The Norwegian Competition Authority has issued a draft paper on abusive (price) discrimination, outlining how the agency plans to approach the matter, which should be most welcome. Essentially, the legal position is at best unclear, creating room for misapplications. The otherwise famous Post Danmark I case, e.g., originated in a misguided national attempt to apply Article 102 to differences offered in terms but not clearly cable of thwarting of competition. Moreover, while promising in 2005 to issue a separate paper on discriminatory abuse, DG COMP has never delivered on this, and in all likelihood, never will, leaving the matter open and unsettled.

 

The Norwegian Competition Authority offers five essential points

In the draft paper, the Norwegian Competition Authority offers five essential lessons. First and foremost, the paper rebuts seeing discrimination in itself as abusive as discrimination often would be benign or pro-competitive. Secondly, the paper refutes seeing dissimilarities in terms offered to large vs small customers as per se problematic. This is important, as the concept can trace its lineage back to a misguided ambition of protecting smaller companies, thus cutting any link with this (false) theory of harm. Thirdly, the paper explains how (price) discrimination may come in many shades and forms, including being price or non-price based and involving (see also Scanlines Sverige AB, recital 276):

  1. applying dissimilar conditions to equivalent transactions, which could involve that some customers get better terms, or
  2. applying equivalent conditions to dissimilar transactions, which could involve all customers offering identical terms regardless of differences, e.g., in the quantum they source.

Fourthly, relying on the Advocate General’s Opinion in MEO, the paper by the Norwegian Competition Authority explains how exclusionary discrimination involves primary and second-line discrimination, where the former is directed at foreclosing a competitor (upstream or downstream), while the latter is directed at thwarting competition downstream (or upstream) and only relevant when the perpetrator is not active here. The focus of the paper is second-line discrimination, as first-line discrimination often would be covered by other concepts of abuse, e.g., margin squeeze or refusal to supply. Fifth, and finally, the paper underscores that an anti-competitive effect must be identified but does not offer anything further on how to undertake this.

 

The paper is most welcome but falls short of providing a coherent approach

The paper from Norwegian Competition Authority should be considered most welcome, as abusive (price) discrimination remains the most misunderstood form of abuse, and essential cases such as MEO and BdKEP/Deutsche Post AG are challenging to read and apply. However, the latter is an overlooked jewel (not even cited in the paper by the Norwegian Competition Authority) when it comes to the matter of discriminatory abuse. E.g. DG COMP (recital 93) uses the opportunity to explain how:

The wording [of Article 102] covers three types of discrimination, the first two of them exclusionary and the last one exploitative: (i) the customer of the dominant firm is placed at a competitive disadvantage vis-à-vis the dominant firm itself; (ii) in relation to other customers of the dominant firm; or (iii) the customer suffers commercially in such a way that its ability to compete in whatever market is impaired. It is obvious that type (i) and (iii) do not require a competitive relationship between the two comparator groups.”

Combined with the Advocate General’s Opinion in Meo, it becomes apparent that the concept of discrimination not only covers three forms of abuse, of which two are exclusionary, and one exploitive, but also that more advanced observations can be provided. These include how Article 102(c) potentially can be applied to:

  1. Horizontal (exclusionary) discrimination (referred to as primary-line discrimination by the Norwegian Competition Authority), initiated for the purpose of foreclosing competitors by targeting actual or potential customers with selective price reductions or other favours. Moreover, this includes foreclosure of upstream and downstream markets secured by preferential treatment of subsidiaries and internal departments of the vertically integrated company. The foreclosure could thus have a vertical element to it, but because the victim is a direct competitor (upstream or downstream), the foreclosure remains horizontal.
  2. Vertical (exclusionary) discrimination (referred to as secondary -line-discrimination by the Norwegian Competition Authority) was initiated for the purpose of twisting competition in other markets e.g., for the benefit of a preferred trading partner (but not a subsidiary or internal department). While also directed upstream or downstream, the potential abuser has no direct interest in the foreclosure as it remains inactive in any of the affected markets. Hence, the foreclosure is (truly) vertical.
  3. Exploitative discrimination that in practice involved national-based discrimination and henceforth potentially individualized pricing. However, the abuse is essentially exploitive and should not be confused with discrimination but reviewed under the legal standards for exploitation.

The Norwegian Competition Authority does not spell things out as clearly as done above, but the meaning is clear. Unfortunately, the paper offers nothing in terms of when to accept horizontal or vertical discrimination as anti-competitive. However, accepting horizontal discrimination as nothing short of a traditional exclusionary practice, it would be logical to turn to the most recent practices from the Court of Justice, including Intel, commanding analysis of the ability to foreclose an as As Efficient Competitor (AEC). In most cases, this would involve evaluating if the offered terms secured coverage for the dominant undertakings LRIC on the markets affected by the (alleged) abuse. Only then could a preferential treatment of customers lead to the foreclosure of an As-Efficient Competitor, and thus, rightly be held anti-competitive. Of course, since this has not been cited directly in discrimination cases, it might represent wishful thinking but would nevertheless represent the next logical evolution – and thus advances the more effect-based process further.

 

How to treat vertical discrimination?

Vertical discrimination covers, as explained in BdKEP/Deutsche Post AG, discrimination of downstream trading parties (or upstream suppliers). However, void of elements thwarting the Single Market, e.g., nationality-based discrimination, practice is limited, indicating that it neither has nor should be, a priority. This is a logical decision, as the dominant undertaking often lacks an incentive to pursue foreclosure of vertical markets. Even the ability might be lacking, unless in a monopolistic position, as any attempt to pursue a foreclosure could undermine the dominant position and thus be unprofitable in a longer perspective. Nevertheless, guidance might be found in Portugal v. Commission (cited by the Norwegian Competition Authority but not detailed substantially). Here a linear and quantum discount had de facto benefitted domestic air operators, making it abusive. Not because some got better terms, as this was an inherent feature in quantum discounts, but due to high thresholds only attainable by a few particularly large partners and the up to 30% differences in the offered terms. Embedded in this differential treatment, even if non-objective, is in itself insufficient for identifying an abuse, as selective discounts always will benefit some at the expense of others.  Moreover, the disadvantages must be assessed in a tangible manner, which in Meo led the Advocate General to recommend looking at the relationship between the levied prices and total costs, making the analysis resemble a margin squeeze test.

 

The logical standard would be a margin squeeze approach

There is some logic to this, as vertical discrimination initiated by a vertically integrated operator, directed as favouring downstream group interests, typically would fall within the notion of a margin squeeze. Usually, abusive margin squeeze would require downstream prices that do allow for recovery of the dominant undertakings LRIC thus capable of foreclosing an As-Efficient Competitor. Translated to vertical discrimination, this means that the evaluation in Portugal v. Commission should have moved on to demonstrate how the 30% differences accounted for a substantial portion of the downstream costs and thus capable of creating a foreclosure. Essentially what the Advocate General suggested in Meo. Unfortunately, the Court in neither Meo nor Portugal v. Commission embraces this, while it might be possible to deduce this from the latter. Not explicitly, but based on the fact that airlines often operate on the margins, making it plausible that even minor discounts would be translated into a competitive advantage in the downstream markets.

 

The paper is also silent when it comes to exploitive discrimination

Further to the lack of a framework for evaluating exclusionary discrimination, the paper from the Norwegian Competition Authority also refrains from touching upon the matter of exploitive discrimination. Not only in general, but in particular, with respect to the matter of perfect price discrimination, and whether Article 102 (c) covers this. Perfect price discrimination (in economic theory referred to as first-degree price differentiation) is when the price is perfectly aligned with each customer’s payment willingness, maximizing the producer surplus at the expense of the consumer. Usually, this is untenable, but the proliferation of the digital economy has provided unprecedented opportunities when it comes to targeting customers. I think DG COMP would see exploitive abuse, but case law does not support this unequivocally. In e.g. United Brands (see paragraph 228), the Court appeared willing to accept local prices fluctuation if reflecting differences in payment ability, and in this rebutting perfect price discrimination as abusive per se. Naturally, United Brand is an old case, that might no longer hold, making it even more unfortunate that the Norwegian Competition Authority did not touch upon the issue.

 

Next step is the final version of the paper

The paper published by the Norwegian Competition Authority is a draft, and comments are welcome until 27 January 2022. Following this, a final version will be published, and even if it does not take any of my suggestions into consideration, it would still represent a most welcomed development. Not only because the legal position is at best unclear, but also because the proliferation of the digital economy might create renewed interest from enforcers in checking discrimination.


________________________

To make sure you do not miss out on regular updates from the Kluwer Competition Law Blog, please subscribe here.

This page as PDF

Leave a Reply

Your email address will not be published. Required fields are marked *