In last year’s edition, we wrote that unilateral pricing behaviour – in that case, conditional discounts – requires a careful approach to enforcement. While under-enforcement could undermine competition, over-enforcement could deter companies from charging low prices – both outcomes that are bad for consumers. The stakes are equally high in alleged predatory pricing. Unlike other conditional discounting arrangements, however, below-cost pricing might justifiably be more easily condemned as lacking a legitimate justification and aimed solely at eliminating competition (although there are exceptions, like promoting new products or clearing old stock). The trick to predatory pricing is recognising it when it occurs and avoiding condemning otherwise beneficial discounting. The recent judgment in Post Danmark1 attempts to do precisely that.
By its preliminary ruling of 27 March 2012, the European Court of Justice (ECJ), in responding to questions from the Danish Supreme Court, ruled that the fact that a dominant firm charges prices below average total cost (ATC) but above average variable cost (AVC) to attract a rival’s customers is not in itself an infringement, even if the discount is selectively made in order to target rivals’ customers. To constitute abuse, the selective price-cutting must be shown to be part of a scheme to dominate and to be capable of having that effect, rather than merely reflecting competition for customers.
On one level, the ruling confirms existing rules laid out in Akzo,2 namely that prices below AVC are presumed illegal and prices above that level but below the ATC are abusive if part of a plan to eliminate competitors. Nevertheless, the judgment sheds light on some important aspects of the EU rules on predation, such as the degree to which the A kzo standards apply to ‘selective price cutting’ and the role of ‘intent’ to eliminate competitors when pricing above ATC and below AVC. The judgment also rejects per se condemnation of form-based conduct and unequivocally adopts an ‘effects-based’ analysis which is both in line with the European Commission’s approach in recent years3 and creates useful precedent for future cases in the Courts.
EU rules on predation and selective price-cutting
The EU courts first dealt with predatory pricing in the 1991 Akzo judgment. The Commission had found that Akzo, a dominant manufacturer of chemical products, had systematically offered flour additives at below-cost prices to damage a competitor’s business viability.4 On appeal, the ECJ set out a two-tiered test for predatory pricing:
- prices below the AVC (ie, costs that vary with the amount of products produced) by a dominant firm will be presumed abusive; and
- prices below the ATC (ie, fixed costs plus variable costs) but above the AVC will be regarded as abusive if they are part of a plan to eliminate competitors.5
Pricing below AVC
In Akzo, the Court created a presumption of illegality when prices are not only below the ATC but also below the AVC. The ECJ reasoned that a dominant firm can only be interested in applying such low prices to eliminate competitors and gain a dominant position which will allow it to exercise market power.6 As the Court explained in France Télécom, the relevant measure of costs are those of the dominant firm, even if they are not below costs incurred by competitors.7 Nevertheless, this presumption could theoretically be rebutted with evidence of alternative reasons justifying the conduct.8 It is hardly surprising that rebutting the presumption is very difficult.9
Pricing below ATC but above AVC
According to the second Akzo test, when prices are below the ATC but above the AVC, they are ‘to be considered abusive only where an intention to eliminate can be shown’.10 Rules that depend on intent, particularly that of a corporation with many different actors, present difficulties for the Commission which, in addition to a cost-based analysis, must assess the dominant firm’s strategic motivations, something which can generally only be done through documentary evidence.
As to the cost-based component of the test, the explanation given by the ECJ to justify the absence of a presumption of illegality for prices that are in between the AVC and the ATC is that ‘such practices can drive from the market undertakings […] which, because of their smaller financial resources, are incapable of withstanding the competition waged against them’.11 Despite the predominance of the ‘as-efficient competitor’ test, remnants of protectionism for less efficient competitors may still be present.
The second component, the proof of the existence of a plan to eliminate a competitor, remains difficult to apply. Any company engaged in price competition is naturally predisposed to prevail in its competitive environment and ultimately to eliminate a competitor. In distinguishing legitimate from illegitimate intent, the Commission has usually considered as illegitimate any intent that shows a proactive eviction strategy that goes beyond aggressive competition. Such intent is therefore usually substantiated mainly with directly incriminating documentary evidence and sometimes by an economic incentive analysis. In Akzo, for instance, the ECJ inferred such an intent based on cost-related reasons (the fact that there was no objective justification for undercutting prices over a long period of time, especially in the absence of price reductions by its competitors),12 but mainly on the basis of direct threats made by Akzo to its rival ECS (evidenced by documents recording meetings held between the directors of the two firms).13 In Tetra Pak II, the General Court (GC) went a step further by suggesting that, absent any directly incriminating evidence, intent to eliminate competition could be found in a series of convergent factors, such as the fact that loss-making prices were charged only where it was necessary to win prospective customers or win back customers14 and the fact that Tetra Pak was not adapting its pricing to its competitor (Elopak) but was increasing the price difference by further reducing its prices.
In Wanadoo Interactive, on the other hand, extensive internal company documents (such as management-level communications) indicating the existence of a plan to pre-empt the market were used by the Commission to show the intent to eliminate competitors in the ADSL market.15 The Commission also demonstrated that the expression of the pre-emption intention was ‘conveyed and corroborated by undisputable economic factors and by the translating of the intention into commercial policy’.16 On appeal, the EU courts upheld the finding solely on the basis of documentary evidence showing intent stemming from the managerial level.17 The GC, drawing a distinction between ‘object-based’ and ‘effects-based’ analysis (unusual for Article 102 TFEU), explained that ‘showing an anti-competitive object and an anti-competitive effect may, in some cases, be one and the same thing. If it is shown that the object pursued by the conduct of an undertaking in a dominant position is to restrict competition, that conduct will also be liable to have such an effect.’18
There is thus an inconsistency in the current decisional practice concerning the type of evidence used to prove intent. While the Court is happy to find direct evidence of anti-competitive intent coming from the actors’ statements, the absence of such evidence will not necessarily absolve the existence of wrongdoing.
Unlike the US, the Commission and EU courts have refused to require the possibility of ‘recoupment’ as an element of the infringement of predation. Instead, the ability to recoup losses can be a key component when assessing the alleged predator’s strategy.
In Akzo, the ECJ acknowledged that ‘a dominant firm has no interest in applying prices below AVC except that of eliminating competitors so as to enable it subsequently to raise its prices by taking advantage of its monopolistic position’.19 The question was also addressed in Tetra Pak II where the Commission found that Tetra Pak had engaged in predatory pricing of its non-aseptic cartons in Italy. On appeal before the ECJ, Tetra Pak argued that sales at a loss could only be predatory when there was a reasonable prospect of subsequently recouping the losses incurred. The Court rejected the claim as ‘it would not be appropriate […] to require in addition proof that Tetra Pak had a realistic chance of recouping its losses. It must be possible to penalise predatory pricing whenever there is a risk that competitors will be eliminated.’ The aim of maintaining undistorted competition ‘rules out waiting until such a strategy leads to the actual elimination of competitors’.20 While the ECJ narrowed this ruling applied ‘in the circumstances of the case’, any possible confusion was eliminated following the France Télécom ruling where the GC and the ECJ reiterated that proof of recoupment was not a condition for predation.21 Interestingly enough, the Commission nevertheless provided extensive evidence showing that recoupment of losses was indeed likely.22 Ultimately, this difference between the EU and the US is unlikely to result in different outcomes. In the US, recoupment is generally proved by showing high entry barriers which thus allow the infringing firm to raise price above competitive levels once it has driven out its rivals. Though the requirement does not exist in Europe, the EU does require that the predating firm enjoys a pre-existing dominant position, a circumstance typically inconsistent with low entry barriers.
The EU antitrust rules on ‘selective price-cutting’
Predation can take the form of selective price-cutting when the predator imposes below-cost prices to a competitor’s customers or to customers that would otherwise desert to a competitor while leaving other prices to other existing customers at a higher level. Selective price-cutting may be more tempting given that it is easier for a dominant undertaking to engage in predatory conduct by selectively targeting specific customers, as this limits the losses incurred23 and enables the financing of predation from that part of the market where higher prices are offered.
Selective pricing has for the first time been analysed in Akzo. Akzo allegedly engaged in selective quotations to its competitor’s customers that were below its ATC, while it continued offering its own regular customers prices above its ATC. The Commission and the EU courts inferred anti-competitive intent from this pricing, holding that Akzo was ‘able, at least partially, to set off losses resulting from the sales to customers of ECS against profits made on the sales to its existing customers’.24 This was sufficient to demonstrate Akzo’s intention to pursue a strategy that could damage its competitor. Yet it would be wrong to conclude that below ATC pricing in itself is always an abuse, as the ruling should be read in the context of Akzo’s more general predation strategy.25
In Compagnie Maritime Belge the dominant undertaking had altered its freight rates to match or beat those of the main independent competitors for vessels sailing. The ECJ noted that the dominant undertaking had never seriously disputed that the purpose of its conduct was to eliminate its sole competitor from the market, without relying on a cost-based analysis.26 Similarly, in Irish Sugar, the dominant undertaking on the retail sugar market in Ireland was found to have given special discounts with no objective economic justification to certain retailers in the frontier region between Ireland and Northern Ireland to meet competition from sugar imports from Northern Ireland. The court concluded that the pricing was not ‘predatory’ but was still illegal, as the financing of those rebates with other sales in Ireland prevented the development of free competition on that market.27 Finally, in Eurofix-Bauco/Hilti, Hilti offered its competitors’ customers particularly favourable conditions by removing quantity discounts for customers who bought nails and nail cartridges from its competitors, and giving away some products free of charge. The Commission held that such pricing practices were abusive, not because they were below cost, but because Hilti could rely on its dominance to offer discriminatory prices with a view to damaging the competitors’ business.
These cases suggested that a selective reduction in prices by a dominant undertaking could be contrary to Article 102 TFEU regardless of whether they were above cost. The EU courts condemned the practice on the basis of internal evidence of an intention to drive out the competitor. Yet this common interpretation of these cases stretched their actual holdings. Not only were the undertakings concerned super-dominant in most cases (with a market share of 80 to 90 per cent), but the conduct could not be treated in isolation. It was part of a more general abusive predatory practice.
The role of ‘intent’
In sum, the current EU law approach entails some inherent inconsistencies that are primarily linked to the use of the subjective criterion of ‘intent’. Article 102 TFEU abuses have traditionally been considered ‘objective’ and therefore dissociated from subjective considerations, such as ‘intent’.28 The additional evidence of an intention to eliminate competition thus sits at odds with this objectivity requirement.
Apart from this conceptual discordancy, decisional practice reveals the inherent difficulties in proving intent. Just as with any other infringement of EU competition law, the existence of intent must be established on ‘cogent and convincing evidence’.29 While the Commission may legitimately rely on a presumption of illegality where prices are below the AVC, in the presence of prices above the AVC but below the ATV, only strong evidence of an exclusionary strategy can meet the requisite legal standard to prove intent.
The Commission’s 2009 Guidance Paper
These difficulties prompted the Commission to propose a clearer effects-based approach for the assessment of predation, thereby mitigating the problems associated with the proof of intent. As a general rule, in order to determine whether a (hypothetical) competitor as efficient as the dominant undertaking would likely be foreclosed, the Commission will examine economic data relating to cost and sales price. If the ‘equally efficient competitor’ can compete effectively with the pricing conduct of the dominant undertaking, the Commission will in principle infer that the conduct is not likely to have an adverse impact on effective competition, and thus on consumers. Otherwise it will integrate this finding ‘in the general assessment of anti-competitive foreclosure, taking into account other relevant quantitative and/or qualitative evidence.’30 Intent is relevant evidence of likely effects, but the Commission will not regard it as an element of the offence.
For predatory pricing, the Commission first proves a so-called ‘sacrifice’ (ie, losses or foregoing profits in the short term),31 and then a risk of ‘anti-competitive foreclosure’ of the actual or potential competitors.32 Sacrifice depends on a cost analysis of the average avoidable cost (AAC) pricing33 (ie, costs which a firm would avoid incurring by ceasing a particular activity) which gives the Commission ‘a clear indication of sacrifice’. Nevertheless, the Guidance Paper demands that the Commission may also investigate whether the conduct led, in the short term, to net revenues lower than could have been expected from a reasonable alternative conduct (ie, whether it incurred a loss that it could have avoided),34 and use direct evidence that ‘clearly show[s] a predatory strategy’.35 This test is more problematic, since it can freeze beneficial discounting merely because the firm is choosing not to maximise its profits.
In demonstrating the ‘anti-competitive foreclosure’, the Commission will apply the ‘as-efficient competitor’ test. The Guidance Paper provides that evidence of intent is relevant but not required: ‘Direct evidence of any exclusionary strategy’, such as an ‘internal document which contain[s] direct evidence of a strategy to exclude competitors […] may be helpful in interpreting the dominant undertaking’s conduct’.36 The discussion on ‘recoupment’ is more ambiguous. On the one hand, the Guidance Paper reaffirms that the Commission does not need to intervene only where the dominant undertaking is likely to be able to increase its prices at exploitative levels after the predation period. On the other hand, however, the Commission does need to show that ‘the conduct would be likely to prevent or delay a decline in prices that would otherwise have occurred.’37
The Post Danmark A/S judgment
Post Danmark (PD) and Forbruger-Kontakt (FK) were the two largest undertakings in the unaddressed mail sector (eg, brochures and telephone directories). PD was also the universal postal service provider, which at the time enjoyed a public monopoly in the delivery of addressed letters and parcels not exceeding a certain weight. For that purpose, PD had a network covering the national territory in its entirety, which was also used for the distribution of unaddressed mail. Towards the end of 2003, PD concluded contracts for the distribution of their unaddressed mail with three supermarkets that were formerly major customers of FK. The offer for one of these (Coop Group) was marginally lower than FK’s price and, while it did not allow PD to cover its ATC, it did allow it to cover its AVC. Following a complaint by FK, the Danish Competition Council held that PD had abused its dominant position by practising a targeted policy of reductions designed to ensure its customers’ loyalty by, firstly, not putting its customers on an equal footing in terms of rates and rebates, and secondly, by charging FK’s former customers rates different from those it charged its own pre-existing customers without valid cost justifications.
After a series of appeals the case was brought before the Supreme Court of Denmark which stayed the proceedings and referred two preliminary questions to the ECJ:
- under what circumstances selective pricing towards certain former customers of a competitor must be considered to amount to an exclusionary abuse; and
- whether the mere fact that the price charged was lower than the ATC but higher than the average incremental costs (AIC) of the business concerned sufficed to prove such an abuse.
The ECJ’s judgment
In responding to the question, the Grand Chamber of the ECJ first recalled that Article 102 TFEU prohibits pricing practices that have an exclusionary effect on ‘as efficient’ competitors and opposes the strengthening of a dominant position through methods other than competition on the merits.38 In determining whether an undertaking has practiced predatory pricing, the ECJ, in a remarkably short judgment, reiterated the Akzo rule that prices below the AVC are in principle abusive and prices below ATC but above AVC are abusive if they ‘are part of a plan for eliminating a competitor’. The Court thus confirmed that the lawfulness of a low-price policy depends on a cost-based analysis as well as on the undertaking’s ‘strategy’.39 Given that, according to the Danish Court it could not be established that PD had deliberately sought to drive FK out of the market, the second condition could not be met. Absent proof of intent, the ECJ’s analysis focused on cost. Upholding the Danish Court’s methodology of using the incremental costs as a relevant cost-benchmark, the ECJ found that the method of attribution of costs adequately identified the ‘great bulk of the costs attributable to the activity of distributing unaddressed mail’.40
To the extent that a dominant undertaking sets its prices at a level covering the great bulk of those costs, it will, as a general rule, be possible for an as efficient competitor to compete with those prices without suffering losses that are unsustainable in the long term.41 While it belongs to the national court to run this ‘asefficient competitor’ test, the ECJ implied that FK had managed to effectively compete on the same market despite losing those three customers, noting in particular that it had even managed to win back Coop Group as a customer.42 The Court goes a step further in advising the national court that, were it to find anti-competitive effects due to PD’s actions, it is open to PD to provide justifications for its behaviour, by either demonstrating that the conduct was ‘objectively necessary’ or that the exclusionary effect produced may be counterbalanced by advantages in terms of efficiency that are also beneficiary to consumers43 (eg, economies of scale due to the increased number of customers).44 Interestingly, the ECJ declined to analyze the risk of cross-subsidisation examined in the opinion of Advocate General (AG) Mengozzi, who explained that selective pricing should be unlawful (even when above the AVC) when it is subsidised by another market given that the only reason for maintaining such behaviour would be to drive competitors out of the market.45
The Post Danmark judgment, underpinned by a more modern and effects-driven analysis, develops the law in three important ways. First, it confirms AG Mengozzi’s analysis that, contrary to a common interpretation of existing case law, selective price cutting by a dominant firm is not automatically unlawful without a detailed cost-based analysis to assess whether it truly forecloses equally efficient competitors. A dominant undertaking that offers its competitors’ customers prices that are higher than its ATC cannot foreclose equally efficient competitors.
Second, the judgment departs from a central need to show anticompetitive intent and underlines the greater importance of assessing the real risks of foreclosing an ‘as-efficient competitor’.46 While this reading is in line with current Commission practice, this element of the Court’s judgment may be due largely to the wording of the preliminary question itself, which explicitly ruled out the presence of an anti-competitive strategy. It remains to be seen whether in future the Court will consider that pricing below the ATC but above the AVC, will be regarded as abusive only to the extent that no other rational explanation for that pricing (other than to eliminate an efficient competitor) can be found.
Third, for the first time the ECJ explicitly holds that a dominant undertaking should be able to defend seemingly predatory conduct with objective justifications as well as efficiencies that could counteract any negative effects on competition. This is consistent with the Commission’s Guidance Paper, which also is quick to say that the defence will be hard to prove.
The Court’s judgment in Post Danmark is thus another step towards condemning behaviour only when it is likely to have an anti-competitive effect, a particularly important requirement when the conduct at issue involves low pricing. Nothing in Post Danmark should be taken to mean that truly predatory pricing will be tolerated, as it will only rarely reflect competition on the merits, but the case does recognise that pricing strategies can be complicated and must be assessed in detail before being condemned.