This blog considers the recent judgment of the Court of Appeal in Deckers UK Ltd v Up & Running (UK) Ltd [2026] EWCA Civ 553 in which Lord Justice Green sets out the test to be applied in determining the existence of an object infringement – in this case, in the context of a selective distribution agreement for Hoka running shoes.
The case concerned a clause in Deckers’ selective distribution agreement (“the Agreement”) which had been relied on to refuse one of its distributors, Up & Running, permission to sell excess stock built up over the Covid period at discounted rates on an anonymised website. The anonymity of the website meant that consumers would not associate it with the bricks and mortar stores of Up & Running or its branded website. Deckers considered that this was inconsistent with its brand strategy and the relevant clause in the Agreement which required permission to be sought for sales on a website and for that website to comply with certain requirements. When Up & Running launched the anonymised website regardless, Deckers terminated the Agreement with notice for breach. Up & Running then brought a damages claim before the Competition Appeal Tribunal (“the CAT”).
The CAT found that the Agreement:
- did not meet the Metro safe-harbour criteria (see Case 26/76 Metro SB-Groβmärkte v Commission [1977] ECR 1875) because the criteria on which it was based were not properly recorded or clear, included some quantitative criteria and were not applied uniformly so could not be said to go no farther than was necessary.
- sought to prevent retailers such as Up & Running from making passive sales to consumers through a specialised channel of clearance retailers which Deckers controlled, amounting to a restriction on internet sales which was the type of hardcore restriction described in Article 4(c) of the EU Vertical Block Exemption then in force as set out in Commission Regulation (EU) 330/2010 (“the VBE”). It was also a by object restriction under the Chapter I prohibition as the relevant clause pursued no plausible material objective other than the restriction of intra-brand competition.
- sought to prevent retailers such as Up & Running from selling Hoka running shoes at a material discount on clearance websites, amounting to resale price maintenance which was the type of hardcore restriction described in Article 4(a) of the VBE. It was also a by object restriction under the Chapter I prohibition as the relevant clause pursued no plausible legitimate aim other than the restriction of intra-brand competition.
- could not take advantage of the VBE despite the relevant market shares of the parties being comfortably below 30% because it contained two hardcore restrictions caught by Articles 4(a) and (c) of the VBE respectively.
- was in breach of statutory duty and liable to Up & Running in damages.
The Court of Appeal disagreed, finding that the CAT erred in concluding that:
- a selective distribution agreement falling outside the Metro safe-harbour amounted to, or was very likely to amount to, a restriction by object.
- the correct legal test for determining an object restriction focused on the objective or purpose of a measure because absent a plausible legitimate justification, a restrictive purpose necessarily constituted a restriction by object.
- a clause conferring an unfettered power was, in itself, a restriction by object because it had the potential, as a matter of contract, to be used for an unlawful, restrictive purpose.
- the classification of a clause as a hardcore restriction under the VBE was sufficient to classify it as a restriction by object.
The Court of Appeal concluded that there was no basis on which Deckers’ act of termination had the object of restricting competition when viewed in its proper legal and economic context and taking into account its content or scope. Further, Articles 4(a) and (c) VBE did not serve to preclude the challenged conduct from benefitting from the block exemption.
There is a lot in the judgment. In my view, there are at least five important points of interest.
First, the Court of Appeal made clear that a selective distribution system which fails the Metro criteria will not necessarily – or even likely – be a restriction by object. The Court considered that the CAT had reasoned that such a system which fell outside the Metro criteria would almost invariably give rise to a restriction by object (at §43 and §108). However, the CAT had (at §41) expressly cast doubt on the suggestion by some commentators that if a selective distribution system fails to meet the Metro criteria, then it will be a “by object” restriction, based eg. on Case 107/82 AEG Telefunken at [37]. Although the CAT stated (also at §41) that in its view, a failure to meet the Metro criteria suggested there may well be a “by object” restriction, relying on AEG Telefunken at [42] and Case C-439/09 Pierre Fabre at [40], it also stated that there was no presumption to that effect, relying in particular on Advocate General Wahl’s opinion in Case C-230/16 Coty at [116], such that it was still necessary to conduct the assessment of whether that was in fact such a restriction. Nonetheless, the Court of Appeal’s approach underscores that a failure to meet the Metro criteria is not the end of the road in assessing whether a selective distribution system is anti-competitive.
Secondly, the Court of Appeal conducted a detailed review of the jurisprudence before setting out a cumulative four-part test for determining the existence of an object restriction (at §105(iii)-(vii)) ie. assessing:
- the content of a measure, assessing the scope of a restriction or agreement and drawing conclusions about the legal scope of a contract or extent of a concerted practice.
- the objective or purpose of a measure, where the analysis was objective. Subjective intention could be taken into account but did not make a restriction unlawful by object if, taking into account the broader test, it would not have a sufficient adverse effect on competition.
- the legal context which included decisional practice, whether the agreement was horizontal or vertical and whether there were legal or regulatory barriers to entry.
- the economic context of a measure, including the nature of the goods or services affected and the actual conditions of the functioning and structure of the markets or market in question, including the nature and extent of inter-brand competition.
The last limb engages the points made by Tom Coates in his recent blog post on Tondela about (1) similar language having been criticised historically as blurring the lines between object and effects infringements and (2) the implications of such an approach for evidence and analysis in object cases.
Thirdly, the Court of Appeal identified that the threshold which has to be met when applying the four-part test is whether the measure revealed “a sufficient degree of harm” to competition (at §105(i)-(ii)). This is not whether a sufficient degree of harm is a likely or non-fanciful consequence or a real (non-fanciful) potential or capacity to restrict competition. It is also a threshold to be applied strictly and restrictively. As a consequence, the fact that a measure might have a restrictive objective is not sufficient to render it an object restriction if, but for that objective, the measure does not reveal a sufficient degree of harm to competition. It remains necessary to consider the content of the restriction and its legal and economic context. As to what is “sufficient” in any given context, the Court accepted (at §113) the existence of a sliding scale of seriousness of different types of case with price fixing at one end and vertical restraints in selective distribution at the other end; at the more serious end eg. in a collusive horizontal price fixing agreement, perhaps not much more additional evidence might be needed beyond the primary facts, whereas at the other end eg. in a vertical distribution agreement, more evidence of harm might be needed as the risk of harm to competition is typically commensurately lower. Potentially, this reduces the evidential burden in object cases in certain types of case. Interestingly, in this case, it being at the less serious end of the spectrum, the result was a debate about whether the case needed to be remitted to the CAT, but the Court ultimately concluded (at §134) that this was not necessary given the findings that had been made.
Fourthly, the Court of Appeal grappled with the question of whether broad contractual discretions are inherently restrictions by object because of their potential to be used for restrictive purposes. The Court answered that the reasoning of the CJEU in Case 333/21 Superleague does not support that proposition, is inconsistent with the need to consider the economic context of a measure when determining whether it constituted an object restriction and would create a situation where any contractual power would become unlawful, even if in an otherwise innocuous vertical agreement or universally exercised in a benign manner (at §98). This approach underscores the focus on the context and effects of a potential object restriction. It remains the case however, that broadly drawn contractual discretions are likely to create more risk than more narrowly circumscribed ones; they may also limit the ability to take advantage of the Metro safe harbour.
Fifthly and finally, the Court of Appeal made clear that even where infringing conduct constitutes a hardcore restriction under the VBE, this does not indicate, in and of itself, that the conduct constitutes a restriction by object. The concepts of hard-core restriction under the VBE and a restriction by object are not co-extensive and the ordinary analysis must still be applied (at §121(iii)). Indeed, the CAT had recognised that they were not co-extensive as a matter of principle. It had noted that resale price maintenance was a hardcore restriction and identified it as restrictive by object “in many, if not in almost all circumstances” while also identifying the need to assess whether there was a sufficient degree of harm to competition in light of the relevant legal and economic context, where the legal context includes its status as a hardcore restriction (at §42(7)-(8) and §53). But when applying principle to the facts of the case, the CAT appeared to lean heavily on its identification of the Agreement as containing a hardcore restriction in identifying an object restriction (at §§196-197 and §§211-212). The Court of Appeal’s reasoning makes clear that the categorisation of a measure as a hardcore restriction will inform the legal context in which the assessment of the measure takes place for the purposes of identifying an object restriction but will not determine that assessment. So if, for example, the economic context is benign because of relevant market shares and the extent of inter-brand competition, even in the case of a hardcore restriction, there may not be a sufficient degree of harm to competition to identify an object restriction.
In this case, the Court of Appeal’s conclusion that there was no restriction by object was driven heavily by the economic context of the Agreement. The limited scale of the arrangements – a refusal of permission to sell one-off surplus Covid stock at discounted rates on a single anonymised website – in a context where the market shares of the parties were already a small proportion of the overall product market made it very difficult to identify a sufficient degree of harm to competition to give rise to an object restriction.
The economic context was also central to the Court of Appeal’s conclusion that the Agreement did not in fact contain hardcore restrictions caught by Articles 4(a) and (c) VBE. The Court focused on the fact that retailers could discount all stock in their stores and branded websites as they wished and consumers could buy that stock actively and passively through those channels. There was therefore no hardcore resale price maintenance restriction or internet sales restriction of the kind excluded from the benefit of the VBE.
So economics, economics, economics. The judgment is a welcome addition to an area where there is limited UK jurisprudence and is good news for suppliers. It will also have been welcomed by the CMA which intervened in the case to seek clarity on the test to be applied to restrictions by object and to prevent the use of contract as a normal commercial tool being hindered in distribution systems. One is, though, left wondering whether the judgment hangs together because the Court found both that there were no hardcore restrictions and that there were no object restrictions. The judgment seems to leave open the possibility of a hardcore restriction that forms part of the context of the assessment of a measure which does not ultimately constitute an object restriction, but it is difficult to imagine what this might look like. It might be considered odd for a something to be excluded from an exemption it does not require in the first place. Future cases in this area will no doubt be illuminating.
