By Hans Vedder
On June 5 the Court has handed down the eagerly awaited judgment in the Kone case. This is one of the several cases that result from the Commission’s decision finding a cartel in the elevators and escalators sector. The decision concerned a bid rigging cartel involving four well-known firms (Kone, Schindler, Otis and ThyssenKrupp) active in the market for the production, installation and servicing of elevators and escalators. Bid rigging is a practice by which the participants in a tender procedure coordinate their bids in order to determine who wins the tender at what price. They will typically determine the cartel member intended to win and ensure that the other bidders put in a higher price. As most of these products are bought by professional buyers that tend to hang on to their purchasing records, civil damages claims resulted from the Commission’s finding that there was cartel. This means that the customers of the companies involved in the cartel seek to claim the supracompetitive part of the price they paid (the cartel mark-up). In keeping with the need for more damages claims fervently voiced by then Competition Commissioner Neelie Kroes, the Court has had to deal with quite a few cases on this issue already, but many more are to be expected. Kone deals with the question to what extent the cartelists are required to compensate the higher price charged not just by the members of the cartel, but also by other companies in the market (the umbrella effect).
Recent developments concering damages claims in cartel cases
A bit of history to start with. It begins with the desire expressed by Neelie Kroes to have more civil damages claims, i.e. the victims of the cartel claiming their damages (comprising but not confined to the mark-up) from the cartelists. This has generated a myriad press releases in which the Commission announces yet another busted cartel and then encourages the consumers to claim damages. It has also resulted in the Ashurst Report that found civil enforcement in the EU to be in a state of ‘total underdevelopment’ and ‘astonishing diversity’. Concerning the latter, judgments like the one in Kone certainly contribute to making the law in the 28 Member States more uniform. Moreover, it is just one of many cases to show that this dearth of civil enforcement was about to change with several high-profile cases that have spun off from Commission and national competition authority decisions involving notably big institutional purchasers. These cases concern gas insulated switchgear, subsea cables and the escalators and elevators central to the Kone case. They have highlighted several interesting features of the interaction between public law enforcement of competition law and the civil law procedures that ensue. The latter type of case is generally characterized as a follow-on damages claim, a typology that will also be used in this contribution for brevity’s sake. Thus far most of the relevant cases (Pfleiderer, Donau Chemie, Energie Baden-Württemberg) have dealt with questions concerning access to the file in general and access to leniency applications and related documents in particular. These cases turn the spotlight on the potentially problematic relationship between follow-on damages claims and public enforcement. To start with the latter: such enforcement is by far the most prominent form of competition law enforcement in the EU. Moreover, the overwhelming majority of these cases start with a so-called leniency application. Immunity from fines is awarded to the first company in a cartel to disclose the existence of this cartel. By awarding complete immunity to the first applicant and at maximum a 50% rebate to all others, the leniency instrument creates an asymmetry of costs between the cartel members that is likely to destabilize the cartel. Instead of balancing just the benefit of the cartel mark-up (and thus extra profitability) with the risk of detection and fining that is identical for all members, a cartel member now has to balance the extra profit with the unequally distributed risk of being detected and fined. This creates a classic prisoners dilemma, where only one cartel member gets full immunity and the others face a nearly 100% certainty of being detected and having to pay a hefty fine.
Most people agree that leniency is a very effective instrument to uncover cartels. However, most people also agree that public access to leniency documents for the purpose of a private follow-on damages claim may reduce the attractiveness of such leniency programs and thus jeopardise the effectiveness of this instrument. What could happen is that the leniency applicant may be discouraged from disclosing the existence of the cartel when the attractive prospect of immunity from fines is offset by the possible liability for damages. Cases like Pfleiderer, Donau Chemie and many national cases like National Grid deal with the balancing of the interests represented by public and private enforcement. According to the Court this balancing is to be done by the national courts in the light of the principle of effectiveness. This means that national law is to ensure the effective enforcement of EU law and thus Article 101 TFEU. While the Court focused in these cases on what could be called the beginning of the process of a follow on damages claim, i.e. the gathering of the information concerning the existence of an infringement of Article 101 TFEU, Kone is different in that it concerns an issue that arises at a later stage, when the preliminaries relating to the gathering of evidence are normally completed.
Kone and the umbrella effect
Kone is a follow-on damages case brought against the four companies involved in the cartel by the infrastructure division of the Austrian State Railway Company (ÖBB). As anyone who has ever been to a railway station can acknowledge, there are quite a few escalators in most railway stations and the stations operated by ÖBB are no exception to that rule. Having bought a large number of escalators both directly and through contractors, ÖBB sought to recover the cartel damages from Kone and the other bid rigging companies. This is obvious in those cases where the escalators, normally purchased in or as part of public tendering procedures, had been bought from one of the members of the cartel. Interestingly, ÖBB pushed the envelope beyond this obvious case and also claimed damages from the cartel in those cases where the escalators were actually purchased from a third company. In doing so it basically relied on what is called the ‘umbrella effect’.
According to the umbrella effect, a cartel mark-up will not only be charged by the cartel, it will result in higher prices across the board, so also when prices are set by companies that are not a part of the cartel. This makes some economic sense. It starts from the premise that a successful cartel needs some market power, otherwise they would be out of business pretty quickly (customers would simply take their business elsewhere). In the cartel at hand this worked out to the extent that the Court notes that more than half of the projects concerned involved rigged bids, meaning that at least one third of the market volume was covered by the cartel. Moreover, the cartel was successful in approximately two thirds of the projects, so that the project was awarded to the company chosen by the cartel who would then get the fixed price. In one third of the tenders, the rigged bid failed either because of cheating cartel members or outsiders. As a result of this, market prices and the cartelists’ market share were stable. If a functioning cartel is able to raise prices in the entire market, why then should not the cartelists be held liable for this price increase that would not have come about if it was not for the cartel? This is the essential question at hand in Kone.
Austrian law prohibited this type of causation outright, much in the way access to the file for follow-on damages claimants was banned categorically in Donau Chemie. Indeed, the fate of the Austrian rule in that case would prove to be a good indicator of what would happen to the categorical rejection of umbrella effects in Austrian civil law. In the words of the referring court as paraphrased by the Court (para. 14):
The referring court submits that, according to the adequate causal link criterion, the person responsible for damage must provide compensation for all consequences that he could foresee in abstracto, including accidental ones, but not for atypical consequences. According to Austrian case-law, when an undertaking not party to a cartel takes advantage of the effect of umbrella pricing, there is no adequate causal link between the cartel and the loss potentially suffered by a buyer, since it consists of an indirect loss: a side effect of an independent decision that a person not party to a cartel has taken on the basis of his own business considerations. The view would be taken that the effect on a competitor of market conditions, as are influenced by the cartel, the economic conclusions he draws from those conditions as regards his undertaking and as regards his goods as well as the business decisions he then takes, particularly as regards pricing, are largely determined by a great number of factors completely unrelated to that cartel.
This is then also connected to a Schutznorm-issue (para. 15):
With regard to the question of unlawfulness, the Oberster Gerichtshof has held that, in accordance with the academic writings relating to the protective purpose of the provision, the act of causing pecuniary loss entails an obligation to provide compensation only if the unlawfulness of the loss stems from a breach of contractual obligations, a breach of absolute rights or a breach of protective provisions. According to the referring court, the decisive factor is therefore whether the provision infringed by the person responsible for the loss had as its object the protection of the injured person’s interests. Such is not the case in the practice of umbrella pricing, which involves no relationship of unlawfulness. The unlawful conduct of the cartel members seeks to injure those who buy their goods at the artificially high prices they charge. The loss caused by the umbrella pricing is merely a side-effect of an independent decision that a person not involved in that cartel has taken based on his own business considerations.
The essential question then becomes how wide the net of liability under Article 101 is cast. Can a successful cartel be held liable for increasing the market price or just their own prices? In a bidding market, these two prices are easily separable, which highlights the effect of accepting umbrella pricing: the cartelists would be liable for the price set by an independent company.
In a remarkably short judgment the Court starts its considerations from the direct effect and full effectiveness of Article 101(1) to find that national procedural rules relating to claims based on EU law must be subject to the principles of equivalence and effectiveness. Whereas the former is clearly not at stake here, it could be argued that the effectiveness of Article 101(1) may be impeded. The reasoning in this regard is framed in the light of the plausibility of an umbrella effect. In the words of the Court ‘[it] is recognised [by the interested parties] as one of the possible consequences of a cartel, in certain circumstances’ (para. 28). If there is possibly an umbrella effect, the categorical rejection of umbrella claims because of insufficient causality will reduce the effectiveness of Article 101(1) TFEU. The categorical rejection is then replaced with a right to compensation ‘where it is established that the cartel at issue was, in the circumstances of the case and, in particular, the specific aspects of the relevant market, liable to have the effect of umbrella pricing being applied by third parties acting independently, and that those circumstances and specific aspects could not be ignored by the members of that cartel’ (para. 34).
Reflections on an umbrella effect
Now, I am not an economist, but I would guess that in bidding markets with a limited number of participants, these conditions could be satisfied. In a market with relatively homogeneous products and a high degree of transparency (the call for bids will specify what type of escalator or elevator is needed and which specifications must be met), all the participants to the tender, including those outside the cartel, will be able to identify the winning price and thus the market price with absolute certainty. Moreover, assuming relatively homogenous production costs, the outsider would then be able to guesstimate the profitability of the bid. If, as an outsider to the cartel, you would then notice a structurally higher price in those tenders where these four big companies have entered bids, it is not too difficult to increase your next bid to a price that is just below that which can be expected as the market price. The outsider would then only just undercut the rigged bid. This in turn, should be easy to detect for the cartelists and one could even imagine tacit collusion or oligopolistic interdependence, depending on how concentrated the market is. In a sense, the cartel would start to function as a barometric price leader, followed by the outsiders in that market. This denotes the situation where, in highly concentrated transparent markets price setting by a company perceived to be the industry leader is followed by the other companies and knows that this is the case. Note that barometric price leadership has been accepted by the Commission, meaning that this behavior does not infringe Article 101 TFEU. This is a very significant causality hurdle to be overcome, but it is not insurmountable.
The judgment further deals with some miscellanea such as the punitive character the damages may thus obtain in view of the fact that the cartelists will be made to pay for more than they have benefitted from the cartel. This is given short shrift by the Court, essentially on the basis that compensation of loss for the consumer does not presume enrichment for the persons responsible for that loss. A second point dealt with in a very succinct manner is the possible negative relation between umbrella effects and leniency. In a nutshell, an umbrella effect will increase the civil liability of the member of the cartel in a symmetrical manner. If an umbrella effect is found,, all members of the cartel will face an identical risk of being held liable for damages. This sits uneasy with the asymmetry in risks that drives the prisoners’ dilemma that powers leniency. Furthermore, and this echoes the punitive damages argument put forward by the cartelists in this case, the prospect of having to pay damages may well offset the immunity from fines. To give an indication: the Austrian authorities imposed a total fine of € 75 million and ThyssenKrupp was given full immunity, whereas Kone got a 50% reduction and had to pay € 22 million. The damages claimed by ÖBB alone amount to € 8 million of which at least € 1.8 million is attributed to an umbrella effect. Given that it is a fair assumption that the elevators and escalators were sold to other customers as well, the potential damages claims may very well outweigh the benefits of immunity, in particular if a member of the cartel is made to pay for the profit incurred by someone else. I can imagine that this will change the game during the next cartel meeting. The Court’s very formalistic way of dealing with this, where it qualifies leniency as a program developed by the Commission without binding effects on the Member States, certainly does not do justice to the uneasy relationship between a very effective detection tool and follow-on damages claims.
So, all in all, what are we to think of Kone? It has certainly been received critically, with many authors finding it overly interventionist and ineffective at the end. Albert Sánchez Graells, for example, deplores the fact that the principle of equivalence was ignored by the Court when it reduced the procedural autonomy of the Member States. However, this judgment fits in a long line of interventionist judgments by the Court, that dates back to famous cases like Van Gend & Loos, Dassonville and Cassis de Dijon. The first one is about empowering individuals in the European legal order. The second and third basically set the framework for the free movement of goods. Interestingly, the debate preceding the judgment in Dassonville centered on the question whether it entailed a ban on discrimination (i.e. an equivalence approach) or also caught rules that applied ‘without distinction’ but still reduced the effective application of the free movement of goods. We all know the Court’s answer and what it resulted in. It gave us a lawyers’ paradise, even after the Court tried to reduce the scope of this provision in Keck, that continues to generate case law and warrants new editions of already quite substantial books. But at the end of the day, it did benefit the free movement of goods and the individuals that relied on it. Mickelsson and Roos now know where they can use their jet skis legally in Sweden and Dusseldorp could export the waste oil filters to Germany. I am not so sure whether this will actually help ÖBB and other claimants. In fact, I am not so sure whether the Court wanted to go beyond paying lip service to encouraging effective private damages claims. This very short judgment makes me feel that much more was said in camera, but later removed as too controversial or too telling of a preference for public or private enforcement. At the end of the day, it boils down to the Schutznorm-issue skimmed over by the Court. Who is intended to be protected from what by competition law? Consumers are certainly protected, but are they effectively protected from supracompetitive market prices or should we rather say that effective protection from these umbrella effects is a matter of public interest, and therefore within the realm of the competition authorities? These fundamental questions will undoubtedly pop up over the coming years in the form of many smaller issues that will be dealt with by the Court, balancing effectiveness with objective justifications.