AGENDA 2025: Reform of Cartel Damages Law

AGENDA 2025: Reform of Cartel Damages Law

This article is part of the D’Kart Spotlights: AGENDA 2025, in which experts from academia and practice comment on aspects of the Competition Policy Agenda presented by the Federal Ministry of Economic Affairs and Climate Action (BMWK). The contributions already published can be found here.

Cartel Damages Law is also part of the agenda of the BMWK. Christian Kersting takes this opportunity to adress the issues of the Antitrust Damages Directive.

The Competition Policy Agenda of the Federal Ministry for Economic Affairs and Climate Action (BMWK) until 2025 was presented in a ten-item paper “for sustainable competition as a cornerstone of the social-ecological market economy” dated February 21, 2022. The ninth item aims at strengthening EU competition policy and increasing transparency. One aspect envisaged is the reform of the Cartel Damages Directive 2014/104/EU:

“[…] we want to strengthen cartel prosecution through an initiative to better protect immunity recipients and reform the EU Cartel Damages Directive.”

In the following, I will consider areas in which the Cartel Damages Directive needs to be reformed. The Commission’s report that was to be submitted to the European Parliament and the Council by December 27, 2020, in accordance with Article 20 (1) of the Directive, does not provide any guidance in this regard. In its report of December 14, 2020, the Commission essentially states that it is still too early for an evaluation in view of the long implementation periods and the fact that there is not yet sufficient case law. At least from a German perspective one has to disagree. Even if there is indeed still a lack of ample case law, a certain need for reform can already be identified. The initiative mentioned by the BMWK to improve the protection of immunity recipients is just one example. Not in every case does an existing need for reform require an amendment of the directive; in some cases, an adjustment of national law would suffice. However, the aim of harmonization argues in favor of adopting improved provisions in a uniform manner throughout Europe. In the following, some suggestions for an improvement of private antitrust enforcement will be made, which can be implemented by a reform of the directive or (within the framework of the existing directive) of national law:

Relationship between primary law and the Directive

From the beginning, the relationship between primary law and the Directive or national law, respectively, was difficult to determine. The discussion about the protection of leniency documents, which are awarded absolute protection under Art. 6(6) of the Directive, while the ECJ in DonauChemie had just rejected such absolute protection, demonstrates this tension (see already here). The situation is further complicated by the fact that the boundaries between European primary law and national law are blurred by the ECJ’s decisions in the Skanska and Sumal cases. In both decisions, the ECJ derived at least the liability of the undertaking and all of its constituent parts, but also other elements of the cartel damages claim directly from Art. 101 TFEU. It is an open question how this fits in with the Directive and national law. The correct understanding would be to continue to consider, the claim for damages as a claim under national law with primary law only requiring a specific result. In the understanding of the ECJ in Sumal primary law would even enforce this result directly, should the national law not be able to.

In order to avoid legal uncertainty and friction, it could be made clear in the recitals of the Directive that the claim for damages is, in all its aspects, assigned to national law. To this end, reference could be made to the statement of the ECJ in Skanska that directives may confer regulatory powers on the Member States (there para. 34). In addition, reference can be made to the settled case law of the ECJ according to which “any national measure taken in an area which has been the subject of exhaustive harmonisation at the level of the European Union must be assessed in the light of the provisions of that harmonising measure and not in the light of the provisions of primary law ” (see TAP at para. 49). In this respect, one may also clarify that the level of harmonization does not leave room for deviation.

Collective redress

The Directive deliberately omits regulations on collective redress. At the same time, it demands the (almost) impossible: on the one hand, all injured parties must be able to claim and to obtain full compensationfor their harm (Art. 3(1) of the Directive). On the other hand, full compensation must not lead to overcompensation (Art. 3(3) of the Directive). This is difficult to achieve in view of the fact that both direct and indirect purchasers or providers may claim damages (Art. 12(1) of the Directive). The question at which level the damage occurred and what amount of damage was incurred constantly presents itself. If the damage is located at the wrong level, a non-injured party will receive compensation while the injured party will be left empty-handed. Matters worsen if, in an action brought by the direct purchaser, the infringer successfully claims a passing-on of the damage and the indirect purchaser does not succeed in showing that the damage has been passed on. Then, despite the fact that harm was caused and damages have certainly occurred, no one receives compensation and the infringing party gets off “scot-free”. For this reason alone, the introduction of elements of collective redress should be considered.

The current system also fails in the case of scattered damage: end consumers, who ultimately suffer damage, receive no compensation. At the same time, either the infringer can successfully defend itself against claims by its direct purchasers by invoking the passing-on defence (and thus keep the illegal profit), or the direct purchasers receive compensation, even though they have passed on the damage to the end consumers.

Consequently, collective redress must be made possible. If this cannot be realized at the European level, it should be considered at the national one. Time to introduce class actions!

Bundling of claims

Unless the German legislator can bring itself to introduce the long overdue class action (–and not only in antitrust law, just think of the diesel scandal!) it should be clarified under which conditions the bundling of claims is permissible. Claimants who have bundled their claims through the assignment of claims in order to be able to develop sufficient clout for the enforcement of their claims in the first place have often experienced nasty surprises: the assignments have been deemed invalid and void due to, among other things, a violation of the Legal Services Act (see only LG München I, judgment dated February 7, 2020, 37 O 18934/17, NZKart 2020, 145). Although this was intended to protect the assignors, they were given cold comfort. On the one hand, the assignors often were not in a position or it did not make economic sense for them to enforce their claims individually. On the other hand, by the time a decision was made on the admissibility of the assignment, the statute of limitations had often expired. The situation bordered on a complete refusal of access to justice. Although the Federal Court of Justice (BGH) made it clear in its AirDeal decision that the bundling of claims does not violate Section 4 RDG, it is not quite certain whether this also applies in the context of cartel damages.

Overall, the legislator should clarify the conditions under which the bundling of claims and the use of claims vehicles (companies that act as assignees in the context of bundling of claims) is permissible. At least as regards antitrust law, the requirements should not be too high. Otherwise, there would be no effective possibility of enforcing antitrust damages claims, despite such a possibility being required by primary law and the Cartel Damages Directive. A clarification in the Cartel Damages Directive may help here. In any case, even without a clarification in the directive, the admissibility should be clarified in national law.

Concentration of jurisdiction

In principle, the general rules on jurisdiction apply to cartel damages proceedings. Section 89 German Act Against Restraints of Competition (GWB) allows for a certain concentration of jurisdiction, but proceedings resulting from large cartels, such as the rail cartel or the truck cartel, are still conducted before many different courts. As a result, there are considerable inefficiencies: Not all courts have the same level of experience with antitrust law. Not all courts are adequately equipped and staffed to swiftly process and conclude the extensive antitrust proceedings, which also frequently require the involvement of economic expertise. The distribution of cases among different courts also means that new judges are constantly having to familiarize themselves with the proceedings. 

Against this background, it makes a lot of sense to think about concentrating jurisdiction. The best solution would be to concentrate jurisdiction at a single court in Germany. Section 39a (5) Securities Acquisition and Takeover Act (WpÜG) can serve as a model in this respect. Therefore, hardly any arguments can be raised against concentrating all these cases at a single court in Germany, e.g. in Düsseldorf. Thus, at least the possibilities Section 89 GWB provides must be exhausted: concentration of cases at one designated court should be carried out in each federal state.

 It also seems conceivable that jurisdiction could be concentrated in such a way that proceedings for damages arising from certain cartels would have to be heard by the court first concerned with the relevant cartel. Proposals in this respect have been made (Kersting/Preuß, Umsetzung der Kartellschadensersatzrichtlinie (2014/104/EU), 2015, p. 34 f., para. 278 ff.). Here, too, it is true that a European provision can be helpful. Even though a European provision is not imperative, it can help to overcome resistance in national law. In any case, there is a need for change in German law.

Presumption of damages

Article 17 (2) of the Cartel Damages Directive contains a rebuttable presumption of damages. It is (merely) presumed “that cartel infringements cause harm”. German law implements this in Section 33a (2) sentence 1 GWB. Ultimately, this presumption proves to be largely toothless, as the injured parties still have to prove the actual amount of damage at great expense in terms of time and money. The case law has yet to succeed in translating the statutory presumption that the damage is in any case greater than zero into manageable and predictable specifications (on this mandate to the judicature, see Kersting/Preuß, Umsetzung der Kartellschadensersatzrichtlinie (2014/104/EU), 2015, para. 58 ff.). Courageous approaches by courts to estimate the amount of damages by using a wide discretion have still remained isolated and are subject to narrow conditions (see LG Dortmund, 30.9.2020, 8 O 115/14 (Kart); see also here). Real progress could be achieved if a genuine presumption of the amount of damages were included in the Directive or at least in German national law. One could think of a presumption that cartels cause an overcharge in the amount of 10%. This would correspond to the provisions in Hungary and Latvia. Romania even presumes damages of 20% (see here). This would significantly facilitate and accelerate private enforcement in Germany. It would also encourage out-of-court dispute resolution. 

As an aside, it should be clarified in the context of a reform of the Directive and/or German law that the presumption of a passing-on of an overcharge (Art. 14(2) of the Directive), which intervenes in favor of an indirect purchaser, includes the presumption that the overcharge has been passed on in full. This is wrongly disputed in German law and inadequately implemented in Section 33c (2) GWB as a presumption only that harm has been caused.

Better protection for leniency witnesses

Cartelists face heavy fines. To encourage them to uncover cartels, cartelists who make themselves available as crown witnesses and uncover a cartel will receive immunity and have their fine reduced to zero. This makes cartels unstable and facilitates the work of competition authorities. A large number of cartels have been uncovered in the past through leniency applications. However, immunity recipients still face high damages claims, which can lower the willingness to file a leniency application (see the ECJ in Pfleiderer [there para. 26 f.] and DonauChemie [there para. 33]). The Cartel Damages Directive has addressed this question by providing for a privilege for immunity recipients. According to its Art. 11(4) immunity recipients are only liable to their direct or indirect purchasers or providers. However, this privilege has not been considered sufficient for some time as leniency applications have declined in numbers (see also here). The German Federal Cartel Office (BKartA) in particular is calling for better protection for immunity recipients (see here).

However, the directive is already misguided in its approach as it protects immunity recipients at the expense of the injured parties. The correct approach is to protect the immunity recipients not externally with regard to the injured parties, but internally against the other cartel participants. It is not the injured parties who are responsible for uncovering the cartel, but the cartel participants! Not the injured parties should bear the costs of uncovering cartels, but the cartelists! 

Such an internal privileging of the immunity recipient could be constructed in such a way that, on the one hand, the immunity recipients are jointly and severally liable to the injured parties for full damages. On the other hand, they could be given a larger claim for contribution against their co-infringers than they would normally have. This internal claim for contribution can be structured in such a way that the immunity recipients receive full compensation from the other cartel participants for any compensation paid by them to injured parties. This would mean that economically they are not burdened at all by the claims for damages. In this respect, the other cartel participants could even be held jointly and severally liable for the immunity recipient’s contribution claim, contrary to the general rules. It is also conceivable to only grant a contribution claim that is not a full contribution claim, but is reduced by a certain percentage. One could even privilege those cartelists who are not immunity recipients but who have received a reduction in fines. Relevant proposals have been on the table for a long time (cf. Kersting, FS Meier-Beck, GRUR 2021, 250 ff., see also here).

It should also be briefly mentioned at this point that the current privileged treatment of immunity recipients in their external relations with the injured parties poses considerable problems in terms of the right to legal protection. After all, granting immunity deprives some injured parties of an existing claim. Therefore, granting immunity means that the legal position of third parties is impaired. They may have to be heard beforehand, but in any case must be granted legal protection, i.e. the opportunity to challenge the immunity decision (cf. Kersting, FS Meier-Beck, GRUR 2021, 250, 251 ff.). These problems do not arise if leniency is granted not in the external relationship with the injured parties but in the internal relationship with the co-infringers. Such a paradigm shift – also required by primary law – away from granting a privilege in the external relationship at the expense of the injured parties and towards granting a privilege in the internal relationship at the expense of the co-infringers who caused the harm will, however, actually require a reform of Art. 11 (4) of the Cartel Damages Directive. 

As long as there is no such reform, it should at least be considered to give undertakings a choice under German law. In this case, the immunity recipient should be granted the right to choose between an external privilege vis-à-vis the injured parties as required by the Directive and a more extensive internal privilege vis-à-vis its co-infringers, i.e. the other cartelists. It can be assumed that the immunity recipient will prefer the full privilege in the internal relationship to only a partial privilege in the external relationship as provided for in Art. 11 (4) of the Directive. Proposals for this have been available for quite a while (Kersting/Preuß, Umsetzung der Kartellschadensersatzrichtlinie (2014/104/EU), 2015, pp. 24, 81 ff.) and would be compatible with the requirements of the directive.

The big picture: coordination of public and private enforcement, of sanctioning and disgorgements of benefits

A major problem of antitrust enforcement is that public and private antitrust enforcement are not sufficiently coordinated. The punishment of the offense and the disgorgement of benefits or the compensation of damages do not work together without friction. Section 81d (3) sentence 1 GWB allows the economic advantage derived from the cartel to be disgorged via the assessment of fines. At the same time, Section 34 (1) GWB permits a disgorgement by administrative means, provided that the surplus has not already been disgorged by way of damages, the imposition of a fine, an order to confiscate the proceeds of the offence or by way of restitution, Section 34 (2) first sentence GWB. In theory, one of these ways of disgorgement must be chosen, but in practice this does not work: 

Although the infringement-related turnover and thus the infringement’s potential to generate profits is taken into account when calculating the fine, the Federal Cartel Office declares its fines to be purely punitive fines in all cases. Nevertheless, there is no subsequent disgorgement pursuant to Section 34 GWB, even though such a disgorgement is mandatory. The fact that the fine is purely punitive without any element of disgorgement should actually lead to lower fines. Yet, this is not taken into account when determining the amount of the fine. This disregards the mandatory requirement of Section 81d (3) sentence 2 GWB. 

The fact that the imposition of fines regularly precedes the assertion of claims for damages further complicates the situation. Although the damages caused by the cartel are not identical to the benefits derived from the cartel, there is an overlap. The benefits derived from the cartel regularly correspond to damages incurred by the opposite side of the market, with the damages possibly exceeding the benefits. If claims for damages are then brought, this must be taken into account. The compensation of damages needs to have an effect on an earlier disgorgement insofar as the damages correspond to benefits derived from the cartel. There is no statutory provision for this, but the correct view is that Section 34 (2) sentence 2 GWB is to be applied by analogy and that reimbursement is to be made to the cartel participants. It is also necessary to coordinate this with confiscation under criminal law (Section 73b Criminal Code (StGB)). Tax issues have also not been solved satisfactorily and in compliance with constitutional requirements. There is a risk of unconstitutional double taxation of cartel participants if they pay taxes on their illegal profits but no account is taken of the (hidden) disgorgement of illegal profits. In this regard see Drüen/Kersting, Steuerrechtliche Abzugsfähigkeit von Kartellgeldbußen des Bundeskartellamts, 2016, p. 76 ff.

Finding a solution is proving to be a Herculean task and requires a major overhaul of the system. A deterrent fine must be imposed, the economic advantage derived from the cartel must be fully disgorged, the injured parties must be fully compensated, and everything must be predictable ex ante for the cartelists. Obvious considerations of assigning a central position to the Federal Cartel Office must be reconciled with the right of injured parties to compensation. Here, no comprehensive and convincing solution can be developed, but some cautious initial thoughts can be presented: 

As a starting point, it does indeed seem sensible to assign a central role to the Federal Cartel Office. As a rule, actions for damages are always brought as follow-on actions, i.e., following the imposition of a fine by the Federal Cartel Office. The Federal Cartel Office has far more comprehensive possibilities to gather information at its disposal than the injured parties. It is therefore in the best position to determine the economic advantage and thus the lion’s share of the damage suffered by the injured parties. In practice, the central dispute about the extent of the damages incurred from the price overcharge could thus be resolved in a single proceeding before the Federal Cartel Office. If the sequence is then reversed and the Office first disgorges the economic advantage and only then imposes the fine, the separation between punishment and disgorgement is clear and also permits tax treatment in line with constitutional requirements. Damages already paid beforehand could be taken into account when determining the amount to be disgorged. 

Subsequently, injured parties can take civil action against the cartelists. In doing so, they can refer to the determination of the economic advantage by the Cartel Office, which reflects the overcharge. To the extent that compensation payments are subsequently made by the cartelists, they will receive reimbursement from the money disgorged from them. One could consider limiting these claims for reimbursement to the amount of damages and excluding ancillary claims, in particular claims for payment of interest. This could lead to an acceleration of the proceedings.

This approach is also likely to work for in-court and out-of-court settlements; if necessary, a summary examination of the justification of the claims by the Federal Cartel Office can be made a prerequisite for the reimbursement claim. The possibility of stand-alone actions would still exist. If the Federal Cartel Office becomes active at a later date, it should be possible to suspend these lawsuits until the Authority reaches a decision. 

More comprehensive solutions are also conceivable, based, for example, on the stock corporation law appraisal proceedings (see Klumpe, NZKart 2019, 405, 406) or on the idea of merging public and private enforcement. This would require coordination with the requirements of European law. However, such coordination is necessary anyway, because ideally a new solution would not be limited to proceedings of the German Federal Cartel Office but would also include proceedings of the European Commission. However, this would require an overall solution at the European level. If a comprehensive, uniform solution is sought there, it may also be possible to justify the compatibility of a comprehensive solution with primary law. 

In summary: think big!


There is a need for reform in the area of antitrust enforcement. In addition to the points mentioned here, one might also think of abolishing the non-sensical privileging of SMEs and of reviewing the disclosure provisions in Sections 33g, 89b GWB, which should also be subjected to an evaluation; for reasons of space, this cannot be addressed in this article. In any case, the Directive should oblige Member States to introduce class actions and provide for better protection of immunity recipients. Such better protection of immunity recipients can be achieved by granting them a comprehensive internal compensation claim against the other cartelists.

Prof. Dr. Christian Kersting, LL.M. (Yale) is director of the Institute for Cartel Law at Heinrich-Heine-University Düsseldorf.

Leave a Reply

Your email address will not be published. Required fields are marked *