Why litigate in Germany?
There are obvious reasons why Germany is one of three European jurisdictions in which private antitrust litigation, in particular private damage claims, have become a significant factor of competition law enforcement. Following significant changes introduced to its national competition legislation in 2005, German courts are building up a body of case law step by step, clarifying issues that needed clarification. Under German law, its civil courts are already bound by a national authority’s decision establishing an infringement of competition law (provided that this decision has become final and unappealable).1 In a growing body of judgments, German courts considered that decisions by competition authorities establishing an infringement constitute prima facie evidence that the sale of cartelised products during the cartel period had been effected by the cartel as found by the competition authority. In addition, there is prima facie evidence that at least long-lasting cartels lead to inflated prices and thus lead to damages sustained by those purchasing the cartelised product.
Whereas German law only allows for mere compensation of damages sustained and thus ignores multiple or punitive damages, German law grants pretrial interest as of the moment in which the damage occurred.2Although interest payment only compensates for the lapse in time, it adds substantial value to damage claims in case of long-lasting cartels. There are a number of cases in which interest due exceeds the nominal damage amount. Germany also offers reasonable rules on limitation periods. The typical three-year limitation period will not start prior to the claimant knowing (or reasonably being expected to know) of the infringement and its damage. Even more importantly, limitation periods are suspended during the proceedings of a competition authority, subsequent appeal proceedings and for an additional six-month period. Finally, the German court system continues to offer some significant advantages. Cartel damage cases are decided by specialised chambers in a limited number of first instance courts. Some of these courts have acquired considerable expertise. With respect to costs, the German system, despite some uncertainties, continues to offer advantages too. Although there is a ‘loser pays’ rule, compensation of the adversary’s fees are limited to statutory attorney fees.
Growing importance of private antitrust litigation
Germany has a long-standing tradition of private antitrust litigation. In the past, these cases had been predominantly dealing with refusals to supply, challenges to the validity of agreements and discrimination by dominant or near-dominant undertakings. In addition, following liberalisation of the electricity and gas markets, network access litigation had initially been based on antitrust rules. While these more traditional cases continue to be litigated, the landscape changed significantly with the emergence of cartel damage cases. There are currently several hundreds of cartel damage cases before the German courts. With respect to the German rail cartel alone, OPPENLÄNDER is handling more than 50 court cases. The German sugar cartel fined by the Federal Cartel Office has probably created an even larger wave of lawsuits.
This change in numbers is primarily due to a change in German corporate culture. There is a growing awareness in corporate Germany that long-lasting cartel infringements typically create damages and thus harm to those purchasing cartelised products or services. As these damages typically reduce profits, it is generally considered that the management of a corporation is under a legal duty to evaluate the possibilities to recover damages from those inflicting the harm. In many cases, this evaluation will lead to the conclusion that it is in the company’s interest (and in the interest of its shareholders) to seek compensation. This is in sharp contrast to an attitude prevailing perhaps a decade ago according to which antitrust damage claims should be avoided as the claimant itself could also become subject to antitrust charges. The change in culture is probably best illustrated by Deutsche Bahn AG, the state-owned railway company. Although being itself charged with antitrust infringements, Deutsche Bahn started to vigorously enforce cartel damage claims. This example has been followed by a number of large corporations that, like Deutsche Bahn, created their own in-house departments dealing with cartel damage claims. The result of this development is an increased number of claims, many of which are dealt with in private settlements but an increasing number of them end up in court.
Damages Directive will lead to significant changes
Without a doubt, the German legislator will be required to introduce a significant amount of changes to the substantive and procedural rules governing private antitrust litigation in Germany. Unlike in other European jurisdictions, implementation of the Directive3 will not establish but improve the system of private antitrust enforcement. It seems unlikely that the German legislator will meet the Directive’s deadline as only a first internal draft from the Ministry of Economics had been made public when this article went to press.
It is widely expected that significant changes to procedural rules will have to be introduced in order to comply with the Directive’s provisions on the disclosure of evidence. Currently, there are no specific rules on evidence for private antitrust litigation. Plaintiffs therefore have to rely on a couple of rarely used provisions which, in theory, allow courts to order defendants or third parties to produce specific documents.4 As the German judicial culture, like in most ‘continental’ European countries, is rather opposed to disclosure, these rules have been sparsely used thus far. Implementation of the Directive with respect to disclosure of evidence will bring about fundamental changes, if not a small revolution.
As these changes will be of a procedural nature, they will affect litigation that starts after the entry into force of the legislation implementing the Directive. With respect to substantive issues, private antitrust litigation will for quite a long time be governed by the rules currently in place. As a matter of fact, Germany could serve as an example for the long time span needed to have private damage claim cases decided under new substantive rules. This type of litigation being essentially formed by follow-on cases investigated by the authorities for years, the rules applicable during the time of the infringement will for a long time remain those currently in force.
Direct and indirect plaintiffs
For all practical purposes, private damage cases in Germany are cases brought by direct plaintiffs (ie, those who purchased the cartelised goods or services directly from the cartelists or their competitors). In its landmark case ORWI,5 the German Federal Supreme Court had clearly established that under German and European competition law rules, not only direct purchasers, but all those harmed by the infringement are entitled to compensation. With respect to substantive rules, it is thus clearly established that indirect purchases are entitled to bring compensation claims.
For indirect purchases it is, however, extremely difficult to substantiate damages caused by the cartel to them. As already mentioned, the German rules of civil procedure are lacking disclosure provisions which would enable indirect purchasers to access means of evidence available to both the cartelists and direct purchasers. It therefore comes as no surprise that claims by indirect purchasers are rare and practically limited to cases in which there is an enhanced level of transparency with respect to pricing in relation to direct customers.
Most frequently, the issues linked to claims of indirect customers are dealt with in the context of attempts to counter a possible passing-on defence. Similar to the European Damages Directive, the flip side of indirect customers being entitled to damages is the ability of cartelists to argue that damages sustained by direct customers had been passed on, wholly or in part, to subsequent levels of the supply chain. Although accepted in principle, many details of a passing-on defence under German law are still unclear. In the context of indirect claims, we would limit ourselves to point out attempts to counter a possible passing-on defence by having damage claims of indirect purchasers assigned to direct purchasers.
There are no per se restrictions regarding the assignment of damage claims under German law. Some restrictions apply with respect to assignments to special purpose vehicles the financial means of which might be regarded as insufficient to cover subsequent cost compensation claims in court proceedings. A major difficulty with respect to the assignment of damage claims of indirect purchasers to direct purchasers is the need to agree on a ratio for the distribution of proceeds. Claims of indirect purchasers will depend on the degree to which damages have been passed on by the direct purchaser. Direct purchasers are generally reluctant to give their customers too detailed information on calculation and cost structures which might be used to their detriment in subsequent commercial negotiations. This favours distribution ratios applying rules of thumb or determined by a jointly appointed expert after collection of the damages. The assignment mechanism, however, only works in relation to a distinct number of indirect customers. It is most likely no solution in relation to widely disbursed private end-customers. In the absence of collective redress mechanisms, they continue to lack meaningful mechanisms to claim their damages under German law.
‘Binding effect’ of infringement decisions, standard of proof
Under general standards, the burden of proving a competition law infringement and the damage caused by this infringement is on the plaintiff. This burden, however, is alleviated by a number of mechanisms. We already mentioned the statutory provision in section 33 paragraph 4 ARC, according to which the court shall be bound by a finding of the European Commission or a national European competition authority that an infringement has occurred. Although there is still some debate around the precise reach of this ‘binding effect’, it is established that a civil court deciding on a damage claim is bound by a final and unappealable decision that a competition law infringement as described in this decision took place. This necessarily includes the participants in the cartel (or more generally in the infringement), the type of competition law infringement, but also the facts upon which the authority relied to establish the infringement. This ‘binding effect’ considerably reduces the difficulties of any plaintiff to prove what generally is a secret infringement the details of which are normally hidden to the outside world. This binding effect, on the other hand, increases the need to obtain access to the decision of the relevant authority. Unlike the European Commission, which is obliged to publish a non-confidential version of its fining decisions, the FCO6 will only publish press releases and, at its own discretion, case reports. We will address possible difficulties to assess the FCO’s fining decisions in the context of ‘access to evidence’ below.
The ‘binding effect’ of decisions by competition authorities is limited to the infringement and does not include damages, even in cases in which the FCO had to estimate the proceeds of the cartel in order to determine the fines imposed. There is, however, according to the German jurisprudence, a prima facie evidence that long-lasting cartels lead to increased prices as this is considered to be the very reason why they are concluded. The plaintiff in cartel damage cases can rely on this prima facie evidence, which will normally enable a court to hand down a declaratory judgment according to which the defendant cartelist is obligated to compensate damages caused by the cartel to the plaintiff.
Depending on the competition authority’s decision, it may be less apparent that a given cartel affected not only ‘many’ customers but also affected the purchases made by a given plaintiff. In many cases, decisions by competition authorities will not enter into a detailed description of transactions or customers affected by the cartel. Competition authorities will typically refer to specific examples but, for their own purposes, are only interested in setting out the infringement, its participants and the duration which will affect the level of the fine. Based on this approach, decisions by competition authorities will normally not set out that a given cartel did affect purchases made by a given plaintiff. Typically, descriptions of the infringement will be of a more generic nature. This will in part reflect a generally cautious approach of the authorities. In particular in settlement decisions it might also be the result of attempts by the addressees to mitigate negative effects by increasing uncertainty for potential plaintiffs.
At least some German courts of first instance have helped plaintiffs to overcome difficulties linked to generic descriptions of infringement in the authority’s decision. According to a judgment of the lower court of Frankfurt/Main, long-lasting cartel arrangements including quota and ‘preferred supply’ arrangements lead to a prima facie evidence for a generally increased price level for these products during the cartel period. This generally increased price level, in turn, creates prima facie evidence for purchases of cartelised products made during the cartel period being affected by cartel-inflated prices. This and similar judgments are under appeal. They are nevertheless indicative of a certain reluctance of German courts to allow cartelists to benefit from more generic descriptions of the infringement in order to escape liability for claims made by specific customers.
There is another important element which reduces the plaintiff’s burden of proof. It is a general feature of German procedural rules which allow the court to estimate whether an infringement caused damages and, in particular, to estimate the amount of damages. Under this general provision,7 the court is under an obligation to estimate at least minimum damages to the extent the person harmed did provide a suitable basis for estimation. Currently, there is, unfortunately, still no clarity as to the minimum requirements for this ‘basis’ to be provided by the plaintiff. There are indications that a court might estimate cartel damages based on a comparison of average product prices during and after the cartel period.8Courts could also rely on econometric evidence introduced by plaintiffs to estimate damages. There seems, however, to be a certain reluctance of German courts to venture into estimation without the help of a court-appointed expert. At this stage, we are still lacking decisions on the use of econometric evidence and the standard to be applied by court-appointed experts to allow for an estimation by the court.
Also based on general rules, it is on the defendant cartelist to substantiate and to prove any pass-on of cartel damages he or she wants to argue. There are apparent difficulties for a cartelist to substantiate and even more to prove a level of pass-on given his or her ignorance with respect to pricing and market conditions on subsequent levels of the supply chain. Nevertheless, the German Federal Supreme Court in its ORWI judgment was only prepared to impose a secondary burden of proof on the cartelist’s customer claiming damages in very limited circumstances. Such secondary burden of proof could only be accepted following a case-by-case balancing of all relevant factors. The court seems more inclined to impose a secondary burden of proof to the extent a plaintiff claiming pass-on is able to substantiate the mechanisms and conditions which made a pass-on likely. On the other hand, the Federal Supreme Court is aware of the possibly sensitive nature of information on pricing to be forwarded by the cartelist’s customer in order to discharge the secondary burden of substantiation.
Access to evidence
By their very nature, cartels are secret infringements. Cartelists not only tried to disguise the existence of a cartel but also typically deployed significant effort in hiding the detailed mechanisms of the cartel’s operation. In addition, data which would allow to estimate overcharges created by the cartel are held by the cartelists. It is therefore obvious that those seeking redress either need access to means of evidence in possession of the cartelists or, at least, need access to information obtained by the cartel authorities during their investigation of the cartel. We already mentioned that the German system of civil procedure has practically no or extremely limited means to allow for some form of disclosure.
Theoretically, plaintiffs could rely on section 142 of the German Code of Civil Procedure and apply to the court to order a defendant cartelist or a third party to produce certain documents. German courts seem extremely reluctant to make use of this provision. In their view, production of documents runs counter to the traditional principle that the parties each have to adduce evidence in their favour. In this area, the procedural rules currently applicable in Germany are clearly at odds with the requirements under article 5 of the Damages Directive. The genuine test for the Directive’s implementation will be to what extent the German legislator provides not only stipulations in line with the Directive, but workable mechanisms changing the deeply rooted traditions of German civil litigation with respect to private antitrust litigation.
In the absence of disclosure mechanisms, it is no surprise that potential plaintiffs turned to the competition authorities’ files as the only other means of access to information on a secret infringement. At first look, there should be no difficulties for persons potentially harmed by a cartel to obtain access to the FCO’s files regarding this infringement. Section 406e of the German Code on Criminal Procedure granting a right of access to file is also applicable to the FCO’s administrative proceedings. Under the wording of section 406e, it is plain that a person harmed by the infringement only needs to show legitimate interest in order to have full access to the authority’s file. Access to the file is only to be restricted if preponderate interests of the accused or third parties worth being protected are at stake. Although the provisions on section 406e of the German Code of Criminal Procedure work perfectly in any case of traffic accident involving police investigations, there are significant difficulties relying on this provision in order to obtain access to the FCO’s file. The FCO has so far held the position that section 406e – contrary to its wording – only allows potential victims access to a redacted version of the FCO’s fining decision. The FCO has constantly refused proper access to its file (ie, to the full file of its investigation). Only recently, in relation to the sugar cartel, the FCO also granted access to protocols on the hearing of witnesses which had been specifically mentioned in the request for access to the file. The FCO’s practice has so far been backed by the Local Court of Bonn, the only court to hear appeals against the FCO’s decisions on access to file.
The FCO’s restrictive attitude has made it impossible to date for potential cartel victims to rely on documents in the FCO’s file which ease either the proof of damages or the proof of specific purchases being affected by the cartel.
Under the general rules of civil procedure there is another possibility to introduce the files of authorities prosecuting cartels into cartel damage proceedings. Based on section 273 of the German Code of Civil Procedure, a Berlin court of first instance had requested the public prosecutor’s office to make its files on criminal investigations relating to the elevator and escalator cartel fined by the European Commission available to this court. The defendant cartelists in the Berlin proceedings appealed against the public prosecutor’s stated intention to make the files available. In this appeal, the FCO supported the cartelists as amicus curiae in order to protect not only those parts of the file formally originating from the FCO, but also documents related to leniency applications. The Higher Regional Court of Hamm rejected the appeal.9 The subsequent constitutional complaint brought by the cartelists remained unsuccessful. The Constitutional Court confirmed the Higher Regional Court’s decision.10 The latter had denied any specific character to leniency applications or parts of the European Commission’s confidential fining decision. These documents, according to the Higher Regional Court, are no extraordinary sort of data, but constitute ‘merely a detailed description of behaviour constituting an infringement’. According to this judgment, it is for the court receiving the files requested to decide to what extent there are restrictions on the parties of the civil litigation having access to the file’s content based on legitimate interests of confidentiality. In the specific case, the court still has to decide to what extent the plaintiffs obtain access to the files of the criminal proceedings.
Many of the fining decisions handed down by the European Commission or national authorities in recent years did concern long-lasting infringements. In addition, the authorities’ own proceedings often took considerable time. Appropriate rules on statutory limitation are thus important to ensure effective redress for those harmed by cartels. As briefly mentioned above, the modifications introduced into the German Act on Restraints of Competition (ARC) in 2005 already introduced a specific suspension mechanism for proceedings by cartel authorities. According to section 33(5) ARC, the limitation period for claims for damages is suspended if proceedings are initiated either by the European Commission or a national competition authority. The suspension ends six months after the proceedings are terminated, which may include several levels of court appeals. The standard three-year limitation period, which starts to run at the end of the year in which the claim arose and the claimant obtained knowledge of the circumstances giving raise to the claim and of the identity of the infringer (or would have obtained such knowledge if he or she had not shown gross negligence), therefore typically does not constitute a major obstacle to bringing successful antitrust damage claims. In many cases the required level of knowledge will only be obtained once the non-confidential version of a Commission decision has been published or, in case of a German proceeding, the plaintiff obtained detailed knowledge on the cartel by publications of the FCO or access to the fining decision.
The situation is more complex with respect to the maximum limitation period of 10 years. This limitation period starts regardless of knowledge or grossly negligent lack of knowledge on the day the claim has arisen. In particular in cases of long-lasting cartels and relatively short investigation periods, possible claimants run the risk of at least part of the oldest claims becoming time barred before they are able to enter into meaningful settlement talks or bring court proceedings. Although the suspension rules equally apply to this maximum limitation period, it remains questionable to what extent this period is compatible with the effet utile requirements under European Union law. Clearly this maximum limitation period is incompatible with the Damages Directive and its application will thus have to be formally excluded in antitrust infringement cases by the legislator.
Typically, potential claimants will try to avoid risks of statutory limitation by entering into tolling agreements in which the infringing party waives the right to invoke statutory limitation of claims.
As briefly mentioned above, German law expressly provides for interest to be paid as from the occurrence of the damage. In particular for long-lasting cartels, this considerably increases the value of claims. There is still some doubt as to the exact interest level to be applied. In many cases courts tend to apply the statutory interest rate under section 288(1) of the German Civil Code, which amounts to five percentage points above the basic rate of interest. Others consider cartel damage claims to be ‘claims for payment’ and thus hold the higher rate of eight percentage points above the basic rate of interest under section 288(2) of the German Civil Code to be applicable. Surprisingly there is still considerable debate as to the grant of interest prior to the entry into force of section 33(3) ARC expressly providing for pretrial interest in cartel damage cases. As the right to full compensation under EU law includes the payment of interest, it seems beyond doubt that interest is also due for damages caused prior to 1 July 2005. Again, there is doubt as to the exact interest level. Even prior to 2002, the interest level cannot be lower than 4 per cent, the then applicable standard legal interest rate. It is undisputed that, like any other payment claim, damage claims for cartel infringements carry trial interest as of the date of pendency.
Courts and procedures
There are courts specifically designated to hear private antitrust litigation, typically one to three for each of the German states. Each of these courts will typically designate one of its chambers to hear first instance cartel damage claims. These chambers being specifically designated does not automatically mean that they are in a similar way experienced to deal with cartel damage claims or cartel matters in general. A couple of courts turned out to be more well-versed in antitrust matters and less impressed by the bulky files of cartel damage cases. Others acquired a reputation for dragging on cases for years. It remains to be seen to what extent the implementation of the Damages Directive will lead to a further degree of concentration in courts competent to hear damage cases which is considered by many to be a prerequisite of courts taking into account other actions for damages by claimants from different levels in the supply chain required under article 15 of the Damages Directive.
Unfortunately, there is still some uncertainty as to the local jurisdiction of courts in antitrust cases. Cases can easily be brought in the courts of the defendant’s domicile and in the courts for the place where the harmful event occurred, which will typically include the place where the purchaser of the cartelised goods is seated. German courts are hesitant though to consider that damage claims against participants in a nationwide cartel can be brought before any German court based on the harmful effects also occurring in the place of that court. In addition, there is no clear equivalent in German procedural law to article 6 No. 1 of the Brussels Regulation.11 In cases involving only German defendants, the fact that a defendant is jointly and severally liable for a cartel infringement does not allow him or her to be sued in the courts of the place where another jointly and severally liable cartelist is domiciled. There are, however, possibilities to have a court designated to be competent to hear a case involving defendants domiciled in various places for which there is no single court competent.
Due to joint and severally liability of cartelists under German law, any cartel damage claim will most likely lead to court proceedings with either several defendants held jointly and severally liable or cartelists receiving third-party notices and likely to intervene in support of the defendant cartelist.
These third-party interventions create additional cost risks for the claimant. According to a strongly criticised judgment of the Higher Regional Court of Düsseldorf,12 all third-party intervenors prevailing in the litigation are entitled to have their legal costs compensated based on the full amount in dispute. Other courts consider this result to be inappropriate. They consider that all intervenors joining the defendant have an equal share in the amount in dispute, which leads to cost compensation claims by third-party intervenors based on their per-capita share.
Cartel damage claims before German courts tend to be complex and lengthy cases. There is a growing tendency by first-instance courts first to decide separately on the general entitlement to damages and to leave the question of quantum to subsequent proceedings.
Joint and several liability
Under general German tort law, cartelists are regarded to be joint tortfeasors and thus jointly and severally liable. Implementation of the Damages Directive will require the German legislator to limit the liability of immunity applicants for infringements following the entry into force of this legislation. Joint and several liability leads to possible compensation claims between cartelists. In principle, a tortfeasor compensating a victim has a compensation claim against other joint tortfeasors. There are, however, a couple of problems attached to these claims. First of all, these compensation claims are likely to be time-barred in many cases as the cartelists have knowledge of the infringement and the three-year standard limitation period thus starts running immediately. The suspension rules in section 33(5) ARC do not apply to these compensation claims.
Furthermore, there is considerable uncertainty as to the respective shares of liability to be applied to the compensation claims. Section 426(1) of the German Civil Code provides for equal portions of all debtors ‘unless otherwise determined’. German jurisprudence tends to favour shares based on the respective responsibility for the infringement which, in cartel cases, is all but easy to determine. In many cases it would seem appropriate to grant contribution claims based on the respective supply shares but this has not been established in German jurisprudence.
- Section 33(4) ARC (Act on Restraints of Competition, GWB).
- Section 33(3) ARC.
- Directive 2014/104/EU of the European Parliament and of the Council of 26 November 2014 on certain rules governing actions for damages under national law for infringements of the competition law provisions of the Member States and of the European Union, OJ L 349, 5 December 2014, p.1.
- Eg, section 142 of the German Code of Civil Procedure.
- Judgment of 28 June 2011, KZR 75/10
- Federal Cartel Office, Bundeskartellamt.
- Section 287 of the German Code of Civil Procedure.
- Eg, the Regional Court of Dortmund, judgment of 1 April 2004 regarding the vitamin cartel.
- OLG Hamm, decision of 21 November 2013, 1 VAs 116/13–120/13 and 122/13.
- Bundesverfassungsgericht, decision of 6 March 2014, 1 BvR 3541/13.
- Council Regulation (EC) No. 44/2001 of 22 December 2000 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters, OJ L012, 16/01/2001 pp. 1–23 as amended subsequently.
- OLG Düsseldorf, judgment of 18 February 2015, VI-U (Kart) 3/14.
Tel: +49 711 60187 0
Fax: +49 711 60187 222
OPPENLÄNDER is a mid-sized German firm providing high-end legal advice. The firm is perhaps best known for its constantly high-rated competition practise. The five partners strong competition team at OPPENLÄNDER has acquired a strong reputation across the board of competition law matters – ranging from German and EU merger control across dominance matters to German and EU cartel cases as well as public procurement matters. OPPENLÄNDER has particular expertise in private competition litigation. Starting with the landmark case for bundled claims in Germany against the Cement-cartel, OPPENLÄNDER has brought numerous claims in follow-on cases for both, large corporations and medium-sized firms. We believe no other German law firm to date has handled more court cases for plaintiffs in private competition litigation. Relying on teams including seasoned litigators, OPPENLÄNDER is familiar with the particularities of all relevant antitrust courts in Germany. Like most German competition practises, OPPENLÄNDER, however, is not restricted to plaintiffs work. The firm has successfully handled private damage claims on the defence side as well. Apart from competition law, OPPENLÄNDER offers particular expertise and specialised advise in corporate, IP and regulatory matters, including energy and pharmaceuticals/life-sciences.