supplementary evidence showing payment of the relevant tax by Royal Mail. A necessary precondition for exercise of the right to deduct therefore remains unsatisfied.

114.

I also fail to see how Zipvit could hope to circumvent this fundamental difficulty by arguing that the requirement for a compliant VAT invoice is one of form rather than substance, and by invoking the discretion which HMRC have to accept alternative evidence under regulation 29(2) of the 1995

Regulations. It is true that Barlis [2016] BVC 43 (at paras 40 and 41) and a number of other cases which we were shown, consistently draw a distinction between the substantive conditions which must be met in order for the right to deduct VAT to arise, and the formal conditions for the exercise of that right. But to describe a requirement as “formal” does not necessarily imply that compliance with it is optional, or that a failure to satisfy it is always capable of being excused. Cases like Barlis show that some of the requirements relating to invoices in article 226 must be dispensed with, if the tax authorities are supplied with the information necessary to establish that the substantive requirements of the right to deduct are satisfied. But the court was careful in Barlis to confine its discussion to the requirements in article 226(6) and (7), and I do not think its reasoning can be extended to cover a failure to comply with the fundamental requirements relating to payment of the relevant tax in article 226(9) and (10). Provision of an invoice which complies with those requirements is essential to the proper performance by HMRC of their monitoring functions in relation to VAT, and is needed as evidence that the supplier has duly paid or accounted for the tax to HMRC.

115.

It needs to be remembered in this context that the amounts for which Zipvit is claiming a deduction have not been paid by Zipvit in response to a request by Royal Mail for payment once the taxable status of the supplies had been established. In that situation, Royal Mail would have rendered an invoice showing the VAT due, and would then have been liable to account for it to HMRC as output tax in the usual way. In those circumstances, there would have been no difficulty about Zipvit deducting the amount shown on the new invoice as input tax. All that has actually happened, however, is that Zipvit now wishes to treat the payments which it originally made to Royal Mail, on the common understanding that the supplies were exempt, as comprising an element of VAT, and to obtain a deduction for that element on the strength of nothing more than the original payment.

116.

Even if it is open to Zipvit to recharacterise the original payment in this way (which at this stage of the argument must be assumed in Zipvit's favour), there would be an obvious detriment to HMRC and the public purse if Zipvit were able to obtain such a deduction without first showing that the tax in question had been paid by Royal Mail. The normal way of fulfilling that obligation is by production of a fully compliant

VAT invoice. Since Zipvit is unable to produce such an invoice, I am unable to see any grounds upon which HMRC could properly conclude that Zipvit should nevertheless be allowed the deductions claimed, to the detriment of the general body of taxpayers. In effect, a retrospective recharacterisation of sums originally paid on the footing that the supplies in question were exempt would now yield an uncovenanted bonus to Zipvit, generated by nothing more than Zipvit's unilateral decision to treat the amounts originally paid as VAT-inclusive. It would, I think, be offensive to most people's sense of fiscal justice if a mechanical accounting exercise of this nature were permitted to generate a very substantial input tax credit, in circumstances where (for whatever reason) none of the tax in question has been paid by the supplier.

117.

Whether the situation is described as one in which HMRC have no discretion, because the requirements of article 226(9) and (10) cannot be dispensed with, or as one where there is in law a discretion but on the facts of the present case it can only be exercised in one way, does not seem to me to matter. The important point is that the inability of Zipvit to produce a compliant VAT invoice in support of its claim to deduct input tax is in my judgment fatal. This was rightly recognised by the two tribunals below, although I would (with respect) not adopt their analysis of the position in terms of the absence of an “economic burden” on Zipvit. That way of looking at the matter seems to me misconceived, because Zipvit did bear the economic burden of paying the original purchase price for the supplies. The real issue, as I see it, is whether Zipvit can claim a deduction for VAT by treating the original price as VAT-inclusive, without producing evidence that the tax in question has been duly paid by the supplier.”

I have emphasised the words relied on by Mr Cordara. He submitted that they demonstrated that a taxpayer could not invite the exercise of the discretion to provide alternative evidence under Regulation 29(2) unless it could show that the tax had been paid by the taxpayer. That was not going to be possible in most cases where the taxpayer has not been given a tax invoice, so the discretion was not an effective alternative remedy to be taken into account in considering whether there was a private remedy to compel an invoice.

I do not consider that Henderson LJ was going that far. He was delivering his judgment in a striking case in which it was actually apparent that the output tax had not been paid. That was highly likely to be the case, if not inevitable, where both parties had treated the supplies as exempt throughout, and the supplier did not change its mind; and it is his express finding in paragraph 116 of his judgment. His judgment focuses mainly on the invoice as evidence of payment and accounting, and the effect of the absence of that evidence. He acknowledged the offence to fiscal justice if the situation were to generate

I believe that that view is supported by the normal way in which VAT operates in practice. Generally, a customer will not have any idea whether its supplier has duly accounted for and paid the output tax on the transaction or not. He generally does not need to know, and as was pointed out in Zipvit an ability to deduct input tax is not generally independent of proper accounting for output tax. Furthermore, in practice a customer is almost never going to be able to demonstrate that output tax was properly accounted for, because it will not have access to the relevant records, and even it did it may be impossible to answer the question if for any reason the supplier has not accounted for all output tax due to the transaction in question. If some output tax remains due it will not be possible to say in respect of which transactions it was not paid. If a customer seeking to invoke HMRC’s discretion in the absence of an invoice had to demonstrate payment then the discretion would never in practice be exercisable. That cannot, in my view, be the position. It would place improper obstacles in the way of the exercise of a discretion which is mandated by the Directives and be contrary to the sort of principles set out in Vadan v Agentia Nacionala de Administrare Fiscala (2018) (Case No. C-664/16):

“41.

The Court has held that the fundamental principle of the neutrality of VAT requires deduction of input VAT to be allowed if the substantive requirements are satisfied, even if the taxable persons have failed to comply with some formal conditions. It follows that the tax authorities cannot refuse the right to deduct VAT on the sole ground that an invoice does not satisfy the conditions required by Article 226(6) and (7) of the VAT Directive if they have available all the information to ascertain whether the substantive conditions for that right are satisfied…

42.

Thus, the strict application of the substantive requirements to produce invoices would conflict with the principles of neutrality and proportionality, in as much as it would disproportionately prevent the taxable person from benefiting from fiscal neutrality relating to his transactions.”

The point at issue in that case was not the same, but the observations demonstrate that the discretion has to be a real one and not one surrounded by obstacles which are in practice impenetrable, which is what would be the case if Mr Cordara were right in his interpretation of Zipvit. Authority demonstrates that the Revenue does not erect impenetrable barriers, though there certainly are serious barriers - see eg Everycar Contracts Ltd v HMRC [2013] UKFTT 405 (TC) at para 76.

There may not be many circumstances in which the discretion is needed, but one can imagine a circumstance in which a customer can demonstrate payment but has, without fault, lost the invoice and cannot get a duplicate from a supplier (perhaps because the supplier has gone into chaotic insolvency and feels it cannot properly issue a duplicate). It would seem to be strange that in such a situation the customer should not be able to seek to prove payment of input VAT in a different way, without an invoice, even though it would be impossible for it to prove the payment of output tax. If for some reason that example does not work, one could imagine others.

I therefore reject the submission that there is in effect no alternative to producing an invoice. There is an alternative in the form of the discretion. That means, as Mr Cordara accepted, that a material plank in his argument has been taken away. He can no longer argue that a statutory duty is required because there is no real alternative. There is an alternative, and it can take its place in what Mr Herberg says is a workable scheme embodying discretions, which points away from there being a statutory duty to provide an invoice.

Accordingly, Mr Herberg has a reasonable armoury in support of his submission that there is no private law remedy based on statutory duty which requires the production of a VAT invoice. In summary he can point to:

Mr Cordara is nonetheless left with some powerful factors. The whole system depends very heavily on the invoice as evidence of payment of tax, for the benefit of both the supplier and the customer. This feature is clearly set out by Henderson LJ where he summarises the authorities. While some authorities emphasise the importance of the invoice as between tax authority and supplier, there is nonetheless a clear recognition of the important interest of the customer as well. At paragraph 112 he adopts a summary of the position from Barlis 06 - Investimentos Imobiliários e Turisticos SA v Autoridade Tributaria e Aduaneira (2016)(Case C-516/14):

“As the Advocate General observes in points 30, 32 and 46 of her Opinion, the objective of the details which must be shown in an invoice is to allow the tax authorities to monitor payment of the tax due and, if appropriate, the existence of the right to deduct

VAT.”

Mr Cordara points out that HMRC could not have been anticipated to get involved in everyday trader disputes as between traders involving the non-rendering of a VAT invoice, and that points to the fact that traders themselves should have the right to compel the rendering of a VAT invoice. A breakdown of the system on the scale that happened in the case of the Royal Mail (at least up to 2009) is contrary to the tax neutrality which the system is supposed to give rise to. The right claimed is simple and straightforward enough, and the necessity and importance of an invoice is recognised by the Revenue itself in one of its VAT notices (VAT Notice 700/21):

“4.2.

The VAT invoices you issue form a very important part of your business records and you must keep a copy of every VAT invoice you issue.

Similarly the VAT invoices you receive are the primary evidence for you to recover VAT you have incurred as input tax and you should make sure you keep them in a way that allows you to find them easily when asked. VAT invoices are crucial to your business customers because the VAT invoice is the primary evidence that’ll allow them to recover the VAT you’ve charged.”

Some of Mr Cordara’s points have merit, but there are answers to them. He may be right in saying that it was not anticipated that the Revenue would get involved in the day to day disputes he referred to, but that is as likely to be because one would generally expect a supplier to supply an invoice – see below. If something has gone wrong on some sort of allegedly grand scale in the present cases then that is because of the highly unusual circumstances which gave rise to them. Nor do I think that any extra weight is added to Mr Cordara’s case by this manual. He can make his point without it. For his

There is no suggestion there that the customer should be pointed in the direction of a private law action; in fact the absence of such an action is implicit in that extract. So far as this is an implicit expression of the Revenue’s view, it is not particularly relevant, because the Revenue may be right and may be wrong. The issue is a legal one for the courts. But it is of some relevance that the Revenue’s view does not seem to be that the system is not properly workable without a civil law remedy being open to the customer.

It was in the context of the debate about discretion that the parties made submissions as to whether success for the claimants would give rise to a windfall at the expense of the taxpayer, of the nature referred to by Henderson LJ in Zipvit. The windfall would arise if the invoices enabled recovery of input tax from HMRC without HMRC having a right to recover the corresponding output tax from Royal Mail. The parties debated the effect of the statutory provisions relating to the recovery of output tax, and in particular whether there were any and if so what time limits. For my part I do not see the relevance

Although this is not an easy question, I do not consider that the case in favour of a statutory duty to provide a VAT invoice, enforceable by a customer as against a supplier, has been made out. I reach this conclusion on two bases.

The first is that there are too many contra-indications of the type referred to in the authorities. There is an alternative method of obtaining input tax in the form of the discretion. Mr Cordara’s own submissions acknowledged the real significance of this point. The obligations in relation to the form and timing of the invoice are too variable at the behest of the Revenue; thus the content of the duty would be variable at the behest of a third party. And there is an enforcement mechanism in the form of the daily penalties. The lack of strength of that as a method of enforcement has to be acknowledged, but Parliament has put it there and it is the existence of some power of enforcement which is more significant than what one might think about its effectiveness. Those factors are, by themselves, sufficient in my view to indicate that Parliament did not intend an individual to have the benefit of a directly enforceable right against a supplier.

The second, which really reinforces the conclusion derived from the first, is the way that commerce would be perceived to work in practice, which in my view is something that Parliament is likely to have had in mind. In practice invoices are rendered as a matter of course. Suppliers generally expect to render them in order to get payment, and customers generally expect to receive them so that they know that payment is now expected of them. Customers would be likely to refuse to pay until they had had a VAT invoice if they needed one, and a supplier would generally understand that and supply one, if only because he wanted to be paid. If a customer was particularly concerned about its position it could contract on the basis that it would only pay against a VAT invoice. In other words, in practice, so far as a customer is concerned, a problem would not be anticipated to arise in practice, or at least not to an extent to make it appropriate (or necessary) to create a statutory duty enforceable at the behest of the customer.

I have not ignored one authority which might be seen to assume the existence of the statutory duty relied on by Mr Cordara. In Europhone International Ltd v Frontel Communications Ltd [2001] STC 1399 Ferris J had to consider whether there was an express or implied duty to render VAT invoices on the facts of that case. He concluded that there was not. I return to that case in the context of considering the contractual duties said to be owed in the present cases, but for present purposes it is relevant to note that paragraph 35 of Ferris J’s judgment might be thought to reflect an assumption on his part that Regulation 13 imposed a statutory duty enforceable by the customer:

“35.

Even if Frontier did not do this voluntarily Europhone could, by making payment on account of the services supplied, bring into existence an obligation to issue VAT invoices in order to comply with reg 13.”

If that is a correct view of the paragraph then I do not consider that it binds me to reach the same conclusion. It is not clear that the point was actually argued (or even in dispute) in Europhone and, having heard full argument, I prefer the contrary view.

I therefore conclude that VATA and the statutory regulations do not create a general statutory duty to provide a VAT invoice enforceable at the behest of an individual customer. That is the general position in law and there is no case for an argument that the facts of this (or any) particular case give rise to such a duty on an individual basis. The answer to Issue 1 is therefore: No, there is and was no such actionable statutory duty.

Issue 6 raises the question of whether Royal Mail was under any contractual duty to provide invoices. This was tested by reference to various sample contracts which, I assume, represent a useful cross-section of the relevant contracts. The contracts related to various delivery services and to franking. Franking was covered by statutory schemes rather than by contractual terms. There was no particular discussion as to whether those schemes led to contractual, rather than scheme, rights, and they were by and large (though not entirely) dealt with as though they were contracts. I do not consider that assumption to be correct.

The relevant terms of each contract (or scheme) are set out in Appendix 1 to this judgment, under the heading or descriptions which were given to them at the time and/or which were given to them by the parties for the purposes of this action. Nothing turns on the details of the nature of the services in question, so I do not need to set them out. Some of the contracts are made up of several documents. Where that is the case the relevant terms of each document are set out under an appropriate heading. The Appendix contains all provisions referring to “invoices”, and other relevant terms.

The Assumptions which are principally relevant for these purposes are nos. 1, 3, 4 and 5. As I have already indicated, the parties did not prepare an agreed description for the purposes of Assumption 4. That Assumption leaves open the possibility of my making some findings of fact, or perhaps some further assumptions, in order to be able to construe the contracts or consider implied terms. In fact it is hardly necessary for me to do so. Where I have done it, it will appear below.

In addition there were some more extensive assumptions (the “Franking Assumptions”) which apply to the franking schemes. I have set those out in Appendix 2 to this judgment.

The law applicable to the processes of construction and implication with which I have to engage was by and large agreed. It is the law as set out in several recent Supreme Court decisions, and I can set it out as follows.

First, so far as construction is concerned, the principles appear in Rainy Sky SA v Kookmin Bank [2011] 1 WLR 2900. The contract must be construed as it would be understood by a reasonable person having the background knowledge that was available to the parties in the situation of the parties, and ascertain what that person would have understood a disputed term to mean (para 21). It is relevant to consider whether a particular construction flouts business sense and lean against that construction so far as the words and the context might allow, to an appropriate extent (though this principle does not really arise in this case).

So far as the implication of terms is concerned, Mr Cordara drew attention to the two principal classes of implied terms - those implied into a particular class of contracts, and those implied into a particular contract on the basis that the parties must be taken to have included them, principally on the grounds of business efficacy – Geys v Societe Generale v Geys [2013] 1 AC 523 at 55. He relied on business-to-business contracts where both parties were VAT registered as being a relevant class of contracts for the purposes of the first of those types of implied terms.

So far as the second of those types of implication is concerned it is not necessary to go beyond Marks & Spencer plc v BNP Paribas Securities Trust Co (Jersey) Ltd [2016] AC 742. Lord Neuberger summarised and commented on the law in the following terms:

“16.

There have, of course, been many judicial observations as to the nature of the requirements which have to be satisfied before a term can be implied into a detailed commercial contract. They include three classic statements, which have been frequently quoted in law books and judgments. In The Moorcock (1889) 14 PD 64, 68, Bowen LJ observed that in all the cases where a term had been implied, “it will be found that … the law is raising an implication from the presumed intention of the parties with the object of giving the transaction such efficacy as both parties must have intended that at all events it should have”. In Reigate v Union Manufacturing Co (Ramsbottom) Ltd [1918]

1 KB 592, 605, Scrutton LJ said that “A term can only be implied if it is necessary in the business sense to give efficacy to the contract”. He added that a term would only be implied if “it is such a term that it can confidently be said that if at the time the contract was being negotiated” the parties had been asked what would happen in a certain event, they would both have replied: “‘Of course, so and so will happen; we did not trouble to say that; it is too clear.’” And in Shirlaw v Southern Foundries (1926) Ltd

[1939] 2 KB 206, 227, MacKinnon LJ observed that, “Prima facie that which in any contract is left to be implied and need not be expressed is something so obvious that it goes without saying”. Reflecting what Scrutton LJ had said 20 years earlier, MacKinnon LJ also famously added that a term would only be implied “if, while the parties were making their bargain, an officious bystander were to suggest some express provision for it in their agreement, they would testily suppress him with a common ‘Oh, of course!’

17.

Support for the notion that a term will only be implied if it satisfies the test of business necessity is to be found in a number of observations made in the House of Lords. Notable examples included Lord Pearson (with whom Lord Guest and Lord Diplock agreed) in Trollope & Colls Ltd v North West Metropolitan Regional Hospital Board [1973] 1 WLR 601, 609, and Lord Wilberforce, Lord Cross of Chelsea, Lord Salmon and Lord Edmund-Davies in Liverpool City Council v Irwin [1977] AC 239, 254, 258, 262 and 266 respectively. More recently, the test of “necessary to give business efficacy” to the contract in issue was mentioned by Baroness Hale JSC in Geys v Société Générale [2013] 1 AC 523, para 55 and by Lord Carnwath JSC in Arnold v Britton [2015] AC 1619, para 112.

18.

In the Privy Council case BP Refinery (Westernport) Pty Ltd v Shire of Hastings (1977) 180 CLR 266, 283, Lord Simon of Glaisdale (speaking for the majority, which included Viscount Dilhorne and Lord Keith of Kinkel) said that:

“for a term to be implied, the following conditions (which may overlap) must be satisfied: (1) it must be reasonable and equitable; (2) it must be necessary to give business efficacy to the contract, so that no term will be implied if the contract is effective without it; (3) it must be so obvious that ‘it goes without saying’; (4) it must be capable of clear expression; (5) it must not contradict any express term of the contract.

…

21.

In my judgment, the judicial observations so far considered represent a clear, consistent and principled approach. It could be dangerous to reformulate the principles, but I would add six comments on the summary given by Lord Simon in the BP Refinery case 180 CLR 266, 283 as extended by Bingham MR in the Philips case [1995] EMLR 472 and exemplified in The

APJ Priti [1987] 2 Lloyd’s Rep 37. First, in Equitable Life Assurance Society v Hyman [2002] 1 AC 408, 459, Lord Steyn rightly observed that the implication of a term was “not critically dependent on proof of an actual intention of the parties” when negotiating the contract. If one approaches the question by reference to what the parties would have agreed, one is not strictly concerned with the hypothetical answer of the actual parties, but with that of notional reasonable people in the position of the parties at the time at which they were contracting. Secondly, a term should not be implied into a detailed commercial contract merely because it appears fair or merely because one considers that the parties would have agreed it if it had been suggested to them. Those are necessary but not sufficient grounds for including a term. However, and thirdly, it is questionable whether Lord Simon’s first requirement, reasonableness and equitableness, will usually, if ever, add anything: if a term satisfies the other requirements, it is hard to think that it would not be reasonable and equitable. Fourthly, as Lord Hoffmann I think suggested in Attorney General of Belize v Belize Telecom Ltd [2009] 1 WLR 1988, para 27, although Lord Simon’s requirements are otherwise cumulative, I would accept that business necessity and obviousness, his second and third requirements, can be alternatives in the sense that only one of them needs to be satisfied, although I suspect that in practice it would be a rare case where only one of those two requirements would be satisfied. Fifthly, if one approaches the issue by reference to the officious bystander, it is “vital to formulate the question to be posed by [him] with the utmost care”, to quote from Lewison, The Interpretation of Contracts 5th ed (2011), p 300, para 6.09. Sixthly, necessity for business efficacy involves a value judgment. It is rightly common ground on this appeal that the test is not one of “absolute necessity”, not least because the necessity is judged by reference to business efficacy. It may well be that a more helpful way of putting Lord Simon’s second requirement is, as suggested by Lord Sumption JSC in argument, that a term can only be implied if, without the term, the contract would lack commercial or practical coherence.”

I shall apply those principles of construction and implication in turn to each of the sets of contractual terms in question.

So far as the first type of implication is concerned, Mr Cordara drew attention to what Lady Hale said in Geys:

“56.

A great deal of the contractual relationship between employer and employee is governed by implied terms of the latter kind. Some are of long-standing, such as the employer’s duty to provide a safe system of work. Some are of more recent discovery, such as the mutual obligations of trust and confidence. This was referred to by Dyson LJ in Crossley v Faithful & Gould Holdings Ltd [2004] ICR 1615 as an “evolutionary process”. He also described the “necessity” involved in implying such terms as “somewhat protean”, pointing out that some well established terms could scarcely be said to be essential to the functioning of the relationship. At para 36, he said:

“It seems to me that, rather than focus on the elusive concept of necessity, it is better to recognise that, to some extent at least, the existence and scope of standardised implied terms raise questions of reasonableness, fairness and the balancing of competing policy considerations.”

There is much to be said for that approach, given the way in which those terms have developed over the years.”

As will appear, I do not think that this type of implication has a part to play in this case.

I have set out the background to the relevant contracts as it appears from the Assumptions. As I have also pointed out, Assumption 4 leaves it open for there to be some further background facts, and Mr Cordara suggested some more in the course of his submissions. They were not all accepted by Mr Herberg but most were. Based on that, and based on the latitude left open to me by Assumption 4, I proceed on the basis of the following facts or factors.

The most significant piece of factual background, in my view, is that contained within Assumption 5, together with the state of UK legislation up to 2009. The Assumption only operates until TNT1 in 2009, but until that time the parties and HMRC all thought that the relevant services were exempt. A common assumption of that nature would normally be expected to be relevant to an estoppel by convention claim, or perhaps rectification, rather than to construction, but in this case it is a relevant and necessary part of the background. It was a standard form of industry assumption against which business (and tax returns) were conducted, and all the more so because it reflected UK legislation. This point permeates the argument hereafter.

Against that background I shall consider construction points first in relation to each contract, and then turn to implication.

For the purposes of this section reference should be made to Appendix 1 for the terms of the various sample contracts. In this section I also deal with two franking schemes which were actually schemes constituted by statute and which did not necessarily have contractual effect (a point which I have to determine). Whatever the effect of those schemes, they are dealt with in this section for the sake of convenience.

Example 1 - Parcelforce

There is one reference to an invoice, and that is in the penultimate clause. However, there are no words of obligation in terms of rendering an invoice. The word “will” in relation to accounts is not capable of imposing an obligation on Royal Mail - it is an indication of what will happen, not what must happen, and there is no comparable word in relation to the invoice anyway. Since there is no obligation to render an invoice at all, there can be no obligation, as a matter of construction, to render a VAT invoice.

This conclusion is in line with the interesting decision of Ferris J in Europhone International v Frontel Communications [2001] STC 1399. In that case, like the present ones, a claimant sought to compel the production of a VAT invoice so that input tax could be claimed, even though it had no intention of accounting for arrears of output tax. A claim was made on the basis of express and implied terms. So far as the express term is concerned the contract provided:

“The customer shall make payment to Frontier in full within 20 calendar days of the date of each valid invoice which Frontier will provide at the end of each month to which it relates…"

Ferris J held that that wording did not give rise to any obligation to issue an invoice at the end of each month, though the mutual expectation of the parties was probably that such invoices would be issued (see the submissions of counsel recorded at paragraph 28, accepted by the judge at paragraph 31). Although each case of disputed contractual construction depends on its own facts and contract, I think that Europhone is sufficiently close to the present case in terms of the structure of the invoicing provision to allow parallels to be drawn.

It will be convenient to refer at this point to a factor which permeates the entire contractual debate and which was not present in Europhone. I have already referred above to the importance of Assumption 5. Bearing that assumption in mind, it would seem to me to be impossible to construe the word “invoice” in this document as meaning “VAT invoice”. It is true that pursuant to Assumption 3 the price is to be taken as including VAT, but that is a legal conclusion flowing from TNT1 (and other legal sources). It does not alter the underlying commonly held perception, enshrined in legislation, that the services in this instance (and in all the other instances in this action) were VAT exempt. It seems to me to be completely impossible to construe the simple word “invoice” as being taken to mean something which the parties (with their beliefs) can never have intended it to mean. That, too, is therefore a bar to there being a contractual duty to render a VAT invoice in the instance of this Example.

Example 2 - Tracked Mail

In this contract there is, of course a reference to VAT - see clause 6.6 of the main contract - “the charges … do not include VAT. You must pay any VAT due on the charges.” However, that does not create an obligation to bill for VAT, or an obligation to render a VAT invoice. It is pointing out what the obligation on the customer is in the event of VAT being due. It is an obligation on the customer, not on the supplier, and does not shift wording to suggest that an invoice can be compelled.

Again, this contract must be viewed against the background of Assumption 5, which makes it even harder to construe anything in the contract as containing an obligation to render a VAT invoice.

Example 3 - Mailsort

However, whatever may be the position in relation to the compulsion of invoices, there can be no argument that, as a matter of construction, the invoices should be VAT invoices which specify a sum for VAT so far as Mailsort is concerned. The Mailsort User Guide, which is part of the contract, is quite inimical to such a suggestion. It is a very long document (some 204 pages) and the only relevant part is 4 pages in from the

Example 4 - Franking Services - 1979 Scheme

The 1979 Scheme was a general scheme covering all postal services and was made under section 28 of the Post Office Act 1969. So far as material that section provided:

“28.— Schemes for determining charges and other terms and conditions applicable to services.

(1)

The Post Office may make, as respects any of the services provided by it, a scheme for determining either or both of the following, namely,—

(a)

the charges which (save in so far as they are the subject of an agreement between it and a person availing himself of those services) are to be made by it; and

(b)

the other terms and conditions which (save as aforesaid) are to be applicable to those services; but so that no provision be included in any such scheme for limiting liability of the Post Office for loss or damage or for amending the rules of law with respect to evidence.

(2)

A scheme made under this section may, as respects the services to which it relates, adopt such system for the determination of the charges or other terms and conditions or (as the case may be) the charges and other terms and conditions that are to be applicable as may appear desirable and, in particular and without prejudice to the generality of the foregoing words, may, in all or any cases, leave the determination thereof to the Post Office subject to such (if any) conditions and limitations as may be provided for in the scheme.

(3)

A scheme made under this section may, as respects the services to which it relates, specify the manner in which, time at which and person by whom the charges that are to be applicable are to be paid.

(4)

A scheme made under this section may make different provision for different cases or classes of cases determined by, or in accordance with, the provisions of the scheme.

(5)

A charge exigible by virtue of this section may be recovered by the Post Office in any court of competent jurisdiction as if it were a simple contract debt.

…

(8)

In the application of subsection (5) above to Scotland, the words “as if it were a simple contract debt” shall be omitted.”

Sections 29 and 30 provide statutory limitations of liability.

There is nothing in the Act (or nothing was drawn to my attention) which suggests that a scheme was to have contractual force. In fact, section 28(5) strongly suggests the contrary. If the Scheme were to have contractual effect it would be unnecessary for Parliament to have to provide that charges were recoverable “as … a simple contract

That view is strongly reinforced when one looks at the Scheme itself. I was shown the 1979 Scheme in the form published in the Belfast Gazette on 1st June 1979. In that form it extends over 33 pages of small print and it covers, in detail, all the postal services made available by the Post Office. There is a complete absence of any reference to any contractual effect. Paragraph 4 (set out in Appendix 1) shows its scope (there are insignificant exemptions) and reinforces the view that it takes effect as a statutory scheme, not as the basis for a consequential contract. The Scheme then goes on in detail to deal with charges, general conditions, delivery, returns, required packaging and many other details. There is nothing at all which suggests contractual effect. I find that it did not have that effect subject to the possibility of there being a separate transaction which might create a contract in an individual case.

That rather limits the scope for argument based on contractual interpretation, as I have indicated. It is not relevant to consider the intentions of the parties, because the Scheme is not a document which is said to embody that intention. One has to conduct a process more akin to statutory interpretation. Conducting that exercise, there is nothing in the Scheme which provides for any invoice, let alone a VAT invoice. In fact, there is no mention of the word in the entire Scheme. The most material provisions are set out in Appendix 1, including those relating to franking, which is the relevant service for present purposes. There is no wording there in relation to which a question of interpretation about VAT invoices (or any invoices) could conceivably arise.

If I am wrong about that analysis of the status of the Scheme, and it falls to be treated as a contract, the same conclusion applies. There are simply no words which are capable, as a matter of construction, as requiring any invoice, let alone a VAT invoice.

It is, however, necessary to complete the picture by reference to the terms of the franking licence which Royal Mail issued to authorise the use of a franking machine - Documents 1 and 2 in Franking Assumption 6(ii). Document 1 is called a “Franking Licence”. The grant of a licence for machines is provided for in paragraph 12(6) of the 1979 Scheme. On the first page it says:

“Your franking machine licence is a promise by the Royal Mail to provide you with services for franked mail. In return, we require you to follow the terms and conditions printed on the reverse of this licence.”

The rest of the first page identifies the customer and its machine.

The conditions printed overleaf are headed up:

“Terms and conditions for franking

Royal Mail Group Plc's ("Royal Mail") terms and conditions for franking letters and parcels and for posting franked mail in the United Kingdom are set out in the Post Office Scheme for Franking Letters and Parcels 2000. These terms and conditions include some of this information for easy reference.”

I am invited to read the reference to the 2000 Scheme as if it were a reference to the 1979 Scheme. I understand this to be because it has not been easy to find examples of all the licences which will have been operating under the 1979 Scheme and the parties consider it would be useful to assume that the terms of this licence had been deployed under the 1979 Scheme, or perhaps words to a similar effect. I shall therefore apply this licence accordingly.

The terms and conditions appearing on the reverse of this licence are terms dealing with matters such as posting franked mail, franked impressions, user authorisation, ending the licence, maintenance and repair and other matters. None of them deal with invoices, payment or VAT.

This licence would be capable of being significant if it were capable of giving rise to a contract and thereby of requiring a process of construction of the 1979 Scheme under contractual principles which would affect my conclusions above in relation to the effect of that Scheme. However, since I have concluded that there is no process of construction of the Scheme, treating it as a contract, which would assist Mr Cordara, it does not matter whether this licence gives rise to some sort of parallel co-extensive contract incorporating the Scheme, or not. I tend to doubt that it does, but I do not have to decide that. Nor is there anything in the terms and conditions on the reverse of the licence which provide Mr Cordara with any words which he could usefully seek to construe in his favour. In the circumstances, therefore, this licence adds nothing.

There is one more licence which was placed before me. It is a short form of licence which invites the user to peel off a sticker and attach it to the machine and which sets out details of the user, the machine and the die number. It says that: “The full terms of

Example 5 - Franking Services - the 2000 Scheme

This Scheme was made under section 89 of the Postal Services act 2000. This provision is very similar to section 28 of the 1969 Act. Section 89 provides:

“89.

Schemes as to terms and conditions for Provision of a universal Postal Service.

(1)

A universal service provider may make a scheme under this section in relation to the services provided by him in connection with the provision of a universal Postal Service or any of those services.

(2)

A scheme under this section is a scheme for determining any or all of the following (so far as not otherwise agreed) –

(a)

the charges which are to be imposed in respect of the services concerned,

(b)

the other terms and conditions which are to be applicable to the services concerned, and

(c)

procedures for dealing with the complaints of persons who use the services concerned.

…

(7)

any charge payable by virtue of this section may be recovered by the universal service provider concerned and in England and Wales and Northern Ireland may be so recovered as a civil debt due to him.”

Under section 89A of the Postal Services Act 2000 Ofcom has certain powers to require modification of any proposed scheme.

My remarks above about the apparent effect of the 1969 Act apply to this Act. It does not look from the statute as though it is putting in force a contractual scheme, though contract may have a part to play under the “(so far as not otherwise agreed)” proviso. Nor is there anything in the Scheme which points to its operating via a contract (absent some specific agreement to that effect) and in fact there are two contra-indications in the Scheme itself:

Accordingly normal principles of contractual construction do not apply here but I will, as in relation to the 1979 Scheme, consider what the position would be were I to be wrong on that once I have considered it on what I consider to be the true position.

Relevant paragraphs (most of which are required only for the purpose of setting the background for the one paragraph which matters) are set out in Appendix 1. The only word which might be capable of importing an obligation to render a VAT invoice, and which requires interpretation for these purposes, is the word “will” in clause 7.7 - “a VAT invoice will be sent to the User in accordance with Royal Mail’s invoicing procedures”. I agree with Mr Herberg that in its context that word does not import an obligation. It is a statement of procedures, not a statement of obligation. Exactly how that happens is clearly a matter for Royal Mail, not for the User. The Scheme is replete with more conventional words of obligation, all imposed on the User - “may not” and “must”. There is nothing in the Scheme which, as a matter of statutory interpretation, suggests that the “will” means a “shall” which may be enforced by the User.

If I am wrong in saying that the Scheme does not have statutory effect, and if it falls to be interpreted as a contract, then in my view the same conclusion applies for the same reason.

The tests for implication are set out above. For these purposes there is nothing materially different between any of the various forms of contract that were before me, treating the statutory schemes as contracts for these purposes. There is, however, one material difference between the circumstances of the 2014 Scheme and the other contracts (or potential contracts). Assumption 5 could not be operating across the board when this Scheme was in force because by 2014 the parties no longer shared it because of TNT1, and the consequential change in the UK legislation. This was not a point developed by the parties, who seemed to think that this Scheme could be lumped together with the other contracts for these purposes. I will have to consider separately what to do about that.

In oral argument Mr Cordara contended for 3 alternative versions of an implied term, each of which was alternatively suggested to be sufficiently obvious to pass the implied term test. They were:

In his skeleton argument his proposed implications were put differently:

As well as the Assumptions, he sought to add a couple of additional pieces to the factual matrix against which the implication should be judged.

Armed with this background, Mr Cordara argued for his implied terms. His case was (as it has to be) an obviousness case, based primarily on the need for a VAT invoice and the foreseeability of changes, which in oral argument he tested via the metaphor of the officious bystander. He also sought to argue that a term requiring a VAT invoice fell to be incorporated into all contracts of this type, that is to say a business-to-business contract with both parties registered for VAT.

I will start by disposing of his latter point (the class of contracts point). If it is to have any force it can only work where there is a class of contracts to which it might apply. I can see that there might be the beginnings of an implication which requires the rendering of a VAT invoice, or a right to demand one, in the case of all contracts where both parties understand VAT to be applicable, though even then the position is not clear, and that is only a start of the reasoning - I do not pursue it further. That is because that class of cases, if it is a class, does not include the cases in this claim. The Assumptions mean that it has to be assumed that VAT was in fact applicable, but they also mean that the parties understood that not to be the case, with some considerable support from the legislation. That is not the same class of contract as the more general class which I

It was necessarily accepted by the parties that obviousness lies at the heart of this class of implication - see the cases cited above. They (and in particular Mr Cordara) tested obviousness by reference to the usual metaphor of the officious bystander. I shall start by doing the same.

As Sir Kim Lewison says in his book in The Interpretation of Contracts, 6th Edition at para 6.09:

“…it is vital to formulate the question to be posed by the “officious bystander” with the utmost care. The “officious bystander” must be equipped with such knowledge as is necessary for him to be able to ask the necessary question. It seems probable that he must be equipped with all the background knowledge that would have been available to the parties and which would have been admissible for the purpose of interpreting the contract.”

It is also necessary to consider the question and answer on a contract by contract (or in this case a type of contract by type of contract) basis because the terms of the contract are capable of informing the question of whether there should be an implication, though as will appear the answer is generally the same. This was not an exercise conducted by Mr Cordara. I shall also focus first on the three alternatives proposed by Mr Cordara in his oral submissions rather than the more elaborate formulations proposed in his skeleton argument, observing at this stage that so many variants do not exactly assist a case based on obviousness.

So I take first Example 1 - Parcelforce. This contract makes no reference to VAT, and the assumption is that the parties did not think that it was chargeable. In order to get to one of his first three types of implied term the officious bystander would have to ask:

“Will a VAT invoice be provided whenever a taxable supply is made?”

And receive the answer “Of course”. But that would not happen. If that question were answered the parties would say:

“Yes, but that does not apply to this transaction because everyone knows it is exempt.”

Anything else would be nonsensical. The response would not generate the term which Mr Cordara requires. The same applies to the formulations based on request or reasonable request.

In fact the question is loaded, because it assumes a state of affairs which is not in the parties’ minds and does not invite them to address it. The more proper way of testing obviousness would be to propose a question which reflects everyone’s state of mind (and the legislation) and ask:

“If, contrary to your shared understanding, it turns out that this transaction is not VAT-exempt, should Royal Mail issue a VAT invoice in the same gross amount, reflecting VAT as part of that

sum?”

Whatever the customer might say, Royal Mail would never say “Of course” to that. The reason for that is that to agree to render a VAT invoice of the kind required by Mr Herberg would expose Royal Mail to financial risk and to upsetting the agreed commercial deal. If it rendered an invoice it would be obliged, or at the very least be exposing itself to an obligation, to render the relevant VAT to HMRC. Mr Cordara sought to say that that was no longer a risk in this case because the limitation periods for liability contained in VATA section 73 (one, two and four years) meant that there was no exposure to such a liability in this case. Mr Herberg sought to say that there was still a potential exposure under Schedule 11, which Mr Cordara in turn disputed. I do not need to deal with the real effect of this legislation (and therefore do not set it out). The fact is that Royal Mail would perceive a risk of a substantial liability (and it must be remembered that the question must be considered as at the date of the contract, not as at the present date after the lapse of so much time). That potential liability, if it eventuated, would upset the commercialities of the transaction, with the customer paying much less than he bargained for, and Royal Mail receiving much less than it had agreed. Royal Mail would never have agreed to that. If one imagines a dialogue (which of course one does not do where the officious bystander is concerned) Royal Mail would doubtless counter a suggestion that it should render a VAT invoice with a suggestion that the customer might like to pay the VAT that no-one was thought was due, as an adjunct to the agreed price. That demonstrates the non-obviousness of the proposed implied term. It would introduce a degree of commercial incoherence which an implied term is supposed to remove, not create - see Lord Neuberger’s 6th point in Marks and Spencer paragraph 21.

Mr Cordara’s position is not assisted by his qualifications by reference to requests and reasonableness. The addition of a request goes only to timing; the same objections apply. The qualification by reference to reasonableness introduces a slippery bar of soap factor which would lead to dispute, not resolution. The officious bystander would get the same “No” from Royal Mail.

Nor is it improved by his alternative formulations in argument. The first three are basically the same as the proposals that I have rejected. (d) is no more appealing. It is not plausible, or obvious, that Royal Mail would accept such a broad obligation as a matter of contract, and (e) would attract expressions of bemusement, not testy acceptance, were it proposed by the officious bystander. Royal Mail would not, as a matter of contract, undertake any obligations “to ensure the proper administration of the VAT system”. That is a matter for HMRC.

That reasoning applies to Example 1. The same conclusion apples to Example 2 (Tracked Mail). That example contemplates invoices, but as a trigger, not an obligation. It is far from obvious that, in the face of such an express term, there should be an implied obligation to render an invoice, and even less obvious that it should be a VAT invoice against the background of Assumption 5. It is significant that in Europhone Ferris J declined to imply an obligation to render an invoice, in part because the supplier would wish to render one anyway, in order to get paid. It was unnecessary to have an implied

Furthermore, this example refers to VAT and refers to any VAT chargeable as being paid in addition to the agreed price. An implied term which requires the rendering of a VAT invoice where the VAT is to be wrapped up in the price offends against the principle that implied terms should not contradict express terms.

The same sort of reasoning applies to Example 3 (Mailsort). Assumption 5 by itself would be enough to dispose of obviousness and the officious bystander, but in addition there are various references to invoices which would stand in the way of implied terms as to invoicing, and most significantly of all the provisions in the User Guide that “Postage charged by Royal Mail is zero-rated for VAT purposes” prevents the implication of any term which requires the rendering of any VAT invoice, let alone one which is based on a VAT-inclusive price.

I have already held that Example 4 (the 1979 Franking Scheme) is not a contract, so there is no room for the concept of the implication of contractual terms. If I am wrong about its status, then the same reasoning would apply to the notional contract incorporating the Scheme terms as applies to the other contracts - there is no basis for implication.

So far as the 2014 Scheme is concerned, the territory is again familiar. By this time the full blown Assumption 5 does not apply in that the understanding that it applied across the board had gone. But translating that Assumption to the 2014 Scheme (which needs to be done in order to comply with the basis on which these preliminary issues are to be decided) one still has an assumption that for this transaction the parties, at least, did not think that VAT was chargeable, and that is fatal to the implication of a term which supposes that it is, for all the reasons given above. Furthermore Mr Cordara’s first three implied terms would be inconsistent with what I have held to be the true construction of the reference to invoicing (see above), which bars the implication.

I have not applied Mr Cordara’s more elaborate implications to the 5 Examples, because the answer is the same as that already expounded.

It follows that the answer to Issue 6 is: No - there was no contractual duty on Royal Mail to issue a VAT invoice.

This issue remains live for the purposes of the proceedings generally but not for the purposes of this trial of issues. The claimants wish to assert that Royal Mail is an emanation of the state and as such is obliged to provide the invoices it seeks because, as such, it has to comply with the relevant provisions in the Directives. Royal Mail disputes that. That question will be decided on a later occasion. However, for the purpose of these issues Royal Mail is prepared to assume a direct liability under the Directive. That is so that issues concerning limitation can be decided at this stage on the assumption of liability, although, as will appear, when it got to the hearing Mr Cordara in fact maintained that there is insufficient evidence of loss to enable me to decide the limitation point.

I therefore proceed on the footing that for the purposes of this phase of the proceedings the answer to this Issue is: Yes, by concession for the purposes of this hearing.

Since I have concluded that there is no cause of action based on a breach of UK statutory duty, this question does not arise in relation to that claim. However, bearing in mind the test status of the issues before me I will express a view on these points in relation to such a claim on the assumption that I am wrong in my conclusions. In any event the point arises in relation to the assumed direct duty under the Directives.

These points all raise questions about the nature of the cause of action in this case for the purposes of limitation. The points arise because of the terms of sections 8, 2 and 9 of the Limitation Act 1980. Section 8 reads:

“8.

Time limit for actions on a specialty

(1)

And action upon a specialty shall not be brought after the expiration of twelve years from the date on which the cause of action accrued.

(2)

Subsection (1) above shall not affect any action for which a shorter period of limitation is prescribed by any other provision of this Act.”

Thus if the claim for breach of statutory duty falls to be treated as an action on a specialty then a twelve year period applies unless it also falls within another of the heads in the Limitation Act to which a shorter period applies (subsection (2)). There are in fact two, and not merely one, alternative candidates proposed. The first is section 2 (actions in tort, to which a six year limit applies) and section 9 (“An action to recover any sum recoverable by virtue of any enactment…”) to which again a six year period applies. These four issues do not expressly cater for the possibility of the application of section 9 but it was debated before me and I shall deal with it.

Some of the submissions in this case seemed to be predicated on the proposition that one had to choose whether the cause of action was either an action on a specialty, or an action in tort, or an action for a sum of money recoverable under a contract. That does not seem to me to be necessarily quite right. Section 8 seems to contemplate that, at least for the purposes of limitation, an action could be an action on a specialty as well as falling within one or other of the other two categories. In the latter event the shorter period applies, but it does not necessarily follow that the cause of action could not also be an action on a specialty. If, for example, the action is one founded in tort then the lesser limitation period applies; it does not necessarily follow that the action has to be completely recharacterised. It is enough to say it is an action in tort, whatever other qualities it might have.

It is tempting to go first to the latter questions and consider whether a lesser period does apply, but since I heard arguments on the classification I shall deal with specialty first, albeit a little more briefly than might otherwise be necessary because (as will appear) even if the action is one on a speciality, I have concluded that the action is also an action in tort whether it is founded on the English statutory regulations or on the European Directive, so the shorter 6 year period applicable to actions in tort applies.

Thus the starting point is to consider whether a claim under the various VAT Regulations is an action on a specialty. It is well established that deeds are specialties for these purposes, but that is not relevant. It is also established that some claims under some statutes are specialties. In Collin v Duke of Westminster [1985] QB 581 the Court of Appeal had to consider the limitation period applicable to the statutory right of a tenant to claim the freehold of his property under the Leasehold Reform Act 1967. It held that the right to claim that interest was a specialty. Having rejected the suggestion that what the court was really enforcing was a contract, Oliver LJ held that it was indeed a specialty. He cited, with apparent approval, a passage from Aylott v West Ham Corporation [1927] 1 Ch 30 at 50:

“where a plaintiff relies and has to rely upon the terms of a statute so that his claim is under the statute the nature of the claim is one of specialty and the 20 years applies.”

and having referred to results of other cases he went on:

“Broadly the test is whether any cause of action exists apart from the statute:…” (p602A)

Having said that, he explained what that meant in terms of that case:

“It seems to me to be quite clear that in the instant case any cause of action which the applicant has derived from the statute and from the statute alone. Apart from the statutory provisions he could have no claim and it is only by virtue of the statute and the regulations made thereunder that there can be ascertained the amount of the price to be paid under the statutory contract the terms of which can be gathered only from the sections of the Act and the Schedules. Subject, therefore, to one question, namely whether the word "specialty" as used in the Limitation Act 1939 and the Act of 1980 has assumed a more limited meaning than it originally bore, I have no doubt at all that the applicant's claim is a claim on a specialty.”

Those are words to which I will have to return. They were applied by Mummery LJ in the Court of Appeal in Rahman v Sterling Credit Ltd [2001] 1 WLR 496. That case concerned the right to re-open an extortionate credit bargain under section 139 of the Consumer Credit Act 1974. That Act expressly gave the court power to re-open the bargain and relieve the debtor in whole or in part from the obligation to make further payments. Mummery LJ held that Oliver LJ’s reasoning applied:

“That reasoning applies to the statutory right of a borrower to make application to the court under section 139. The cause of action arises out of and only out of those provisions of the 1974 Act. Apart from those provisions Mr Rahman would have no right to have the loan agreement reopened in that manner.

It follows that, in so far as Mr Rahman seeks, whether by counterclaim or by separate action, to make a claim to reopen the loan agreement under section 139, that claim is not barred by limitation: that cause of action arose in 1989, less than 12 years ago. If he is successful in his claim the court may make an order relieving him in whole or in part from the obligation to make future payments. That in turn would make it necessary for the court to reconsider whether it was appropriate to leave the 1990 possession order in place.” (502B-D).

Section 8 was considered in Re Priory Garage (Walthamstow) Ltd [2001] BPIR 144 in the context of a claim to set aside a transaction under sections 238 and 239 of the Insolvency Act 1986. One does not get much reasoning from this; the real question was whether the claim was taken outside the effect of section 8 by its being a claim for payment of money under a statute under section 9. The principle of the case was followed by HH Judge Weeks QC in Re Nurkowski [2005] BPIR 842, without further reasoning.

I therefore turn to consider whether these analyses mean that the claim based on a breach of the Regulations is an action on a specialty.

The first point taken by Mr Herberg is that the claim on the English legislation is a claim under statutory regulations and not under the statute itself. I do not consider that this would stand in the way of the application of section 8 if it were the only question. If the claim in Collin were a claim based on regulations to the same effect as the actual statute, made under a statutory power, I find it hard to believe the decision of the Court of Appeal would have been different. It is clear that Oliver LJ was aware that part of the claim in Collin depended on regulations - he refers to them. I doubt if Mr Herberg’s point is correct.

However, even given that, I do not consider that the claim in the English regulations is a specialty. Mr Cordara points to the dicta in Collins which he says provide that absent the statutory provisions there would be no claim, and that, he says, is what makes our action an action on a specialty, because absent the regulations there would be no claim. I do not consider that that is the correct analysis. The fact that the action stems from a statute originally cannot, by itself, make the action one on a specialty, because otherwise all actions which have a statutory root would fall into that category. What distinguishes the cases above from many other cases is the fact that the whole package - the right to claim and the appropriate remedies - comes from the statute - see Oliver LJ’s longer paragraph cited above in which he lists all those matters. The same is true of the subsequent cases which follow Collin. They are all cases in which the circumstances giving rise to the claim and the remedies are all built into the statute. In that sense there would be no claim but for the statute, when one regards the claim as what I call the whole package.

That does not describe the present (deemed) cause of action under the VAT Regulations. Only the “wrong” is described. Nothing is said about civil remedies. Part of the package is missing. So in my view the claim is not one on a specialty for the purposes of section 8.

The same is true for the claim under the Directive, which I am invited to assume to exist, and for the same reasons.

Knocking out the possibility of the claim being one on a specialty is not good enough for Mr Herberg because he still needs a limitation period. Furthermore, if I am wrong in my conclusion in specialty he needs a shorter period to avoid the 12 year period of section 8.

Mr Herberg’s first candidate is section 9 (“An action to recover any sum recoverable by virtue of any enactment …”) to which a 6 year period applies. He submitted that the claims for breach of statutory duty were claims to recover a sum recoverable by virtue of an enactment, namely the VAT Regulations.

The trouble with this argument is that the Regulations say nothing about money, or recovery of money. On the hypothesis of this part of this judgment they impose an obligation without saying more about its effect. The cases cited to me were where the statute in question specified not only the cause of action but also specified a remedy, and that remedy was money or an equivalent. Thus in Re Farmizer Products Ltd; Moore and another v Gadd and another [1997] BCLC 589 a claim under section 214 of the Insolvency Act 1986 (summary remedy against directors and officers of a company for defaults) was held to fall within section 9 because the section provided for a “contribution” to be made and that meant money. In Rahman v Sterling Credit Ltd [2001] 1 WLR 496 the court held that a claim to re-open a transaction under the Consumer Credit Act 1974 was an action on a specialty but a claim for repayment of moneys, which was a remedy specifically provided for under the Act, was thought by Mummery LJ to be a claim to which section 9 applied, though it was not necessary for him actually to decide the point.

In my view the facts and decisions in these cases confirm what one gets from the wording of section 9 itself. To fall within that section there has to be a claim prescribed by statute, and the statute has to prescribe a form of monetary payment (or an equivalent) payable in respect of that claim. The present case does not fall within that description. On the assumption that the Regulations give rise to a civil claim, they provide an obligation (to provide an invoice) but say nothing about any civil remedy for a disappointed customer, whether monetary or otherwise. (There are, of course, remedies or sanctions available to HMRC, but they do not provide for sums to be payable to the disappointed customer.) The claimants claim damages, but that is not a prescribed remedy under the section. It is a remedy provided for in the general law. So the damages are not “[a sum] recoverable under a statute”. This conclusion is consistent with the decision of HHJ Toulmin QC in R v Secretary of State for Transport Ex parte Factortame Ltd (No 7) [2001] 1 WLR 942 at para 163:

“I therefore construe the words “any sums recoverable by virtue of any enactment” in section 9 of the 1980 Act as referring to cases where those sums which are recoverable by the claimant are specified in or directly ascertainable from the enactment. This is to be contrasted with damages recoverable under section 2 of the 1980 Act which are compensatory damages assessed under common law principles and which cannot therefore be directly ascertained from the statute.”

That leaves Mr Herberg with section 2 - “An action founded in tort …” to which a six year limitation period is applied. He submits that authority tells us that claims for breach of statutory duty are actions in tort, and if that is right then a six year period applies under section 2 even if a 12 year period would have otherwise applied under section 8. Mr Cordara did not accept that.

Since the preliminary issues seem to contemplate there may be a difference between claims for declarations and other claims I shall first consider those other claims (damages and an injunction, though it is accepted that by reason of section 36 any limits that would otherwise apply do not apply to the injunction claim).

Mr Herberg’s starting point for saying the claims are claims in tort was a paragraph from Clerk & Lindsell on Torts, 22nd Edition - paragraph 9-01. This is the first paragraph of a chapter headed “Breach of Statutory Duty”. The presence of a chapter with that title in a book on torts clearly assumes that the wrong is a tort, but does not prove it.

In fact that paragraph in question does not contain any reasoning as to why the wrong should be treated as a tort. The paragraph says:

“A person who has suffered damage as a result of the breach of a statutory duty may have an action in tort, classified by Lord

Browne-Wilkinson in X (Minors) v Bedfordshire CC as an “action for breach of statutory duty simpliciter”).”

I think that Mr Herberg would wish to suggest that that sentence suggests that the idea that it was tort came from Lord Browne-Wilkinson, but a careful study of the relevant part of his judgment reveals that he did not say anything about tort - indeed, in that part of his speech (as opposed to later parts when he is dealing with something else) he does not even mention the word. In fact, that is not what the paragraph is doing. The paragraph in the textbook is in fact laying the groundwork for a consideration of the circumstances in which a breach of statutory duty can give rise to a private law claim by a “victim” (the point dealt with above). It assumes that the wrong is a tort, but does not of itself go further than that.

However, other authority is of rather more assistance to Mr Herberg. In Factortame (No 7) (supra) HHJ Toulmin had to consider the limitation period applicable to a direct claim against the UK government for breaches of European Community law. He concluded that section 2 was applicable. He concluded that he had to give section 2 a wide interpretation (paras 132 and 149) and found that the claim before him fell within it. He did so by determining, without a lot of reasoning, that:

“An action for breach of English statutory duty is properly classified as an action founded on tort.” (para 145)

He then extended that to the European law-based “wrong”. On the way he defined a tort as:

“…a breach of non-contractual duty which gives a private law right to the party injured to recover compensatory damages at common law from the party causing the injury.” (para 150)

The principle that damages against the UK for infringing European-based legislation amounts to a tort for the purposes of section 2 was accepted by the parties and by Sir Andrew Morritt V-C in Phonographic Performance Ltd v Department of Trade and Industry [2004] 1 WLR 2893, and by the parties and the Court of Appeal in Spencer v Secretary of State for Work and Pensions [2009] QB 358. The same applies to F J Chalke v HMRC [2009] 3 CMLR 14. It seems to me clear that if infringement of European legislative provisions amounts to a tort for the purposes of section 2, then a fortiori so must infringement of a UK provision. It is significant that in Sevcon Ltd v Lucas Ltd [1986] 1 WLR 462 a claim for damages for infringement of a patent, awardable under the Patents Act 1949 (but not expressly provided for under the Act),

In another sphere claim based on a statute has been held to be a tort. In American Express Co v British Airways Board [1983] 1 All ER 557 it was held that a claim based on the theft of a postal packet amounted to “proceedings in tort” for the purposes of section 29(3) of the Post Office Act 1969, which provided for an exclusion of such claims. One of the bases of the claim was under the Warsaw Convention on the International Carriage of Goods by Air, which was given statutory effect. Lloyd J assumed that there was a claim and held that it amounted to “proceedings in tort”. He arrived at that conclusion by relying on textbook and case-law citations which characterised such claims as being generally tortious claims. It is unnecessary to set out his extensive citations. It is sufficient to point out that the case reflects a view that statutory claims were capable of having that character. To the same effect is the general assumption of the whole of a book on “Statutory Torts” by Stanton, Skidmore, Harris and Wright (2003). The authors of that work state, as apparently obvious, that section 2 applies (see paragraph 9.012).

The considerable preponderance of authority is therefore to the effect that, if there is a statutory wrong of the kind relied on in this case, it falls to be treated as a claim in tort. In my view it is correct to treat it as a tort for the purposes of section 2 of the 1980 Act so far as the damages and injunctions claims are concerned.

Is the claim for a declaration any different? The declaration sought is in the following terms:

“A declaration that the Defendant is obliged to issue VAT invoices and/or corrective VAT Invoices in accordance with its statutory duty, and/or contractual duty, and or the European Duty, under the Principal VAT Directive.”

It is difficult if not impossible to see what good a declaration would do in the present case if any claim for damages is statute barred. The claim for an injunction may or may not be barred by lapse of time - the statutory time limits do not per se apply to that claim (section 36). If an injunction is ordered the declaration is pointless. If it is not ordered then the declaration cannot be granted, as it seems to me, because there is no basis on which it could be said to be correct when all remedies are barred by lapse of time. As Webster J said in National Bank of Commerce v National Westminster Bank [1990] 2 Lloyds Rep 514:

“But I cannot think of a situation in which a Court in this country would make a declaratory judgment of a right which could not be enforced here because a claim to enforce it would be statute barred.”

Mr Cordara was unable to say what useful purpose the declaration would serve.

However, this is not a striking out application. The question is whether the declaration is within section 8 or section 2 of the 1980 Act, and I shall answer that question briefly. (Issue 3 does not propose section 5 as a further alternative.) The claimants’ argument, advanced more in their skeleton argument than in oral argument, was that the claim for a declaration was founded in a debate about underlying rights which were fiscal, not tortious (or contractual), and that argument was said to be fortified by the fact that unlike a claim in tort, the claim for the declaration did not require damages to be proved. Their case was that the claim for a declaration was an action on a specialty.

The point is dealt with, adversely to the claimants’ case, by the decision in Woodeson v Credit Suisse [2018] EWCA Civ 1103. In that case various declarations were sought in relation to the state of account between the parties to the action. The Court of Appeal held that the Limitation Act could apply to a claim for a declaration by reference to the nature of the underlying claim, relying on a decision of Colman J in P & O Nedloyd BV v Arab Metals Co [2005] 1 WLR 3733. The relevant part of Longmore LJ’s judgment is at paragraph 24:

“24.

In P & O Nedlloyd (paras 20-21) Colman J seems to have agreed with this criticism and set out the way he saw it:-

"The function of the Limitation Act 1980 is to identify those periods of time within which a claimant is permitted to invoke the jurisdiction of the court to grant relief. Those periods of time vary according to the nature of the grounds for relief. Those grounds comprise both the factual foundation and the assertion of a legal or equitable right consequential upon those grounds. A claim for a declaration that a contractual right has accrued or that a breach of contract has occurred is thus a claim, or, in the words of section 5 of the Limitation Act 1980, an action, for relief founded on grounds an essential part of which is a simple contract.

Moreover, a declaration, although a discretionary remedy, is not an equitable remedy but is a creation of the Judicature Acts 1873 (36 & 37 Vict c 66) and 1875 (38 & 39 Vict c 77): see Chapman v Michaelson [1909] 1 Ch 238. Accordingly, the claims for declaratory relief in this case do not engage section 36(1)(b) of the Limitation Act 1980."

I agree with this approach and would hold that the claimants' cross-claims in the present case have as their "basis", to use Professor McGee's word, claims in tort and breach of statutory duty which must be brought within 6 years of the accrual of the cause of action. The claimants' position cannot be improved by making claims for declarations rather than for damages and the claims must, therefore, be regarded as timebarred unless time can be extended by virtue of the bank's deliberate concealment.”

In the present case the claims (assuming for these purposes that there are some) are based in tort for the purposes of the Act, so any claim for a declaration would fall to be treated under that head. (The same would apply if the declaration were based on contractual rights.) That answers the question raised in Issue 3, which is the question I am invited to address. It may not quite answer the question of whether a declaration could be granted if enforcement via an injunction were available, but as I have observed the declaration would be pointless in that event anyway.

In relation to Issues 3, 4, 11 and 12, I therefore conclude that the actions are not actions on a specialty, and are actions in tort, and that section 2 of the Limitation Act 1980 does indeed apply to the claim for a declaration. Those questions should be answered accordingly.

This issue goes to limitation. As with the other issues, it does not actually arise because there is no cause of action in the first place, but again in order to assist the furtherance of the resolution of this dispute, and in case I am wrong about the statutory cause of action, I shall consider this issue on the footing that I am indeed wrong and there is a statutory duty to provide a VAT invoice as claimed. I shall deal with accrual of cause of action before dealing with whether its nature is continuing or not.

I approach this issue on the assumption that the relevant classification of the claim based on the assumed statutory duty is that it is a breach of duty amounting to a tort for the purposes of section 2 of the Limitation Act, and not an action on a specialty within section 8 (with no lesser period applying to it for the purposes of section 8(2)). I shall also deal with the possibility of a contractual cause of action.

Mr Cordara’s submissions on the tort point can be boiled down to the following propositions. In an action for breach of statutory duty no cause of action, at least in damages, accrues until damage is suffered. Damage has not been suffered until it is plain that loss has been caused, and loss is not suffered until one gets to the point of time at which it becomes apparent that the ability to recover VAT is fatally prejudiced by the lack of a VAT invoice; that is to say, damage accrues when, but for the want of an invoice, VAT would have been reclaimable. Until TNT1, the Commissioners would not have paid out on any VAT invoice anyway, so no loss was suffered in that period. Until TNT2 decided the VAT status of some services, HMRC would not have paid on those either, so there was no loss then. One needs an evidential inquiry to determine the earliest time when HMRC would have paid out on any given service; until then, and in relation to those services, there was no loss and therefore no cause of action. In fact, the position is more subtle than that. Until Zipvit is finally decided, it will not (on this line of reasoning) become apparent that there is no alternative route, which means that arguably the possibility of loss is not crystallised until then, so the limitation clock has not yet started running even now. Mr Cordara said that all this required an evidential inquiry which I could not conduct in this case.

Mr Herberg started from the proposition that a breach of statutory duty was classed as a tort for limitation purposes, and he submitted that because it was a claim in tort a cause of action did not accrue until some damage accrued. Damage was necessary to complete a cause of action in tort. However, that date was much earlier than Mr Cordara submitted. The question was when some measurable damage accrued, and that was the date from which the claimants were first unable to recover their input tax. That, in turn, was the date of the first VAT return in which, but for the absence of an invoice, they could have claimed it. It did not matter that HMRC would in fact have rejected the claim. TNT1 declared what the law had been, not what it was to be for the future. A decision by HMRC to refuse the claim would have been challengeable in law, and that challenge, on the basis of the law as we know it to be, would have been successful (in the case of a VATable service, which has to be assumed for these purposes).

Thus both parties started from the proposition that what one was looking for so far as an action in tort is concerned is the date when damage accrued from the breach of duty in question, because until that happened there was no complete tort. I am afraid I disagree with both of them on the principle of that matter. I do not consider that it follows from the fact that a breach of statutory duty is regarded as a tort that damage is an essential ingredient in all cases before the cause of action accrues. Both parties treated that as a given. I am afraid I do not. In some (indeed many, or even most) cases that may be true of actions in tort, but it is not a universal requirement. As pointed out in Clerk & Lindsell on Torts, 22nd Edition at paragraph 1-52, there are a number of torts actionable per se and in relation to which damage is not a necessary ingredient - the editors list trespass, certain nuisance claims, libel and some old cases relating to breach of duty by public officers (the latter perhaps being significant in the present context). The requirement of damage is therefore not a given. Furthermore and in any event, the fact that a breach of statutory duty is classified as a tort by textbook writers, and should be held to be a tort for the purposes of limitation, does not mean that it

It is true that in many cases of breach of statutory duty there is no cause of action until damage accrues, and Lord Browne-Wilkinson’s description seems to require that damage be incurred, but it would be wrong to take those factors and require damage in every case. In all justifiable actions for breach of statutory duty an individual will have to establish an undesirable effect on him or her, and that will generally be synonymous with the causing of some damage, but I do not see why that should lead to a rigid requirement that there be damage (as opposed to some other detriment which does not necessarily sound in damages at the time) before there is an accrued cause of action.

I consider the present matter to be a case in point (assuming there is an actionable breach of statutory duty in the first place). Under Regulation 13(1) there is a (deemed) obligation to supply an invoice. The time when it is to be provided is in paragraph (5) - within 30 days of when the supply is treated as having taken place (ie when the services were provided, for practical purposes). On the 31st day after the supply, if there is no invoice, there is a breach of duty. Mr Cordara and Mr Herberg both accepted that an injunction could be applied for on that date. They are both right - there is at that point a breach, which is actionable. A cause of action has accrued. Mr Herberg, who seemed curiously resistant to the idea that the availability of an injunction meant that a cause of action had accrued, suggested that the injunction was a form of quia timet injunction, but that cannot be right. A quia timet injunction is one granted in anticipation of a breach. In the present case the injunction would be granted because there has already been a breach.

Mr Cordara lent heavily on Phonographic Performance Ltd v Department of Trade and Industry [2004] 1 WLR 2893 as demonstrating that the present claim was one for breach of statutory duty, that that meant it was a tort and that accrued damage was a necessary ingredient of that tort. That was a case of a Francovich claim against the UK government. It is true that Sir Andrew Morritt V-C determined that damage was a necessary ingredient of the claim so that no cause of action accrued until damage was sustained, but that was a different claim from that which is made in the present cases. It was a Francovich claim in respect of which the third requirement is that damage shall have been caused to the claimant. In other words, damage is a necessary ingredient of that particular cause of action because of the way it is formulated. See in particular paragraph 28 of the judgment:

“In my view its claims are not statute-barred because they are both claims in respect of a continuing breach of duty and a cause of action in which damage is an essential ingredient.” (my emphasis)

The case is therefore distinguishable and does not assist Mr Cordara.

In my view that is an end of the argument as to when the (deemed) cause of action for breach of statutory duty first accrued. It is unnecessary to go on and look for a date when damage first accrued. However, once again I express a view on that in case my conclusion on the absence of a need for damage is wrong.

I do so on the potentially artificial basis of the Assumptions. It may be that in a real world inquiry there would be some merit in Mr Cordara’s analysis. His submission was that I cannot reach a view on this point because of the sort of variables that he identified (see above), but that is not the world in which I am currently operating. I am determining these issues on the basis of the Assumptions.

The damages claimed in the common form of Particulars of Claim is “in particular” the inability to claim credit for, or to recover, input tax. I shall treat that, for these purposes, as being the sole damages claimed or claimable. At one level the question of when that damage was suffered is relatively easy to answer. On that footing, damage accrues (in relation to each occasion on which a VAT invoice ought to have been rendered) on the date of the next VAT return (generally required quarterly) when the customer could have made a claim for input tax but could not in fact do so for want of a VAT invoice. It is at that point that the customer is worse off in the amount of the unclaimable VAT.

That is Mr Herberg’s case, and I consider that he is right about that. If one is to assume that the price did, unknown to the parties and the Revenue, include VAT, and if one assumes that there was a statutory duty to provide an invoice notwithstanding that belief (which is my assumption for these immediate purposes, contrary to my analysis above), then I think one concludes that damage was incurred when the customer could have made a reclaim of input tax had it had an invoice but did not do so. It may or may not have been the case that the Revenue would (on this artificial factual basis) have refused to allow the input tax, but it would (on the Assumptions) have been wrong to do so. One cannot simply stop the inquiry with such a refusal. Since the Revenue would have been wrong, one has to assume it would have been put right by a court somewhere - this question should not be answered on the footing that the Revenue was right to be wrong.

That means that damage was suffered at each of the return dates when the relevant input tax could have been reclaimed. Whether that conclusion, reached on that basis, is of any use to the parties, is for them to ascertain.

In the course of identifying when the cause of action accrued I have also identified the nature of the wrong. It was a failure, on a given date, to provide an invoice which ought to have been provided. That underpins the next question, which is whether the breach was continuing or not. The significance of a finding that it was continuing is that if it was then the claim is not statute barred because there is a breach day by day.

Mr Cordara’s case was that the breach, once it had occurred, was a continuing one - he submitted that the situation was analogous to that of a former employee who competes with the employer contrary to a covenant, save that the present situation is tripartite. His client suffered more damage for every day that the reclaim could not be made and used, and there was more inconvenience every day. He relied on the fact that under the Regulation there was a daily penalty for 100 days, which he said demonstrated that the breach was a continuing one, and pointed out that in the Phonographic case above the breach was held to be continuing. For his part Mr Herberg said that Phonographic concerned a different wrong and that VAI Industries (UK) Ltd v Bostock & Bramley

The question of whether the breach is once and for all or continuing is a question of the construction of the obligation and the nature of the act in question. In my view it is plain that if the failure to provide an invoice was a breach of a statutory obligation it was a breach on the day when it ought to have been provided and not a further breach on the day after that and thereafter day by day. The obligation under Regulation 13 is clear - it provides for the provision of an invoice where a taxable supply is made and paragraph (5) provides for when it is to be done by - within 30 days of the supply. When that is not done in time, there is a breach, and the nature of the breach is a failure to supply an invoice within the 30 days. That remains the breach on the first day after the 30 days, the second day, and thereafter. There is no separate breach on the second and subsequent days. The breach is the same as it always was. If one asks the question on (say) day 5 after the 30 days, “Did the supplier commit a breach today?”, the answer would be: No - it committed a breach by the end of 30 days from the supply. In my view the conclusion contended for by Mr Cordara cannot be maintained in the light of that.

I do not overlook the provisions of Regulation 69 (see above in the section of other regulatory material) which provides for a penalty of a rate “multiplied by the number of days on which the failure continues …”. The “failure” can be said to continue after the 30 days, but it does not follow that there is a breach on every day after that 30 days. The position is analogous to that of the solicitor in Bell v Peter Browne & Co [1990] 2QB 495, where Nicholls LJ rejected the suggestion that a failure to remedy a breach of duty itself amounted to a fresh breach in respect of which an action could be brought. At page 500H he said:

“Despite this, it was in 1978 that the breach occurred. Failure thereafter to make good the omission did not constitute a further breach. The position after 1978 was simply that, in breach of contract, the solicitor had failed to do what he ought to have done in 1978 and, year after year, that breach remained on remedied. Nor would the position have been different if in, say, 1980 the plaintiff's solicitor had been asked to remedy his breach of contract and had failed to do so. His failure to make good his existing breach of contract on request would not have constituted a further breach of contract: it would not have set a new six-year limitation period running. Once again, the position would have been simply that the solicitor remained in breach.”

In my view the same reasoning applies in the present case, notwithstanding the imposition of a penalty. The penalty reflects the fact that there is an unremedied breach; it does not reflect the fact that there is a breach for every day on which the penalty can be incurred.

VAI Industries is a case which was ultimately one on its own facts. It concerned a warranty. Carnwath LJ and Newman J held that the breach was not a continuing one. Carnwath LJ said (para 51) that:

“… clearer words would be needed, certainly in a form of contract provided by the purchaser, to create a continuing obligation such as that relied on by the appellant.”

That tends to suggest that there is some sort of elevated starting point if one is to establish a continuing obligation. Newman J (in para 56) was clearer:

“The relationship of landlord and tenant (Bell v Peter Browne &

Co [1990] 2 QB 495) and husband and wife (Shaw v Shaw [1954] 2 QB 429) have a continuing quality and character giving rise to an "exceptional" obligation and, in my judgment, are of no assistance when considering a commercial contract for the supply of goods.”

The reference to exceptionality probably comes from a later passage of the judgment of Nicholls LJ in Bell v Peter Browne & Co at page 501D:

“For completeness I add that the above observations are directed at the normal case where a contract provides for something to be done, and the defaulting party fails to fulfil his contractual obligation in that regard at the time when performance is due under the contract. In such a case there is a single breach of contract. By way of contrast are the exceptional cases where, on the true construction of the contract, the defaulting party's obligation is a continuing contractual obligation. In such cases the obligation is not breached once and for all, but it is a contractual obligation which arises anew for performance day after day, so that on each successive day there is a fresh breach. A familiar example of this is the usual form of preparing clause in a tenancy agreement. Non-repair for six years does not result in the repairing obligation becoming statute-barred while the tenancy still exists. The obligation of the tenant or the landlord to keep the property in repair is broken afresh every day the property is out of repair…”

In my view those words can be simply transposed to cover the (deemed) breach of statutory duty in the present matter. The obligation to render an invoice by the end of 30 days after the supplying question does not “arise anew for performance day after day”. It arose as soon as the supply was made and was complete, as an obligation, at the end of the 30th day. On the 31st day there was not a new obligation to render an invoice by the 30th day. That makes little sense.

The Phonographic case does not assist Mr Cordara. That case concerned an entirely different obligation on different legislative provisions and there is nothing in it which can be usefully transposed into this case to assist the claimants.

I therefore conclude that if and insofar as the claimants have claims for breach of statutory duty, those claims were not continuing claims as posited in Issue 2.

Once again this Issue does not actually arise in the light of my conclusions as to the non-existence of the underlying claims. On this occasion there are so many potential variables to consider on the assumption that I am wrong in my conclusions that I decline to express a conclusion on each of the possible alternatives on each of the contracts. In case it assists the parties I will merely say:

Because of Assumption 2, the answer to this question is Yes. However, the significance of that must be made clear. Royal Mail does not accept that as a final conclusion to the matter. Whether it should actually be treated as an emanation of the state for the purposes of a direct cause of action is a very live issue and is to be determined at a later date. The parties are using the assumed position so as to put themselves in the position of being able to test the effect of limitation in the event that there is a direct claim.

In order to address these questions it is necessary to ascertain the scope of the duty and the time for its performance. Mr Cordara’s case (advanced more in his skeleton argument than orally) was that this duty arose from the Directives, and the benefits of it are not exhausted by subordinate legislation passed by Member States (Marks & Spencer Plc No 1 [2002] STC 1036 at para 27). He submitted that the question of whether an obligation was a continuing one was inevitably bound up with whether it could be repeatedly exercised rather than defeated by passage of time. The 30 day period prescribed by the UK legislation said nothing about the ending of the period of the breach, and Member States should do nothing to impede the due and proper operation of the VAT system. Therefore there is a constant accrual of causes of action while the breach (a failure to provide an invoice) goes unremedied. No time limit should be implied into Article 220. He relied heavily on what was said in Phonographic at paras 24 and 25 as demonstrating that each time a claimant rendered a VAT return its right to deduct VAT was frustrated by Royal Mail’s continuing failure to issue a VAT invoice. To the extent that limitation periods operate to defeat a taxpayer’s rightful claims, they must be disapplied, appropriately moulded or construed so as not to defeat a taxpayer’s right to deduct input tax.

It is not always easy to disentangle the various legal elements bound up together in Mr Cordara’s reasoning. In my view it fails to analyse the source and nature of the duty, and how it might be legitimately qualified, and confuses the issues of whether there is a limitation period applying with the question of the nature (continuing or otherwise) of the breach of duty. Mr Herberg’s approach in his skeleton argument was briefer - he

Spencer is not on all fours with the present case, at least so far as the formulation of the conceded duty is concerned. Spencer was a claim for failure to implement EU legislation properly. The concession arising (pro tem) from the accepted answer to Issue 9 is that there was a direct duty to provide an invoice, not a failure to implement obligations to provide an invoice. It is more direct. However, I think that the reasoning in Spencer nonetheless applies. In paragraph 35 Waller LJ said:

“35.

The fact that the United Kingdom could have remedied the situation is not to the point. The position seems to me clear - the United Kingdom was in breach if it failed to implement the Directive. I would accept that so far as Mr Moore was concerned the United Kingdom must be assumed to have been in continuous breach up until the moment he was injured by an untraced driver. Once Mr Moore suffered damage from the breach, that completes his cause of action against the government. The obligation of the government to Mr Moore, once he had suffered damage for their failure, was to pay damages.”

In the same way, in the present case the cause of action was complete when damage accrued (damage being necessary in this instance, because it is a Francovich claim) and there is no scope for any day by day continuing duty thereafter. That damage accrued when the customer could not claim the input tax in the appropriate return, as in the case of the breach of UK statutory duty. This is not, as Mr Cordara submitted, a matter of an illegitimate implication of a time limit as was held inappropriate in British Telecommunications plc v Revenue and Customs Commissioners [2014] STC 1926 (CA). It is a question of construing the nature of the obligation and the breach.

I also think that Mr Herberg is probably right in his reliance on the UK provisions which introduce the 30-day time limit. The (presently deemed) EU duty is set out in the two provisions identified in Issue 9 (Articles 22.3 and 220). The terms of those Articles are set out above. So far as Article 220 and the Principal VAT Directive are concerned, it is apparent that Member States have certain discretions in the operation of the VAT system in their respective territories. In ascertaining the scope of a provision having direct effect, it would be wrong (in these circumstances anyway) to ignore qualifications to that effect which Member States are entitled to introduce. One of the

That would seem to allow the UK to do exactly what it has done in the Regulations and impose the 30 day time limit. It has not been suggested that the UK was not entitled to introduce that time limit, and if that is the case then it was not explained to me why it should not qualify the scope of the direct obligation assumed by the agreed answer to Issue 9. I fail to see why the direct obligation should be treated as free from intra vires qualifications which the UK was expressly allowed to introduce. When combined with the obligation in the Directive, that gives rise to an obligation to the same effect as that contained in the Regulations - one which has to be fulfilled within the 30 days. It is a one-off, and non-continuing, obligation, for the reasons given above.

The legislative position in relation to Article 22.3 is slightly different. There is no equivalent of Article 222 of the Principal VAT Directive. However, it has not been suggested that the 30-day limit (first introduced by regulations in 1977) were outside the scope of the UK’s powers, so the absence of an equivalent to Article 222 ought not to matter for these purposes. So although the legislative position is slightly different, the result is the same.

The answer to Issue 10 is therefore that the cause of action accrued when damage was suffered as a result of the inability to include the input tax in the next return, and the breach is not continuing.

I do not need to answer this question myself because Mr Herberg accepted that the answer to this question was: Yes (ie the periods in sections 2 and 8 of the Limitation Act are disapplied by section 36).

Likewise, Mr Herberg accepted that the answer to Issue 8 was: Yes (ie section 36 disapplied other limitation periods to claims for specific performance).

Since the claim for breach of statutory duty does not fall within section 8, this question does not arise.

My answers to the questions, in the order in which they are raised, are therefore as follows:

regulation 8(1) of the Value Added Tax (General) Regulations 1975, SI 1975/2204 (“the 1975 Regulations”) (for the period from the beginning of each Claim Period until 31 December 1977);

regulation 8(1) of the Value Added Tax (General) Regulations 1977, SI 1977/1759 (“the 1977 Regulations”) (for the period from 1 January 1978 until 16 November 1980);

regulation 8(1) of the Value Added Tax (General) Regulations 1980, SI 1980/1536 (“the 1980 Regulations”) (for the period from 17 November 1980 until 31 August 1985);

regulation 12(1) of the Value Added Tax (General) Regulations 1985, SI 1985/886 (“the

regulation 13(1)(a) of the Value Added Tax Regulations 1995, SI 1995/2518 (“the 1995 Regulations”) (for the period from 20 October 1995 until the end of each Claim Period)

Article 22.3 of the EEC Sixth Council Directive of 17 May 1977 on Value Added Tax (77/388/EEC), to provide an invoice containing the particulars prescribed in that Article (for the period from the beginning of each Claim Period until 31 December 2006); or

Article 220 of the EU Council Directive of 28 November 2006 of the Common System of Value Added Tax (2006/112/EC), to provide an invoice containing the particulars prescribed by Article 226 of that Directive (for the period from 1 January 2007 until the end of each Claim Period)?

Appendix 1

Contracts – extracts

Franking Equipment must not be used before the User has obtained a Licence. All

If Royal Mail approves an application to use Franking Equipment it will issue a Licence to the User. Franking Equipment must not be used:(i) before a Licence has been issued to the User; (ii) after a Licence has either expired or been ended; or (iii) during any period of suspension of the Licence.

The User shall carry out all reasonable instructions given to it by Royal Mail in relation to recording details of the Franking Equipment, the issue of a Licence and the identity of the User.

If an application for a Licence is rejected, Royal Mail will tell the applicant and/or the Authorised Supplier or Authorised Manufacturer by whatever means Royal Mail deems appropriate. The requirement for a User to obtain a Licence applies whether or not the User has entered into a contract with Royal Mail to receive postal services. If an applicant has neither received a Licence nor a rejection then the application has been rejected.

The Licence will include the following information [details of the user and of the franking machine]

It is the User’s responsibility to ensure that a then current Tariff is at all times loaded onto, or is used in respect of, the Franking Equipment. Tariffs are available from Authorised Suppliers and Authorised Manufacturers and must be loaded onto compatible Franking Equipment in accordance with the instructions (if any) relating to them. The failure to use the correct Tariff may stop the Franking Equipment working and/or may lead to Royal Mail ending the Licence for that Franking Equipment.

Franking Equipment must always show the correct amount of Postage and Fees which the User has prepaid or has agreed to pay.

The value of Postage paid, shown by the Franking Mark on any Letter, must match the applicable rate set out in the relevant Scheme. If Postage on any Letter is found to be underpaid for the selected service, the Letter will be sent by the selected service and, except to the extent that the provisions of another Scheme apply, the User shall be charged for the difference between the payment franked and the correct amount for that service, together with such administrative charge as Royal Mail may decide from time to time. Together the amount of the underpayment and the administrative charge are referred to as the “Underpayment Surcharge”.

Appendix 2

Franking assumptions