European Union: Abuse of Dominance

Has the Commission’s Guidance been saved from going the way of the dodo?

The European Commission’s (the Commission) ‘Guidance on the Commission’s enforcement priorities in applying article 82 of the Treaty¹ to abusive exclusionary conduct by dominant undertakings’ (the Guidance Paper)² has, at best, had mixed success. The long-awaited paper advocated a more ‘economic’ approach to the enforcement of article 102 of the Treaty on the Functioning of the European Union (article 102). While some within the Commission may have used the Guidance Paper internally to screen cases and not investigate pricing practices that arguably could have fallen foul of older more restrictive case law, the EU courts and the Commission’s own legal service showed reluctance and even hostility towards the Guidance Paper and continued to apply the older and at times inconsistent case law. This led practitioners to question the value of the Guidance Paper and arguably it risked becoming irrelevant. However, much appears to have changed with the September 2017 judgment in Intel v Commission³ in which the Grand Chamber of the Court of Justice of the European Union (CJEU) seemingly resuscitated the approach advocated in the endangered Guidance Paper. Nonetheless, while the Grand Chamber judgment emphasises a more economic approach to enforcing article 102, it leaves important questions unresolved.

The Guidance Paper

The Guidance Paper was part of a series of reforms designed to modernise the substantive rules of European competition law.⁴ However, unlike other Commission modernisation instruments, the Guidance Paper does not formally aim to state the Commission’s views as to what conduct is likely to constitute an infringement of competition law. Rather, it indicates the enforcement priorities that should guide the Commission when considering whether to initiate a case under article 102.⁵ Its purpose is to clarify ‘the general framework of analysis which the Commission employs in determining whether it should pursue cases concerning various forms of exclusionary conduct’.⁶ Nevertheless, it aims to help undertakings better assess whether their decisions risk infringing EU law.⁷

Practitioners and companies needed guidance on article 102; in particular, EU law on rebates, grounded in a handful of mostly fairly old judgments of the CJEU, diverged markedly from US law and practice. The enforcement of article 102 had long been criticised for being too formalistic and legalistic and for placing greater emphasis on characterising particular kinds of conduct rather than examining their actual effects on the market.⁸ In response to this criticism, the Guidance Paper states that the purpose of enforcement is to ‘protect an effective competitive process and not simply competitors’.⁹ Moreover, the Guidance Paper states the Commission will intervene under article 102 only where, on the basis of cogent and convincing evidence, the conduct is likely to lead to anti­competitive foreclosure.¹⁰ ‘Anticompetitive foreclosure’ is defined as when ‘effective access of actual or potential competitors to supplies or markets is hampered or eliminated as a result of the conduct of the dominant undertaking whereby the dominant undertaking is likely to be in a position to profitably increase prices to the detriment of consumers’.¹¹ The Guidance Paper states that the Commission will act only if it is proved that the effect of conduct is actual or likely harm to consumers. Moreover, the Commission will take into account the economic context rather than the exact form of the conduct. The Guidance Paper therefore advocates a more ‘economics’ or ‘effects-based’ approach (albeit for some economists and lawyers the Guidance Paper may not have gone far enough)¹² rather than a traditional ‘formalistic’ or ‘per se illegality’ based approach.

One of the Guidance Paper’s central themes is its emphasis on avoiding potential foreclosure of ‘equally efficient competitors’.¹³ Accordingly, it states that when assessing conditional rebates (namely rebates granted in return for a particular type of purchasing behaviour such as purchasing a stated quantity over a defined period), it will apply an ‘as efficient competitor’ (AEC) test, which examines whether a competitor that is as efficient as the dominant firm can profitably price at the same level as the dominant firm.¹⁴ The Guidance Paper thus sets out the cost benchmarks that the Commission will use in its analysis, namely the average avoidable cost (AAC) and the long-run average incremental cost (LRAIC).¹⁵

The Commission’s practice since 2006

The Tomra decision was adopted in March 2006 and nicely illustrates the Commission’s pre-Guidance Paper approach to enforcing article 102 in the rebates context.¹⁶ Relying on the preexisting case law such as Michelin II,¹⁷ the Commission effectively states that it can find that a company infringed article 102 without showing that its conduct affected the market: ‘the effect referred to in the case law cited in the preceding paragraph does not necessarily relate to the actual effect of the abusive conduct complained of. For the purpose of establishing an infringement of Article [102 TFEU], it is sufficient to show that the abusive conduct of the undertaking in a dominant position tends to restrict competition or, in other words, that the conduct is capable of having that effect.’¹⁸ The Commission finds that the rebates were conditioned on customers exclusively purchasing from Tomra and that ‘exclusivity has, by its nature, the capability to foreclose because it requires the customer to purchase all or almost all its requirements from the dominant supplier’.¹⁹

To bolster its conclusion, the Commission claims to have ‘completed its analysis . . . by considering the likely effects of Tomra’s practices’;²⁰ however, this additional analysis was mainly an examination of Tomra’s stable market shares during the period in which its rebates were in force rather than a more thorough analysis such as that required under an AEC test. Moreover, before the EU courts on appeal, the Commission’s Legal Service appears to have distanced itself from the effects analysis and instead emphasised that its decision was consistent with the courts’ case law.

In the 2009 Intel decision,²¹ the Commission notes that the Guidance Paper was a statement regarding its future enforcement intentions and it was not therefore applicable to the Intel proceedings, which had been initiated before the Guidance Paper’s publication.²² Moreover, the Commission considers that the relevant Intel rebates were ‘in themselves sufficient to find an infringement’ under article 102 TFEU and it therefore is not obliged to analyse any of the surrounding context nor prove actual foreclosure.²³

Nonetheless, while noting that this was ‘not indispensable’ the Commission seeks to demonstrate that Intel’s exclusivity rebates were capable of causing or likely to cause anticompetitive foreclosure.²⁴ In particular, the Commission (almost bashfully) notes that an AEC analysis was ‘one possible way’ of examining whether the rebates were capable of foreclosing a hypothetical competitor who was as efficient as Intel. Using a modified version of the Guidance Paper’s AEC test, the Commission concluded that this efficient competitor would have had to offer its products at a price that was below its AAC to compete against Intel’s rebates, which would not be viable.²⁵ Again, however, before the General Court on appeal, the Commission’s Legal Service de-emphasised this aspect of its decision.

This Commission’s approach in Intel is similar to what it did in previous cases, such as Wanadoo,²⁶ Microsoft²⁷ and Telefónica,²⁸ where the Commission examined the impact of conduct on the market only after having stated that there was no requirement for it to demonstrate that the abuse in question had concrete effects on competition or the market. In effect, the Commission covers all its bases and concludes that either under the EU courts’ case law or under an economics-based approach, the conduct infringed article 102. One can hardly imagine that, having found an infringement based on the EU courts’ formalistic approach, the Commission’s economic analysis could subsequently reach the opposite conclusion that conduct already condemned as abusive would in fact be unlikely to adversely impact the market. The usefulness of conducting an economic analysis under these circumstances is therefore questionable.

The Commission recently fined Qualcomm €997 million for making exclusivity payments to Apple in return for Apple exclusively purchasing Qualcomm’s LTE baseband chips for Apple’s iPhones and iPads.²⁹ The Commission is yet to publish its decision but it does not appear to have conducted its own economic analysis of Qualcomm’s practices (perhaps this was not necessary on the facts); rather it appears to have rejected a ‘price-cost’ test that Qualcomm submitted.³⁰

The courts’ pre-September 2017 reaction

Rebates

The Guidance Paper itself recognises that it is not intended to be a statement of the law on article 102 and that it is without prejudice to the European courts’ interpretation of article 102.³¹ This follows from the principle of separation of powers and the EU courts’ prerogative to interpret EU law. Nonetheless, if the Guidance Paper is to be of optimal use to companies and their advisers, it would seem appropriate for the courts to uphold the Commission’s methodology in the Guidance Paper; this was not what happened. Instead, the courts mostly remained attached to the earlier rebates case law, which often does not require an economic analysis proving damage to competition or consumers.

In Tomra, the CJEU upheld the General Court’s dismissal of Tomra’s appeal against the Commission’s decision.³² It expressly stated that under article 102 it was sufficient to show that Tomra’s conduct tended to restrict competition or that its conduct was capable of doing this.³³ It was not therefore necessary for the Commission’s decision to go further and analyse the actual impact of Tomra’s rebates on competition.³⁴ Moreover, the CJEU noted that Tomra had highlighted that the Guidance Paper provided for a comparison of prices and costs but the CJEU considered this of ‘no relevance’ since the Tomra decision pre-dated the Guidance Paper.³⁵

In Intel, the General Court stated that exclusivity rebates granted by an undertaking in a dominant position were, ‘by their very nature’, capable of restricting competition and foreclosing competitors from the market.³⁶ For the General Court it was thus once again unnecessary under article 102 to show that the rebates were capable of restricting competition.³⁷

Turning to the Commission’s AEC analysis, the General Court reiterated that for exclusivity rebates there was no requirement to show an anticompetitive effect and therefore no need to use an AEC test.³⁸ Furthermore, the General Court stated that an AEC test is not required even if the rebates do not require exclusivity provided ‘the mechanism for granting the rebate’ has a ‘fidelity-building effect’;³⁹ this directly contradicts the Guidance Paper’s endorsement of the AEC test. Moreover, even if an AEC test is applied and it shows that an equally efficient competitor could compete with the dominant company, the General Court considered that this would not be conclusive evidence of article 102 not being infringed.⁴⁰ This last statement effectively meant that dominant companies could not rely on the Guidance Paper to determine if their conduct complies with article 102; there was no safe harbour. While the General Court correctly noted that the Commission was not required to follow the Guidance Paper in the Intel case (and therefore leaves open the possibility that the courts could endorse the Commission’s approach in future cases) the broad nature of the criticism of the AEC test did not augur well for this possibility.

In its Post Danmark II judgment,⁴¹ the CJEU had the opportunity to give further guidance on rebates, including the potential applicability of the Guidance Paper’s AEC test. The CJEU concluded that the AEC was of no relevance in the particular factual circumstances owing to Post Danmark’s very large market share and its structural advantages on the market in the form of a statutory monopoly; in such circumstances the emergence of an as efficient competitor was practically impossible.⁴² More generally, the CJEU confirmed what the General Court had stated in Intel, namely that the Guidance Paper sets out the Commission’s administrative practice and is not binding on national competition authorities or courts⁴³ and that there is no legal obligation to assess rebates according to the AEC test where it can be established that rebates are capable of restricting competition based on qualitative elements.⁴⁴ At best, the AEC is ‘one tool amongst others’ for assessing whether article 102 has been infringed and recourse to the AEC test is not excluded.⁴⁵

The courts’ scepticism towards the Guidance Paper was perhaps best summed up by Advocate General Kokott in her opinion in Post Denmark II where she cautioned against adopting ‘a more economic approach’ in these words: ‘the Court should not allow itself to be influenced so much by current thinking (Zeitgeist) or ephemeral trends, but should have regard rather to the legal foundations on which the prohibition of abuse of a dominant position rests in EU law’.⁴⁶

Other pricing practices

Perhaps because there were no relevant older precedents, in cases concerning margin squeeze and selective price cutting, the courts adopted an approach and tenor that was more aligned with the Guidance Paper albeit without endorsing it expressly. The judgments in Intel and Post Danmark II thus sit somewhat uneasily with three CJEU judgments concerning margin squeeze and selective price cutting.

In Deutsche Telekom, the CJEU endorsed the Commission (and General Court’s) application of an AEC test to assess the legality of an alleged margin squeeze.⁴⁷ TeliaSonera also concerned an alleged margin squeeze.⁴⁸ Here the CJEU again emphasised that article 102 prohibits pricing practices that could potentially exclude equally efficient competitors from the market.

In Post Danmark I, it was alleged that a dominant company had selectively reduced its prices in order to win certain customers from a competitor.⁴⁹ In particular, the prices offered to one customer were below Post Danmark’s average total costs but above its average incremental costs.⁵⁰ The CJEU considered that this did not amount to an abuse since an as efficient competitor could, as a rule, compete against prices that covered the bulk of the dominant company’s costs of providing services to that one customer.⁵¹ In language that strongly echoed the Guidance Paper, the CJEU noted that article 102 did not seek to ‘ensure that competitors less efficient than the undertaking with the dominant position should remain on the market’.⁵²

Intel

The CJEU’s judgment in Intel’s appeal was eagerly awaited in particular since Advocate General Wahl, while not expressly endorsing the Guidance Paper (he only cited to it in a footnote), appeared to endorse its approach when he criticised the General Court for dismissing the relevance of the Commission’s AEC test.⁵³

The relevant section of the CJEU’s judgment⁵⁴ begins by stating that article 102 does not prohibit the exclusion of less efficient competitors.⁵⁵ In contrast, the judgment states that article 102 prohibits dominant firms from adopting pricing practices that have an exclusionary effect on competitors that are as efficient as the dominant firm.⁵⁶ Already, this echoes the prominent theme in the Guidance Paper.

The judgment then recalls the CJEU’s older Hoffmann-La Roche⁵⁷ case law holding that a dominant firm infringes article 102 if it ties purchasers by exclusivity requirements or implements loyalty rebates (‘discounts conditional on the customer’s obtaining all or most of its requirements . . . from the undertaking in a dominant position’). Then – in its key paragraphs – the judgment states that this existing case law ‘must be further clarified’ if a company under investigation submits evidence to the Commission ‘that its conduct was not capable of restricting competition and, in particular, of producing the alleged foreclosure effects’.⁵⁸ In these circumstances, the Commission – and the EU courts on appeal – are obliged to analyse all relevant circumstances to determine if the rebates infringe article 102.⁵⁹ This analysis should include considering the extent of the dominant position, the market share affected by the rebate, the conditions applicable to the rebate, its duration and amount and whether there is evidence of a strategy aimed at excluding as efficient competitors.

Similarly, the CJEU holds that the Commission (or court on appeal) should analyse the intrinsic capacity of conduct to foreclose as efficient competitors.⁶⁰ This requires analysing all relevant circumstances. Since in the Intel decision the Commission carried out an AEC test, the General Court was required to consider all of Intel’s arguments concerning that test.⁶¹ The General Court had not done this so the CJEU ruled that its judgment should be set aside and that the General Court should re-examine all of Intel’s arguments.

Potential implications

While the CJEU suggests that it is merely clarifying its existing case law, this judgment does far more than that. Like the Guidance Paper, it emphasises the centrality, under article 102, of analysing whether rebates foreclose as efficient competitors.

Nonetheless, the CJEU’s judgment leaves a number of questions open (not least for Intel).

• First, it remains unclear what a dominant company must do to require the Commission to consider all relevant circumstances. As noted, the judgment states that the company must provide evidence that ‘its conduct was not capable of restricting competition and, in particular, of producing the alleged foreclosure effects’. But this does not clarify whether the dominant company must present a complete analysis of the conduct’s effect or just satisfy some lower threshold that would shift the burden of proof back on to the Commission to prove – taking account of all relevant circumstances – that the conduct would likely exclude as efficient competitors.
• Second, the judgment does not explicitly require the Commission to carry out an AEC test in all cases. In practice, however, it is likely to have to do this unless the company under investigation does not adduce evidence to contest the finding that its conduct was capable of foreclosing as efficient competitors.
• Third, it would appear that even if a company’s rebate system would allow an as efficient competitor to compete, this will not necessarily be a safe harbour. This would seem to follow from the requirement that the Commission analyse the potential existence of a strategy designed to exclude as efficient competitors. In other words, even if a company carries out an elaborate AEC test before implementing a new rebate scheme and concludes that the system would not exclude as efficient rivals, it may still infringe article 102 if the Commission were later to discover ‘bad’ documents suggesting that the company planned to foreclose as efficient competitors.
• Fourth, while the key paragraph 138 of the judgment refers to ‘that case law’ – and this most directly refers to Hoffmann-La Roche mentioned in the preceding paragraph and that case concerns rebates – arguably the CJEU’s ‘clarification’ of article 102 could apply to any potentially exclusionary conduct and not just rebates. Notably, paragraph 136 of the judgment refers to ‘pricing practices that have an exclusionary effect’ and Post Danmark I and not only rebates.
• Lastly, while the CJEU judgment refers to concepts such as as efficient competitors and the potential need to analyse whether such competitors are foreclosed, it does not openly endorse the Guidance Paper. As noted, this may just be a question of timing since the Intel investigation preceded the Guidance Paper. Companies and their advisers must therefore review rebate systems in light of both the Guidance Paper and the CJEU’s judgment.

Conclusion

One of the Guidance Paper’s purposes was to provide greater clarity and predictability to potentially dominant companies.⁶² Until now, one certainly could not say this had been achieved.

To a degree, companies could look to the Guidance Paper for reliable guidance on whether their proposed strategies comply with article 102.

With the CJEU’s endorsement of price- or cost-based AEC tests in Deutsche Telekom, TeliaSonera and Post Danmark I, this seemed to be the case for alleged margin squeezing and allegedly selectively low prices. The CJEU has also long applied such a test to alleged predatory pricing. For tying or bundling, neither the Commission nor the EU courts expressly applied the cost-based test suggested in the Guidance Paper to multi-product rebates.⁶³ Nonetheless, given that this test is closely linked to the established tests for analysing allegedly predatory prices, it seems reasonable to assume that, insofar as the Guidance Paper emphasises the need to consider equally efficient competitors, the courts would do the same. (On the other hand, the fact that both the Commission and the General Court analysed the effects of tying in Microsoft⁶⁴ offers little comfort, given that both, and in particular the General Court, took pains to make it clear that this analysis was not required under law.)

When, however, it came to rebates, the situation was very different and potentially dominant companies could not just rely on the Guidance Paper. Faced with divergence between the Commission (or at least DG Comp), which stated that it will normally only intervene when there is concrete proof of harm to competition (albeit leaving room to intervene in other exceptional cases) and the EU courts (apparently spurred on by the Commission’s own Legal Service),⁶⁵ which seemingly disregarded the Guidance Paper and continued to apply their well-­established case law, dominant companies continued to lack the necessary legal certainty when devising rebate programmes. If they satisfied the Guidance Paper’s AEC test, they could be reasonably confident that the Commission was unlikely to open a formal investigation or fine them. However, since national competition authorities and national courts are not bound by the Guidance Paper and national courts – in particular – would be slow to disregard the EU courts’ adherence to its case law, prudent dominant companies had no alternative but to ensure that they complied with the courts’ case law. This risked companies deciding not to introduce rebates that may actually be pro-competitive, which chills competition. The CJEU’s Intel judgment is a huge step away from earlier case law but, as noted above, it still leaves questions open and it may take other judgments (most immediately the General Court’s judgment when it reconsiders Intel’s appeal) or a greater body of Commission decisional practice to clarify fully the extent to which the Commission must take account of all relevant circumstances and the extent to which dominant companies can rely on an AEC test or other economic evidence.

Meanwhile, the Commission needs to commit wholeheartedly to the economics-based approach. The ‘have your cake and eat it’ approach in cases like Intel (although again that investigation admittedly was initiated pre-Guidance Paper) – where, on the one hand, the Commission demonstrates that rebates are unlawful under the courts’ case law and, on the other, purports to apply an AEC test while making it clear it feels such a step is not necessary – is all too comfortable with the Commission all but ensuring that it will win either way. Moreover, such an approach perpetuates the relevance of the older case law. If it believes in the Guidance Paper’s approach to rebates, the Commission should act consistently and, in a suitable case, only apply an AEC test to find an infringement of article 102 and defend its findings in court based on this sole approach. This would appreciably increase the confidence that companies could place in the Guidance Paper as a whole and would also encourage national authorities and national courts to apply the AEC test to rebates.