The Intel Effect

The General Court Judgement in Case T-286 / 09 RENV Intel Corporation v Commission

Overview

On 26 January 2022 the European General Court (GC) partially annulled a 2009 European Commission (Commission) decision, in which the Commission found that Intel had abused its dominant position by implementing a strategy intended to exclude its competitors from the market for x86 central processing units.^[1]

The GC’s findings are, in many ways, a natural next step following the 2017 judgment of the European Court of Justice (CJ), which made clear that the Commission should have analysed the effects of Intel’s conduct when assessing whether it had infringed Article 102 of the Treaty on the Functioning of the European Union (TFEU) (see here).^[2]

Nonetheless, the judgment also provides some welcome clarity on the degree of analytical rigour expected of the Commission when performing such an analysis.

Case history

In 2009 the Commission fined Intel a then-record €1.06 billion for implementing exclusivity rebates and other “naked restrictions” in arrangements with five original equipment manufacturers (OEMs) and one retailer. The Commission took the view that it did not need to prove any anticompetitive effects arising from the arrangements, on the basis that they were per se (automatically) abusive under Article 102 TFEU.^[3]

The Commission did, however, still conduct an economic analysis of the effects of Intel’s conduct by applying the so-called ‘as efficient competitor’ (AEC) test, which measures whether a competitor that was as efficient as Intel could have competed with Intel notwithstanding the exclusivity rebates.

Intel appealed the Commission’s decision to the GC, which dismissed the appeal in 2014 and upheld the Commission’s finding that the exclusivity rebates were per se abuses of dominance.

Intel appealed the GC’s judgment to the CJ. In a much anticipated judgment in 2017, the CJ overturned the GC, ruling that exclusivity rebates are not per se abuses under Article 102 TFEU. The CJ also held that where a dominant undertaking submits evidence that its conduct was not capable of restricting competition and, in particular, of producing the alleged foreclosure effects, the Commission must analyse the effects of the conduct taking into account (as a minimum) the following relevant criteria: the extent of the undertaking’s dominant position, the share of the market covered by the conduct, the conditions and arrangements for granting the rebates, their duration and amount, as well as the possible existence of a strategy aimed at excluding competitors that are at least as efficient as the dominant undertaking.^[4]

Since the GC had not considered Intel’s arguments regarding the Commission’s application of the AEC test, the CJ remitted the case to the GC for reconsideration.

Key findings of the GC

Commission’s decision on exclusivity rebates was vitiated by an error of law

The GC found that the Commission’s decision regarding the exclusivity rebates at issue was vitiated by an error of law because it had treated Intel’s rebates as automatically anticompetitive, and had therefore concluded that no effects analysis was required to establish an abuse. The GC found that this position was inconsistent with earlier case law – as clarified by the CJ in its 2017 judgment, while a system of rebates imposed by a dominant undertaking may be presumed to restrict competition, it is not a per se infringement that relieves the Commission of its obligation to conduct an effects analysis where a dominant undertaking challenges the presumption.

The GC reached this finding despite the Commission having conducted an AEC analysis in any event, since the Commission had – in reaching its decision - expressed (repeatedly) the view that it did not need to take that analysis into account in order to conclude the conduct was abusive.

Commission’s AEC analysis was vitiated by numerous errors

The GC confirmed that, although the Commission is not required to conduct an AEC analysis to assess the capability of exclusivity rebates to foreclose an as-efficient competitor, since the AEC test had played an important role in the Commission’s decision against Intel, it was required to examine Intel’s arguments concerning that test. The GC therefore conducted a detailed examination of the Commission’s AEC analysis, and found that it was vitiated by numerous errors. These included using the incorrect proportion of contestable demand,^[5] the incorrect value of costs,^[6] an unsubstantiated extrapolation of its analysis based on limited time periods,^[7] and an unsubstantiated extrapolation of its analysis based on a limited sample size.^[8]

The GC concluded that the Commission’s AEC analysis had not established to the requisite legal standard that Intel’s exclusivity rebates were capable of foreclosing a competitor as efficient as Intel.

Insufficient consideration of relevant criteria

The GC found that the Commission had failed to take into account all the relevant criteria identified by the CJ in its 2017 judgment as required for an effects analysis.^[9] In particular, the GC found that the Commission had not considered properly the criterion relating to the share of the market covered by the rebates and had also not correctly analysed their duration.^[10]

Annulment of the entire fine

Although the Commission’s findings were only annulled insofar as they related to the exclusivity rebates at issue (its finding regarding the naked restrictions was upheld by the GC), the GC considered that it was not in a position to identify the amount of the fine relating solely to the naked restrictions.

As a result, and for only the third time in an Article 102 TFEU case, the fine imposed by the Commission was annulled in its entirety.^[11]

Key Implications

The end of per se infringements?

The CJ’s 2017 judgment had been heralded by some as a triumph of effects over form under Article 102 TFEU.

Although potentially dominant companies may be well advised to argue that the principles contained in the 2017 judgment are of general application, and to put forward evidence as to a lack of foreclosure capability, the Commission may well look for ways to apply these judgments as narrowly as possible. This may therefore prove a key battleground in a number of cases currently before the European courts.

Evidentiary standards

Following the CJ’s ruling that the Commission should have analysed the effects of Intel’s conduct, question marks remained over the degree of rigour that would be expected of the Commission in performing such an analysis. The GC’s latest judgment has not altered the applicable evidentiary standard in this respect - the GC confirmed that the Commission must prove infringements on the basis of a body of “sufficiently precise and consistent evidence”.^[12] Furthermore, if there is another plausible explanation for the facts relied on by the Commission to establish an infringement, then insufficient proof of that infringement has been adduced.^[13]

That said, in finding that the Commission failed to meet this standard in a number of respects, the judgment provides clarity as to what is expected of the Commission in practice – the standard of proof must be met, and reliance on incomplete or inaccurate evidence is not acceptable.

The GC has firmly demonstrated that the European Courts will not uphold Commission decisions where the necessary legal and evidentiary analysis is lacking. In future investigations, one might also see the Commission learning from its experience in Intel, by seeking to ensure that any evidence that it relies on is sufficiently robust to withstand scrutiny.

AEC test

The GC’s findings on the Commission’s AEC analysis provide welcome clarity on certain aspects of the AEC test. Importantly, the GC found that the relevant hypothetical competitor for the Commission’s AEC analysis should not have been less efficient than Intel in terms of the costs it would incur whilst competing.^[14]

The case demonstrates, however, the scope for potential disagreement over the substance of the AEC test and how it should be conducted, and it remains to be seen how far the Commission will perform (and rely on) AEC analyses in future Article 102 TFEU cases concerning rebates. Nevertheless, for potentially dominant companies assessing the compliance of their rebate arrangements, a robust and accurate (and favourable) AEC analysis ought to be an effective method of rebutting (at the least) the starting presumption that exclusivity rebates are anticompetitive.

Conclusions

The GC’s latest judgment represents a significant blow for the Commission, which will have to repay the fine in its entirety and potentially also interest accrued over the previous 13 years.

The Commission also now has to decide whether to appeal the case to the CJ, revisit its original investigation to assess whether the errors identified by the GC may be rectified (and, if so, seek to re-impose its decision on Intel) or prioritise its resources for other investigations.

Regardless of the next steps in this long running saga, the latest instalment delivers a number of important lessons for both the Commission and potentially dominant companies.