The European Commission’s remedial practice displays important differences in the type of remedies accepted in merger as opposed to antitrust cases. This paper briefly reviews the Commission’s remedies practice over the last 14 years highlighting the differences and discussing inconsistencies. It raises the question why predominantly behavioral remedies are chosen in antitrust cases and how this practice could be reconciled with the approach in merger control where the risks to effective competition are viewed as deriving from changes in the structure of the market and where therefore structural remedies are typically considered necessary.
By Frank Maier-Rigaud, Benjamin Loertscher1
Both antitrust and merger investigations at the EU level regularly conclude with the European Commission (“Commission”) accepting commitments or, in antitrust cases, imposing remedies.
In antitrust investigations, Article 9 of Regulation 1/20032 (“Article 9”) provides that the Commission can render commitments binding. In addition, Article 7 of Regulation 1/2003 (“Article 7”) endows the Commission with power to impose remedies to bring an infringement of Articles 101 or 102 of the Treaty on the Functioning of the European Union3 (“TFEU”) effectively to an end.4
In the merger control context, the Commission can accept commitments modifying a notified concentration and enabling the Commission to declare the concentration compatible with the internal market and the EEA Agreement pursuant to Articles 6(2) and 8(2) of the Merger Regulation.5
“Remedies” are therefore understood not only to include modifications to notified concentrations but also measures intended to bring an infringement of Article 101 or 102 to an end, or to meet concerns regarding such an infringement and remove grounds for action by the Commission.
The purpose of remedies is either to reduce or eliminate the ability or incentives of the undertakings concerned to follow a conduct that would impede or even eliminate effective competition, or, in particular, also to increase the ability of third parties to compete.
The theories of harm underlying antitrust and merger investigations are often similar if not identical; the difference being that antitrust investigations are typically concerned with actual (or at least alleged) infringements of competition law, whereas merger investigations consider potential harm to competition in the future.6
Despite these similarities for at least a subset of cases, the measures applied to remedy concerns in these two areas of competition law vary substantially. The predominance of behavioral remedies in antitrust cases stands in contrast to structural remedies mostly relied upon in merger investigations. This is surprising and begs the question of the consistency of the analytical framework used and the factors driving the Commission’s remedies practice.
The Merger Regulation notes the requirement that commitments accepted by the Commission “should be proportionate to the competition problem and entirely eliminate it.”7 Similarly, Article 7 empowers the Commission “to impose any behavioural or structural remedies which are proportionate to the infringement committed and necessary to bring the infringement effectively to an end.”
Remedies therefore must meet the following two requirements:
The remedies must be effective in removing the competition concerns and restore or maintain competition without the need for a (merger) prohibition decision or an infringement decision under Article 7.
The remedies must be proportionate to the competition concerns identified.
The legal hurdles for imposing structural remedies under Article 7 continue, however, to be considered to be higher than those for behavioural remedies. At least this is what a cursory look at Article 7(1)3 would suggest. Moreover, according to recital 12 of Regulation 1/2003, the imposition of a structural remedy “would only be proportionate where there is a substantial risk of a lasting or repeated infringement that derives from the very structure of the undertaking.” Reformulating the relevant sections of Article 7(1)3 while preserving their meaning, that is, ensuring the logical consistency between the original and the reformulated sentence, however, reveals that the subsidiarity of structural remedies is merely an impression based on the convoluted wording chosen. A logically equivalent, i.e. content preserving reformulation of Article 7(1)3 is as follows:
Behavioral remedies can only be imposed either where there is no more effective structural remedy or where any equally effective structural remedy would be equally or more burdensome for the undertaking concerned than the behavioral remedy.8
B. Types of Remedies
As indicated by Regulation 1/2003,9 the broadest classification for remedies distinguishes between structural and behavioral remedies.
Structural remedies seek to directly influence the competitive structure of the relevant market(s) in order to maintain or improve the conditions for competition. Behavioral remedies seek to address the identified competition concerns by requiring certain conduct from the undertakings concerned, which can of course include the requirement to refrain from certain actions.
Access remedies, which will also be introduced in more detail below, are another type of remedy often accepted in the past that has both structural and behavioral aspects. Despite various efforts to clearly distinguish between types of remedies,10 the distinction is not always clear cut in practice.
- Structural Remedies
A generally accepted definition of “structural remedies” does not exist.11 The divestiture of a business to a suitable purchaser is a very common structural remedy in conditional merger clearance decisions by the Commission. While stating that other types of commitments may also be acceptable, the Commission has declared its clear preference for structural remedies in merger cases,12 and, as shown below, this preference is also evidenced by the Commission’s practice.
The reason for this preference lies, on the one hand, in the perceived capability of structural remedies to durably prevent the competition concerns raised by the notified merger by changing the incentives of the firm(s) in the market. On the other hand, structural remedies have the advantage that they do not require ongoing monitoring over the medium or long term.
The removal of links with competitors is another type of structural remedy. This can for example include commitments to exit from a joint venture or to sell a minority shareholding in a competitor.
- Behavioral Remedies
Behavioral remedies are designed to regulate the ongoing conduct of the undertakings concerned and can come in very different forms. In antitrust cases, behavioral remedies are often designed to mirror the abuse: for example, a refusal to supply would be remedied with a commitment to supply or anticompetitive tying would be addressed with a commitment to untie.13
Behavioral remedies can be “positive” in the sense that they require a certain conduct or “negative” in the sense that they prohibit certain conduct. Examples of behavioral remedies that have been accepted in the past include:
- amending corporate governance provisions;14
- refraining from limiting capacity of certain infrastructure available to competitors;15
- making investments into gas interconnection capacity;16
- enabling customers to switch;17
- introducing a new pricing system;18
- capping of prices;19 and
- ring-fencing of strategic information within a consortium.20
Under Article 7, the Commission can also order the addressees of the decision to cease and desist the conduct infringing Articles 101 or 102. Such an order can also be viewed as a behavioral remedy
In the context of merger investigations, behavioral remedies are rarely favored. The Commission states in the Merger Remedies Notice that “[c]ommitments relating to the future behaviour of the merged entity may be acceptable only exceptionally in very specific circumstances.”21 The main reason for this is that their effectiveness is sometimes questioned because, in contrast to structural remedies, they leave the firms’ incentives essentially unchanged.22 In addition, there may be risks regarding the workability of behavioral remedies as an effective implementation and monitoring may be difficult to ensure, and risks regarding distorting effects on competition.23
Designing behavioral remedies that are effective and do not entail the risks noted above over the typically long periods during which they are in effect is very challenging. Especially in fast-moving industries, changing circumstances that could not have been foreseen can mean that behavioral remedies become less effective or even irrelevant, or more difficult to monitor.24 Moreover, the remedies can have the consequence of distorting the behavior of affected parties in unintended ways such as providing a negative incentive to innovate.25 Behavioral remedies are often accepted to complement other remedies in a “commitments package” to ensure their effectiveness.
- Access Remedies
Access remedies seek to eliminate the competition concern identified by requiring that access is granted at appropriate terms to an asset necessary to enable third parties to compete. The asset could be key infrastructure or intellectual property, for example:
- natural gas;28
- capacity in gas import infrastructure;29
- airport slots;30
- mobile telecommunications network;31
- network for supplying traction current;32
- technical information;33 and
- data under copyright.34
The granting of access is aimed at removing or lowering a barrier to entry or expansion to enable third parties to enter the market or to compete for a larger part of the market.
Access remedies do not neatly fall into the categories of structural or behavioral remedies and are therefore presented here as a separate category.35 The Merger Remedies Notice discusses them under the heading “other remedies”36 but also refers to the granting of access to key infrastructure or inputs as a structural remedy.37 Others suggest that access remedies are behavioral,38 non-structural39 or “quasi-structural”40 remedies. While access remedies a capable of achieving a structural effect on the market concerned, such an effect is not always guaranteed. A key aspect in this regard is whether the undertakings concerned actually commit to grant access to one or more third parties or whether they commit to offer to grant access in the event of a request from a third party.41 Access remedies can include the transfer of an asset.42 More often, however, they are designed to enable access through a supply, lease, license or other type of agreement that leaves the ownership of the assets unchanged. Moreover, even where such an agreement enables access for an unlimited duration, it will not necessarily be as permanent as a transfer of ownership since the agreement may be terminated. In this regard, access remedies are less contentious as property rights are less affected than in a divestiture. In contrast to divestiture commitments, behavioral and access remedies often need to be implemented over many years43 and can even be unlimited in duration44 and therefore require monitoring measures over a long-term period.
While divestitures, once implemented, are definitive,45 there remains a possibility for the undertakings concerned to request an amendment or waiver of long-term behavioral or access commitments under the review clause long after the original decision.46
III. THE COMMISSION’S PRACTICE
In order to gain a better understanding of the Commission’s remedies practice,47 we have reviewed the Commission’s decisions to accept or impose remedies in antitrust and merger cases over the period from November 2004 to December 2018.48 This period is split into three periods (November 2004 to January 2010, February 2010 to October 2014 and November 2014 to December 2018) which roughly corresponds with the terms of DG Competition’s Commissioners.49
Overall, we have reviewed 309 Commission decisions, including 55 antitrust decisions and 254 merger decisions. These decisions were reviewed with a view to categorizing each case according to the type(s) of remedies accepted or imposed. This work relied on the text of the decisions and commitments published by the Commission by the end of January 2018 as well as in some cases on the Commission’s press releases and Q&A documents. All cases were categorized according to whether the remedies included structural, behavioral or access remedies. Since remedies packages often consist of several types of remedies a decision can fall into more than one category so that the total number of remedies over all types exceeds the total number of cases.
It should be noted that this categorization is not always straightforward and there are of course instances in which a different categorization could arguably be used. Despite these inherent difficulties, this review provides some insight into the key differences in the Commission’s remedies practice in antitrust and merger cases as well as the development of this practice over a longer period of time.
A. Comparison of Antitrust and Merger Decisions
Figure 1 presents the share of the different types of remedies accepted (or imposed) by the Commission in antitrust and merger cases in the three periods defined. The share of decisions that included cease-and-desist-orders but do not impose specific other remedies are represented with crosshatched columns.
Source: Loertscher & Maier-Rigaud (2020) based on public Commission decisions.50
The overwhelming preference for structural remedies in merger cases stands in stark contrast to the very low share of antitrust cases involving structural remedies. Figure 1 also shows that this practice has not changed over time.
In line with the Commission’s Merger Remedies Notice, over 80 percent of conditional merger clearances in either Phase I or in Phase II involve a structural remedy.51 Conversely, only a limited number of mergers have been cleared subject to behavioral remedies. In this regard it is worth recalling that the small share of merger decisions including behavioral remedies indicated in Figure 1 includes decisions in which behavioral remedies were accepted as part of a remedy package that also included other types of remedies.
Figure 1 reveals the importance of behavioral remedies in Article 9 decisions. This prevalence is even stronger in Article 101 cases than in Article 102 cases. Only in the period from February 2010 to October 2014 did not all Article 9 decisions relating to infringements of Article 101 include behavioral remedies.52
No case has been identified in which a decision relating to an infringement of Article 101 led to structural remedies. The occurrence of structural remedies is also very rare in Article 102 cases.
Figure 1 also indicates an increasing occurrence of access remedies in Article 9 decisions rising from four out of 13 cases in the period from November 2004 to January 2010 to nine out of 17 cases in the period from February 2010 to October 2014 and three out of six cases in the period from November 2014 to December 2018.
While no Article 7 decisions were categorized as imposing access remedies, this is due to the fact that cease-and-desist-orders do not specify how compliance should be ensured. In certain cases, it would, however, be expected that compliance requires providing (better) access to essential assets to competitors.
Figure 1, however, shows no decrease in behavioral remedies accepted in Article 9 decisions. The higher occurrence of Article 7 decisions may indicate a reduced willingness by the Commission to accept remedies under Article 9 or firms may find it less appealing to offer commitments under Article 9 to the Commission.
With respect to access remedies, we observe a similar share of antitrust and merger cases that include such remedies and those shares also developed similarly over time. Their share increased in the period from February 2010 to October 2014 compared to the previous period but fell again in the period thereafter.
The review of the Commission’s case practice in antitrust and merger cases over a period of more than 14 years demonstrates stark differences in the Commission’s approach to remedies in these two areas of competition enforcement. The Commission mainly relies on behavioral remedies to address the competition issues underlying antitrust infringements while it displays a strong preference for structural remedies in merger investigations.
These different approaches to solving competition problems seem inconsistent and difficult to reconcile. On the one hand, while the legal hurdle for imposing structural remedies under Article 7 is certainly perceived to be higher than accepting them under Article 9 or in merger cases, the overwhelming use of behavioral remedies in antitrust cases suggests that behavioral remedies are in the Commission’s view an effective tool to resolve competition concerns. In order to address antitrust concerns, the Commission requires structural changes affecting firms’ incentive structures only in exceptional cases.
On the other hand, merger decisions are based on the idea that the risks to effective competition derive from changes to the structure of the market and therefore typically require structural remedies. The Commission’s merger remedies practice therefore suggests that behavioral remedies are not “as effective” in preventing restrictions to effective competition. If behavioral remedies are, however, an effective tool and if it is not necessary to change the structure of the undertaking in question to prevent effective competition from being restricted, then proportionality would suggest that they should be used more often in merger cases as well. This would in particular have to be the case considering the somewhat more uncertain nature of any forward-looking analysis indicating negative merger effects and despite the fact that pre-merger assets may not be integrated yet.
The review of the Commission’s remedies practice across antitrust and merger cases raises the question whether the theory underlying the Commission’s merger remedies practice or the theory underlying the Commission’s antitrust remedies practice is flawed or whether its reluctance to require structural remedies in antitrust cases and its corresponding reluctance to accept behavioral remedies in merger cases is driven by factors other than the underlying economics of the cases.
The Commission’s remedies practice may rather be a result of differences in the Commission’s negotiation position and pragmatism. While the Commission may balk at the suggestion that it is involved in negotiations when deciding to impose or accept remedies, there is no denying that firms’ incentives are to try and propose the least burdensome remedies that the Commission will accept and they will typically make several proposals to the Commission over the course of the proceedings to design the remedies in a manner acceptable to the Commission. This is as true in merger cases as it is in Article 9 cases. Even in Article 7 cases, commitment discussions may have taken place earlier in the proceedings. In addition, not only the Commission but also the parties are aware of the possibility of judicial review of the Commission’s decisions. The presumed higher legal threshold for imposing structural rather than behavioral remedies under Article 7 also reduces the likelihood of structural remedies being proposed under the Article 9 procedure. Firms may not wish to propose remedies for an Article 9 decision that are substantially tougher than the remedies the Commission could impose under Article 7. By contrast, in merger control proceedings the merging parties are in most cases under intense pressure to complete the review process quickly.
The stark differences identified between merger and antitrust remedies is, however, in part driven by a different distribution of competition concerns within these two areas of competition law. While the theories of harm underlying antitrust and merger investigations are often ultimately the same, their distribution is different. While in merger cases the concerns are often horizontal, Article 102 cases often involve concerns that are of a vertical nature and, even though Article 101 cases often also raise vertical concerns, it may be possible to resolve these concerns without resorting to structural remedies. Nevertheless, the stark contrast found in the review suggests that there is at least a subset of cases where concerns are identical and where one therefore would have expected an identical approach to resolve the identified competition concern.
1 Prof. Dr. Frank Maier-Rigaud, Managing Director and Head of Competition Economics Europe, NERA Economic Consulting; Department of Economics and Quantitative Methods at IESEG School of Management and Université Catholique de Lille; Researcher at the LEM CNRS. firstname.lastname@example.org. Benjamin Loertscher, Consultant, NERA Economic Consulting, email@example.com. This paper is based on a presentation given at the 14th annual conference of the GCLC in Brussels on January 31 – February 1, 2019 on Remedies in EU Competition Law: Substance, Process & Policy and the longer and more detailed analysis provided in Loertscher and Maier-Rigaud (2020).
2 Council Regulation (EC) No 1/2003 of December 16, 2002 on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty, OJ L 1/1 (4 January 2003) (“Regulation 1/2003”).
3 Consolidated Version of the Treaty on the Functioning of the European Union  OJ C 83/47.
4 See Lowe & Maier-Rigaud (2008) for an overview on antitrust remedies. Lowe, P., & Maier-Rigaud, F. (2008). Quo Vadis Antitrust Remedies. In B. Hawk (ed), International Antitrust Law & Policy: Fordham Competition Law 2007 (pp. chapter 20, 597–611). Fordham University School of Law.
5 Council Regulation (EC) No. 139/2004 of January 20, 2004 on the control of concentrations between undertakings (OJ L024, 29/1/2004) (“Merger Regulation”).
6 See for example Maier-Rigaud, F. P., & Tosini, N. (2019), Mergers between Backward Integrated Firms: Insights from BASF/Solvay’s Polyamide Business. Journal of European Competition Law & Practice, forthcoming, for a vertical theory of harm that could arise not only in a merger setting.
7 See Merger Regulation, paragraph 30.
8 See Maier-Rigaud (2012) and Maier-Rigaud (2016), which include not only a proposed test to choose remedies, but also a treatment of the likely intent of the European Commission as traced back to the initial proposals for Regulation 1/2003. Maier-Rigaud, F. P. (2012). The Idea of the Subsidiarity of Structural Remedies in European Competition Law (Zur Idee Der Subsidiarität Struktureller Maßnahmen Im Europäischen Wettbewerbsrecht). Wirtschaft und Wettbewerb, 5, 485-500. Retrieved from https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1906335. Maier-Rigaud, F. P. (2016). Behavioral versus Structural Remedies in EU Competition Law. In P. Lowe, M. Marquis, & G. Monti (eds.), European Competition Law Annual 2013, Effective and Legitimate Enforcement of Competition Law (pp. chapter 7, 207-224). Hart Publishing. Retrieved from https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2457594.
9 See Regulation 1/2003, preamble recital 12 and Article 7.
10 See for example the Merger Remedies Notice, the International Competition Network. (2016). Merger Remedies Guide. Retrieved from https://www.internationalcompetitionnetwork.org/portfolio/merger-remedies-guide, or Motta, M., Polo, M., & Vasconcelos, H. (2007). Merger remedies in the European Union: An overview. The Antitrust Bulletin, Vol. 52, Nos 3 & 4, 603-631.
11 For an overview of the discussion of the definition of structural versus behavioral remedies see Maier-Rigaud, F. P. (2016). Behavioural versus Structural Remedies in EU Competition Law. In P. Lowe, M. Marquis, & G. Monti (eds.), European Competition Law Annual 2013, Effective and Legitimate Enforcement of Competition Law (pp. chapter 7, 207-224). Hart Publishing. Retrieved from https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2457594.
12 See Merger Remedies Notice, paragraphs 15 and 17.
13 See Hellström, P., Maier-Rigaud, F., & Bulst, F. W. (2009). Remedies in European Antitrust Law. The Antitrust Law Journal, 76(1).
14 E.g. Case M.3817 – Wegener/PCM/JV.
15 E.g. Case AT.39316 – GDF foreclosure.
16 E.g. Case M.4180 – Gaz de France/Suez.
17 E.g. Case AT.39654 – Reuters Instrument Codes.
18 E.g. Cases AT.39678 & 39731 – Deutsche Bahn I & II.
19 E.g. Case AT.39398 – Visa MIF.
20 E.g. Case M.6844 – GE/Avio.
21 See Merger Remedies Notice, paragraph 17.
22 See Hellström, P., Maier-Rigaud, F., & Bulst, F. W. (2009). Remedies in European Antitrust Law. The Antitrust Law Journal, 76(1), or Kwoka, J. (2017). Merger Remedies: An Incentives/Constraints Framework. The Antitrust Bulletin, Vol 62, Issue 2.
23 See Merger Remedies Notice, paragraph 17.
24 See e.g. Hoehn, T., & Lewis, A. (2013). Interoperability remedies, FRAND licensing and innovation: a review of recent case law. European Competition Law Review, Vol. 34, Issue 2, p. 101.
25 See note 24 above.
26 E.g. Case AT.38636 – Rambus.
27 E.g. Case M.6564 – ARM/Giesecke & Devrient/Gemalto/JV.
28 E.g. Case M.3696 – E.ON/MOL.
29 E.g. Case AT.39316 – GDF foreclosure.
30 E.g. Case AT.39596 – British Airways/American Airlines/Iberia or Case M.5335 – Lufthansa/SN Airholding (Brussels Airlines).
31 E.g. Case M.6992 – Hutchison 3G UK/Telefonica Ireland.
32 E.g. Cases AT.39678 & 39731 – Deutsche Bahn I & II.
33 E.g. Case AT.39230 – Rio Tinto Alcan.
34 E.g. Case M.7337 – IMS Health/Cegedim business.
35 For example, the Merger Remedies Guidance of the Competition and Markets Authority in the UK (last updated in 2018) distinguishes between “IP remedies,” which it considers may have features of structural or behavioral remedies depending on their formulation, and access remedies, which it considers to be behavioral remedies. See also Sependa, P. A. (2013). Structural remedies under European Union antitrust rules. Concurrences, No. 2.
36 See Merger Remedies Notice, section 3.
37 See Merger Remedies Notice, paragraph 17.
38 See for example the Commission’s contribution to OECD. (2007). Remedies and Sanctions in Abuse of Dominance. Policy Roundtables.
39 See Motta, M., Polo, M., & Vasconcelos, H. (2007). Merger remedies in the European Union: An overview. The Antitrust Bulletin, Vol. 52, Nos 3 & 4, 603-631, or the Merger Remedies Guide published by the International Competition Network (2016), Annex 2. Retrieved from https://www.internationalcompetitionnetwork.org/portfolio/merger-remedies-guide/.
40 See for example Motta, M., Polo, M., & Vasconcelos, H. (2007). Merger remedies in the European Union: An overview. The Antitrust Bulletin, Vol. 52, Nos 3 & 4, 603-631, OECD. (2011). Remedies in Merger Cases. Policy Roundtables, or Competition Bureau Canada. (2006), Information Bulletin on Merger Remedies in Canada.
41 E.g. Case M.5440 – Lufthansa/Austrian Airlines.
42 E.g. in Case M.6607 – US Airways/American Airlines airport slots would effectively be transferred to the new entrant if appropriate use had been made of the slot by the new entrant through the access remedy over a certain period.
43 For example, five years in AT.39592 – Standard & Poor’s.
44 E.g. the interfacing commitment in Case M.3083 – GE/Instrumentarium or the airport slot release commitments in Case M.3770 – Lufthansa/Swiss.
45 In merger cases, commitments accepted by the Commission will normally require the merged entity not to re-acquire material influence over the whole or part of the divestment business for a period of 10 years, unless the Commission finds that the structure of the market has changed to such an extent that the absence of influence over the divestment business is no longer necessary to render the proposed concentration compatible with the internal market. After this 10-year period, the merged entity may (re-) acquire the divestment business. Such a transaction may not necessarily be subject to merger control. See Merger Remedies Notice, paragraph 43.
46 In Case M.3280 Air France/KLM for example, the Commission agreed to waive a slot release remedy as well as other related commitments more than 15 years after the conditional clearance decision.
47 A more detailed analysis of the Commission’s practice can be found in Loertscher, B., & Maier-Rigaud, F. P. (2020). On the Consistency of the European Commission’s Remedies Practice. In D. Gerard, & K. A., Remedies in EU Competition Law: Substance, Process, Policy, Kluwer Law International.
48 Decisions relating to cartel investigations have not been considered for this purpose.
49 The last of the three periods does not fully cover the current Commissioner’s first term.
50 Loertscher, B., & Maier-Rigaud, F. P. (2020). On the Consistency of the European Commission’s Remedies Practice. In D. Gerard, & K. A., Remedies in EU Competition Law: Substance, Process, Policy. Kluwer Law International.
51 Overall, 212 out of 254 conditional merger decisions reviewed included structural remedies.
52 The exceptions are two aviation cases (AT.39596 – British Airways/American Airlines/Iberia and AT.39595 – Continental/United/Lufthansa/Air Canada) in which airport slot release and other commitments were accepted that have been categorized as access remedies.