Major victory for Intel as CJEU sends case back to General Court for re-examination
On 6 September 2017, the EU’s highest court, the Court of Justice (CJEU), released its long-awaited decision in the Intel case, in which the Commission imposed a fine of €1.06 billion – at the time, the largest fine ever imposed by a competition regulator on an individual company. This is a very important decision in light of the CJEU’s holding that authorities must examine evidence of the anti-competitive effect of allegedly abusive conduct, instead of relying simply on form-based infringement findings. The judgment may play an important role in a number of high-profile Commission investigations of alleged abuses of dominant positions, particularly in the technology and pharmaceutical sectors.
The Commission’s decision in the Intel case related to the legality of rebate schemes and “naked restrictions” which the Commission found Intel had employed in its customer contracts to exclude competition by Intel’s rival, AMD, in the manufacture of a particular type of computer microprocessor (x86 CPUs). In its judgment, the CJEU disagreed with several aspects of the lower General Court’s judgment upholding the Commission’s decision, but the key point leading the CJEU to refer the case back to the General Court was the need to assess Intel’s economic evidence that the allegedly abusive practices did not foreclose competition. From the perspective of businesses considering the implications of this landmark ruling, the key takeaway is the CJEU judgment’s effect on the legal test for an infringement finding under Article 102, but the CJEU’s discussion of jurisdictional and procedural issues are also noteworthy.
The legal test for an infringement finding under Article 102
A consistent theme of Intel’s arguments was that there was no compelling evidence of actual competitive harm arising from the conduct sanctioned by the Commission. The Commission took the view that under prior case law no showing of competitive harm was required, but it did in fact conduct an “as efficient competitor” (AEC) analysis as provided for in its guidelines on enforcement priorities in exclusionary abuse cases. On appeal, the General Court agreed that no such economic analysis was required and did not address Intel’s criticisms of the Commission’s AEC analysis.
The CJEU held instead that, where evidence is put forward by a dominant company that its conduct in fact could not foreclose competition, the Commission is required to examine that evidence. Although the Commission did in fact make an effects-based assessment, the General Court had not assessed Intel’s arguments that the Commission’s analysis was defective. This is the first time that the CJEU has required an effects-based analysis in an exclusivity rebate case. The Intel judgment means that an effects-based analysis will become much more central not only in future rebates cases, but by analogy also to other abuse of dominance cases.
On the other hand, the CJEU confirmed the long jurisdictional reach of the Commission based on the novel “qualified effects test” and rejected its own Advocate-General’s advice in his October 2016 opinion to take a narrower approach to EU jurisdiction. This will reinforce the enthusiasm of the Commission to act as global enforcer and could have implications for the UK post-Brexit.
Intel argued that the Commission did not have jurisdiction over its agreements (as a US-based company manufacturing CPUs in Taiwan) entered into with Lenovo (a Chinese based manufacturer of computers), because the agreements were concluded in China and involved non-EU companies, and the CPUs in question were sold outside the EU.
In determining the Commission’s jurisdiction to apply EU law to the conduct in question, the CJEU did not apply the standard “implementation test” (i.e., was the conduct in question implemented in the EU?). Instead the CJEU confirmed that the “qualified effects test” (i.e., did the conduct have an immediate and substantial effect in the EU?) provided jurisdiction. The fact that some computers with Intel CPUs were sold in the EU was, therefore, sufficient to satisfy the qualified effects test.
The CJEU also noted that Intel’s deal with Lenovo was part of an overall strategy to exclude rivals and the Commission was right to look at Intel’s behaviour as a whole when determining whether it had jurisdiction regarding the Lenovo deal.
The CJEU also rebuked the Commission for not having recorded an interview carried out during its investigation, and agreed with Intel that the Commission could not rely on a distinction between “formal” and “informal” interviews under Article 19 of Regulation 1/2003. However, the CJEU disagreed with Intel that the Commission’s approach damaged Intel’s interests.
Intel had complained that an informal interview between a senior executive at Dell and EU Commission officials should have been recorded and disclosed as evidence that could have helped its defence. Under Article 19 of Regulation 1/2003, all interviews conducted by the Commission in the context of an investigation must be recorded. In its defence, the Commission sought to draw a distinction between formal and informal interviews, but this distinction was rejected by the CJEU. In this context, an internal note of the meeting subsequently disclosed by the Commission to remedy Intel’s concerns was held to be insufficient. On the other hand, the CJEU held that the Commission’s procedural violation did not vitiate the Commission’s decision.
Although this point did not affect the outcome of the case, the Commission is now on notice that in its investigations it is required to follow the same procedural protections with respect to all interviews, “formal” or not. This could mean that many more interactions during such investigations, potentially including phone calls and state of play meetings, will be more thoroughly documented by the Commission future. This could also have implications for ongoing cases where “informal” investigations not meeting the CJEU’s standards have already been carried out and have been relied upon by the Commission.
Conclusion and implications
The 17-year Intel saga continues – now that the case has been remitted back to the General Court, it could be several more years before a final decision is reached, particularly because either Intel or the Commission – or both – could appeal a future General Court judgment. Nevertheless, the CJEU’s decision is a victory for Intel, and there is a real prospect of the Commission’s decision being ultimately overturned or at least of a significant reduction in fine.
Moreover, the Intel decision could have significant implications on potentially dominant companies as well as on the Commission’s jurisdiction and its evidence gathering powers.
In particular, this decision changes the framework for future dominance investigations, making clear that, where a dominant company puts forward economic evidence that its conduct has not negatively affected competition, the Commission must address this information and the economic context of the alleged infringement. This is a significant change, as it opens the door to companies submitting economic evidence in cases that have previously been addressed under older precedent case law which emphasised the formalistic qualities of conduct or agreements, rather than compelling the Commission to consider and demonstrate actual competitive harm through an effects analysis. The Commission still has no positive obligation to do so, but this ruling means well-advised parties will put forward their own evidence on effects - and the Commission will then be forced to address this prior to reaching an infringement decision, knowing that its approach will then be subject to scrutiny by the EU courts.
This approach may well impact some of the Commission’s current high-profile Article 102 investigations, to the extent that the companies involved have put forward evidence that their conduct was not capable of restricting competition and, in particular, of producing the alleged foreclosure effects. The impact of this decision is less clear for current cases where no such evidence has been put forward and which are far advanced in the EU court process.
In terms of other implications, the effects doctrine applied by the CJEU (in place of the more recognised implementation doctrine) will make the Commission more comfortable in asserting its jurisdiction in cases with no direct nexus to the EU, reinforcing its enthusiasm for acting as the lead global enforcer and could result - for example - in the Commission continuing to have jurisdiction over the UK post-Brexit, where the relevant conduct has an effect in the EU even if it took place in the UK and involved UK companies.
Finally, the decision’s impact on the Commission’s information gathering powers is also important. For those businesses involved in some of the Commission’s long-running investigations, the Commission’s extensive information gathering powers can be very disruptive and any limitation on those powers will be welcomed. The Commission will now be required to apply the same procedural protections with respect to all interviews (formal or informal) and may be required to document all interactions during investigations much more thoroughly.
It is very rare for cases of this sort to reach the CJEU let alone be referred back to the General Court for re-examination, but overall the Intel case should be seen as a welcome development in an overall trend of holding the Commission to account in requiring more consideration of anticompetitive effects of any conduct before reaching an infringement decision. This ruling follows the 2014 Cartes Bancaires CJEU decision which confirmed the Commission needed to be careful in pushing the boundaries of the 'by object' (akin to per se) classification in anticompetitive agreement cases, reminding the Commission of the need to carefully consider the 'legal and economic context' of arrangements. The Intel case can be seen as acting as a similar reminder for unilateral conduct/dominance cases. This decision therefore sets an important precedent for future cases investigated by the Commission as well as national agencies in the EU, which are bound by CJEU precedent - and indeed for the many authorities beyond the EU with laws modelled on the EU framework and which are influenced by the decisions of its courts.