Relaxing whilst doing Competition Law is not an Oxymoron

The Unintended Consequences of the Case Law on Restrictions by Object

with 10 comments


On second thoughts, the recent bad case law on the notion of “restriction by object” may incidentally, and unexpectedly, fetter the margins of the Commission, and in particular its ability to handle all cases under the obese Article 9 procedure.

As hinted recently by DG Italianer in a most interesting speech, cases such as Irish Beef (C-209/07), Expedia (C-226/11), Slovak Banks (C-68/12) and Allianz Hungary (C‑32/11) all suggest that agreements which are devoid of the “obvious” capacity to harm competiton are nonetheless “restrictions by object” because they are “serious” violations of the law.

In other words, those infringements are restrictions by object because they are sinful – in competition cases, the Court says “by nature injurious to the proper functioning of normal competition” – so sinful that they should be prosecuted even absent anticompetitive intent or effects.

This line of judicial precedents brings restrictions of competition close to the notion of “infractions objective” known in criminal law. Take murder: you can be convicted of homicide even if you unsuccessfully or unwittingly try to kill someone (in some legal orders, it is even illegal to shoot a dead body). Closer to competition law, take insider trading. You can be sentenced even if you use insider information unknowingly or ineffectively.

The bottom line: such infringements are so morally sinful that they should be prosecuted just for the sake of it. Regardless of their impact. Regardless of their motives. Full stop.

Now, let us revert to competition law. If the concept of a restriction by object means infringements that are morally sinful or morally so “serious” that they should be unlawful for the sake of it, then the Commission should no longer be free to settle such cases and decline to reach a finding of infringement.

This is the rule applied in most criminal law systems, where in principle the gravest infringements (eg crimes such as homicide, rape, etc.) cannot be subject to settlements.

The same should apply in competition proceedings. Restrictions by object ought to be treated under Article 9.  If restrictions by object are that bad, then the sole procedural way to handle them is under Article 7 . Full stop.

Written by Nicolas Petit

3 December 2013 at 4:59 pm

Posted in Uncategorized

10 Responses

Subscribe to comments with RSS.

  1. Hi Nicolas, I read this quickly so maybe I am missing some points but …
    – Which sinful antitrust infringements were ended with commitments?
    – In most national jurisdictions I see the opposite, killers can say yes to procedural shortenings and thus obtain reductions of their sentences. There are frequent amnesties having killers get out of jail after three years instead of twenty. And so on. Unfortunately even the most sinful criminal cases can be settled.
    – Finally unintentional hard-core infringements will be sanctioned because the error of law cannot be excused as the court of justice said earlier this year in Schenker.
    Best wishes. G.


    3 December 2013 at 5:40 pm

    • Hi Gianni, thanks for the useful post: on question 1, all the recently settled 101 cases involve restrictions by object (E-Books, Siemens/Areva, Airlines cases, etc). On question 2, you are very right, with this little yet important caveat that the settlements your are talking about do give rise to a finding of infringement, unlike in Article 9. Moreover, this tendency is true mostly in anglo saxon countries. In continental countries, settlements are excluded in serious criminal offenses (US lawyers would talk of serious felonies v misdeamenors). On 3, I fully agree.

      Nicolas Petit

      5 December 2013 at 10:12 am

  2. While I completely take your point on the current state of the case law on the object/effect distinction (or, should I say the lack of any meaningful distinction in practice…. ), I don’t agree that object cases can never be suitable for commitments. While it is true that the most serious forms of infringement should not be closed by commitments, given the need to ensure punishment and maintain a deterrence effect, I would limit this category to hard core cartels and (possibly) serious and deliberate abuse of dominance. Preventing authorities accepting commitments that adequately meet their competition concerns in a wider category of cases would not be a practical solution and could have the perverse effect of leading to more cases dragging on and ultimately being closed without any form of resolution, due to resource contraints.

    Becket McGrath

    3 December 2013 at 6:26 pm

    • Becket, thanks for this interesting comment. I agree that commitments can be suitable in some circumstances. And I am certainly not saying that object cases can never be suitable for commitments in the current state of EU law. What I am looking at is possible evolutions in the case law and decisional practice. And what I am saying is that if – the if is important – the concept of object ought to be understood as sinful behavior – as implicitly suggested in recent case-law – then the logical, and possibly unintended, consequence of this should be to rule out dealing with object cases under Article 9. This is the rule that applies in other areas like criminal law, etc (recall that we antitrust lawyers should not “believe that [ou] beloved field of law shld get special treatment”…). The recital 13 of Regulation 1/2003 that excludes cases suitable for fines from Article 9 confirms this, as recalled by Alfonso below (with the caveat though that cases suitable for fines are not only cartel cases).

      Nicolas Petit

      6 December 2013 at 11:50 am

      • I think we’re all on the same page that, if the object box was properly limited to clearly anticompetitive and harmful conduct, it would be harder to justify commitments in object cases. Interesting to compare approaches in Ebooks – in the US it was prosecuted as per se (albeit civil, not criminal) but settled for all except Apple; similarly the EC viewed the conduct as anticompetitive by object yet settled by way of commitments in all instances. I think it’s pretty clear that Recital 13 is simply being ignored in practice. Surely the Commission would have fined in Ebooks without settlements and would rightly consider fines if it can’t settle with Google?

        Becket McGrath

        6 December 2013 at 12:03 pm

  3. The proposal sounds good in theory, but with Allianz, you simply do not know if there is a restriction by object or effect and thus you do not know if you would be able to settle your case or not. The proposal would only work in a world where the object category is clear-cut and foreseeable…


    5 December 2013 at 1:11 am

  4. Nico, you write “infringements that are morally sinful or morally so “serious” that they should be unlawful for the sake of it, then the Commission should no longer be free to settle such cases and decline to reach a finding of infringement (…) This is the rule applied in most criminal law systems, where in principle the gravest infringements (eg crimes such as homicide, rape, etc.) cannot be subject to settlements”

    But we do have that rule in EU competition law too! (although arguably in a soft-version):

    According to recital 13 of Reg 1/2003, commitment decisions are not appropriate when the Commission considers that the nature of the infringement calls for the imposition of a fine.

    Therefore, in practice, “the Commission does not apply the Article 9 procedure to secret cartels that fall under the Notice on immunity from fines and reduction of fines in cartel cases” (para 13 of the chapter on commitments in the ManProc). This rule is exactly what I believe Becket and Asimo considered desirable in their comments.

    As regards non-cartel conduct, I don’t think anyone is (or should be) saying that infringements by object should be condemned as sinful or morally wrong; at least in theory, the object label doesn’t equate to condemnation, and should only take us to 101(3). Now, the problem comes with what we once called “the slow death of Article 101(3)” (i.e. in practice, once you fall within the object box you’re done..; but that’s a problem of practical -not moral-condemnation).

    To be clear, my point is that only hardcore cartels are viewed as morally wrong, and these aren’t settled via Art. 9 decisions.

    You may recall that we also had interesting discussions about tboth the enlargement of the object category ( ) as well as about the death of 101(3) ( )

    Alfonso Lamadrid

    5 December 2013 at 8:50 pm

    • @Beckett: fully agreed, the Commission does close cases suitable for fines. Just to add another example: Microsoft WMP (2004) gave rise to a fine. 5 years later, the copycat Microsoft IE gave rise to …… an article 9 decision…

      Nicolas Petit

      6 December 2013 at 12:11 pm

      • Yes, that case came to my mind too. Though of course, they got their pound of flesh from MS in the end, due to non-compliance with the commitments…

        Becket McGrath

        6 December 2013 at 12:13 pm

      • Maybe the MSFT IE case was different on the facts, and maybe the legal arguments made were more compelling, as a result of which an infringement decision would have been an uphill struggle.


        7 December 2013 at 1:55 am

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: