Chillin'Competition

Of the many judgments addressing the notion of ‘by object’ infringement that came out last year (and which I discuss here), FIFA v BZ is particularly significant, not just because of the issues at stake and the way the Court addressed them, but because of the hints it gives about the future of Article 101(1) TFEU.

In several respects, the enforcement of EU competition law is entering a new era, and FIFA v BZ heralds the new times. There are, in my view, three main lessons to draw from the judgment.

The resilience of the Court’s approach to restrictions by object

The first lesson is that the Court’s methodological approach to identify restrictions by object is proving to be remarkably resilient. It is clear from the case law that, in order to determine whether an agreement infringes Article 101(1) TFEU by its very nature, it is necessary to consider not just its content, but also its objective aims and the economic and legal context of which it is a part.

More importantly, this methodological framework is not an abstract formula. As FIFA v BZ exemplifies, the Court walks the talk. You will find in the judgment a meticulous contextual evaluation of, inter alia, the content of the rules (in that regard, the Court found that there was a mismatch between the legitimate aims pursued and the breadth of the restraints), the discretion they gave FIFA (and therefore the ability to decide which aims to pursue case by case) and the latter’s position.

The resilience of the case law is notable considering the tendency among some actors in the system to rely on shortcuts that disregard the methodology mandated by the case law. The instinct to establish a ‘by object’ infringement on the basis of form alone (e.g. the fact that the practice amounts to ‘price-fixing’ or ‘market sharing’) remains particularly strong, even thought it has been consistently rejected by the Court.

Every now and then, some actors also display a tendency to infer a ‘by object’ infringement from the effects of an agreement. Object and effect are two separate, fundamentally different stages. However, it is tempting to claim that a practice is restrictive by its very nature where it is deemed to have substantial effects on competition, and conversely, to claim that no ‘by object’ violation exists based on the absence of an appreciable impact.

FIFA v BZ confirms that the two stages must not be conflated, even when doing so comes across as intuitively right. Object is not, and has never been, a presumption of effects (as the Expedia judgment clarified). The effects of an agreement (or the unlikelihood of such effects) is not a reliable guide of its object.

From public to private enforcement

The judgment also reveals that a new generation of cases is reaching the Court. The landmarks of the past decade addressing the notion, including Cartes Bancaires, Generics, Budapest Bank and Super Bock, all originated in cases before a competition authority, whether the Commission or an NCA.

The new generation of cases, including not just FIFA v BZ, but also Superleague and the recent ruling in Booking, originated in disputes between private parties. The rise of private enforcement is testament to the success of Regulation 1/2003.

The change in the nature and profile of litigants is not without consequences. By necessity, the objectives and incentives of private litigants differ from those of a public authority. This difference is likely to be manifested across a number of fronts, from the nature of the cases that reach the Court to the way the arguments are framed.

There is a chance, accordingly, that issues pertaining to the notion of restriction by object, will be approached in novel and unprecedented ways (and thus that the resilience of the case law will be tested from new angles).

A second consequence of the rise of private litigants is decentralised enforcement. The latter, in turn, risk affecting the uniformity in the interpretation and application of Article 101 (and 102) TFEU. For the same reason, the resilience of the Court’s methodology to identify ‘by object’ infringements is likely to be put to the test more frequently than in the past.

In a decentralised landscape, the inclination to rely on discredited methodological approaches is likely to be more frequent (by the same token, the risk of substantive fragmentation is likely to increase).

A more explicit style when ruling?

A salient aspect of FIFA v BZ (and Booking) relates to the approach when delivering rulings. In the context of a preliminary reference, it is for the national court to apply the law, as interpreted by the Court, to the facts of the case. However, the Court, in both judgments, was explicit about the legal characterisation of the practices at stake (holding, in FIFA v BZ, that the rules were restrictive by object and, in Booking, that the clauses were not ancillary restraints).

This tweak in terms of style is reasonable if one considers the risk of fragmentation mentioned above. It is an effective means to ensure that a preliminary ruling is not interpreted differently from one court to another and, similarly, to avoid that the same issues keep coming back before Court.

There is a chance that, for these very reasons, the Court will also be more explicit (and/or more structured) about certain aspects of the methodology it follows to identify restrictions by object. At the very least, doing so would contribute to ensuring that national courts do not fall into the old trap of formalism when conducting their analysis.

One of the most significant developments of the past few years is the comeback of competition on the merits (as I had the chance to discuss here). For a long time (and certainly during the past decade), competition on the merits was nothing other than an abstract aspiration (a ‘general umbrella’, as Heike Schweitzer and Simon de Ridder crisply put it here), with no real substantive content.

There has been a push, very recently, to turn this abstract idea (with which it is hard to disagree) into an operational one. Instead of a ‘general umbrella’, it was hoped that competition on the merits would become a working tool allowing courts and authorities to draw the line between lawful and abusive conduct in individual cases.

The origins of this attempt are easy to trace in the case law. Competition on the merits was invoked by defendants in some cases (namely Google Shopping and Servizio Elettrico Nazionale) in the hope that it would (i) reduce the scope of Article 102 TFEU and/or (ii) make it harder for competition authorities to discharge their burden of proof.

The hopes behind this push have not materialised. It is clear after Google Shopping that conduct that is not inherently at odds with competition on the merits can be caught by Article 102 TFEU in a particular economic and legal context.

It is also clear that a competition authority cannot be required to show, always and everywhere, that the potentially abusive practice departs from competition on the merits. Where there is an established legal test, showing that the conditions enshrined therein are satisfied will be sufficient to prove an abuse to the requisite legal standard.

The attempt to turn competition on the merits into an operational concept has failed and might have backfired. This nebulous notion can be easily used against defendants (casting conduct as departing from competition on the merits is effective, if only as a rhetorical tool, to imply that it is necessarily abusive).

It may have failed, but the attempt to revive of competition on the merits has left a trail of confusion with which the competition law community has to live. The Commission, for one, could not avoid engaging with the concept in its Draft Guidelines. It was not an easy hand.

The Draft Guidelines are particularly helpful when they clarify that it is not necessary to show that a practice departs from competition on the merits in all circumstances. The question will typically be subsumed into the legal test if there is one already in place. This is an important, if uncontroversial, clarification.

On the other hand, the Draft Guidelines do not dispel the impression that the notion of competition on the merits is a source of confusion and duplication. This is not a criticism of the document: it is a criticism of the attempts to turn competition on the merits into something that it is not.

We have known for decades that competition on the merits cannot and will never be an operational concept. As I had the chance to discuss with Heike Schweitzer, Ordoliberals became aware, early on, of the limits of the concept and that it could not assist the sort of granular assessment that individual cases demand. It did not take long before it was abandoned (other than as a ‘general umbrella’, that is).

The Draft Guidelines illustrate the uselessness of competition on the merits as an operational concept. The overlap between Section 3.2. (‘Conduct departing from competition on the merits’) and Section 3.3. (‘Capability to produce exclusionary effects’) and the confusion to which this overlap gives rise is one of the most salient aspects of the document.

There is, for instance, duplication between the idea of competition on the merits and that of ‘naked restriction’ (or ‘by object’ abuse). Is showing that a practice departs from expression of competition on the merits not the same as showing that it is abusive by its very nature (in the sense that it has no plausible explanation other than a restrictive one)? If the former is established, what is the point of requiring proof of the latter too?

The duplication that the notion creates is, if anything, more apparent after Google Shopping. As pointed out above, the Court clarified in the judgment that some practices are only against competition on the merits in certain circumstances. This is true, for instance, of a scenario where a vertically-integrated dominant firm discriminates against rivals.

In such instances, a case-by-case assessment is warranted. The assessment, according to the ruling, will have to consider the relevant economic and legal context within which the practice is implemented.

The future Guidelines will have to account for this reality, and acknowledge that the issue of competition on the merits is context-dependent. Any future version of the document will have to distinguish between conduct that is inherently at odds with competition on the merits and that which demands a case-by-case assessment of ‘all the relevant circumstances’. This distinction necessary mirrors the framework laid down in Section 3.3.

A second, related consequence is that the issue of competition on the merits will overlap in theory and in practice with the analysis of effects (as is apparent from Google Shopping itself).

Is there a way forward, against this background? A cleaner approach that would avoid duplications would be to arrange differently the ideas currently found in Sections 3.2. and 3.3.

Google Shopping shows that it would be artificial to separate between competition on the merits and the assessment of effects. Both issues are part of the evaluation of ‘all the circumstances’ and will typically consider aspects pertaining to the economic and legal context.

It would therefore be more straightforward, more intuitive and more faithful to the case law to structure the analysis along the lines of the following:

First, naked restrictions which, by their very nature, are against competition on the merits and therefore inherently abusive. In this case, the dominant undertaking would only be able to escape the prohibition in exceptional circumstances (as discussed here).

Second, conduct which, by its very nature, is an expression of competition on the merits and as such incapable of having actual or potential effects on competition. These practices, if anything, show well how artificial it is to distinguish between Sections 3.2. and 3.3. and how intertwined the questions are. Pricing above average total costs is a valid expression of competition on the merits because it cannot be expected to exclude rivals, and vice versa.

Third, conduct that is presumptively abusive and thus at odds with competition on the merits, which would include exclusive dealing, loyalty rebates and tying in traditional (that is, non-digital) scenarios. In this context, the dominant firm may avoid the prohibition if it shows that the practice is incapable of restricting competition.

Fourth, conduct that requires a case-by-case assessment. Conduct falling within this category comes in two flavours. There are, on the one hand, those subject to a specific legal test, in which case showing that the conditions of the test will be sufficient to establish an abuse; and, on the other, those practices subject to an assessment of ‘all the circumstances’.

Even without trying, we find ourselves looking back (and forward to what is coming) by the end of every year.

As far as I am concerned, this natural inclination is compounded by the fact that it is almost exactly a year since the publication of The New EU Competition Law. I am pleased with how it has been received, and have the best memories of the few trips to discuss it with friends and colleagues.

On this occasion, the book is also an invitation to look forward. As I was working on it, I quickly realised that The New EU Competition Law is but the first part of a trilogy examining the transformation of our field. If all goes well, next year, by this time, I will be announcing the publication of the second volume.

Book aside, I have been able to get some papers and book chapters out. This year’s vintage features the following:

Competition on the merits, published in the Common Market Law Review. It came out in January, but I have not changed my mind since. Competition on the merits, as a relic of times past, is an ‘irritant’ in the case law (as apparent in the Draft Guidelines). The notion has undergone a surprising revival which (I like to believe) will be short-lived.

Restrictions by object under Article 101(1) TFEU: from dark art to predictable framework, forthcoming in the Yearbook of European Law. This is a paper that pretty much wrote itself, as the Court of Justice kept issuing major rulings interpreting the notion of ‘by object’ restriction. The central idea behind the piece stems clearly from its title: it is no longer possible to argue that the notion is unclear. The Court has developed an analytical framework that has proved resilient to pressures from all sides.

Resale price maintenance in EU competition law: understanding the significance of Super Bock, recently published in World Competition. In a sense, this piece is a companion to the preceding one. It is also a reminder that EU law cannot be fully understood without considering how it is applied at the national level by courts and authorities (as I pointed out here).

Remedies in EU antitrust law, accepted in the Journal of Competition Law & Economics. Antitrust law is now in the age of remedies: never before had the adequate design of measures ending the infringement been so central to effective enforcement. In spite of this fact, remedies remain relatively misunderstood. I discuss their nature and purpose in the article, and propose some adjustments now that Regulation 1/2003 is being reviewed.

This paper (like all the rest, really) is dedicated to the memory of Heike Schweitzer. It is impossible for me not to think of her when working on all things competition law. Honouring her memory and legacy is high up on the list for 2025 and beyond.

Last, but not least, I published a chapter on data leveraging in energy markets (jointly written with Alexis Brunelle, Adrien de Hauteclocque and Juliette Ogez) for the Research Handbook on EU Competition Law and the Energy Transition, edited by Leigh Hancher and Ignacio Herrera Anchustegui.

As to what is in store for 2025, there is, among others, a piece on Hoffmann-La Roche forthcoming in Landmark Cases in EU Law (and on which I did a great deal of research into the concept of competition on the merits) and the forthcoming and much awaited edition of EU State Aid Control: Law and Economics, where I contribute a chapter with Damien Neven on State aid as a successful legal export.

Before I forget: 2024 will always be the year the blog reached 3 million views. Thanks you all so much for reading and sharing your thoughts over so many seasons!

The latest contributions to this series (see here) dealt with the analysis of effects. The main point it made is that we need clarity about the meaning of the notion of effect: the litmus test of the whole exercise depends on whether there is clarity about what an effect is and what it is not.

There are more aspects pertaining to the analysis of effects that could benefit from greater clarity. Over and over, we see the some issues that keep coming back, Sisyphean in their persistence, even though they should have long been behind us.

The Guidelines on which the Commission is working provide an ideal chance to address some of these common misconceptions around the issue of effects. I cannot think of a better forum to convey clarity, certainty and, above all, a clean, discernible analytical framework on which national courts and authorities can rely.

As far as the fundamental issues that could be addressed, I can think of the following:

Actual and potential effects are about the time dimension of the analysis

The distinction between actual and potential effects refers to the temporal dimension of the analysis. There should be little doubt, after the Court’s careful analysis in the Servier saga (see here), that references to potential effects relate to instances in which the assessment is prospective (that is, about an effect that may materialise in the future).

Contrary to what is sometimes (read: often) suggested, ‘actual’ and ‘potential’ do not refer to different substantive thresholds (the former being more demanding than the latter). In fact, they are not really, or not exactly, about the substantive threshold (keep reading till the end for more on this).

Clarifying this point is key for a number of reasons. The most salient of these, I would say, is that it makes sense to use the vocabulary consistently across issues and provisions. Using the same words to mean different things is a recipe for opacity and confusion.

When we talk about actual or potential competitors, we refer unquestionably to the time dimension: a potential competitor is one that has not yet entered the market but may do so in the future. This is, by the way, the manner in which the concept is used in the Draft Guidelines.

It would make little sense to refer to ‘actual’ or ‘potential’ in a different way when talking about the analysis of effects.

Appreciability, de minimis and significant effects

The Court of Justice has held unambiguously that, in the context of Article 102 TFEU, it is not necessary to show that the effects on competition are significant or appreciable. In other words, there is no such thing as a de minimis doctrine when the concept of abuse is at stake.

I do not believe there is anything controversial in this position. Because the application of Article 102 TFEU presupposes a finding of dominance, there is by definition no scope for de minimis doctrine.

This is a point on which the final version of the Guidelines could be even more explicit (ideally with a reference to Völk, where the Court explained that the doctrine applies on account of the ‘weak position’ of the parties in the relevant market, and which is not mentioned in the Draft).

The fact that the de minimis doctrine has no role to play in the context of Article 102 TFEU does not mean (this is another major source of confusion and misunderstandings) that every practice implemented by a dominant firm and/or every competitive disadvantage it inflicts upon rivals necessarily has an anticompetitive effect.

The rejection of the de minimis doctrine simply means that effects, when established, will necessarily be appreciable. But this fact does not dispense the authority or claimant from the need to establish these effects to the requisite legal standard in light of factors such as the coverage of the practice.

The substantive threshold of effects

An additional aspect on which the Draft Guidelines is silent, and about which a conversation is as necessary as it is indispensable, relates to the substantive threshold of effects. The issue is bound to be discussed sooner or later and it would be best if it were openly addressed by the Commission.

It has already been mentioned that effects may be potential. The Court of Justice has clarified, moreover, that the practice must be ‘capable’ of having such effects. These points do not say anything, however, about the requisite threshold of probability.

When the analysis is prospective, is it enough to show that the practice is a plausible source of anticompetitive effects? Is it necessary to show that it is likely to do so (that is, a probability of >50%)?

The case law is not unequivocal on this point and there is room for interpretation. One thing is clear: it is not necessary to show that the practice is certain to affect competition. I have always been of the Opinion that AG Kokott is on the money here and that the substantive threshold is one of likelihood (probability >50%).

As I say, the Draft Guidelines would be the right forum to address this matter once and for all. It would also be a great opportunity to make it clear (if only in a footnote) that the substantive threshold of effects is different from the standard of proof (even if both are expressed in probabilistic terms and even though they are, all too often, conflated).

I very much look forward to your comments, on this post and the rest of the series (see here, here and here).

The LSE Short Courses on Advanced EU Competition Law and on State Aid and Subsidies Regulation were launched a bit more than four years ago. I am really pleased that they have been received so enthusiastically by our community (so much so that we launched an extra edition of the competition law course in 2024, which finished a couple of weeks ago).

The next in line is the Short Course on State Aid and Subsidies Regulation, which will run between January and February of this coming 2025. A particularly exciting time with the many changes that the discipline is undergoing. There are still places available and it would be wonderful if you could join: all the info can be found on the link to the course here.

As usual, the short course will take place online and it is designed with full-time professionals in mind. Attendance will be capped at around 25 participants to maximise interaction (which, if you ask me, is one of the big pluses of this format: every edition is unique, and very much shaped by the discussions).

The sessions will be on four consecutive Thursdays: 6th, 23rd and 30th January, as well as 6th February (at the usual time: 2pm to 6pm London time).

An LSE Certificate will of course be available upon completion, along with CPD points for practitioners.

If you have any questions about the organisational aspects of the two courses, do not hesitate to contact my wonderful colleague Amanda Tinnams: A.Tinnams@lse.ac.uk.

ASCOLA needs no introduction in the competition law community. It has become the privileged forum for the exchange of ideas among scholars specialised in the field. The expansion it has undergone in the past few years is truly remarkable, and testament to its growing importance and the hard-work of the leaders steering the ship.

Next year’s conference (26-28th June) will take place in one of the landmarks (both in the geographic and the intellectual sense) of antitrust and competition law: Chicago. Participants will be hosted by renowned Professor Spencer Weber Waller at Loyola University Chicago School of Law.

If you are interested in taking part in the conference, the call for papers can be accessed here. As usual, the call is open to diverse disciplinary approaches: diversity is, in fact, one of ASCOLA’s trade marks.

On top of that, I very much like that organisers have identified three streams for potential participants: (i) academic papers (the general stream, for papers that have the typical length of a research paper); (ii) early-career researchers; and (iii) early project ideas (for very short submissions that seek to test new concepts or ideas).

For more information (other than that provided in the call for papers), it is probably best to get in touch with the organising team at ASCOLA2025@luc.edu. Best of luck with your submission!

This month’s issue of World Competition comes with a short piece of mine (the submitted version of which is available on ssrn here) on resale price maintenance in the aftermath of Super Bock.

I thought it interesting to say a word on it, since it is, I feel, one of these rulings that is only fully appreciated when read in conjunction with developments at the national level. Those of you who read Portuguese might want to take a look, in particular, at the judgment of the Tribunal da Relação de Lisboa following the preliminary ruling.

How does Super Bock change or refine our understanding of Article 101(1) TFEU? Prior to this judgment, resale price maintenance was considered to be restrictive of competition always and everywhere (that is, without the need to consider the relevant economic and legal context and without the need to take into account the aims pursued by the parties).

In Binon, in fact, the Court expressly dismissed the relevance of any contextual considerations when vertical price-fixing is at stake. Such an approach to the assessment of restrictions of competition was, unsurprisingly, the one followed by the Portuguese first-instance court in Super Bock.

The problem? Binon is at odds with subsequent case law, which has in turn been consistently followed by the Court in the past decade. The significance of Super Bock lies precisely in the fact that it addresses this friction and makes it explicit that the orthodox framework is also extensive to resale price maintenance.

Thus, resale price maintenance is no longer the ‘odd one out’ in the aftermath of Super Bock. Whether or not this practice restricts competition by its very nature depends on an analysis of the content of the agreement, its aims and the economic and legal context of which it is a part. The analysis, in other words, is not formalistic, but context-specific.

The impact of the preliminary ruling in Super Bock is best appreciated in light of the judgment of the Tribunal da Relação de Lisboa, which ruled on appeal in the national dispute. Following the guidance of the Court of Justice, the Tribunal found that the first-instance court had erred in law by following the Binon approach and meticuously ascertained the object of the practice in light of the factors identified in the case law.

The main lesson to draw from Super Bock is that resale price maintenance may not always restrict competition by object. It is not difficult to think of scenarios where the practice pursues a ‘legitimate aim’ within the meaning of the case law and might therefore not be caught, in and of itself, by Article 101(1) TFEU. Whether these hypotheticals will find their way in concrete cases is, alas, a different story.

This week’s post focuses on what is probably the single most complex point of law considered in the Draft Guidelines on exclusionary abuses. It is also the keystone of the whole venture, in the sense that it could mark its fate.

The case law of the past decade has very much emphasised the need to engage in a meaningful assessment of the effects of certain practices. It is now undeniable that some conduct is only caught by Article 102 TFEU where it can be shown to have an actual or potential impact on competition.

This so-called ‘effects-based approach’, now enshrined in the case law, is not a capricious hurdle aimed at making enforcement more difficult or less effective. It is rather the acknowledgement that many, if not most, practices do not necessarily pursue an exclusionary aim and, similarly, do not invariably deteriorate the competitive process (even when implemented by a dominant firm).

The (daunting) challenge, against this background, is to get the definition of anticompetitive effects right. Getting the definition right, in this context, means, first, providing the basis for a meaningful assessment (that is, one that is not merely a formality); and, second, making it possible to discern what an effect is and what it is not.

What an effect is not: lessons from the case law

Part of the difficulty that comes with fleshing out the concept is that the case law is far more illuminating about what an anticompetitive effect is not (as opposed to what it is). The issue, as the law stands, is best approached in a negative manner.

It is clear from the case law, to begin with, that an anticompetitive effect is not synonymous with harm to consumer welfare. Therefore, the former can be established without showing the latter (and, more importantly, without the latter necessarily occurring, whether actually or potentially).

Second, a mere competitive disadvantage and/or a limitation of a firm’s freedom of action do not amount, in and of themselves, to an anticompetitive effect within the meaning of Article 102 TFEU.

The latter point has long been clear. If anything, it has become even more difficult to dispute following Servizio Elettrico Nazionale and the appeal judgment in Google Shopping. As reminded in para 186 of the latter, the fact that a vertically-integrated dominant firm discriminates against its non-integrated rivals (which necessarily places them at a disadvantage) is not abusive in and of itself.

Similarly, the fact that a practice (say, a system of standardised rebates) limits the freedom of action of a dominant undertaking cannot, in and of itself, substantiate a finding of anticompetitive effects. The clarification of this point is arguably the most valuable contribution made by Post Danmark II to the body of case law.

The challenge, in theory and in practice, is to identify the point at which a competitive disadvantage and/or limitation of a firm’s freedom of action are significant enough to amount to anticompetitive effects.

Third, the fact that the rivals of a dominant undertaking lose customers as a result of the behaviour of the dominant firms does not necessarily mean that the said behaviour has actual or potential anticompetitive effects.

According to the case law, there will be no anticompetitive effects, whether actual or potential, for as long as rivals remain willing and able to compete. Post Danmark I provide an ideal case study (if only because the Court unambiguously signalled the absence of effects in light of the facts of the case and the evolution of the relevant market).

The definition of anticompetitive effects in the Guidelines (and its practical consequences)

Capturing the essence of the case law is not an easy task. Crafting a definition that is operational (in the sense that its meaning can be grasped by national courts and authorities) is even more difficult.

The Commission could not avoid, alas, trying its hand at the challenge. In para 6 of the Draft Guidelines, it proposes to define the notion of anticompetitive effects as referring to:

‘any hindrance to actual or potential competitors’ ability or incentive to exercise a competitive constraint on the dominant undertaking, such as the full-fledged exclusion or marginalisation of competitors, an increase in barriers to entry or expansion, the hampering or elimination of effective access to markets or to parts thereof or the imposition of constraints on the potential growth of competitors‘.

The fundamental point to make in relation to this definition is that it does not fully shed light on what an effect is (and, indeed, what an effect is not).

More precisely, the definition enshrined in the Draft Guidelines could be reasonably interpreted as suggesting that virtually any competitive disadvantage amounts to an anticompetitive effect. According to the current drafting, any ‘hindrance’ to the exercise of a competitive constraint would be sufficient, in and of itself, to establish an abuse.

Such an approach may not be obvious to square with the case law described above. As rulings like Servizio Elettrico Nazionale and Google Shopping show, ‘hindering‘ rivals’ ability to exercise competitive pressure does not necessarily lead to actual or potential anticompetitive effects.

The practical consequences of such an expansive understanding of the notion of anticompetitive effects are arguably more significant than the tension with some aspects of the case law.

Accepting that any ‘hindrance‘ can amount to an anticompetitive effect means accepting that pretty much any strategy implemented by a dominant firm amounts to an abuse of a dominant position. With such a broad definition, one could always make a plausible case that the requisite threshold is met.

The assessment of effects, in other words, would become a formality (actual or potential effects would always be shown to exist), rather than a meaningful one.

One can expect, in the same vein, actors in the system to exploit the expansive understanding of the notion of abuse before national courts (which, unlike competition authorities, lack the ability to prioritise cases).

If enforcement moves down this road, it may take a while before the Court of Justice is given the chance to refine the definition of effects and correct deviations from the case law. In the meantime, the legal community will have to learn to live with a significant degree of legal uncertainty around the meaning and scope of Article 102 TFEU.

A proposed definition of anticompetitive effects

One can think of a definition of anticompetitive effects that more faithfully reflects the essence of the case law and ensures, in the same vein, that the assessment remains meaningful. The notion can be construed, for instance, as follows:

‘Anticompetitive effects exist where the practice reduces rivals’ ability or incentive to compete to such an extent that the competitive pressure to which the dominant firm is subject is reduced as a result’.

Next week, I will address the ways in which this definition can be made operational (and administrable) so that it can be applied in concrete cases. In the meantime, I would very much welcome your thoughts.

‘Remedies in EU Antitrust Law‘, my latest paper, is already available on ssrn and can be downloaded here. It is based on a keynote speech delivered at the University of Mannheim upon the invitation of Heike Schweitzer. The piece is dedicated to her memory.

Remedies are now central to successful enforcement. The correct design and adequate implementation of measures ceasing the infringement are as important, if not more, as detecting and establishing a breach of Articles 101 and/or 102 TFEU.

The days of yore, where the remedy was self-executing, are long gone. With the focus of enforcement on digital markets, effective enforcement depends on specifying, in a regulatory-like way, the measures that firms need to positively (and not just negatively) implement to bring the infringement to an end.

Against this background, the paper makes one point about current administrative practice and another one about the legal landscape.

The first point is that a ‘principles-based approach’ to remedial intervention, whereby an authority refrains from specifying the ways in which the infringement is to be brought to an end, is not compelled by law and is likely to lead to suboptimal outcomes.

Such an approach to the administration of remedies can be expected to delay effective enforcement, is opaque for third parties and is inevitably a source of legal uncertainty: it may never be entirely clear whether the firm subject to the obligations is fully complying with its duties.

The second point the paper makes is that Regulation 1/2003 was probably not designed with regulatory-like intervention in mind. This reality is neither a surprise nor a criticism: as Andriani Kalintiri and I showed empirically (see here), regulatory-like remedies were a rarity under Regulation 17.

Article 7 (unlike Article 9) of Regulation 1/2003 appears to be premised on the idea that behavioural remedies are necessarily negative in nature (that is, they impose an obligation to refrain from engaging in certain conduct) and implemented on a one-off basis.

It is suggested that Article 7 could be reformed so as to introduce a structured framework giving third parties the chance to express their views on the proposed measures and to give addressees the certainty that the obligations they implement comply with the terms of the decision.

Finally, the paper also addresses some of the ongoing debates. Competition authorities are right to be cautious about the implementation of so-called ‘restorative remedies’. It is unclear that there is a legal basis for such remedies (as a matter of EU antitrust and EU law at large).

Even if there were, it is not clear that it would be desirable to embrace restorative intervention. The system already struggles with the administration of complex, regulatory-like remedies.

Adding a layer of complexity would absorb even more resources and may venture beyond what competition authorities can realistically and consistently achieve. Restorative intervention, it is submitted, is best undertaken by means of ad hoc regulation (such as the DMA).

I would very much welcome your thoughts (nothing to diclose).

Last week’s post focused on the tripartite categorisation of practices in the Draft Guidelines on exclusionary abuses. It concluded that, while the categorisation is sound and finds support in the case law, it misses a fourth family of practices (that of presumptively lawful conduct).

‘Naked restrictions’ in the Draft Guidelines

This week’s post zooms into a sui generis category, which the Commission labels ‘naked restrictions’ in the Draft Guidelines. Practices falling within this group share two distinctive features. First, they lack redeeming virtues (that is, they can only be explained as a device to exclude actual or potential rivals). This feature is the one that makes them stand out from the rest.

Second, ‘naked restraints’ are prohibited as abusive without the need to consider their effects on competition. Contrary to what is true in the default scenarios, such effects can be safely presumed. Which makes sense. It is reasonable to expect that a practice that pursues no plausible purpose other than the restriction of competition is (at least) capable of exclusion.

‘Naked restrictions’ as a reality and a necessity

The first point to make about ‘naked restrictions’ is that they are both a reality and a necessity. That they are a reality need not be explained at length. There are concrete examples in the case law showing that these practices are an integral aspect of the legal landscape.

Predatory pricing within the meaning of AKZO is one mentioned last week, but not the only one. For instance, providing objectively misleading information to regulatory authorities, which was at issue in AstraZeneca (and is topical again in the context of the Teva decision) is also abusive by its very nature.

‘Naked restrictions’ are also a necessity. According to a well-established doctrine, a practice can be subject, either alternatively or cumulatively, to both Articles 101 and 102 TFEU.

If the behaviour under consideration is restrictive by object under Article 101(1) TFEU, it stands to reason that it is also prohibited without the need to show effects under Article 102 TFEU. Reverse payment settlements, which may amount to a ‘by object’ infringement and have been scrutinised under the two provisions (including in Generics), illustrate this point particularly eloquently.

‘Naked restrictions’ or ‘by object’ abuses?

The second point relates to the labelling of practices. While a relatively minor issue in a field that places substance above form, it makes sense to say a word about it.

The question, in essence, is whether to label these practices ‘naked restrictions’ or ‘by object’ abuses. It seems to me that the latter (‘by object’) is preferable (and the one that the final version of the Guidelines would ideally adopt).

This is so, to begin with, because it is the label that the Court has consistently used since Generics (and then Superleague, and then Google Shopping). The reiteration of the formula suggests that it is now part of the acquis on Article 102 TFEU.

A second reason why the ‘by object’ label would be preferable is that its scope of application is very similar, if not identical, to the scope of ‘by object’ infringements within the meaning of Article 101(1) TFEU.

The Court has reiterated since Generics (most recently in the Servier saga and Banco BPN), that an agreement is restrictive by object under Article 101(1) TFEU when it cannot be explained other than as a means to restrict competition.

As the Court put it in para 56 of Banco BPN: ‘[…] [A]n exchange of information which, although not formally presented as pursuing an anticompetitive object, cannot, in the light of its form and the context in which it occurred, be explained other than by the pursuit of an objective contrary to one of the constituent elements of the principle of free competition must be regarded as constituting a restriction by object‘.

It is sufficient to compare and contrast the Court’s position in Banco BPN with the definition of ‘naked restriction’ given in the Draft Guidelines (in turn borrowed from para 71 in AKZO) to realise that they both concern the same range of practices. The case law would be cleaner and easier to navigate if the same phenomenon were labelled identically irrespective of the provision.

Rebutting the presumption of anticompetitive effects: ‘real and concrete possibilities’

The third, related issue concerns the rebuttal of the presumption of anticompetitive effects underpinning the qualification of a ‘naked restriction’ as an abuse of dominance. The Draft Guidelines point out that it should only be possible to rebut this presumption in rare, if not exceptional circumstances.

It seems to me that this point of principle cannot be disputed. Where the Draft Guidelines could be more specific is in relation to the (exceptional) instances where the presumption can be rebutted.

The currrent version of the document helpfully (and rightly) mentions that the bar is significantly higher than in Intel-type scenarios. However, it does not elaborate much further and leaves a great deal of scope for speculation.

The most elegant solution, and the one that is wholly aligned with the existing body of judgments, would be to draw inspiration straight from the case law on restrictions by object under Article 101(1) TFEU. One should bear in mind, in this sense, that ‘Articles 101 and 102 TFEU must be interpreted consistently‘.

The recent Servier saga has helpfully clarified (in light with the preceding cases and with AG Kokott’s own analysis in her Opinion in Generics) that there is no restriction, whether by object or effect, where there is no actual or potential competition in the relevant economic and legal context (that is, where there are no ‘real and concrete possibilities‘ of entry).

There would be no ‘real and concrete possibilities’ of entry, for instance, where the regulatory regime prevents actual or potential competition (an example at stake in E.On Ruhrgas and cited with approval by AG Kokott in the abovementioned Opinion to make this very point). The same would be true where no other firm has the ability or incentive to enter the market (for instance, because none has not taken the requisite preparatory steps to place competitive pressure on actual competitors).

As far as ‘naked restrictions’ are concerned, it is submitted that the presumption of anticompetitive effects should only be rebuttable in these same (and very narrow) scenarios. These are, incidentally, the scenarios where (as in E.On Ruhrgas) the presumption of effects underpinning a ‘by object’ infringement has been successfully rebutted by the parties in Article 101(1) TFEU disputes.

As it happens, the case law on exclusionary abuses already provides scenarios where there were no ‘real and concrete possibilities‘ of entry (and thus no abuse). One of these scenarios is found in the BEH ruling, where the General Court found that the regulatory context precluded actual or potential competition. An example of the second scenario can be found in Qualcomm (exclusivity). In that case, the customers of the dominant undertaking had no realistic alternatives.

Refining the Guidelines to make it explicit that the presumption of anticompetitive effects underpinning a ‘naked restriction’ can only be rebutted where the dominant undertaking can show that there are no ‘real and concrete possibilities‘ of entry in the market affected by the practice and therefore no actual or potential competition would greatly contribute to the clarification of the existing law of exclusionary abuses.