A new dawn for collective actions in Germany? Federal Court of Justice in Germany sets new boundaries for bundled antitrust claims in Financialright

On 12 May 2026, the German Federal Court of Justice (“FCJ”) handed down a groundbreaking judgment which has the potential to be a competition law decision named in the same row with, e.g. for enthusiasts of European law, van Gend & Loos, Dassonville or for European Corporate Lawyers with Centros, Überseering and Inspire Art. The FCJ drew the line this case on (i) how to structure and (ii) bundle various claims in one case (case no. KZR 6/24 – Sammelklage-Inkasso). 

The assignment model as a response to the lack of an opt-out regime

The German legislator has decided not to introduce a full opt-out class action regime, in contrast to the UK, Portugal or the Netherlands. The German regime, the so-called “Abhilfeklage” is not applicable for B2B cases, provides only an opt-in mechanism for B2C claims and small companies. In addition, for international funders, it is not super attractive, as a success fee of up to 10% can be agreed (Sec 4 para 2 No. 3 VDuG).

Having said that, another option was developed: The Assignment model: Multiple claimants assign their damages claims to a claim’s vehicle, an SPV, which is registered according to the German Legal Services Act (“RDG”), which enforces them jointly with the backing of third-party funding on a contingency basis.

Abuse of process

As a rule of thumb, the FCJ confirmed that cartel damages claims may generally be bundled and pursued collectively by the SPV, in other words: The Assignment model is valid. And funders can use this model to generate turnover (para 41). But the FCJ drew a line when the point of no return is reached re the applicability of the Assignment model, and this line is the “abuse of process” (para 25). This is basically the case where bundling makes it practically impossible for courts to grant effective legal protection (wirkungsvollen gerichtlichen Rechtsschutz). Three dimensions are important in this regard: First, the size of the claim (claimants / claims). Second, the heterogeneity of the claims and third, the way of presenting the claim to the court. In other words: An abuse of process exists, when a single court chamber cannot properly adjudicate the claims within the framework of the procedural rules and the typical time constraints.

In para 53 of the judgment, the FCJ then sets out the following calculation, which had already been explained when the judgment was delivered in person in Karlsruhe: It was assumed that a judge on a chamber (at the Regional Court) requires one hour of processing time per truck (taking into account that the claims are often governed by foreign law); at 1,800 hours per year, a judge would therefore need 38 years to deal with the Financialright case. In my view, this comparison is flawed and also questionable in substance, given today’s technical capabilities. With a license for an AI tool, a judge needs only a fraction of that time if, for example, they upload all the invoices to Legora and create a prompt to analyze all the documents in a structured manner. This puts the scale of a case into perspective to a considerable extent. In addition, help could be on the way from the ECJ. Following  requests for preliminary rulings from the Hoge Raad der Nederlanden (Cases Stichting Truck Cartel Recovery, C-425/25, and Stichting Cartel Compensation, C-426/25), the CJEU will decide whether a single cartel gives rise to the view that a single claim for damages. Furthermore, the case could also be handled by the sample analysis proposed by Prof. Dr. Klumpe from the Regional Court in Dortmund (Klumpe, WuW 2022, 596, 599 ff.) and OLG Stuttgart, Urteil vom 20. November 2025 - 2 U 263/21-, Rn. 58, juris.

Even though the FCJ does not expressly specify what is permissible in terms of size (although the court costs threshold of EUR 30 million may in any event be used, cf. para. 26, 41) or according to which criteria bundling must take place. From our point of view, it seems sensible to consolidate cases at least by country of procurement and by size (e.g. maximum of 2,000 claims if the ruling of the FCJ will be applied. Taking into account the AI aspect, the number can be much higher).

Countries of procurement make sense in that, ideally, the Max Planck Institute would only need to be commissioned once for the proceedings and the presiding (single?) judge could take particular account of the specific features of the relevant national law. This is particularly relevant given that the relevant documents in the individual countries are at least similar, if not identical (in certain European countries, for example, there is an invoice – sometimes issued in parallel with the lease agreement – confirming the lessee’s acquisition of the vehicle at the start of the lease term).

Finally, a grouping of up to 2,000 acquisition transactions also appears sensible. Given the estimated time of one hour per acquisition transaction and the annual working hours of around 1,800 (incidentally, in line with the billable hour targets of most big law firms), it is realistic to expect the claim to be processed within a year. In view of the claims already filed and those yet to be filed (particularly taking into account the fact that the claims are now likely all to be brought before the Mainz Regional Court), this appears necessary. Given that the limitation on such a number serves the purpose of ensuring that the proceedings can be concluded within a reasonable timeframe, consideration must be given not only to the sheer number of acquisition transactions but also to how difficult they are to review. In this respect, therefore, the limit of 2,000 does not represent an actual limit, but serves, firstly, merely as a guide and, secondly, has in mind average cases that actually require the one hour of processing time envisaged by the FCJ.

(No) conflict of interests

A lot was discussed after the hearing in Financialright re the various RDG topics. The FCJ has confirmed that most of the playballs of the defendants are now outside the play court, e.g. it was confirmed that no registration is needed with regard to claims from foreign jurisdictions in the meaning of Sec 10 para 1 No. 3 RDG (para 66). This makes sense, when German law is agreed e.g. in the Assignment and in the proceedings. The last playball in the ground is Sec 4 RDG, the conflict-of-interest provision. Under this provision, legal services that may have a direct influence on the fulfilment of another contractual obligation may not be provided if this jeopardizes the proper provision of the legal service.

Instructively, the Federal Court of Justice first notes that bundling does not per se lead to a conflict of interest in breach of Section 4 para 1 RDG, as the interests of the funder, the claimant and the assignor are all directed towards achieving the highest possible satisfaction (para 80). Such a conflict of interest is conceivable, however, where, as in the present case, claims relating to different market levels are asserted jointly, in which the existence of damage is already logically mutually exclusive (lessor and lessee, for example), or where particularly promising claims are bundled with less promising ones, making it considerably more difficult to reach a settlement.

In addition, the conflict of interests could arise from the funder’s interest in pursuing a specific litigation strategy. This in this case seems evident in the funding agreement. In this respect, the funder’s interest in pursuing the most profitable course of action and the assignors’ interest in the most effective enforcement may not run entirely parallel, as a maximally profitable approach could, for example, be achieved through a swift settlement (and thus typically lower legal fees), whereas a final judgment (including the award of interest) would be necessary for the maximally effective enforcement of the interests of individual assignors. This is particularly the case when claims with varying degrees of success are asserted jointly (in this respect, the proceeds from the more promising claims in both cases are likely to suffer from being bundled with less promising claims).

This appears reasonable in principle but neglects the specific features of the individual proceedings. As is known from the market, purchasers of trucks during the years affected by the cartel – and in some cases well beyond that period – receive far more than one offer to sell or assign their claims. In this respect, it is up to the truck owner to choose the offer that suits them best. It is therefore by no means impossible that interests were already aligned at this stage and that, consequently, there is no conflict of interest at all, but rather a full alignment of interests. For instance, some assignors are interested in quick proceedings (not least due to current fuel price shocks), whilst others are waiting for the highest possible proceeds, partly because compound interest continues to accrue.

Admittedly, according to the wording of the provision, the mere risk of a conflict is sufficient, so that an actual conflict of interest need not have materialized. However, this must also be the case in the specific instance, and the defendant, as the party raising this objection, bears the burden of proof and presentation (Grunewald/Römermann, BeckOK RDG, Sec 4, para. 26). In this regard, the FCJ refers in para. 80 to possibilities of influence going beyond mere rights to information and expressly mentions veto rights with regard to procedural acts but convincingly states that it depends on the individual case. This is not least because, in 2022, it further held that the existence of a power of attorney to conclude (revocable) settlements is not sufficient to assume a conflict of interest (BGH VIa ZR 418/21). If the assignor has in the end the contractual power to direct the proceedings this could lead to a conflict of interest.

If the funding contracts contains only information obligations between the Funder and the SPV and the assignor can decide by himself if we want to file an appeal or enter into a settlement, the corporate of the SPV does not play any role, e.g. if the funder is the direct shareholder of the SPV which concludes the agreements with the assignors. The only right, which cannot lead to a Conflict of interests, is the right to terminate the agreement, if the Assignor is rejecting to be part of a settlement (para. 85). He can then proceed at this own risk, but the funder should be able to terminate the agreement.

The interpretation should also take into account the fact that the European legislator has deliberately relied on private enforcement of competition law as a means of fostering competition and continues to do so (see para. 47). Restricting class actions by dismissing the claim on the basis of any (alleged) formal defects significantly curtails this. For, as the FCJ itself rightly notes, smaller businesses have little interest in enforcing their claims themselves (see para. 47). Insofar as the FCJ argues that there were also actions brought by individual businesses which, amongst other things, formed the basis of the LKW-Kartell I-VI proceedings, it must be objected that one cannot generalize from individual cases to the general ones. After all, every company is subject to different circumstances; in some cases, claims are asserted by insolvency administrators who are compelled to do so in the interests of creditors, or the sums in dispute are so large that there is no longer any lack of interest in pursuing them.

Takeaways and Outlook

The FCJ has defined the boundaries for mass claims in Germany based on the Assignment model. Not everything is crystal clear and must be defined by the lawyers and courts afterwards, e.g. what kind of bundles in mass claims should be made and how to divide the group.

The FCJ itself suggests a solution to at least the most recently discussed problem: statutory provisions for (mass) class-action lawsuits, in addition to the provision of sufficient personnel and other resources (para. 55). This would, on the one hand, pool expertise (handling a class action, particularly in antitrust law, requires entirely different skills than an appeal proceeding does) and, on the other hand, create clear lines of responsibility. Furthermore, the BGH states that the division of the proceedings into individual claim groups may also be left to the plaintiff, so that in the future as well, a large class action may be filed and only subsequently divided according to the rules imposed by the court following the court’s instruction (para. 30).

Bundling, however, is here to stay, even with a slightly different role of the funder, depending on the decision of the OLG München that is now in the driving seat again.

* The authors, Dr. Lars Maritzen and Henrik Oltersdorf (both Schalast, Düsseldorf), were not involved in the proceedings that led to the FCJ-judgment which is discussed in this blogpost.