Even before covid-19 disrupted the world as we knew it, competition law was at a crossroads, facing far-reaching and sometimes contradictory calls for reform – including with respect to monopolisation and abuse of dominance.
Some, such as President Macron and Chancellor Merkel, have argued that there is too much competition from abroad, and advocate for more permissive enforcement to facilitate 'European champions' to emerge: 'We need to adapt the EU competition law: [It's] too focused on consumer rights and not enough on EU champions' rights.'
Others maintain that there is too little competition, enforcement has been too permissive, and the rules should be tightened. Senator Elizabeth Warren, for example, has argued that 'competition is dying. Consolidation and concentration are on the rise in sector after sector. Concentration threatens our markets, threatens our economy, and threatens our democracy. Evidence of the problem is everywhere.' Similarly, Professor Joseph Stiglitz contends that 'current antitrust laws, as they are enforced and have been interpreted, are not up to the task of ensuring a competitive marketplace'.
A third set of commentators believes that competition policy is misdirected, that the historic focus of competition law has been too narrow, and that the consumer welfare standard should be expanded to take account of social, industrial, environmental, and other considerations (sometimes referred to as 'hipster antitrust').
And a fourth critique, voiced by Maurice Stucke and Ariel Ezrachi, maintains that many of today's problems result from too much 'toxic' competition overall, driven by ideologues, lobbyists, and privatisation, and that we need to promote a kind of 'noble competition', where rivals mutually strive for excellence.
To address these challenges, a dizzying array of reports has emerged commissioned by governments in the US, EU, UK, Germany, France, Australia and elsewhere. And from those reports, a constellation of ideas has emerged to overhaul competition law, including: reorientating the goals of antitrust policy away from the consumer welfare standard towards a broader societal test; reversing the burden of proof; per se bans on certain categories of conduct (including prophylactic controls on vertical integration); lowering the standard of judicial review; injecting political oversight into competition law enforcement; loosening the standard to impose duties to share data with rivals; introducing market study regimes; allowing authorities to impose remedies without formally establishing an infringement; and establishing mandatory codes of conduct for digital platforms.
Where does this all leave busy practitioners and businesses that are trying to navigate the complex and constantly-evolving rules concerning abuse of dominance? Helpfully, this eighth edition of The Dominance and Monopolies Review seeks to provide some respite, providing an accessible and easily-understandable summary of global abuse of dominance rules. As with previous years, each chapter – authored by specialist local experts – summarises the abuse of dominance rules in a jurisdiction; provides a review of the regime's enforcement activity in the past year; and sets out a prediction for future developments. From those thoughtful contributions, we identify three notable points from last year's enforcement.
Exploitative abuses pre- and post-covid-19
Exploitative abuses have in recent years enjoyed somewhat increased attention from regulators. The covid-19 pandemic intensifies that trend. It is leading to extreme demand and price volatility for certain products, as well as fluctuations in firms' costs. As firms struggle to manage these changes, agencies are aggressively seeking to show they are preventing consumer exploitation during the crisis. Charging excessive prices or imposing unfair terms and conditions constitutes an abuse of dominance in many countries, including almost all OECD members. In the US, excessive prices are not in and of themselves a matter for competition enforcement at the federal level, but many states have laws that prohibit price gouging and the current administration recently issued an executive order designed to prevent hoarding and price gouging.
Governments across the world have indicated that they will remain vigilant to sudden and significant price hikes during the pandemic. For example, in March 2020 the European Competition Network issued a statement identifying excessive pricing as a particular concern during the outbreak, noting that 'it is of utmost importance to ensure that products considered essential to protect the health of consumers in the current situation (e.g., face masks and sanitising gel) remain available at competitive prices'. In a similar vein, on 27 March, Commissioner Vestager explained that 'a crisis is not a shield against competition law enforcement' and that the European Commission (EC) 'will stay even more vigilant than in normal times if there is a risk of virus-profiteering'. Several national authorities have opened investigations or created task forces dedicated to preventing excessive prices during the crisis.1
Even before covid-19, however, EU agencies were increasingly pursuing exploitation theories. In 2016, Commissioner Vestager stressed that the EC would seek to 'intervene directly to correct excessively high prices'. So far, most recent exploitation cases have been in the pharmaceutical sector, but the French and German agencies have pursued exploitative abuse theories in the technology sector. We pick out four developments over the last year.
First, the Court of Appeal judgment in Pfizer/Flynn, discussed in the UK chapter of this book, brings helpful clarity to evidence required to bring an excessive pricing case. As a recap: in 2016, the Competition and Markets Authority (CMA) imposed record fines on Pfizer and Flynn for charging excessive prices for phenytoin sodium capsules, an anti-epileptic drug. In July 2018, that decision was quashed by the Competition Appeal Tribunal (CAT) on the basis that the CMA had applied the wrong legal test and had failed to consider appropriately the economic value of the product. In March 2020, the Court of Appeal upheld the CAT's judgment that the case should be remitted to the CMA, though it agreed with the CMA on some issues (which will affect the remitted investigation) and the CMA welcomed the judgment as a 'good result.'
In a nutshell, the Court of Appeal held that competition agencies have a 'margin of manoeuvre' in deciding how to prove their cases, including the 'Cost Plus' method that the CMA had used. Importantly, though, if a defendant adduces evidence that challenges the agency's methodology (as the defendants did in this case), the agency should consider that evidence. The extent of the agency's duty to consider the evidence adduced by the defendant will depend on the extent and quality of the evidence (i.e., there is no need to investigate each and every claim the parties bring up if those claims are not sufficiently substantiated). On the facts of the case, the Court held that there was an obligation on the CMA to evaluate the defendants' evidence regarding the prices of phenytoin capsules because it was prima facie evidence that prices were fair.
Second, in the Sanicorse case, discussed in the France chapter, the Paris Court of Appeal annulled the French Competition Authority's (FCA) decision of imposing a €199,000 fine on Sanicorse for imposing excessive price increases for medical waste treatment. The FCA had found that Sanicorse had abruptly, significantly, and durably increased the waste disposal prices it charged hospitals and clinics. In its ruling of November 2019, the Paris Court of Appeal clarified the conditions for establishing an exploitative abuse. Repeating the dictum from the United Brands ruling, the Court emphasised that an exploitative abuse arises in a situation where a dominant firm 'has made use of the opportunities arising out of its dominant position in such a way as to reap trading benefits which it would not have reaped if there had been normal and sufficiently effective competition'. The Court of Appeal found that the authority had failed to demonstrate that Sanicorse's price increases were unfair, and it accordingly annulled the decision.
Third, in December 2019, the FCA found in its Gibmedia decision (also discussed in the France chapter of this book) that Google's termination of three advertisers' Google Ads accounts was abusive. The authority's theory is that termination policies that allegedly lack objectivity and transparency, and are discriminatory, are a form of exploitation of customers. An apparent problem with the theory, however, is that a decision to terminate supply cannot, by definition, exploit the customer – it does not 'reap a trading benefit' from the trading partner, as required by United Brands and stressed by the Paris Court of Appeal in its Sanicorse decision.
Fourth, in February 2019, the Bundeskartellamt found that Facebook's terms and conditions relating to its collection of user data constitute an abuse (discussed in the Germany chapter). The Bundeskartellamt held that Facebook's terms and conditions, under which users agreed to the combination of their data from, for example, WhatsApp, Instagram and Facebook, violated the GDPR. Relying on German law principles that unlawful terms and conditions can constitute an abuse of dominance, the Bundeskartellamt held that Facebook committed an exploitative abuse by combining data from different sources. In August 2019, however, the Düsseldorf Court of Appeal granted suspensive effect to Facebook's appeal against the decision, holding that there are serious doubts about its legality. The Court found that users are not exploited by Facebook's use of data because, unlike financial payments, the data can be replicated and used again. Users freely decide whether to allow use of their data by balancing pros and cons of using ad-funded social network. The Court also held that the Bundeskartellamt had failed to prove the required causal link between Facebook's abuse and its market power: it failed to show that Facebook's terms deviated from the terms that would exist in a more competitive scenario. The judgment on the merits is pending.
Despite the renewed appetite to bring exploitation cases, these cases should in our view – in line with Advocate General Wahl's warning in the Latvian Banks case – remain rare and exceptional. Otherwise, there is a risk that the concept of exploitative abuse is stretched to address policy issues beyond the scope of competition law and that require broader discussion outside individual cases.
A greater push for interim measures
The second notable development in abuse of dominance enforcement in 2019 was the EC's decision – for the first time in an antitrust case in almost 20 years – to impose interim measures on Broadcom (this decision is discussed in the EU chapter). The decision orders Broadcom to cease to apply exclusivity provisions in six agreements with manufacturers of TV set-top boxes and modems, while the Commission's full investigation continues. On announcing the decision, Commissioner Vestager stressed that interim measures decisions are 'so important', especially in 'fast-moving markets'. The Commissioner emphasised that she is 'committed to making the best possible use of this important tool' so as to enforce competition rules 'in a fast and effective manner'.
Like other developments at EU level, push for greater use of interim measures has been encouraged by national authorities, particularly in France, with the Commissioner citing France as a source of inspiration. The UK CMA has also stated that greater use of interim measures is 'essential if the CMA is to respond to the challenges thrown up by rapidly changing markets', and Germany is adopting new rules to accelerate proceedings and apply interim measures.
Two examples discussed in the French chapter illustrate the FCA's expansionist approach to interim measures, both in cases involving Google. First, in Amadeus, the authority found Google's decision to suspend the Google Ads accounts of a paid phone directory services operator to be an exploitative abuse (similar to the theory in the Gibmedia case discussed above). The Paris Court of Appeal subsequently partly annulled the decision. Second, in early 2020, the authority found that Google's refusal to pay news publishers for showing preview snippets in search results alongside a link to the publisher's site may also amount to an exploitative abuse. The decision orders Google to enter into good faith negotiations with publishers, although it also makes clear that the negotiations may result in zero monetary compensation to publishers (considering that Google sends traffic to the publishers that they can monetise via ads on their page or convert users to paid subscribers).
Several points of caution should be heeded from the appetite to bring interim measures cases. Interim measures decisions should focus on the most egregious and clear-cut abuses, such as exclusivity clauses by obviously dominant firms, rather than seeking to create new law or go against existing precedent. The efficiency and effectiveness of competition procedures should not come at the expense of investigative rigour, due process, and the right to be heard. Interim measures should not prejudge the final decision from the authority on the merits. Accordingly, they should be tailored to implementing measures that are possible in principle to reverse, if it subsequently turns out that on a full merits review there is no case to answer. Finally, the new appetite to impose interim measures should not slow down the speed of the main proceedings, as agencies get caught up duplicating investigations and satellite appeals.
Per se bans on self-preferencing
The third development is the wide-ranging proposals to overhaul competition rules to address the perceived challenges of the digital economy. Proposals in the pipeline include the EC's suggestion for further regulation of digital platforms; mandatory codes of conduct in Australia to address perceived bargaining power imbalances between platforms and media companies; and, in the UK, the CMA's aim to develop 'a coherent and innovation-friendly approach to governing digital technologies to ensure their benefits are shared far and wide'.
Describing all these proposals is beyond the scope of the present editorial. We instead focus on one eye-catching suggestion: the suggestion – included in several of the reports commissioned by governments and agencies, such as the EU Special Advisors' Report, the Furman Report in the UK, the German ARC Amendments, and the Stigler Report – to introduce per se bans on digital platforms or companies that perform a 'regulatory function' from engaging in 'self-preferencing.' The reports, however, do not explain precisely what they mean by 'self-preferencing'. Self-preferencing is a generic expression that covers a range of different practices, for example, margin squeezing, tying and refusal to supply.
For example, keeping an indispensable asset to oneself and refusing to supply it to rivals is an example of abusive self-preferencing. But the refusal to deal in case law makes clear that it is, so far, not abusive for a dominant company to favour itself by reserving for its own use an asset that is not indispensable, but merely 'advantageous.' On the contrary, it is generally pro-competitive for companies to develop their own innovations, and use those innovations as the tools to compete against one another. As Advocate General Jacobs explained in Bronner:
it is generally pro-competitive and in the interest of consumers to allow a company to retain for its own use facilities which it has developed for the purpose of its business . . . Thus the mere fact that by retaining a facility for its own use a dominant undertaking retains an advantage over a competitor cannot justify requiring access to it”.
This makes sense, for several reasons. First, there is an inherent contradiction between competition and duties to supply rivals; competition rules seek to encourage companies to compete vigorously against each other, not cooperate. Second, a duty to supply interferes with fundamental rights to dispose of property and to conduct business. Third, duties to supply reduce incentives to innovate for both the supplying company and the company that receives supply. Fourth, in industries with fast innovation cycles, a duty to integrate rivals into constantly-evolving technologies may delay – or preclude – new developments.
The Courts, therefore, only allow interference with the freedom to contract in exceptional and limited circumstances. By contrast, we are concerned that a per se ban on self-preferencing could have several unintended consequences: hampering vertical integration, which is presumptively efficient; eliminating synergies; and leading to delayed or mothballed product improvements.
Consider Google's introduction of a thumbnail map on its results pages in response to location-based queries: the UK High Court held that this was 'pro-competitive' and an 'indisputable' product improvement. Not only was Google's introduction of the thumbnail map not likely to harm competition, but the conduct was also objectively justified. This was because showing rival maps would have degraded the overall quality of Google's search services, for example, via delays in returning results. Under the contemplated presumptions against self-preferencing, however, companies would have to ask themselves before launching this type of improvement whether they could prove the negative (i.e., that it would not lead to long-run exclusionary effects). That appears to be a difficult threshold to cross before launch.
Accordingly, we believe we should be looking at measures that make a real improvement to consumer welfare and avoid chilling innovation and investment. Neat-sounding slogans – such as a presumptive and generic ban on self-preferencing – can prove harmful in practice. As a recent CMA report into competition and regulation recognised, 'greater regulation is – on average – associated with less competition. For instance, countries with lower levels of product market regulation tend to have more competitive markets and enjoy higher rates of productivity and economic growth.' Similarly, in her speech on 'Remembering Regulatory Misadventure', FTC Commissioner Wilson recalled that attempts to prescribe 'fairness', 'non-discrimination', and 'reasonable and just' prices in the airline and railroad industries led to distortions of competition and restricted output. Removing these regulations 'significantly reduced consumer prices and increased output, generating billions of dollars in consumer surplus'. This is not to say that regulation is not desirable for objectives other than fostering competition, but regulation to encourage competition is likely to result in outcomes that any pro-competition and pro-innovation regime should avoid.
As in previous years, we would like to thank the contributors for taking time away from their busy practices to prepare insightful and informative contributions to this eighth edition of The Dominance and Monopolies Review. We look forward to seeing what the next year holds.
Maurits Dolmans and Henry Mostyn
Cleary Gottlieb Steen & Hamilton LLP
1 For further discussion, ee Cleary Gottlieb, Exploitative Abuse of Dominance and Price Gouging in Times of Crisis, 31 March 2020.