The bundling of claims in cartel damages litigation – Germany v. Netherlands v. UK v. Italy v. France

When pursuing damages in so called (follow on) cartel damages claims claimants can anticipate fierce resistance from defendants. The resistance is caused by the sheer volume of claims in these kind of cases. Collective claims can mount to several billions of euros (see for instance the claims asserted in Air Cargo[1] and Trucks[2]), sometimes even hundreds of billions, thus triggering defendants to counter with as many arguments (im)possible. Normally, the only way for claimants to survive this legal battle is to combine forces. Let’s tackle the defendants together. This is what is called the bundling of claims. Often litigation is impossible without the bundling of claims. One might argue that jurisdictions that forbid this bundling of claims are in breach of the doctrine of the useful effect. The useful effect doctrine constitutes a specific branch of the EU state action doctrine that serves to prevent Member States from enacting state measures that enable undertakings to escape antitrust accountability.[3] In other words damaged parties must be able to pursue compensation of their damages. We have seen some recent case law of the European Court of Justice (ECJ) in which it applied the useful effect doctrine to the benefit of claimants.[4] We did a comparative law research on the possibility of working together as claimants, thus the bundling of claims in various European jurisdictions.

Germany

Contributed by Konstantin Seifert, Oppolzer | Seifert Kartellrecht

In Germany, there is no specific regulation concerning the bundling of claims in cartel damage cases. As a result, there are no clear rules allowing or prohibiting the purchase or assignment of such claims and there are also no provisions allowing for a genuine opt-out antitrust class action. Thus, claimants (or “class action” organizers) that want to enforce damage claims of various injured parties in one and the same legal proceeding currently have to choose one of mainly three options under “regular” German civil law, civil procedure and – in one case – the general regulations concerning legal service providers:

Option 1 is going to court as a “joinder of parties” (“Streitgenossenschaft”), which means that every and all injured parties participating in the proceedings will be claimants next to each other. While this is undoubtedly permissible, the group of claimants (and the organizer/litigation funder behind them) face two challenges: The first one is that being a party to the proceedings every individual claimant can – in theory – take an active role in the trial, thus undermining the intended efficiencies of the “class” proceedings and potentially breaking up the united front of the claimants. Therefore, an elaborate contractual design may be needed in order to keep everyone aligned without overly imposing on the individual claimants (and risking an invalidity of the entire construct). The second challenge is that the court might simply decide to break up the proceedings into smaller pieces (up to one case per claimant), so that the claimants (and the claim organizer/funder) end up with dozens or hundreds of individual proceedings (each with their own – in Germany quite substantial – costs and cost risks). This has not been tested much, so the risk of a split-up is hard to gauge. However, the risk appears to be lower for smaller (still manageable) or very large groups of claimants (which would cause thousands of separate trials if split up, thus swamping the court) than for a mid-size group of claimants.

Option 2 is the assignment model, in which the injured parties assign their claims to a claim vehicle which then goes to court as sole claimant. The injured parties in turn bear no cost risk whatsoever and receive the lion’s share of the recovery (usually around 70%) whereas the rest goes to the claim vehicle/funder. While this is a very practical construct for all parties involved, it has had some teething problems and has – initially- been met with opposition on district court level. Multiple courts have found the assignments (i.e. the transfer of the claims to the claim vehicle) to be invalid due to either (i) insufficient financial resources of the claim vehicle or (ii) a violation of Germany’s Legal Services Act (“Rechtsdienstleistungsgesetz”). As a consequence, the trials inevitably ended in a fiasco (because the claim vehicle never possessed any claims it could enforce in court).

The first issue, however, practically arises only in special circumstances and can also be solved quite simply legally (but not cheaply). There has only been one landmark case in which the courts (in Düsseldorf) have held the assignments to be invalid due to insufficient funding of the claim vehicle.[5] That case was particular in that the claim vehicle itself had publicly communicated its sparse funding before the assignments were made so that the injured parties (i.e. the assignors) were aware or at least had to have known about the lack of funding, which under law is a prerequisite for the invalidity of the assignments.[6] Naturally, this will very rarely happen in other cases where clients simply have no reason to doubt the claim vehicles’ funding. Also, any (residual) risk can be excluded by providing sufficient financial resources to the claim vehicle to cover all costs reasonably expected in the proceedings (including, in particular, any potential cost reimbursements for defendant’s counsel). Because then, the defendant(s) would not be exposed to the undue risk of being unable to recover its costs for legal defense in case it won the case. As stated above this is simple in a practical sense, but it might be costly, because Germany has embedded the so-called loser pay principle in its procedural law, so counsels’ statutory fees and court fees might mount up to appr. 2 million euros. However, based on the Düsseldorf decisions, one can assume that taking out ATE (“after-the-event”) insurance (covering any adverse costs if the proceedings are lost) will also be sufficient and much more economical than depositing funds for all potential adverse costs in the claim vehicle’s bank account.

The second issue is that the German Legal Services Act (“Rechtsdienstleistungsgesetz”) considers the enforcement of claims for the account of a third party to be a legal service, which may only be rendered by companies with a debt collection license. It is not impossible to obtain such a license, but even then, several district courts had held that the license would not cover a business model directly aimed at legal proceedings in court (as opposed to out-of-court).[7] This, however, has recently been overturned by a landmark ruling of the Federal Supreme Court.[8] The FSC has also further clarified another contested issue, which is the conflict of interests provision in the Legal Services Act. Some district courts saw a violation of that provision in the fact that (i) the interests of different (groups of) injured parties might not be aligned (so the claim vehicle or the settlement could favor some injured parties at the expense of others) and (ii) the interests of the injured parties on the one hand and the claim vehicle on the other hand might also not be aligned (since the claim vehicle bears all cost risks and thus might be inclined to accept a settlement sooner than the injured parties would for themselves).[9] The FSC, by contrast, held that when designed properly (and particularly, if the claims are sufficiently homogeneous), these issues could be overcome.[10] This should also hold true with regard to a potential conflict of interests between the injured parties and a third party funder (providing the resources to the claim vehicle), at least if the funder’s influence in the proceedings (i.e. on the claim vehicle) is limited (and/or if any settlement can be revoked by the injured parties). The latter point (influence of the funder) has even been picked up by the legislator who has adopted a corresponding change in the relevant provision (§ 4, 1) of the Legal Service Act, which entered into force on 1 October 2021. There are several appeal proceedings pending in those cases, where a violation of the Legal Services Act had been ruled on District Court level. In those proceedings, a number of higher regional courts (in Munich,[11] Stuttgart,[12] Schleswig,[13] and Braunschweig[14])– even though they have not handed down final judgments – have issued orders and/or given opinions in oral hearings that strongly indicate they will overturn the first instance decisions and restore the validity of the assignments.

Given these developments, we can expect to see a renaissance of the assignment model in Germany. The extent of it will also depend on the outcome of an appeal case in the trucks cartel pending at the higher regional court of Munich[15] which is much anticipated (and which will probably go up to the FSC to finally settle the matter and provide even more guidance for future permissible designs of the assignment model).

Finally, it should be borne in mind that the German Legal Services Act will not apply to foreign injured parties assigning their claims to a foreign claim vehicle which then brings the case to court in Germany (against, for example, a German defendant).

Option 3 is the purchase of claims, which then are also assigned to a claim vehicle. Since there is a fix purchase price and therefore the injured parties have no further interest in what becomes of “their” claims, the claim vehicle then goes to court entirely for its own benefit. Therefore, there is no question of rendering a legal service to someone else and consequently no issue with the Legal Services Act. However, there are practical disadvantages: the claim vehicle/funder requires significantly more financing in order to be able to purchase the injured parties’ claims up-front – assuming even more risk than in the assignment model (Option 2) or in the financing of a joinder of parties (Option 1). Inversely, the injured parties usually get only a fraction of what they would have received had they chosen for option 2 with a favorable outcome.

The Netherlands

Dutch civil procedural law allows special claim vehicles to act as a plaintiff in proceedings. Dutch law provides the following options to enable a special purpose vehicle (SPV) to engage in proceedings.

Option 1 is that the injured parties can assign their claims to the SPV, which subsequently litigates these claims. The injured parties and the SPV can also enter into a contract of mandate which will entitle the SPV to litigate the concerned claim. The SPV can subsequently litigate these claims either in its own name, or in the name of the injured parties. This is what we call the “opt-in route”. Any participating party has to actively decide for themselves to join this collective action.

In various judgments in cartel follow-on cases in connection to the Air Cargo cartel, the Amsterdam District Court and Court of Appeal held that the assignment by individually injured parties of claims to a claim vehicle is in principle valid under Dutch law.

The burden of proof regarding the legal validity of the assignment of the claims lies with the claim vehicle, the Amsterdam District Court ruled. In a damages action initiated by claim vehicle Stichting Cartel Compensation (SCC), the court clarified that claimants will fulfil their obligation to furnish facts by submitting an extract of the deed of assignment and the title:[16]

“4.14. The District Court starts from the premise that if the documentation brought into the proceedings contains per shipper: (i) the assignment agreement (title) and (ii) the deed of assignment and (iii) it is clear that these were signed/issued by the assignor, it is sufficiently established that SCC is the party entitled to the claims, unless there are concrete indications, to be put forward by the airlines, that a legally valid assignment has not taken place in spite of this.”

“The court assumes that if the documentation per shipper submitted to the proceedings contains: (i) the assignment agreement (title) and (ii) the deed of assignment, and (iii) that it is clear that these have been signed/furnished by the assignor, then it is established to a sufficient degree that SCC is the entitled party to the claims, unless there is concrete evidence, to be submitted by the airline companies, showing that nonetheless, no legally valid assignment has taken place. It is important in this respect that the assigned debtor (and the court) can establish on the basis of the documentation that the assignor and assignee did actually intend to assign the claim In this case, the Amsterdam court held (on the basis of the assignment documents provided by SCC) that the assigned claims were described in a sufficiently clear and precise manner:[17]

“4.21. The court finds that it is sufficiently clear from the aforesaid assignment documentation and bailiff’s notifications that this concerns claims from the shippers for compensation for all damages resulting from the cartel, including overcharge, interest, lost profit and costs. […] It has also been taken into account that this concerns claims arising from tort by virtue of a cartel (the size and duration of which was apparent to the cartel members but not to the shippers). The position of the airline companies basically boils down to the fact that the shippers did not wish to assign their entire claim for damages arising from the cartel but excluded parts of it which, in the absence of evidence to support this, seems to the court to be implausible. It is furthermore clear that this concerns claims from all the members of the cartel referred to in the decision (as debtors of the claims). The deeds of assignment (due in part to the reference to the assignment agreements) accordingly contain sufficient details to be able to determine which claims are concerned. Contrary to what the airline companies have argued, it is not necessary for the determinability of the assigned claims, namely the claims for damages that are based on tort (participation in the cartel), that it can be established (now already) which shippers purchased which flights (which routes).

4.32. Based on all of the preceding, the conclusion is that the assignments that are governed by Dutch law are legally valid. This implies that the defence of the airline companies that the litigation assignments are not valid does not need to be discussed.”

This criterion was applied in another likewise judgement of the District Court of Amsterdam[18] and later confirmed in appeal, by the Amsterdam Court of Appeal (2020);[19] if the claim vehicle has fulfilled its above mentioned burden of proof, it is then up to the airlines to argue, in the context of substantiating their defence, how and why the validity of the assignment should be questioned on reasonable grounds, the appeal court ruled. The appeal court added that the debtors had a limited right to information and that they could not claim additional documents, referring to the parliamentary history of Article 3:94(4) DCC, which provision provides that the person against whom the debt-claim is to be exercised may demand that a written summary of the notarial or private deed and of its legal basis are handed over to them.[20]

The appeal court further added that at (this stage of) these proceedings it is, under Dutch law, not necessary to establish whether or not the assignment of the claims was validly made, since that is not necessarily important in the relationship between assignee (claim vehicle) and debtor (cartelists). The important thing is whether the debtor has to accept the effectiveness of the assignment against him. If the debtor subsequently were to pay to a party who was not authorised to receive the payment, the debtor can object to the party to which the payment was to be made that he paid in full, if he reasonably assumed that the payment was made to the recipient.[21]

This means that cartelists’ defenses in cartel cases, seeking access to all the underlying documents, will thus be brushed aside at first instance. The documents in question may become relevant if there are reasonable grounds to doubt the validity of the assignment, but such discussions often focus on one or a few assignment agreements rather than on all assignment agreements submitted by the claimant. Overall, the abovementioned judgements have substantially reduced the possibilities of cartelists to call into question the validity of assignments when governed by Dutch law.

Option 2 is that the SPV can bring a so-called “collective action” on the basis of article 3:305(a) of the Dutch Civil Code (DCC), either old legislation, or new legislation, depending on whether the events that are subject to the action occurred prior or after 15 November 2016. Article 3:305(a) DCC, old legislation, enables the SPV to demand declaratory relief with regard to liability and causal relationship, for the benefit of groups of injured parties as far as their claims are sufficiently similar and insofar as the claim vehicle promotes these interests pursuant to its articles of association. The options under the old regime are limited to declaratory decisions only, however, these collective actions can provide the momentum necessary to force the injuring party to accept a collective settlement. A SPV can commence a collective action under Article 3:305(a) DCC (old) without the cooperation of the injured parties, but is subject to other limitations. This is the “opt-out” route. All injured parties are included, when finally a settlement has been reached, parties can opt out of this settlement and pursue their own goal. More strict rules with regards to corporate governance apply in the “opt-out” system, naturally because damaged parties can be drawn into this kind of litigation without prior consent. The old Article 3:305(a) DCC only remains available when the relevant events took place before 15 November 2016.

Recently the Dutch legislator updated the collective action regime, with the amendment of the Act on the Resolution of Mass Claims in Collective Action (Wet afwikkeling massaschade in collectieve actie, “WAMCA”),^[22] and has introduced a mechanism to claim payment of damages in collective actions on behalf of the injured parties, as well as a lead plaintiff system. The legislator further added a number of additional safeguards and requirements for claim vehicles to constitute a viable representative claim as is required under Article 3:305(a) DCC to ensure the standing of the claim vehicle. For example, the SPV has to be sufficiently representative, has to have sufficient experience and expertise to commence and conduct the action and has to have a supervisory body. The merits of a case will only be assessed after the court has established that the claim is admissible under Article 3:305(a) DCC and the plaintiff has made it sufficiently plausible that pursuing the collective action is more efficient than individual claims (Article 1018(c) par 5 sub b Dutch Code of Civil Procedure).

If it has been established that a claim is admissible and successful, the court is also allowed to determine the amount of damages and the way in which the damages will be paid.[23] As the WAMCA rather new, there is no case law yet that gives guidance on how the courts will in practice deal with these competencies as there are no cases yet that have reached this phase.

The “opt-out route” is also an option for collective actions for damages under the WAMCA. If the collective action is on an opt-out basis, a court decision granting or dismissing the collective action will be binding on all injured parties who are member of the class, who reside in the Netherlands and who did not opt out.[24] The court decision will also be binding on members who reside abroad but in principle only if these parties opt in within a time period to be set by the court after the lead plaintiff has been appointed and announced, although the court is allowed to rule otherwise at the request of the plaintiff.[25] The WAMCA regime is limited to claims concerning events that took place after 15 November 2016.

Option 1 and 2 can be combined. Collective actions, option 2, can only be brought by a Dutch foundation or association and claims as described above under option 1 can also be brought by other vehicles than Dutch foundations and associations, provided the plaintiff’s law of incorporation empowers it to bring legal actions (article 10:119 (a) DCC).

The UK

In the UK, the ancient rules against “trafficking” of litigation: the law of ‘maintenance and champerty’, still has effects on the possibility to bundle claims.

Law of maintenance and champerty

Historically, English law did not recognize and enforce arrangements which qualified as ‘maintenance’ and/or ‘champerty’. ‘Maintenance’ entails the support of litigation in which the supporter has no legitimate concern without just cause or excuse. ‘Champerty’ is an aggravated form of maintenance in which the party who maintains the litigation funds the litigation and in return receives a share of the proceeds of a successful claim. The conclusion of such agreements was punishable under criminal law and constituted a tort.

Meanwhile, English law and UK courts have become more flexible and have adopted in principle a favorable perception towards the funding of litigation. Maintenance and champerty have been abolished as crimes and torts for a few decades, but the general rule has been left in place that a contract that breaches the rule against maintenance and champerty is considered to be contrary to be public policy and therefore unenforceable (section 14(2) of the CLA).

UK case law

Recent case law of the UK courts clarified that this rule does not necessarily invalidate a third party litigation funding agreement, unless there is some other element that is contrary to public policy. That might be for example, when the funder has undue control over litigation and/or when the funder receives a disproportionate share in the proceeds as opposed to the claimant(s).

The UK Courts have however shown to be less tolerant when claims are assigned to a third party which pursues the claim in its own name, as opposed to third party litigation funding. Assignment agreements may still be labeled as ‘champerty’ or ‘maintenance’ in particular when the assignee has no legitimate personal interest in pursuit of the assigned claims.

Leading cases regarding the assignment of claims have been for a long time Trendtex Trading Corp v Credit Suisse (1982)[26] and Jennifer Simpson (as assignee of Alan Catchpole) v Norfolk & Norwich University Hospital NHS Trust (2011)[27]:

• The Trendtex case established that the assignment of a bare right of action is considered to be against public policy where the assignee does not have a “sufficient interest” to justify pursuit of the proceedings for his own benefit. The court considered that the aim to profit from the litigation does not amount to a “sufficient interest”, and neither does the pursuit of litigation as part of a personal campaign.
• The court applied the principles from Trendtex also in the Simpson The cause of action assigned in this case concerned a cause of action in tort for personal injury. The claim was pursued by Mrs Simpson against the hospital where her husband had been a patient. While a patient at the hospital, her husband had contracted an infection called ‘MRSA’. Another patient at the hospital had also contracted the infection while treated in the hospital. The other patient assigned her claim to Mrs Simpson for the consideration of £1, and she pursued the claim in her own name and for her own benefit. The court considered that the assignment of a bare cause of action in tort for personal injury remains unlawful and void (para 24), stating that even though Mrs Simpson might have had ‘honorable motives’ in pursuing the claim, to demand attention for the (supposed) failing of the hospital, this was not the sort of interest the law recognized as “sufficient” as required by section 14(2) CLA. According to the court, the assignment in this case amounted to champerty, because it involved the purchase of a claim which, if it would be successful, would lead to Mrs Simpson recovering damages in respect of an injury she had not suffered. In the view of the court, that is an assignment of a bare right of action, in the sense the assignee has no legitimate interest, and is therefore void (para 28).

This case law makes clear that it is not so easy for a third party to pursue claims in its own name, when there’s no ‘legitimate’ personal interest and profit is the only goal.

There are however some recent cases which might suggest that there might be a development towards a more liberal approach by the UK courts with regards to the assignment of claims:

• In the case of JEB Recoveries LLP v Binstock (2015)[28], claims that were assigned to a special purpose vehicle ‘JEB Recoveries’, were accepted by the UK High Court. The original claim concerned was a debt due from defendant under a contract between the defendant an Mr Wilson. Mr Wilson assigned his claim to the claimant SPV, which was established by him and two others. The court held that although the assignment of a bare cause of action had long been recognized as champertous and that, without more, assignment of the claim of a nominal sum would be likely to offend public policy, in this particular case there was more: the assigned claims were not a bare right of action but included debts, and also a connection remained between the assignor and the SPV pursuing the claims: Mr. Wilson had a one-third interest in JEB Recoveries and also assisted in the pursuit of the claims. This was different in the Simpson case, in which Mrs Simpson only pursued the claim in order to pursue a separate (personal) campaign. The court also cited the court in Simpson, which had held that the law on maintenance and champerty is open to further development as perceptions of public interest change, with reference to, inter alita, new statutory regulation since the Simpson case allowing certain damages-based agreements.
• Two years later, in the case of Casehub Ltd v Wolf Cola Ltd (2017)[29], the UK High Court also allowed the bundling of claims by claim purchase agreements. The claims assigned were claims of customers against the defendant, who operated a software business, to be refunded the cancellation fees paid by them on the ground that the cancellation fee provisions in the terms and conditions of the defendant, were unlawful. The claims were assigned either in return for a percentage of the proceeds, or in return for a fixed amount. The court considered that in this case, the assigned claims did not amount to a bare cause of action because the assigned claims qualified as the right to the sum in question and the assignment of the right to bring a restitutionary claim to recover the sum, which would be incidental and subsidiary to that right (para 25).

Consequently, the question remained whether the claimant had a legitimate interest in the pursuit of the claims and whether there was a risk the integrity of the legal process would be impugned in some way (para 27). The court concluded that there were no public policy grounds which would lead to the conclusion that the assignment is invalid, to the contrary, the court held that there were strong public policy grounds in favour of upholding the agreement. In brief, because the assignments enhanced access to justice for the customers, while the court did not see a risk in this case of the litigation process being abused. The court also recognized that the claimant had a legitimate commercial interest in being able to pursue the claims assigned to it in order to protect the liquidates sums it acquired (para 28).

While these cases might suggest a shift in the UK courts perception towards the assignment of claims, it remains uncertain whether these cases will be upheld in all situations of assignment of claims. The acceptance by the courts of the assignment agreements in JEB Recoveries and Casehub was very case and fact specific, while the courts also made clear that the Simpson and Trendtex cases remain leading cases regarding this subject matter.

Therefore caution is still in order when it comes to the assignment of claims under UK law, as it will have far reaching consequences when a court will hold that an assignment is in breach of the law of maintenance and champerty: the assignment will be void and the claims cannot be pursued.

Italy

Contributed by Giovanni Scoccini, Scoccini & Associati

Italian law allows to bundle claims from different claimants in one lawsuit provided that these claims have the same causa petendi. This is the case with cartel damages claims, which stem from a single unlawful behaviour. Where the causa petendi is the same it is possible to file either a joint action under Article 103 of the code of civil procedure (c.c.p.) or a class action under the recently introduced Article 840 bis c.c.p

Option 1. The joinder of parties under Article 103 c.c.p is the traditional procedural instrument to realize economies of scale in the legal proceedings. It allows to bundle claims from claimants irrespective from where they are located (in Italy or abroad), provided that the lawsuit is filed with the court of the defendant’s domicile. On the other hand, if the lawsuit is filed with the court of the place where the harmful event occurred, the joinder of only the parties that are in the same jurisdictional district is possible. This could limit the use of the joinder of the parties if the defendant is not domiciled in Italy. However, the possibility that there are no defendants domiciled in Italy appears rare following the judgment of the Court of Justice in the Sumal case (C-882/19) in which judgement the Court established the liability of the local subsidiaries of the parent company which is the addressee of the infringement decision. Likewise in Germany, the Court may decide to break up the proceedings of the bundled claim either if it is requested by all the parties or if the joint management of the case may delay the proceedings or it may be too burdensome. In the truck cartel litigation, the court decided to break up the claims concerning the Scania vehicles from the other claims because the pending appeal of Scania against the Commission decision may result in a stay of the proceedings. The joinder of parties allows savings in the court fees and of the other costs of the proceedings.

Option 2. The new class action regime in Italy has been introduced by Law n. 31/2019 and is applicable to unlawful conducts that have taken place after November 19^th 2020. The new class action regime is open both to consumers and undertakings directly or through an association. The proceedings is divided in three phases: 1) admissibility of the action; 2) judgment on the merit of the case 3) in case of success, payment phase of the compensation to members.

The class action can be filed by a single plaintiff which can also be an association provided that it is enrolled in the list kept by Ministry of Justice. In the first phase the Court will decide on the admissibility of the action. The action shall be declared inadmissible if it is blatantly ungrounded, the claims are not homogeneous (i.e. different causa petendi), there is conflict of interest between the plaintiff and the defendant or the applicant does not have the resources to adequately pursue the claim.

If the Court established that the action is admissible, the interested parties that have homogeneous claims can join the action and the case will go to trial. The Court shall not apply the formal rules of the code of civil procedure. It can ascertain the liability of the defendants taking into account statistical data and simple presumptions. The cost of the technical assessment shall be paid temporarily  by the defendant.

In case of success the Court awards compensation to the plaintiff, that kicked off the class action, establishes which are the injured parties, opens the possibility again to new members to file an application to join the class action, appoints the judge in charge of the payment procedure and a representative of the members of the class action.

The defendant can file a reply to the applications of the members that can be assisted by their lawyers. After the assessment of the applications and of the replies of the defendant, the representative of the members shall draft a payment project to be submitted to the judge that will decide whether to grant the applications or not. The defendants shall pay the legal costs to the original plaintiff, the fees of the representative of the members and the legal costs of the single members. These costs can be significant.

The new class action regime shall be welcomed for its efficiency because it allows to file a pilot case easy to manage by both the court and the lawyer of the claimants and to postpone the heavy work of book building of members, the collection and the assessment of the documents only when and if the pilot case is successful. On the other hand, it can be very burdensome for the defendant that may be found guilty and condemned to pay compensation to an indefinite number of members following a judgment based on statistical data and simple presumptions. In case of defeat the legal costs for the defendant can be significant.

Another option available for bundling the claims is the assignment of the claims. The assignment of receivables is provided by article 1260 of the Italian civil code, and the Supreme Court made it clear that damage claims may be transferred like any other receivables. However, when assigning/purchasing claims under Italian law, there might be another spanner in the works in the form of a potentially required banking authorization pursuant to article 106 of the Consolidated Law on Banking (TUB). This provision may be applicable to (inter alia) the acquisition of cartel damages claims, and even though this is not a clear-and-cut case, this provision deserves consideration as the consequences of violating article 106 TUB might have far-reaching consequences.

Article 106 TUB provides that “the exercise towards the public of the activity of granting loans in any form is reserved for authorized financial intermediaries, registered in a special register held by the Bank of Italy.”.

The activities envisaged by Article 106 TUB require that they are carried out with some degree of professional manner towards third parties (“towards the public”)[30] and one of the examples of what is meant by the “activity of granting loans” is the ‘purchase of receivables for consideration’.[31] This means that receivables, which might cover cartel damages claims of cartel victims, when acquired in a professional manner, might constitute a reserved financial activity within the meaning of article 106 TUB.[32]

When no authorization has been granted, such qualification is a risk because violation of article 106 TUB may have far reaching consequences. Not only will the (unauthorized) transaction(s) be considered invalid, the ‘lender’ may risk criminal sanctions pursuant to Article 132 TUB, which provides for criminal sanctions ranging from pecuniary sanctions to imprisonment.

It is held in case law that for specific financial activities to be considered abusive and therefore criminal, it is necessary that the activity is professionally organized with methods and tools such as to foresee and allow the systematic granting of an indefinite number of loans, addressing a potentially vast number of people.[33]

Even though the scope of article 106 TUB and the abovementioned case law within the specific context of the acquisition of cartel damages claims is not entirely clear (yet), the provision deserves consideration given the potential consequences when it may turn out that the provision does apply to (certain) models by which claims are bundled; being invalidity of the (assigned) claims and possible criminal sanctions pursuant to Article 132 TUB.

France

Contributed by Marc Barennes, bureau Brandeis Paris

In France, there are no specific rules authorizing or prohibiting the bundling of claims in cartel damage cases. However, there are three main mechanisms allowing cartel victims to bring large damages claims.

Option 1 is the joint actions model, whereby cartel victims bring individual (separate) claims at the same time, before the same court, to which they request that these claims be dealt with jointly pursuant to Article 367 of the procedural civil code. Such actions will normally be dealt with jointly as the claims are connected and it is in the best interest of justice that they be decided together. The challenges plaintiffs will face are the same ones as those identified above for the option 1 in Germany. While there is no doubt that joint actions have in fact successfully been brought in other fields than competition law, none of these actions have been brought yet before the French courts in the specific area of cartel damages claims.

Option 2 is the assignment model, in which the injured parties assign their claims to a private entity which acts as claim vehicle. The claim vehicle, rather than the injured parties, brings the case in its own name. There are two potential scenarios. According to a first scenario, the claim vehicle buys the claim for a price which is paid once and for all at the time of the assignment. In such a case, the assignor does not have any financial interest in the outcome of the case which is brought by the assignee in its own name. According to a second scenario, the claim vehicle buys the claim for a price which is totally or partially paid once the damages are recovered only. In this second scenario, the assignor keeps a stake in the outcome of the case brought by the assignee. However, as the assignees have transferred their claims to the assignors who will bring the claim in its own name, they bear no cost nor any risk and normally receive the lion’s share of the recovery (usually around 70%) whereas the rest goes to the claim vehicle/funder.

The French courts have not yet had the opportunity to decide in a cartel damage claim whether either of these two types assignments are valid under French law (let alone EU law). However, pursuant to the principle of contractual freedom (“principe de liberté contractuelle”) and that claims may be validly transferred pursuant to Article 1321 of the Civil Code, no rule prevents such an assignment of cartel damages claims under French law.

Assignors should however be particularly careful in two regards. Firstly, to the extent that the second type of assignment described above could be considered as an activity carried out by a recovery agency pursuant to Articles R124-1 to R124-7 of the civil procedural execution code, the assignee should hold a specific authorization to recover these damages. Secondly, assignees should pay specific attention to the right provided for in Article 1699 of the civil code. Pursuant to this Article, in cases where an assignor transfers a « claim » which is disputed before a court to an assignee, the debtor of the claim may put an end to the dispute by paying to the assignee the price (plus fees and costs) the assignee paid to the assignor to acquire the claim (the so-called « droit de retrait »). In other words, a claim vehicle which would buy a cartel damage claim which was disputed by the debtor before a court may end up being entitled to claim only the price (plus fees and costs) it paid to the assignor for that claim, instead of the full value of the claim.

Option 3 is the fiduciary model (“fiduciaire”), which allows injured parties to assign their claims to a fiduciary entity which sole role is to recover their damages. Unlike the assignment model in which assignors transfer theirs claim to a private company acting as a claim vehicle, injured parties become constituents and beneficiaries of the fiduciary entity which brings the claim in its own name. If the fiduciary entity is funded by a litigation funder, all the costs and risks of bringing the claim may be borne by the fiduciary entity rather than the injured victims. While Articles 2011 to 2030 of the civil code provide for the conditions pursuant to which the fiduciary entity may act, the courts have not yet had the opportunity to decide a case in which a fiduciary entity claims cartel damages. There is however no reason to think that the fiduciary mechanism, which presents some similarities with a US “trust” or Dutch “Stichting”, is not particularly adapted to bringing large cartel damages claims.

Conclusion

All in all we come to the conclusion that in the various member states of the European Union there is a variety in the possibility to ascertain claims by bundling them. Since we feel that it is adamant for claimants to bundle their claims (otherwise they would effectively be excluded from the possibility to pursue damages at all) we are curious whether in any jurisdiction the argument of the useful effect doctrine will be accepted to facilitate the bundling of claims.

Hans Bousie (editor), with contributions from:

Marc Barennes,

Tessel Bossen,

Giovanni Scoccini and

Konstantin Seifert

[1] European Commission decision of 9 October 2010 and 17 March 2017 Case AT.39258 (Airfreight).

[2] European Commission decision of 19 July 2016 Case AT.39824 (Trucks).

[3] ECJ 16 November 1977 case C-13/77, ECLI:EU:C:1977:185 (INNO/ATAB).

[4] ECJ 14 March 2019 case C-724/17, ECLI:EU:C:2019:204 (Skanska); ECJ 28 March 2019 case C-637/17, ECLI:EU:C:2019:263 (Cogeco).

[5] District Court Düsseldorf 17 December 2013 file no 37 O 200/09 (Kart) U; Higher Regional Court Düsseldorf 18 February 2015 file no VI-U (Kart) 3/14.

[6] Higher Regional Court Düsseldorf (fn 3), at rec. 104 seq.

[7] See, e.g., District Court Hannover 4 May 2020 file no 18 O 50/16.

[8] Federal Supreme Court 13 July 2021 file no II ZR 84/20.

[9] See, e.g. District Court Munich 7 February 2020 file no 37 O 18934/17.

[10] Federal Supreme Court 13 July 2021 file no II ZR 84/20, at rec. 55.

[11] File no 21 U 5563/20.

[12] File no 5 U 173/21.

[13] File no 7 U 130/21.

[14] File no 8 U 40/21.

[15] File no 29 U 1319/20.

[16] District Court of Amsterdam 2 August 2017 ECLI:NL:RBAMS:2017:5512, para. 4.14.

[17] District Court of Amsterdam (fn 14), para 4.12 et seq.

[18] District Court of Amsterdam 13 September 2017 ECLI:NL:RBAMS:2017:6607.

[19] Court of Appeal of Amsterdam 10 March 2020 ECLI:NL:GHAMS:2020:714, para. 4.10.5.

[20] Court of Appeal of Amsterdam (fn 17), para. 4.10.2.

[21] Court of Appeal of Amsterdam (fn 17), para. 4.10.3.

[22] The new legislative proposal for the Act on the Resolution of Mass Claims in Collective Action (Wet afwikkeling massaschade in collectieve acties, “WAMCA”) was adopted by the House of Representatives on 29 January 2019 and entered into force on 1 January 2020.

[23] Article 1018(i) par 2 Dutch Code of Civil Procedure.

[24] Article 1018(f) par 1 Dutch Code of Civil Procedure.

[25] Article 1018(f) par 5 Dutch Code of Civil Procedure.

[26] Trendtex Trading Corp v Credit Suisse (1982) AC 679 HL. 

[27] Jennifer Simpson (as assignee of Alan Catchpole) v Norfolk & Norwich University Hospital NHS Trust [2011] EWCA Civ 1149.

[28] JEB Recoveries LLP v Binstock [2015] EWHC 1063 (Ch).

[29] Casehub Ltd v Wolf Cola Ltd [2017] EWHC 1169 (Ch).

[30] The decree of the Ministry of Economy and Finance of 2 April 2015 no. 53 (MD 53), Article 3.

[31] MD 53 (fn 21), Article 2

[32] see Court of Venice 13 February 2013 No. 316; Court of Venice 2 September 2014 No. 1758; GdP of Rom 18 July 2016 No. 24510; GdP of Prato 1 February 2016 No. 80.

[33] Italian Criminal Court Section V No. 18317/2016.

Of apples and oranges – the similarity test in Dutch collective actions

Dutch law allows for foundations and associations to file claims on behalf of a class of third parties if their position is ‘sufficiently similar’ and provided certain other criteria are met. The mechanism of collective redress is applied in the Netherlands in cases where substantial financial interests are at stake, such as in securities cases and cartel damages cases, in data breaches cases under GDPR, as well as fundamental rights cases. In all of these cases, the question comes up to what extent the interests of the represented class are sufficiently similar, such that litigating the matter on a collective basis is more efficient than litigating this on an individualized basis.

A question of what is ‘sufficiently similar’ is topical for several reasons:

• The Dutch courts have had to decide on this in a flurry of collective claims over the past 6 months.
• Internationally, and in particular in the UK, courts or at least litigants have struggled with the same matter resulting in relevant guidance for the Dutch practice.
• Collective actions are on the rise in Europe ^[1]Directive (EU) 2020/1828 of the European Parliament and of the council of 25 November 2020 on representative actions for the protection of the collective interests of consumers and repealing … Continue reading, and the question of which cases are, and which are not suitable for collective action will come into play in all European jurisdictions.
• Practically: failing to meet this test will result in a rejection of certification of the claim, and the claimant being found to have no standing. Considering that it may take a year or more of costly litigation, and even longer to prepare the case, not passing the certification test is a huge budgetary blow for the claimant. Claimants who are dependent on external funding will find that often time, adverse costs orders are not covered under the funding agreement. In particular ‘ad hoc’ claimants who have not accrued a war chest through previous cases may run a risk of bankruptcy as they are unlikely ^[2]This may be different of the class action vehicle has charged members of the represented class a fee for signing up to the class action. However, for various reasons, this is no longer common in the … Continue reading to have access to any other source of funding. By the same token, the willingness to settle on the part of the defendant may increase dramatically with certification.

A little background

As of 1 January 2020, the Dutch regime for collective actions was overhauled when WAMCA ^[3]Wet Afwikkeling Massaschade in Collectieve Actie, or WAMCA for short. entered into force. WAMCA applies to facts after 15 November 2016 and brought before the court after 1 January 2020. Unlike the former regime referred to as WCAM ^[4]Wet Collective Afwikkeling Massaschade, or WCAM for short. Under WCAM, it was not possible for a claim vehicle to claim specified damages. In liability cases, this meant that the vehicle would seek a … Continue reading, allows for the class action vehicle to claim specified damages on behalf of the represented class.

At the same time, the law was changed in the following respects, which are relevant to the topic of similarity:

1. Under WAMCA, the case will be litigated on the merits only after the court has confirmed that the claimant has standing to bring the case, and after the court will have defined what the claim will be exactly, and for the benefit of whom. In line with Anglo-Saxon practices, this confirmation is often referred to as ‘certification’.
2. Under WAMCA, damages can be claimed on an opt-out basis for, briefly put, residents of The Netherlands only. Others can choose to participate in the case, on an opt-in basis. This domestic scope rule did not exist under WCAM and indeed, under WCAM the represented class could and often did consists of members from a global background.
3. The fact that damages can be claimed under WAMCA brings into play the issue of damages assessment on a group level whilst properly including individual circumstances, and the principle that the defendant should be found liable to compensate the actual damage (proven by the claimant or estimated by the court), and nothing more.

To date, approx. 30 cases have been filed under WAMCA, not all of which are aimed at seeking damages, but none of which have come to a resolution in first instance. Nevertheless, the Dutch court has ruled on the ‘sufficiently similar’ requirement under WAMCA in one instance, which will be discussed below.

At the same time, several WCAM cases were still pending in the courts after introduction of WAMCA, and indeed: some cases can only be tried under WCAM as the relevant facts predate the temporal scope of WAMCA. Over the past six months or so, the courts have ruled in a handful of cases on the similarity test under WCAM. All of these will be discussed below, in respect of the similarity test.

The applicable standard of similarity under Dutch law.

Art. 3:305 of the Dutch Civil Code (DCC) holds that a foundation or association can file a claim in the interest of protecting similar interests of other persons, if this is in furtherance of the statutory purpose of such foundation or association and to the extent the interests of those persons are sufficiently protected ^[5]This last requirement was introduced in the WAMCA, which entered into force on 1 January 2020 and which applies to cases dealing with facts which occurred after 15 November 2016..

A landmark decision for the definition of the similarity test, is the Baas in Eigen Huis v. Plazacasa case ^[6]HR 26 februari 2010, ECLI:NL:HR:2010:BK5756 (Baas in Eigen Huis/Plazacasa).. In this case, it was set out that the similarity test holds that the interest which are meant to be protected by the collective action, must lend themselves for bundling, such that efficient and effective protection of the interested persons is promoted. If this is the case, the issues can be resolved in one case, without having to involve the specific circumstances relating to the individual interested parties. A sufficient similarity of interests does not necessarily imply that the positions, backgrounds and interests of those on whose behalf a collective action is brought are identical or even essentially the same. In a collective action, therefore, a certain abstract test is appropriate. Whether or not (a substantial part of the) interested persons agrees with the collective action, or even takes an adverse position, in itself is not an obstacle for finding that the case does serve the protection of similar interests. A sufficient level of similarity in position, background and interest of the group members suffices. These do not need to be identical or even predominantly the same. The norm is whether an abstract assessment of the issues a stake and the claims, is possible.

Similarly, and only just preceding the Baas in Eigen Huis case, in the World Online case, the Dutch Supreme Court confirmed ^[7]Hoge Raad 27 november 2009, ECLI:NL:HR:2009:BH2162, NJ 2014/201. the finding of the court of appeals that ‘even if there are great differences between the investors in terms of, amongst other, level of professionalism, level of information, care that was applied when taking the investment decision and the amount of the loss, and the fact that some are not a natural person but a corporation’ this does not stand in the way of finding that the interests are sufficiently similar. The court held that these circumstances can be taken into account if and when damages are assessed for each of the individual investors, in follow on proceedings.

The similarity test applied over the past six months.

Stichting Volkswagen Car Claim v. Volkswagen AG et al. (WCAM)

On 14 July 2021, the Amsterdam District court ruled in respect of the ‘diesel fraud’ case against a group of Volkswagen dealers, largely in favor of the claimant. The foundation in this case is seeking compensation for a class of persons who have bought new and second hand cars from several Volkswagen dealers.

The court held that the fact that the case was filed on behalf of several classes of interested parties, against several groups of defendants, did not mean that the case did not lend itself for bundling, as these distinctions were made sufficiently clear in the claim, and the common denominator for the claimants was that all possessed or had leased a Volkswagen car, and that all defendants had either sold or procured Volkswagen cars.

In the preceding interim decision from 2019 ^[8]20 November 2019, Amsterdam district court, C/13/647072 / HA ZA 18-432., the court had also ruled that the claim for nullification of the contracts with the car dealers on grounds or error, lent itself for bundling: although the decision on the merits would require an assessment of all facts and circumstances which may have had an influence on the decision-taking process of the members of the represented class, the case deals with in essence the same issue, being the presence of software which manipulates the outcome of emissions tests, which was not known to any of the class members. The court said it did not want to rule out in advance that this circumstance would be so relevant, that irrespective of other facts and circumstances, this can be decisive for the outcome of the case and the possible awarding of the declaration of law.

The court in this case had previously also ruled that the claim on the basis of unfair businesses lent itself to bundling in respect of the installation of the software mentioned above. The court did not find this to be so in respect of the information provided by the defendants, as this may vary from case to case.

In the recent 2021 decision, the court contemplates that one of the defendants argued that the claims against it had not been tolled by the claiming foundation, and that they had become time barred under applicable German law. The court finds that indeed, claims of individual car owners have become time barred, and that therefore the claims do not serve to protect ‘similar interests which lend themselves for bundling’. It seems, however, that the court could have sufficed by stating that there simply is no interest whatsoever, and that similarity of the (lacking) interests was not the issue.

FNV v. Wibra Supermarkt BV (WCAM)

On 5 July 2021, the Amsterdam district court ruled in the matter brought before it by the labor union FNV against Wibra supermarkets.

FNV is an association which can bring claims on the basis of art. 3:305a DCC. It filed a claim against Wibra for making workers compensate certain working hours, during which the shops had not been able to retain their services due to Covid-19 related restrictions.

The court ruled that the interest of the workers represented by FNV were sufficiently similar, in the way that the collective labor agreement had been applied in respect of the matter at stake, regardless of how this played out in practice for individual workers.

Data Privacy Stichting v. Facebook (WCAM)

On 30 June 2021, the Amsterdam district court issued an interim decision in the case brought by the Data Privacy Stichting against Facebook on the matter of (amongst other) admissibility of the case. The case relates to facts which only to a limited extent fall within the temporal scope of WAMCA. However, in view of Section 119a Transition Act New Civil Code in conjunction with Section III paragraph 2 WAMCA, the WAMCA does not apply to this case, because the legal actions in this case were instituted before the date of entry into force of the WAMCA (1 January 2020). The case is therefore covered (largely) by WCAM.

The Foundation is seeking a declaration that Facebook acted unlawfully against the class represented by the Foundation by, amongst other:

1. not complying with the information obligations of art. 12, 13 and 14 GDPR and its equivalents under the national law that preceded the GDPR;
2. By giving third parties access to personal data of the class members without properly and timely informing them thereof;

and several other breaches, including the Unfair Business Practises Act.

The represented class constitutes of all natural persons who, in a non-commercial capacity, are or were on Facebook in the period 1 April 2010 through 1 January 2020, whilst residing in the Netherlands.

According to Facebook, the claims of the Foundation do not lend themselves to collective proceedings, because the factual and legal questions are not the same, or at least not sufficiently similar, and the interests of the individual interested parties differ. Facebook contends that there are various factual allegations that cover almost a decade; there are different groups of users and different legal provisions apply; the use of the Facebook service involves a considerable amount of fine-tuning at individual level, as different user agreements, policies and information provisions apply to users, depending on the time period during which each of them used the Facebook service.

The court considered that the question of whether the interests involved in the claims lend themselves to bundling also depends on the nature of the claims made. The claims of the Foundation are limited to declarations for law that there is unlawful act, unfair trade practices and unjustified enrichment. Unlike in, for instance the Elco Foundation v. Rabobank et al. case (see below) the Foundation does not claim a declaratory judgment relating to error, in the assessment of which individual circumstances are rather important. The court finds that the claims of the Foundation relate to various sufficiently narrowly defined actions of Facebook. These claims are essentially based on the fact that Facebook has violated the privacy of its users by processing personal data without the necessary consent. With the claims, the Foundation seeks a judgment on the question of whether the personal data of (certain) users of the Facebook service have been processed in accordance with the applicable regulations. Such a judgment on the lawfulness or otherwise of the actions of Facebook et al. with regard to the processing of personal data lends itself to a collective action. This is not altered by the fact that over time there have been different terms of use and different statutory regulations.

Milieudefensie et al. v. Royal Dutch Shell (WCAM)

On 26 May 2021, the district court of The Hague ruled in the case of environmental NGO Milieudefensie and various other Dutch and foreign environmental NGO’s and individuals, against Royal Dutch Shell (‘RDS’e). Milieudefensie filed claims on a collective basis against RDS, seeking an order for it to bring its CO2 emissions policies in line with Paris Climate Agreement.

The court held that the interests of ‘current and future generations of the world’s population’, as served principally with the class actions, is not suitable for bundling. It held that although the entire world population is served by curbing dangerous climate change, there are huge differences in the time and manner in which the global population at various locations will be affected by global warming caused by CO2 emissions. Therefore, this principal interest does not meet the requirement of ‘similar interest’.’

The court held, however, ‘that interests of current and future generations of Dutch residents and [with respect to one of the Dutch NGO’s the Waddenvereniging, aimed at protecting the interests of (those living on) the Dutch group of Wadden-islands, FMP] of the inhabitants of the Wadden Sea area, a part of which is located in the Netherlands, as served in the alternative with the class actions, are suitable for bundling, even though in the Netherlands and in the Wadden region there are differences in time, extent and intensity to which the inhabitants will be affected by climate change caused by CO2 emissions. However, these differences are much smaller and of a different nature than the mutual differences when it concerns the entire global population and do not stand in the way of bundling in a class action.’

The court therefore declared not admissible the claims insofar as they serve the interest of the world’s population, except for the interest of Dutch residents and the inhabitants of the Wadden region.

It is interesting to see that the court finds that the interests of anyone outside the Netherlands, is not sufficiently similar to the interests of those living within. The court does not explain why this is so, other than that it finds that the differences within the Netherlands are ‘much smaller and of a different nature’. Considering that the merits of the case have been decided on a global scale, and on the basis of an abstract reasoning (impact of CO2 emissions on global warming, statistical models, activities of RDS all over the world), and if anything: the interests of those living in countries which are less well-off than those in the Netherlands are even more affected by global warming, it seems that the interests of all would benefit from this case. At the same time, the fact that the case has formally been decided for the benefit of Dutch nationals only, does not seem to have had any bearing on the actual decision, in the sense that it is not apparent in what way the decision would have been different if the court would have allowed the case to be heard for the wider class.

Stichting Elco Foundation v. Rabobank et al. (WCAM)

The Stichting Elco v. Rabobank et al. case ^[9]District Court of Amsterdam, 9 December 2020, ECLI:NL:RBAMS:2020:6122. related to the scandal of manipulation by a group of panel banks of the interbank interest rates LIBOR and EURIBOR. The foundation, according to its articles of association, represented all financial institutions in the EU, to the extent they had had exposure to LIBOR or EURIBOR in non US transactions, excluding from the class those parties which participated in the LIBOR and EURIBOR determination process, such as the defendants themselves. Defendants included the Dutch anchor defendant Rabobank, and four other, foreign panel banks.

The court held that the interests of the represented group, met the similarity test:

“The represented class, according to the articles of association, is a wide group of persons who have in common – briefly put – that they have a place of establishment or office in the EU and who have engaged in transactions outside the US which in one way or the other are related to the interest benchmarks. The fact that a multitude of laws may apply to the claims of the foundation, in itself does not stand in the way of giving a judgement on the issues raised by the foundation, considering that the foundation has asked for declaratory relief only. These issues are, as the foundation rightly argued, in general separate from the specific circumstances relevant to the interested persons. That means that, [the claims] lend themselves for bundling.”

This seems to be in line with existing case law under WCAM. As damages assessment is a matter for the court to deal with after certification, it seems that the outcome would not have been different under WAMCA.

The court finds, however, that in this case, to the extent the claims are based on the legal theory of unjust enrichment, these are not suitable for collective redress, as an individual assessment is required of both enrichment and impoverishment.

Considering that declaratory relief was sought, it is understandable that the court finds that it cannot give a relevant declaration for each of the individual situations imaginable for this very wide group of both interested parties, but also defendants.

However, the court declines to allow the case as a whole to proceed, on the basis in particular that the claim relates to such a variety of facts over a very long period of time, that it will not be able to give a meaningful declaration in the end, meaning that the interests of the interested persons are not served by proceeding on a collective basis. The facts of the LIBOR and EURIBOR case, as presented to the court, related to a pattern of behavior covering a period of 10 years. For individual parties to claim damages, however, they would need to show individual transactions and the specific effects of individual manipulations of these interbank rates, for them to be able to prove they suffered a loss. The court expressly considers that the facts of this case deviate from the World Online situation discussed above, where the legal matters could be decided abstract of the specific circumstances of individual investors.

In addition, the court deemed relevant that the interested parties were all professional parties, who, if substantial financial interests are at stake, are capable of filing individual claims, meaning that collective redress is not necessary to allow access to an effective remedy.

As said, the case was decided under WCAM. The reasoning may also have applied under WAMCA (apart from that the class in that case cannot consist of non-Dutch financial institutions), considering the typical circumstances of this case: an enormous variety of possible individual positions.

Conclusion as to Dutch law

The standard as developed in the Baas in Eigen Huis and World Online cases under WCAM stand and it seems that it will continue to be the standard under WAMCA. The standard is not a high threshold, nor is it meant to be a high threshold.

A quick look at the UK provides a similar picture: Merricks v. Mastercard

On 11 December 2020, the U.K. Supreme Court handed down its ruling in Merricks v Mastercard ^[10]Mastercard Incorporated and others (Appellants) v Walter Hugh Merricks CBE (Respondent) [2020] UKSC 51, dismissing Mastercard’s appeal against the English Court of Appeal’s April 2019 decision. Class representative Merricks commenced follow-on damages litigation after the EC’s Decision in respect of breaches of competition law, by commencing a U.K. class action in September 2016 on behalf of approximately 46.2 million U.K. consumers. The action seeks approx. GBP 14 billion in damages for the inflated prices paid by U.K. consumers. Collective competition claims must be certified to be able to continue as a collective action. Certification requires that the claims are brought on behalf of an identifiable class of persons, raise common issues, and are suitable to be brought in collective proceedings.

At first glance, the ‘common issues’ and ‘suitable for collective proceedings’ tests are not unlike the ‘similarity test’ applied by the Dutch courts. The UK Supreme Court’s majority decision was that the applicable test whether or not claims are “suitable” to be brought in collective proceedings is a relative judgement, meaning one needs to determine whether the claims are more appropriately brought as collective proceedings rather than individual proceedings. This is also decisive when contemplating whether the claim is “suitable for an award of aggregate damages”. In essence, what is required is that an aggregate damages award is more suitable than “a multitude of individually assessed claims for damages”. This is much more efficient than going through the burden of thousands of individual claims.

Conclusion

The question as to whether or not claims are suitable for collective redress, because they are sufficiently similar, seems to be approached benevolently by the Dutch courts, be it in labour cases, GDPR damages cases or cartel damages cases, or whatever the nature of the collective action. In the UK this does not appear to be any different. Cases where individual circumstances are decisive for a finding of legal wrongdoing, nevertheless are not suitable for collective action. The distinction becomes blurred in cases where claims are based on grounds of error or unjust enrichment, but in all events: the facts of the particular case will be decisive.

Comparative analysis between the UK, Dutch and French approach to passing-on in competition cases

Cartel damages litigation is an increasingly hot topic in Europe. For those who are not familiar with this topic a short explanation. Under normal market conditions, enterprises set their own market prices for their products. Under cartelized conditions, however there is some form of concerted practice (either explicit or tacit) which could lead to an agreement on prices for instance. If competitors agree on a price, this will normally lead to higher prices than under normal circumstances. Competition law prevents competitors this kind of behavior, setting high penalties (potentially in the billions of Euros for world-wide players) when trespassers on the EU competition laws are caught. However, these penalties, severe as they might appear, nevertheless pale into insignificance compared to civil damages claims. The gap between the normal market price and the artificial cartelized price is the so-called cartel damage. In addition, to give you an idea of the enormity of these kind of damages we give as an example the trucks case. Six truck companies were caught red handed by the European Commission in a cartel that lasted (at least) from 1997 to 2011. They were fined 3.8 billion euros. However, estimates are that the total of cartel damages amount to a staggering figure of 200 billion euros. These stakes are high enough to ensure massive court battles. The defendants have no other option than to put forward any possible or impossible argument to prevent the court from a decision that might cause their bankruptcy. For the claimants on the other side it is inevitable to go to court, since their losses have been so high.

Therefore, when stakes are as high as they are, there is no other option than to be very thorough in all the arguments and defenses that are brought to the court. Let alone to be meticulous in the first place as where to litigate.

Several legal topics have drawn special attention over the last few years. In this series of articles, we will shed light on the most hotly debated. Today we discuss the passing on defense. We made a comparative analysis between the Netherlands, France and the United Kingdom. The Netherlands and England (along with Germany) are considered mature jurisdictions in cartel damages litigation, while France (along Spain and Portugal for instance) are on the move to join this lawyer’s paradise.

Passing-on

“Passing-on” in competition cases is where overcharges caused by a cartel, which affect the customers of the cartelists (direct purchasers), are passed-on by these purchasers to buyers further down the supply chain (indirect purchasers). The pass-on argument as a defense may be invoked by a cartel member as a (partial) shield against a claim for damages and by an indirect purchaser as a sword to support the argument that it has suffered damages and/or to evidence cartel collusion.

Legal background

The passing on defence is valid under both EU and national laws.

The Damages Directive ^[1]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 … Continue reading (and implementation laws in the Member States) has set two important presumptions reversing the burden of proof:

– as far as direct purchasers are concerned, it is presumed that they have not passed the overcharge on to their own customers. Thus, it is up to the defendant in the antitrust action for damages to prove that the overcharge has indeed been passed on and that its direct purchasers have not suffered any (or less) damage (Article 13 Damages Directive);

– concerning the indirect purchasers, it is presumed that their supplier has passed on the overcharge. Therefore, the burden of proof is here again placed on the defendant in the action for antitrust damages (Article 14 Damages Directive).

Those in itself contradictory presumptions could potentially apply to all claims. The presumptions in itself apparently are designed to help the (potential) claimants in a case.  Everyone familiar with civil litigation knows that a sentiment of wrongdoing is not the same as proving you were wronged. Therefore, it is extremely important to have the burden of proof shifted to the wrongdoers.  In that regard, national courts have established a common understanding for the enforcement of the passing on defence it is for the defendants to prove passing on and the extent thereof as well as the absence of volume effects.  This outcome is consistent with the acquis communautaire on the burden of proving pass-on (i.e. in line with the EU principle of effectiveness) that has been codified in Article 13 Damages Directive. So how do courts apply these presumptions in practice in their case law? We believe that this would lead to the conclusion that the odds should favour the claimants. However, do courts indeed apply this presumption? We compared the United Kingdom, the Netherlands and France.

United Kingdom approach

In two recent landmark cases the UK Supreme Court (SC) and the UK High court rendered decisions on the UK evidential standard in connection to the passing on defense. The Courts emphasized that claimants should be neither undercompensated nor overcompensated and therefore the evidential burden in relation to mitigation of loss on defendants / cartel members should not be ‘unreasonably high’. This is an approach in which no apparent choice seems to be made in favor of the claimants or the defendants.

In June 2020, the SC overturned a decision of the Court of Appeal in which it had decided that it required defendants (cartel members) to prove the (virtually) exact amount of loss mitigated in order to reduce claimed damages. ^[2] UK Supreme Court 17 June 2020, case refere The SC decided that the law does not require such high evidential standard and that the Court of Appeal had erred insofar that it required ‘unreasonable precision’ from the defendants in the proof of the amount of loss that the claimants had passed on to end customers. This decision was rendered in the context of a damages lawsuit brought by several supermarkets against Mastercard, Visa and several other large banks for the use of certain payment card schemes, in particular the multilateral interchange fees (MIFs)[3] applicable in the EEA, which the European Commission found to be anti-competitive in 2007.

Key considerations of the SC regarding the evidential burden of defendants:

• Claimants should not be under-compensated but neither overcompensated when suffering harm from a competition law breach.
• In the UK, pass-on is an element in the quantification of damages that is required by the compensatory principle and required to prevent double recovery through claims in respect of the same overcharge by a direct purchaser and by subsequent purchasers in a chain. Against this background, the SC considered that “justice is not achieved if a claimants receives less or more than its actual loss”. (217)
• A balance is required between the compensatory principle and the principle that disputes should be dealt with ‘at a proportionate cost’. In that light “the court and the parties may have to forego precision, even where it is possible, if the cost of achieving that precision is disproportionate, and rely on estimates”. (217).
• The SC considered that it sees no reason why in assessing compensatory damages there should be a requirement of greater precision in the quantification of the amount of an overcharge which has been passed on to end consumers because there is a legal burden on defendants in relation to mitigation of loss. (219).

The preceding approach does not offend the principle of effectiveness of EU Law, according to the SC:

“As we have said, the relevant requirement of EU law is the principle of effectiveness. The assessment of damages based on the compensatory principle does not offend the principle of effectiveness provided that the court does not require unreasonable precision from the claimant. On the contrary, the Damages Directive is based on the compensatory principle.” (220)

“ As the regime is based in the compensatory principle and envisages claims by direct and indirect purchasers in a chain of supply it is logical that the power to estimate the effects of passing-on applies equally when pass-on is used as a sword by a claimant or as a shield by a defendant.” (224)

On 25 February 2021, the UK High Court rendered a decision in which it validated the approach to pass-on of the SC decision of June 2020.[4] The UK High court referred to several key considerations of the SC in its decision, among which the compensatory principle and the fact that claimants should not be overcompensated for their damages as much as they should not be undercompensated. In addition, the decision implies that the pass-on approach can / should also be applied in complex and a-typical cases of passing on of overcharges, as the underlying case, which would pose a difficult and costly evidential burden on both parties.

This case stems from the foreign exchange cartel and a damages lawsuit filed by Allianz Global Investors and other claimants against several banking groups for their participation in the aforementioned cartel. The case involves investment funds who seek to generate a return for their investors and in so doing make use of the foreign exchange services provided by banks. Pass-on in this case is said to occur when an investor redeems or withdraws his investment from the fund. The High Court made clear that even though this is not a typical pass-on case involving the sale and purchase of goods in a supply chain, the compensatory principle equally applies. Defendants should not be subject to double recovery (12).

Claimants argued that the defendants’ pass-on defense should be stricken out because there was no real prospect of success, mainly because investors would have no cause of action against the banks. They also argued that it would have great impact on the future scope of claims, in terms of disclosure of documents and provision of evidence. The UK High Court did not agree that there would be no real prospect of success. The UK High Court said that the defense was appropriate and therefore it considered it necessary to “investigate precisely how the alleged wrongdoing of the Defendants impacted upon the investment fund (…) and how that affected the sum payable to the investor”, by disclosure and by factual and perhaps expert evidence. The pass-on defense could therefore be advanced to trial.

Our preliminary conclusion is that the UK courts do not favor one party over the other. It seems as if the courts feel that over-compensation is just as bad as under-compensation. In addition, in this argument there seems to be a deviation from the choice for the principle of effectiveness. So how do the Dutch approach this?

The Dutch approach

In Dutch case law the threshold for an effective passing-on defense has been set relatively high, contrary to the standard that has been set in the UK. The principle of effectiveness and the scope of the Cartel Damages Directive have served as normative and guiding principles for Dutch courts in this regard. The landmark judgements stem from a follow-on damages case between electricity transmission operator TenneT and electricity equipment corporation ABB.

On 8 July 2016, the Supreme Court ruled in the cartel damages case between ABB and TenneT that ABB was liable for the damage suffered by TenneT through the cartel on the market of gas-insulated switchgear.[5]

Amongst others, the Dutch Supreme Court considered in this case that even without retroactive effect for material law, the Damages Directive nevertheless has to be taken into account in order to sure the European l’effet utile and the principle of equality (4.3.1 and 4.3.4). In other words, it was clear that the Damages Directive was not applicable in this case but the Supreme Court did consider it. With reference to article 12 (3) of the Damages Directive, the Supreme Court decided that the evidential burden in connection to passing on is in principle on the cartel member. Furthermore, the Supreme Court confirmed that the court is authorized to estimate damages if it is not possible to determine the amount of damages precisely. So with reference to the not applicable Damages Directive the Supreme Court clearly decided in favor of the claimant.

The District Court of Gelderland delivered judgment on 29 March 2017 and ABB was ordered to pay € 23 million in compensation for the damage caused by the cartel.[6] ABB argued against the extent of the damage by invoking the passing-on defense. The District Court did not agree with this and found that the question whether this defence is reasonable, the principle of equality, the principle of effectiveness and the scope of the Damages Directive serve as normative and guiding principles (4.17). The court considered that “the object of the Damages Directive is not that the infringer should be given a hook to get out his liability of damages. The intention is that the compensation to be paid by the infringer should accrue to the direct and indirect customers in the chain to whom the additional costs were charged” (4.18). The court also ruled that the chance of end consumers bringing their own damages actions (and thus the risk of double compensation) was negligible. Once again and even clearer than the Supreme Court the court here argued in favor of the claimant, especially because they took into account the possibility of actual passing on, but that it was unlikely that further down the line any consumer would collect these scattered damages.

The so-called efficiency defense has been paid particular attention to in this context. Nowadays, almost every claimant advances this defence. Parties claim the harm suffered and alternatively claim compensation by invoking the efficiency defence. Briefly stated, the efficiency defence results in the court nevertheless awarding compensation to the claimant even if strictly speaking, the claimant is unable to prove the harm. This is of course a slippery slope from the point of view of legal certainty. However, it does seem to follow the European starting position, which is that the process should not be made too difficult for claimants and which forms the basis for the Damages Directive.It is a means of ensuring that private litigation is not made impossible from the very start.

The French approach

In France, in cases in which Pre-Damages Directive rules apply, the question of the burden of proof regarding the passing on defense is not entirely settled yet.

According to the Circular of 23 March 2017, the new Article L481-4 of the French Commercial Code (including burden of proving passing on) does not apply to damages claims resulting from an infringement, which took place before its entry into force, that is on 11 March 2017.

Prior to the adoption on 9 March 2017 of the rules implementing the Damages Directive in France, there was no specific legal provision on the issue of pass-on of overcharge in cartel damages cases. The general civil law provisions, namely Art, governed the issue. 1315 of Old French Civil Code (Art. 1353 of the New French Civil Code with the exact same wording).

In accordance with  a ruling of the Cour de cassation[7], several French civil and commercial courts of first instance have handed down judgments putting the burden on the claimant to prove the absence of passing on.^[8] While in two cases the Paris Court of Appeal considered that it was for the defendant to prove the passing on after the plaintiff had brought some indicia showing that there was no passing-on^[9], it recently quashed a ruling of the first instance court where it found that the plaintiff which had been granted damages had not provided any evidence that there had not been any passing on.[10] Most recently however, the Paris Administrative Court of Appeal^[11] and the French Court of Cassation^[12] held that, in accordance with EU law and principles, the burden of proof regarding the passing on of the illegal overcharge lies with the defendant. Besides the economic aspects (additional damages in form of loss of profit and the difficult proof of causality), there are therefore strong legal arguments to counter any potential passing-on defence. As the case law stands, it therefore seems highly recommendable to the plaintiffs that they bring as much evidence as possible about the absence of passing-on, allowing then the French Courts to shift the burden of proof on the defendant to establish passing-on.

Therefore, overall it appears that the French courts are slowly marching away from Albion to get closer to the Dutch approach regarding cases in which Pre-Damages Directive rules apply.

On the other hand, with regards to cases where post-Damages Directive rules apply, there is no doubt that the plaintiffs will fully benefit from the presumptions set by the Damages Directive.

Next time we discuss the bundling of claims!

bureau Brandeis, 4 June 2021

Marc Barennes, Tessel Bossen, Hans Bousie & Sarah Subremon

[1] 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 Text with EEA relevance (OJ L 349, 5.12.2014, p. 1–19)

[2] UK Supreme Court 17 June 2020, case references UKSC 2018/0156 Sainsbury’s Supermarkets Ltd and others (Respondents) v MasterCard Incorporated and others (Appellants); UKSC 2018/015 Sainsbury’s Supermarkets Ltd (Respondent) v Visa Europe Services LLC and others (Appellants).

[3] MIFs are charged by a cardholder’s bank (issuer) to a merchant’s bank (the acquirer) for each transaction made to the merchant with a payment card. In practice, the acquiring banks passed these fees on to merchants by charging a merchant services charge (MSC), for which they were seeking damages. The banks however argued that the supermarkets had passed-on the overcharges by the MIFs/MSCs to their end consumers by raising retail prices.

[4] UK High Court 25 February 2021, Case no CL-2018000840.

[5] Supreme Court 8 July 2016, ECLI:NL:HR:2016:1483.

[6] District court of Gelderland 29 March 2017, ECLI:NL:RBGEL:2017:1724.

[7] Cour de cassation, 15 May 2012, Le Gouessant.

[8] Paris Commercial Court, 26 March 2018, Provera; Paris Commercial Court, 20 February 2020, Cora; Rennes High First Instance Court, 7 October 2019, FRSEA.

[9] Paris Court of Appeal, 20 September 2017, JCB; Paris Court of Appeal, 6 February 2019, Doux.

[10] Paris Court of Appeal, 14 April 2021, Johnson & Johnson.

[11] Paris Administrative Court of Appeal, 13 June 2019, SNCF Mobilités.

[12] French Court of Cassation, 12 February 2020, Collectes valorisation énergie déchets. This case does not relate to antitrust damages and may only be referred to by analogy.