Despite the pandemic, 2020 has been a very busy year for the Federal Cartel Office (“FCO”) and courts in the field of competition law in Germany. The following is merely a selection of interesting developments. It does not include the recently adopted reform of the Competition Act, which merits its own blog.
Generally, the FCO has continued its vigorous cartel enforcement, even if overall fines decreased compared to the previous year, and dealt with interesting questions in merger cases. The digital economy played an important role throughout antitrust, merger and abuse of dominance cases. Private damages claims are increasing in Germany, while the Federal State Court has continued to clarify fundamental questions like applicable evidence rules.
Like other authorities, the FCO provided a lot of informal guidance to companies due to the challenges posed by the pandemic, notably regarding possible cooperation among competitors to ensure continuous supplies, but also on possibly restrictive agreements within supply chains. A good example is the guidance to VDA (German automotive industry association) on the cooperation of industry stakeholders regarding restructuring measures for suppliers in distress (here). The cooperation had to be limited in time, on a voluntary basis for suppliers, and the information exchange among stakeholders should be limited to what was necessary (both in substance and format, i.e., aggregated data only).
Cartel enforcement continued to be a top priority for the FCO. It issued fines in total of € 358 million on 19 companies and 24 individuals. The technical building services cartel (€ 110 million fine) concerned bid rigging, and the conduct inter alia involved to compensate the assigned “losers” through offering them subcontracting agreements for those bids ( see here). This is also a reminder that the line between subcontracting among competitors and a cartel may be fluid, and that a company’s practice to frequently or systematically engaging in subcontracting with competitors may merit closer internal review.
The aluminum forging cartel (€ 175 million fine) involved the exchange of detailed individual costs, the discussion of general cost developments as well as the agreement to pass on any cost increases to customers (see here). While the case primarily concerns price fixing on the sales side, it also reflects that agencies have shown increased interest in cartel conduct on the purchasing side – which companies may wish to reflect in their compliance efforts.
The plant protection cartel (€ 155 million fine) concerned price fixing at the wholesale level (see here). It is noteworthy that one of the cartelists, BayWa AG – following leniency cooperation and a settlement with the FCO – claimed civil damages from the FCO (the reduced fine plus legal costs), based on public liability claims: Following an anonymous complaint, the FCO contacted three selected companies in the industry, suggesting they submit leniency applications, but not BayWa or others. BayWa saw this as discrimination and a violation of its procedural rights. It argued that if it had also been contacted by the FCO, it would have applied for leniency earlier and received immunity. In contrast, the FCO referred to its investigatory discretion, noting that BayWa could have applied for leniency at any time. The court rejected the damages claim in substance. BayWa has announced to challenge the judgement. The case raises interesting questions on the authority’s investigative powers in cartel cases. It is unclear whether the FCO intends to continue selected “fishing for leniency applications” in the future.
In the beer cartel, on July 13, 2020, the Federal State Court (“BGH”) quashed a decision of the Düsseldorf Court of Appeals lifting a fine on Carlsberg and referred the case back to another court chamber for a final decision (see here). The BGH clarified the concept of concerted practice under German law: in line with EU law, if there is an exchange among competitors of information on possible future price increases, no additional element of agreement or certainty is required. The BGH also confirmed that there is a factual presumption that participants take this type of information into account in subsequent market conduct. This presumption must be reflected when assessing the facts in cartel proceedings, but falls short of a reversal of the burden of proof. The latter deviates from EU competition law, which presumes causality between contact and subsequent market conduct and shifts the burden on the participants to demonstrate that they did not consider the information exchanged. The BGH also found that a concerted practice is one single infringement including contact and subsequent market conduct, and that termination of the infringement (and beginning of the statute of limitations) only starts once the market conduct (here the price increase) ends.
2. Horizontal cooperation
Digital trading platforms. The FCO did not raise objections against B2B trading platforms Unamera for agricultural products, with three agricultural trading companies as platform shareholders (see press release of February 5, 2020, see here). Similarly, on May 14, 2020, the FCO did not oppose B2B trading platform for mineral oil products, operated by a joint venture between Shell and OnlineFuels, for short term spot market sales ( see here). The FCO clarified that these platforms generate efficiencies, but should neither lead to price fixing nor discrimination. The transparency generated through the platform should be limited, both between the platform and its shareholders, including through Chinese Walls, as well as between platform users that are competitors – regarding the type and extent of information exchanged, the publication of market statistics and the identity of the contracting partners to be revealed at a late stage in the purchasing process.
The FCO also greenlighted the centralized B2C-online sales platform of Intersport, a large group of independent medium-sized sports retailers active under the Intersport brand ( see here). The platform acts as seller and enters into contracts with consumers, setting the sales price. (If there is an order, the platform purchases the relevant product from an Intersport retailer, which sends it directly to the consumer.) The FCO concluded that this set-up only marginally restricts competition among Intersport retailers: online sales represent a minimal proportion of their overall sales, and they can decide at which price they are willing to sell to the platform. They remain free to engage in parallel online sales, through their own website or third-party platforms. The FCO noted that the platform was the only way for many of these retailers to engage in viable and competitive online activities, which in turn strengthened competition in online sports product sales. At the same time, the FCO clarified that the platform must be accessible to all Intersport retailers meeting the criteria, which must not be discriminatory.
3. Vertical agreements
Best price clauses. The FCO’s December 2015 decision to prohibit hotel reservation platform Booking.com to apply narrow best-price clauses (i.e., that participating hotels are not allowed to offer better prices on their own website) is still not final. Booking.com appealed, and the Düsseldorf Court of Appeals quashed the decision in June 2019 (see a blog on this here): it held that the narrow best-price clause restricted competition, but amounted to a legitimate ancillary restraint to the main contract between Booking.com and the hotels on the intermediation of hotel room bookings. That meant the clause would not even fall within the scope of Article 101 TFEU, and that there was no need to review whether an individual exemption under Article 101(3) TFEU would apply – which was the route the FCO (and all other NCAs reviewing Booking.com’s practices) had taken. The court also excluded further appeal to the Federal State Court (“BGH”).
The FCO requested the BGH to quash the ban on further judicial review. On June 14, 2020, the BGH granted the request, saying that the questions raised concern many digital platforms (see here). It held that the precedents regarding the narrow exception of ancillary restraints on which the OLG Düsseldorf relied seemed to concern different scenarios than the price restrictions in the current case, with the narrow best price clauses having similar effects to minimum price agreements, i.e., hardcore restrictions. The BGH noted that that the legality of such restrictions aimed at preventing free-riding in vertical relations has so far only been reviewed under Article 101(3) TFEU. The case is thus still ongoing – now the BGH needs to decide in main proceedings in substance.
In parallel, the FCO has published the results of an additional investigation into the effects of narrow best price clauses on online sales of online hotel platforms, which it carried out in the context of the appeal proceedings, and its own interpretation of these (see here). In short, the FCO says that free-riding is only a minimal problem for Booking.com, because the vast majority of consumers would not compare prices, but book a hotel on the first forum/platform where they found the hotel.
III. Merger control
The FCO clarified its approach on the assessment of state-owned companies from planned economies, adapted retail product market definitions in several cases with respect to including online sales, and further developed its practical approach to identify markets that may raise unilateral effects concerns, and assessed digital platform markets.
CRRC/Vossloh. The FCO cleared the acquisition of the German shunter producer Vossloh by Chinese state-owned rail vehicle manufacturer CRRC in second phase without commitments on April 27, 2020 (see here). The FCO raised concerns as Vossloh’s current market share in shunters in a European-wide market was between 40-50%. CRRC is primarily active in China, and has just entered the market, which is characterized by the longevity of the products, tenders and complex product certification processes, resulting in significant time laps between orders and actual supply. The FCO acknowledged that the past market shares were not necessarily meaningful for the future, given that Vossloh had not invested in technological developments and would thus likely lose market share, while CRRC’s development was difficult to predict.
The case is interesting as the FCO explains how it assesses state-owned companies from regulated, planned economies: for the purposes of German merger control, the entire group of any state-owned companies is relevant, i.e., here all companies owned by the state of China. (Unlike at EU level, this group approach is not limited to the level of the same state-owned holding company). That does not only impact the turnover thresholds, but all other information to be provided as part of the filing (e.g., up- and downstream or neighboring activities) and in response to information requests, which can render these exercises quite vast. The FCO used the stop-the-clock mechanism several times in the proceedings.
In addition, the FCO acknowledged that China’s ability to grant open or hidden subsidies is largely uncontrolled. Its state-owned players are thus viewed as having quasi unlimited financial resources, enabling them to heavily invest in technology. The planned-economy background means they can follow a long-term low-price strategy that is not necessarily only based on, and thus also not limited by, commercial reasons, and may serve as a major driver to gain market share. Ultimately, the FCO assessed various alternative scenarios of possible future market development absent and with the merger, and looked at a five to ten year period (instead of its typical three-year prognosis). Given that it could not find with requisite certainty that the merger would create dominance (or any significant coordinated or unilateral effects), it cleared the transaction.
Retail mergers. The FCO reviewed several mergers at retail level, including grocery, furniture stores, DIY, sports and outdoor apparel and bookshops. Regarding groceries, the FCO continued its established practice and market definition (stationary grocery sales only), and cleared Kaufland and Globus acquiring 92 and 24 former Real grocery stores, respectively, on December 22, 2020 (see here). The clearance was subject to commitments, i.e., divestitures of several stores (up-front-buyer solution) to alleviate concerns in individual local sales markets, but also in order to reduce the combined entity’s purchasing volume and thus power.
In the other retail areas, the FCO increasingly includes online sales channels into the relevant product market, which can facilitate mergers of retailers with predominant stationary activities. On February 10, 2020, the FCO cleared the acquisition of sports retailer Sportcheck by Signal Retail (see here), and on November 19, 2020, the largest stationary bookstore chain in Germany, Thalia got phase-I-clearance to acquire the Osiander chain (see here). The FCO concluded in both cases that the relevant product market comprised both stationary and online sales, but also reviewed these two segments separately.
Another example is the joint venture between furniture retailers XXXLutz and the Tessner Group companies Roller and Tejo Möbel, all primarily active in discount furniture stores. The FCO cleared it in a second phase decision on November 26, 2020, with a commitment to divest 23 retail outlets (up-front buyer solution) (see here). The merger combined the number two and four in the furniture retail industry in Germany. The FCO defined the product market as overall retail sales of a basic furniture portfolio, including full-portfolio retailers’ stationary and online sales, as well as specialized stationary trade with a partial portfolio (like kitchen, upholstery, etc.). Nevertheless, the FCO reviewed the stationary segment separately, and further distinguished segments by portfolio.
It is also interesting how the FCO approached the unilateral effects analysis in practice. The FCO applied alternative market-share-thresholds by segments, considered indicative to identify regional markets (by catchment area) that could significantly impede effective competition: The threshold were (i) in the stationary discount segment a combined share of 2/3 and an increment through the merger of medium single-digit percentage points; or (ii) in the overall stationary and online segment (a) involving all full-portfolio furniture retailers a combined share of 30%, or (b) involving all furniture retailers, including partial portfolio, a combined share of 25%, and an increment through the merger in the low to mid-single-digit percentage points. All the markets identified as problematic based on these thresholds were then subject to further individual analysis.
Digital platforms. The FCO reviewed several mergers involving digital platforms, and a recurring theme is the aim to keep these markets open for competition. The FCO cleared ProSiebenSat.1/Meet Group on July 6, 2020 (see here), combining leading online dating platforms Elite Partner and Parship with Lovoo. Despite finding that the merging parties had market power, the FCO cleared the transaction. It concluded that the online dating market in Germany is dynamic and growing, and stressed recent market entries, in particular by dating apps for mobile devices (like Tinder). There is multihoming, and the FCO also considered Facebook’s imminent market entry into online dating services.
On October 21, 2020, the FCO cleared Allianz/ControlExpert, with the insurer acquiring the leading provider of automated IT-based services to settle motor vehicle damages claims (see here). While there was no horizontal overlap, the FCO scrutinized whether the deal would financially and strategically strengthen ControlExport’s position in a way that it would become an indispensable service provider for other insurance companies, at a time when the provision of claim settlement services is at the brink of dramatic change towards digitization and the use of AI. However, the investigation revealed that other competitors have sufficient innovative potential to develop similar solutions.
IV. Abuse of dominance
Facebook. The BGH dealt with the FCO’s Facebook (“FB”) decision. In February 2019, the FCO found that FB abused its dominance in the social network platform market in Germany by making access to facebook.com conditional upon agreeing that FB could combine the user data collected on third party websites (and other FB websites and apps) with the data collected on the facebook.com platform (see a blog on this here). The FCO prohibited this practice FB (and imposed remedies to terminate the infringement within a certain time period). FB appealed the prohibition with the Düsseldorf Court of Appeals. In parallel, it requested in interim relief proceedings that the Court order the appeal to have suspensive effect, i.e., that FB would not need to comply with the prohibition pending appeal proceedings. The Court of Appeal granted FB’s request in August 2019, expressing serious doubts as to the legality of the FCO’s decision following a summary review.
Upon the FCO’s appeal, the BGH quashed the Court of Appeal’s interim relief decision on June 23, 2020 (see here). It found that there was “no doubt” that FB had a dominant position in the social network market in Germany and abused it through the relevant user conditions. However, the BGH expressly deviated from the FCO’s conclusion that the infringement of data protection law through the user conditions led to an abuse of dominance. In contrast, the BGH found abusive that FB did not offer users any choice regarding access to the platform. The BGH concluded that under competitive conditions, FB would presumably offer an alternative, less restrictive version, i.e., to be able to use facebook.com and enjoy the promised “personalized experience” only based on the user data collected on the very same platform – given that there was apparently consumer demand for that.
The BGH qualified FB’s current practice as abusive exploitation of consumers in terms of abusive conditions. It clarified that it is sufficient to show a causal link between dominance and the abusive market result, rather than the actual market conduct. (This addresses the argument that other non-dominant platforms may apply similar user conditions.) The BGH also held that in the context of platforms, there may be causality if the exploitation is capable of impeding competition in the dominated market as well as on the other platform side (a third market). The BGH concluded that in the current case FB’s conduct might strengthen the lock-in effect for consumers to facebook.com, and that this might foreclose competing social networks in Germany, which require access to user data. It could also not exclude negative effects on competition in online advertising, through which social network platforms are typically financed. The BGH decision is remarkable because it includes rather bold language, some “final” findings and the BGH replaced the FCO’s reasoning on the abuse with its own, different assessment, which is unusual for interim relief proceedings with only a summary review.
FB filed another request for interim relief and on November 30, 2020, this time focusing on the FCO’s remedy decision to change user conditions. At the same time, it requested the Court to order suspensive effect for this (new) action, to prevent the implementation period for the remedies to start. The Court issued a “temporary” interim suspension until it would decide upon the new interim relief request (in March 2021), and excluded further judicial review (see here). The FCO filed a request to lift that judicial review ban with the BGH on December 2, 2020, which the BGH granted on December 15, 2020 (see here).
The saga will continue, and the apparent stand-off between the Court of Appeals and the BGH is remarkable. The case illustrates the complexity of abuse of dominance enforcement in digital markets in practice. Supposedly, the case triggered a last-minute change for the FCO’s new prohibition power vis-à-vis digital gatekeepers under the recently adopted competition law reform: judicial review of decisions will be limited to one single instance with the BGH (i.e., skipping the Court of Appeals).
Facebook – Oculus. The FCO has opened new abuse of dominance proceedings into FB’s plan to link the use of its Oculus virtual reality (“VR”) products to its dominant social network facebook.com on December 10, 2020 (see here). Following the integration of FB’s previous stand-alone VR platform Oculus into the social network platform, the use of the latest Oculus VR glasses will require registration to facebook.com.
Amazon. The FCO reportedly opened new abuse of dominance proceedings against Amazon because of the allegation that Amazon intervened against “price gouging” of sellers on its platform. The allegations first occurred in the context of Covid 19, but continued afterwards. The FCO reportedly reviews whether Amazon generally engages in price controlling mechanisms.
V. Sector inquiries
The FCO finalized sector inquiries under its consumer protection powers and published final reports on smart TVs in June 2020 (see here), mainly focusing on the issues of personal data collection by smart TVs, their usage and to which the extent consumers are informed about this. In October 2020, it published the final report on online user reviews (see here), explaining the functioning of user review systems, the risk of non-authentic reviews in certain systems and possible consumer protection law violations. The FCO is still conducting its antitrust sector inquiry into online advertising (started in 2018).
VI. Private enforcement
The BGH rendered several landmark judgments in 2020, clarifying mainly questions regarding follow-on damages litigation on evidence rules, legal standing, the role of economic opinions, joint and several liability and umbrella damages and passing on. The cases all concern “old” law, i.e., prior to the implementation of the EU private damages directive, given that the cases concern potential damages that occurred before 2017. The BGH mainly issued judgments regarding follow-on damages claims regarding the tracks cartel (Schienenkartell). (For its ruling on the trucks cartel, see the blog by Thomas Thiede here).
In Schienenkartel II (decision of January 28, 2020, see here) it reiterated that there is no prima facie evidence that a cartel leads to higher prices, but only a factual presumption in that regard, thereby quashing a general practice at lower civil courts throughout Germany. The claimants (and courts) cannot any longer rely on an abstract evidence rule that – absent any extraordinary circumstances – the cartel led to increased prices. They need to fully prove that the cartel led to increased price. In assessing whether the cartel caused any damages, the courts need to examine all relevant facts of each case. In that exercise, they can consider the said factual presumption, which becomes stronger the longer the duration of a cartel and the broader its scope (market coverage). In practice, it remains to be seen whether this will affect the actual outcome or rather the amount of reasoning and work to be included in claims and the judgments.
The BGH also reversed and clarified previous caselaw that for customers of cartelists to establish legal standing for bringing damages claims, it is sufficient to show that the cartelists engaged in conduct that is capable of leading to damages through subsequent sales or other means for direct or indirect customers. Claimants do not need to show on top that each product purchase in the damages claim was individually affected by the cartel infringement. This is rather a question of the amount of damages, which can ultimately be estimated by the court.
While a court is relatively free in estimating the damages amount, the BGH stressed that this exercise needs to be based on facts, to be set out in the judgment. It is not mandatory that the court appoints an economic expert for the estimate: expert opinions do not account as evidence for the existence or lack of damages, as they only provide estimates, and they cannot replace the court’s own estimate. Nevertheless, the court needs to deal with any private expert economic opinions submitted by the parties in its assessment.
In Schienenkartell III (decision of May 19, 2020, see here) the BGH held that in a single continuous infringement, there is joint and several liability of all cartelists for all resulting damages. The FCO qualified the tracks cartel as a single infringement: a basic quota agreement between all cartelists, which was implemented in numerous subsequent projects. The BGH held that even if some cartelists were not involved in a particular project during the cartel period for which damages were sought, they would still be liable for the damages caused. The BGH thus aligned the rules on civil liability with the principles on infringement participation in cartel proceedings.
In Schienenkartell IV (decision of May 19, 2020, see here), the BGH confirmed that cartelists may be liable for umbrella damages, i.e., based on purchases from cartel outsiders. There is equally no prima facie evidence rule that a cartel leads to umbrella damages, but a complex assessment of all facts is necessary. Relevant aspects include duration and coverage of the cartel, product homogeneity, market transparency, the possibility of cartel outsiders to expand capacity, competition among cartel outsiders, buyer power, etc. On the passing-on-defense, the BGH reiterated that defendants need to prove that a claimant engaged in pass-on of all or part of the damages to its own customers. It concluded that there is no duty for the defendant to disclose details of its own price setting, if in the case any pass-on would likely have been marginal, and it is unlikely that the downstream market side would claim damages from the defendant (e.g., because they are consumers who suffered small damages only).