From the Courtroom to the Opinion: AG Wahl resets the standard for rebates in Intel

United Kingdom

This article was produced by Nabarro LLP, which joined CMS on 1 May 2017.

For those that were in the Intel Courtroom on 21 June 2016, the direction of AG Wahl’s Opinion four months later on 20 October 2016 - suggesting that the General Court’s analysis and findings on loyalty rebates were wrong, and the case should be remitted to it - comes as no surprise. It is the clarity and balance in the analysis, and the way Wahl binds his views into an interpretation of the existing case law, that is so ground breaking.

As with my previous briefings on this case, I will concentrate here on the side of the case concerned with rebates and exclusionary abuse on the part of dominant firms. The important further issues of procedural fairness and extraterritoriality are for discussion elsewhere.

To start in the middle, what Wahl essentially says is that in order to condemn as abusive certain types of rebate schemes, including at least loyalty and target rebates, while you may not necessarily have to carry out a full effects analysis, you do have to analyse them in all the circumstances to see if they really are anti-competitive.

That analysis of “all the circumstances” includes looking at “coverage”, i.e. the proportion of the market to which the rebate scheme applies. And this needs to be looked at from the perspective of the rivals – what portion of their opportunities are foreclosed. It includes looking at duration. And it may include other things such as looking at the famous “as efficient competitor” test, in particular where analysis of other aspects is “equivocal” or where, as in Intel, the Commission had done so.

While there is a starting presumption of unlawfulness for these types of schemes, the analysis of those factors needs to “confirm” anti-competitive foreclosure. “More likely than not” is specifically stated to be insufficient for this purpose. If the analysis of the various factors is “inconclusive”, a full effects analysis must be carried out to determine whether the rebate programme qualifies as abusive.

One can quibble with language about “actual” or “likely” effects, or “likelihood” and “capability” of effects (Wahl, apparently in an effort to reinterpret rather than contradict existing case law, equates the two). But semantics should not distort Wahl’s clear message. No type of rebate is immune from analysis of its terms and full context in order to determine whether it is anti-competitive, and if the result is not that it is clearly anti-competitive a full effects analysis must be carried out.

This notion of a two-stepped approach, where the threshold for a conclusion of unlawfulness at the first stage is high, and where an “inconclusive” result at the first stage requires moving to the second stage of analysis, can be traced back to Wahl’s Opinion on ‘object’ restrictions under Article 101(1) in Cartes Bancaires. See for example the concept of agreements with “ambivalent” effects not qualifying under restriction by object at paragraph 56 of that Opinion.

There is a very real chance in practice that the first stage of analysis does lead to a result that is inconclusive – in Wahl’s view that is the result on the facts of Intel, where coverage of the market was only 14% (compared with much higher figures in previous cases) and duration of the arrangements was short.

Wahl makes several other common sense points that are refreshing given the restrictive approach in the case law to date. For example, just because customers stay with the dominant firm, that does not itself justify an assumption that a scheme is anti-competitive. The dominant firm’s product may be better. Or they may be responding to the wishes of their own customers.

Similarly refreshing is Wahl’s recognition, long overdue in this area, of the need for some consistency of approach between unconditional and conditional pricing practices – and the need therefore for some alignment with the more effects-based approach taken by the Courts with respect to the former. And Wahl exposes the circularity of the General Court’s approach in fastening on a “strategy of foreclosure” in linking two types of practices under review (rather than analysing them to see if they infringed first).

This is a highly complex area, and Wahl’s Opinion, in focusing in particular on loyalty rebates, is not completely comprehensive. One issue not fully examined is volume rebates. Wahl says (more than once) that they are presumptively lawful, but his citations are to comments in the case law about quantity rebates creating scale efficiencies which are “given back” to customers. As I have argued previously, that is not of much use in the real world. Wahl observes, in one of his footnotes, the General Court’s qualification in Michelin II that the presumption does not apply when a volume scheme has a “loyalty inducing effect” – but he does not explain how this squares with his view that there are only two types of rebates (presumptively lawful and presumptively unlawful).

Ultimately, however, the key point is that under Wahl’s approach, presumptions are just that: starting presumptions. They are but the beginning from which the real analysis needs to take place. And at least for loyalty and target rebates there are potentially two separate stages to that analysis (if the first stage is inconclusive), before such rebates can be condemned as abusive.

Of course, in the end, what happens in the Courtroom or even on the pages of the AG’s Opinion is not what counts most. We will have to wait for the judgment to see if Wahl’s balanced and clear approach is carried through to the final ruling.