Relaxing whilst doing Competition Law is not an Oxymoron

The Commission sends an SO to Apple: common carrier antitrust picks up speed

with 10 comments

A Short History of the Spotify–Apple Music Beef - The Ringer

As readers will know, the Commission sent a Statement of Objections to Apple last week (for the press release, see here). The investigation focuses on the firm’s practices in relation to the conditions under which it provides access to its app store; it is confined to music streaming services. Most of you probably remember that the case traces its origins back to the complaint brought by Spotify, which concerned the 30% commission Apple ask on sales taking place via its app store.

According to the press release, the Commission has reached the preliminary view that the abovementioned practice amounts to an abuse of a dominant position. The document also mentions the ‘anti-steering provisions’ whereby app developers are restricted in their ability to inform users about alternative purchasing options.

The case has long intrigued the competition law community. The theory of harm that the Commission would pursue was not immediately obvious to infer from the publicly available information. Was the investigation really about exclusion, considering that Spotify is by some distance the market leader on the relevant market? If so, what are the conditions to assess the legality of the conduct? Was the case about exploitation instead?

The press release hints at some answers in this regard. Generally speaking, it suggests that common carrier antitrust (an emerging interpretation of competition law provisions and a new approach to enforcement) is picking up speed. I would note three aspects in this regard:

  • First, the press release suggests that the lawfulness of potentially exclusionary conduct does not depend (or, rather, no longer depends) on an assessment of anticompetitive effects.
  • Second, the press release signals that the Commission is ready to question firms’ business models in the digital sphere: what EU competition law, for decades, was reluctant to challenge absent exceptional circumstances, has now become a central feature of enforcement.
  • Finally, the press release hints at an additional feature of common carrier antitrust: it would seem that each ecosystem is deemed a market on its own.

A farewell to anticompetitive effects?

Modern EU competition law, as interpreted by the Court of Justice, is based on the idea that the vast majority of potentially abusive practices are unproblematic where they are unlikely to have anticompetitive effects (in refusal to deal cases, the test is even stricter). This is the interpretation underpinning the most recent case law (including Deutsche Telekom, TeliaSonera, Post Danmark I and II, Intel, MEO) and the Commission’s Guidance Paper (in which the authority committed to prioritising cases that would likely lead to anticompetitive foreclosure).

The principles of contemporary EU competition law are not obvious to reconcile with the Apple investigation. In this sense, the case hints at the rise and consolidation of common carrier antitrust.

Rulings like Deutsche Telekom and MEO show that a raising rivals’ costs strategy is not anticompetitive in and of itself. An evaluation of the likely impact of the conduct is a precondition for the application of Article 102 TFEU (or, indeed, merger control and Article 101 TFEU). According to this case law, the Commission would need to establish the anticompetitive effects of Apple’s app store-related behaviour.

A cursory look at the relevant market suggests that showing the exclusionary impact of Apple’s practices on the market for music streaming services is anything but an easy task. The fact that the original complainant in the case is (and has been for a while) a market leader one of the factors in this regard. The fact that music streaming is accessible in many ways (that is, not only via Apple’s app store or, more generally, Apple devices) is another one.

The press release is interesting in that it suggests that the Commission believes that it can establish an abuse of an exclusionary nature without showing that the practice is likely to have anticompetitive effects. It would seem that the case is predicated on the idea that the Commission can discharge its burden of proof merely by showing that the practices distort competition by raising rivals’ costs. In this sense, it signals a move away from Deutsche Telekom (with all the implications and/or complications that follow).

The move away from anticompetitive effects does not seem to be an isolated instance. I have the impression that other cases in the digital sphere will be conducted following the same approach. One can think, in particular, of the Amazon case (which I discussed here). As is true of the Apple investigation, the use of non-public data by Amazon cannot be assumed to lead to the exclusion of rivals on the relevant markets (and might very well inject competition in Amazon’s marketplace).

This new approach to the assessment of practices comes at a time when the fundamental tenets of modernisation are being challenged. It has become relatively frequent to read that the effects-based approach has gone too far, or that it would place an undue burden on authorities in digital markets.

The move away from anticompetitive effects raises a number of fascinating questions. I will mention just two here. The first is whether the Court will agree to depart from the case law mentioned above and embrace the Commission’s expansive interpretation of its powers. The second is whether this new approach makes it possible to meaningfully constrain administrative action. To the extent that it equates a competitive disadvantage with harm to competition, its scope of application seems to lack clear boundaries (it is a criterion that seems fulfilled pretty much always and everywhere).

In this sense, the post-modernisation approach reminds one of the pre-modernisation times, when the Commission had a tendency to equate a limitation of a firm’s freedom of action with a restriction of competition.

Competition law and business models: what is the counterfactual?

The Apple investigation is also an example of another aspect of the emerging approach to enforcement: the Commission is no longer reluctant to challenge firm’s business models. Traditionally, the EU competition law system was deferential to firm’s strategies. It did not question, absent exceptional circumstances, a company’s decision to produce exclusively in-house. Similarly, the core of distribution methods like selective distribution and franchising are deemed prima facie lawful irrespective of their effects.

Common carrier antitrust is much less deferential to the central aspects of a firm’s business model. This approach to enforcement is fraught with challenges for an authority. First and foremost, one cannot simply assume that the undesirable aspects of a company’s core strategy can be removed without consequences. For instance, one cannot take for granted that a manufacturer will rely on franchising if it is required to accept competing products in the franchisees’ premises. Similarly, one cannot simply assume that forcing a company to deal with third parties will have no impact on incentives to invest and innovate.

Generally speaking, tweaking a firm’s business model via competition law enforcement requires a careful evaluation of the counterfactual. When it is said that a business model restricts competition, the question should be: compared to what? What would the world look like if the business model was a different one? How are the different aspects of a business model (both the desirable and undesirable) intertwined? I was reminded of these questions when reading the press release in the Apple case.

The press release suggests that a central aspect of the Commission’s case is that Apple’s 30% fee leads to higher prices for consumers (as the fee is passed on to subscribers). It is unlikely that challenging Apple’s ability to charge a fee to app developers will come for free; it will most probably be compensated elsewhere (possibly in the form of more expensive devices and/or services for end consumers). Once again: one cannot assume that the undesirable aspects of a business model can be removed without consequences.

I look forward to seeing how these central questions are addressed in the case. And I look forward to your comments, in particular if you see things differently. As you know, I have nothing to disclose in this or indeed in any other matter.

Written by Pablo Ibanez Colomo

3 May 2021 at 2:09 pm

Posted in Uncategorized

10 Responses

Subscribe to comments with RSS.

  1. Thank you for this very interesting and much appreciated piece. On the idea that this may represent a move back from an effects-based approach to Article 102, would you not expect the Commission to treat the pass on of part of the 30% fee as an anti-competitive effect in its ultimate analysis? To me, that seems to go beyond simply raising a rival’s costs as it makes the end product that competes with Apple downstream more unattractive to consumers and has – presumably – led to some diversion to Apple Music (even if that stops far short of foreclosure given Spotify’s clear market leading status). I look forward to receiving your thoughts.


    3 May 2021 at 2:31 pm

    • Thanks, Charlie! You describe effectively the analysis under the common carrier antitrust approach.

      Such an interpretation amounts to equating a competitive disadvantage (‘some diversion to Apple Music’) with an anticompetitive effect. It is not, however, how anticompetitive effects have been interpreted by the Court. A competitive disadvantage indeed leads to some diversion by its very nature, but this is not enough, according to the case law, to trigger Article 102 TFEU (think of Deutsche Telekom, where the Commission made an argument to this effect but did not win the day, Post Danmark I or MEO).

      If such an approach (ie ‘some diversion’ being enough to trigger Article 102 TFEU) captured the essence of the case law, then the scope of the prohibition would be much broader and, crucially, it is not obvious to see how administrative action would be meaningfully constrained.

      Pablo Ibanez Colomo

      3 May 2021 at 6:25 pm

      • Thank you for the response! I can see the parallels with Post-Danmark I (where the complainant stayed on the market, kept fighting and even won back a customer). However, I would be very interested in your views on why, from a consumer welfare standard (which would be a higher threshold than that which needs to be applied in any event) an ultimate price rise for end consumers of music streaming services should not be characterized as an anti-competitive effect of the relevant practice? Many thanks again


        3 May 2021 at 9:45 pm

      • Thanks again, Charlie!

        If we were to apply the consumer welfare standard in this context, the question would be whether the competitive pressure on the dominant firm is reduced to such an extent that it gains the ability to raise prices (or otherwise affect the conditions of competition). This assessment (which you can find in the classic raising rivals’ cost literature by Salop and others) strikes me as different from the approach at which the press release hints. We will have to wait and see, I guess!

        By the way: we fully agree that evidence of harm to consumers is not required in the EU competition law system.

        Pablo Ibanez Colomo

        4 May 2021 at 8:14 am

  2. The resolution of these two questions will indeed be critical and jumped out at me when reading the Press Release as well and especially the press briefing Q&A (, where the EC repeatedly referred to App Store customers as belonging to Spotify, and hence putting Spotify at a disadvantage vis-a-vis these customers being “unfair” (see discussion starting at minutes 13:50, and 17:47).

    Two other points of note: (1) the Commission defined the market incredibly narrowly, i.e. according to the type of business user (“the market for the distribution of music streaming apps through its App Store.”); and (2) the Commission appears to be leaning towards not imposing a specific remedy, but instead just saying “cease the infringement” (Q&A, minute 8:00). As noted in your piece “Legal Tests in EU Competition Law: Taxonomy and Operation” ( this could impact the application of the legal test and its constituent elements.


    3 May 2021 at 3:49 pm

  3. I entirely agree. The apparent expansion of the boundaries of Article 102 into a tacit lex specialis that we are seeing has made it increasingly difficult to conduct any form of self-assessment for large tech companies, or indeed anyone else. Since the ability to self-assess infringement risk is a major plank of the whole competition law system, this materially undermines the integrity of the regime. Ultimately, this development persuaded me of the merits of moving to an ex ante regulatory approach for big tech, as the lesser of two evils. Of course, much could go wrong on that front but it will hopefully not bend antitrust out of shape. As you note, in the meantime it remains to be seen how these Article 102 cases develop and hence what the wider impact on the law will be.

    Becket McGrath

    5 May 2021 at 1:35 pm

  4. […] had sent a statement of objections (“SO”) to Apple on 30 April is very different from the reaction of Pablo Ibanez Colomo on his Chillin’ Competition Blog. I was pleased with the issuance of this […]

  5. […] scholars such as Ibáñez Colomo have argued that competition law has traditionally been careful not to interfere with the business model of […]

  6. […] was not surprised by the Commission’s announcement. As I explained a while ago (see here), the exclusionary claims in the case were not obvious to reconcile with the case law. The modest […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: