Minimum Damages under EU Law in Antitrust Damages Actions? Conclusions from the Recent Case Law of the German Supreme Court
“The estimated damage cannot be less than 5% of the purchase price paid for reasons of effectiveness under EU law.” This was recently stated by the German Federal Court of Justice about claims for damages due to the Dieselgate scandal. However, the Court’s reasoning in favour of this legal lower limit for damages might equally, if not even more so, be applied to cartel damages claims – an analysis.
Regarding the well-known diesel emissions scandal (Dieselgate), the German Federal Court of Justice (Bundesgerichtshof, BGH) explained in the summer of 2023 that the buyer of a motor vehicle equipped with a defeat device prohibited under EU law is entitled not only to compensation but may also expect compensation of at least 5% of the purchase price paid. The reasons for this could also apply beyond the diesel complex and, perhaps, even beyond Germany. The latter has recently been confirmed by the Austrian Supreme Court. In essence, the BGH relies on the requirement of the Court of Justice of the European Union (CJEU) to sanction the relevant infringement of EU law through a national claim for damages in an effective, proportionate and dissuasive manner, using the principle of effectiveness. On this basis, would it not be consistent to also assume a minimum amount of damages be compensated in the event of a breach of EU competition law?
We argue that a lower limit to the judge’s discretionary power of estimation when quantifying damages would also be favourable in hardcore cartel cases if it were established that the cartel in question caused damage. In the European Union, the Directive 2014/104/EU on antitrust damages (Damages Directive)[1] provides for a presumption of harm in the case of cartels. Therefore, the occurrence of damage is proven in the absence of a well-established theory of no harm. In such cases, a minimum damage of a certain percentage appears to be justified.
Recent Dieselgate case law of the Federal Court of Justice: minimum damages 5% of purchase price paid for reasons of effectiveness under EU law
On 21 March 2023, the CJEU found that Member States must provide that the buyer of a vehicle equipped with a prohibited defeat device, within the meaning of Regulation (EC) No 715/2007, may claim damages from the vehicle manufacturer when that device has damaged the buyer. This right essentially follows from the fact that, in accordance with Directive 2007/46/EC (Framework Directive), Member States must sanction such an infringement effectively, proportionately and dissuasively. In the absence of EU law provisions, the details are to be governed by national law, which, subject to the principle of effectiveness, must not lead to unjust enrichment of the buyer.[2]
Therefore, in a series of landmark judgments on 26 June 2023,[3] the Civil Senate No. Via of the BGH held that the purchaser of a diesel vehicle can claim damages under German law for having purchased the vehicle at too high a price, due to the risks associated with the illegal defeat device (e.g. official operating restrictions) compared to the hypothetical situation without the infringement. The amount of this (‘differential’) damage must be estimated by the trial judge under Section 287 of Germany’s Code of Civil Procedure – but, as the BGH emphasised, in compliance with the ‘requirements of EU law’.
The BGH provided details about these requirements:[4]
‘The [CJEU] has held that the sanctions […] must be effective, proportionate and dissuasive and that national provisions must not make it practically impossible or excessively difficult for the purchaser to obtain adequate compensation ([…]). This gives rise to requirements of EU law for the application of national law about both the lower limit and the upper limit of the damages to be awarded under [national law], which legally limit the discretion in the estimation within a range of between 5% and 15% of the purchase price paid.’
In particular, the BGH justified the lower limit of the estimation discretion of 5% by linking it to the behavioural control of potential infringers:
‘For reasons of effectiveness under EU law, the estimated damage cannot be less than 5% of the purchase price paid. Otherwise, the sanctioning of an even merely negligent infringement of [Regulation (EC) No 715/2007] would not be sufficiently effective in the promotion of the objectives of EU law due to its insignificance. The damage estimation must lead to a sanction that is also tangible for the vehicle manufacturer in terms of amount. However, the sanction is not only tangible in this sense if the damages awarded are suitable in themselves to bring about a behaviour change. This would hardly be possible given the scope of the manufacturer’s business activities on the one hand and the maximum possible amount of damages in an individual case on the other. Rather, it is sufficient if, on the one hand, each sanction taken individually is associated with a not insignificant loss in relation to the manufacturer’s earnings associated with the transaction and, on the other hand, the sanctions for many legal infringements as a whole can bring about a change in behaviour in terms of compliance with all legal acts. This is the case with a lower assessment limit for damages of 5% of the purchase price paid.’
In contrast, the BGH considers damages of more than 15% to be ‘disproportionate’. For this, the BGH points out that the prohibited defeat devices would be ‘objectively comparatively minor infringements of the law’. Besides, it has to be taken into account that the vehicle manufacturer is equally liable to every buyer in the event of a resale of the same vehicle (‘cumulative effect’).
According to the BGH, the estimation of damages still follows ‘original aspects of tort law’, and the trial judge is not obliged to obtain an expert opinion when estimating within the aforementioned scope. Moreover, the BGH states:
‘The parties’ submission that […] the damage in the specific case amounts to less than 5% or more than 15% of the purchase price paid is irrelevant for the estimation of the trial judge. Such assertions are irrelevant because the principles of effectiveness on the one hand and proportionality on the other limit the compensation of differential damages for legal reasons and cannot justify taking evidence by obtaining an expert opinion.’
Another Senate of the BGH and many other German courts have meanwhile confirmed this case law.[5] Beyond Germany, the Austrian Supreme Court (Oberster Gerichtshof, OGH) recently transferred this case law to Austrian law and referred to the requirements of EU law for compensation as well.[6]
The courts further clarify that those principles do not rule out considering circumstances that occur later and reduce the damage via ‘benefit equalisation’ (Vorteilsausgleichung). For example, if the vehicle manufacturer invokes the subsequent improvement of the vehicle through a software update, and the software update significantly reduces the risk of operating restrictions, this may reduce the damage.[7]
Transferability of this case law to cartel damages
Lower assessment limit for damages
At first glance, the prohibited defeat device of a diesel vehicle and the infringement of EU competition law do not have much in common. However, under the conditions of Article 17(1) Damages Directive, damages resulting from the latter are also subject to judicial estimation in Germany, as in the case of the defeat device under Section 287 Code of Civil Procedure.
In particular, the ‘requirements of EU law’ from which the BGH derives the lower limit for the damages to be estimated are by no means due to special legal features of the diesel complex but apply equally to damages resulting from antitrust infringement:
- The individual’s right to compensation is based directly on the premise of the effectiveness of EU law, both in the case of prohibited defeat devices (Framework Directive and Regulation (EC) No 715/2007) and in the case of infringements of competition law (Articles 101, 102 TFEU).[8] Sanctions for breaches of EU law must always be effective, proportionate and dissuasive.[9]
- In both cases, EU law does not provide specific requirements for the estimation of damages as such but leaves the details to the national laws of Member States, subject to the principle of effectiveness (for antitrust damages, see Articles 4 and 17(1) of the Damages Directive).[10]
- The idea of preventative behavioural control of potential infringers, from which the BGH derives the lower damage limit in the Dieselgate cases (‘change in behaviour in terms of compliance with all legal acts’), is even more important for antitrust damages. The CJEU leaves no doubt about this:[11]
‘Beyond the compensation itself for the harm alleged, the establishment of such an entitlement [to damages; added by this Author] contributes to the objective of dissuasion, which is at the heart of task of the Commission, which is under a duty to pursue a general policy designed to apply, in competition matters, the principles laid down by the TFEU and to guide the conduct of undertakings in the light of those principles […] It follows from the foregoing that, just as is the case for the implementation of the EU competition rules by public authorities (public enforcement), actions for damages for infringement of those rules (private enforcement) are an integral part of the system for enforcement of those rules, which are intended to punish anticompetitive behaviour on the part of undertakings and to deter them from engaging in such conduct.’
An upper assessment limit for damages?
The BGH has also derived from EU law an upper damage limit of 15%. The question is whether the principles of proportionality that are decisive for the upper damage cap apply to antitrust cases as well. Firstly, as the above CJEU statement suggests, cartel offences are not objectively minor infringements of the law, in contrast to the Dieselgate cases (in the opinion of the BGH). On the contrary, the more harmful the cartel, i.e. the more the overcharge exceeds 15%, the higher the interest in effective sanctions to protect the European internal market. Secondly, antitrust damages do not bear the risk of the cumulative effect that the BGH fears in the Dieselgate cases. Cartel offenders must only compensate for the harm they have caused once. Hence, there is no reason to suggest that a similar upper limit should be applied to cartel cases.
On parallel antitrust damages claims of several distribution levels
Another question is who is harmed and to what extent. Overcharges caused by an antitrust infringement may have been passed on in whole or in part from one market level to the next, e.g. from persons who acquired a cartelised product directly from an infringer (direct purchaser) to their own customer (indirect purchaser). The Damages Directive devotes an entire chapter to this issue (Articles 12 et seqq.). Accordingly, both direct and indirect purchasers shall be able to effectively claim damages, but only to the extent that they have suffered harm because of the overcharge at their respective distribution level. Non-liability of the infringer shall be avoided, as well as double claims for the same damage or over-compensation of the respective customers.
However, this should not affect the lower limit of damages as defined by the BGH. This limit primarily concerns the assessment of the overcharge itself, irrespective of any pass-on. Consequently, Dieselgate case law does not preclude the principle of benefit equalisation (Vorteilsausgleichung, see above), in the context of which the passing-on defence, within the meaning of Article 13 Damages Directive, is examined under German law. The behavioural control intended by the lower limit on damages is also in line with the goal of avoiding (de facto) non-liability of antitrust offenders.
Commentary and categorisation
This is not the place to ask whether the conclusions of the BGH in the Dieselgate case are correct in themselves (which is frankly controversial in the literature). At the very least, they bring up a painful subject in the context of antitrust damages.
The practical need for a lower damage limit
Many a plaintiff who has been awarded antitrust damages complains that they were given a stone when they asked for bread. The much-noticed, long-running legal dispute before the Mannheim Regional Court over the sugar cartel, which the Federal Cartel Office had classified as a ‘well-functioning’ hardcore cartel, in the form of a market-sharing agreement that covered the entire national market for almost 15 years, ended in 2023 with the finding that the cartel-related overcharge had only amounted to 2%.[12] The Higher Regional Court of Schleswig-Holstein is assuming an overcharge of only 0.5% in the case of a drugstore cartel – while the plaintiff invested around €325,000 in its expert opinion on damages alone.[13] The damages calculated from these overcharges hardly even cover the plaintiffs’ legal costs.
Such results, if assumed to be correct, are not tangible sanctions that deter undertakings from committing antitrust infringements. The objective of effective compensation for damages (Articles 3 and 4 Damages Directive) is also missed. The prospect that a cartel victim will only achieve a Pyrrhic victory characterised by a negative cost-benefit ratio increases the risk that it will not assert its rights at all.
In this way, a ban on cartels which relies not only on public enforcement but also private enforcement of individual rights is devalued.
However, this risk would be counteracted by a lower limit for damages, in the sense of the BGH, and thus fixed minimum damages (which should be tangible). There would also be an incentive to sue for damages where it is unclear whether an expensive quantification and presentation through economic expert opinions is worthwhile. This would serve the effectiveness of the cartel prohibition and the objective of deterrence. Besides, cartel damages proceedings would be more predictable for all parties involved, promoting legal certainty. Given the large number of injured parties and individual proceedings typical of cartels, this outcome would have a significant impact. In fact, the existence of mass claims is said to have played a role in the BGH’s fixing of the estimate range of 5-15% in the Dieselgate cases.
Compatibility of the lower loss limit with the compensation principle
Criticism that such minimum damages constitute unjust enrichment or punitive damages is misguided, in principle at least.
The minimum damage within the meaning of the BGH case law only becomes relevant if it has already been established that the infringement (the cartel offence) has caused damage. However, if it is established in the first step that a claimant has suffered harm, the amount of harm must be estimated in a further step, unless it is either practically impossible or excessively difficult to quantify it precisely, see Article 17(1) Damages Directive. This estimate, which is essentially based on mere assumptions about hypothetical market conditions absent the infringement, naturally gives the courts some discretion. The lower damage limit formulated by the BGH only narrows this discretion (in the second step) and shifts the risk of inaccuracy inherent in any estimate to those who are responsible for the damage: the infringers.[14] The estimate itself, as the BGH made clear, is still based on original aspects of tort law.
In fact, the judicial findings on the sugar and drugstore cartels mentioned above also do not claim absolute truth in the sense of mathematical accuracy. They are probably not even at the upper end of reasonable compensation. Plaintiffs’ economic experts have determined far higher overcharges than the courts in Mannheim and Schleswig-Holstein. In practice, it is unlikely that the justification for a reasonable legal minimum amount of damages will ever be called into question because it is clear that the actual damage in individual cases is significantly lower.
Key points of an antitrust damages floor
Assuming a minimum damage prescribed by EU law requires some clarifications, however:
- Occurrence of harm must be established independently. To emphasise this once again, the minimum damage assumption would only concern the amount of harm. It must therefore be proven beforehand that the plaintiff has suffered harm. Article 17(2) Damages Directive provides for a presumption of harm in the case of cartels, but this is rebuttable. In other words, the minimum damage presupposes – but is also justified by the fact – that the defendant has not succeeded in proving a ‘theory of no harm’.
- Hardcore restraint of competition. The minimum damage is likely to be considered at least for the prevention of serious infringements of competition law, such as a ‘cartel’ within the meaning of Article 2(14) Damages Directive (e.g. price-fixing; market-sharing) that violates Article 101 TFEU. This would also be in line with the said presumption of harm under Article 17(2) Damages Directive, which applies exclusively to those cartels. In contrast, it is questionable whether other restrictions of competition, which may be prohibited under Article 101 or 102 TFEU as well, also require this type of behavioural control.[15]
- Difference to any presumption of the amount of harm. The minimum damage in question should not be confused with a (rebuttable) presumption of the amount of harm, which – unlike the Damages Directive (Recital 47) – is provided for in individual national legal systems for cartel offences (e.g. Romania: 20% overcharge; Hungary and Latvia: 10%). The much-discussed ‘broad axe’ when estimating the amount of damage[16] is also irrelevant in this context. Instead, the BGH derives from the CJEU case law a normative (lower) limit of damages directly under the principle of effectiveness, which narrows the range of estimation from the outset. It is therefore not possible for a defendant to refute this minimum damage based on the actual circumstances of the individual case, irrespective of the possibility to prove that any damage has been passed on to the next market level (see above).
- Direct and indirect customers. The minimum damage would apply to claims for damages from direct customers of a cartel as well as from indirect customers operating at downstream market levels. This is logical both because of the underlying concept of effectiveness and the fact that the question of passing on losses from direct to indirect customers would be examined independently.
Role models in national law
The above considerations are by no means revolutionary. On the contrary, one can find examples of this in current European antitrust law developments.
For instance, the 11th Amendment to the German Competition Act (CA), which just came into force on 7 November 2023, stipulates that the ‘economic benefit’ from an antitrust infringement, which the competition authorities can skim off from an infringer, ‘amounts to at least 1% of the turnover achieved in Germany with the products or services related to the violation’ (Section 34(4)(4) CA). It cannot be argued that no economic benefit or merely a minor benefit has accrued (Section 34(6) CA).[17]
These provisions indeed aim at the disgorgement of benefits by the competition authorities, but they can in principle be applied to private damages as well (irrespective of the too-low percentage value). The economic benefit of a cartel offender from its infringement reflects the damage caused by it to cartel victims. Accordingly, the disgorgement of economic benefits is expressly subsidiary to private damages (Section 34(2)(1) no. 1 CA) and, like the latter, is based on an estimate by civil law standards (Section 34(4)(3) CA refers to the above-mentioned Section 287 Code of Civil Procedure). The (overly cautious) generalisation of the economic benefit is intended by the German legislator to be preventive and to serve the need for effective enforcement of competition rules[18] – just like private antitrust damages.
Determination of the lower damage limit
The question remains as to the amount of minimum damage. Its preventive purpose can only be fulfilled if the amount is not merely symbolic. The BGH demands a tangible sanction, and therefore sets the estimated damage in the Dieselgate cases at a minimum of 5% of the purchase price paid.
In fact, in the antitrust arena, in several decisions in 2023, the Spanish Supreme Court (Tribunal Supremo) set the overcharge on the Truck Cartel at 5% as well, especially for cases in which the plaintiffs were unable to prove higher damages due to a lack of suitable economic expert opinions.[19]
On the other hand, the BGH only had a minor and negligent breach of EU law in mind when setting the estimation range of 5-15% for the Dieselgate cases. Even in this context, other German courts and legal scholars consider this estimation range to be too low for an effective and dissuasive sanction. The CJEU will soon also be dealing with this issue.[20] In any event, it is reasonable to assume higher values in the case of serious, wilful breaches of competition law that can shake entire markets to their foundations. In the case of contractual liquidated damages that the plaintiff and defendant have agreed in advance in the event of a cartel infringement, with reference to general cartel studies and empirical economics, the Cartel Senate of the BGH already considers values of up to 15% to be ‘justifiable and appropriate’ for compensation.[21] With the prospect of such damages, private enforcement would indeed prove to be an effective sanction for cartel offences.
* This entry is a re-post of the contributor’s blog post here.
[1] [2014] OJ L349/1.
[2] CJEU, ECLI:EU:C:2023:229, Mercedes-Benz Group, paras. 90 et seqq.
[3] E.g., BGH, ECLI:DE:BGH:2023:260623UVIAZR335.21.0.
[4] The following according to BGH, ibid., paras. 73 et seqq. Translation, underlining, as well as omissions and additions in brackets in the quotations, by this Author.
[5] E.g. BGH, ECLI:DE:BGH:2023:200723UIIIZR267.20.0 (20/07/2023), para. 34.
[6] OGH, 10 Ob 27/23b (28/09/2023), paras. 39-40.
[7] Recently, e.g., Appeal Court of Celle, ECLI:DE:OLGCE:2023:1122.7U40.23.00 (22/11/2023), at II 1 f.
[8] See, on the one hand, AG Rantos’ Opinion in Mercedes-Benz Group, ECLI:EU:C:2023:229, paras. 53-55, and, on the other hand, CJEU, ECLI:EU:C:2001:465, Courage and Crehan, paras. 25 et seqq.
[9] Already, CJEU, ECLI:EU:C:1989:339, Commission v Greece, paras. 23-24; see also CJEU ECLI:EU:C:1984:153, Von Colson and Kamann, paras. 23, 28; ECLI:EU:C:1997:208 , Draehmpaehl v Urania Immobilienservice, para. 25.
[10] See again AG Rantos’ Opinion, loc. cit.; see also CJEU, ECLI:EU:C:2022:494, Volvo and DAF Trucks, para. 81.
[11] CJEU, ECLI:EU:C:2021:800, Sumal, paras. 36-37.
[12] E.g., Regional Court of Mannheim, ECLI:DE:LGMANNH:2023:0623.14O103.18KART.00 (23/06/2023), para. 291.
[13] Higher Regional Court of Schleswig-Holstein, ECLI:DE:OLGSH:2023:1012.16U97.22KART.00 (12/10/2023), paras. 77, 86, 104.
[14] See also for “estimation windows” in the quantification of antitrust damages: Schweitzer/Woeste, Zum Umgang mit ökonomischer Unsicherheit bei der Schätzung von Kartellschäden: Eckpfeiler eines kartellschadensersatzspezifischen Beweisrechts, Zeitschrift für Wettbewerbsrecht (ZWeR) 1/2022, pp. 46-79 [here].
[15] On the other hand, it is exactly the principle of effectiveness, which the BGH considers relevant for the minimum damage, that also leads to damages in case of the abuse of a dominant position under Article 102 TFEU: see CJEU, ECLI:EU:C:2019:263, Cogeco Communications, paras. 39-41.
[16] E.g., Royal Mail Group v DAF Trucks & Ors [2023] CAT 6 (07/02/2023), paras. 173 et seqq., 479 et seqq.
[17] Sect. 34 of Germany’s 2023 Competition Act is available in English here.
[18] In detail: Printed Matter of the Bundestag no. 20/6824 (2023) pp. 39-40.
[19] For this, see in detail https://carteldamageclaims.com/2023/06/29/tribunal-supremo-trucks-cartel.
[20] See the last question referred to the CJEU in: Regional Court of Ravensburg, ECLI:DE:LGRAVEN:2023:1027.2O232.20.00 (27/10/2023); see also Thiede, “Thermofenster”, Sanktionen, Verbotsirrtümer und Nutzungsersatz; zugleich Besprechung von BGH Urt. v. 26.6.2023 – VIa ZR 335/21, Europäische Zeitschrift für Wirtschaftsrecht (EuZW) 21/2023, pp. 990-992 [here].
[21] BGH, ECLI:DE:BGH:2021:100221UKZR63.18.0, Rails Cartel VI, paras. 43-45.
