European competition law has a remedy problem. As digitisation changes how markets work, it becomes harder to find effective remedies. In this week’s blog, Jasper van den Boom discusses the papers [1][2] he has written together with Fiona Scott Morton, in which the authors theorise on how the EU remedy design can be recalibrated to empower the Commission and Make EU Competition Law Remedies Great Again (catchy!). Read a summary of the papers on the topic here first!
By Jasper van den Boom
Introduction
The European Commission recently released its ex post evaluation of the effectiveness of remedies following antitrust cases. The report shows that the Commission still favours cease-and-desist orders in most competition law cases (including cartels), that it will impose behavioural remedies in some cases, and that structural remedies remain exceedingly rare. It also shows that in Art. 102 TFEU cases, it is difficult to find remedies that are actually effective. Out of the 12 highlighted cases, only half of the imposed remedies achieved their desired results.
Over the past months, Fiona Scott Morton and I have conducted an in-depth look into the need for effective remedies in antitrust cases involving digital giants. Inspired by the US Department of Justice’s win in the Google case, we set out to find out what is needed next to actually restore competition. We started by looking at what the US can learn from experiences in remedy design in the EU. However, it turned out to be just as relevant to look at what the EU can learn from the remedy procedure in the US.
This is not the place to rehash the details of the cases against Google, nor to explain (again) why remedies in the EU did not work. Instead we turn to the future, with a call to action for both the European Commission and the US courts. Our study finds that the focus of the cases in the EU and the US differs: the first focuses on what happens in the markets around search, while the latter focused on what happened within search. However, none of the cases are perfect, each has their own merits and blind spots.
Despite the imperfect design of cases on both sides of the Atlantic, we found that US antitrust authorities are – at least juristically – significantly better positioned to introduce effective remedies. In fact, we find that the European Commission never really had the tools to successfully tackle Google’s anti-competitive acts in Google Shopping and Android. The Commission is held back by the procedures governing it: Already doomed to be David in a David v. Goliath scenario, DG Comp is fighting Big Tech also with a broken sling.
Remedies in EU competition law
The Commission leads in investigating powerful firms. It is supremely skilled at fact-finding and the development of theories of harm, establishing anti-competitive effects and finding liability. Once the Commission has established that the dominant firm has abused its dominance it is within its rights to end the abuse. This is where EU competition law differs greatly from US antitrust: the latter prohibits monopoly in itself, while the first only prohibits the abuse of one’s monopoly. The latter will try to end the monopoly following a finding of liability, while the first will try to end that anti-competitive conduct investigated.
How much space the Commission has to do so is largely determined by Regulation 1/2003, the Commission’s procedural rulebook. According to Article 7 of this Regulation, the Commission may take measures that are necessary and proportional to bring the infringement to an end. This leaves very little legal space for the Commission: not only can it not fight the source of the market power held by entrenched digital monopolists. It must fight the symptoms of that power in the least intrusive way possible (from the perspective of the investigated firm).
Such an approach was perhaps sensible when firms were still smaller than nations. Today’s digital giants however have more users than many countries have inhabitants, and more money than many states in their coffers. Not wholly unimportant, these firms have their own armies… of lawyers. After years of investigations, the Commission must render judgement on an opaque and ever changing business model to see what should be changed to resolve the investigated abuse. The firms, having convinced everyone of the enormous complexity of such an assessment, persuade us that there is more than one way to comply. The Commission, bound by precedent, lets the firm choose its method of compliance. The firm in question of course goes out of its way to find the best (read: worst) compliance strategy and to present it as if it were in fact a great sacrifice.
The laws currently governing remedies in the EU are simply not up-to-date for digital markets, and unable to safeguard both effectiveness and proportionality. Structural remedies are locked behind proportionality assumptions based on a false dichotomy: they can only be imposed if behavioural remedies cannot resolve the abuse in question or if more burdensome than structural remedies. In an era of doubt, it is difficult to prove that something cannot achieve its goals or is more burdensome, unless the firm in question allows you to make that finding. To make matters worse, the Commission must find a remedy that is proportional to the investigated behaviour. So, if the Commission finds that Google has made billions of euros over a decade of monopolising the comparison-shopping market, a divestment further up the value chain (Search, Chrome, Android) is not even on the table.
As a result of these constraints, the Commission is stuck between a rock and a hard place. On the one hand, it must protect its reputation as a leading investigator by doing what it’s good at: conducting substantive investigations and proving abuses of dominance. On the other, it’s reputation as a competition law enforcer is at risk if it cannot impose effective remedies. All those years investigating Google Shopping, all those years litigating, and what do we have to show for it? How then to escape this uncomfortable position, and where to look for guidance? It is becoming increasingly unlikely that the EU will look at the US on how to regulate digital giants, especially if EU antitrust interventions against large US firms are viewed as politically motivated hostility by the Trump administration.
Remedies in the US: a second-mover advantage
Antitrust authorities in the United States have an inverted position compared to their European counterparts. As we will show, US antitrust authorities have great capabilities to impose far-reaching remedies. The problem in the US is that of creating liability. Judges are not keen on agreeing with the DOJ that there is monopolisation. Cases under Section 2 of the Sherman Act are rare, and successful proceedings even rarer.
However, where the DOJ convinces the judge that monopolisation has taken place, a world of opportunity opens up. The DOJ gets to propose whatever remedies it would like to see, and it becomes up to the court to decide which ones are actually implemented. Structural remedies are definitely on the table. In fact, in the two largest monopolisation cases in US (and perhaps global) history, Standard Oil and AT&T were broken up. Alphabet could soon be added to that list. In the case of DOJ v. Google, the DOJ has proposed remedies that:
- Prohibit self-preferencing
- Require the divestiture of Chrome and Android
- Allow for the syndication of search results
- Create data-sharing, access, and portability obligations
- Impose transparency obligations
- Establish a technical committee
In many instances, these prohibitions go further than what is even in the DMA. As part of its structural remedies, Google would not be allowed to invest or acquire shares in any AI development, it must provide real-time access to search data, and pay for its own technical committee.
Proportionality means something completely different in the US. It is not necessarily about what is least intrusive for the firm, but it looks at finding a remedy that is best suited to effectively end the monopoly, and is not disproportionate in doing so. Proportionality is not the starting point, but rather the final check. Moreover, as it is the judge that picks the remedies, the DOJs hands remain relatively clean in this whole process.
Another important feature of the US remedy system is procedural: there is first the finding of liability, and then the case of remedies. Between the end of the substantive investigation and the imposition of the remedies, stakeholders are invited to share their views about what is needed, documents of discovery are made publicly available, and the dialogue is held out in the open. The latter, in turn, helps to inform private litigation that inevitably follows the public authority’s case. All of this gives the US a significant second-mover advantage: if the EU does most of the fact-finding and proves that anti-competitive activity is going on, the US can build on their work and step in with more effective remedies.
This second-mover advantage only works one way however: a look at the Microsoft cases shows that, even if the US has successfully proven liability and unsuccessfully tried to remedy the harm, the European Commission simply cannot impose more impactful remedies with the legal space granted to it. In some cases, like making the infringing firm pay for their own monitoring, it cannot even match the remedies imposed in the US. So, in terms of lessons for the US, we mostly found that the DOJ’s space for remedies is able to make up for its obstacles in establishing liability. So in a case like Google’s, where liability is established, it would be a shame to squander the judgement with ineffective remedies (here, of course, many minds will drift to the 2001 Microsoft judgement).
What does EU competition law need to restore balance?
The EU’s understanding of proportionality in competition law cases is currently out of balance. Sure, there is reason to say that no party should carry undue burdens and costs. However, this protection cannot only be extended to the investigated firm that has richly benefitted from its anti-competitive activity. If the harms to competition are not resolved, it become the public institutions, injured private parties, and society as a whole that carry these burdens on behalf of the monopolist. This is simply not right.
For what must be done to regain balance, we see three ways forward.
Option 1: amending the treaties to prohibit monopolisation
One, Article 102 TFEU is revised to prohibit monopolisation. With the European Commission’s investigatory fervour and pro-enforcement culture, such a change may better align its willingness to investigate and its ability to remedy anti-competitive market power. However, such an amendment of the Treaties would require unanimity in the Council and it is unlikely that such a revolution happens. It would also require us to reinvent the wheel of antitrust enforcement, as we must develop new standards and principles.
Option 2: revising Regulation 1/2003
The second option is to revise Regulation 1/2003, especially Article 7, to recalibrate the Commission’s toolbox with market realities and bring the meaning of proportionality back into focus. If the starting point for interventions is not merely ending the infringement, but actually restoring competition, we can see proportionality in that light.
First, we propose that the wording of Article 7 of Regulation 1/2003 should be changed from “ending the infringement” to “restoring competition”. Pursuing the restoration of competition broadly has two implications: first, the remedies should be able to effectively restore the competitive process going forward. Second, the Commission should see whether it can restore competitive conditions looking backwards by disgorging illicitly obtained profits and correcting the harms to injured third parties where possible.
We believe that the recitals could stipulate three elements of restoring (lost) competition. First, the recitals should explain that the Commission should have the discretion to restore competition to the state that it was in before competition was harmed by the anti-competitive conduct. After all, Art. 102 TFEU declares any abuse incompatible with the internal market. This means that any consequence of that abuse should be reversed, at least where it is possible to do so. Second, restoring lost competition may require that the illicitly obtained profits are disgorged from the monopolist. Not just through fines, but also by making the monopolist bear the costs of restoring competition going forward and back. This again helps to recalibrate the idea of proportionality: remedies should not cause unnecessary burdens for the investigated entity, but it should be possible to make the monopolist carry the costs of its own compliance, at minimum to the point that the illicitly obtained gains have been disgorged. Finally, a shift from ending the infringement towards restoring competition inverts the hierarchy of options by means of which the Commission can intervene: the starting point should be that the Commission does whatever is necessary to effectively restore lost competition. In a cartel case, this may simply be a cease-and-desist order, dissolving the cartel and reintroducing competition. However, it is not the assumption that cease-and-desist is enough. Instead, the norm becomes to have an in-depth investigation as to what is actually needed to restore competition.
Option 3: recalibrating the assessment on a case-by-case basis
There is also a third way. Rather than revising the law, the Commission could try to extract such a reinterpretation from the Court ad hoc. The Commission could propose more creative remedies, or argue that allowing defendants to choose their own method of compliance is no longer suitable to create effective remedies. This would allow the Commission to extract a more favourable interpretation of their legal space independently from the Council. However, trying to convince the Courts requires the Commission to take risks, in already long and complex procedures, and may ultimately lead to losses or unwanted precedent.
Our policy recommendation
We strongly endorse a revision of Regulation 1/2003 as the best route forward. By going through the relevant legislative procedure, the Regulation can be changed with a simple majority in the Council. A reinterpretation would renew the Commission’s mandate to enforce the law and find effective remedies. It would put the Commission in a strong position, and ensure that EU competition law remains a relevant alongside the DMA, and that the EU is not dependent on US antitrust authorities to step in and introduce effective remedies in the future. This route forward is more practicable than amending the Treaties, which would require unanimity of the Council. At the same time, it offers more legal certainty and is quicker than attempting to get a new interpretation of the court through enforcement.
Conclusions
This blog provides a sample of the two articles we will be publishing, the first preprint is available on SSRN already. In our first paper, we discuss what the US can learn from the EU. Here, we explore the second-mover advantage held by the US and the implications for the Google v. DOJ case. In our second paper, we propose the recalibration of Reg. 1/2003, and even suggest suitable wording. In short, competition problems in modern markets require modern solutions. In the face of powerful digital undertakings, entrenched monopoly, and new dependencies, the EU cannot stick to the same routines and practices. We are not the first to explore the EU remedy problem. However, our research comes at a critical inflection point. With the new Commission, ongoing revision of EU competition rules, and geopolitical shifts, now is the right time to critically assess our approach to remedy design and update our rules so that Europe can protect its markets. Our papers make a strong case for such a recalibration of EU law, and we provide clear policy suggestions in terms of the goals, text, and methods of assessment in the EU remedy procedure.
If you enjoy our blog, make sure to subscribe to our newsletter and/or share it!
