On 17 June 2020 the Supreme Court unanimously upheld an earlier Court of Appeal ruling that Mastercard and Visa’s multilateral interchange fees (MIFs) restricted competition contrary to Article 101(1), resolving several disputes between the card schemes and retailers that have run in the UK courts since as early as 2012, and have at times resulted in conflicting outcomes.
The Supreme Court’s judgment represents a significant step towards the end of the long-running MIFs saga.[1] It also raises three points of more general interest: (i) the standard of evidence required in order to benefit from the so-called “efficiencies defence” under Article 101(3); (ii) the correct application of the “fair share” test under Article 101(3); and (iii) the degree of precision required by a defendant to establish the extent of pass-on. This update relates to the first two of these points of interest. Our briefing dated 24 June 2020 provides an update on the third.[2]
The history of the case
The case has its origins in proceedings brought by the European Commission against Visa and Mastercard in the early 2000s. Whilst the Visa proceedings were settled, the Commission issued a final infringement decision against Mastercard in 2007, [3] which Mastercard unsuccessfully appealed to the General Court and then again to the Court of Justice (CJ). In its 2014 judgment[4] (Mastercard CJ), the CJ upheld the Commission’s and the General Court’s findings that the MasterCard MIFs applicable to cross-border bank card payments within the EEA restricted competition since they limited the pressure merchants could exert on acquiring banks when negotiating the costs charged by those banks.
In light of these proceedings, in 2012 several retailers filed claims for damages against the two card schemes in the High Court in relation to both their EEA MIFs (as considered in Mastercard CJ) and their domestic UK MIFs. The first substantive trial of these issues before the UK Courts was heard by the CAT in early 2016.[5] This was followed by two further sets of proceedings before the High Court in 2017, one before Mr Justice Popplewell [6] and one before Mr Justice Phillips.[7]
These three sets of proceedings, despite relating to materially identical facts, resulted in three conflicting decisions. The CAT found that Mastercard’s MIFs restricted competition by effect and awarded damages of £68.5m to Sainsbury’s. By contrast, each of the High Court judges dismissed the retailers’ claims, but for different reasons: Popplewell J on the basis of the so-called “death spiral argument” (whereby he found that Mastercard’s MIFs were not restrictive since, in their absence, Mastercard would have exited the market given the competition it would have faced from Visa),[8] and Phillips J on the basis that Visa’s MIFs did not restrict competition in the acquiring market (since, in his view, the level of competition was the same regardless of whether the MIF was positive or zero).
All three judgments were appealed. Unsurprisingly, the Court of Appeal directed that all three appeals should be heard together, to avoid any further conflicting outcomes. The Court of Appeal overturned the reasoning behind all of the judgments in the lower courts, although it agreed with the conclusion of the CAT that there was a restriction of competition, finding that Mastercard and Visa’s MIFs infringed Article 101(1).[9] Unlike the lower courts,[10] the Court of Appeal considered that it was bound by Mastercard CJ which found as a matter of law that: (a) the correct counterfactual for schemes like the Mastercard and Visa schemes was a “no default MIF” and a prohibition on ex post pricing (or a settlement at par rule); and (b) against that counterfactual, Mastercard’s (and by extension Visa’s)[11] MIFs were restrictive of competition.
Mastercard and Visa appealed the Court of Appeal’s judgment. The Supreme Court heard the appeal in January 2020 and issued its judgment on 17 June 2020.
The Supreme Court’s judgment
The Supreme Court unanimously upheld the conclusion of the Court of Appeal that Mastercard CJ is binding, on the basis that “the essential factual basis upon which the Court of Justice held that there was a restriction on competition is mirrored in these appeals”.[12] The MIFs were thus a restriction of competition contrary to Article 101(1). The Supreme Court went on to say that, even if it was not bound by Mastercard CJ, it would still have followed it and concluded that there was a restriction of competition (in particular because the effect of the MIFs, which were determined by collective agreement rather than by competition, was to create a minimum price floor for the merchant service charge (MSC) which “immunised” a significant portion of the MSC from competitive bargaining). By contrast, in a counterfactual with no MIFs, the whole of the MSC would be determined by competition.
The Supreme Court also concluded that the Court of Appeal was correct to find that there was a requirement for “cogent empirical evidence” in determining whether a claim for exemption under Article 101(3) has been made out, as explained further below.
The standard of evidence required under Article 101(3)
Article 101(3) allows agreements that would otherwise restrict competition to be exempted from Article 101(1) if they generate objective economic benefits (otherwise known as “efficiencies”) that outweigh the negative effects of the restriction of competition.
To satisfy Article 101(3), an agreement must satisfy four cumulative conditions:
- it must contribute to improving the production or distribution of goods or contribute to promoting technical or economic progress;
- consumers must receive a fair share of the resulting benefits;
- the restrictions must be indispensable to the attainment of these objectives; and
- the agreement must not afford the parties the possibility of eliminating competition in respect of a substantial part of the products in question.
It is well established that the burden of satisfying these four conditions lies on the parties to the agreement.
In considering how this burden can be discharged by the parties to an agreement, the Court of Appeal had followed the approach of Phillips J’s (strictly obiter) judgment on Article 101(3),[13] which in turn referred back to the Commission’s 2007 Mastercard decision, in concluding that any claimed efficiencies must be based on robust analysis and cogent empirical evidence, meaning that it was not sufficient for the parties to rely on economic theory alone. Any claim under Article 101(3) must also be based on empirical data and fact.
Mastercard and Visa disagreed with this approach, arguing before the Supreme Court that it created an unduly onerous standard of proof on the parties to an agreement, which went over and above the usual civil standard of the balance of probabilities. According to Visa, a judge should be entitled to conclude that “nothing more is required than the expert evidence of economists” in order to be satisfied that the relevant efficiency has been made out.
The Supreme Court rejected this argument. In its view, the complaint made by Visa and Mastercard did not relate to the standard of proof but rather to the level of evidence required to meet that standard of proof. Siding with the Court of Appeal, it found that, in discharging the burden that an Article 101(3) efficiency existed, the usual civil standard of proof applied (i.e. it had to be demonstrated that, on the balance of probabilities, the claimed efficiencies existed). However, the Supreme Court considered that the standard of evidence required to discharge this burden was higher than that claimed by Mastercard and Visa. According to the Court, it is not sufficient for the parties to an agreement to rely on economic theory alone: any claim under Article 101(3) must be supported by empirical data and fact (referring to a requirement for “detailed, empirical evidence and analysis”).[14]
The practical implications of this approach are likely to be challenging. It is worth recalling that an assessment under Article 101(3) needs to be carried out at the time of entering into the relevant agreement, when empirical or real world evidence might not be available to the parties. This point was raised by Visa during the proceedings, but the Supreme Court’s judgment largely avoids the issue by referring back to the Commission’s Article 101(3) Guidelines which require that, in cases where an agreement has yet to be fully implemented, “the parties must substantiate any projections as to the date from which the efficiencies will become operational so as to have a significant positive impact in the market”.[15] What is interesting is that if these “projections” are based on economic theory and/or modelling alone then this is unlikely to be sufficient to establish the relevant efficiencies: it seems that empirical data and facts will always be required.
The fair share test under Article 101(3)
The Supreme Court was also required to consider the correct application of the second limb of Article 101(3) and the requirement for consumers to receive a “fair share” of the benefits resulting from the agreement. The Court was clear that to satisfy this test Visa and Mastercard were required to demonstrate that customers who were impacted by the MIFs (i.e. merchants, who had to pay higher card acceptance fees) were compensated by a corresponding benefit. It was not sufficient to point to benefits arising elsewhere (in this case, to cardholders).
This finding has broader application to two-sided markets. When considering whether an efficiencies claim under Article 101(3) has been made out, the parties to an agreement involving a two-sided market will need to be satisfied that customers on the side of the market that is impacted by the relevant restriction are compensated for that restriction by the claimed benefit or efficiency. Whilst a benefit to participants on the other side of the market can be considered in assessing whether the relevant agreement satisfies the first limb of Article 101(3) (i.e. whether the agreement is efficiency enhancing more generally), it cannot be taken into account when assessing whether the second limb has been satisfied (i.e. whether customers receive a fair share of that efficiency).[16]
[1] Whilst the Supreme Court has resolved the question of whether Mastercard’s and Visa’s MIFs infringed Article 101(1), two of the three sets of proceedings have been referred to the CAT in order to determine whether the MIFs are exempt under Article 101(3). The question of quantum of damages also remains (which will be decided by the CAT). [Back to article]
[5] Sainsbury’s Supermarkets Ltd v MasterCard Inc [2016] CAT 11. This case was issued in the High Court. However, in December 2015 the High Court decided that the host of complex competition law issues at hand warranted a transfer to the CAT. [Back to article]
[7] Sainsbury’s Supermarkets Ltd v Visa Europe Services LLC [2017] EWHC 3047 (Comm). [Back to article]
[9] Sainsbury’s Supermarkets Ltd & Others v Mastercard Incorporated & Others [2018] EWCA Civ 1536. [Back to article]
[10] Broadly speaking, the lower courts had considered that the CJ’s conclusions in Mastercard CJ were based on materially different facts than those relevant to the UK proceedings (in particular that Mastercard CJ related to Mastercard’s intra-EEA MIFs, whereas the present proceedings related primarily to UK MIFs applicable during a largely distinct time period). [Back to article]
[11] On the basis that Visa’s MIFs were “materially indistinguishable” from Mastercard’s. [Back to article]
[12] Sainsbury’s Supermarkets Ltd v Visa Europe Services LLC and others and Sainsbury’s Supermarkets Ltd and others v Mastercard Incorporated and others [2020] UKSC 24, para. 93. [Back to article]
[13] Sainsbury’s Supermarkets Ltd v Visa Europe Services LLC and others [2018] EWHC 355 (Comm), para. 24. [Back to article]
[14] Sainsbury’s Supermarkets Ltd v Visa Europe Services LLC and others and Sainsbury’s Supermarkets Ltd and others v Mastercard Incorporated and others [2020] UKSC 24, para. 118. [Back to article]
[15] Commission Guidelines on the application of what is now Article 101(3) TFEU (2004/C 101/8) (the “Commission Guidelines”), para. 131. [Back to article]
[16] Unless the two sets of customers are substantially the same – but in many contexts that will unlikely be the case. [Back to article]
