The Impact of EU Antitrust Procedure on the Role of the EU Courts (1997–2016)

The Impact of EU Antitrust Procedure on the Role of the EU Courts (1997–2016) Key Points The decentralisation of EU antitrust enforcement did not result in a drop in the number of applications for annulment of Commission decisions, nor in an increase in the number of preliminary references. The introduction of the formal commitments procedure did not, but the introduction of the settlement procedure in cartels did cause a significant drop in the number of applications for annulment. The low number of preliminary references in competition cases in recent years may be influenced by the soft law instruments issued by the Commission. I. Introduction For better or worse, the EU courts have played an important role in the development of EU competition law. This is not surprising: since its foundation, the CJEU (then ECJ) has had a profound impact on European integration in general and the development of European law in particular. In the early days of antitrust enforcement, ‘the support given by the Court [of Justice] to the work of the Commission was of critical importance in enabling its main lines of policy to be established.’1 But also in subsequent years, the CJEU opened and closed enforcement paths of the Commission and for private plaintiffs before national courts, both through direct actions (applications for annulment) against Commission decisions and through its preliminary rulings procedure.2 After its creation in 1989, the Court of First Instance (now General Court) assumed its own role in shaping EU competition law. The Court of First Instance was set up in response to the increased number of cases brought before the CJEU (and a growing backlog) because of the geographical and substantial enlargement of the European Communities in the 1970s and 1980s. In particular in the field of competition law, it was also felt that the CJEU ‘was not in the best position to review … serious issues of fact or economic assessment’.3 The Court of First Instance would allow ‘a fuller investigation of factual issues [to] take place’ in these cases4 or even an ‘exhaustive review’ of the ‘complex technical and economic assessments’ of the Commission.5 Much has been written about the quality of the case law of the CJEU and General Court and how their influence has sometimes been a boon for and sometimes an obstacle to effective antitrust enforcement. The CJEU’s 1973 judgement in Continental Can, for example, basically opened up the entire field of merger control as an area of enforcement.6 Similarly, the General Court’s annulment of three merger decisions in 20027 resulted in more robust economic reasoning in the field of merger control and the creation of the Chief Economist team at DG Competition in 2003. On the other hand, the ‘toothless fallacy’ accepted by the CJEU in United Brands continues to discredit the use of market definitions in competition cases.8 In recent years, several practitioners have also criticised the CJEU and the General Court for softening their review of Commission decisions: the EU courts would take a too deferential approach to the Commission’s fine calculations in cartel cases9 and would leave too much margin of appreciation for the Commission’s economic assessments.10 The aim of this paper is not to add another gloss to the discussion on the appropriate review powers of the General Court and the CJEU. By way of an analysis of the statistics of the EU courts, it rather aims to respond to the empirical question: what is the actual role of the EU courts in competition cases today? This question should precede the discussion on the review powers of the EU courts, lest the latter be based on false premises.11 The focus of this paper is limited in scope: its statistical assessment does not cover the outcome of proceedings before the General Court and the CJEU (the number and content of judgements) but merely its input (the number of new cases brought before the EU Courts).12 To some extent the CJEU and General Court themselves can influence that input. Through their case law, they set the boundaries of their review, in terms of which cases are admissible and which arguments have a good chance of success. The more interventionist the EU courts are, the more likely it is that undertakings will bring applications for annulment against Commission decisions.13 To a lesser extent, the CJEU’s approach to preliminary questions can also influence the number of questions put to it. Also the duration of proceedings before the EU courts impacts on how attractive procedures before them are. This paper focuses on another factor that may impact on the workload of the EU courts and the role they play in competition law enforcement. This is the procedural framework for enforcement at the level of the Commission and national competition authorities (NCAs), in particular in light of the changes brought about since the so-called ‘modernisation’ of European antitrust in the late 1990s and early 2000s. I will examine in turn the following three procedural changes14: the decentralisation of antitrust enforcement; the introduction of (formally) negotiated resolutions to cases; and the use of soft law instruments by the European Commission. The conclusions of my analysis are as follows: Contrary to common belief and expectations, the decentralisation of EU antitrust enforcement did not result in a drop in the number of applications for annulment of Commission decisions, nor in an increase in the number of preliminary references. While there does not seem to be any evidence of a reduction in the number of applications for annulment following the introduction of the formal commitments procedure in Article 9 of Regulation 1/2003, the introduction of the settlement procedure in cartels in 2010 did cause a significant drop in the number of applications for annulment. Although the number of preliminary references in competition cases has always been small, it cannot be excluded that the reduction in the number of references in recent years is influenced by the soft law instruments issued by the Commission. II. Decentralisation Regulation 17/62 provided that undertakings who wanted to avail themselves of the exception of (what is now) Article 101(3) TFEU, needed to notify their agreements to the Commission.15 This system was abolished when Article 1 of Regulation 1/2003 introduced a system of legal exceptions meaning that (what is now) Article 101(3) TFEU became directly applicable16 and undertakings had to self-assess whether their agreements were in accordance with it. This change also implied that national courts could now rule on all aspects of antitrust disputes (whereas they before had to refuse jurisdiction if a party invoked what is now Article 101(3) TFEU). In parallel, Article 3(1) of Regulation 1/2003 imposed an obligation on NCAs to apply (what is now) Article 101 and 102 TFEU each time they applied their national competition law to an agreement or practice that affected trade between the Member States. National courts were therefore empowered and NCAs enlisted to apply the European antitrust rules alongside the Commission. This decentralisation of EU antitrust enforcement has significantly changed the enforcement architecture. In July 2014, the Commission reported to the European parliament and the Council that in the first 10 years since the entry into force of Regulation 1/2003, the vast majority of decisions on Article 101 and 102 TFEU had been taken by NCAs.17 This development has continued unabated since.18 How did this affect the EU courts? In light of the increased role of national courts in EU antitrust enforcement both for disputes between private parties and on appeal against decisions of the NCAs, many predicted a rise in the number of preliminary references in this field.19 However, this is not borne out by the statistics of the CJEU. Figure 1 shows how the number of preliminary questions in competition cases (excl. State aid) has fluctuated over the last 20 years (dated by the year in which the question was received by the CJEU). While the numbers are limited and, in line with the law of small numbers, it is therefore difficult to draw firm conclusions from peaks and troughs, there has not been a significant increase in the period after 2004, compared to the period before. On the contrary, since many national competition cases are handled by NCAs and only come before a court or tribunal in the sense of Article 267 TFEU after a few years, one would expect to see a particular increase from, say, 2007 onwards, but precisely from then on, the figures have been particularly low (with the exception of 2016). Figure 1: View largeDownload slide Preliminary questions in competition cases (excl. State aid). Figure 1: View largeDownload slide Preliminary questions in competition cases (excl. State aid). This (lack of) development is even more striking if compared to the increase of the number of preliminary references in competition matters over previous decades20 as well as the continuous increase of preliminary references in other fields of European law since the mid-2000s, as shown in Figure 2. Figure 2: View largeDownload slide Preliminary questions (all subjects). Figure 2: View largeDownload slide Preliminary questions (all subjects). I will explore below possible explanations for the perceived evolution of the number of preliminary references in competition cases. It has been argued that the decentralised application of EU competition law, has also led to a drop in the number of direct actions before the EU courts (in first instance, mainly at the level of the General Court).21 But it is not obvious to interpret the evolution of the number of direct actions (contained in Figure 3) in this sense. Figure 3: View largeDownload slide Direct actions in competition cases (excl. State aid). Figure 3: View largeDownload slide Direct actions in competition cases (excl. State aid). If anything, the number of direct actions seems to have increased after the entry into force of Regulation 1/2003 and only starts to fall from 2011 onwards. As we will see further on, there is a specific explanation for the drop in the last five years. The idea that decentralisation would mean fewer direct actions is based on the premise that the Commission would adopt fewer decisions after decentralisation. But that is not what the Commission did. In fact, already in the years prior to the entry into force of Regulation 1/2003, the Commission had started to refocus its investigations. The ‘more economic approach’ to antitrust propagated since the middle of the 1990s had led the Commission to turn its attention away from vertical restraints (which Chicago School economists viewed as generally unproblematic) and towards the most egregious of antitrust offences: cartels. While in the 1990s, the Commission adopted only 2 cartel decisions on average every year, since 2000 it has adopted on average 6.22 Decentralisation has therefore not simply led to fewer decisions by the Commission but to different decisions (fewer on verticals, more on cartels). Not only did the Commission increase its cartel enforcement, also the level of fines went up significantly: while the total amount of cartel fines in 1995–1999 was EUR 293 million, this increased more than tenfold in 2000–2004 to EUR 3.5 billion and again almost tripled in 2005–2009 to EUR 9.4 billion.23 It is this increasing number of cartel decisions and the increasing level of fines imposed for cartel infringements which coincides with the increased number of applications for annulment to the General Court.24 As a consequence, the number of direct actions has increased from 25 in 1997 (the first year for which statistics on the judicial activity of the courts are available online) to a peak of 82 in 2006. III. Negotiated procedures A second development in the enforcement of EU competition law, relates to the use of the negotiated conclusion of investigations. The European procedural rules for antitrust enforcement provide for the negotiated conclusion of cases in two forms: through the commitment procedure pursuant to Article 9 of Regulation 1/2003 and through the settlement procedure in cartel cases introduced in 2008.25 Under Regulation 17/62 it was already possible for the Commission to attach conditions and obligations to its decisions, but this assumed that it had established an infringement. In parallel, the Commission developed a practice of informally concluding investigations by means of negotiated amendments to the agreements or practices it investigated, but the enforcement of such negotiated amendments was problematic, since it was not foreseen by Regulation 17/62. Article 9 of Regulation 1/2003 formally introduced a commitments procedure pursuant to which parties can offer commitments to address concerns the Commission has as regards an agreement or practice to which they are a party. If the Commission considers that these commitments are adequate, it can make them binding on the undertaking and conclude that there is no longer ground to continue its investigation. The undertaking in this way avoids a finding of infringement (which entails reputational damage and facilitates third-party damage actions) and a fine. The advantage for the Commission is that it avoids protracted litigation over the alleged infringement while still having enforcement tools at hand to ensure compliance with the commitments offered (in particular Article 23(2)(c) of Regulation 1/2003 which allows it to fine an undertaking which does not comply with a commitment made binding by a decision). According to recital (13) of Regulation 1/2003, commitment decisions are not appropriate in cases where the Commission intends to impose a fine and the Commission has stated that it will not use the procedure if the nature of the infringement calls for the imposition of a fine, such as in the case of secret cartels.26 The commitment procedure is therefore used to combat practices which the Commission believes are harmful to competition but where this harm may not be self-evident to the undertaking(s) concerned. This allows the Commission to be more flexible in enforcement but at the same time this implies that its enforcement practice becomes less transparent (both as regards the undertaking which is the subject of the commitment decision and third parties).27 Of specific relevance for this paper, is the prediction and affirmation that the increasing use of the commitment procedure would reduce the number of applications to the EU courts resulting in less judicial scrutiny of Commission decisions.28 However, it is hard to demonstrate this by way of the statistics of the EU courts. Figure 4 shows that, after the introduction of the commitments procedure in 2004 and the first commitment decisions in 2005 and 2006, the number of applications for annulment to the General Court did not decrease—on the contrary. Figure 4: View largeDownload slide Commitments decisions v direct actions in competition cases. Figure 4: View largeDownload slide Commitments decisions v direct actions in competition cases. More importantly, it is hard to compare the situation following the introduction of the formal commitment procedure through Article 9 of Regulation 1/2003 with the prior situation, because, as indicated above, the Commission already informally accepted commitments under Regulation 17/62 (even though there were doubts as to the enforceability of these commitments). There are no full statistics of informal commitments under Regulation 17/62 but there is no doubt that informal commitment procedures were also common under Regulation 17/62, so that the introduction of the formal commitment procedure of Article 9 of Regulation 1/2003 may not have significantly reduced the number of infringement decision or the number of applications for annulment to the EU courts.29 Also the drop in recent years cannot be explained by an increased use of commitment decisions: under Commissioner Almunia (2010–2014) the number of commitment decisions was not significantly higher than under Commissioner Kroes (2005–2009),30 nor has it increased since the appointment of Commissioner Vestager.31 Let us then turn to the second negotiated procedure in antitrust, the settlement procedure. This procedure was introduced, not by Regulation 1/2003 but by an amendment to the Commission’s Implementing Regulation32 in 2008. The settlement procedure only applies to cartel cases: it allows cartel participants to obtain a fine reduction of 10 per cent if they acknowledge their participation in the cartel. In this way, the Commission obtains a swifter procedure, as it no longer needs to proceed through the entire evidentiary process to establish the infringement. The settlement procedure rapidly became the norm in cartel cases: between 2010 (when it was first used) and 2016, over half of the cartel decisions adopted by the Commission were settlement decisions. The settlement procedure allows the Commission to wrap up investigations faster and with less effort: undertakings can no longer argue about the ins and outs of every allegedly anticompetitive contact and have to focus on a number of key parameters that will affect their fine level. In most cases, this has allowed the Commission to settle with all cartel participants, although it does not fail to pursue cases where one or two cartel members try to escape fines by staying out of the settlement process.33 For the undertakings involved, the 10 per cent discount of the fine is not the only advantage they obtain. Since they know that the settlement process entails efficiencies for the Commission, they can also try to negotiate key parameters for the fine that are relevant for them—even though the Commission has denied that the fine is negotiable.34 Furthermore, the Commission decision of a settlement is much less detailed than ordinary cartel decisions: it is usually only one or two dozen pages, while ordinary cartel decisions can run in the hundreds of pages. This makes a settlement decision much less useful for damage claimants. Let us then turn to the effect of the settlement procedure on the applications before the EU courts. Although there is no legal bar to bringing an application for annulment against a settlement decision, the number of such actions has been extremely limited.35 As a consequence, cartel cases concluded through settlement are much less likely to result in cases before the EU courts than ordinary cartel decisions. The dotted orange line in Figure 5 shows the number of undertakings which concluded a settlement (rather than the number of cases in which a settlement was concluded) since this better reflects the potential number of direct actions that were avoided through settlement.36 Figure 5 indeed shows that the number of applications for annulment before the General Court dropped significantly in 2011, the year after which the first settlement decisions were concluded.37 This drop is even more striking if compared to the number of direct actions in other fields (in particular in the field of intellectual property), which continue to increase year after year. Figure 5: View largeDownload slide Cartel settlements v direct actions in competition cases. Figure 5: View largeDownload slide Cartel settlements v direct actions in competition cases. A study of the German Centre for European Economic Research estimated that, excluding other factors that might affect the willingness to appeal, the introduction of the settlement procedure has reduced the number of applications for annulment in cartel cases by 20–55 per cent.38 This indeed demonstrates that the drop in the number of applications in recent years is at least partially due to the introduction of the settlement procedure. IV. Soft law instruments The third development discussed in this paper is the increased use by the Commission of soft law instruments, i.e. guidelines and notices to clarify its policy. This tool has been used by the Commission since the middle of the 1990s and has been one of the hallmarks of the modernisation of EU competition law. Indeed, the ‘more economic approach’ was difficult to reconcile with the formalism of block exemption regulations which was a key enforcement tool for the Commission under Regulation 17/62 (if only to keep the number of notifications of agreements and practices under control).39 Furthermore, although the EU courts and the Commission in its decisions had provided clarifications on many aspects of EU competition law, there were other fields where guidance was lacking or at least not straightforward. By issuing formal guidance, the Commission facilitated the task of undertakings (and their external advisers) who had to self-assess the compliance of agreements and practices with the EU antitrust rules under the regime of legal exception of Regulation 1/2003. It also allowed for greater consistency in the application of Article 101 and 102 TFEU by the NCAs and national courts which were at the forefront of enforcement after decentralisation. While the more economic approach was obviously also apparent in the field of merger control, the other two drivers of soft law discussed above (regime of legal exception and decentralisation) only applied in relation to EU antitrust and I will therefore focus on the soft law in this field here.40 A first wave of soft law guidance was initiated with the 1997 Green Paper on vertical restraints41 and resulted in the following soft law instruments on the interpretation of Article 101 TFEU in the period 2000–2004:42 The 2000 guidelines on vertical restraints,43 The 2001 guidelines on horizontal cooperation agreements,44 The 2004 guidelines on technology transfer agreements,45 and The 2004 guidelines on the application of Article 101(3) TFEU.46 The first three of these guidelines were updated a decade later following a consultation process initiated in 2009 and resulting in the following instruments: The 2010 guidelines on vertical restraints,47 The 2011 guidelines on horizontal cooperation agreements,48 and The 2014 guidelines on technology transfer agreements.49 In between these two waves of substantive guidance on Article 101 TFEU, the Commission also held a lengthy consultation process on the interpretation of Article 102 TFEU which resulted in the 2009 guidance on the enforcement priorities in exclusionary conduct.50 How did the increased use of soft law by the Commission affect the role of the EU courts in antitrust enforcement? First of all, as pointed out above, the issuance of soft law guidance was both a precondition and a consequence of the abolition of the notification system by Regulation 1/2003. Like the latter, it allowed the Commission to refocus its investigations to hardcore offences, and to devote less attention to less egregious offences of competition law. This also implied fewer decisions in the fields were soft law was issued.51 However, as we already saw when discussing decentralisation, this did not necessarily imply fewer applications to the EU courts: the refocusing of the Commission increased the number of cartel decisions it issued and this in turn increased the number of applications for annulment before the EU courts (see Figure 3). What about preliminary questions? As we saw when discussing decentralisation, while the number of preliminary questions in competition cases is limited and it is therefore hard to draw statistically significant conclusions from it, the number of questions seems to be in decline and this, contrary to the trend in other fields of European law. Since there have virtually been no preliminary references in the field of mergers,52 the answer must be sought in the area of antitrust, but this is paradoxically the area in which the NCAs and national courts have been empowered by Regulation 1/2003. At first sight, the increased use of soft guidance by the Commission instead of block exemption regulations can also not explain the reduction in preliminary questions. Contrary to block exemption regulations, the guidance is not binding on the NCAs and the national courts. One would therefore expect that disagreement amongst parties to litigation would continue, despite the guidance, and that preliminary references would continue to be necessary to obtain authoritative interpretation of EU antitrust law. Nevertheless, the Commission’s guidance has been very influential for NCAs and national courts (and even the CJEU has been deferent to it) such that it cannot be excluded that it is precisely the Commission’s soft law that stopped national courts from making preliminary references to the CJEU.53 It is difficult to come to affirmative conclusions on this. But it is interesting to note that in the most recent year for which statistics are available (2016), the number of preliminary questions is on the rise again.54 This could suggest that the CJEU is re-emerging as rule-maker in the field of competition law. V. Conclusion: the changing the role of the EU courts in EU antitrust enforcement As we have seen above, the modernisation of EU antitrust law has allowed the Commission to refocus its attention to cartel enforcement and this led to an upswing in the number of direct actions before the EU courts. The introduction of the settlement procedure in 2009 has, however, counteracted this development: with many cartel cases now being concluded by plea bargaining, fewer cartel cases are brought before the General Court. The reduced number of applications before the General Court of course also reduces the number of appeals before the CJEU which will therefore have fewer opportunities to pronounce itself on Commission procedures. But this does not necessary mean that the importance of the judgements of the EU courts in competition cases diminishes. First of all, in cartel cases, the existence of the infringement was often not in dispute: in their pleas before the courts, the undertakings mostly argued about the calculation of the fine they received (and factors affecting the fine such as parental liability or duration). The role of the EU courts in shaping the substance of EU antitrust law is limited in those cases. To the extent that the drop in direct actions to the EU courts is therefore the result of the introduction of the settlement procedure, this should not affect their role significantly. Of more concern for the role of the EU courts is the reduced number of preliminary references which seems to be apparent in the statistics in recent years. Indeed, preliminary rulings are generally more important for the development of the law than appeals, as was demonstrated in a recent study by Derlén and Lindholm. They assessed CJEU case law both by reference to inward citation (i.e. how many later judgements refer to a particular judgement) and outward citation (to how many previous judgements does a particular judgement refer). They concluded that preliminary rulings have higher inward and outward citations than (appeals against judgements in) direct actions.55 This is also true specifically for competition cases.56 The relative importance of preliminary rulings as compared to appeals is also apparent from the higher number of preliminary questions that result in Grand Chamber judgements. The CJEU sits in Grand Chamber only if a Member State or EU institution so requests and in particularly complex or important cases, which is also reflected in the higher inward and outward citation index they have.57 The statistics of the EU courts in competition cases confirm this: while there were 227 appeals brought before the CJEU in competition cases in the period 2007–2016, in the same period only 15 Grand Chamber judgements in appeals in competition cases58 were delivered (i.e. 7 per cent).59 By comparison: there were 13 Grand Chamber judgements on preliminary questions in competition cases and only 66 preliminary questions sent to the CJEU (20 per cent).60 Optimistically, one could argue that the drop in the number of preliminary references, may be the result of national judges becoming increasingly knowledgeable about EU antitrust law, so that they need to refer fewer questions. Or there may be less national opposition to EU antitrust rules (also as a consequence of harmonisation) so that fewer judges question the EU antitrust rules. Given the variations in the number of preliminary references from individual Member States (and even from individual courts within a Member State), the reasons for referring or not referring questions to the CJEU may in fact be quite idiosyncratic.61 It seems to me that the soft law guidance of the Commission has in any event played an important role in harmonising the interpretation of the antitrust rules across different enforcers, and thereby has helped reduce the need for preliminary references. In this way, the Commission seems to have become the main rule-making institution in EU antitrust law, even for NCAs and national courts. This development is certainly questionable as a matter of EU institutional law where the CJEU should have the last word in speaking the law.62 But quantity is not everything. While the total number of cases and even the number of preliminary references may have been lower than in the past, the CJEU in recent years still pronounced itself on a number of key questions of interpretation of Article 101 and 102 TFEU.63 Many of these judgements follow preliminary references in cases that originated in NCA proceedings and this may well explain why they concern fundamental questions of EU antitrust law which are furthermore fundamental to the outcome of the case.64 While decentralisation may not have resulted in more references, it may therefore have improved their quality. Through the preliminary reference procedure, the CJEU can remain, if not the main, at least the ultimate rule-maker in EU competition law.65 Annexes Cases before the EU Courts (1997–2016)    1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  General Court   Application to GC in competition (excl. State aid)  24  31  34  36  39  62  53  36  40  81  62  71  42  79  39  34  23  41  17  18   Total new cases before GC  644  238  384  398  345  411  466  536  469  432  522  629  568  636  722  617  790  912  831  974   Total GC completed cases in competition (excl. State aid)  21  41  43  61  22  miss.  38  26  35  42  38  31  31  38  100  61  75  72  52  36   Total GC completed cases  186  348  659  344  340  331  339  361  610  436  397  605  555  527  714  688  702  814  987  755  Court of Justice   Direct action to CJEU in competition (excl. State aid)  1  1  9  2  5  1  6  5  1  1  3  0  0  0  0  0  0  0  0  0   Appeal to CJEU in competition (excl. State aid)  9  14  13  14  6  11  8  18  2  15  20  7  10  13  52  23  32  15  32  23   Preliminary questions to CJEU in competition (excl. State aid)  14  13  7  6  15  1  6  9  7  14  8  3  5  5  7  6  6  8  6  12   Interim measures to CJEU in competition (excl. State aid)  0  0  0  0  4  0  1  1  0  2  0  0  2  2  1  1  1  0  2  0   Total new competition cases before CJEU (excl. State aid)  24  28  29  22  30  13  21  33  10  32  31  10  17  20  60  30  39  23  40  35   Total preliminary questions to CJEU  239  264  255  224  237  216  210  249  221  251  265  288  302  385  423  404  450  428  436  470   Total new cases before CJEU  445  485  543  503  504  477  561  531  474  537  580  592  561  631  688  632  699  622  713  692  Combined   Total new direct actions in competition law before GC and CJEU  25  32  43  38  44  63  59  41  41  82  65  71  42  79  39  34  23  41  17  18   Total new competition cases before GC and CJEU  39  45  50  44  59  64  65  50  48  96  73  74  47  84  46  40  29  49  23  30  Cases before the EU Courts (1997–2016)    1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  General Court   Application to GC in competition (excl. State aid)  24  31  34  36  39  62  53  36  40  81  62  71  42  79  39  34  23  41  17  18   Total new cases before GC  644  238  384  398  345  411  466  536  469  432  522  629  568  636  722  617  790  912  831  974   Total GC completed cases in competition (excl. State aid)  21  41  43  61  22  miss.  38  26  35  42  38  31  31  38  100  61  75  72  52  36   Total GC completed cases  186  348  659  344  340  331  339  361  610  436  397  605  555  527  714  688  702  814  987  755  Court of Justice   Direct action to CJEU in competition (excl. State aid)  1  1  9  2  5  1  6  5  1  1  3  0  0  0  0  0  0  0  0  0   Appeal to CJEU in competition (excl. State aid)  9  14  13  14  6  11  8  18  2  15  20  7  10  13  52  23  32  15  32  23   Preliminary questions to CJEU in competition (excl. State aid)  14  13  7  6  15  1  6  9  7  14  8  3  5  5  7  6  6  8  6  12   Interim measures to CJEU in competition (excl. State aid)  0  0  0  0  4  0  1  1  0  2  0  0  2  2  1  1  1  0  2  0   Total new competition cases before CJEU (excl. State aid)  24  28  29  22  30  13  21  33  10  32  31  10  17  20  60  30  39  23  40  35   Total preliminary questions to CJEU  239  264  255  224  237  216  210  249  221  251  265  288  302  385  423  404  450  428  436  470   Total new cases before CJEU  445  485  543  503  504  477  561  531  474  537  580  592  561  631  688  632  699  622  713  692  Combined   Total new direct actions in competition law before GC and CJEU  25  32  43  38  44  63  59  41  41  82  65  71  42  79  39  34  23  41  17  18   Total new competition cases before GC and CJEU  39  45  50  44  59  64  65  50  48  96  73  74  47  84  46  40  29  49  23  30  Source: annual reports of the CJEU and the General Court: <https://curia.europa.eu/jcms/jcms/Jo2_7000/en/> accessed 19 October 2017. (II) Commission decisions in antitrust (2004–2016)    2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  Source  Cartel decisions (incl. settlements)  6  5  7  8  7  6  7  4  5  4  10  5  6  (1)(2)(3)(4)  Settlement decisions  0  0  0  0  0  0  2  3  2  2  8  2  3  (2)(5)  Number of undertakings settling  0  0  0  0  0  0  15  12  8  10  30  5  7  (2)(5)  Other 101/102 decisions  5  2  1  4  1  1  1  1  0  3  4  0  1  (3)(4)(5)  Commitment decisions  0  2  4  5  1  5  5  2  4  4  2  2  3  (3)(4)(5)  Other decisions (e.g. rejection of complaints)  1  1  1  0  4  1  0  1  1  1  5  3  1  (3)(4)(5)  Total Commission decisions in antitrust (excl. concentrations and State aid)  12  10  13  17  13  13  14  8  10  12  21  10  11  (3)(4)(5)  (II) Commission decisions in antitrust (2004–2016)    2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  Source  Cartel decisions (incl. settlements)  6  5  7  8  7  6  7  4  5  4  10  5  6  (1)(2)(3)(4)  Settlement decisions  0  0  0  0  0  0  2  3  2  2  8  2  3  (2)(5)  Number of undertakings settling  0  0  0  0  0  0  15  12  8  10  30  5  7  (2)(5)  Other 101/102 decisions  5  2  1  4  1  1  1  1  0  3  4  0  1  (3)(4)(5)  Commitment decisions  0  2  4  5  1  5  5  2  4  4  2  2  3  (3)(4)(5)  Other decisions (e.g. rejection of complaints)  1  1  1  0  4  1  0  1  1  1  5  3  1  (3)(4)(5)  Total Commission decisions in antitrust (excl. concentrations and State aid)  12  10  13  17  13  13  14  8  10  12  21  10  11  (3)(4)(5)  Sources: Commission cartel statistics <http://ec.europa.eu/competition/cartels/statistics/statistics.pdf> accessed 19 October 2017 R. Snelders, ‘The EU cartel settlement procedure: the first years’ (2016) Presentation to the Studienvereinigung Kartellrecht<https://www.studienvereinigung-kartellrecht.de/sites/default/files/14h_30_snelders_sv_kartellrecht_-_cartel_settlements_final_0.pdf> accessed 19 October 2017 OECD, ‘Commitment decisions in antitrust cases—Background paper by the Secretariat’ (2016) <https://one.oecd.org/document/DAF/COMP(2016)7/en/pdf> accessed 19 October 2017 Commission, ‘Ten years of antitrust enforcement under Regulation 1/2003: achievements and future perspectives’ COM(2014)453 <http://ec.europa.eu/competition/antitrust/legislation/antitrust_enforcement_10_years_en.pdf> accessed 19 October 2017 Own research for cases in 2014-2016 using the Commission’s search tool <http://ec.europa.eu/competition/elojade/isef/index.cfm?clear=1&policy_area_id=1> accessed 19 October 2017 Footnotes 1 DG Goyder, EC Competition Law (Clarendon 1993) 71. 2 On the latter in particular, see BJ Rodger, ‘Article 234 and competition law: a comparative analysis’ (2008) 15 Maastricht Journal of European and Comparative Law 149, 151-152 and references there, and DP Domenicucci, ‘Preliminary rulings and competition law: some reflections for national judges’ in B Cortese (ed), EU competition law: Between public and private enforcement (Wolters Kluwer 2014) 179, 211. 3 G Slynn, ‘Court of First Instance of the European Communities’ (1989) 9 Northwestern Journal of International Law & Business 542, 543. 4 Idem, 545. 5 AG Cosmas in case C-344/98 Masterfoods v HB ECLI:EU:C:2000:249, 54. 6 Case C-6/72 Europemballage and Continental Can v Commission ECLI:EU:C:1973:22. 7 Cases T-342/99 Airtours v Commission ECLI:EU:T:2002:146; T-310/01 Schneider Electric v Commission ECLI:EU:T:2002:254; T-5/02 Tetra Laval v Commission ECLI:EU:T:2002:264. 8 Case 27/76 United Brands v Commission ECLI:EU:C:1978:22. See S Bishop and M Walker, The economics of EC competition law (Sweet & Maxwell 2010) 4-027. 9 See I Forrester, ‘A challenge for Europe’s judges: the review of fines in competition cases’ (2011) European Law Review 185, in particular 194-196, and D Gerard, ‘Breaking the EU antitrust enforcement deadlock: re-empowering the courts?’ (2011) 36 European Law Review 457, 469-471. 10 See, for example, I Forrester, ‘A bush in need of pruning: the luxuriant growth of light judicial review’ (2011) Competition Law Annual 407; Gerard above n 10, 472 and J Derenne, ‘The scope of judicial review in EU economic cases’ in J Derenne and M Merola (eds), The Role of the Court of Justice of the European Union in Competition Law Cases (Larcier 2012) 73, in particular 83–85. 11 This is for example the case for the claim that the General Court shows more deference to the Commission in cases under Article 102 TFEU than in cases under Article 101 TFEU, a claim made in D Geradin and N Petit, ‘Judicial review in European Union competition law: a quantitative and qualitative assessment’ (2010) Tilburg Law and Economics Center (TILEC) Law and Economics Discussion Paper No. 2011-008 <https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1698342> accessed 23 October 2017, in particular 31. Contrary to what the authors claim, their data does not show a statistically significant difference in the number of annulments of decisions adopted under Article 102 TFEU than under Article 101 TFEU, taking into account the number of decisions contested (which is much lower for Article 102 TFEU than for Article 101 TFEU). The number of Article 102 decisions that is annulled in full or in part is even slightly higher than the number of Article 101 decisions: 28.6 per cent (4/14) as compared to 26.5 per cent (31/117). And the fact that there was no full annulment of any of the 14 Article 102 decision in their data sample is not surprising since also less than 1 in every 14 Article 101 TFEU decisions in their sample was annulled in full (8 out of 117). 12 The Annexes contain all the data used in the figures in this paper, as well as their source. Note that the statistics of the EU courts are based on the case numbers allocated by the Court even though one case number does not always correspond to one case. For example, it is quite common for the European Commission to impose a cartel fine on a number of legal entities who are jointly and severally liable for the fine because they are considered part of the same undertaking. The entities may bring one joint application for annulment against this fine or they may bring separate applications: which of these choices they make, may depend on a number of factors, such as whether they have different interests or different arguments which they want to bring to the court. In the statistics of the General Court and the Court of Justice each application will be assigned a case number and therefore be viewed as a single new case (even if they are afterwards joined for the hearing or for judgement): see the annual reports of the ECJ in 2005 and previous years; for example, the annual report of 2005, page 191. Also in the case of preliminary questions, what counts as one case does not depend on the Court of Justice. If a number of cases are joined at the national level, the national court will send only a single request for a preliminary ruling to the Court of Justice, resulting in a single case number. See for example, case C-419/12 Crono Service scarl and Others ECLI:EU:C:2014:81, which concerned an application by Crono Services scarl and 111 other applicants against Roma Capitale for the annulment of local legislation on car and driver hire services. On the other hand, if cases are not joined at the national level, the national court may send several requests for a preliminary ruling (even if they contain the exact same questions), resulting in several case numbers. See for example, cases C-159/12 to C-161/12 Venturini and Others ECLI:EU:C:2013:791, which concerned application by three pharmacists against decisions of the local health authorities refusing the sale of certain medicinal products in para-pharmacies. 13 A good example of this is in the field of merger control where the interventionism of the General Court resulted in a peak of applications for annulment of 4–5 per year in the early 2000s, while the more hands-off approach in subsequent years led to a decline in the number of cases to 0-3 per year since the end of the 2000s until today. 14 The three factors are discussed in more detail in P Ibáñez Colomo, ‘Three shifts in EU competition policy: towards standards, decentralisation, settlements’ (2013) 20 Maastricht Journal of European and Comparative Law 363. 15 Article 4 of Regulation 17/62 First Regulation implementing Articles 85 and 86 of the Treaty, [1962] OJ 13/204. Undertakings could also request a negative clearance from the Commission on the basis of Article 2 of Regulation 17/62 to the effect that (what is now) Article 101(1) TFEU and Article 102 TFEU were not applicable to an agreement or practice in which the undertaking engaged. 16 This is probably a better term than ‘directly effective’ because the direct effect of the provision results directly from the Treaty of Rome: its application was merely suspended when Article 9(1) of Regulation 17/62 granted the Commission the sole power to apply (what is now) Article 101(3) TFEU. See P Landolt, Modernised EC Competition Law in International Arbitration (Kluwer 2006) 37. 17 In the period May 2004 to December 2013, the Commission adopted 122 decisions applying the Article 101 and/or 102 TFEU and the NCAs 665. See Commission, ‘Ten years of antitrust enforcement under Regulation 1/2003: achievements and future perspectives’ COM(2014)453, 4. 18 See the Commission, ‘Proposal a Directive of the European Parliament and of the Council to empower the competition authorities of the Member States to be more effective enforces and to ensure the proper functioning of the internal market’ COM(2017)142 final, 2 which continues to report that the NCAs were responsible for 85 per cent of enforcement decisions. 19 See, for example, C-D Ehlermann and I Atanasiu, ‘The modernisation of E.C. antitrust law: consequences for the future role and function of the E.C. courts’ (2002) European Competition Law Review 72, 76; K Lenaerts and D Gerard, ‘Decentralisation of EC competition law enforcement: judges in the frontline’ (2004) 27 World Competition 313, 341 and T Cowen, ‘’Justice delayed is justice denied’: the rule of law, economic development and the future of the European community courts’ (2008) European Competition Journal 1, 39. 20 See Rodger above n 3, 159–160. 21 M Prek and S Lefèvre, ‘Competition litigation before the General Court: quality if not quantity?’ (2016) 53 Common Market Law Review 65, 66. 22 See the Commission’s own cartel statistics which refer to 10 cases in the period 1990–1994, 10 cases in 1995–1999 but 30 cases in 2000–2004, 33 cases in 2005–2009 and 30 cases in 2010–2014. In 2015 there were 5 cases and in 2016 6. See <http://ec.europa.eu/competition/cartels/statistics/statistics.pdf> accessed 12 October 2017. 23 Idem. 24 As predicted by some. See Ehlermann and Atanasiu above n 20, 74 and Lenaerts and Gerard above n 20, 341. 25 Commission Regulation 622/2008 of 30 June 2008 amending Regulation (EC) No 773/2004, as regards the conduct of settlement procedures in cartel cases [2008] OJ L171/3; Commission Notice on the conduct of settlement procedures in view of the adoption of Decisions pursuant to Article 7 and Article 23 of Council Regulation (EC) No 1/2003 in cartel cases [2008] OJ C167/1. Note that negotiations can also take place outside this formal framework (see for example the Commission decision of 20 September 2016 in case 39.759 ARA foreclosure where ARA received a 30 per cent fine reduction for cooperation, including acknowledging the infringement and proposing a structural remedy). The discussion of the use of commitments in EU merger procedures is outside of the scope of this paper. 26 Commission, ‘Notice on best practices for the conduct of proceedings concerning Articles 101 and 102 TFEU’ [2011] OJ C308/6. 27 See D Waelbroeck, ‘Le développement en droit européen de la concurrence des solutions négociées (engagements, clémence, non-contestation des faits et transactions): que va-t-il rester aux juges?’ (2008) GCLC Working Paper 1/08 <https://www.coleurope.eu/system/files_force/research-paper/gclc_wp_01-08.pdf?download=1>accessed 19 October 2017, 3 and F Jenny, ‘Worst decision of the EU Court of Justice: the Alrosa judgement in context and the future of commitment decisions’ (2015) 38 Fordham International Law Journal 701, 735–736. 28 See E Barbier de La Serre, ‘Competition law cases before the EU courts: is the well running dry?’ in J Derenne and M Merola (eds), The Role of the Court of Justice of the European Union in Competition Law Cases (Larcier 2012) 87, 98–99 and Prek and Lefèvre above n 22, 66. See also AG Wahl, as reported in ‘Increase in EU settlements risks sidelining courts, victims, Wahl says’ MLex (20 October 2014). 29 See in particular W Wils, ‘Ten years of commitment decisions under Article 9 of Regulation 1/2003: too much of a good thing?’<https://papers.ssrn.com/sol3/Papers.cfm?abstract_id=2617580>accessed 22 August 2017, 10. The importance of informal commitments to close cases under Regulation 17/62 is also emphasised for instance by Waelbroeck above n 28, 5 and Ibáñez Colomo above n 15, 368. 30 Wils above n 30, 6–9. 31 See the statistics in the Annexes. 32 Regulation 773/2004 relating to the conduct of proceedings by the Commission pursuant to Articles 81 and 82 of the EC Treaty [2004] OJ L123/18. 33 Up to 2016 there were 5 so-called hybrid settlement cases were one or more undertakings were fined under the ordinary procedure while the other cartel members settled the case: cases 38.866 Animal feed phosphates, 39.861 Yen interest rate derivatives, 39.914 Euro interest rate derivatives, 39.965 Mushrooms and 39.792 Steel abrasives. 34 See the Q&A on settlements on the Commission’s website:<http://ec.europa.eu/competition/cartels/legislation/cartels_settlements/settlements_en.html>accessed 23 October 2017. 35 At the moment of concluding this paper, there were only two examples. The first was the application for annulment filed by Société Générale in case 39.914 Euro interest rate derivatives on grounds of differential treatment (case T-98/14 Société générale v Commission) which was withdrawn after Société Générale provided corrected value of sales data to the Commission and subsequently saw its fine reduced by the Commission. The second case is the application for annulment filed by Printeos in case 39.780 Envelopes which resulted in a judgement of the General Court on 13 December 2016 (case T-95/15 Printeos and others v Commission ECLI:EU:T:2016:722) in which the General Court annulled the Commission decision because of insufficient reasoning concerning discretionary fine reductions; but the Commission readopted its decision and imposed an identical fine on Printeos on 16 June 2017. 36 However, as pointed out in footnote 14, there is not a one-to-one relationship between the number of undertakings subject to a cartel investigation as identified by the Commission and the number of applicants before the EU courts. 37 Although there appears to be a direct correlation between the evolution of the number of settlements and the evolution of the number of direct actions in 2013-2016, it is hard to see what causal link there might be between the two, except perhaps the external factor that 2014 was the last year of the tenure of Joaquín Almunia as Commissioner for Competition, which may have resulted in a peak in the number of settlement decisions (8) as well as other decisions (13). 38 M Hellwig, K Hüschelrath and U Laitenberger, ‘Settlements and appeals in the European Commission’s cartel cases: an empirical assessment’ (Zentrum für Europäische Wirtschaftsforschung GmbH, January 2016)<http://ftp.zew.de/pub/zew-docs/dp/dp16010.pdf> accessed 16 August 2017. 39 See on this aspect: Ehlermann and Atanasiu above n 20, 73. 40 Note that some guidance, such as the ‘Notice on the definition of the relevant market for the purposes of Community competition law [1997] OJ C372/5, is relevant for both merger control and antitrust. 41 Commission, ‘Green Paper on vertical restraints in EC competition policy’ COM (96) 721, <http://europa.eu/documents/comm/green_papers/pdf/com96_721_en.pdf> accessed 9 October 2017. 42 There were also several notices on jurisdiction and procedure (such as on fines, leniency, effect on trade between Member States and the de minimis notice) but these again have less importance from the perspective of the regime of legal exception and decentralisation. 43 Commission, ‘Guidelines on vertical restraints’ [2000] OJ C291/1. 44 Commission, ‘Guidelines on the application of Article 81 of the EC Treaty to horizontal cooperation agreements’ [2001] OJ C3/2. 45 Commission, ‘Guidelines on the application of Article 81 of the Treaty on technology transfer agreements’ [2004] OJ C101/2. 46 Commission, ‘Guidelines on the application of Article 81(3) of the Treaty’ [2004] OJ C101/97. 47 Commission, ‘Guidelines on vertical restraints’ [2010] OJ C130/1. 48 Commission, ‘Guidelines on the application of Article 101 of the Treaty on the Functioning of the European Union to horizontal cooperation agreements’ [2011] OJ C11/1. 49 Commission, ‘Guidelines on the application of Article 101 of the Treaty on the Functioning of the European Union to technology transfer agreements’ [2014] OJ C89/3. 50 Commission, ‘Guidance on tis enforcement priorities in applying Article 82 of the EC Treaty to abusive exclusionary conduct by dominant undertakings’ [2009] OJ C45/7. 51 Ibáñez Colomo above n 15, 370. 52 There has only been one question relating to mergers: case C-248/16 Austria Asphalt ECLI:EU:C:2017:643. 53 This is also the belief of Barbier de La Serre above n 29, 96. 54 A look at the figures for the first half of 2017 is inconclusive as to whether this trend will continue: there were four requests for preliminary rulings in the field of competition (excluding State aid) in January–June 2017: cases C-27/17, flyLAL-Lithuanian Airlines, C-144/17, Lloyd’s of London, C-152/17, Consorzio Italian Management and C-267/17 Rhenus Veniro. 55 M Derlen and J Lindholm, ‘Characteristics of precedent: the case law of the European court of justice in three dimensions’ (2015) 16 German Law Journal 1073, 1082-1085. This applied even to different segments of direct actions, with the exception of outward citations (so references to previous case law), where unsuccessful infringement proceedings top the chart. 56 Idem, 1088–1090. 57 Idem, 1092. 58 Note that five of these judgements concerned two related cases, so one could also say that in only 12 competition cases on appeal a Grand Chamber judgement was issued. 59 Note that this is a comparison of the number of appeals brought in 2007–2016 and the number of judgements on appeal in that same period (some of which relate to appeals brought prior to 2007), so the figures do not relate to the same cases. The figures listed therefore merely indicate a tendency. 60 The previous footnote applies mutatis mutandis. 61 See, for a discussion on the practice of individual Member States to refer preliminary questions in competition cases and speculation on the reasons for this: Rodger above n 3, 149. 62 K Lenaerts, ‘The rule of law and the coherence of the judicial system of the European Union’ (2007) 44 Common Market Law Review 1625, 1652. 63 For example, in the field of Article 101 TFEU, on the conditions for unlawful exchange of information in case C-74/14 Eturas and Others ECLI:EU:C:2016:42, in the field of Article 102 TFEU, on the relevance of the as-efficient-competitor test in cases C-209/10 Post Danmark v konkurrencerådet ECLI:EU:C:2012:172 (and later in C-413/14 P Intel v Commission ECLI:EU:C:2017:632) and in procedural matters on the circumstances were exculpatory evidence can be accepted in case C-681/11 Schenker & Co. and Others ECLI:EU:C:2013:404. 64 I thank an anonymous reviewer of the Journal of European Competition Law & Practice for this suggestion. 65 See also N Petit, ‘The future of the court of justice in EU competition law’ in Court of Justice of the European Union (ed), The Court of Justice and the Construction of Europe: Analyses and Perspectives on Sixty Years of Case-law - La Cour de Justice et la Construction de l’Europe: Analyses et Perspectives de Soixante Ans de Jurisprudence (T.M.C. Asser Press 2013) 397–421. © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of European Competition Law & Practice Oxford University Press

The Impact of EU Antitrust Procedure on the Role of the EU Courts (1997–2016)

Loading next page...
 
/lp/ou_press/the-impact-of-eu-antitrust-procedure-on-the-role-of-the-eu-courts-1997-yvf5NS5XAG
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com
ISSN
2041-7764
eISSN
2041-7772
D.O.I.
10.1093/jeclap/lpy004
Publisher site
See Article on Publisher Site

Abstract

Key Points The decentralisation of EU antitrust enforcement did not result in a drop in the number of applications for annulment of Commission decisions, nor in an increase in the number of preliminary references. The introduction of the formal commitments procedure did not, but the introduction of the settlement procedure in cartels did cause a significant drop in the number of applications for annulment. The low number of preliminary references in competition cases in recent years may be influenced by the soft law instruments issued by the Commission. I. Introduction For better or worse, the EU courts have played an important role in the development of EU competition law. This is not surprising: since its foundation, the CJEU (then ECJ) has had a profound impact on European integration in general and the development of European law in particular. In the early days of antitrust enforcement, ‘the support given by the Court [of Justice] to the work of the Commission was of critical importance in enabling its main lines of policy to be established.’1 But also in subsequent years, the CJEU opened and closed enforcement paths of the Commission and for private plaintiffs before national courts, both through direct actions (applications for annulment) against Commission decisions and through its preliminary rulings procedure.2 After its creation in 1989, the Court of First Instance (now General Court) assumed its own role in shaping EU competition law. The Court of First Instance was set up in response to the increased number of cases brought before the CJEU (and a growing backlog) because of the geographical and substantial enlargement of the European Communities in the 1970s and 1980s. In particular in the field of competition law, it was also felt that the CJEU ‘was not in the best position to review … serious issues of fact or economic assessment’.3 The Court of First Instance would allow ‘a fuller investigation of factual issues [to] take place’ in these cases4 or even an ‘exhaustive review’ of the ‘complex technical and economic assessments’ of the Commission.5 Much has been written about the quality of the case law of the CJEU and General Court and how their influence has sometimes been a boon for and sometimes an obstacle to effective antitrust enforcement. The CJEU’s 1973 judgement in Continental Can, for example, basically opened up the entire field of merger control as an area of enforcement.6 Similarly, the General Court’s annulment of three merger decisions in 20027 resulted in more robust economic reasoning in the field of merger control and the creation of the Chief Economist team at DG Competition in 2003. On the other hand, the ‘toothless fallacy’ accepted by the CJEU in United Brands continues to discredit the use of market definitions in competition cases.8 In recent years, several practitioners have also criticised the CJEU and the General Court for softening their review of Commission decisions: the EU courts would take a too deferential approach to the Commission’s fine calculations in cartel cases9 and would leave too much margin of appreciation for the Commission’s economic assessments.10 The aim of this paper is not to add another gloss to the discussion on the appropriate review powers of the General Court and the CJEU. By way of an analysis of the statistics of the EU courts, it rather aims to respond to the empirical question: what is the actual role of the EU courts in competition cases today? This question should precede the discussion on the review powers of the EU courts, lest the latter be based on false premises.11 The focus of this paper is limited in scope: its statistical assessment does not cover the outcome of proceedings before the General Court and the CJEU (the number and content of judgements) but merely its input (the number of new cases brought before the EU Courts).12 To some extent the CJEU and General Court themselves can influence that input. Through their case law, they set the boundaries of their review, in terms of which cases are admissible and which arguments have a good chance of success. The more interventionist the EU courts are, the more likely it is that undertakings will bring applications for annulment against Commission decisions.13 To a lesser extent, the CJEU’s approach to preliminary questions can also influence the number of questions put to it. Also the duration of proceedings before the EU courts impacts on how attractive procedures before them are. This paper focuses on another factor that may impact on the workload of the EU courts and the role they play in competition law enforcement. This is the procedural framework for enforcement at the level of the Commission and national competition authorities (NCAs), in particular in light of the changes brought about since the so-called ‘modernisation’ of European antitrust in the late 1990s and early 2000s. I will examine in turn the following three procedural changes14: the decentralisation of antitrust enforcement; the introduction of (formally) negotiated resolutions to cases; and the use of soft law instruments by the European Commission. The conclusions of my analysis are as follows: Contrary to common belief and expectations, the decentralisation of EU antitrust enforcement did not result in a drop in the number of applications for annulment of Commission decisions, nor in an increase in the number of preliminary references. While there does not seem to be any evidence of a reduction in the number of applications for annulment following the introduction of the formal commitments procedure in Article 9 of Regulation 1/2003, the introduction of the settlement procedure in cartels in 2010 did cause a significant drop in the number of applications for annulment. Although the number of preliminary references in competition cases has always been small, it cannot be excluded that the reduction in the number of references in recent years is influenced by the soft law instruments issued by the Commission. II. Decentralisation Regulation 17/62 provided that undertakings who wanted to avail themselves of the exception of (what is now) Article 101(3) TFEU, needed to notify their agreements to the Commission.15 This system was abolished when Article 1 of Regulation 1/2003 introduced a system of legal exceptions meaning that (what is now) Article 101(3) TFEU became directly applicable16 and undertakings had to self-assess whether their agreements were in accordance with it. This change also implied that national courts could now rule on all aspects of antitrust disputes (whereas they before had to refuse jurisdiction if a party invoked what is now Article 101(3) TFEU). In parallel, Article 3(1) of Regulation 1/2003 imposed an obligation on NCAs to apply (what is now) Article 101 and 102 TFEU each time they applied their national competition law to an agreement or practice that affected trade between the Member States. National courts were therefore empowered and NCAs enlisted to apply the European antitrust rules alongside the Commission. This decentralisation of EU antitrust enforcement has significantly changed the enforcement architecture. In July 2014, the Commission reported to the European parliament and the Council that in the first 10 years since the entry into force of Regulation 1/2003, the vast majority of decisions on Article 101 and 102 TFEU had been taken by NCAs.17 This development has continued unabated since.18 How did this affect the EU courts? In light of the increased role of national courts in EU antitrust enforcement both for disputes between private parties and on appeal against decisions of the NCAs, many predicted a rise in the number of preliminary references in this field.19 However, this is not borne out by the statistics of the CJEU. Figure 1 shows how the number of preliminary questions in competition cases (excl. State aid) has fluctuated over the last 20 years (dated by the year in which the question was received by the CJEU). While the numbers are limited and, in line with the law of small numbers, it is therefore difficult to draw firm conclusions from peaks and troughs, there has not been a significant increase in the period after 2004, compared to the period before. On the contrary, since many national competition cases are handled by NCAs and only come before a court or tribunal in the sense of Article 267 TFEU after a few years, one would expect to see a particular increase from, say, 2007 onwards, but precisely from then on, the figures have been particularly low (with the exception of 2016). Figure 1: View largeDownload slide Preliminary questions in competition cases (excl. State aid). Figure 1: View largeDownload slide Preliminary questions in competition cases (excl. State aid). This (lack of) development is even more striking if compared to the increase of the number of preliminary references in competition matters over previous decades20 as well as the continuous increase of preliminary references in other fields of European law since the mid-2000s, as shown in Figure 2. Figure 2: View largeDownload slide Preliminary questions (all subjects). Figure 2: View largeDownload slide Preliminary questions (all subjects). I will explore below possible explanations for the perceived evolution of the number of preliminary references in competition cases. It has been argued that the decentralised application of EU competition law, has also led to a drop in the number of direct actions before the EU courts (in first instance, mainly at the level of the General Court).21 But it is not obvious to interpret the evolution of the number of direct actions (contained in Figure 3) in this sense. Figure 3: View largeDownload slide Direct actions in competition cases (excl. State aid). Figure 3: View largeDownload slide Direct actions in competition cases (excl. State aid). If anything, the number of direct actions seems to have increased after the entry into force of Regulation 1/2003 and only starts to fall from 2011 onwards. As we will see further on, there is a specific explanation for the drop in the last five years. The idea that decentralisation would mean fewer direct actions is based on the premise that the Commission would adopt fewer decisions after decentralisation. But that is not what the Commission did. In fact, already in the years prior to the entry into force of Regulation 1/2003, the Commission had started to refocus its investigations. The ‘more economic approach’ to antitrust propagated since the middle of the 1990s had led the Commission to turn its attention away from vertical restraints (which Chicago School economists viewed as generally unproblematic) and towards the most egregious of antitrust offences: cartels. While in the 1990s, the Commission adopted only 2 cartel decisions on average every year, since 2000 it has adopted on average 6.22 Decentralisation has therefore not simply led to fewer decisions by the Commission but to different decisions (fewer on verticals, more on cartels). Not only did the Commission increase its cartel enforcement, also the level of fines went up significantly: while the total amount of cartel fines in 1995–1999 was EUR 293 million, this increased more than tenfold in 2000–2004 to EUR 3.5 billion and again almost tripled in 2005–2009 to EUR 9.4 billion.23 It is this increasing number of cartel decisions and the increasing level of fines imposed for cartel infringements which coincides with the increased number of applications for annulment to the General Court.24 As a consequence, the number of direct actions has increased from 25 in 1997 (the first year for which statistics on the judicial activity of the courts are available online) to a peak of 82 in 2006. III. Negotiated procedures A second development in the enforcement of EU competition law, relates to the use of the negotiated conclusion of investigations. The European procedural rules for antitrust enforcement provide for the negotiated conclusion of cases in two forms: through the commitment procedure pursuant to Article 9 of Regulation 1/2003 and through the settlement procedure in cartel cases introduced in 2008.25 Under Regulation 17/62 it was already possible for the Commission to attach conditions and obligations to its decisions, but this assumed that it had established an infringement. In parallel, the Commission developed a practice of informally concluding investigations by means of negotiated amendments to the agreements or practices it investigated, but the enforcement of such negotiated amendments was problematic, since it was not foreseen by Regulation 17/62. Article 9 of Regulation 1/2003 formally introduced a commitments procedure pursuant to which parties can offer commitments to address concerns the Commission has as regards an agreement or practice to which they are a party. If the Commission considers that these commitments are adequate, it can make them binding on the undertaking and conclude that there is no longer ground to continue its investigation. The undertaking in this way avoids a finding of infringement (which entails reputational damage and facilitates third-party damage actions) and a fine. The advantage for the Commission is that it avoids protracted litigation over the alleged infringement while still having enforcement tools at hand to ensure compliance with the commitments offered (in particular Article 23(2)(c) of Regulation 1/2003 which allows it to fine an undertaking which does not comply with a commitment made binding by a decision). According to recital (13) of Regulation 1/2003, commitment decisions are not appropriate in cases where the Commission intends to impose a fine and the Commission has stated that it will not use the procedure if the nature of the infringement calls for the imposition of a fine, such as in the case of secret cartels.26 The commitment procedure is therefore used to combat practices which the Commission believes are harmful to competition but where this harm may not be self-evident to the undertaking(s) concerned. This allows the Commission to be more flexible in enforcement but at the same time this implies that its enforcement practice becomes less transparent (both as regards the undertaking which is the subject of the commitment decision and third parties).27 Of specific relevance for this paper, is the prediction and affirmation that the increasing use of the commitment procedure would reduce the number of applications to the EU courts resulting in less judicial scrutiny of Commission decisions.28 However, it is hard to demonstrate this by way of the statistics of the EU courts. Figure 4 shows that, after the introduction of the commitments procedure in 2004 and the first commitment decisions in 2005 and 2006, the number of applications for annulment to the General Court did not decrease—on the contrary. Figure 4: View largeDownload slide Commitments decisions v direct actions in competition cases. Figure 4: View largeDownload slide Commitments decisions v direct actions in competition cases. More importantly, it is hard to compare the situation following the introduction of the formal commitment procedure through Article 9 of Regulation 1/2003 with the prior situation, because, as indicated above, the Commission already informally accepted commitments under Regulation 17/62 (even though there were doubts as to the enforceability of these commitments). There are no full statistics of informal commitments under Regulation 17/62 but there is no doubt that informal commitment procedures were also common under Regulation 17/62, so that the introduction of the formal commitment procedure of Article 9 of Regulation 1/2003 may not have significantly reduced the number of infringement decision or the number of applications for annulment to the EU courts.29 Also the drop in recent years cannot be explained by an increased use of commitment decisions: under Commissioner Almunia (2010–2014) the number of commitment decisions was not significantly higher than under Commissioner Kroes (2005–2009),30 nor has it increased since the appointment of Commissioner Vestager.31 Let us then turn to the second negotiated procedure in antitrust, the settlement procedure. This procedure was introduced, not by Regulation 1/2003 but by an amendment to the Commission’s Implementing Regulation32 in 2008. The settlement procedure only applies to cartel cases: it allows cartel participants to obtain a fine reduction of 10 per cent if they acknowledge their participation in the cartel. In this way, the Commission obtains a swifter procedure, as it no longer needs to proceed through the entire evidentiary process to establish the infringement. The settlement procedure rapidly became the norm in cartel cases: between 2010 (when it was first used) and 2016, over half of the cartel decisions adopted by the Commission were settlement decisions. The settlement procedure allows the Commission to wrap up investigations faster and with less effort: undertakings can no longer argue about the ins and outs of every allegedly anticompetitive contact and have to focus on a number of key parameters that will affect their fine level. In most cases, this has allowed the Commission to settle with all cartel participants, although it does not fail to pursue cases where one or two cartel members try to escape fines by staying out of the settlement process.33 For the undertakings involved, the 10 per cent discount of the fine is not the only advantage they obtain. Since they know that the settlement process entails efficiencies for the Commission, they can also try to negotiate key parameters for the fine that are relevant for them—even though the Commission has denied that the fine is negotiable.34 Furthermore, the Commission decision of a settlement is much less detailed than ordinary cartel decisions: it is usually only one or two dozen pages, while ordinary cartel decisions can run in the hundreds of pages. This makes a settlement decision much less useful for damage claimants. Let us then turn to the effect of the settlement procedure on the applications before the EU courts. Although there is no legal bar to bringing an application for annulment against a settlement decision, the number of such actions has been extremely limited.35 As a consequence, cartel cases concluded through settlement are much less likely to result in cases before the EU courts than ordinary cartel decisions. The dotted orange line in Figure 5 shows the number of undertakings which concluded a settlement (rather than the number of cases in which a settlement was concluded) since this better reflects the potential number of direct actions that were avoided through settlement.36 Figure 5 indeed shows that the number of applications for annulment before the General Court dropped significantly in 2011, the year after which the first settlement decisions were concluded.37 This drop is even more striking if compared to the number of direct actions in other fields (in particular in the field of intellectual property), which continue to increase year after year. Figure 5: View largeDownload slide Cartel settlements v direct actions in competition cases. Figure 5: View largeDownload slide Cartel settlements v direct actions in competition cases. A study of the German Centre for European Economic Research estimated that, excluding other factors that might affect the willingness to appeal, the introduction of the settlement procedure has reduced the number of applications for annulment in cartel cases by 20–55 per cent.38 This indeed demonstrates that the drop in the number of applications in recent years is at least partially due to the introduction of the settlement procedure. IV. Soft law instruments The third development discussed in this paper is the increased use by the Commission of soft law instruments, i.e. guidelines and notices to clarify its policy. This tool has been used by the Commission since the middle of the 1990s and has been one of the hallmarks of the modernisation of EU competition law. Indeed, the ‘more economic approach’ was difficult to reconcile with the formalism of block exemption regulations which was a key enforcement tool for the Commission under Regulation 17/62 (if only to keep the number of notifications of agreements and practices under control).39 Furthermore, although the EU courts and the Commission in its decisions had provided clarifications on many aspects of EU competition law, there were other fields where guidance was lacking or at least not straightforward. By issuing formal guidance, the Commission facilitated the task of undertakings (and their external advisers) who had to self-assess the compliance of agreements and practices with the EU antitrust rules under the regime of legal exception of Regulation 1/2003. It also allowed for greater consistency in the application of Article 101 and 102 TFEU by the NCAs and national courts which were at the forefront of enforcement after decentralisation. While the more economic approach was obviously also apparent in the field of merger control, the other two drivers of soft law discussed above (regime of legal exception and decentralisation) only applied in relation to EU antitrust and I will therefore focus on the soft law in this field here.40 A first wave of soft law guidance was initiated with the 1997 Green Paper on vertical restraints41 and resulted in the following soft law instruments on the interpretation of Article 101 TFEU in the period 2000–2004:42 The 2000 guidelines on vertical restraints,43 The 2001 guidelines on horizontal cooperation agreements,44 The 2004 guidelines on technology transfer agreements,45 and The 2004 guidelines on the application of Article 101(3) TFEU.46 The first three of these guidelines were updated a decade later following a consultation process initiated in 2009 and resulting in the following instruments: The 2010 guidelines on vertical restraints,47 The 2011 guidelines on horizontal cooperation agreements,48 and The 2014 guidelines on technology transfer agreements.49 In between these two waves of substantive guidance on Article 101 TFEU, the Commission also held a lengthy consultation process on the interpretation of Article 102 TFEU which resulted in the 2009 guidance on the enforcement priorities in exclusionary conduct.50 How did the increased use of soft law by the Commission affect the role of the EU courts in antitrust enforcement? First of all, as pointed out above, the issuance of soft law guidance was both a precondition and a consequence of the abolition of the notification system by Regulation 1/2003. Like the latter, it allowed the Commission to refocus its investigations to hardcore offences, and to devote less attention to less egregious offences of competition law. This also implied fewer decisions in the fields were soft law was issued.51 However, as we already saw when discussing decentralisation, this did not necessarily imply fewer applications to the EU courts: the refocusing of the Commission increased the number of cartel decisions it issued and this in turn increased the number of applications for annulment before the EU courts (see Figure 3). What about preliminary questions? As we saw when discussing decentralisation, while the number of preliminary questions in competition cases is limited and it is therefore hard to draw statistically significant conclusions from it, the number of questions seems to be in decline and this, contrary to the trend in other fields of European law. Since there have virtually been no preliminary references in the field of mergers,52 the answer must be sought in the area of antitrust, but this is paradoxically the area in which the NCAs and national courts have been empowered by Regulation 1/2003. At first sight, the increased use of soft guidance by the Commission instead of block exemption regulations can also not explain the reduction in preliminary questions. Contrary to block exemption regulations, the guidance is not binding on the NCAs and the national courts. One would therefore expect that disagreement amongst parties to litigation would continue, despite the guidance, and that preliminary references would continue to be necessary to obtain authoritative interpretation of EU antitrust law. Nevertheless, the Commission’s guidance has been very influential for NCAs and national courts (and even the CJEU has been deferent to it) such that it cannot be excluded that it is precisely the Commission’s soft law that stopped national courts from making preliminary references to the CJEU.53 It is difficult to come to affirmative conclusions on this. But it is interesting to note that in the most recent year for which statistics are available (2016), the number of preliminary questions is on the rise again.54 This could suggest that the CJEU is re-emerging as rule-maker in the field of competition law. V. Conclusion: the changing the role of the EU courts in EU antitrust enforcement As we have seen above, the modernisation of EU antitrust law has allowed the Commission to refocus its attention to cartel enforcement and this led to an upswing in the number of direct actions before the EU courts. The introduction of the settlement procedure in 2009 has, however, counteracted this development: with many cartel cases now being concluded by plea bargaining, fewer cartel cases are brought before the General Court. The reduced number of applications before the General Court of course also reduces the number of appeals before the CJEU which will therefore have fewer opportunities to pronounce itself on Commission procedures. But this does not necessary mean that the importance of the judgements of the EU courts in competition cases diminishes. First of all, in cartel cases, the existence of the infringement was often not in dispute: in their pleas before the courts, the undertakings mostly argued about the calculation of the fine they received (and factors affecting the fine such as parental liability or duration). The role of the EU courts in shaping the substance of EU antitrust law is limited in those cases. To the extent that the drop in direct actions to the EU courts is therefore the result of the introduction of the settlement procedure, this should not affect their role significantly. Of more concern for the role of the EU courts is the reduced number of preliminary references which seems to be apparent in the statistics in recent years. Indeed, preliminary rulings are generally more important for the development of the law than appeals, as was demonstrated in a recent study by Derlén and Lindholm. They assessed CJEU case law both by reference to inward citation (i.e. how many later judgements refer to a particular judgement) and outward citation (to how many previous judgements does a particular judgement refer). They concluded that preliminary rulings have higher inward and outward citations than (appeals against judgements in) direct actions.55 This is also true specifically for competition cases.56 The relative importance of preliminary rulings as compared to appeals is also apparent from the higher number of preliminary questions that result in Grand Chamber judgements. The CJEU sits in Grand Chamber only if a Member State or EU institution so requests and in particularly complex or important cases, which is also reflected in the higher inward and outward citation index they have.57 The statistics of the EU courts in competition cases confirm this: while there were 227 appeals brought before the CJEU in competition cases in the period 2007–2016, in the same period only 15 Grand Chamber judgements in appeals in competition cases58 were delivered (i.e. 7 per cent).59 By comparison: there were 13 Grand Chamber judgements on preliminary questions in competition cases and only 66 preliminary questions sent to the CJEU (20 per cent).60 Optimistically, one could argue that the drop in the number of preliminary references, may be the result of national judges becoming increasingly knowledgeable about EU antitrust law, so that they need to refer fewer questions. Or there may be less national opposition to EU antitrust rules (also as a consequence of harmonisation) so that fewer judges question the EU antitrust rules. Given the variations in the number of preliminary references from individual Member States (and even from individual courts within a Member State), the reasons for referring or not referring questions to the CJEU may in fact be quite idiosyncratic.61 It seems to me that the soft law guidance of the Commission has in any event played an important role in harmonising the interpretation of the antitrust rules across different enforcers, and thereby has helped reduce the need for preliminary references. In this way, the Commission seems to have become the main rule-making institution in EU antitrust law, even for NCAs and national courts. This development is certainly questionable as a matter of EU institutional law where the CJEU should have the last word in speaking the law.62 But quantity is not everything. While the total number of cases and even the number of preliminary references may have been lower than in the past, the CJEU in recent years still pronounced itself on a number of key questions of interpretation of Article 101 and 102 TFEU.63 Many of these judgements follow preliminary references in cases that originated in NCA proceedings and this may well explain why they concern fundamental questions of EU antitrust law which are furthermore fundamental to the outcome of the case.64 While decentralisation may not have resulted in more references, it may therefore have improved their quality. Through the preliminary reference procedure, the CJEU can remain, if not the main, at least the ultimate rule-maker in EU competition law.65 Annexes Cases before the EU Courts (1997–2016)    1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  General Court   Application to GC in competition (excl. State aid)  24  31  34  36  39  62  53  36  40  81  62  71  42  79  39  34  23  41  17  18   Total new cases before GC  644  238  384  398  345  411  466  536  469  432  522  629  568  636  722  617  790  912  831  974   Total GC completed cases in competition (excl. State aid)  21  41  43  61  22  miss.  38  26  35  42  38  31  31  38  100  61  75  72  52  36   Total GC completed cases  186  348  659  344  340  331  339  361  610  436  397  605  555  527  714  688  702  814  987  755  Court of Justice   Direct action to CJEU in competition (excl. State aid)  1  1  9  2  5  1  6  5  1  1  3  0  0  0  0  0  0  0  0  0   Appeal to CJEU in competition (excl. State aid)  9  14  13  14  6  11  8  18  2  15  20  7  10  13  52  23  32  15  32  23   Preliminary questions to CJEU in competition (excl. State aid)  14  13  7  6  15  1  6  9  7  14  8  3  5  5  7  6  6  8  6  12   Interim measures to CJEU in competition (excl. State aid)  0  0  0  0  4  0  1  1  0  2  0  0  2  2  1  1  1  0  2  0   Total new competition cases before CJEU (excl. State aid)  24  28  29  22  30  13  21  33  10  32  31  10  17  20  60  30  39  23  40  35   Total preliminary questions to CJEU  239  264  255  224  237  216  210  249  221  251  265  288  302  385  423  404  450  428  436  470   Total new cases before CJEU  445  485  543  503  504  477  561  531  474  537  580  592  561  631  688  632  699  622  713  692  Combined   Total new direct actions in competition law before GC and CJEU  25  32  43  38  44  63  59  41  41  82  65  71  42  79  39  34  23  41  17  18   Total new competition cases before GC and CJEU  39  45  50  44  59  64  65  50  48  96  73  74  47  84  46  40  29  49  23  30  Cases before the EU Courts (1997–2016)    1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  General Court   Application to GC in competition (excl. State aid)  24  31  34  36  39  62  53  36  40  81  62  71  42  79  39  34  23  41  17  18   Total new cases before GC  644  238  384  398  345  411  466  536  469  432  522  629  568  636  722  617  790  912  831  974   Total GC completed cases in competition (excl. State aid)  21  41  43  61  22  miss.  38  26  35  42  38  31  31  38  100  61  75  72  52  36   Total GC completed cases  186  348  659  344  340  331  339  361  610  436  397  605  555  527  714  688  702  814  987  755  Court of Justice   Direct action to CJEU in competition (excl. State aid)  1  1  9  2  5  1  6  5  1  1  3  0  0  0  0  0  0  0  0  0   Appeal to CJEU in competition (excl. State aid)  9  14  13  14  6  11  8  18  2  15  20  7  10  13  52  23  32  15  32  23   Preliminary questions to CJEU in competition (excl. State aid)  14  13  7  6  15  1  6  9  7  14  8  3  5  5  7  6  6  8  6  12   Interim measures to CJEU in competition (excl. State aid)  0  0  0  0  4  0  1  1  0  2  0  0  2  2  1  1  1  0  2  0   Total new competition cases before CJEU (excl. State aid)  24  28  29  22  30  13  21  33  10  32  31  10  17  20  60  30  39  23  40  35   Total preliminary questions to CJEU  239  264  255  224  237  216  210  249  221  251  265  288  302  385  423  404  450  428  436  470   Total new cases before CJEU  445  485  543  503  504  477  561  531  474  537  580  592  561  631  688  632  699  622  713  692  Combined   Total new direct actions in competition law before GC and CJEU  25  32  43  38  44  63  59  41  41  82  65  71  42  79  39  34  23  41  17  18   Total new competition cases before GC and CJEU  39  45  50  44  59  64  65  50  48  96  73  74  47  84  46  40  29  49  23  30  Source: annual reports of the CJEU and the General Court: <https://curia.europa.eu/jcms/jcms/Jo2_7000/en/> accessed 19 October 2017. (II) Commission decisions in antitrust (2004–2016)    2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  Source  Cartel decisions (incl. settlements)  6  5  7  8  7  6  7  4  5  4  10  5  6  (1)(2)(3)(4)  Settlement decisions  0  0  0  0  0  0  2  3  2  2  8  2  3  (2)(5)  Number of undertakings settling  0  0  0  0  0  0  15  12  8  10  30  5  7  (2)(5)  Other 101/102 decisions  5  2  1  4  1  1  1  1  0  3  4  0  1  (3)(4)(5)  Commitment decisions  0  2  4  5  1  5  5  2  4  4  2  2  3  (3)(4)(5)  Other decisions (e.g. rejection of complaints)  1  1  1  0  4  1  0  1  1  1  5  3  1  (3)(4)(5)  Total Commission decisions in antitrust (excl. concentrations and State aid)  12  10  13  17  13  13  14  8  10  12  21  10  11  (3)(4)(5)  (II) Commission decisions in antitrust (2004–2016)    2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  Source  Cartel decisions (incl. settlements)  6  5  7  8  7  6  7  4  5  4  10  5  6  (1)(2)(3)(4)  Settlement decisions  0  0  0  0  0  0  2  3  2  2  8  2  3  (2)(5)  Number of undertakings settling  0  0  0  0  0  0  15  12  8  10  30  5  7  (2)(5)  Other 101/102 decisions  5  2  1  4  1  1  1  1  0  3  4  0  1  (3)(4)(5)  Commitment decisions  0  2  4  5  1  5  5  2  4  4  2  2  3  (3)(4)(5)  Other decisions (e.g. rejection of complaints)  1  1  1  0  4  1  0  1  1  1  5  3  1  (3)(4)(5)  Total Commission decisions in antitrust (excl. concentrations and State aid)  12  10  13  17  13  13  14  8  10  12  21  10  11  (3)(4)(5)  Sources: Commission cartel statistics <http://ec.europa.eu/competition/cartels/statistics/statistics.pdf> accessed 19 October 2017 R. Snelders, ‘The EU cartel settlement procedure: the first years’ (2016) Presentation to the Studienvereinigung Kartellrecht<https://www.studienvereinigung-kartellrecht.de/sites/default/files/14h_30_snelders_sv_kartellrecht_-_cartel_settlements_final_0.pdf> accessed 19 October 2017 OECD, ‘Commitment decisions in antitrust cases—Background paper by the Secretariat’ (2016) <https://one.oecd.org/document/DAF/COMP(2016)7/en/pdf> accessed 19 October 2017 Commission, ‘Ten years of antitrust enforcement under Regulation 1/2003: achievements and future perspectives’ COM(2014)453 <http://ec.europa.eu/competition/antitrust/legislation/antitrust_enforcement_10_years_en.pdf> accessed 19 October 2017 Own research for cases in 2014-2016 using the Commission’s search tool <http://ec.europa.eu/competition/elojade/isef/index.cfm?clear=1&policy_area_id=1> accessed 19 October 2017 Footnotes 1 DG Goyder, EC Competition Law (Clarendon 1993) 71. 2 On the latter in particular, see BJ Rodger, ‘Article 234 and competition law: a comparative analysis’ (2008) 15 Maastricht Journal of European and Comparative Law 149, 151-152 and references there, and DP Domenicucci, ‘Preliminary rulings and competition law: some reflections for national judges’ in B Cortese (ed), EU competition law: Between public and private enforcement (Wolters Kluwer 2014) 179, 211. 3 G Slynn, ‘Court of First Instance of the European Communities’ (1989) 9 Northwestern Journal of International Law & Business 542, 543. 4 Idem, 545. 5 AG Cosmas in case C-344/98 Masterfoods v HB ECLI:EU:C:2000:249, 54. 6 Case C-6/72 Europemballage and Continental Can v Commission ECLI:EU:C:1973:22. 7 Cases T-342/99 Airtours v Commission ECLI:EU:T:2002:146; T-310/01 Schneider Electric v Commission ECLI:EU:T:2002:254; T-5/02 Tetra Laval v Commission ECLI:EU:T:2002:264. 8 Case 27/76 United Brands v Commission ECLI:EU:C:1978:22. See S Bishop and M Walker, The economics of EC competition law (Sweet & Maxwell 2010) 4-027. 9 See I Forrester, ‘A challenge for Europe’s judges: the review of fines in competition cases’ (2011) European Law Review 185, in particular 194-196, and D Gerard, ‘Breaking the EU antitrust enforcement deadlock: re-empowering the courts?’ (2011) 36 European Law Review 457, 469-471. 10 See, for example, I Forrester, ‘A bush in need of pruning: the luxuriant growth of light judicial review’ (2011) Competition Law Annual 407; Gerard above n 10, 472 and J Derenne, ‘The scope of judicial review in EU economic cases’ in J Derenne and M Merola (eds), The Role of the Court of Justice of the European Union in Competition Law Cases (Larcier 2012) 73, in particular 83–85. 11 This is for example the case for the claim that the General Court shows more deference to the Commission in cases under Article 102 TFEU than in cases under Article 101 TFEU, a claim made in D Geradin and N Petit, ‘Judicial review in European Union competition law: a quantitative and qualitative assessment’ (2010) Tilburg Law and Economics Center (TILEC) Law and Economics Discussion Paper No. 2011-008 <https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1698342> accessed 23 October 2017, in particular 31. Contrary to what the authors claim, their data does not show a statistically significant difference in the number of annulments of decisions adopted under Article 102 TFEU than under Article 101 TFEU, taking into account the number of decisions contested (which is much lower for Article 102 TFEU than for Article 101 TFEU). The number of Article 102 decisions that is annulled in full or in part is even slightly higher than the number of Article 101 decisions: 28.6 per cent (4/14) as compared to 26.5 per cent (31/117). And the fact that there was no full annulment of any of the 14 Article 102 decision in their data sample is not surprising since also less than 1 in every 14 Article 101 TFEU decisions in their sample was annulled in full (8 out of 117). 12 The Annexes contain all the data used in the figures in this paper, as well as their source. Note that the statistics of the EU courts are based on the case numbers allocated by the Court even though one case number does not always correspond to one case. For example, it is quite common for the European Commission to impose a cartel fine on a number of legal entities who are jointly and severally liable for the fine because they are considered part of the same undertaking. The entities may bring one joint application for annulment against this fine or they may bring separate applications: which of these choices they make, may depend on a number of factors, such as whether they have different interests or different arguments which they want to bring to the court. In the statistics of the General Court and the Court of Justice each application will be assigned a case number and therefore be viewed as a single new case (even if they are afterwards joined for the hearing or for judgement): see the annual reports of the ECJ in 2005 and previous years; for example, the annual report of 2005, page 191. Also in the case of preliminary questions, what counts as one case does not depend on the Court of Justice. If a number of cases are joined at the national level, the national court will send only a single request for a preliminary ruling to the Court of Justice, resulting in a single case number. See for example, case C-419/12 Crono Service scarl and Others ECLI:EU:C:2014:81, which concerned an application by Crono Services scarl and 111 other applicants against Roma Capitale for the annulment of local legislation on car and driver hire services. On the other hand, if cases are not joined at the national level, the national court may send several requests for a preliminary ruling (even if they contain the exact same questions), resulting in several case numbers. See for example, cases C-159/12 to C-161/12 Venturini and Others ECLI:EU:C:2013:791, which concerned application by three pharmacists against decisions of the local health authorities refusing the sale of certain medicinal products in para-pharmacies. 13 A good example of this is in the field of merger control where the interventionism of the General Court resulted in a peak of applications for annulment of 4–5 per year in the early 2000s, while the more hands-off approach in subsequent years led to a decline in the number of cases to 0-3 per year since the end of the 2000s until today. 14 The three factors are discussed in more detail in P Ibáñez Colomo, ‘Three shifts in EU competition policy: towards standards, decentralisation, settlements’ (2013) 20 Maastricht Journal of European and Comparative Law 363. 15 Article 4 of Regulation 17/62 First Regulation implementing Articles 85 and 86 of the Treaty, [1962] OJ 13/204. Undertakings could also request a negative clearance from the Commission on the basis of Article 2 of Regulation 17/62 to the effect that (what is now) Article 101(1) TFEU and Article 102 TFEU were not applicable to an agreement or practice in which the undertaking engaged. 16 This is probably a better term than ‘directly effective’ because the direct effect of the provision results directly from the Treaty of Rome: its application was merely suspended when Article 9(1) of Regulation 17/62 granted the Commission the sole power to apply (what is now) Article 101(3) TFEU. See P Landolt, Modernised EC Competition Law in International Arbitration (Kluwer 2006) 37. 17 In the period May 2004 to December 2013, the Commission adopted 122 decisions applying the Article 101 and/or 102 TFEU and the NCAs 665. See Commission, ‘Ten years of antitrust enforcement under Regulation 1/2003: achievements and future perspectives’ COM(2014)453, 4. 18 See the Commission, ‘Proposal a Directive of the European Parliament and of the Council to empower the competition authorities of the Member States to be more effective enforces and to ensure the proper functioning of the internal market’ COM(2017)142 final, 2 which continues to report that the NCAs were responsible for 85 per cent of enforcement decisions. 19 See, for example, C-D Ehlermann and I Atanasiu, ‘The modernisation of E.C. antitrust law: consequences for the future role and function of the E.C. courts’ (2002) European Competition Law Review 72, 76; K Lenaerts and D Gerard, ‘Decentralisation of EC competition law enforcement: judges in the frontline’ (2004) 27 World Competition 313, 341 and T Cowen, ‘’Justice delayed is justice denied’: the rule of law, economic development and the future of the European community courts’ (2008) European Competition Journal 1, 39. 20 See Rodger above n 3, 159–160. 21 M Prek and S Lefèvre, ‘Competition litigation before the General Court: quality if not quantity?’ (2016) 53 Common Market Law Review 65, 66. 22 See the Commission’s own cartel statistics which refer to 10 cases in the period 1990–1994, 10 cases in 1995–1999 but 30 cases in 2000–2004, 33 cases in 2005–2009 and 30 cases in 2010–2014. In 2015 there were 5 cases and in 2016 6. See <http://ec.europa.eu/competition/cartels/statistics/statistics.pdf> accessed 12 October 2017. 23 Idem. 24 As predicted by some. See Ehlermann and Atanasiu above n 20, 74 and Lenaerts and Gerard above n 20, 341. 25 Commission Regulation 622/2008 of 30 June 2008 amending Regulation (EC) No 773/2004, as regards the conduct of settlement procedures in cartel cases [2008] OJ L171/3; Commission Notice on the conduct of settlement procedures in view of the adoption of Decisions pursuant to Article 7 and Article 23 of Council Regulation (EC) No 1/2003 in cartel cases [2008] OJ C167/1. Note that negotiations can also take place outside this formal framework (see for example the Commission decision of 20 September 2016 in case 39.759 ARA foreclosure where ARA received a 30 per cent fine reduction for cooperation, including acknowledging the infringement and proposing a structural remedy). The discussion of the use of commitments in EU merger procedures is outside of the scope of this paper. 26 Commission, ‘Notice on best practices for the conduct of proceedings concerning Articles 101 and 102 TFEU’ [2011] OJ C308/6. 27 See D Waelbroeck, ‘Le développement en droit européen de la concurrence des solutions négociées (engagements, clémence, non-contestation des faits et transactions): que va-t-il rester aux juges?’ (2008) GCLC Working Paper 1/08 <https://www.coleurope.eu/system/files_force/research-paper/gclc_wp_01-08.pdf?download=1>accessed 19 October 2017, 3 and F Jenny, ‘Worst decision of the EU Court of Justice: the Alrosa judgement in context and the future of commitment decisions’ (2015) 38 Fordham International Law Journal 701, 735–736. 28 See E Barbier de La Serre, ‘Competition law cases before the EU courts: is the well running dry?’ in J Derenne and M Merola (eds), The Role of the Court of Justice of the European Union in Competition Law Cases (Larcier 2012) 87, 98–99 and Prek and Lefèvre above n 22, 66. See also AG Wahl, as reported in ‘Increase in EU settlements risks sidelining courts, victims, Wahl says’ MLex (20 October 2014). 29 See in particular W Wils, ‘Ten years of commitment decisions under Article 9 of Regulation 1/2003: too much of a good thing?’<https://papers.ssrn.com/sol3/Papers.cfm?abstract_id=2617580>accessed 22 August 2017, 10. The importance of informal commitments to close cases under Regulation 17/62 is also emphasised for instance by Waelbroeck above n 28, 5 and Ibáñez Colomo above n 15, 368. 30 Wils above n 30, 6–9. 31 See the statistics in the Annexes. 32 Regulation 773/2004 relating to the conduct of proceedings by the Commission pursuant to Articles 81 and 82 of the EC Treaty [2004] OJ L123/18. 33 Up to 2016 there were 5 so-called hybrid settlement cases were one or more undertakings were fined under the ordinary procedure while the other cartel members settled the case: cases 38.866 Animal feed phosphates, 39.861 Yen interest rate derivatives, 39.914 Euro interest rate derivatives, 39.965 Mushrooms and 39.792 Steel abrasives. 34 See the Q&A on settlements on the Commission’s website:<http://ec.europa.eu/competition/cartels/legislation/cartels_settlements/settlements_en.html>accessed 23 October 2017. 35 At the moment of concluding this paper, there were only two examples. The first was the application for annulment filed by Société Générale in case 39.914 Euro interest rate derivatives on grounds of differential treatment (case T-98/14 Société générale v Commission) which was withdrawn after Société Générale provided corrected value of sales data to the Commission and subsequently saw its fine reduced by the Commission. The second case is the application for annulment filed by Printeos in case 39.780 Envelopes which resulted in a judgement of the General Court on 13 December 2016 (case T-95/15 Printeos and others v Commission ECLI:EU:T:2016:722) in which the General Court annulled the Commission decision because of insufficient reasoning concerning discretionary fine reductions; but the Commission readopted its decision and imposed an identical fine on Printeos on 16 June 2017. 36 However, as pointed out in footnote 14, there is not a one-to-one relationship between the number of undertakings subject to a cartel investigation as identified by the Commission and the number of applicants before the EU courts. 37 Although there appears to be a direct correlation between the evolution of the number of settlements and the evolution of the number of direct actions in 2013-2016, it is hard to see what causal link there might be between the two, except perhaps the external factor that 2014 was the last year of the tenure of Joaquín Almunia as Commissioner for Competition, which may have resulted in a peak in the number of settlement decisions (8) as well as other decisions (13). 38 M Hellwig, K Hüschelrath and U Laitenberger, ‘Settlements and appeals in the European Commission’s cartel cases: an empirical assessment’ (Zentrum für Europäische Wirtschaftsforschung GmbH, January 2016)<http://ftp.zew.de/pub/zew-docs/dp/dp16010.pdf> accessed 16 August 2017. 39 See on this aspect: Ehlermann and Atanasiu above n 20, 73. 40 Note that some guidance, such as the ‘Notice on the definition of the relevant market for the purposes of Community competition law [1997] OJ C372/5, is relevant for both merger control and antitrust. 41 Commission, ‘Green Paper on vertical restraints in EC competition policy’ COM (96) 721, <http://europa.eu/documents/comm/green_papers/pdf/com96_721_en.pdf> accessed 9 October 2017. 42 There were also several notices on jurisdiction and procedure (such as on fines, leniency, effect on trade between Member States and the de minimis notice) but these again have less importance from the perspective of the regime of legal exception and decentralisation. 43 Commission, ‘Guidelines on vertical restraints’ [2000] OJ C291/1. 44 Commission, ‘Guidelines on the application of Article 81 of the EC Treaty to horizontal cooperation agreements’ [2001] OJ C3/2. 45 Commission, ‘Guidelines on the application of Article 81 of the Treaty on technology transfer agreements’ [2004] OJ C101/2. 46 Commission, ‘Guidelines on the application of Article 81(3) of the Treaty’ [2004] OJ C101/97. 47 Commission, ‘Guidelines on vertical restraints’ [2010] OJ C130/1. 48 Commission, ‘Guidelines on the application of Article 101 of the Treaty on the Functioning of the European Union to horizontal cooperation agreements’ [2011] OJ C11/1. 49 Commission, ‘Guidelines on the application of Article 101 of the Treaty on the Functioning of the European Union to technology transfer agreements’ [2014] OJ C89/3. 50 Commission, ‘Guidance on tis enforcement priorities in applying Article 82 of the EC Treaty to abusive exclusionary conduct by dominant undertakings’ [2009] OJ C45/7. 51 Ibáñez Colomo above n 15, 370. 52 There has only been one question relating to mergers: case C-248/16 Austria Asphalt ECLI:EU:C:2017:643. 53 This is also the belief of Barbier de La Serre above n 29, 96. 54 A look at the figures for the first half of 2017 is inconclusive as to whether this trend will continue: there were four requests for preliminary rulings in the field of competition (excluding State aid) in January–June 2017: cases C-27/17, flyLAL-Lithuanian Airlines, C-144/17, Lloyd’s of London, C-152/17, Consorzio Italian Management and C-267/17 Rhenus Veniro. 55 M Derlen and J Lindholm, ‘Characteristics of precedent: the case law of the European court of justice in three dimensions’ (2015) 16 German Law Journal 1073, 1082-1085. This applied even to different segments of direct actions, with the exception of outward citations (so references to previous case law), where unsuccessful infringement proceedings top the chart. 56 Idem, 1088–1090. 57 Idem, 1092. 58 Note that five of these judgements concerned two related cases, so one could also say that in only 12 competition cases on appeal a Grand Chamber judgement was issued. 59 Note that this is a comparison of the number of appeals brought in 2007–2016 and the number of judgements on appeal in that same period (some of which relate to appeals brought prior to 2007), so the figures do not relate to the same cases. The figures listed therefore merely indicate a tendency. 60 The previous footnote applies mutatis mutandis. 61 See, for a discussion on the practice of individual Member States to refer preliminary questions in competition cases and speculation on the reasons for this: Rodger above n 3, 149. 62 K Lenaerts, ‘The rule of law and the coherence of the judicial system of the European Union’ (2007) 44 Common Market Law Review 1625, 1652. 63 For example, in the field of Article 101 TFEU, on the conditions for unlawful exchange of information in case C-74/14 Eturas and Others ECLI:EU:C:2016:42, in the field of Article 102 TFEU, on the relevance of the as-efficient-competitor test in cases C-209/10 Post Danmark v konkurrencerådet ECLI:EU:C:2012:172 (and later in C-413/14 P Intel v Commission ECLI:EU:C:2017:632) and in procedural matters on the circumstances were exculpatory evidence can be accepted in case C-681/11 Schenker & Co. and Others ECLI:EU:C:2013:404. 64 I thank an anonymous reviewer of the Journal of European Competition Law & Practice for this suggestion. 65 See also N Petit, ‘The future of the court of justice in EU competition law’ in Court of Justice of the European Union (ed), The Court of Justice and the Construction of Europe: Analyses and Perspectives on Sixty Years of Case-law - La Cour de Justice et la Construction de l’Europe: Analyses et Perspectives de Soixante Ans de Jurisprudence (T.M.C. Asser Press 2013) 397–421. © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com

Journal

Journal of European Competition Law & PracticeOxford University Press

Published: Feb 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off