Procedural issues with the new test for conditional rebates: Case C-240/22 P Commission v Intel
Year: 2025
Type: article
Abstract: Judgment of 24 October 2024, Commission v Intel, C-240/22 P, EU:C:2024:915. The Court of Justice provides detail on how to calculate costs when applying the ‘As-Efficient Competitor’ test with flexibility left to the Commission as to when it should be applied. The Commission decision against Intel was decided on the 13/5/2009. It found that Intel had abused its dominant position by agreeing conditional rebates with four of its main customers and a major retailer. The Commission applied the law as it was then understood from Hoffmann-La Roche (EU:C:1979:36); ‘An undertaking which is in a dominant position on the market and ties purchasers … by an obligation or promise on their part to obtain all or most of their requirements exclusively from the said undertaking abuses its dominant position within the meaning of Article [102], [when] it is undertaken in consideration of the grant of a rebate’ (para 89). It was also known that a rebate could be objectively justified and that all circumstances should be considered as to see whether it restricted the freedom of the undertakings concerned (Michelin I, EU:C:1983:313, para 14). However, in Intel the Commission went further and sought to apply the ‘as-efficient competitor test’ (AEC test) to show that the rebates would exclude an ‘as-efficient competitor’ even though this was not strictly required by the law at the time. Intel appealed to the General Court (EU:T:2014:547). The GC applied the test from Hoffmann-La Roche and dismissed Intel’s appeal. However, on appeal to the CJEU (EU:C:2017:632), the Court stated that ‘where the undertaking concerned submits, during the administrative procedure, on the basis of supporting evidence, that its conduct was not capable of … producing the alleged foreclosure effects … the Commission is not only required to analyse, first, the extent of the undertaking’s dominant position on the relevant market and, secondly, the share of the market covered by the challenged practice, as well as the conditions and arrangements for granting the rebates in question, their duration and their amount; it is also required to assess the possible existence of a strategy aiming to exclude competitors that are at least as efficient as the dominant undertaking from the market’ (paras 138–139). This formalized the application of the AEC test in the law on conditional rebates. The case was returned to the GC to apply the (effectively) new test. The GC (EU:T:2022:19) applied the test and found that there was a great number of mistakes made by the Commission in relation to assessing the ability of the behaviour to foreclose competition by excluding ‘as-efficient’ competitors. This often revolved around incomplete data, which the Commission struggled to obtain (paras 292 and 297) or arguments that the Commission had made that were based on the original test rather than the new. Consequently, the case against Intel failed. The Commission has therefore challenged this decision before the CJEU once again, almost wholly on procedural grounds. Intel is a supplier of ×86 Central Processing Units (CPUs). At the time of the relevant conduct, there were only two companies manufacturing ×86 CPUs; Intel, which was dominant and AMD, which held a significantly smaller proportion of the market, but had started to make superior CPUs to Intel and was consequently in a position to grow their market share. The Commission alleged that Intel had granted rebates to four OEMs (Original Equipment Manufacturers)—Dell, Lenovo, HP and NEC—on the condition that they purchased all or almost all their ×86 CPUs from Intel. Intel also made payments to MSH, a retailer, on condition they sold exclusively computers containing Intel CPUs. According to the Commission, these were abusive loyalty rebates. The Commission carried out an analysis using the AEC test to discern whether the rebates had the ability to foreclose a competitor as efficient as Intel, but not dominant like Intel. This appeal (EU:C:2024:915) was always going to be an uphill struggle for the Commission. The original decision that the Commission was trying to defend was written before the new test that it was being held to existed. So, it is not a surprise that the Commission’s decision struggles to address it fully. This means the weaknesses of the original decision were impossible to resolve without the Commission being able to present new data or new arguments to address the original decision’s flaws. This is a problem because procedurally, the Commission is not permitted to do either (EU:C:2008:375, EU:C:2022:478). Therefore, the Commission seems to have gone for the ‘attack is the best form of defence’ approach and tried to use procedural rules to allow it to introduce new data, new or reformulated arguments or to challenge the GC’s analysis of certain facts. It also tried to challenge the level of review of its methods to suggest that the Court should leave the application of certain economic tests within its discretion. To this end the Commission argued that the GC ruled ultra petita, misinterpreted and misapplied the criteria, infringed the Commission’s right of defence, committed errors of law concerning the standard of proof, distorted evidence and misunderstood the level of discretion the Commission had in the application of complex economic analysis. Each one of the claims is highly case specific and would often be supported by a number of alleged instances. The first set back was when the Advocate General, (Opinion of Advocate General Medina, Case C-240/22 P Commission v Intel EU:C:2024:65) having been asked to analyse only the fourth and fifth grounds of appeal (para 56), proposed dismissing both (para 169). The CJEU’s decision followed suit. Every part the appeal was rejected due to it being a misinterpretation of the GC’s judgment, a veiled attempt by the Commission to revisit the GC’s unappealable assessment of the evidence, an attempt to construct a new line of reasoning that would enable the Commission to fulfil the AEC test because the original assessment had been deemed deficient/incomplete or an attempt to rely on new information that had not been mentioned in the original decision. However, the final two grounds of appeal have scope for general application. The fifth ground of appeal related to non-cash advantages given by Intel to Lenovo to secure exclusivity. One advantage was extending Intel’s standard 1 year warranty and the second was providing the use of Intel’s supply hub in Shenzhen, China, to Lenovo. The question was raised as to how to integrate a non-cash advantage into the assessment of the size of a conditional rebate. Was the cost to the dominant undertaking or the value to the customer to whom it was provided the relevant figure? The Court stated that the purpose of the AEC test is to compare whether a hypothetical competitor as efficient as Intel could provide CPUs at a price would allow it to stay in the market (paras 307–308). The CJEU explained ‘It is true that a competitor as efficient as Intel must offer its products at a price that compensates for the loss of the conditional benefits granted by the dominant undertaking. However, that compensation does not necessarily have to take the form of a cash advantage equal to the value of the non-cash advantage for the customer concerned.’ (para 310). In other words, the Court suggested that rather than the hypothetical competitor having to drop its prices by the equivalent value of the advantage to the customer, they could just provide the same non-cash advantage themselves. As a result, this means the relevant measure will usually be the cost to the dominant undertaking not the value to the customer. The Court did concede that there can be a certain adjustment to allow for the fact that the hypothetical as-efficient competitor can only fulfil the contestable share. (para 313) But how this adjustment should take place, and in what context was not explained. The sixth ground of appeal was that Intel’s use of average avoidable cost was a conservative, favourable cost benchmark for Intel, since it disregards the industry’s very high fixed costs (para 323). The Commission also argued that even if the application of the AEC test was flawed, it was not significant enough to lead to the annulment of that decision (para 325). Further the Commission submitted that the GC’s conclusion disregarded the fact that the concept of an ‘as-efficient competitor’ is not based solely on considerations relating to price and cost but also depends on parameters such as the choice, quality and innovation demonstrated by that competitor’s products (para 326). None of these arguments were accepted by the Court. It was the Commission that relied upon average avoidable cost in its assessment (para 335) and didn’t include any other alternative measure of cost in its assessment (para 340). It was the Commission that considered the market share covered, the duration of the rebates and the AEC test and devoted parts of its decision to those factors (para 336). It was the Commission that had not applied the test correctly (para 337) and didn’t include any other method to prove foreclosure (para 340). Finally, AMD’s ‘efficiency’ in other areas such as choice, quality and innovation was irrelevant (para 342). The AEC test is a cost focused comparison between Intel and a hypothetical competitor, AMD’s costs or any other qualitative metric was not important (para 343). The vast majority of the decision confirms that unless the Commission gets the application of the AEC test and the assessment of the share of the market covered and the duration of the rebates correct in its decision, the Court will be very unforgiving when it comes to either amending arguments or adding data and analysis later on. It is now reasonably clear that when considering non-cash advantages given in exchange for exclusivity, the correct measure to use is the cost to the dominant undertaking, not the value of the advantage to the customer. Even that however, still retains some ambiguity as the Court accepted that the costs of the dominant undertaking can be amended to take into account the fact that the hypothetical competitor may have different costs compared to the dominant undertaking, due to it being limited to the contestable share of the market. How and when was not explored. Finally, the Commission’s last plea for the AEC test to be applied in a way that takes into account factors beyond price, such as quality, choice and innovation, may signal that the Commission is now suffering from ‘buyer’s remorse’. Having finally convinced the Court to incorporate the AEC test into the law, it is now choosing to point out the flaws with it (some of which had just been published around the time of the Commission’s Intel decision). This change of tack is also being reflected elsewhere. When the Commission originally wanted to incorporate the AEC test into the law, it did so through the Guidance on the Commission’s Enforcement Priorities in Applying [Art 102 TFEU] ([2009] OJ C45/2). This Guidance was ostensibly an expression of the law as it stood combined with an explanation as to how the Commission would prioritize the application of its resources, but in reality it was the Commission’s best way of influencing the law from its executive role. Tellingly, the new draft guidelines, proposed to replace the 2009 version, whilst incorporating the test in Intel broadly, makes no mention of the AEC test (paras 78–83). Whether this change will be embraced by the Court is unknown. On the one hand, the Commission is ‘required to assess the possible existence of a strategy aiming to exclude competitors that are at least as efficient as the dominant undertaking’, which sounds similar to, but does not explicitly require the AEC test. On the other hand, the Court has endorsed the AEC test as ‘one tool amongst others’ to assess the foreclosure potential of conditional rebates (Case C-23/14 Post Danmark II EU:C:2015:651, para 61) and has repeatedly stopped short of requiring it as an essential step (Case C-377/20, Servizio Elettrico Nazionale EU:C:2022:379, paras 80–81, Case T-286/09 RENV, Intel EU:T:2022:19, para 126). The final CJEU judgment follows in this pattern (EU:C:2024:915, para 340). This strongly suggests that the law still gives the Commission the flexibility to choose the method used to determine whether there is anticompetitive foreclosure and consequently, may mean that the AEC test is going to diminish in relevance in future conditional rebate cases.
Author: Matthew Cole
Institution: University of Exeter
Language: English
Subfield: Strategy and Management
Domain: Social Sciences
Open Access status: hybrid
Estimated APC paid: $4,850
OpenAlex ID: W4410064553