1. It was welcome that the European Commission published its long-awaited Communication on exclusionary abuses under Article 82 EC42 in late 2008. So long had passed since the DGCOMP Discussion Paper in December 2005,43 there were fears that the Commission might not continue with the whole exercise.

2. The Communication’s ambitions are scaled back compared to the approach when reform of Article 82 EC was first raised in the 2005 Discussion Paper. It relies less on economic and legal jargon, and thus is more accessible. However, there is an intellectual contradiction at the heart of the Communication: the formalism of the past co-exists with a more economics-based analysis. This prevents the Communication from successfully modernising the enforcement of Article 82 EC and taking the debate forwards – and the contradiction does not help predictability of outcome.

3. The main points of the Guidelines can be summarised as follows:

  • Dominance cases will be investigated where market shares are above 40 percent, but action by the Commission can also be envisaged in cases of lower market shares;
  • In reaching its decision, the Commission will examine direct evidence of any exclusionary strategy, including internal documents which may be helpful to interpret the dominant company’s conduct; in other words, the intention of a company will matter;
  • The stronger the dominant company in terms of market share, the higher “the likelihood” that conduct may lead to anticompetitive foreclosure; although the Commission does not say so expressly, it will be more proactive in situations of “super-dominance”;
  • There may be circumstances where it may not be necessary for the Commission to carry out a detailed assessment before concluding that the conduct in question is anticompetitive; examples of such conduct include: preventing customers from testing a competitor’s products; or paying distributors to delay the introduction of a rival’s product; again, the Commission may not spell it out, but this is reminiscent of a per se approach;

As for the potential justifications on which a dominant company may rely, the Communication indicates that the company must show to a “sufficient degree of probability” that (i) efficiencies have been/are likely to be realised as a result of its conduct; (ii) its conduct is indispensable; (iii) the likely efficiencies will outweigh any likely negative effects on competition and consumer welfare; and (iv) its conduct does not eliminate effective competition, i.e. does not eliminate most sources of potential competition.

I. Critique

4. The antitrust community had expected that the Communication would move a step closer to a more “effects–based” approach. The introduction of language that meets that expectation is not as such particularly groundbreaking. And this step forward is contradicted by the survival and prominence of all the old formalistic mantras. Whereas one would have expected one of the world’s two most important antitrust authorities to powerfully and convincingly describe to the global antitrust community where it stands on unilateral conduct in a positive rather than in a negative way, what we see is a mixed text which has a sound and respectable intellectual basis but which, at the same time, is too timid to break with the formalism of the past. It is as if the economist’s first draft was hijacked by the old-school lawyer, who does not wish to put victory in pending cases at risk and thus wants to ensure the language in the text can be of no prejudice to him. Yet the result of this impossible exercise can only disappoint.

5. In fairness, it was always going to be difficult for the Commission to “restate” the law on unilateral conduct. Stating or “restating” the law is normally the job of the courts. However, a competition authority has an important role to play in this process, by choosing the right cases and therefore by deciding on its priorities. Indeed, the European Commission is much more powerful in this regard than its US counterparts. In contrast to the US, most antitrust enforcement in Europe is public, so it is normally the Commission that decides which case to bring and, thus indirectly, on how to develop Article 82 EC.44 And the Commission has a very high success rate in defending its case before the European Courts (it has not lost a case in 20 years and has a 98% success rate according to one recent article),45 so the Commission could have been relatively comfortable that any enforcement action would be subsequently endorsed by the European Courts, even if it conflicts with present precedent. Besides, the Communication does not profess to be a summary of how the law currently stands; rather, it is a statement of what the Commission considers its “enforcement priorities” to be over the following years.

6. Unfortunately, however, the Communication, rather than speaking about the future, i.e. what the Commission’s priorities should be under the effectsbased approach, attempts to reconcile the formalism of some old case law with economics. This is evident already in paragraph 1 of the Communication where the Commission repeats the old mantra of the “special responsibility” and that “it is not in itself illegal for an undertaking to be in a dominant position”, as if there should be a doubt that the fact of being dominant is not per se illegal.46

7. In paragraph 5 the Commission states that it will focus on conduct that is “most harmful to consumers”, but even this overture is quickly qualified through references to “safeguarding the competitive process” in paragraph 6 and to ensuring that competitors are not excluded. Indeed, in paragraph 23 it suggests that the position of a less efficient competitor must be protected: the constraint of a less–efficient competitor may in certain circumstances be taken into account when judging if price–based conduct leads to anti-competitive foreclosure. There is a hint of “efficiency offense” about this stance. Such a defensive attitude (in terms of breaking with the formalism of the past) is certainly disappointing.

8. Furthermore, there are far too many presumptions working against the dominant company. This was the case with the 2005 Discussion Paper and continues to be the case now.

9. Most importantly, as in the 2005 Discussion Paper, the importation into Article 82 EC of the four conditions of Article 81(3) EC, while making the objective justification and efficiency defences more systematic, it makes them more difficult to succeed. The problem here is that the Article 81(3) EC methodology cannot easily be imported into Article 82 EC (under defences). This methodology was adopted for Article 81 EC, where the principle of prohibition applies and is not suitable for Article 82 EC, where the principle of control of abuse applies.47 Under a system of control of abuse, it should not be an abuse of a dominant position in the first place if specific conduct brings efficiencies. Then, the efficiency defence based on the four cumulative conditions is not a flexible one: the last (negative) requirement for conduct not to eliminate competition means that a dominant firm’s conduct, although socially desirable because of accruing efficiencies, will still be prohibited. There is also a certain inconsistency in the fact that the efficiency defence is based on the principle of “no net harm to consumers” (paragraph 29), yet as seen above the notion of abuse is not itself solely based on consumer harm.

10. Another disappointing element, which can only be seen as retrogression is the resurgence of intention as a critical component of Article 82 EC. We have seen this in the Microsoft and AstraZeneca cases48 and in the context of the on-going pharmaceutical sector inquiry (where extensive use has been made of selected extracts from emails and internal documents), and now we see it in the Communication. This is a serious mistake: the resolve to win over or even to eliminate competitors is the driving force of competition; what should matter is whether there is a plausible consumer harm theory. It is a pity that the Communication appears here to contradict standard principles accepted even under the more formalistic approach, that an intention even by a dominant firm to prevail over its rivals should not be viewed as unlawful.49

II. How the guidelines treat refusal to deal in particular

11. We will look at how the Communication deals with refusals to deal by way of example. Such cases offer a good example of how an antitrust authority perceives itself as enforcer and what kind of antitrust law it subscribes to.

12. The Communication deals with refusal to supply and margin squeeze abuses together. In reality, it mainly refers to refusal to supply.

13. Interestingly, the Commission here departs from the scheme followed in the 2005 Discussion Paper. In the latter, the Commission had distinguished between (a) discontinuation of supplies, (b) refusal to supply a new customer with an indispensable input, and (c) refusal to license an intellectual property right. There was a graduation in this distinction in that the conditions for antitrust intervention were more permissive in the first case (it need only be shown that the customer risked elimination), narrow in the second case (the denied input must also be indispensable) and very stringent in the latter case (in addition to the above, the refusal to license must block the emergence of a new product for which there is consumer demand).

14. This is now all in the past. The Commission, heartened by its Microsoft victory, now follows a different approach: For all cases of refusal to supply, including refusals to license IP rights, the conditions are the following: (a) the refusal must relate to a product or service that is objectively necessary or indispensable (the two terms are used interchangeably) for a competitor to be able to compete effectively on a downstream market; (b) the refusal is likely to lead to the elimination of effective competition on the downstream market;50 (c) for consumers, the likely negative consequences of the refusal outweigh over time the negative consequences of imposing an obligation to supply in the relevant market.

15. The last condition echoes essentially the Commission’s balancing test in its Microsoft Decision. It is noteworthy that the third condition seems to depart substantially from the ECJ’s ruling in IMS Health and even from the CFI’s ruling in Microsoft. Prevention of the emergence of a new product (per IMS Health) and even prevention of follow-on innovation (per Microsoft) are only considered by the Commission as indicative examples of the third condition, which the Commission calls “consumer harm”. The problem here is also that the Commission fails to require any concrete showing of specific consumer harm to be established and merely holds sufficient the abstract possibility of an increase in innovation and consumer choice. The IMS Health test of a concrete “new product” reflected the balancing between the short-term benefits of antitrust intervention, i.e. the need to protect competition, against the long-term prejudice of that intervention, i.e. the reduction of companies’ incentives to engage in research and development and innovate, for fear that they may be obliged to give their competitors access to their assets. It also reflected a policy choice in favour of a system based on competition by substitution and not on competition by imitation.

16. Of course, the Commission does not purport to state the existing case law – this is after all guidance on the Commission’s enforcement priorities, but the watering down of the test will be of concern to potentially dominant undertakings.

17. What’s more, the Commission appears to apply a negative presumption for discontinuation of supply cases. In such cases, the Commission states that it is more likely to find that the indispensability condition is satisfied in favour of a finding of abuse, for example if the recipient had made “relationship– specific investments”. If there has been a previous supply by the dominant firm, the latter will have to “demonstrate why circumstances have actually changed in such a way that the continuation of its existing supply relationship would put in danger its adequate compensation” (paragraph 83). In other words, not only is the burden of proof reversed, but the dominant company’s right to dispose freely of its property is reduced to an economic right to receive adequate compensation. This creates serious disincentives for dominant companies to enter into commercial agreements in the first place, for fear that once they supply someone they will be stuck in that relationship forever. This may create significant problems for companies doing business in Europe, as well as for their potential customers.

18. The Communication opens up an intellectual divide between US and EU approaches to compulsory licensing. Much has been written about the differences between the FTC and the DOJ during the Bush years when it came to single firm conduct. But on the topic of compulsory licensing of IP rights, the FTC and DOJ did speak with one voice. In their joint paper, Antitrust Enforcement and Intellectual Property Rights: Promoting Innovation and Competition, the two agencies stated:

“Antitrust liability for mere unilateral, unconditional refusals to license patents will not play a meaningful part in the interface between patent rights and antitrust protections. Antitrust liability for refusals to license competitors would compel firms to reach out and affirmatively assist their rivals, a result that is “in some tension with the underlying purpose of antitrust law.” [Citing Trinko, 540 U.S. at 407-408] Moreover, liability would restrict the patent holder’s ability to exercise a core part of the patent—the right to exclude.”51

The Commission’s Communication thus takes a significantly different position from the joint DOJ and the FTC approach to refusals to license.

III. Do the guidelines increase predictability?

19. Let us first offer thanks that the Commission did make the effort to finalise and publish the Guidelines. The fact of publishing a set of Guidelines is always positive—having published guidance can never be worse than having no guidance. So the Commission is to be applauded for having made this effort.

20. However, the benefit in terms of predictability that may be derived from the Guidelines is likely to be modest at best for three reasons.

  • First, the intellectual incoherence at the heart of the Guidelines – the combination of formalism alongside increased economic analysis – will make it difficult to analyse how the Commission will judge any specific situation, particularly a novel set of facts.
  • Second, these are only guidelines as to the Commission’s enforcement priorities. There is a significant risk that they will not be followed by national courts or the European Courts, in particular in areas where the Guidelines depart from recent precedent by the European Courts – see above for the discussion of refusals to supply and the Microsoft judgment as one example.
  • Third, a number of the substantive tests proposed in the Guidelines will not be easy to apply. For example, the third test for refusals to supply— consumer harm, whether “for consumers, the likely negative consequences of the refusal outweigh over time the negative consequences of imposing an obligation to supply in the relevant market.” Here the test that the Commission will apply is written in black and white, but applying it in practice will be anything but obvious. Similar considerations apply to the use of the Article 81(3) conditions for judging objective justification/ efficiency under Article 82.

So while the publication of Guidelines is a step forward, it is likely to only be a small step.

IV. Conclusions

21. The new Article 82 Guidelines do not succeed in modernising European law on single firm conduct by replacing the formalism of the past with a convincing approach driven by consumer harm. The Commission had the possibility of achieving this result – and there are indications in the text that efforts were made in this direction. But they were not successful. We are left with an unhappy marriage between the old and the new, which will not increase predictability.

22. The text thus also fails to promote convergence between US and EU rules on single firm conduct. In the future, European and US enforcers may converge towards a middle ground, with Europe abandoning some of its formalism and the US becoming slightly more interventionist. But old habits in Europe die hard. Indeed, the publication of this Communication may even have succeeded in widening the gap between the US and the EU when it comes to refusals to supply and compulsory licensing.?

This article was published in a slightly different format in Concurrences Review N° 2-2009.