Les dommages et intérêts dans les affaires de position dominante : l’exemple de Google Shopping
Si une multitude d’actions en réparation sont régulièrement déposées devant les juridictions de divers pays de l’Union européenne en rapport avec l’article 101 TFUE (pensez, par exemple, à l’abondance des litiges contre les constructeurs de camions), il n’en va pas encore de même pour les affaires d’abus de position dominante. Bien que la Directive régissant les actions en dommages et intérêts soit également applicable aux affaires liées à l’article 102 TFUE, le nombre de litiges (clôturés) en matière de position dominante est assez limité, en particulier ceux dans lesquels une estimation des dommages a été effectuée dans le cadre d’une procédure judiciaire. Nous n’avons connaissance que de quelques affaires, notamment Cardiff Bus et Albion Water au Royaume-Uni ; Arzneimittelpreise et Wasserpreis der Stadt Mainz en Allemagne (bien qu’avec des estimations relativement simplistes). En particulier, les tribunaux français semblent plutôt ouverts à l’estimation (et à l’octroi) de dommages et intérêts dans les affaires de position dominante. Voyez, par exemple, l’affaire Orange/Digicel, un cas d’importance, ou des plus petits litiges comme Le Berry Républicain/Aviscom.[1] Le fait qu’il y ait peu d’actions en réparation exercées à la suite d’un abus de position dominante devant les tribunaux nationaux en Europe n’est pas surprenant pour au moins deux raisons. Premièrement, les affaires de position dominante nécessitent une analyse (plus) économique de la part des autorités de la concurrence, y compris, au moins, une évaluation de savoir si l’entreprise faisant l’objet de l’enquête est dominante sur certains marchés antitrust.[2] Deuxièmement, la quantification des dommages dans les affaires de position dominante est moins évidente que dans les affaires d’entente. Cependant, comme nous le montrons dans ce post avec l’exemple de l’affaire Google Shopping de la Commission européenne (ci-après, « Commission »), un concept théorique clair peut être appliqué. Verrouillage vertical, cadre théorique[3] Le cadre théorique pour le calcul des dommages dans les affaires d’abus de position dominante concernant le verrouillage vertical peut être démontré comme suit. Premièrement, il faut préciser que l’on peut s’attendre à des incitations à l’éviction verticale de concurrents seulement si une entreprise est active à la fois sur les marchés en amont et en aval et si, sur l’un de ces marchés au moins, elle est dominante. Dans cette section, nous supposons que, comme dans l’affaire Google Shopping, l’entreprise est dominante en amont et qu’il y a verrouillage vertical. Lorsque l’abus débute, les concurrents en aval touchés par l’abus commencent à perdre de l’activité, leurs marges diminuent ou même deviennent négatives (phase d’attrition, voir le côté gauche de la figure 1 ci-dessous). La réduction des bénéfices subie par les concurrents en aval peut conduire à une marginalisation ou à une sortie complète du marché. Ces deux conséquences peuvent entraîner des dommages financiers importants. Dans cette phase d’attrition, cependant, il n’est pas certain qu’il y ait des effets négatifs sur les consommateurs. Par exemple, si l’attrition se produit par l’intermédiaire de l’entreprise dominante qui applique des prix anticoncurrentiels (inférieurs aux coûts), les consommateurs en sortiront gagnants pendant un certain temps, car ils paieront des prix plus bas. Au contraire, si l’entreprise dominante réduit de manière anticoncurrentielle la demande des entreprises rivales en refusant l’accès à une grande partie des consommateurs (comme c’est probablement le cas pour Google Shopping), il est fort probable que les consommateurs soient également affectés négativement. Par exemple, les consommateurs ne bénéficient pas d’un plus grand nombre de choix disponibles, et ils ignorent peut-être l’existence d’offres qui sont meilleures.[4] Figure 1 – Analyse de la marge pour les affaires de dommages et intérêts relevant de l’article 102 du TFUE, cadre simplifié Source : version modifiée de la figure 2 dans Fumagalli et. al. (2010) La phase d’attrition est suivie d’une phase de récupération («recoupment phase», partie centrale de la figure 1 ci-dessus), au cours de laquelle l’entreprise dominante récolte les bénéfices de l’éviction réussie des concurrents, en appliquant des prix monopolistiques maximisant les profits sur le marché en aval (manifestement nuisibles pour les consommateurs). Ce faisant, cependant, l’entreprise dominante encourage l’entrée sur le marché à un stade ultérieur : c’est ce qu’on appelle la phase de croissance du marché («market growth»), illustrée à droite de la figure 1 ci-dessus. Des marges élevées peuvent attirer l’entrée dans la phase de croissance, mais il n’est pas certain que les concurrents précédemment évincés puissent revenir ou reprendre des parts de marché (au cas où ils n’auraient pas totalement quitté le marché). Même dans ce cas, il se peut qu’ils ne soient pas en mesure de fournir aussi efficacement après avoir perdu des bénéfices au cours des périodes précédentes, ce qui affecte, entre autres, leur capacité à innover ou à réaliser des économies d’échelle. Bien qu’il ne soit pas habituel que les trois phases puissent être clairement distinguées dans le cas d’un abus de position dominante, la figure 1 ci-dessus représente l’idée qui sous-tend les dommages liés aux affaires relevant de l’article 102 du TFUE. En ce qui concerne la quantification des dommages, la différence entre les affaires d’entente et de position dominante ressort également clairement de la figure 1 ci-dessus : alors que dans les affaires d’entente, l’évaluation contrefactuelle se concentre sur les prix hypothétiques, l’évaluation des dommages dans une affaire de position dominante est effectuée sur les bénéfices contrefactuels des concurrents lésés. Les prix sont toujours importants dans le contexte d’un calcul de la marge, dans lequel on peut évaluer si l’entreprise lésée pourrait maintenir une marge suffisamment élevée pour «survivre» en présence du comportement abusif. Application à Google Shopping En ce qui concerne l’affaire Google Shopping, les domaines en concurrence avec Google Shopping peuvent avoir droit à une réparation de leur préjudice ou à des dommages-intérêts sur la base de la perte des profits ou du manque à gagner imputable au comportement abusif de Google (c’est-à-dire que l’algorithme de Google a rétrogradé la visibilité de ces sites de comparaison de produits). Il y a deux éléments principaux à prendre en compte : (i) la perte des profits ou le manque à gagner était-il imputable au comportement anticoncurrentiel de Google ; et (ii) quel
Paris Court of Appeal provides guidance on damage estimation in competition law cases and awards EUR 250 million in damages
Orange Caraïbe et al. vs Digicel Antilles Françaises Guyane On 17 June 2020, the Paris Court of Appeal (Court) delivered its judgment (Orange judgment) in which it ordered Orange to pay EUR 249.5 million (EUR 181.5 million in damages plus EUR 68 million in interest) to Digicel Antilles and Guyana, following the implementation by the former of a number of anti-competitive practices in the early 2000s.[1] This is one of the most important compensation amounts granted by French courts in the context of an action for damages related to anti-competitive practices. The Orange judgment provides key clarifications on how damages are estimated and quantified in relation to abuse of dominance cases. In particular, the Court addressed in the judgment three forms of damages that relate to: Factual background On 9 December 2009, the French competition authority (FCA) adopted a decision by which it sanctioned Orange Caraïbe and France Telecom (together referred to as Orange) a total amount of EUR 63 million for implementing anti-competitive practices in the mobile telephony market in the Antilles-Guyana (Decision).[2] In the early 2000s, Orange Caraïbe was the established and main operator in the area holding more than 75% of the cell phone market.[3] In its decision, the FCA considered that Orange Caraïbe and France Telecom’s anti-competitive practices resulted in hindering potential competitors’ (Bouygues Telecom Caraïbe (BTC), now Digicel, together referred to as Digicel) from entering and developing on that market.[4] The practices consisted of exclusivity agreements with distributors, a subscriber loyalty program, and abusive price discrimination.[5] Following the Decision, Digicel brought an action for damages before the Paris Commercial Court. In December 2017, the Paris Commercial Court upheld most of Digicel’s claim and awarded an amount of EUR 179.64 million in damages, applying to this sum an actualisation rate (taux d’actualisation) of 10.4% from the date of notification of the writ of summons.[6] Subsequently, Orange appealed this ruling, ultimately resulting in the judgment by the Paris Court of Appeal subject of this article. Brief discussions The Court’s reasoning on the estimation and quantification of harm covers three main forms of damages that are further addressed below. 1 Digicel’s potential development/growth assessed on the basis of comparative time and geographic methods The Court qualified damages caused to development or growth as lucrum cessans (loss of profit).[7] Concretely, this is to be understood as the non-realisation of a profit or Digicel’s loss of developing itself on the Antillean-Guyanese mobile telephony market because of the anti-competitive practices. Loss of profit is the type of damage which is primarily compensated by French courts in abuse of dominance cases.[8] The Court stated that the abusive practices[9] committed by Orange had ‘cumulated over time and reinforced each other, all contributing to a single overall result: hindering the development of Digicel on the Antillean-Guyanese cell phone market’.[10] Given the interaction of the different practices and the difficulty in isolating the effects of each one of them, the Court justified that an ‘overall’ comparative approach can be taken to evaluate the damage caused to Digicel’s lost turnover. To evaluate the damage pertaining to Digicel’s growth, the claimant proposed two complementary comparison-based methods. The first method is an estimation of the market’s progression or evolution in the absence of the anti-competitive practices, by comparing the period not concerned by the offence to the period when the anticompetitive practices were operating (i.e., comparison over time). The second method is based on the growth experienced by entrants in mobile telecommunications markets comparable to that of the Antillean-Guyanese market in terms of structure and concentration (i.e., comparison between geographic markets). The Court agreed with the claimant’s approach and referred to the European Commission’s Communication on quantifying harm in actions for antitrust damages to uphold the use of comparative methods.[11] The Court asserted that both comparative methods resulted in a very close result in terms of quantification of the damage pertaining to Digicel’s development,[12] demonstrating the relevance of both methods. 1.1 Method of comparison over time The Court upheld the analysis of Digicel identifying the damage related period (2002-2005) and the post-damage related period (2006-2009). The first period starting in 2002 was characterised by Digicel’s unusual stagnation of its market share in the mobile telephony market due to the anti-competitive practices put in place by Orange. In the post-damage related period (2006-2009), Digicel was considered no longer to be impacted by the anti-competitive practices.[13] Orange attempted unsuccessfully to identify pitfalls of this comparative method. In particular, it argued that Digicel failed to account for major changes which happened just between both periods such as (i) Digicel acquiring BTC in April 2006, (ii) the development of new technologies and consumption habits, (iii) the entry of other operators on the market, etc. Orange argued that these particular changes accounted for Digicel’s better growth in the post-damage related period (2006-2009), not the termination of the anti-competitive practices. Orange also pretended that Digicel’s market-share stagnation during the damage related period (2002-2005) could not have solely resulted from the anti-competitive practices. The Court rejected Orange’s arguments by stating that Digicel’s stagnation in the damage related period could not simply have been the result of a general market evolution but instead resulted from the anti-competitive practices. In this regard, the Court confirmed that Digicel’s better growth in the post-damage period was linked to the termination of some of the anti-competitive practices rather than caused by the ‘major changes’ identified by Orange between both periods. 1.2 Method of comparison between geographic markets This comparative method consisted of a comparison analysis with other geographic markets having similar structure than Antillean-Guyanese mobile telephony market. It aims at estimating the market share that Digicel would have had in the absence of the anti-competitive practices, namely the number of customers and earnings lost. Orange argued that this method was erroneously drawn up since the counterfactual scenario was based on a sample of only four countries (namely Bulgaria, Ireland, Luxembourg and Malta), which according to Orange (i) is too limited to be representative, and (ii) presents no similarities with
EU Court of Justice confirms right to claim damages in relation to after cartel effects
By judgment of 14 January 2021 in Case C-450/19, Kilpailu- ja kuluttajavirasto (‘Judgment’), the Court of Justice of the European Union (‘CJEU’) stated that an infringement of Art. 101 TFEU in bid-rigging cases ends with the conclusion of the works, goods or services contract, even if the damage occurs at a later stage. To reach this conclusion, the CJEU made a distinction between the practices distorting competition and their damaging effects. Although related to public enforcement, the judgment provides interesting guidelines regarding damage claims. In particular, the judgment confirms that the EU right to full compensation also encompasses harm which materialises after the infringement of Art. 101 TFEU has ended, i.e. lingering or after-effects. Background The referral for a preliminary ruling originates from the Finnish Administrative Supreme Court (Korkein hallinto-oikeus) and contains a single question asking when exactly the infringement of Art. 101 TFEU in a bid-rigged tender procedure ceases. According to the referring court, four moments could eventually be considered as the end of the infringement: (1) the submission of the tender, (2) the conclusion of the procurement contract (3) the payment of the last instalment of the agreed price or (4) the completion of the works (para. 19). Given the novelty of the legal issue, the judgment was preceded by an opinion of Advocate General (‘AG’) Pitruzzella delivered on 10th of September 2020. In paragraph 32 of his opinion, the AG explains that the determination of the end of an Art. 101 TFEU infringement must be contextualized and might differ from one case to another. The opinion hints towards a more structural difference in relation to the end of the market distortion between bid-rigging cases and other types of infringements of Art. 101 TFEU. A brief discussion After recalling that agreements to share customers and on fixing prices fall into the category of the most serious restrictions of competition, and that Art. 101 TFEU forbids bid-rigging practices, the CJEU addresses the end of infringement concerning a manipulated tender procedure. The CJEU gives paramount importance to the moment of the conclusion of the contract between the undertaking and the contracting authority, provided that the essential characteristics, in particular the overall price to be paid, have been definitively determined (para. 41). According to the CJEU, this is the ultimate moment until which the effects on the restriction of competition resulting from the prohibited conduct continue to persist (para. 34). To reach this conclusion, the CJEU draws up a key distinction between on one hand the cartel’s restrictive effects on competition, which in bid-rigging cases deprive the contracting authority of the opportunity to obtain the agreed goods, works or services under competitive conditions, and, on the other hand, the resulting wider damaging effects, on the basis of which harmed persons may seek compensation before national courts (para. 37). Comments With its judgment, the CJEU correctly confirms that the right to full compensation encompasses harm resulting from lingering or after-effects. Such damage positions are also caused by the infringement, but only occur after the end of the infringement. This is in line with the established CJEU’s case law (e.g. Case C‑435/18Otis and Case C-557/12Kone) according to which any person has the right to ask full compensation for any harm that s/he has suffered as a result of an infringement of Art. 101 TFEU, provided that a causal link between the infringement and the harm is established. By Juraj SISKA & Vasil SAVOV
Trucks Cartel: German Supreme Court confirms the binding effect of settlement decisions and the presumption of damages
By judgment of 23 September 2020 (KZR 35/19, translation into English) the German Federal Court of Justice (‘Bundesgerichtshof’, ‘BGH’) provides detailed guidance on legal principles and standards of proof regarding several key aspects of actions for damages resulting from the European trucks cartel. The Bundesgerichtshof confirms that the scope of the infringement established in the binding European Commission’s (‘Commission’) settlement decision is significantly broader than a mere exchange of information. Further, the BGH confirms that a factual presumption for higher prices caused by the trucks cartel applies. Factual Background The judgment concerns a damage action brought by the parent company of two construction companies against Daimler AG. The subsidiaries purchased trucks from Daimler AG during the cartel period and had assigned their cartel-related damage claims to the plaintiff. According to the findings of the Commission in its settlement decision of 19 July 2016, Daimler and at least four other truck manufacturers had infringed Article 101 TFEU and Article 53 of the EEA Agreement by agreeing on prices and list price increases for medium- and heavy-duty trucks, as well as on the timing and passing-on of costs for the introduction of emission technologies in accordance with the EURO emission standards. The EEA-wide infringement covered 14 years, from 1997 to 2011. Based on the findings of the Commission, the plaintiff claimed that its subsidiaries had to pay inflated purchase prices and higher risk premiums due to the cartel. It seeks compensation amounting to a total of EUR 305,000, plus interest. The first and second instance courts decided in favour of the plaintiff and dismissed compensation only for one truck purchased in 1997. Daimler challenged the second instance judgment of the Court of Appeal to the BGH, in particular in relation to the application of the factual presumption that the trucks cartel resulted in higher prices and hence a damage. The Bundesgerichtshof held that the Court of Appeal was right to find that such presumption exists. But the BGH found that the Court of Appeal did not sufficiently consider the application of the presumption in the case at hand, taking into account the nature of the cartel as established by the Commission in its settlement decision. On that ground, the Bundesgerichtshof annulled the first appeal judgment and referred the case back to the Court of Appeal as second instance court. However, the BGH in its judgment provides the Court of Appeal with clear guidance on the standards for applying the presumption of damages in the trucks case that are also relevant beyond the individual case. Findings of Commission in settlement decision and binding nature for national court At the outset, the Bundesgerichtshof confirms that Daimler infringed Art. 101 TFEU and acted deliberately wrongful in this respect. It summarises the findings of the infringement as established by the Commission with a focus on price collusion and on how the infringement was implemented by the trucks manufacturers. a) Infringement consists of collusive arrangements on pricing and gross price increases, not a mere exchange of information Pursuant to the Bundesgerichtshof, the infringement as found by the Commission consists of “collusive arrangements on pricing and gross price increases for medium and heavy duty trucks and in the coordination of their market behaviour on the timing and passing on of the costs of introducing emission technologies for such trucks according to the EURO 3 to EURO 6 emission standards” (para. 19). The BGH rejects the truck manufacturers’ argument that these findings would merely concern an exchange of information: “As stated above (recitals 19 et seq.), the cartel participants discussed their future list prices and their increases with each other and coordinated their future pricing both through agreements and through concerted practices. (…) [S]uch behaviour is fundamentally different from a mere exchange of information” (para. 42 et seq.). According to the Bundesgerichtshof, the agreements as set out in the Commission decision enabled the companies involved to take the information exchanged into account in their internal planning processes and in planning future list price increases for the coming calendar year. The truck manufacturers’ coordination of list prices were the starting point for any individual pricing at the end customer level: “The list prices set by the respective headquarters were the starting point for pricing for all truck manufacturers involved in the agreements; then the invoice prices for importing the trucks into different markets by own or third party distributors and then the prices to be paid by the dealers on national markets were set. Finally, the retail prices were negotiated and set either by a dealer or, in the case of direct sales to dealers or fleet customers, directly by the manufacturer” (para. 22). b) Binding effect extends to all findings of fact and law With reference to its established case law (judgment of 12 July 2016, KZR 25/14), the Bundesgerichtshof (para. 24) stresses the binding nature of the findings of the Commission: “The binding effect therefore extends to all findings of fact and law with on which the competition authority bases its reasons for the infringement of substantive competition law.” It rejects the defendant’s argument that this binding effect is excluded or limited in the case at hand because the Commission decision of 19 July 2016 was issued in the context of settlement proceedings. Also, the Bundesgerichtshof refers to paragraph 3 of the settlement decision from which follows that all addressees «accepted» the findings of the Commission. Trucks purchasers were affected by the infringement The Bundesgerichthof then confirms that the claimant’s subsidiaries as purchasers of trucks were affected by the cartel, and – thus – are entitled to claim damages. It held that “the precondition for a claim for damages under antitrust law (…) is that the opponent of the claim is guilty of conduct restrictive of competition which – by concluding transactions or in another way – is capable of directly or indirectly causing damage to the claimant (…). Nothing else applies to a claim based on a violation of Article 101 TFEU or Article 81 EC
Beyond ORWI: German Supreme Court continues clarifying the ‘passing-on defence’ in cartel damages cases
Cartel members sued for damages regularly invoke as a defence that the plaintiff passed on the overcharge resulting from their unlawful behaviour to its own customers (the ‘indirect purchasers’) and was hence not entitled to claim damages for it. In the well-known ORWI judgment of 28 June 2011 (KZR 75/10), the Federal Court of Justice in Germany (Bundesgerichtshof, ‘BGH’) relying on primary EU law allowed that passing-on defence, as well as own claims for damages by the indirect purchasers. It was the first decision in those matters by a Supreme Court in Europe, and of great impact. The ORWI principles are widely reflected by the later Directive 2014/104/EU and hence by the national laws of all EU Member States that have implemented this Directive. Recently, on 23 September 2020, the BGH rendered even two further landmark judgments on antitrust damages, one (the BGH’s first) concerning the international trucks cartel (KZR 35/19, Trucks), and another one (out of many) on a domestic cartel of rail producers (KZR 4/19, Rails V). In both cases, albeit from different angles, the BGH again takes the opportunity to provide fundamental, helpful, clarifications on the passing-on defence. Especially in Rails V, where in fact it may be assumed that overcharges had been passed on from the plaintiff even down to the end-consumers, it becomes relevant in that regard that ‘the legal concept of damage must be distinguished from the economic concept’. General requirements of the passing-on defence In Trucks, the defendants asserted that, following the payment of an overcharge, the plaintiff’s subsidiaries had charged higher prices to their own customers for the construction services they had provided. Similarly, in Rails V, the rail producers argued that damages of the there-plaintiff, the City of Essen, should be reduced by benefits that accrued to the latter from its fare increases for local transport services in the cartel period. Confronted with these objections, the BGH recalls ORWI at first. Accordingly, passing on costs would not eliminate the damage (‘in the legal sense’, the BGH clarifies in Trucks), because the buyer’s damage (‘finally’) arises with conclusion of the contract obliging it to pay the overcharge. However, under the general principles of the set-off of benefits, an infringer can invoke that any benefit received by the claimant in an adequate causal context of the harmful event has to be deducted from the damages claim. This forms the basis of the passing-on defence in private antitrust cases under German law. Reading Trucks and Rails V together, the BGH then draws the following conclusions: Burden of proof The cartel members bear the burden to demonstrate and prove the prerequisites of the passing-on defence. A ‘secondary burden of proof’ on the part of the cartel victim with regard to the detailed circumstances of its price calculation vis-à-vis its own customers shall ‘only be considered in exceptional circumstances’. Sufficient evidence on a cartel-induced price increase on the downstream market on the basis of the relevant market conditions In principle, for a successful passing-on defence, it is up to the cartelists to show that, based on the general conditions on the relevant downstream market (‘in particular the elasticity of demand, the price development, and the product characteristics’), passing-on of the cartel-induced price increase (i.e. the overcharge) is, at least, a serious possibility. The complexity of the economic context in the individual case may be taken into account. The ‘strict requirements’ of that defence shall not be applied in a way that makes it impossible for the cartelist from the outset to sufficiently present it. The BGH rather points to the power of the courts to estimate the passing-on of an overcharge. No presumption of passing-on in favour of the cartel defendant On the other hand, given that pricing can be influenced by numerous factors of market structure and commercial strategy, neither price increases on the downstream market temporally close to the cartel, nor the fact that a corporate plaintiff has a typical interest in aligning the price of its goods or services with the costs, are sufficient to justify the passing-on defence, generally. The empirical proposition of price effects of a cartel, on which direct purchasers of the cartelised good may base their damage claims, does also not allow conclusions on the probability of a pass-on to indirect purchasers by itself. Rather, the cartelist must provide sufficient evidence that a price increase on the downstream market precisely resulted from the cartel and not from other factors. Individual assessment of different downstream markets If a pass-on to different downstream markets is possible in the case in question, especially because the plaintiff resold the cartelised product into more than one distribution channel, the cartelist must show ‘separately’ for each of those markets ‘that, and in what way,’ a cartel-induced price increase influenced this market. In fact, the Trucks defendants did not meet this requirement, given different downstream markets in that case, namely one for construction services and another one for used trucks, for any of those two markets. For this reason, the BGH fully rejected the passing-on defence in Trucks. Refusal of the passing-on defence to avoid de facto absence of liability (especially in case of passing-on of overcharges to end-consumers) In Rails V, in contrast, based on an in-depth analysis of the market and financing structures relevant for the pricing of local public transport, the BGH accepted that the defendants had sufficiently shown that a passing-on of overcharges may be considered seriously. But only ‘from an economic perspective’. Holding that the concept of damage has both an economic and a legal side, the BGH then elaborates a well-reasoned answer to the legal question, still left open in previous decisions, of what happens if effects of passing on to the downstream market are at best marginal and determinable only at great expense, while there is a risk of unreasonable relief for the cartel infringers due to a lack of enforcement of possible damage claims by the customers at the downstream market level. This is especially, but not
Private damages in dominance cases: the example of Google Shopping
While a plethora of damages claims are regularly filed in front of jurisdictions in various countries of the European Union related to Article 101 TFEU (think, for example, the abundance of litigation against truck manufacturers), the same is not yet true for abuse of dominance cases. Notwithstanding that the Damages Directive is equally applicable to Article 102 TFEU cases, the number of (closed) dominance litigations is fairly limited, especially the ones in which a damage estimation was carried out in court proceedings. We are only aware of a few cases, including Cardiff Bus and Albion Water in the United Kingdom; Arzneimittelpreise and Wasserpreis der Stadt Mainz in Germany (although with relatively simplistic estimations). Notably, French courts seem to be rather open to estimate (and award) damages in dominance cases. See, for example the landmark case of Orange/Digicel or smaller cases like Le Berry Républicain/Aviscom.[1] The fact that there are few dominance damages claims in front of national courts in Europe is not surprising for at least two reasons. First, dominance cases require (more) economic analysis from the competition authorities including, at least, an assessment of whether the company investigated is dominant in certain antitrust markets.[2] Second, the damage quantification in dominance cases is less straightforward than in cartel cases. However, as we show in this post with the example of the Google Shopping case of the European Commission (‘Commission’), a clear theoretical concept can be applied. Vertical foreclosure, theoretical framework[3] The theoretical framework underlying damage calculations in abuse of dominance cases concerning vertical foreclosure can be demonstrated as follows. First, we need to clarify that one can only expect incentives to vertically foreclose competitors if a company is active both in the upstream and in the downstream markets, and at least on one of these markets the company is dominant. In this section we assume that, similarly to the Google Shopping case, the company is dominant upstream, and vertical foreclosure takes place. When the abuse begins, downstream competitors affected by the abuse start losing business, their margins are shrinking or even turn negative (attrition phase, see left hand side of Figure 1 below). The reduction of profits suffered by downstream competitors may lead to marginalization or a complete exit from the market. Both outcomes may lead to substantial financial damages. In this attrition phase, however, it is not clear whether there are adverse effects on consumers. If, for example, the attrition happens via the dominant company applying anti-competitively low (below-cost) prices, consumers may be better off for a while, as they pay lower prices. On the contrary, if the dominant company anti-competitively reduces rival firms’ demand by denying access to a large proportion of consumers (as it is likely in the case for Google Shopping), consumers are most likely negatively affected too. For example, consumers do not benefit from a larger number of available choices, and perhaps they become unaware of better deals.[4] Figure 1 – Margin analysis for Article 102 TFEU damages cases, simplified framework Source: modified version of Figure 2 in Fumagalli et. al. (2010) The attrition phase is followed by a recoupment phase (middle part of Figure 1 above), during which the dominant company reaps the benefits of the successful foreclosure of competitors, by applying profit-maximizing monopoly prices in the downstream market (clearly harmful for consumers). By doing so, though, the dominant company encourages market entry at a later stage: this is called the phase of market growth, shown on the right hand-side of Figure 1 above. High margins may attract entry in the growth phase, but it is uncertain that the previously foreclosed competitors can re-enter or regain strength (in case they did not fully exited the market). Even so, they may not be able to supply as efficiently having lost out on profits during the preceding periods affecting, among other things, their ability to innovate or to realise economies of scale. While it is not typical that all the three phases can be clearly distinguished in the case of an abuse of dominance, Figure 1 above represents the idea behind damages related to Article 102 TFEU cases. Concerning damage quantification, the difference between cartel and dominance cases is also clear from Figure 1 above: while in cartel cases the counterfactual assessment focuses on but-for prices, when assessing damages in a dominance case the analysis is carried out on the counterfactual profits of the harmed competitors. Prices are still important in the context of a margin calculation, in which one can assess whether the damaged company could maintain a high enough margin to “survive” in the presence of the abusive behaviour. Application to Google Shopping Turning to the Google Shopping case, the domains competing with Google Shopping can be entitled to damages based on the lost profits attributable to Google’s abusive behaviour (i.e., that Google’s algorithm demoted visibility of these product comparison websites). There are two main things to consider: (i) was the loss of profits attributable to Google’s anti-competitive behaviour; and (ii) what would have been the counterfactual level of profits of the competing price comparison websites? Loss of business While it is the level of margins that matters, there are other metrics that can be used as indicators of profits (or losses) from running price comparison websites. One of these metrics is the Sistrix Visibility Index (which has also been used by the Commission in the Google Shopping decision[5]). Based on Sistrix data, we generated Figure 2 below showing the Visibility Index of certain high-volume French comparison-shopping websites (domains are selected according to Graph 5 of the Google Shopping decision, the difference is that the lines below are shown to date). We use French data for demonstration, but visibility indices of other comparison websites from other countries show a similar picture. Figure 2 – Sistrix Visibility Index for selected comparison-shopping services, France Source: Sistrix.com The reasons for showing the visibility indices to date are twofold: first, Figure 2 shows more evidently that Google’s demoting algorithm may have caused a long-lasting negative impact on
Full steam ahead: German Supreme Court provides further guidance on ‘umbrella claims’ and the ‘passing-on defence’
Are there no public benefits from a hardcore cartel? The German Rails Cartel, at least, continues to contribute to the general development of private claims for damages resulting from an infringement of the EU cartel prohibition (Article 101 TFEU) and its national equivalent. From this perspective, it supports consumers, by promoting legal certainty for an effective compensation of cartel-based losses. Several decisions of Germany’s Supreme Court (Bundesgerichtshof, ‘BGH’) originating from that cartel have already answered questions on applicable law, joint and several liability of the cartel infringers, as well as the standard of proof for causality and quantification, in antitrust damages actions. Now, for a further landmark judgment dated 19 May 2020 (KZR 8/18, Rails Cartel IV), the BGH has followed the rails on a curvy route to the stations ‘umbrella claims’ and ‘passing-on defence’. Again, the Court’s engine is richly fuelled by principles of EU law. Providing a route map In 2013, the German competition authority fined producers of rails, points, and sleepers for infringing national as well as EU cartel prohibition by agreements to fix prices and allocate customers between 2001 and 2011. This has led to many follow-on actions before different civil courts in Germany. The plaintiff in the case at hand is the City of Munich. It claims that the cartelists are jointly and severally liable for damages because it had to pay overcharges for products affected by the anticompetitive agreements to both members and non-members of the cartel. The Higher Regional Court (Oberlandesgericht) of Munich as the appeal court found for the plaintiff on the merits, assuming damage caused by the cartel, but leaving open its extent. On further appeal, the BGH rejected this assumption. Remanding the decision back to the appeal court, the BGH gave a route map, i.e. instructions, for the new damage assessment in that case, however. This sets the course for any similar litigation for antitrust damages as well. Station 1: umbrella claims – a regular companion Grounds for liability The ride starts with the plaintiff’s right to claim damages from the defendants. The BGH held, as it did with regard to prior actions against the rails producers, that a defendant is liable if it is guilty of an anticompetitive conduct which – through sales transactions or otherwise – is ‘capable’ of directly or indirectly harming the plaintiff. The question of whether the cartel actually affected a certain transaction is not relevant for the causality giving rise to liability. It is sufficient that the plaintiff acquired from cartel members goods which were subject to the cartel agreement. Following the tracks laid by the Kone judgement of the European Court of Justice from 2014 (C-557/12), the BGH added that the purchase of such goods from persons not participating in the cartel (non-cartelists) was likely to cause damage to the plaintiff as well. Non-cartelists benefiting from the protection of the cartel’s practices may have set their own prices higher than they would have done under competitive conditions. Thus, possible effects of a cartel include overcharges resulting from such ‘umbrella pricing’. They may also be claimed from the defendants. Occurrence and extent of a cartel-caused loss resulting from umbrella pricing Turning to the damage assessment, the BGH denied prima facie evidence, contrary to the Munich appeal court. The question of whether and to what extent particularly umbrella pricing was present in a cartel-affected market depends on a variety of factors, given the competitive interdependency of the market players, and typically can only be answered by a comprehensive assessment of the circumstances of the individual case. For this, the BGH refers to the principles formulated in Kone. Even so, under ‘certain conditions’, the non-cartelists ‘can be expected’ to be able to charge higher prices on the market than they would have done without the cartel. According to the BGH, ‘this probability is higher, the greater the market coverage of the cartel and the longer the infringement lasts.’ The BGH continued that the ‘extent’ of umbrella effects depends on several factors. The ‘umbrella effect is higher’, the lower the supply elasticity of the non-cartelists, the greater the market transparency, the greater the degree of substitutability of the products offered on the market and the lower the intensity of competition between the non-cartelists and the competitive pressure from the demand side. Trial judges shall rule on this at their own discretion and conviction, based on an evaluation of all circumstances, under their statutory power to estimate damages. Travelogue Article 17(2) of the European Directive 2014/104/EU on antitrust damages provides a rebuttable presumption that cartel infringements cause harm. The BGH did not have to deal with the question of whether this includes umbrella pricing. The statutory presumption is not yet applicable to the Rails Cartel case. Even so, under the legal standard of the BGH, umbrella claims are likely to succeed in typical cartel cases. Station 2: passing-on defence – a ghost train? The rail producers tried to build a buffer to protect themselves against liability by arguing that the City of Munich (the plaintiff) could not claim damages for paying any overcharge because it had received from the Free State of Bavaria public subsidies for the transactions in question. Furthermore, the plaintiff had allegedly passed on the overcharge, through a subsidiary, to the passengers using the local transport services, by increasing the fares. The ‘passing-on defence’ is a common tool of antitrust defendants. In Europe, even before Directive 2014/104/EU, not only a direct purchaser may claim damages, but also an indirect purchaser, provided it has suffered harm because upstream purchasers passed on the overcharge. In turn, the infringer can argue that the claimant reduced its loss by passing it on to its own customers. Particularly in Germany, it is settled case law that, under the general principles of the set-off of benefits, a cartel infringer can invoke as a defence that any benefit the claimant has received in an adequate causal context of the harmful event has to be deducted from the claim, provided that this
Spanish Trucks Cartel Litigation: Temporal application of EU Damages Directive provisions on limitation and quantification of harm
In the aftermath of Directive 2014/104/EU (Directive), Spain appears to be one of the most active jurisdictions dealing with antitrust damages cases. In particular, a multitude of Spanish courts have been seized with actions for compensation in relation to the European Trucks Cartel (Case AT.39824-Trucks). In that context, preliminary ruling requests were addressed by Spanish judges to the Court of Justice of the European Union (CJEU) on key legal aspects such as the establishment of alternative forum delicti under Article 7(2) of Brussels I bis Regulation (C-30/20, RH vs Volvo) and the civil liability of legal entities that are part of the same economic unit but not addressees of fining decisions (C-882/19, Sumal). Recently, a reference (C-267/20, RM vs Volvo et DAF Trucks) was addressed to the CJEU by the Regional Court of Léon (Court) on the temporal application of the provisions implementing Articles 10 (limitation periods) and 17 (qualification of harm) into national law (Request). In its Request, the referring Court seeks to determine, if the application of these national provisions to claims that are presented to a court after the implementation of the Directive but relate to an antitrust infringement which had ended prior to the Directive’s entry into force would be precluded by Article 22 (non-retroactivity) of the Directive. The response by the CJEU might have significant implications for pending and future actions across the European Union (EU), and could bring additional legal certainty for claimants, defendants and judges involved in complex antitrust damages litigation. Factual context The Directive entered into force on 26 December 2014. The deadline for implementation was 27 December 2016. The Kingdom of Spain implemented the Directive on 27 May 2017 by Real Decreto-Ley 9/2017. On 19 July 2016, the European Commission (EC) found in a settlement Decision (Decision) that several trucks’ manufacturers participated in a single and continuous infringement of the EU cartel prohibition (and the EEA equivalent provision) for the period between 17 January 1997 to 18 January 2011 (European Trucks Cartel). A short press release on the adoption of the Decision was made public on 19 July 2016. A non-confidential version of the Decision has been published on the website of the EC’s Directorate-General for Competition on 6 April 2017. On 1 April 2018, RM, who had acquired three trucks during the period between 2006 and 2007, brought an action for damages before the Commercial Court of Léon (Action) against AB VOLVO and DAF TRUCKS NV (Defendants) for the harm caused as a result of their participation in the European Trucks Cartel. In its judgment of 15 October 2019, the Commercial Court of Léon decided inter alia that: no claim is time-barred under the new five-year limitation period; existence of harm is presumed in relation to the European Trucks Cartel; overcharge could be estimated at 15% of the trucks’ purchase price. This judgment was appealed to the referring Court where the Defendants argue that: (i) the Directive and the national law implementing its provisions on limitation periods and quantification of harm are not relevant in relation to claims resulting from an infringement which ended in January 2011, and (ii) the applicable one-year Spanish pre-Directive limitation period had started to run from the date of the press release on the adoption of the Decision and had expired prior to the initiation of the Action. Discussion With the Request, the CJEU is asked in substance to clarify: (i) whether Articles 10 and 17 of the Directive constitute the Directive’s substantive provisions, and (ii) whether Article 22 of the Directive precludes the immediate application of the respective implementing measures to claims resulting from an infringement which ended prior to the Directive’s adoption. Limitation periods Regarding limitation periods, the national judge would like to determine if the new five-year limitation period, under the rule that implements the Directive into national law, could be applied immediately to claims for which the one-year pre-Directive limitation period would have started to run, but did not yet expire. In substance, the CJEU is asked to establish whether Article 10 of the Directive constitutes a substantive provision and, if this is the case, to determine whether the intended application of the national implementing measures to the claims resulting from the European Trucks Cartel would be precluded under Article 22 of the Directive. In that context, it should be mentioned that in some legal orders limitation periods are of procedural nature (e.g. United Kingdom). In other legal orders, where limitations are of material nature, transitional provisions were adopted on the immediate application of the five-year period to claims not yet time-barred under pre-Directive rules (e.g. Germany). Importantly, in Spain the intended application of the five-year rule is not specifically regulated by the Real Decreto-Ley 9/2017 and the legal basis thereof seems unclear. Abundant case law exists in Spain already in the context of damages actions against the European Trucks Cartel. These judgments inter alia deal with questions on the application and the starting point of the pre-Directive Spanish limitation period. In that regard, Spanish courts, including appeal courts, have found that the one-year limitation period started to run only on 6 April 2017 with the publication of the non-confidential version of the Decision, and not on 19 July 2016 with the publication of the press release on the adoption thereof. This was for instance the outcome of the judgment by Valencia Court of Appeal of 16 December 2019, in Don Jose Pedro. In the present case, should the interpretation of the pre-Directive rule, in line with the EU principle of effectiveness, require that the one-year limitation does not start running prior to 6 April 2017, then the claims subject to the Action, which was initiated on 1 April 2018, would not be time-barred. Guidance on the interpretation of pre-Directive limitation rules in relation to claims rooted directly in EU competition law has already been provided by the CJEU in its landmark judgment in C‑637/1, Cogeco. Quantification of harm Regarding the quantification of harm, the national judge would like to
Paris court of appeal awards EUR 249 million in antitrust damage action against Orange
On 17 June 2020, the Paris Court of Appeal (hereinafter the “Court”) handed down its judgment (hereinafter the “Orange ruling”) in which it ordered Orange to pay EUR 249.5 million (EUR 181.5 million in damages plus EUR 68 million in interest) to the telecom operator Digicel for anti-competitive practices implemented in the West Indies in the early 2000s . This is one of the most significant sanctions granted by French courts in the context of a compensation dispute relating to anti-competitive practices. The Orange ruling is the outcome of a “follow-on damages” action which was brought following the adoption, on 9 December 2009 by the French Competition Authority (FCA), of a decision sanctioning the two companies Orange Caraïbe and France Telecom (now Orange) for a total amount of EUR 63 million. In this decision, the Competition Authority concluded that these companies had implemented a number of anti-competitive practices such as exclusivity agreements, a subscriber loyalty program and price differentiation practices at the time when Bouygues Telecom (now Digicel) was attempting to enter the Antillean-Guyanese market. This decision of the FCA was upheld on appeal and in cassation. The Orange ruling is remarkable for several reasons: First of all, this judgment illustrates particularly well the finding that civil compensation can far exceed the fines imposed by the FCA. Secondly, in this judgment the Court draws very largely on the case law of the Court of Justice of the European Union and the provisions resulting from the transposition of the Damages Directive, which are not applicable to the case in point, to consider that civil liability is established by the final decision adopted by the FCA. Among the principles identified by the Court of Justice and the Damages Directive, the Court of Appeal recalls that “the right to compensation covers not only the losses suffered and the loss of profit but also the payment of interest. The Court of Justice specified that full reparation must include compensation for the negative effects resulting from the passage of time since the occurrence of the damage caused by the infringement, namely monetary erosion, but also the loss of chance suffered by the injured party as a result of the unavailability of capital” (page 30). The court of Appeal also raises the principles set out by the European Commission in its Communication on quantifying harm in actions for damages based on breaches of Article 101 or 102 of the TFEU, “When foreclosed competitors seek redress, they may seek compensation not only for the profits they did not make during the infringement, but also for the profits they were deprived of after the infringement ceased. Such compensation makes sense, in particular, when they have not been able to return to the market or to regain full market share because of the ongoing appreciable effects of the infringement that has been terminated. Compensation would thus be sought for future profits, that is those that are likely to be lost once the damages case is judged. “(page 29). Several other interesting points relating to procedure, substance and the general approach of the Court are to be noted in this judgment: Regarding the statutory limitation, the starting point of the limitation period is the date of adoption of the Decision (i.e. 9 December 2009). Indeed, the Court specifies that “it is on the day the Authority ruled on the merits of the objections, the liability of the practices in question, their duration and the convictions that the victim was able to act in full knowledge of the facts, it being of little importance that the victim had itself referred the matter to the Competition Council” (page 17), even though Digicel had denounced the said practices to the FCA as early as July 2004; Concerning the wrongdoing, the Court then confirmed the wrongful nature of the anti-competitive practices by stating that “the violation of the law resulting from this anti-competitive practice necessarily constitutes a civil fault (…) of such a nature as to engage [the] civil liability [of Orange and Orange Caraïbe] on the basis of Article 1240 of the Civil Code” (page 21). This is a significant statement by the Court as it departs from the judgment of the Paris Commercial Court on the issue of exclusivity, whereby it was deemed that Digicel had indeed been the victim of the exclusive distribution practice; Lastly, in relation to the damage, Digicel had requested the condemnation in solidum of Orange Caraïbe and Orange to compensate its loss of profits, the extra costs and its financial prejudice: With regard to the valuation of the loss of profits, the Court validates the use of an overall assessment considering that “the different practices have accumulated over time and strengthened each other” (page 41), all contributing to a single overall result. The Court also specifies that this global assessment is justified by the difficulty of isolating the specific effects of each practice (page 41); Concerning the assessment of the damage, the Court approves the use of comparative methods (comparison between geographical markets and comparison over time) which are in line with the recommendations of the European Commission. In this respect, it considers that “these methods appear particularly suitable for assessing loss of profits linked to exclusionary practices, since, by virtue of the principle of full reparation, the injured party must be placed in the situation in which it would have been had the infringement not occurred, which leads to the construction of a “counterfactual” scenario of what the normal evolution of the market would have been if the practices had not existed” (page 42). The Orange decision also teaches us that Digicel submitted two comparative methods that the Court considered to be complementary since they lead to an assessment that is very close to the damage suffered excluding discounting; In relation to the additional costs paid by the victim as a result of the practices in question, the Court acknowledges the existence of damages linked to the additional costs resulting from, on the one hand, the effects of the exclusivity which continued after
Actions for damages and unlawful State Aids
Under EU law, State aids are in principle prohibited unless authorised by the European Commission. In cases where Member States grant a State aid to an undertaking without obtaining the prior authorization of the European Commission, competitors of the beneficiary of this unlawful State Aid may obtain damages for the financial harm they may suffer as a result of this Aid. Our former colleague Marc Barennes and Benjamin Cheynel, a référendaire with the Court of Justice of the European Union examine in a post published on the leading French legal think tank blog Le Club des juristes, two issues: firstly, under what conditions undertakings harmed by an unlawful State aid may obtain redress and, secondly, why the number of damage actions in the State aid field should increase in the coming years further to the Covid-19 Crisis. This post (in French) is available on the Club des juristes’s blog at this link.