…with Isabelle de Silva
In this month’s edition of CPI Talks… we have the pleasure of speaking with Isabelle de Silva, President of the French Competition Authority (the Autorité de la concurrence).
In this month’s edition of CPI Talks… we have the pleasure of speaking with Isabelle de Silva, President of the French Competition Authority (the Autorité de la concurrence, or “AdC”).
Thank you, Ms. de Silva, for sharing your time for this interview with CPI.
1. In January of this year, the AdC published a report (the “AdC Report”) reflecting on the AdC’s use of behavioral remedies in competition enforcement. What, in your view, are the key conclusions to be drawn from this report?
The AdC Report sets out how the AdC has used behavioral remedies in merger and antitrust cases so far. Its purpose is twofold: to provide technical guidance for companies, competition law practitioners and academics, and to drive thinking forward on the use of behavioral remedies by the AdC in the future.
The report acknowledges that behavioral remedies have several important advantages. They can take many forms and adapt to many situations. The decision-making practice in antitrust cases has also shown that, when proposed remedies are discussed with third parties, they can, sometimes, bring to light difficulties with existing regulations, so the process is helpful in that it makes it possible to also address these problems.
However, behavioral remedies also raise some specific challenges. First, they may be insufficient for certain types of anti-competitive effects, such as creating a reduction in horizontal competition. In that case, behavioral remedies might need to remain in place for a very long time, without guarantee that, in the end, the market will function well by itself. Another aspect is that monitoring the implementation of behavioral remedies often creates a heavy burden for enforcers and companies alike.
As the challenges of behavioral remedies may often outweigh their benefits, a key conclusion of the report is that the AdC will likely make more limited use of such remedies and favor, instead, quasi-structural remedies in antitrust cases, as well as structural remedies in merger cases.
2. The classic objection to behavioral (as opposed to structural) remedies lies in difficulties relating to their monitoring and enforcement. This concern is perhaps most clearly reflected in practice under the EU Merger Regulation, where the EU Commission displays a clear preference for structural remedies, on this basis. While it is difficult to draw bright-line rules in this regard, would it be worthwhile to issue more detailed guidance documents to identify those situations where behavioral remedies may be more appropriate and proportionate (despite their drawbacks in terms of monitoring and enforcement)?
The AdC Report is designed as a form of guidance for companies, as well as legal and economic advisors. They will be able to find in the report a global synthesis of the different cases in which the AdC accepted behavioral remedies. We don’t want to be too prescriptive because remedies must always be tailor-made for a specific problem, market and competitive situation. But companies will find a wealth of information in our report, for example regarding cases in which behavioral remedies can be a viable avenue and those in which they will be difficult to consider.
The approach taken by the report is to present general principles governing the choice of remedies, as well as the objectives related to each category of remedies (behavioral and structural), to conduct a case-by-case analysis of the AdC’s decision-making practice and to share the lessons learned from this experience. The report will hopefully provide valuable information on the subject since it is fairly comprehensive.
3. As far as judicial review of remedies is concerned, the AdC Report notes that the role of the competent French courts is limited, for the most part, to assessing the proportionality and legality of any remedy imposed, and that courts do not enjoy broad discretion to reformulate detailed aspects of a given remedy imposed by the regulator. Similar rules apply in most jurisdictions. In your view, is this the correct approach? Alternatively, should courts enjoy broader flexibility to reformulate or otherwise correct remedy design, or is this role best left to specialized regulators such as the AdC?
The applicable judicial review is well balanced and works well. Depending on the decision at stake, two different judges intervene: the Paris Court of Appeals for antitrust decisions and the administrative Supreme Court (Conseil d’État) for merger decisions. They both review the legality of remedy decisions: while the Conseil d’État assesses a potential abuse of power and can annul the contested decision in whole or in part, the Paris Court of Appeals can annul or amend the contested decision (noting that the Court does not have the power to discuss new remedies with companies to make them binding, nor can it order them to comply with other specific conditions not provided for in the contested decision). Regarding the assessment of the proportionality of remedies, the Paris Court of Appeals’ review is limited to a so-called “manifest error control,” while the Conseil d’État exercises a so-called “normal control,” which is quite thorough. In at least one case, for example, the Conseil d’État decided that a remedy was not sufficient to compensate for the lessening of competition, and then went on to annul the AdC’s decision in that respect. The AdC had to come up with a new, more stringent remedy.
I find it very wise that the judges refuse to define themselves what a particular remedy should be. First, each remedy is based on a detailed analysis of the different parameters of competition, and it is, first and foremost, the AdC’s job to conduct such an analysis. Second, a remedy should be proposed by the parties and must, in many cases, be fine-tuned after discussions with the AdC and feedback received from third parties. A court is not well-equipped to monitor such a process, and should then be only asked to review its final outcome.
4. The AdC Report notably concludes that, like other competition authorities, the AdC will consider having more limited recourse to behavioral remedies in future, and will favor so-called “quasi-structural” commitments for anticompetitive practices, and “true” structural commitments in merger law, whenever they provide a better resolution to the competition issues identified. What criteria should be used to determine which type of remedy is most appropriate? Does the AdC propose to issue guidance in this regard? Should other bodies also update their guidance (notably, should the EU Commission consider revising its Remedies Notice in line with your conclusions)?
The AdC Report provides ample information to allow stakeholders to make informed decisions. The report underlines that the nature of remedies depends on the specific competition concerns at stake. For merger cases, behavioral remedies have been accepted in 36 percent of clearance decisions in which anti-competitive effects were identified (55 percent if mixed remedies – behavioral and structural – are included) since 2009. The proportion of behavioral remedies out of the total remedies accepted by the AdC is among the highest in Europe. In antitrust matters, because concerns are not related to market structure but to market behavior, remedies were mostly behavioral so far or quasi-structural in some instances. The report details the circumstances of each case in order to explain in which situations a particular remedy may be warranted. The report also notes the difficulties raised by behavioral remedies and the clarification, in the ECN+ Directive, that structural injunctions can indeed be issued in antitrust cases.
We currently feel no need to take the guidance exercise a step further by imposing allocation criteria. First, as illustrated by the report, remedies must be decided on a case-by-case basis. Second, as the report also recalls, the effectiveness of remedies lies in the fact that they are proposed by companies and developed jointly with relevant market players, which reinforces companies’ accountability vis-à-vis the remedies they have proposed and will need to implement.
5. Remedies must be tailored to the facts of the individual case and the industry at hand, the nature of the specific infringement identified, and the theory of harm relied upon by the regulator. New business models in the so-called “digital economy” raise novel issues for enforcers, whether engaged in antitrust enforcement, or merger control. Notably, questions such as access remedies to “big data,” or other complex conduct raised by the practices of big tech companies raise issues that are not necessarily analogous to issues raised in classic antitrust cases. Might complex behavioral remedies necessarily be a feature of enforcement with regard to such practices? What is to be learned from recent high-profile remedy negotiations at both the national and international levels in this industry?
It is true that digital technologies involve increasingly complex remedies and therefore require additional resources and expertise for competition agencies. It is one of the reasons why the AdC created a Digital Economy Unit. One specific task of this specialized unit is to assist our Investigation services in assessing remedies in digital markets and monitoring their implementation.
An emblematic case featuring behavioral remedies in the digital economy sector is the Google Shopping case in which the European Commission fined Google €2.42 billion for abusing its market dominance as a search engine by giving an illegal advantage to another Google product, its comparison shopping service. The Commission decision required Google to stop its illegal conduct and to apply the same processes and methods to rival comparison shopping services in Google’s search results pages as it gave to its own comparison shopping service. Google has sole responsibility to ensure compliance and is under an obligation to keep the Commission informed of its actions. This case is an example of behavioral remedy applicable to an antitrust infringement decision.
More generally, the current debate about platforms illustrates, according to some, the need to apply, very early on, behavioral remedies to maintain a level-playing field, before anticompetitive behavior has wiped out competition.
This is something we take into account in our enforcement. In that respect, interim measures can also incorporate, quite early on, some injunctions to address specific behavior in order to prevent the consequences of a possible abuse of dominance for instance. The recent decision taken by the AdC on the matter of press content ancillary rights is a good example, and illustrates that it is possible to act quickly by issuing interim measures. Following a complaint, the AdC found that Google may have abused its dominant position on the market for general search services by imposing unfair transaction conditions on publishers and news agencies. Google was ordered to negotiate with publishers and news agencies that request it for the remuneration due to them for any use of protected content based on transparent, objective and non-discriminatory criteria. The injunction required that the negotiations actually result in a remuneration proposal from Google. The company will have to provide the AdC with monthly reports on its compliance with the decision.
