This case – which can be found here – concerns multilateral interchange fees (‘MIF’) yet again. This is one of a number of cases where courts had to decide whether such fees were lawful (for examples, see the cases reviewed in my posts of 30 September 2016, 10 February and 24 March 2017).
This one, Sainsbury’s v Visa, is a decision about the lawfulness of Visa’s scheme. The case was brought under the shadow of decisions by the European Commission and a number of courts holding that MasterCard’s MIF scheme was unlawful.
In order to understand this case, it is important to first understand how the various credit card systems operate. There are two main credit card models:
- On the one hand we have three-party schemes, like the ones operated by American Express and Dinner’s Club. In a three-party scheme, the operator (such as American Express) both issues cards and settles transactions with merchants. In other words, when an American Express cardholder enters into a transaction with a merchant, American Express transfers the transaction amount, less a fee, to the merchant. American Express then retrieves the full transaction amount from the cardholder.
- On the other hand, there are four-party schemes – such as the ones run by MasterCard and VISA. Four-party schemes work as follows: a merchant accepts certain credit and debit cards pursuant to an agreement with an “Acquirer”, i.e. a bank or financial institution belonging to the MasterCard or VISA scheme. The card will have been issued by another bank belonging to the scheme (the ‘Issuer’). The Acquirer will charge a fee to the Merchants for the services it provided in respect of a transaction. This fee is known as the Merchant Service Charge (“MSC”), and covers (i) the fee the Acquirer pays MasterCard or Visa (“the Scheme Fee”); (ii) a fee charged by the Issuer to the Acquirer (the Interchange Fee or “IF”) and (iii) the Acquirer’s own fee (“the Acquirer Margin”).
At stake in all VISA and MasterCard cases have been the interchange fees charged by one bank (the Issuer) to the other (the Acquirer). Interchange fees may be needed in four-party schemes, but not in third-party schemes. This is because four-party give rise to “two-sided markets” – one in which card Issuers compete with each other for cardholders (the issuing market); and another in which Acquirers compete for the business of merchants (“the acquiring market”). The two sides are closely linked and dependent on each other: the value of a Visa or MasterCard card to a cardholder is dependent on the extent to which it is accepted by merchants; correspondingly, the benefit merchants gain from accepting Visa and MasterCard cards is dependent on the extent to which consumers have and use those cards.
Interchange fees have the effect of moving value from one side of the system to the other – they rebalance the incentives between the Issuers and the Acquirers, such that Issuers are motivated to provide payment cards to consumers, thereby enabling consumers to transact at the requisite level to allow for the maximisation of benefits to both sides of the market. This is, however, subject to a de facto cap, determined by merchants’ willingness to pay fees for card payments. This cap may be sub-optimal: in order to remain competitive in their market, merchants may find themselves in a situation whereby they are effectively compelled to accept payment cards even if the cost of doing so negates any transactional benefit they might otherwise enjoy; simultaneously, issuer banks have incentives to increase interchange fees since is their main source of revenue under the scheme. As a result, it may be optimal for the scheme (i.e. VISA or MasterCard) to set the fee at a level that is optimal.
A very clear overview of the economics of interchange fees, along the lines just described, are very clearly explained in paras. 20-60 of the judgment. This overview is then followed by a review of the various decisions in Europe and the UK on the topic, including the new European Interchange Fee Regulation (para. 61-79).
The court then assesses the lawfulness of Visa’s interchange fee system as follows:
- It begins by noticing that an interchange fee is not merely a payment from the Acquirer to the Issuer. It is also a mechanism to adjust the amount at which the underlying principal payment is settled – in a way that creates incentives for the use of the system by both sides of the market. In order to create such incentives, the fee may even be negative in some cases (or, in other words, the Issuer may technically have to pay a fee to an Acquirer) (para. 80-82).
- The judgment then provides an overview of European and English law on restrictions to competition by effect (para. 83-97). A particularly important discussion for the outcome of the case takes place at paras. 88-97, where the court rejects a submission that showing that higher prices follow from an agreement suffices to prove that the agreement has a restrictive effect on competition. Instead, increased competition can on occasion lead to higher prices, as is the case with markets undergoing liberalisation or emerging markets. In other words, competition law does not seek to prevent higher prices per se – it seeks to prevent conduct that restricts competition, which usually leads to higher prices.
- The decision then seeks to identify the appropriate counter-factual to determine whether an interchange fee is anticompetitive (para. 98-101). In this regard, the claimant accepted that a world in which each Issuer and each Acquirer entered into bilateral agreements between them was not possible. As a result, the counter-factual had to include a collectively-agreed default rule for the settlement of transactions, because such a mechanism was essential for the Visa scheme to function. On the other hand, it was possible for such a mechanism to provide for transactions to be set at par – in other words, it is possible for the interchange fee to be set at zero.
A different question was whether the counter-factual should presume that MasterCard is constrained not to charge interchange fees because they were illegal (a symmetrical counterfactual) or whether the court should assume that MasterCard would be free to continue setting interchange fees regardless of Visa not doing so (the asymmetrical counterfactual) (para. 162-170). The Court found that this was not relevant to the analysis of the case, but I’ll return to it below.
- Given the counterfactual it identified, the question for the court was not whether interchange fees are anticompetitive – on the contrary, they were found to be necessary for competition between card systems to exist – but whether interchange fees above zero are anticompetitive. The claimant argued that the VISA interchange fee limits the ability of Acquirers to compete on price, and acts as a floor to the amounts that Merchants may pay to Acquirers. VISA did not dispute this, but argued instead that the market would not be any more competitive in a counterfactual world where no MIFs were set – because parties would never depart from the default setting, even if in this setting the interchange fee was zero and payment was settled between banks at par (para. 102-105).
This submission on the part of VISA went against a number of decisions by UK courts, which the judgment reviews (paras. 105-111) before turning to the evidence put before the court (para. 112-125). The main conclusion from this review is that VISA’s submission hold up, i.e. that its interchange fees do not restrict competition (para. 171-173). If the interchange fee was set at zero, there would still be no bilateral agreements departing from that value (para. 125-129). As a result, the competitive process would operate similarly in both the actual and counter-factual worlds as regards interchange fees (para. 130-151). And even if an interchange fee acts as a floor that prevents competition on price as regards the amount of merchant fees charged by Acquirers, this does not restrict competition since any settlement mechanism such as interchange fees, which are necessary for the VISA scheme to operate, would create such a floor. As a consequence, the competitive dynamics would be identical in both the actual and the counterfactual worlds (para. 152-160).
- Despite finding that the interchange fee was not a restriction of competition, the Court nonetheless assessed whether it would be objectively justified if it amounted to such a restriction (para. 174-191). Given the previous findings that the absence of an interchange fee would merely lead to a different default settlement rule, it should come as no surprise that the court found that an interchange fee would not be objectively necessary for the VISA scheme to operate.
In short, the court found that Visa’s UK interchange fees do not restrict competition within the meaning of Article 101(1).
This is a very interesting judgment. I was particularly taken by the quality of the discussion on the requirements for a finding of anticompetitive effects, and by the clarity of exposition of complex matters when discussing the counter-factuals. I found the discussion of how to identify the relevant counter-factual – whether if it should reflect the actual conditions under which competition occurred, or whether it should disregard conduct in the market which might infringe competition law (i.e. unlawful MasterCard MIF’s) – particularly interesting.
At the same time, I think this is a good example of how complicated it is to identify the relevant counter-factual – and of how choosing between counterfactuals grants courts significant leeway in how they ultimately decide a case. By selecting a different counter-factual by reference to the specific evidence before the court, this judgment implicitly finds that previous decisions – by the European Commission and by other English courts – were wrong on the facts (this finding is made expressly as regards the Sainsbury’s v MasterCard case – see para. 154). In particular, the court found that bilateral agreements and negotiations on the level of interchange fees are not really alternatives to interchange fees set by default (which broadly goes against the findings of the European Commission and the Competition Appeal Tribunal); and that the absence of competition on merchant charges beneath a certain price as a result of interchange fees is not a reduction of competition by reference to the relevant counter-factual world (which goes against a previous decision by the High Court decisions).
Given the various decisions reaching different conclusions as regards the lawfulness of very similar conduct, this judgment is likely to be appealed and there is a certain risk of it being overturned. This was a thought apparently shared by the judge and parties. Since it was found that Visa was not liable, the judgment did not need to address the matter of damages. However, I have read (in one of the various competition news services) that the court suggested – and the parties agreed – that a decision on damages would be issued before the end of the year. Since this decision would only be useful if this judgment were to be overturned on appeal, I think it is quite impressive that the judge suggested this – and, as such, I thought I should mention it as an example of a very best practice.