More on AG Wahl and restrictions by object: ‘pay-for-delay’ settlements as a case study
A while ago I wrote a post and engaged in some follow-up comments on the issue of restrictions by object. But since Alfonso is busy these days and has shown some persistence in chasing me to have me write another guest post, I thought it a good idea to add a few more thoughts on the matter. I see value in doing so given that the discussion in the preceding post remained (to my regret) overly abstract. I tell myself that if I illustrate my points by relating them to some on-going disputes/investigations, they may become clearer, and might even spark more discussion.
I explained back in March that the ECJ does not see the notion of restriction by object as a presumption of the likely effects of the agreement. I know this is a very popular understanding of Article 101(1) TFEU, but I see a clear difference – and so does the Court, may I add – between understanding what the agreement is all about (Article 101 TFEU refers explicitly to its ‘object’) and establishing its likely (negative) effects on the market. A ‘naked’ price-fixing agreement between competitors is prohibited irrespective of whether collusion can realistically be sustained on the relevant market (that is, irrespective of whether there are reasons to believe that the parties will ever be credibly committed to restricting competition). When reading the case law, it is pretty clear to me that the real question is whether the agreement is a plausible source of efficiency gains (there are myriad examples where this approach has been followed, some of which I mentioned in the other post). Put differently, the true issue is whether it is realistic to expect pro-competitive effects from the agreement in light of the context in which it is implemented.
Allow me to illustrate these ideas by reference to the on-going debates around ‘pay-for-delay’ settlements (Alfonso already wrote about this some time ago). It is fairly clear that a ‘naked’ (and the word ‘naked’ cannot be emphasised enough) agreement between two competitors whereby one of them agrees to delay the launch of a product amounts to a restriction by object within the meaning of Article 101(1) TFEU. The question is whether the agreements at stake in cases like Lundbeck can be likened to such ‘naked’ restrictions. Addressing this issue requires understanding, first and foremost, the point of these agreements in their context. What becomes immediately apparent in this sense is that they cannot be said to be ‘naked’. There is something else to these agreements, namely a background dispute between the parties relating to the validity or to the infringement of a patent. From this perspective, the question could be rephrased as one of whether putting an end to such a dispute by means of a settlement can be likened to a cartel agreement.
To me, the answer is a clear no. Nobody would deny that out-of-court settlements are an efficient way to deal with disputes. In paragraph 235 of the recently issued Guidelines on technology transfer agreements, the Commission is very explicit in this regard. If this is so, and to the extent that there is genuine uncertainty about the ability of a generic producer to enter the market, the applicable case law suggests that Lundbeck-like settlements should only be deemed to restrict competition after a careful assessment of their effects under Article 101(1) TFEU. By the same token, the ‘object’ category would only be appropriate where it is clear beyond doubt that the generic producer would have been able to enter the market without infringing the patent(s) in question or it is clear beyond doubt that the said patent(s) are invalid. Only then would it be justified to assess them in the same way cartels are (in such a scenario, the restraints would in reality be ‘naked’, as there would be no actual dispute to settle).
Chillin'Competition 
i would say that the arrangements you talk about are all clear market sharing arrangements,which are – due to the “type” of restriction inherent in them,ie the exclusion of a competitor from the market by agreement – by object.and then in certain special cases,you may still try to show they satisfy 101(3) and argue with efficiencies,proportionality,etc.
A general difficulty with turning around object restrictions as you propose is that you lose the criminal character of such restrictions: why do you punish extremely severely arrangements where “no real good” is coming from them as opposed to arrangements where “big harm is expected”?
Asimo
25 May 2014 at 11:43 am
From a more general constitutional / phylosophical perspective: the overall rule in market economies is freedom of contract and the prohibition to make a cartel “contract” is an exception,which is to be construed narrowly and where you need pressing reasons to make a prohibition.Hence you have to see cartels as being harmful and not merely as “not efficient”…there are lots of ineffocient,economically useless contracts and – for the above reasons – you cannot prohibit them!
Asimo
25 May 2014 at 12:30 pm
Hi Asimo,
Great points. Note that, if your logic were to be followed, then non-compete obligations such as the one examined in Remia would be object restrictions (remember it was an agreement whereby one of the parties agreed not to compete for a particular period after the sale of a business). Someone might cry ‘market sharing’, but we know very well from that judgment that non-compete obligations may fall outside Article 101(1) TFEU altogether when they are ancillary to a transaction that may otherwise not have taken place. Again, the main point of this series of posts is to remind readers that thinking about ‘categories’ of agreements or putting ideal ‘types’ of restrictions into pigeon holes does not capture the essence of the case law, which is far more subtle and sophisticated.
In a different vein, you seem to take too many things for granted in your comment. You assume that the generic producer is a ‘competitor’, but this is already far from clear. Is a generic producer a ‘competitor’ if the only way in which it can enter the market is by infringing a patent? Is a generic producer a ‘competitor’ if there is no decision about the validity of a patent? If these questions remain open, the case cannot be about a ‘clear market sharing’ agreement. This is precisely what makes pay-for-delay exciting and very different from plain vanilla cartels.
Pablo Ibanez
25 May 2014 at 2:17 pm
I have to admit that in case of object type restrictions i am very much in favour of pigeon holes and clear, strict well-defined categories and – yes – less (!) economic analysis: these are cases where the authority is exempt from its overall obligation to prove harmful effects and thus it should always be foreseeable for market players to know whether they are in an illegal area or not.
You are though fully correct to state that remia is also an important exception from 101 – ancillary restraints are indeed special cases,which i should have mentioned next to 101(3).And remia is an exception…will write more if i have driven my family back to the capital:)
Asimo
25 May 2014 at 5:08 pm
Apologies for not coming back – this week was much longer than expected.
Lets see remia first:I think you would agree with me that if the horizontal non-compete had been signed by the buyer (as opposed to the seller for a business that is unrelated to the transfer) or simply the seller were to agree to an unlimited term noncompete for the entite Galaxy:)),these would be clear object type restrictions(as forms of market sharing),which would be contrary to 101(1) even if no one implemented or observed them…this shows that remia is nothing more than a narrow sub-exception from the overall exception of object restrictions from the global prohibition of anti-comp agreements.
I agree that there could be seemingly object type arrangements,which in the end (i) turn out to be ancillary restraints (remia),(ii) not at all object type arrangments if properly seen in economic context,ie under the overall test of what is in the object box and what is not or (iii) could escape by arr 101(3).but this should not derail us from the necessity to see object arrangements as what they are,ie particularly serious,exceptional infringements where the authority is exemptred to prove hamrful effects.
As a result,for example,the analysis that is to be undertaken has to be “easy” and should not result in a full-blown economic analysis as required for effect type infringements.if we do not have a clear object category,we would end up in a world where all cartelists would be rushing to prove why their market context is “special” while all competition authorities would be rushing to show why a any arrangement that is hard to assess by effects is by object restrictive to to the “special” economic context (see allianz hungary).
Asimo
1 June 2014 at 2:58 pm
Pablo – I agree with your ideas on this post, even more so post-Groupement des Cartes Bancaires. For once, I think para.54 of the judgment (“in order to determine whether an agreement between undertakings or a decision by an association of undertakings reveals a sufficient degree of harm to competition that it may be considered a restriction of competition ‘by object’ within the meaning of Article 81(1) EC, regard must be had to the content of its provisions, its objectives and the economic and legal context of which it forms a part. When determining that context, it is also necessary to take into consideration the nature of the goods or services affected, as well as the real conditions of the functioning and structure of the market or markets in question”) makes perfect sense when analyzing pay-for-delay arrangements. Pharmaceuticals is a market driven by innovation and intellectual property rights. In this context, like you say, it is also far from clear whether all generic producers can be considered competitors of original manufacturers. If authorities merely focus on the object of delaying the generic entrant in every pay-for-delay case and conclude that this is an object infringement, without considering the actual/potential effects of delaying the specific generic manufacturer and its credibility as a competitor, this could have a greater negative future impact on the manufacturers’ motivation to innovate.
CB judgment makes clear that object restrictions should be interpreted restrictively – with such mixed views in academia/jurisdiction on pay-for-delay agreements (which also differ from one case to the other), how can we conclude, especially in the aftermath of Groupement des Cartes Bancaires, that “pay-for-delay agreements are object restrictions”?
competitiongal
26 November 2014 at 4:55 pm