This paper, available here, deals with a judgment of 11 December 2018 by Germany’s Federal Court of Justice’s Cartel Panel dealing with the applicability of the principle of prima facie evidence (in German ‘Anscheinsbeweis’) regarding causation of damage.

German Gavel

Section II focuses on the law of evidence regarding competition damages in Germany.

Under German law, each party bears the burden of proving all elements of any statutory provision on which they rely. For cartel damages cases, this means that a claimant asking for damages needs to prove: (i) an infringement of competition law, (ii) the existence of such damages and (iii) that the damages were caused by the infringing behaviour.

This situation changed, as regards follow-on claims, as a result of the implementation of the EU Damages Directive. With the goal of ensuring that neither the burden nor the standard of proof render the exercise of the right to compensation practically impossible or excessively difficult, Article 17(2) of the Damages Directive sets out a legal presumption to the effect that cartel infringements cause loss. Furthermore, the implementation of the Damages Directive has also led to a reversal of the burden of proof regarding causation. It is no longer upon the claimant to prove that the cartel caused damages. Instead, a defendant in follow-on claims will have to prove that the cartel did not cause damages to the claimant (see sec. 292 of the Code of Civil Procedure).

However, before Germany adopted its 9th Amendment to the German Competition Act which came into force on 9 June 2017 – or, to be more precise, before the Directive became directly effective on 27 December 2016, the time limit for its implementation – causation of loss was a matter solely regulated by national law, subject to the principles of effectiveness and equivalence. As such, for the pre-Damages Directive era – which covers many of the cases now making their way through Germany’s courts – the central question under German law remains how and when a party meets its burden of proof.

According to sec. 286 (1) of the Civil Process Code, the standard of proof is only met where the claimant can demonstrate to the conviction of the court that the infringement in fact caused harm. In practice, this means that the standard of proof is met when the judge is convinced that certain facts can be reasonably deemed true.  In coming to such a conviction, the judge must (ex officio) take rules of experience into consideration. Such rules of experience are relevant when certain effects typically emerge from certain causes (and vice versa). According to established case law, a sequence of events can be considered typical if it occurs so frequently that there is a very high chance that an individual case belongs to a group of typical cases. Where such established rules of experience exist, the ‘only’ task for the party bearing the burden of proof is to show that the particular case in front of the court matches such an established rule of experience.

Case law in the lower courts consistently held that experience shows that cartels typically lead to harm, and that, as a result, evidence of a cartel infringement is prima facie evidence of loss. Defendants being confronted with prima facie evidence must challenge the typicality of the individual case before the court and show that it does not match the relevant rule of experience. Unlike with the reversal of the burden of proof adopted to implement the Damages Directive, defendants are not required fully to prove the opposite of the facts relied upon by the claimant. In any event, when combined with the statutory binding effect of a final decision of the European Commission or the German Federal Cartel Office, the claimant had a solid basis for successfully claiming damages.

Section III and IV look at how a recent decision by the German Federal Court of Justice significantly modified previous practice.

The judgment was issued in the context of the rail cartel, which was the subject of multiple decisions in the lower courts. The rail cartel involved eight undertakings convicted of dividing tenders or projects amongst themselves over the course of many years. The cartel mostly aimed at protecting longstanding customer relationships, and it was often predetermined which cartel member would win a tender.

The claimant – a local transport company – pursued a follow-on claim against one of the cartel members for overcharges. Both the court of first and second instance largely endorsed the claimant’s line of argument. The infringement was established as a consequence of the binding effect of the competition authority’s decision. The courts also applied the principle of prima facie evidence, and relied on a rule of experience according to which anticompetitive agreements involving hard-core restrictions increase prices and thus cause damages. The defendant argued that such an approach was not justified. They explained that the agreements did not increase prices, but merely served to ensure an even utilisation of the cartelists’ manufacturing plants. Further, the agreements covered only approximately 10 to 15% of total market share and were regional in nature, and as such unable to affect prevailing (competitive) prices. However, the lower courts dismissed such defences, finding that it is no precondition to cartel damage claims that the transaction at issue was subject to specific anticompetitive agreements; the crux of the matter was that cartel leads to overall price increases which the claimant could not escape.

The Federal Court of Justice, on appeal, rejected this approach. It acknowledged that economic experience exists with regard to the effects of cartels which means that they typically lead to increases in prices. However, the court emphasised that no two cartels are alike. Anticompetitive agreements are often very complex and multifaceted. Various factors will affect how a cartel will impact market prices, such as the number of market participants, the number of cartel members, the ability to exchange relevant information for the implementation of the agreement, market coverage, cartel discipline, and the ability of the customers to meet their demand through different providers. These factors may change throughout the life of a cartel, and become more relevant the longer the cartel exists. It must also be recognised that all participants ultimately pursue individual economic interests and are therefore not necessarily prepared to comply with cartel discipline. As a result, the implementation and effects of a cartel lacks the typicality required for the application of the principle of prima facie evidence. Instead, decisions finding cartel infringements may provide the basis for factual presumptions, i.e. they merely provide a strong indication of increased prices which must be considered when assessing the evidence.

Section V discusses these developments.

The authors note that it is unfortunate that the court raised the bar for claimants bringing claims for cartels preceding 2016 at the same time as the legislator introduced a statutory presumption of harm that, according to the explanatory notes of the new rules, were simply intended to replace the prima facie approach that had previously been established case-law.

On the other hand, one should not read too much into the decision. By abolishing prima facie evidence while, at the same time, stressing the existence of a factual presumption, the court is simply increasing the depth of analysis required of the facts of each case. In particular, the court required claimants to substantiate their claim based on the facts of the case and not based solely on mere economic theory detached from the case. Claimants will not have to start from scratch – given the factual presumption, claimants will have high chances of success to gain the judges’ conviction. This can be seen in some lower court decisions subsequently to the Supreme Federal Court’s judgment, which have replaced the words ‘prima facie evidence’ with ‘factual presumption’, but otherwise weighed the facts in the exact same manner.

The paper then concludes with some practical advice on how to bring and defend against claims for follow on damages given these developments. I do not believe this is relevant for our purposes here.


This is a very good introduction to the specifies of German law as regards bringing follow-on claims. In an article I circulated earlier, it was noted that there are currently over 630 pending claims before the German courts, a significant proportion of which I assume will be affected by the developments described in this paper. Furthermore, despite what the authors argue, it is hard to shake the feeling that the Supreme Federal Court’s decision will make it harder successfully to prove damages – which will likely lower the rate of success of such claims, which currently hovers around 75%.

A close analogue to the Supreme Federal Court approach may perhaps be found in the English High Court’s ABB v Britned decision I reviewed here. To an extent, these judgments may speak to some judicial discomfort with the underlying assumptions of competition law, and to how importing these assumptions into tort law may feel akin to attaching a mechanical prosthetic into an otherwise healthy body.

From a policy perspective, another relevant question raised by these decisions concerns the trade-off between adopting a methodology to calculate harm based on presumed average overcharges or based on the details of the specific relationship between the infringing parties. I do not want to express an opinion in favour of either approach – even if, by opting for a presumption of harm, the European legislator seems to tilt towards the former. The point I would like to make here is that I believe this trade-off is intrinsically connected with the availability of collective claim mechanisms as a valid and effective option to obtain compensation for competition harm. After all, pursuing a detailed analysis in a single collective claim, even if in a complex one, is unlikely to be more onerous than pursuing such an analysis in hundred or even thousands of individual cases – with the latter scenario supporting arguments for the adoption of presumptions of harm that facilitate compensation and lighten the burden imposed on courts.

Author Socials A weekly email with competition/antitrust updates. All opinions are mine

What do you think?

Note: Your email address will not be published