Case No: C/1999/0654
IN THE SUPREME COURT OF JUDICATURE
COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE VAT & DUTIES TRIBUNAL
Royal Courts of Justice
Strand, London, WC2A 2LL
Date: 13 July 2000
THE
COMMISSIONER FOR CUSTOMS |
Applicant | |
- and - |
||
FDR |
Respondent |
"B. Other exemptions
Without prejudice to other Community provisions, Member States shall exempt
the following under conditions which they shall lay down for the purpose of
ensuring the correct and straightforward application of the exemptions and of
preventing any possible evasion, avoidance or abuse ...(d) the following
transactions:
(1) the granting and negotiation of credit and the management of credit by
the person granting it...
(3) transactions, including negotiation, concerning deposit and current
accounts, payments, transfers, debts, cheques and other negotiable instruments,
but excluding debt collection and factoring;
(4) transactions, including negotiation, concerning currency, bank notes and
coins used as legal tender, with the exception of collectors' items...
(5) transactions, including negotiation, excluding management and
safekeeping, in shares, interests in companies or associations, debentures and
other securities...
(6) management of special investment funds as defined by Member
States."
FDR's case is that the supplies made by them in the course of their business,
at any rate what has been called the "core" or "principal" supply, are exempt
by force of Article 13B(d)(3).
THE FACTS
3 The Tribunal provided a thumbnail summary of the nature of FDR's business,
which it is useful to set out at once. It is in these terms:
"FDR's business consists of providing services which fulfil the obligations
of banks which issue payment cards to cardholders and of banks contracting with
merchants (normally retailers) to acquire vouchers accepted by those merchants
in payment for goods or services, in effect acting as a clearing house for card
transactions involving its clients whether issuers or acquirers or both, and
effecting settlement of the net positions of those clients and supplying
connected services to them." (paragraph 32 of the decision)
4 But that is the barest outline. It is necessary to set out the facts in much
more detail; a good deal of the hearing before us, and yet more time before the
Tribunal, was taken up with their investigation and exposition, and much of the
Tribunal's decision is devoted to their description. However for my part (and
I acknowledge the fault may be mine) I have not found the passages in question
as clear as I would wish; though I have certainly found the distillation at
paragraph 35 to be helpful, compressed as it is. That passage, as the Tribunal
themselves indicate, was in effect a reproduction of a document prepared by Mr
Paines QC for the Commissioners, which was agreed to with amendments by Mr
Casey, FDR's tax manager, in the course of his cross-examination. I will not
with respect set it out, since, if only so that I may be confident that I have
understood them, I propose to describe the facts in my own words, drawing on
the Tribunal's decision, the parties' skeleton arguments and submissions in
court, and some of the evidence. There is no dispute as to any matter of
primary fact.
5 I have said that the appeal involves the VAT treatment of credit card
services provided by FDR to banks. But for clarity's sake it is first
convenient to describe the relevant credit card services which banks oblige
themselves to provide to their customers, leaving FDR's involvement on one
side. These are of two relevant kinds. The holder of a credit card, which has
been issued to him by a bank, may use it to purchase goods or services. Or he
may hold a card which he uses to draw cash: typically, nowadays, from an
automatic cash-dispensing machine or "hole in the wall", into which he inserts
the card and types in a PIN number which has been notified to him. In many
cases the same card may be used for both purposes. There is another kind of
card, called a debit card or (I think less confusingly) an account card. This
is issued to the holder by the bank where he maintains his ordinary bank
account, and bears his bank account number. It may be used to guarantee any
cheque he draws (up to a given value) or to draw cash from a hole in the wall,
in which case, again, the user will have a PIN number[1]. FDR, whose role of course I have yet to explain, are
involved in all of these kinds of services save that where the holder uses his
account card to support a cheque. It is agreed between counsel that nothing
turns for the purposes of the appeal upon the difference between the use of a
card to obtain goods or services on the one hand and to obtain cash on the
other, or on the distinction between a credit card and an account card. For
the sake of relative simplicity and brevity, I shall therefore in large measure
limit the description in what follows to the steps by which the use of a credit
card to purchase goods or services is put into effect, and as I have
foreshadowed I shall first deal with the relatively straightforward case in
which FDR is not involved at all.
The Simple Case: FDR not involved
6 As I shall describe it this is, I should make clear, a notional case as a
matter of fact. The credit card is issued to the holder by a bank. The
issuing bank, in the vocabulary of the credit card world, is called an
"Issuer". Another bank, or it may be the same one, has a contract, whose
effect I will explain shortly, with a shop or other retail outlet, to whom I
will refer as the "Merchant". This bank, or if it is the same bank as the
Issuer then to the extent that it acts by virtue of its contract with the
Merchant, is called an "Acquirer". The holder of the credit card goes into the
shop. She wants to buy a roll of curtain material, and to use her credit card
to do so. She hands the card to the Merchant's shopkeeper. He records the
transaction, either manually by means of a voucher on which are stamped the
credit card details and the shop's own details and which is signed by the
customer, or (more often nowadays) electronically by swiping the card through a
machine which produces a chit for the customer to sign. Before the transaction
is finalised the Merchant may be required (as a term of his contract with the
Acquirer) to obtain the Issuer's authorisation, if the sale is at a price over
a limit agreed between Merchant and Acquirer (the "floor limit"). The purposes
of this process are to check whether the card is on a warning or stop list
showing that it is or may be stolen, and whether the cardholder's credit limit
is or will be exceeded. If the transaction is being done manually, the
authorisation will be sought by a telephone call to the Issuer; if
electronically, it is effected by an exchange of messages initiated by the
swipe of the card through the machine. When that has been done, and the
customer has signed the chit or voucher and received the roll of curtain
material, the contract of sale between customer and Merchant is complete: she
has the goods, and has discharged her obligation to pay the price by proffering
her credit card which is processed as I have described.
7 But the Merchant has not, of course, so far received any money. He is paid
by the Acquirer. The essence of the latter's contract with the Merchant is
that the Acquirer agrees to pay the price of the curtain material in return for
a commission. The sale in the shop is notified to the Acquirer. In the case
of a manual transaction that is done by the Merchant taking the signed voucher
to the Acquirer's local High Street branch, which effects payment, against the
voucher, into the Merchant's bank account (often at the same branch). In the
case of an electronic transaction, the swipe of the card through the machine
notifies the transaction to the Acquirer's bank which again effects payment to
the Merchant's account.
8 So the Merchant has now been paid. But the Acquirer is pro tanto out
of pocket. The commission he receives is, of course, his profit. He needs to
recover the price of the curtain material, which he has paid out to the
Merchant. He recovers it from the Issuer. If Acquirer and Issuer are the same
bank, this would no doubt be effected by parallel entries in two accounts held
in the respective capacities of Acquirer and Issuer: the Acquirer account
showing a credit, the Issuer account showing a debit. Where the Acquirer and
Issuer are different banks, there is a payment or transfer of the relevant
amount by Issuer to Acquirer.
9 At this point, of course, the Issuer is out of pocket. He is paid by the
cardholder. Typically that is done every month, on presentation by the Issuer
to the cardholder of an account which sets out all the items of expenditure
executed on the credit card within the month past, and any credit items
occurring in the same period - usually the cardholder's payment to meet all or
part of his previous month's bill.
10 Thus the circle is completed. There have been three particular money
transactions or transfers: Acquirer to Merchant, Issuer to Acquirer, cardholder
to Issuer. That is the simple, notional case, with no involvement on FDR's
part.
The Actual Case: FDR involved
11 FDR, which is not itself a bank, is engaged when some or all of the banks'
services, performed as Issuer or Acquirer or both, are "outsourced" to FDR, to
use the barbarous expression apparently current in the trade. There were two
FDR contracts put in evidence before the Tribunal. Under the "Issuer and
Acquirer Agreement", FDR provides services to its client bank in the latter's
dual capacities as Issuer and Acquirer. FDR has four clients who operate in
the payment card market as both Issuers and Acquirers and a further 24 clients
who operate as Issuers only. In respect of these latter, FDR enters into the
other contract which was put in evidence.
12 The essence of FDR's case, as Mr Cordara QC advanced it before the Tribunal
and in this court, is that the services which the Issuer and Acquirer provide
obviously fall within the description given in Article 13B(d)(3):
"transactions, including negotiation, concerning deposit and current
accounts, payments, transfers, debts...", and therefore, since under the
Issuer and Acquirer Agreement FDR do the selfsame things, the services which it
provides must fall to be treated as exempt supplies under the Article. Mr
Paines submitted that this proposed read-across from a state of affairs in
which the banks do all the relevant transactions themselves to one in which FDR
act for them demands some little caution, if only because VAT is (elementarily)
a tax on supplies made by A to B: and in the first state of affairs (where the
banks do everything themselves) some relevant acts or steps may not be supplies
at all because the bank will, as it were, be supplying to itself. But I will
not at this stage anticipate the legal analysis which falls to be carried out
in relation to FDR's activities; what FDR actually does has first to be
carefully described. The account which follows concerns FDR's activities in a
case where the credit card is used for retail purchases, FDR acts both for
Issuer and Acquirer, and Issuer and Acquirer are separate banks. I have taken
that state of affairs purely for clarity's sake. In fact the transactions in
which FDR acts for both Issuer and Acquirer are in a minority; and the
Issuer and Acquirer Agreement before the court is geared to the case where
Issuer and Acquirer are the same. However nothing turns on these differences,
and the correct application of the law to the facts most clearly appears upon
the factual model which I have identified and will now describe.
13 In such a case FDR maintains two accounts, the cardholder account and the
Merchant account, whose functions I will explain shortly. (There was much
controversy in the course of argument as to their status, and I will come to
that in due course.) It is convenient first to return to the purchase of the
roll of curtain material, to see how far the situation differs from one where
FDR is not involved. Where the Merchant executes the sale transaction
manually, receiving the signed voucher from the cardholder, there is at the
very first stage of the cycle of payments no difference: the Merchant takes the
voucher to the High Street branch of the Acquirer's bank, by whom he is paid
without intervention by FDR. However FDR is notified of the payment and then
does two things. It posts the transaction to the Merchant account - as Mr
Paines put it: that is, it makes an entry in that account to the credit of the
Merchant in the relevant sum. And it causes a debit to be entered at the
Acquirer bank's head office, and a corresponding credit at the High Street
branch, so that the payment to the Merchant is, as it were, logged centrally.
It will be appreciated that the Acquirer will have contracts with very many
Merchants, with each of whom there may be very many transactions. All payments
out to the Merchants must no doubt for efficiency's sake be shown in a central
account.
14 The means by which FDR causes the parallel entries to be made in the local
and central branches of the Acquirer involve the use of what are called BACS
tapes. BACS is an automated clearing house. How it works is significant for
the purpose of seeing how or whether Art.13B(d)(3) applies to the facts in this
appeal: as I shall show, FDR also engages BACS at other important stages in the
cycle of transactions in the case I am describing, and one of Mr Paines'
central submissions is to the effect that when FDR "instructs" (as Mr Paines
puts it) BACS to effect a transfer of funds, that is not a transfer by
FDR for the purposes of Art.13B(d)(3). But I shall postpone the
description I must give of BACS, as well as my treatment of Mr Paines'
submission. It will be much clearer if at this stage I return to the curtain
material.
15 Where the transaction in the shop is done electronically - that is, where
the Merchant's shopkeeper swipes the card through a machine which produces a
chit for the cardholder to sign - the position is as follows. The technology
is such that the shopkeeper's swiping the card (a) notifies FDR of the
transaction, and (b) seeks FDR's authorisation of the transaction. The message
sent by the swipe does not go to the Acquirer bank at all; it goes only to FDR.
FDR, then, has the function of authorising the transaction, which is effected
by an exchange of automated messages through the machine. FDR no doubt
obtains, or more likely already possesses, the necessary information from the
Issuer; in any event given the numbers of transactions involved the movement of
any such information is no doubt by electronic means. If the sale is then
authorised, FDR posts the transaction to the Merchant's account. In addition
FDR also provides for (I intend this expression as an entirely neutral form of
words, given Mr Paines' argument as to BACS) BACS to effect a credit in the
Merchant's own bank account, there being created a corresponding debit in the
Acquirer's central accounts. Thus the Merchant is paid by the Acquirer. In
the case where the transaction in the shop is effected electronically, the
Acquirer, as opposed to FDR, has not had to do anything at all to arrive at
that position.
16 The next stage involves payment by the Issuer to the Acquirer, just as in
the notional case where FDR is not involved. In order to describe it properly
it is necessary to say farewell to the customer in the shop buying the roll of
curtain material, and to confront the reality, which is of course that there
are thousands upon thousands such transactions every day. FDR enters, on the
books that it maintains electronically, every claim of each of its Acquirer
clients against an Issuer arising out of a credit card transaction, and every
liability of an Issuer client to pay an Acquirer. There are also situations
called "chargebacks", whose genesis it is unnecessary to describe, whereby an
Issuer may have a claim against an Acquirer and vice versa. FDR
proceeds to establish the net position of each client bank (Issuers and
Acquirers) and the net amount needing to be transferred from or to that bank as
the case may be. In the evidence and argument this has been called
"netting-off", and is carried out on a "pooled" basis: that is, FDR acts as
clearing house: compare the role of IATA in British Eagle v Air France
[1975] 1 WLR 758, to which the Tribunal was referred. That having been
done, FDR makes a payment out of its own funds to each client bank which is a
net claimant, and receives (later the same day) a payment from each client bank
which is a net debtor. This is effected through the banking system by a
mechanism called CHAPS. It is unnecessary to describe CHAPS because Mr Paines
accepts that the payments made by FDR by means of it constitute, if viewed in
isolation (and thereby hangs a different tale), "transfers" within the meaning
of Art.13B(d)(3). Subject to error, the payments and receipts by FDR will be in
balance. All this is done every working day.
17 Thus the accounts between Acquirers and Issuers are reconciled. The last
stage in the cycle consists in the payment by cardholder to Issuer. FDR posts
the debit to the cardholder account (to which it also posts credit entries, as
where the cardholder pays his or her monthly bill). And where the cardholder
pays by direct debit, FDR also provides for BACS to debit the cardholder's
ordinary bank account, and credit the Issuer's account, with the relevant sum.
Where there is no direct debit arrangement, FDR's role at this stage is limited
to updating the cardholder account (for which it has all necessary information
from the Issuer).
18 That completes the cycle of payments. But I should give some further
details of FDR's activities undertaken in the course of performing their
contracts with Issuer and Acquirer in the case I have been describing. First,
the credit card itself: for some but not all of its Issuer clients FDR arranges
for the card to be embossed with the cardholder's name and account number.
Then FDR prepare and send to the cardholder periodic (typically monthly)
statements of his indebtedness to the Issuer: that is, of course, a statement
of the balance on the cardholder account. Enclosed with the mailing will be any
promotional leaflets, circulars and so forth required to be included by the
Issuer.
19 Then at the end of each month FDR will calculate the aggregate fee which the
Merchant owes to the Acquirer by way of commission on transactions, and send a
statement of the Merchant account to the Merchant. Ten 10 days after that they
will make an entry in a BACS tape to effect an appropriate debit to the
Merchant's bank account in favour of the Acquirer.
Other Matters of Fact
(1) Non-FDR Clients
20 Though I have chosen to describe the instance where Issuer and Acquirer
are different banks but both clients of FDR, I should make it clear that on the
evidence before the Tribunal that was by no means the commonest state of
affairs. FDR had at the material time four clients who were both Issuers and
Acquirers and a further 24 who were Issuers only. Mr Casey told the Tribunal
that in March 1998 FDR's market share was 45% of the Issuer market and 27% of
the Acquirer market. The Tribunal, on the basis of certain assumptions,
concluded that of the overall Acquirer market 12% was constituted by
transactions involving an FDR Issuer and an FDR Acquirer, 33% by transactions
involving an FDR Issuer only, 15% by transactions involving an FDR Acquirer
only, and 40% by transactions involving neither an FDR Issuer nor an FDR
Acquirer (decision, paragraph 38).
21 Where an Issuer or Acquirer is not a client of FDR, it seems that it will
belong to one of what are called the "payment systems" - MasterCard, Europay,
Visa and Barclaycard. In that case there will be an account between the
payment system and its Acquirer/Issuer client. FDR will only be involved if it
acts for the Issuer where the Acquirer is a client of the payment system or
vice versa. The principal effect on what is actually done by FDR
relates to the netting-off process. This is how the Tribunal described that
process in paragraph 55 of the decision:
"...sums due to FDR Acquirers from a payments system are set against sums due
from FDR issuers to the same system. This is not undertaken on a client by
client basis but on a global payment system by payment system basis. FDR pays
its Acquirer banks before this process is completed, recouping itself from the
sums received on behalf of those clients from the payments systems. This type
of set-off has a triple effect. It discharges the liability of FDR's Issuer
client to the payment system, it discharges the liability of the payments
system to FDR's Acquirer client and it reduces the amount of FDR's own money
tied up in that system by reducing the CHAPS flow in each direction. The net
payment from each system which is a net payer on each working day is received
as a global payment by FDR on behalf of its various clients and the payments to
each system which is a net recipient on each working day are made as a global
amount..."
22 It is unnecessary to understand the operation of the payment systems in any
greater detail, since, just as it is agreed between counsel that nothing turns
for the purposes of the appeal upon the difference between the use of a card to
obtain goods or services on the one hand and to obtain cash on the other, or on
the distinction between a credit card and an account card, so also it is common
ground that the presence or absence of the payment systems in part of the cycle
of payments can make no difference to the tax treatment of FDR's activities.
In short, therefore, it is irrelevant whether FDR acts for Issuer, Acquirer or
(as in the case I have described) both. I should also notice that in the
notional case which I first set out, that in which FDR acts neither for Issuer
nor Acquirer, the reality would presumably be that both of them would be in
account with a payment system.
(2) BACS
23 As the Tribunal stated (paragraph 67) the BACS procedure is explained at
paragraphs 548 to 565 of the Encyclopaedia of Banking Law. At 548 this
appears:
"BACS Ltd is an automated clearing house which provides bulk electronic
clearing for credit and debit transfers such as standing orders, direct credits
and direct debits. Typically, BACS deals with high volume, but low value,
transfers of funds. It is commonly used for the payment of monthly salaries
and the collection of regular payments, eg utility bills, insurance premiums
and mortage payments.
There are currentlt 15 BACS members [April 2000: all the members are banks].
Each member has direct access to the BACS system and will supply BACS with
credit and debit instructions as computer data (`input data') for
processing..."
24 In paragraph 67 the Tribunal accurately summarised the account given in the
Encyclopaedia thus:
"... BACS receives the input data by 9 pm on the first day of a 3 day cycle,
processes it overnight by 6 am on day two by which time it despatches credit
and debit instructions to each member by telecommunications link or courier.
Every instruction to credit an account is accompanied by an instruction to make
a corresponding debit to the account from which payment is made. Individual
magnetic tapes or discs are produced for each member, the members being banks
or building societies. Once inter-bank clearing is complete the members'
overall balances are adjusted accordingly at the Bank of England."
I do not understand it to be in controversy that once the input data are
supplied by the appropriate means, the process thereafter is wholly automatic,
and results in an actual transfer of money between accounts. And I should
notice that supplies made by BACS are treated by the Commissioners as exempt,
as I understand it, under the terms of Art.13B(d)(3).
25 At paragraph 68 of the decision the Tribunal said this:
"FDR's tapes for a day will supply BACS with instructions to debit the account
of each Acquirer and credit the accounts of its Merchant customers. BACS in
effect breaks the instructions down to credit instructions for a large number
of separate Merchant's accounts presumably consolidating these with
instructions from other sources in respect of those accounts."
THE ISSUES
26 In very considerable measure the arguments advanced before us have been
concentrated upon three questions. The first is: on the facts, do FDR make
"transfers" within Art.13B(d)3)? The second is: given that on any view FDR
carry on a number of different activities in the execution of their contractual
obligations to Issuers and Acquirers, should at least some of those activities
be treated as a "core" or "principal" supply, and thus a single supply (so as
to attract a unitary tax treatment for the purposes of VAT even though the same
activities, if treated individually, would or might attract differing tax
treatments), and, if so, how should the core supply be described? The third
is: if there is a core supply, what should its tax treatment be - taxable or
exempt?
27 As it happens both parties contend that there is a core supply (question
2), but differ fundamentally as to how its content should be categorised and,
therefore, what its tax treatment should be (question 3). Mr Paines says it is
in the nature of book-keeping, which is no more an exempt supply than are
accountancy services in general. Mr Cordara submits that it is as the Tribunal
held it to be in paragraph 163, where they described it as FDR's "principal
service":
"In relation to both Issuers and Acquirers we find that the principal
service provided by FDR consists of processing all their card transactions and
settling their liabilities and claims under these transactions in accordance
with the obligations of the Issuers and Acquirers."
Mr Cordara would say that this articulation of the core supply was correct and
inevitably led to the conclusion that it fell to be treated as an exempt supply
under Art.13B(d)(3).
28 In my view the first of these three questions, as I have set them out, also
falls logically to be addressed first. The reason is that the Tribunal held
(a) (as I read the decision) that the daily "netting-off" procedure itself
amounted to the making of transfers and so fell within Art.13B(d)(3)
(paragraphs 178, 181, 183, 184); (b) that the "issue of instructions" by FDR to
BACS amounted to a transfer, with the same consequence (186, 190, 192, 196,
201); and (c) that the cardholder (185) and Merchant (195) accounts were
current accounts within Art.13B(d)(3), so that, for example, payment from an
ordinary cheque account to a cardholder account would constitute a transfer
(185). These findings are as I see it independent of and logically prior to
the Tribunal's view (or, therefore, anyone's view) of the nature and extent of
a "core" supply and if they are wrong, the whole basis of the Tribunal's
conclusion in FDR's favour would in my judgment be undermined. The
Commissioners' appeal would in that event fall to be allowed (I leave aside the
question what relief might be appropriate), whatever position one might take
as to the existence or description of a core or principal supply. Putting the
matter much more shortly, the tax treatment of such a core supply could not
possibly be assessed properly without a prior decision upon the question
whether FDR makes transfers; and that was question (1).
Issue (1): Do FDR Make Transfers within Article 13B(d)(3), and if so, by
What Transactions are they Constituted?
29 In fact Mr Paines accepts, as plainly he must, that the CHAPS payments made
by FDR to their client banks or to the payment systems, after completion of the
netting-off procedure, amount to transfers within the Directive. Viewed on
their own, therefore, they are admittedly exempt supplies. Mr Paines'
submission about them is that they should be taken to be "ancillary" (an
expression to which I shall have to return) to the core supply of book-keeping
services, and as such fall to receive the same tax treatment and thus be
regarded as part of a taxable supply; alternatively, if they are to be viewed
in isolation, they may be treated as exempt supplies. But in that case Mr
Paines would stoutly insist that everything else done by FDR is or is part of a
taxable supply. In particular, FDR's using or creating tapes to effect an
input to BACS is or is part of a taxable supply, and so is everything it does
in relation to the netting-off procedure, and in relation to the maintenance of
the cardholder and Merchant account. None of these activities, says Mr Paines,
constitutes a transfer within Art.13B(d)(3).
SDC
30 I shall consider separately whether FDR make transfers by virtue of the BACS
entries, the netting-off procedure, or their operation of the cardholder and
Merchant accounts. But first I must pay attention to a leading decision of the
Court of Justice upon which Mr Paines' argument about transfers very heavily
depends. Sparekassernes Datacenter (SDC) v Skatteministeriet [1997] ECR I-3017 ("SDC") was a Danish Article 177 reference concerning the tax
treatment of the activities of an organisation, SDC, which was described
throughout the Court's judgment as a "data handling centre". Having re-read
SDC in the light of counsel's arguments, I accept Mr Cordara's
submission that the company's activities were similar to those of FDR, although
(but this would if anything tell in Mr Cordara's favour) it did not at any
stage make payments or transfers of its own money. It was not itself a credit
establishment or financial institution, but a company (registered for VAT)
owned by the Danish Association of Savings Banks, to whose members its services
were for the most part provided. In paragraph 17 of the Opinion of Mr Advocate
General Ruiz-Jarabo Colomer the "four spheres of activity" in which its
services were performed are summarised as being credit transfers, advice on
management of securities, management of deposits, purchase contracts and loans,
and tasks concerning members' internal administration. The services themselves
are thus described in paragraph 16:
"- the plaintiff performs the services only upon request from a member
savings bank, one of such bank's customers or others who, by agreement
with the customer, are authorized to requisition, for example, specific
payments;
- the request is made by the electronic transmission of information which
may result in the immediate performance of a service or involve several
successive services over a shorter or longer period;
- a customer can only transmit information after authorization by the
financial institution, for example in the form of the issue of a cash card or a
credit card;
- the plaintiff's name is not made known to the individual savings bank
customers and the plaintiff has not entered into legal obligation towards them;
- the plaintiff does not demand payment from the individual customers but
only from its members;
- the plaintiff's services are essentially performed wholly or in part
electronically".
31 The reference from the municipal court engaged Art.13B(d)(3) and (5) of the
Sixth Directive. In considering what was SDC's correct tax treatment, it is
plain that the Advocate General attached crucial significance to the fact that
SDC had no legal links with any of the banks' customers (see for example
paragraphs 49 and 50 of his Opinion). That was the basis, or at least a
principal basis, for his view that the supplies made by SDC were not exempt (as
appears from the answer he would have given to the referring court's second
question: p.I-3039). The Court of Justice did not agree; however that did not
confer outright success on SDC, since, unusually, the Court sent the case back
for the municipal court to find further facts. We were told that the case was
then settled between SDC and the Danish tax authorities, on terms which were at
least relatively favourable to SDC; though I agree with Mr Paines that that
circumstance is of no relevance to anything we have to decide.
32 Turning to the Court's judgment, I should first notice paragraph 20, where
it was pointed out that
"the terms used to describe the exemptions envisaged by art 13 of the Sixth
Directive are to be interpreted strictly since these constitute exceptions to
the general principle that turnover tax is to be levied on all services
supplied for consideration by a taxable person";
and paragraph 22, which is in these terms:
"Finally, a comparison of the various language versions of art 13B(d)(3)
reveals that there are differences in terminology with regard to the phrase
`transactions ... concerning'. In view of those linguistic differences, the
scope of the phrase cannot be determined on the basis of an interpretation
which is exclusively textual. In order to clarify its meaning, reference must
therefore be made to the context in which the phrase occurs and consideration
given to the structure of the Sixth Directive (see the judgement in
Skatteministeriet v Henriksen Case 173/88 [1990] STC 768,[1989] ECR 2763
(paras 10-11))".
Then paragraph 37 is of some importance:
"It must be stated in regard to this point that the specific manner in which
the service is performed, electronically, automatically or manually, does not
affect the application of the exemption. The provisions in question make no
distinction in this regard. Accordingly, the mere fact that a service is
performed entirely by electronic means does not in itself prevent the exemption
from applying to that service. If, on the other hand, the service entails only
technical and electronic assistance to the person performing the essential,
specific functions for the transactions covered by art 13B(d)(3) and (5), it
does not fulfil the conditions for exemption. That conclusion follows,
however, from the nature of the service and not from the way in which it is
performed."
33 Mr Paines lays particular emphasis on paragraphs 53 and 66 (whose context
requires me also to set out paragraphs 61 - 65). Paragraph 53 appears in a
section headed:
"The contractual links between the person performing the service and the
person receiving it",
and is in these terms:
"On this point, it must be noted first of all that a transfer is a
transaction consisting of the execution of an order for the transfer of a sum
of money from one bank account to another. It is characterised in particular
by the fact that it involves a change in the legal and financial situation
existing between the person giving the order and the recipient and between
those parties and their respective banks and, in some cases, between the banks.
Moreover, the transaction which produces this change is solely the transfer of
funds between accounts, irrespective of its cause. Thus, a transfer being only
a means of transmitting funds, the functional aspects are decisive for the
purpose of determining whether a transaction constitutes a transfer for the
purposes of the Sixth Directive."
Paragraphs 61 - 66:
"Transfers and payments
61. It is necessary to consider first of all whether the operations carried
out by a data-handling centre such as SDC in the effecting of a transfer can in
themselves be described as transactions concerning transfers within the meaning
of art 13B(d)(3) of the Sixth Directive.
62. The Danish Ministry for Fiscal Affairs argues that the services provided
by SDC are in fact composed of various administrative or technical components
which are invoiced individually. No price is fixed in advance for the
transfer, the transfer of funds or the services in their entirety.
Consequently, the services provided by SDC are different from those covered by
art 13B(d)(3) of the Sixth Directive.
63. SDC, on the other hand, states that, in order for the exemption to
apply, it is not necessary for the services supplied to be complete services
but it is sufficient that the supply in question should be an element of a
financial service in which various operators participate and which, taken as a
whole, constitutes a complete financial service.
64. Given this difference of view, it must be noted first of all that the
wording of art 13B(d)(3) does not in principle preclude a transfer from being
broken down into separate services which then constitute `transactions
concerning transfers' within the meaning of that provision and which are
invoiced by specifying the elements of those services. The invoicing is
irrelevant for the application of the exemption in question, provided that the
actions necessary for effecting the exempt transaction can be identified in
relation to the other services.
65. However, since art 13B(d)(3) must be interpreted strictly, the mere fact
that a constituent element is essential (indispensable in the French
text) for completing an exempt transaction does not warrant the conclusion that
the service which that element represents is exempt. The interpretation put
forward by SDC cannot therefore be accepted.
66. In order to be characterised as exempt transactions for the purposes of
art 13B(d)(3) and (5), the services provided by a data-handling centre must,
viewed broadly (apprecie de facon globale in the French), form a
distinct whole, fulfilling in effect the specific, essential functions of a
service described in those two points. For `a transaction concerning
transfers', the services provided must therefore have the effect of
transferring funds and entail changes in the legal and financial situation. A
service exempt under the directive must be distinguished from a mere physical
or technical supply, such as making a data-handling system available to a bank.
In this regard, the national court must examine in particular the extent of the
data-handling centre's responsibility vis-à-vis the banks, in particular
the question whether its responsibility is restricted to technical aspects or
whether it extends to the specific, essential aspects of the
transactions."
34 Mr Paines submits that a "transfer" is constituted by the execution
of an instruction that the transfer should take place, and never merely by
the instruction itself: paragraph 53. In line with this is his
submission appearing at paragraph 28 of his skeleton argument:
"...the distinction drawn by the ECJ in paragraphs 65 and 66 of the
judgment is between a service which is indispensable for the performance of an
exempt supply by another (which is insufficient for exemption) and a
service which itself contains the essential elements of an exempt supply
defined in article 13B(d) and thus is an exempt supply. It is only the
latter service which qualifies for exemption. In particular, a `transaction
concerning transfers' is one that has the effect of transferring funds."
Mr Paines was concerned to emphasise that there may be many commercial and
professional services which on the face of it seem well within some or other
part of the Art.13B(d)(3) rubric: thus general accountancy services such as
negotiation on a client's behalf with the Inland Revenue would, as a matter of
ordinary language, readily fall within "transactions, including negotiation,
concerning... payments, transfers"; but it is beyond contest that such
services are not exempt. Something altogether more intimate to the actual
process of moving money is required.
35 In general terms, I agree with this. It is plain that ordinary accountancy
services are not exempt from VAT, and that the exemptions granted by the
provisions contained in Art.13B(d) are much more narrowly confined. It is well
recognised that commercial transactions whose essence involves the movement of
money are in many cases, for conceptual reasons, ill-suited for the application
of the VAT regime, and it seems likely that this is what lies behind the
Art.13B(d) exemptions. Mr Paines was in my judgment right to submit that while
the Court's reasoning in SDC relating specifically to "transfers"
implies a narrow approach to the exemption's reach, it would be no less
inappropriate to open the statutory exemptions to services which are distant
from the actual movement of money merely by reference to other words in the
provision, such as "transactions, including negotiation, concerning deposit
and current accounts... debts, cheques and other negotiable instruments".
In particular I think he was right to submit that FDR's instructions to BACS
would not constitute a transaction concerning a current account: if FDR does
not effect transfers through BACS, it enjoys no other route (vis-a-vis
BACS) to exemption under Art.13B(d)(3). But all this, I conceive, is no
more nor less than the consequence of the well-established requirement to read
the statutory exemptions strictly; and I do not suppose that Mr Cordara would
disagree. I must test what FDR actually do against the reasoning in
SDC, and do so in three areas, (a) transfers and BACS, (b) transfers and
netting-off, and (c) transfers and the cardholder/Merchant accounts.
Transfers in Domestic Law
36 First, however, there is something to be said about the very meaning of a
"transfer" of money, a concept not defined in the Directive. It is, even
nowadays, not difficult to be beguiled by the old model of a transfer in
specie, when money in the shape of tangible coin was moved from one place,
and one owner, to another place and another owner. We were shown two cases
decided in the Commercial Court which describe the modern reality. The first
was Momm [1977] QB 790. A credit entry had been made in the plaintiffs'
bank account on 26 June. The next day it was purportedly reversed in
circumstances which for present purposes I need not describe. At 799G ff Kerr
J as he then was said:
"The issue is whether or not a completed payment had been made by the
defendants to the plaintiffs on June 26... If there were no authorities on
this point, I think that the reaction, both of a lawyer and a banker, would be
to answer this question in the affirmative. I think that both would say two
things. First, that in such circumstances a payment has been made if the
payee's account is credited with the payment at the close of business on the
value date, at any rate if it was credited intentionally and in good faith and
not by error or fraud. Secondly, I think that they would say that if a payment
requires to be made on a certain day by debiting a payor customer's account and
crediting a payee's customer's account, then the position at the end of that
day in fact and in law must be that this has either happened or not happened,
but that the position cannot be left in the air. In my view both these
propositions are correct in law.
Kerr J proceeded to refer to Eyles v Ellis (1827) 4 Bing. 112. There
Best CJ, delivering the judgment of the court, said (114):
"... on the 8th a sum was actually placed to the plaintiff's account; and
though no money was transferred in specie, that was an acknowledgement from the
bankers that they had received the amount from Ellis. The plaintiff might then
have drawn for it, and the bankers could not have refused his draft."
The case remains good law, as Kerr J made clear; and he said at 800G in
Momm:
"The important feature of the case for present purposes is that the payment
was held to be complete when the payee's bank account was credited and before
the payee had had any notice that this had happened."
The other case decided in the Commercial Court is Libyan Arab Foreign Bank
[1989] QB 728. Staughton J as he then was said this at 750E-H:
"Any account transfer must ultimately be achieved by means of two accounts
held by different beneficiaries with the same institution. In a simple case
the beneficiaries can be the immediate parties to the transfer. If Bankers
Trust held an account with the A bank which was in credit to the extent of at
least $131m., and the Libyan Bank also held an account at the A bank, it would
require only book entries to achieve an account transfer. But still no
property is actually transferred. The obligation of Bankers Trust is
extinguished, and the obligation of A bank to Bankers Trust extinguished or
reduced; the obligation of A bank to the Libyan Bank is increased by the like
amount.
On occasion a method of account transfer which is even simpler may be used.
If X Ltd. also hold an account with Bankers Trust London, and the Libyan Bank
desire to benefit X Ltd., they instruct Bankers Trust to transfer $131m. to the
account of X Ltd. The obligation of Bankers Trust to the Libyan Bank is
extinguished once they decide to comply with the instruction, and their
obligation to X Ltd. is increased by the like amount. That method of account
transfer featured in Momm [1977] QB 790.
In a complex transaction at the other end of the scale there may be more
than one tier of intermediaries, ending with a Federal Reserve Bank in the
United States..."
37 The value of these statements (which have, according to counsel's
researches, never been doubted) is that they show that, if one leaves aside
transfers in specie (of coin, goods or other property), a transfer of
money means no more nor less than the entry of a credit in the payee's
account and the entry of a corresponding debit in the payor's account. There
may be - will be - problems in cases of error or fraud in the posting of
entries to the accounts. But however those may fall to be resolved, there is no
further, elusive, event by which the money is really transferred: no
Platonic Form, of which day-to-day transfers are only shadows. The pro
and con entries constitute the transfer. There is nothing
else. I recognise, of course, that this reasoning boils down the reality to
the simplest case. In truth, creditor and debtor may have accounts at banks A
and B respectively; banks A and B may themselves have accounts at banks C and D
respectively; and it may be only when one comes to banks J and K that one finds
both of them having accounts at the Bank of England. But the logic is
unaffected.
38 If this reasoning is right it is, I think, very significant for a sensible
and intelligent understanding of SDC. It demonstrates that what the
Directive imports by the term "transfer" inheres in the notion of a "change
in the legal and financial situation" - an expression used in both
paragraphs 53 and 66 - where that is a reference to the effects of the
corresponding credit and debit entries in the accounts of the paying and
receiving parties. This is a point which in my judgment possesses particular
resonance when one comes to counsel's submissions relating to "netting-off".
39 Now I may turn to the three areas to which I referred at the end of
paragraph 34, in whose context the question of transfers by FDR falls to be
considered.
(a) Do FDR Make Transfers by Means of BACS?
40 I have already given an account of the BACS system (paragraphs 23 and 25).
There are up to four (not four in every case) stages in FDR's activities where
BACS may be deployed: (1) where the Acquirer's head office is debited, and its
High Street branch credited, with a payment made out of the local branch to the
Merchant's bank account (paragraph 14 above), (2) where (the purchase of the
curtain material in the shop having been effected electronically) a credit
entry is posted by BACS to the Merchant's bank account and a corresponding
debit entry posted to the Acquirer's account (paragraph 15); (3) where the
cardholder's ordinary bank account is debited with the amount which represents
his payment (in full or part) of his monthly bill, and the Issuer's account is
credited accordingly (paragraph 17); (4) where the Merchant's bank account is
debited, and the Acquirer's credited, with an amount to represent the
commission due from the former to the latter.
41 The Tribunal held, specifically in relation to (3) above (paragraph 186 of
the decision):
"We also hold that the issue of instructions by FDR to BACS to notify direct
debits to cardholder's banks in order to cover the amounts due to the card
account is a transfer, whether or not there is only the one bank involved. If
BACS as intermediary is making a transfer it would be illogical if the person
initiating the direct debit is not. It would be theoretically possible for FDR
to give an instruction for a direct debit to BACS which is not netted off by
BACS in any way because there is no other transaction involving that person on
that day. In such a case BACS would merely be passing on the instruction. It
would be anomalous if the activity of BACS was exempt but not that of the real
transferor."
And at paragraph 196 they stated;
"The BACS activities in respect of Acquirers are more extensive that those
involving Issuers but we see no difference in principle."
42 On this aspect of the case, it is in my judgment of the first importance to
recognise that BACS for its own part exercises no judgment or discretion
whatever. Once the relevant tape is prepared (and that is admittedly done by
FDR) and delivered to BACS, the process is, as I have said, automatic.
Moreover the inevitable outcome is a redistribution of the rights and
obligations of payor and payee - a change in the legal and financial
situation" - the very circumstances which in my judgment constitutes a
transfer of funds for the purposes of Art.13B(d)(3). As far as I can see that
result would only not be arrived at if the BACS hardware or software
were to break down, or if (assuming this were possible) FDR were to countermand
its instructions during the BACS payment cycle. In those circumstances BACS is
in my judgment merely the agency by which FDR effects transfers, in the four
situations I have identified. Any other conclusion would be contrary to the
good sense of the general law: Qui facit per alium facit per se. And I
cannot in this see the least affront to the reasoning in SDC: quite the
contrary: it is a conclusion which conforms to the letter and spirit of
Art.13B(d) as it was explained in that case.
(b) Do FDR Make Transfers by Means of the Netting-Off Process?
43 The Tribunal held (paragraph 183):
"The fact is that the daily netting off procedure involves an account being
struck of the debits and credits of each client bank. The netting off procedure
involves a credit in that daily account and in economic terms clearly involves
both a payment and a transfer. In any event on any normal use of language the
satisfaction of the Issuer's obligation to the payment system or the Acquirer
is clearly a transaction involving the debt (or creance) to which the
creditor is entitled. It would be wholly illogical if netting off fell within
point 3 when viewed from the creditor's side but not when viewed from the
debtor's."
44 The pooled arrangements by which FDR net off the mutual liabilities of
Issuers, Acquirers and payment systems were described by Mr Paines in his reply
as amounting to no more than a "calculation". Their eventuation in day-to-day
practice was "merely declaratory of what the true legal and financial situation
is". In his skeleton argument he had submitted (paragraph 38) that netting-off
"is simply the striking of an account of mutual debits and credits. It
avoids pro tanto the need to make a transfer or payment". Mr
Paines' submissions upon this point are very elegant but entirely misconceived.
They depend upon a hidden premiss, namely that for a transfer to take place
something has to happen over and above a change (to put the matter in
summary form) in the relevant parties' legal relationship constituted by
corresponding credit and debit entries in their respective bank accounts: there
has to be, in some sense, a real transfer, which must, presumably, be
different in kind from the change in parties' bank account entries. But the
premiss is false. There is, as I have explained in paragraph 36, no such extra
happening. It is not unlike the story of the Emperor's New Clothes, in which
the little boy realised that what everyone else said they saw - the Emperor's
supposed finery - was not there at all.
45 The reality is that the netting-off process achieves precisely the same
result as would be attained - unspeakably more laboriously - if, as between all
the Acquirers, Issuers and payment systems, each debt owed by any one to any
other were the subject of individual credit and debit entries in the bank
accounts of the two of them. If FDR effected such transactions, then subject
to his argument about BACS Mr Paines would as I understand it accept that FDR
indeed made transfers. But that is wholly unreal. It cannot be right that the
most inefficient way of doing X constitutes an exempt supply, but the most
efficient way of doing it constitutes a taxable supply. On this issue the
Tribunal was in my judgment entirely right.
(c) Do FDR Make Transfers by their Operation of the Cardholder and Merchant
Accounts?
46 Dealing with the cardholder's account, the Tribunal said (decision,
paragraph 185):
"We accept Mr Cordara's submission that these are current accounts within
point 3 and we consider the debits and credits to these accounts to be just as
much transactions concerning current accounts as if the accounts were normal
cheque accounts. They are accounts of the relationship between the cardholder
and the Issuing bank, and indeed in relation to card transactions are the only
such accounts. The cardholder will no doubt normally have a conventional
account with the Issuer if the issuer is a clearing bank. But it does not
follow that he will use that account to meet his liabilities under the card
account... Payment from a normal cheque account to a card account is clearly a
transfer involving a debit to the cheque account and a credit to the card
account."
As regards the Merchant account they said (195)
"Just as we consider the cardholder accounts to be current accounts within
point 3, so also we consider Merchant accounts to be current accounts. Again
the Merchant may or may not have a normal cheque account with the Acquirer. If
he does have such account, card transfers to and from the one account are
balanced by corresponding transfers from and to the other. If he does not, the
transfer to his own bank is balanced by the transfer from his Merchant
account."
47 Mr Paines submitted that these conclusions were wrong in principle. He said
that the cardholder and Merchant accounts were not accounts "postings to which
constituted a transfer of funds". Rather, they had the same status as a
person's account with a shop: each month a statement arrives; the statement is
no more nor less than a record of the state of account between shop and
customer. An account of this kind is "merely historic". When the customer
pays the bill at the end of the month by cheque drawn on his bank, the only
transfer properly so called is between his and the shop's bank account into
which the cheque is paid.
48 Mr Paines advanced a series of heady, Alice in Wonderland anomalies in
support of this argument. He submitted that if a credit entry in the
cardholder account (that is an entry which represents or constitutes a
reduction in the cardholder's indebtedness) is regarded as a transfer, then the
conclusion must be that the cardholder has received funds, which is a
nonsense: he has on the contrary paid out. Further, Mr Paines submitted in
reply that since both cardholder and Merchant will have conventional bank
accounts as well as the cardholder and Merchant accounts properly so called,
and postings to and from the conventional accounts undoubtedly constitute
transfers, then if postings to and from the cardholder and Merchant accounts
are also to be treated as transfers the result is that at every transaction the
cardholder or Merchant pays or is paid twice.
49 This last submission - double payment - in my judgment provides a clue to
the right answer to this part of the case. The truth is that to the extent
that FDR indeed effect transfers of money, they do so by the CHAPS and BACS
transactions. The former are admittedly transfers, and the latter I have found
to be so. For Mr Cordara's purposes the cardholder/Merchant accounts are icing
on the cake: or, perhaps better because it expresses the futility of the point,
the fifth wheel of the coach. There is no transfer of money in which FDR is
actually or allegedly involved which is not effected through CHAPS or
BACS. I can see that one might categorise the cardholder/Merchant accounts as
current accounts, as Mr Cordara urges, with no great offence to linguistic
usage; but in the present context it seems to me simply unnecessary and
confusing to do so.
Issue (2): Should Some or All of FDR's Activities be Treated as a Core or
Principal Supply (and thus a Single Supply) for the Purposes of VAT?
50 The notion of a core or principal supply, and indeed the conception of a
dichotomy between single and multiple supplies, are not I think themselves
creatures of the Sixth Directive (or the national legislation). They have been
developed by the courts as a technique for use in the task of ensuring the fair
and proper tax treatment, in light of the VAT legislation properly construed,
of a taxpayer's activities in cases where that would not be achieved if the
activities were viewed singly and in isolation. The process is not unlike the
deployment of the concept of an "ancillary use" in the planning cases; that is
done to achieve a fair and reasonable categorisation of the use of land where
more than one activity is being carried on, for the purpose of planning control
under the Town and Country Planning Act 1990.
51 Counsel on both sides have referred to the decision of the Court of Justice
in Card Protection Plan Ltd v CCE [1999] STC 270. Paragraphs 26 - 30 of
the Court's judgment are in these terms:
"26. By its first two questions, which should be taken together, the
national court essentially asks, with reference to a plan such as that offered
by CPP to its customers, what the appropriate criteria are for deciding, for
VAT purposes, whether a transaction which comprises several elements is to be
regarded as a single supply or as two or more distinct supplies to be assessed
separately.
27. It must be borne in mind that the question of the extent of a
transaction is of particular importance, for VAT purposes, both for identifying
the place where the services are provided and for applying the rate of tax or,
as in the present case, the exemption provisions in the Sixth Directive. In
addition, having regard to the diversity of commercial operations, it is not
possible to give exhaustive guidance on how to approach the problem correctly
in all cases.
28. However, as the Court held in Faaborg-Gelting Linien v Finanzamt
Flensburg [1996] STC 774, paragraphs 12 to 14, concerning the
classification of restaurant transactions, where the transaction in question
comprises a bundle of features and acts, regard must first be had to all the
circumstances in which that transaction takes place.
29. In this respect, taking into account, first, that it follows from
Article 2(1) of the Sixth Directive that every supply of a service must
normally be regarded as distinct and independent and, second, that a supply
which comprises a single service from an economic point of view should not be
artificially split, so as not to distort the functioning of the VAT system, the
essential features of the transaction must be ascertained in order to determine
whether the taxable person is supplying the customer, being a typical consumer,
with several distinct principal services or with a single service.
30. There is a single supply in particular in cases where one or more
elements are to be regarded as constituting the principal service, whilst one
or more elements are to be regarded, by contrast, as ancillary services which
share the tax treatment of the principal service. A service must be regarded as
ancillary to a principal service if it does not constitute for customers an aim
in itself, but a means of better enjoying the principal service supplied (see
Customs and Excise Commissioners v Madgett and Baldwin (Cases C-308/96
and C-94/97)[1998] STC 1189, paragraph 24)."
In paragraph 24 of the Court's judgment in Madgett (referred to in
paragraph 30 in Card Protection Plan) the Court had approved an
observation of Mr Advocate General Leger in the same case, at paragraph 36 of
his Opinion:
"I consider that a service is ancillary if, first, it contributes to the
proper performance of the principal service and, second, it takes up a marginal
proportion of the package price compared to the principal service. It does not
constitute an object for customers or a service sought for its own sake, but a
means of better enjoying the principal service."
52 I should also set out a passage from the judgment of Millett LJ as he then
was in CCE v Wellington Private Hospital Ltd [1997] STC 445, 462:
"The issue is not whether one element of a complex commercial transaction is
ancillary or incidental to, or even a necessary or integral part of, the whole,
but whether one element of the transaction is merely ancillary or incidental
to, or a necessary or integral part of, any other element of the transaction.
The reason why the former is the wrong question is that it leaves the real
issue unresolved; whether there is a single or a multiple supply. The proper
enquiry is whether one element of the transaction is so dominated by another
element as to lose any separate identity as a supply for fiscal purposes,
leaving the latter, the dominant element of the transaction, as the only
supply. If the elements of the transaction are not in this relationship with
each other, each remains as a supply in its own right with its own separate
fiscal consequences. In determining whether what would otherwise be two
supplies should be regarded as a single supply the court has to ask itself
whether one element is an `integral part' of the other, or is `ancillary' or
`incidental' to the other; or (in the decisions of the Court of Justice)
whether the two elements are `physically and economically dissociable'. This,
however, merely replaces one question with another. In order to answer this
further question, the court must consider `what is the true and substantial
nature of the consideration given for the payment' (see Bophuthatswana
National Commercial Corp Ltd v CCE [1993] STC 702 at 708 per Nolan LJ).
There are, however, limits to this process. Where supplies are made by
different suppliers, they cannot be fused together to make a single supply; and
it is probably only in relatively simple transactions that the reduction of
multiple to single supplies is appropriate."
53 I should with diffidence like to make one or two observations about this
learning. I am sure with very great respect that Lord Millett did not intend,
in the first four sentences of the passage I have just cited, to indicate that
in every case where multiple supplies properly fall to be treated as a single
supply for fiscal purposes there is always a single or unitary
dominant supply to which all the other supplies in question are then
regarded as ancillary. That, certainly, is one case; but there may be others
where the single supply that is arrived at for VAT purposes consists, not in
one supply to which others are ancillary, but in a bundle of supplies none of
which predominates over the others; the single supply may, as it were, be an
apex or a table-top. There is thus a difference between what is "ancillary"
and what is "integral": several supplies may be "integral" to one another, with
none predominating - the table-top - and this I think is the situation
contemplated by the phrase "physically and economically dissociable", quoted by
Lord Millett and appearing in some of the Court of Justice jurisprudence, and
by Lord Nolan's expression "the true and substantial nature of the
consideration given for the payment". The services of a hotelier (compare the
facts of the Madgett case) are perhaps an example. And whether the
single or core supply is unitary or consists in a number of supplies integral
to one another, there may be other supplies which lie outside the core but are
ancillary to it (in the sense given in paragraph 30 of the judgment in Card
Protection Plan and paragraph 36 of the Advocate General's Opinion in
Madgett); they will, of course, be subject to the same tax treatment as
the core. There may also, of course, be supplies made to the same customer by
the same supplier on the same occasion as the core and ancillary supplies are
made, but which are themselves in neither such category; they will be viewed in
isolation for the purpose of ascertaining their proper tax treatment.
54 While I hope these observations are helpful I think there is some danger of
over-elaboration and needless complexity in this field. We are not here
concerned with deep legal principle, but with the articulation of a fair and
reasonable approach to those cases where there is a question how should the
consideration given by a supplier for his reward be categorised for the
purposes of VAT, when there are multiple acts of supply involved The simpler
it is the better, so long as it is kept consistent with the doing of justice.
With respect I apprehend (but I by no means propose to lay down any rule) that
where this sort of issue arises, the first question to be asked may be couched
as Lord Nolan put it: what is "the true and substantial nature of the
consideration given for the payment". That will identify the apex or the
table-top. The second question will be whether there are other supplies which
are ancillary to the core.
55 But there is, I think, one further complication. Where the core supply is
on the table-top model - a congeries of supplies which are integral to each
other or "indissociable" - it may not be self-evident from the description of
the core supply at which the court or tribunal arrives what its tax treatment
should be. In that case, it will be necessary to look again at the elements
which comprise the core, and arrive at a decision on the facts whether,
numerically if nothing else, the taxable or exempt elements predominate.
Necessarily no such difficulty arises where the core supply is on the apex
model.
56 With those observations I turn back to the present case. The Tribunal
identified what it held to be the core supply in paragraph 163 of the decision,
whose first sentence I have already set out at paragraph 26. It is convenient
to cite the whole paragraph:
"In relation to both Issuers and Acquirers we find that the principal service
provided by FDR consists of processing all their card transactions and settling
their liabilities and claims under these transactions in accordance with the
obligations of the Issuers and Acquirers. This includes opening and maintaining
accounts, authorising transactions, ascertaining the credits and debits and
statementing. All of the above are either integral parts of the principal
supply or necessary for its performance. We consider that card embossing for
Issuers is ancillary to the principal supply having no independent aim or
serving no independent purpose, but that the insert of circulars is not
ancillary, except possibly in so far as the circulars themselves relate to card
transactions. It may be that collection of vouchers is ancillary. The extent to
which the provision of management information to FDR's clients is to be
regarded as ancillary to card processing and settlement depends on the nature
of the information; clearly if it is to identify prospects for a mail shot for
other financial services it is not ancillary. If on the other hand information
is to identify customers who are bad risks for card transactions, it may be
ancillary. We accept Mr Cordara's submission that the purpose of the core
service is the settlement of the credits and liabilities of FDR's clients. That
is the clients' basic requirement in order to enable them to perform their
obligations as Issuers `and where relevant as Acquirers."
Then I should go to paragraph 173:
"In order to consider whether the principal supply comes within the exemption,
it is necessary to set out the ambit of that supply in somewhat more detail
than in paragraph 163. It includes opening and maintaining accounts for each
cardholder, responding to authorisation requests, recording all card
transactions, calculating the total due from the Issuer each day, settling the
amount due either by CHAPS transfers or by netting it off, providing the Issuer
with details of all transactions, mailing statements to cardholders, raising
direct debits for cardholders where relevant, recording payments by cardholders
and sending cardholders' cheques to Issuers."
Paragraph 175 is of some importance:
"Clearly many of the elements listed at paragraph 173 if considered in
isolation would not come within the wording of point 3; however almost any
exempt supply can be broken down into elements which would not be exempt if
viewed in isolation. What matters is not the individual parts but the essential
features of the whole. The issue is whether viewed broadly the principal supply
forms a distinct whole which fulfils the specific, essential functions of a
service described in the statutory exemption. It is clear from paragraph 66
that the essential functions do not all have to come within the same point of
Article 13B(d) since paragraph 66 refers to a service described in both point 3
and point 5. Here we are only concerned with point 3. A fortiori, it is
sufficient if some essential functions fall under "payment", some under
"transfer", some under "debts" and so on."
The Tribunal then proceeded to examine the individual supplies comprising the
whole, and I have already dealt with those so far as any of them were asserted
to amount to transfers within Art.13B(d)(3).So far as the Tribunal expressed an
overall conclusion as to the tax treatment of the core supply, it is to be
found at paragraph 187:
"We conclude that the payment, netting off, account operating and direct debit
activities performed by FDR for Issuers all fall within point 3. In our
judgment those functions lie at the heart of the services performed by FDR."
Then at 190:
"In our judgment the basic services supplied by FDR to Issuers fall within the
exemption at point 3."
There is not I think a parallel statement relating to the services supplied to
Acquirers looked at overall on a "core" basis, but it is implicit that the
Tribunal took the same view. They said at 201:
"We therefore allow the appeal in respect of the services which we have
identified as included in the principal supply."
58 Mr Cordara submits that there is no error of law (and certainly no
appealable error of fact) in the Tribunal's conclusions as to the core supply
in paragraphs 163 and 173. Mr Paines did not in truth dissent; his argument
was that it was unnecessary for him to seek to undermine the Tribunal's finding
in the first sentence of paragraph 163, which was to no other or greater effect
than that there was a single supply consisting of a number of elements which,
by virtue of being treated as a single supply, would fall to receive the same
tax treatment. That, however - and here was the submission's bite - did not
determine what the tax treatment should be.
59 In my judgment Mr Paines was quite right, at least to this extent. On the
Tribunal's findings this was a table-top case. That being so, it is not
necessarily self-evident what the core supply's tax treatment should be: see
paragraph 54 above. But what the tax treatment should be falls under Issue 3,
with which I have yet to deal.
60 As to the nature or proper description of the core supply itself, I cannot
see how the Tribunal's conclusion at the first sentence of paragraph 163 of the
decision can, so far as it goes, be faulted. It was quintessentially a
conclusion of fact, and there is nothing in the nature of a Wednesbury
challenge conceivably available to undermine it. Mr Paines' argument
amounted, rather, to an assault on the Tribunal's view as to what supplies were
ancillary to the core. He said, correctly I think, that the Tribunal had
identified five particular elements within the activities of FDR as being of
special significance. They were (1) the netting-off procedure, (2) the
maintenance and operation of the cardholder/Merchant accounts, (3) the use of
BACS tapes, (4) the authorisation of transactions, and (5) the provision of
regular statements of the cardholder and Merchant accounts. All five figure in
the Tribunal's factual summary at paragraph 35 of the decision. As I have
shown the Tribunal held that (1), (2) and (3) involved exempt supplies, as it
were in their own right, because in each of them FDR executed transfers within
Art.13B(d)(3). That I have dealt with under Issue (1); but it will be crucial
for Issue (3) as well. However Mr Paines says also that the Tribunal was wrong
to hold that activities (4) and (5) were ancillary or integral to FDR's core
supply. The Tribunal plainly found them to be so in paragraphs 188 and 189.
Mr Paines submits that the Tribunal there committed the very error warned
against by Millett LJ in the opening sentences of the passage from
Wellington Private Hospital which I have cited.
61 I do not consider this criticism to be justified. The potential error to
which Lord Millett drew attention would arise where the court or tribunal
starts by asking whether some particular element in what the taxpayer does is
ancillary to the whole of his relevant activities, for that would be to beg the
question whether the case falls to be treated as one of a single supply at all.
That is not what the Tribunal did. Paragraph 188 opens with the words:
"The authorisation activity itself does not of itself fall within the terms of
point 3, it is however clearly essential to the transaction giving rise to the
liability of the Issuer the discharge of which liability is within point 3, and
indeed gives rise to a contingent liability on the part of the Issuer."
Paragraph 189 begins:
"We hold also that the statementing both to cardholders and to Issuers is also
part of the main supply or ancillary thereto."
The Tribunal have begged no questions, in these or other passages. Their
earlier holding, notably in paragraph 163, as to the nature and extent of the
core supply, is logically prior to and independent of these later findings.
62 In my judgment the Tribunal's conclusions (a) that there was here a single
or core supply, and (b) that its nature was as they described it in paragraph
163 of the decision, are well justified. There might be argument as to some of
the detail in paragraph 163, for example as to whether the insertion of
circulars in the statements of cardholders accounts is ancillary to the
principal supply; but as I understand it we are asked only to decide the
principle of the thing.
Issue (3): What is the Correct Tax Treatment of the Core Supply?
63 If my conclusions upon Issues (1) and (2) are correct, they provide the
answer to this last question with which I may accordingly deal very shortly.
Given that the core supply is a table-top not an apex, it is necessary to look
at its constituent elements to ascertain what its tax treatment should be: see
paragraphs 54 and 58 above. But the result of that enquiry is given by my
conclusions on Issue (1): the making of transfers within Art.13B(d)(3) is at
the centre of the core supply. Mr Paines' arguments to the effect that the
core supply should be categorised as book-keeping or the like rest entirely on
the proposition, which I have rejected, that FDR makes no supplies amounting to
transfers (save in the CHAPS transactions).