Chillin'Competition

I have uploaded a new paper on ssrn (see here) on Anticompetitive Effects in EU Competition Law.

What was the impetus behind the paper? The ‘effects-based approach’ have long been part of discussions, yet there are only a handful of judgments addressing the analysis of effects in a meaningful way.

There is still some uncertainty about what anticompetitive effects are and how they are measured in concrete cases. On the other hand, the Court has already clarified the key issues. Against this background, I felt I could contribute to the debate by bringing together the various strands of the case law and present them in a single framework.

The paper tries to organise what we know about effects around the main variables that shape its meaning and scope, and in particular:

• The time dimension: actual and potential effects (and retrospective and prospective analysis).
• The dimensions of competition (inter-brand and intra-brand) and the counterfactual (ex ante and ex post).
• The meaning of effects: effects can mean many things, from a competitive disadvantage, to harm to the market structure to harm to consumers. The analysis is particularly sensitive to the way this variable is defined.
• The threshold of effects: again, the analysis would vary substantially depending on whether it is enough to show that effects are plausible or whether instead it is necessary to establish that effects are likely to happen.

It would not be obvious to summarise the paper in a single post (I will probably add some dedicated entries). It may make sense to mention, however, that the notion seems to have acquired a clear meaning over the years.

In particular, it seems like effects are more than a competitive disadvantage and/or a limitation of a firm’s freedom of action (think of Deutsche Telekom, Post Danmark I, MEO, Maxima Latvija, Generics and, in the context of merger control, Tetra Laval and GE/Honeywell).

We know from experience that having an edge over rivals is not necessarily fatal for competition; it may even spur rivalry (think of Post Danmark I, where rivals were able to withstand a below-cost price campaign). A limitation of a firm’s freedom of action is not enough either (think of Generics, where the Court held that the effects should be more than the impact of each individual agreement).

What matters, the case law suggests, is whether firms’ ability and/or incentive to compete is affected by a practice or transaction, and this to such an extent that competitive pressure is reduced. Thus, no effects would exist where firms on the market are still willing and able to compete.

The threshold of effects is another contentious issue. We all know that effects can be actual or potential. The real question, when the analysis of potential effects is at stake, is whether it is sufficient to show that harm is plausible or it is ‘more likely than not’ (to use AG Kokott’s expression in Post Danmark II).

If we pay attention to what the Court does, it appears that, regarding ‘by effect’ conduct under both Articles 101 and 102 TFEU (as well as mergers), the negative impact of a practice should be probable (‘more likely than not’) and not simply plausible. Just think of how the analysis was actually conducted in, inter alia, Delimitis, TeliaSonera, Post Danmark II, Kali & Salz and Microsoft/Skype. 

As far as ‘by object’ conduct is concerned, it is sufficient to show that harm to competition is ‘plausible’ (the threshold, in other words, is much lower). Bananas or Toshiba are clear examples in this regard, and reveal how much the analysis differs between the ‘by object’ and ‘by effect’ stages.

As I say, I will probably follow up with more posts on the paper. In the meantime, I very much look forward to your comments.