The German Federal Cartel Office (FCO) continues to be an active, sophisticated and efficient enforcement authority. Fines rose to €848 million overall in 2019, in six different cartels. Against the background of the steep rise in private follow-on cartel damages claims, the number of leniency applications and dawn raids has declined. Most cartel cases are still settled rather than appealed as the FCO and undertakings both aim to avoid lengthy and burdensome oral hearings in court. Undertakings may also fear that their fine could be substantially increased by the Appeals Court. In an important development for the balance of power in settlement negotiations, two Appeals Court judgments that had increased fines imposed by the FCO have been annulled by the Federal Court of Justice. In turn, the Appeals Court has rendered an important judgment annulling a fine on a member of the beer cartel for lapse of time. The imminent next amendment of the German Act against Restraints of Competition will also change cartel procedure.
- Trends in cartel enforcement at FCO, leniency and settlements
- Appeals Court proceedings and their impact on undertakings’ willingness to settle or to appeal
- Damages claim against FCO for alleged discrimination in leniency procedure
- Impact of pending changes to procedural law in antitrust cases
Referenced in this article
- Confectionery cartel proceedings
- Liquid gas cartel proceedings
- Rossmann case
- Beer cartel proceedings
- Sausage gap
- 10th amendment to the German Act against Restraints of Competition
Germany’s Federal Cartel Office (FCO) is a very active, efficient and sophisticated enforcement authority which has proactively picked up various matters and, not infrequently, has been a frontrunner in Europe with its enforcement. It is operating in a legal setting where substantive and procedural laws are reviewed and amended regularly. There have been three fundamental reforms of the German Act against Restraints of Competition (GWB) since 2005.1
Because antitrust fines can only be based on substantive law that was in force at the time the competition law infringement was committed, most of the cases pending before or decided by German courts and the FCO nowadays are still dealing with former versions of the GWB. To understand the enforcement trends and precedents of the reporting period, some background information, therefore, is essential (see ‘Background – particularities of German cartel enforcement procedure’). We then touch on numbers and trends in fining decisions issued by the FCO (see ‘Proceedings and fines at FCO level – number crunching’) and leniency applications (see ‘Leniency – flattening the curve?’). Furthermore, we highlight the most relevant cartel fines cases that have gone to appeal (see ‘Appeal of fine decisions – no risk, no fun?’). We then report on a damages claim against the FCO for allegedly having triggered leniency applications in a discriminatory way (see ‘Action for damages against the FCO for triggering of leniency applications’) and the state of the famous sausage gap (see ‘Sausage gap closed’).
Finally, another major reform affecting cartel enforcement is imminent with the 10th amendment to the GWB (referred to as the GWB Digitalisation Act).2 The draft bill is still under discussion but the legislative process is expected to be completed in spring 2021 at the latest. We therefore also touch on what to expect in cartel procedure in the coming years (see ‘Pending reform of the GWB and the effects on cartel proceedings’).
Background – particularities of German cartel enforcement procedure
German cartel enforcement – be it enforcement of article 101 of the Treaty on the Functioning of the European Union, or the German equivalent in section 1 of the GWB – follows the procedural rules of criminal enforcement. Other than the European Commission, the FCO needs to address (and may fine) at least one individual in a leading position who has either been participating in the cartel activity or has (negligently or intentionally) not fulfilled supervision obligations (instruction, organisation and control) thereby facilitating an infringement committed by an individual acting for or on behalf of the undertaking. No fine can be imposed if the individual’s cartel conduct cannot be identified. In practice, the FCO fines between one and three individuals from senior management as well as the undertaking (provided that their involvement can be proven).
If a company files for leniency, it usually includes those individuals who have acted for the company, thereby extending the leniency benefits to them. Any individual can, in principle, blow the whistle and request leniency just for himself or herself, although this rarely happens.
If the cartel activity concerns tenders, individuals involved are subject to criminal sanctions according to the Criminal Code, instead of just an administrative fine. In these types of cases, a leniency application will not relieve the individuals of pending criminal sanctions. Individuals’ contributions to the cooperation are likely to be taken into account, however, when determining the criminal sanction for the individual concerned. If an individual is to be criminally prosecuted, the FCO will go after the undertaking and the criminal prosecution services will target the individual; both authorities can work in parallel or consecutively. A recent example is the technical building services cartel, in which the FCO ended its proceedings by imposing fines on undertakings while the criminal proceedings pursued by the prosecution service in parallel against the individuals has not yet been concluded.
Only since June 2005 has the FCO had the power to impose sanctions of up to 10 per cent of the overall group turnover. Before then, it could impose sanctions above €1 million against undertakings only, if it could prove a ‘cartel surcharge’. In practice, this has mostly induced the FCO not to sanction cartel conduct before 2005, to avoid the (massive) burden of proving the cartel surcharge. That topic is nowadays only relevant for private enforcement, where it is key.
Infringement decisions of the FCO can only be contested in front of the Higher Regional Court of Düsseldorf’s specialist cartel senates (the Appeals Court). Unlike the European Union courts, the Appeals Court will not just review whether the fining decision was justified and whether the FCO has stayed within the scope of its discretion. Instead the Appeals Court has to apply the standard of criminal proceedings and hence has to establish every single fact in a public hearing – including full witness hearings. Cartel appeal trials are therefore quite burdensome on all parties. They usually see between 20 and 50 days in court – in extreme cases, they can take up more than 100 hearings3 (former examples include 136 days in the liquid gas cartel).4 This procedural situation combined with the possibility of the Appeals Court to increase the fine (reformation in peius, see ‘Appeal of fine decisions – no risk, no fun?’, below) has been a strong incentive for undertakings to settle rather than to take the case to court or to withdraw the appeal. A prominent example is the beer cartel, in which the FCO had fined six undertakings a combined total of €336 million. Two undertakings appealed. One of them, Radeberger brewery, withdrew its appeal against a €154 million fine the day before the hearing was scheduled, for fear that its fine could soar to €1 billion, should its appeal not succeed.5
In the following sections, we highlight some of these cases and exceptions to that rule. The examples explain how these cases have affected the ‘balance of power’ between undertakings and the FCO when negotiating settlements and deciding whether to appeal the case.
Proceedings and fines at FCO level – number crunching
While the number of behavioural proceedings appears to have remained more or less stable in the past few years,6 the number of typical hard core proceedings in which dawn raids have been executed, and the number of leniency applications, have dropped. The overall fines peaked at a total of €848 million in 2019, involving six cartel cases and 23 undertakings:
|Dawn raids (number of undertakings involved)||11 (69)||7 (51)||5 (32)|
|Proceedings closed with fines||7||4||6|
|Overall fines imposed (€)||60 million||376 million||848 million|
The names of individuals fined and the amounts of the fines imposed on them are not published by the FCO. Hence, specific up-to-date data from the reporting period is not available. However, from FCO statistics, it can be established that between 2008 and 2015, the FCO fined 328 individuals with an overall volume of €24.3 million, the individual fines ranging from €500 to €800.000.7
Among the cases closed in 2019, the largest overall fine by far (€646 million) was imposed on steel manufacturers for cartelising the production of quarto plates.8 Other fines imposed in 2019 include those for the purchasing of steel for automotive production (€100 million) and those for steel production – for the automotive industry. The FCO furthermore imposed fines totalling €110 million on the 12 members of the technical building services cartel, for colluding on bids for major contracts.9 Smaller fines have been imposed, inter alia, on a horizontal cartel between providers of magazine lending services (€8 million)10 and a bicycle wholesaler for vertical price fixing (€13.4 million).11
Like many other antitrust authorities, the FCO offers a settlement bonus of 10 per cent. The vast majority of cases – virtually all that do not go on appeal – will therefore be ended by way of settlement. Owing to various factors, undertakings perceive ‘not settling’ as a risky option – in particular against the background that the Appeals Court has significantly increased the fines in a number of cases (see ‘Appeal of fine decisions – no risk, no fun?’, below); furthermore, a settlement decision is relatively short compared to an ordinary fining decision and may therefore provide less information for cartel damages claimants. Hence, an undertaking will endeavour to appeal only (and not settle) if it sees a substantial chance that it could escape the fine altogether – for example, because the statute of limitation has lapsed. Owing to the burden of the appeals procedure with a large number of oral hearing days, the FCO, in turn, has a strong incentive to settle, so as to spend its resources on pursuing cartels rather than on defending its decisions in lengthy proceedings. While an appeal may result in higher fines, if the Appeals Court finds the undertakings to be guilty, more recent procedural history has shown that the Appeals Court also sides with the applicants on substance in some cases (see also ‘Appeal of fine decisions – no risk, no fun?’).
Leniency – flattening the curve?
For two decades, the FCO has employed a Leniency Programme (inaugurated in 2000 and overhauled in 2006). Private practitioners have frequently criticised a somewhat opaque practice that differs from case to case, depending on which unit within the FCO deals with the case, while also praising the flexibility of the FCO. In specific cases, the impression could arise that leniency applications received after dawn raids (ie, of applicants ranked 2 et seq.) may still lead to comparatively high reductions. This has triggered the speculation that the FCO may have deliberately wanted to discourage perpetrators from seeking refuge to the Appeals Court and instead to enter into a settlement, thereby closing the case at that level. This type of speculation may or may not have its grounds but undertakings who have just done their best to reduce their fine, will certainly not complain and the FCO legitimately has no interest in having its resources tied up by many days in court. In some respects, the pending reform of the law is likely to lead to a higher degree of legal certainty for leniency applicants by incorporating the cornerstones of the leniency programme into the law.
It seems that the dust stirred up by the European Court of Justice’s Pfleiderer decision has settled, at least in Germany. Since 2011, when the judgment was passed, the number of leniency applications to the FCO increased steadily from year to year, peaking in 2015 (76 applications). Since then, it has been constantly decreasing: 59 in 2016; 37 in 2017; 21 in 2018; 16 in 2019. This is most likely because of the reverse trend of rapidly increasing numbers of (successful) private damages claims. These were sparked by legal amendments some years ago, one being the rule that in a damages action, a civil court may not question whether a company participated in an antitrust infringement if the FCO has held so.
While the FCO will not pass on leniency applications and accompanying material to third parties seeking to pursue damages claims against cartelists, it will hand out the infringement decisions. These decisions may be relatively short (with often only a couple of pages on the infringement itself) if an undertaking enters into a settlement with the FCO. Other than a settlement, these decisions are much longer and contain much more information on the infringement itself; whether these are helpful to quantify damages, for example, is a quite different matter. The number of applications to the FCO to hand out infringement decisions has risen significantly during the past few years. While Parliament has laid the legal basis for damaged parties to claim information from cartelists, leaving access to the FCO’s files only as a last resort, in 2017, it is currently a matter of dispute whether this new rule applies only to damages actions that are based on infringements that occurred after 27 December 2016. Despite these developments, still more than half of all cartel proceedings are triggered by information from leniency applicants.
Appeal of fine decisions – no risk, no fun?
During the past few years, some fine decisions issued by the FCO have been appealed to the Higher Regional Court of Düsseldorf (Appeals Court). The Appeals Court has – after conducting its own proceedings – substantially increased the overall fines in various cases. Famous decisions that led to significantly higher fines than those imposed by the FCO include liquid gas and wallpaper (overall increase to more than 130 per cent of the initial fines),12 confectionary products (overall increase to 150 per cent),13 and the drug store chain Rossmann (600 per cent).14 This reformatio in peius is compliant with German constitutional law and stems from the specific German procedural system, according to which the Appeals Court assesses all facts of the case from scratch and decides on a possible fine completely independently of the fine imposed by the FCO.
Nevertheless, those decisions of the Appeals Court increasing the fines to such an extent have been widely (and harshly) criticised by the industry and private practice, as they have been perceived to de facto hinder companies in appealing against the FCO’s fine decisions (and at an earlier stage urge them into settlement agreements with the authority or to withdraw the appeal). Moreover, whereas the FCO has issued (self-binding) guidelines for the setting of fines, the Appeals Court is not bound by them and only has to take into account that fines cannot exceed 10 per cent of the company’s group global turnover. Criticism has been voiced to the effect that appealing a fine decision would bring a systemic change of the calculation15 of the fine and therefore automatically lead to an increase of it.
Yet, the Appeals Court has also significantly reduced or even cancelled fines (eg, for lapse of time). Some of these decisions may not have received the same attention as those raising the fines.16 However, there have recently been some important decisions backing the fined undertakings’ points of view, which have been perceived as a slight counter trend. These decisions show that the Appeals Court, and on further appeal the Federal Court of Justice (FCJ), takes a close look at fine decisions and that appealing can in fact be helpful if the fined companies have substantive arguments apart from the calculation of the fine. These developments evidence that appeals to cartel proceedings in Germany are alive and kicking.
For instance, the FCO has been overruled by the Appeals Court in the beer cartel. This is one of the rare cases in which the undertaking concerned (Carlsberg) firmly held that it had not participated in the cartel at all, although a couple of leniency applicants had settled on the basis that Carlsberg had been involved in their admitted wrongdoing. After 20 days of oral hearing, the prosecutor requested that the fine be raised from €65 million to €250 million; however, the Appeals Court cancelled the fine. The Court was of the opinion that the prosecutor (the FCO) could, if at all, only prove Carlsberg’s participation in a mere information exchange. Owing to the statute of limitations, this could no longer be prosecuted. The case is presently under appeal to the FCJ.
Fines have also been reduced (from €18.4 million to €5.1 million) by the Appeals Court in November 2019 in a case concerning agreements between a beer manufacturer and its retailers, following an in-court settlement between the companies and the public prosecutor.17
Whereas the Appeals Court has substantially increased most of the fines imposed by the FCO in the confectionary proceedings,18 the FCJ overruled the Appeals Court and remanded the proceedings to a different senate of the Appeals Court in June 2019. The FCJ held in particular that the Appeals Court did not sufficiently consider the defending statements of the appealing companies but, in contrast, too uncritically followed statements of crown witnesses.19 The case is of specific interest to all industries and receives major attention as it concerns the limits of a lawful information exchange (in particular in meetings of industry associations), which has significant importance beyond the confectionary industry. In this case, the FCO had followed a very broad concept of the ban on cartels, which reflects paradigmatically the authority’s increased vigorous competition law enforcement during the past few years.
Likewise, the FCJ has quashed the Appeals Court’s judgment in the Rossmann case on the basis of a purely procedural flaw within the appeals procedure. This decision underlines the relevance of the German courts applying criminal procedural rules, which are not to be taken lightly when deciding on cartel fines. The case been remanded to a different senate of the Appeals Court,20 which has to fully try the case from scratch, including hearing all witnesses again.
In the liquid gas cartel case, the cartelists had agreed not to actively solicit customers and not to make (attractive) offers to other cartelists’ current liquid gas customers. Because the cartel took place before 2005, the legal framework in force before 2005 applies for a large part of the sanction. Hence, the fine of €180 million imposed by the FCO had to be largely based on a cartel surcharge and could not rely on the group turnover. On that basis, the Appeals Court imposed an increased fine of €244 million on the six undertakings concerned.21 On further appeal, the FCJ has confirmed that the six undertakings had formed an illegal cartel, but considered the Appeals Court’s fine calculation based on the cartel surcharge to be faulty.22 It annulled the judgment and sent the case back to be tried before a different senate at the Appeals Court, where the case is currently pending. While the elaborations on the cartel surcharge calculation are more or less legal history for the purpose of calculating fines (as no further cartel fines will be based on the cartel surcharge), these could still be interesting for cartel damages claimants in the future. Even though procedural rules for establishing the cartel surcharge may vary between a civil procedure and a quasi-criminal procedure, plaintiffs still have to prove the cartel surcharge as well so as to be rewarded cartel damages.
Action for damages against FCO for triggering leniency applications
In March 2020, a cartelist filed an action against the FCO to a civil court seeking damages amounting to €73 million. The claimant, who was fined by the FCO in 2020 for anticompetitive agreements, asserts that the FCO tipped off other cartelists, thereby placing them in a better position to apply for leniency.
The matter started at the end of 2014, when the FCO received an anonymous complaint. At the beginning of 2015, an FCO official contacted three companies by phone, two of which filed for leniency within 24 hours. The official later stated that he called the three companies because he knew them from a merger case he had been dealing with. He did not call any other company involved in the cartel. The first applicant received immunity from fines, and the fine of the second company was substantially reduced because of its leniency application. Following a dawn raid in March 2015, most of the other cartelists also filed for leniency. Nearly all of them, including the claimant, admitted having violated antitrust law and agreed to a settlement. The claimant now bases its damages claim (approximately equal to the €70 million fine already paid and other costs) on the accusation that the FCO gave preferential treatment to the three companies and thus violated the German constitution. This case is completely novel in German antitrust law. And, if proven true, it may add awkward aspects. The files of the FCO are reported to contain a memo drafted by the official about his phone calls to the three companies, which allegedly was drafted one month after the calls, but backdated as if it had been drafted only two weeks afterwards. The FCO initiated internal investigations and an individual was withdrawn from the case team.
Sausage gap closed
The infamous German sausage gap seems to be (finally) closed as a result of amendments to the law in June 2017, which introduced a parental liability corresponding to the European role model.
The sausage gap derives its name from a case in which the FCO prosecuted a cartel of sausage manufacturers but was not able to enforce fines of approximately €238 million owing to internal restructurings of some of the cartelists. The gap resulted from the fact that the FCO was entitled to impose a fine not on groups of companies but only on the legal person who by acts of its employees had committed the infringement – similar to the principle of criminal law that one family member cannot be imprisoned for a crime committed by another member of the family. As a consequence, several groups of companies were able to avoid liability by making the infringing legal entity disappear while shifting the assets to another group company; if the family member who committed the crime dies, the prosecutor will stop the prosecution and the court will not ask the deceased, or the deceased’s family, to serve the sentence. The restructuring was not as simple as it sounds (for example, it was not possible to go free by shifting the assets to another subsidiary, which basically continued exactly the same economic activity of the company that had been involved in the cartel), but in several cases – such as the sausage cartel – it worked. In 2013, the legislator narrowed down the gap but still left a loophole. Now, the liability for fines is extended to the economic unit to which the infringing company belongs (ie, to the group of companies). This has been the legal situation in EU law all along. So it can be stated that at least for infringements committed after 2017, the sausage gap is closed. Battles that are fought now are circulating around pre-2017 infringements, with proceedings against various companies being terminated by the FCO this year for ‘reasons of discretion’ (or, in fact, the unenforceability of the fines).
One of these old and as yet unsettled cases is pending with the European Commission. When – after three years of pursuing the case – the FCO detected that several metal packaging manufacturers subject to the proceedings undertook internal restructuring, it found ‘increasing evidence that the cartel was not limited to Germany but also concerned other EU Member States’. It referred the investigation to the Commission, which does not need to be concerned about the sausage gap that does not apply under EU law. One undertaking’s challenge of the Commission’s decision to initiate the infringement proceeding was rejected by the European Court of Justice at the beginning of 2020.23 Thus, the investigation is still pending with the Commission.
Pending reform of the GWB and the effects on cartel proceedings
As already mentioned, the legislative process of enacting the 10th amendment to the GWB has been started. In January 2020, the Federal Ministry for Economic Affairs published a draft bill that proposes new provisions of the GWB. Besides implementing the ECN+ Directive (EU) 2019/1 (which entails, inter alia, an expansion of the procedural rights of antitrust authorities), the reform shall bring major changes for the digital economy and new provisions regarding fines, damage claims and merger control.
The draft reform provides, among other things, for the following important changes affecting cartel procedures and, overall, providing the FCO with more powers than it has at present:
- codification of some cornerstones of fine calculations as binding for the FCO as well as the Appeals Court, which are albeit unlikely to completely remove the different approaches in setting the fine;
- codification of cornerstones of the leniency programme, making it binding for the Appeals Court as well;
- specific rules for fines on associations, and the possibility to have them eventually enforced against an association’s members;
- additional investigative powers for the FCO (aligning with EU law), while still not granting full ‘legal privilege’ for attorneys’ correspondence; and
- extending the statute of absolute limitation to the detriment of undertakings under investigation.
The draft is still under debate and it remains to be seen which of the changes will ultimately find their way into the GWB.
1 7th amendment to the German Act against Restraints of Competition (Gesetz gegen Wettbewerbsbeschränkungen [GWB]) of 2005 (inter alia, taking account of the changes prompted by Regulation 1/2003), 8th amendment to the GWB of 2013 and 9th amendment to the GWB of 2017 (inter alia, taking account of the changes prompted by the Damages Directive 2014/104/EU).
2 10th amendment to the GWB (implementing, inter alia, the ECN+ Directive (EU) 2019/1, and making amendments to account for the specific challenges to antitrust by the digitalisation).
3 Ost/Breuer, NZKart 2019, p. 119 et seq.
6 39 in 2017 and 40 in 2018 – cf. ‘Tätigkeitsbericht des Bundeskartellamtes 2017/2018’, pp. 135 and 136, at https://www.bundeskartellamt.de/SharedDocs/Publikation/DE/Taetigkeitsberichte/Bundeskartellamt%20-%20T%C3%A4tigkeitsbericht%202017_2018.pdf?__blob=publicationFile&v=6; number for 2019 not yet published.
7 ‘Tätigkeitsbericht des Bundeskartellamtes 2015/2016’, pp 32 to 34, at https://www.bundeskartellamt.de/SharedDocs/Publikation/DE/Taetigkeitsberichte/
8 FCO, press release, 12 December 2019, at https://www.bundeskartellamt.de/SharedDocs/Meldung/EN/Pressemitteilungen/2019/12_12_2019_Quartobleche.html.
9 FCO, press release 27 March 2020, at https://www.bundeskartellamt.de/SharedDocs/Meldung/EN/Pressemitteilungen/2020/11_03_2020_TGA.html?nn=3591568.
10 FCO, press release 13 February 2020, at https://www.bundeskartellamt.de/SharedDocs/Meldung/EN/Pressemitteilungen/2019/13_02_2019_Lesezirkel.html.
11 FCO, press release 29 January 2019, at https://www.bundeskartellamt.de/SharedDocs/Publikation/EN/Pressemitteilungen/2019/29_01_2019_Fahrrad.pdf?__blob=publicationFile&v=2.
15 Mäger, NZKart 2019 p. 361 et seq.
16 Ost/Breuer, NZKart 2019, p. 119 et seq. – the authors from the FCO emphasise that fine reductions had gone unnoticed.
21 FCO, press release, 16 April 2013, at https://www.bundeskartellamt.de/SharedDocs/Meldung/DE/Pressemitteilungen/2013/16_04_2013_Fluessiggaskartell-OLG.html; Higher Regional Court of Düsseldorf, Decision of 15 April 2013, Az. VI-4 Kart 2 - 6/10 (OWi).
22 BGH, Decision of 9 October 2018 – KRB 51/16, at https://www.juve.de/nachrichten/verfahren/2019/01/bussen-falsch-berechnet-olg-duesseldorf-muss-verfahren-gegen-fluessiggaskartellanten-neu-aufrollen.
23 European Court of Justice, Decision of 29 January 2020, Case C-418/19 – Silgan, at http://curia.europa.eu/juris/document/document.jsf?text=&docid=223088&pageIndex=0&doclang=