The non-disclosure rules in EU Directive 2014/104/EU. An unsuccessful attempt to protect the EU Leniency Programme

The non-disclosure rules in EU Directive 2014/104/EU: An unsuccessful attempt to protect the EU Leniency Programme

The European Union recently enacted Directive 2014/104/EU on “certain rules governing actions for damages under national law for infringements of the competition law provisions of the Member States and of the European Union” (Damages Directive). One of the main purposes of this directive is to balance private and public competition law enforcement, particularly to solve the tension between the Commission’s leniency programme and private enforcement.^[1] The tension between these two institutions is caused by the fact that potential leniency applicants might be discouraged from applying for the programme if the danger exists that they could find themselves in a worse position than non-applicants in relation to potential damage actions against them.^[2] It can be noted that there are generally two ways of solving this tension: (1) to prevent disclosure of leniency material from potential damage claims and (2) to reduce the amount of damages successful leniency applicants have to pay.^[3] The EU legislator decided to combine both methods and implemented three provisions to protect the leniency programme. These are the Articles 6 (6) (a), 7 (1), which take the first way, and 11 (4), which limits the damages successful applicants have to pay to the ones suffered by direct or indirect customers and cartel victims which could not obtain damages from other cartelists. Additionally a new Art 16a (2) was very recently (3 August 2015) implemented into Regulation 773/2004 in order to prevent leniency documents in possession of the Commission from being disclosed. In this essay however, I will focus mainly on the non-disclosure rule of Art 6 (6) (a) Regulation 2014/104/EU since this provision does seem to be most controversial. Nevertheless it cannot be analysed without taking into account the other provisions. Article 6 (6) of the Damages Directive states that

“ Member States shall ensure that, for the purpose of actions for damages, national courts cannot at any time order a party or a third party to disclose any of the following categories of evidence:

(a) Leniency statement …”

This provision and the underlying system of protection of leniency documents is opposed by many authors. Some of them ^[4] hold the opinion that it does not protect leniency applicants effectively since many documents and information could still be obtained, which lets leniency applicants worse off than co-infringers. Others^[5] argue that the provisions are inconsistent with recent ECJ case law and therefore with principle of effectiveness in consumption with Art 101, 102 TFEU. Both groups of authors have in common that they propose a different way to protect leniency applicants while not discouraging possible private damage claims.

Private Enforcement was historically more important in the US, where private actions account for at least 80 % of all antitrust filings each year,^[6] than in the EU.^[7] However in recent years especially the Netherlands, Germany and the UK developed a working practice of private Enforcement.^[8] Furthermore the ECJ established the possibility for damage claims following infringements of EU Competition Law in Courage v. Crehan^[9] and Manfredi^[10]. In Courage the court stated that private damage claims significantly contribute to secure effective competition within the EU and therefore pose a necessity for an effective competition law enforcement.^[11] Following that it was decided in Manfredi that Art 101 TFEU must be interpreted in a way which allows any individual to claim damages for an infringement of the provision if a ‘casual relationship’ between the harm suffered and the infringement exists.^[12] After these decisions the Member States of the EU developed an unequal private enforcement practice.^[13]

However, private enforcement cannot be seen in isolation to public enforcement as also described by recital (6) of Directive 2014/104/EU. It can have positive as well negative effects on public enforcement in general and on leniency programs in particular.^[14] The EU leniency program is an extremely important tool of the Commission since 88 % of the Commissions’ infringement decisions are based on leniency applications.^[15] It has been put into danger by the ECJ’s decisions in Pfleiderer^[16] and Donau Chemie^[17] . The ECJ ruled that no rigid rule, whether leniency documents may be disclosed to damage claimants or not, exists. The court decided that the Member States’ courts have to conduct a weighing exercise between the interests of the claimant to disclose leniency documents as evidence and the interests in non-disclosure, one of which is to provide effective leniency programmes. This, so it is argued, caused uncertainty for possible leniency applicants as they could not predict whether they had to disclose leniency statements to possible claimants or not. This uncertainty was consequently illustrated by decisions of English and the German Courts. The English High Court partly granted access in the first National Grid^[18] by ordering the disclosure of the Defendants’ reply to the Statement of Objections besides other documents and refusing the disclosure of documents particularly prepared for the leniency application. However the Dusseldorf Appeal Court rejected the disclosure of any documents after weighing the opposing interests against each other.^[19] Following this decisions, the Commission feared that its leniency programme could suffer severe damages in its efficiency and therefore implemented Art 6 (6) (a) in the new directive.

This provision however is not the right way to protect the leniency programme, although protection from a possible negative impact of private enforcement is certainly necessary, as I will show in the first part of this essay. It does not combine the different possibilities to protect leniency programmes from private enforcement efficiently. As I will show in this essay, it would be the best way forward to walk the second way described above: to reduce the damages which immunity recipients have to pay. In order to support this claim I will show different flaws of Art 6 (6) (a) and the other provisions. First of all the rigid non-disclosure rule is inconsistent with the ECJ’s decisions in Pfleiderer and Donau Chemie. Furthermore it is also not desirable from a competition policy point of view, which I will illustrate in the third part. Also does the way taken by the Directive leave a number of loopholes, through which damage claimants could obtain the information contained in a leniency statements which is why the directive does not reach its goal of protecting leniency applicants from private enforcement. I will demonstrate this by providing examples of possibilities to obtain the information in the fourth part of this essay. Thereafter I will describe another, more advantageous way to protect the leniency programme and facilitate private enforcement at the same time.

I. The need to protect the leniency programme

As regards the perspective of potential leniency applicants it is to note that the weighing exercise created in Pfleiderer caused great uncertainty as well for them as for damage claimants.^[20] The judgement does neither grant full protection to a claimants’ right nor does it protect the leniency programme.^[21] It is argued that Leniency programmes may take significant harm due to the insecurity,^[22] since leniency programmes generally require certainty and cost effectiveness to work and that no rational risk neutral or averse agent would seek the benefits of a leniency programme without being able to calculate the risks ex ante.^[23] For this reasons the Advocate General already opposed the court in Donau Chemie and the ECJ recognized the leniency privilege and inherent danger of limiting the effectiveness of the program in Pfleiderer.

It has to be noted that a protection of the leniency programme is also demanded by Art 101, 102 TFEU. The principle of effectiveness does not only support private enforcement. This general principle of EU Law^[24] articulates that Member States may not render the enforcement of EU law impossible or extensively difficult.^[25] On the one hand the principle helped the ECJ to establish the right to claim damages in Courage and Manfredi. ^[26] On the other hand the principle demands that the effectiveness of Art 101 TFEU in total is not undermined.^[27] Consequently the principle can narrow the possibilities for private damage claimants since it also demands an effective public competition law enforcement and therefore a functioning leniency programme.^[28]. Additionally an efficient leniency programme contributes to the effectiveness of private enforcement through easier follow-on actions. These actions still represent the high majority of actions brought in the EU,^[29] Thus disclosure of leniency documents could even hinder the effectiveness of private enforcement.^[30] If one takes into account all the parts of the principle of effectiveness (effective enforcement and effective judiciary protection) the principle of effective enforcement argues for and against the non- disclosure provisions. Only the principle of effective judiciary protection solely argues against the new provisions. Because the right for damage claims is based on the principle of effectiveness the exercise of the right must be limited if the full effectiveness of EU Law would otherwise be reduced.^[31] This means that private enforcement of competition law might have to step back if it inflicts leniency programmes to an extent which harms the effectiveness of public competition law enforcement. The principle of effectiveness has to be seen as a whole and cannot be separated in public and private cartel enforcement since Art 101, 102 TFEU do not protect individual interests. If they would do so, it would be necessary for the claimant to prove that he belongs to the category of person the provision intents to protect (protective purpose doctrine) - a “casual connection to the infringement” as stated in Manfredi^[32] would not be sufficient.^[33]

If leniency documents are not protected somehow, private damage actions pose a severe danger for the effectiveness of public enforcement since leniency applicants might reduce the amount of information they are willing to provide due to the thread of disclosure.^[34] Furthermore the applicant would be worse off than the co-infringers since he produced written corporate statements and witness statements that explain anti- competitive arrangements and admit participation.^[35] Hence it is very difficult for a leniency applicant to dispute its liability in front of a court, an applicant is unlikely to appeal against a competition authority’s infringement-decision.^[36] Because of this and because the leniency applicant is jointly and severally liable with all co-infringers, it is argued that claimants have a strong incentive to suit the leniency applicant for the entire loss.^[37] This incentive could be strengthened by the opportunity of obtaining evidence in possession of the leniency applicant.^[38] Although there is no hard evidence for it,^[39] it is highly probable that this situation would deter possible leniency applicants from applying. This effect could be enhanced by the inability of potential applicants to assess the risks connected to an application,^[40] and would lastly lead to less cartels being uncovered.

Another problem which could occur due to the uncertainty is the loss of an equal “level playing field” within the European Union. The courts of the different member states courts developed different practices regarding the disclosure of leniency documents, as seen in National Grid^[41] and the German Pfleiderer^[42] cases.^[43] This might lead to “forum-shopping” of damage claimants who would seek damages in the most claimant friendly Member State.

II. Consistency of the Directive’s approach with primary EU Law

Although, as this makes clear, leniency applicants have to be protected from private damage claimants, the Directive’s way is not the right one for several reasons. First of all it is important to note that the directive cannot change the way primary EU law is interpreted by the ECJ. Primary Law includes the European Treaties as well as the methods which the ECJ developed for interpreting this treaties.^[44] One of this methods is the Principle of effectiveness.^[45] The ECJ interprets the treaties and other sources of primary EU law in preliminary rulings (Art 267 TFEU)^[46] as the ones of Pfleiderer and Donau Chemie. Furthermore the treaties can only be changed following the formal procedure of Art 48 TFEU.^[47] This demonstrates that a directive, which is secondary EU law within the meaning of Art 288 TFEU, cannot change primary law and is ranked lower than the interpretation of the treaty by the ECJ. Otherwise the procedure of Art 48 TFEU would be unnecessary. Therefore the provisions Art 6 (6) (a) of Directive 2014/104/EU is inconsistent with primary EU law if they are inconsistent with the judgments of Pfleiderer and Donau Chemie. The adoption of the draft directive was, in opposition to Wood’s statement^[48] not capable of “consigning the judgements to history”. The directive would not be automatically void, but the European Court of Justice could declare them as such in an action for annulment (Art 263 TFEU). Therefore the importance of the question whether the provisions at hand are consistent with the ECJ’s recent judgements cannot be overestimated.

In Pfleiderer the ECJ admitted the disclosure of leniency documents can harm the effectiveness of leniency programmes.^[49] However the court noted that according to settled case law that any individual has the right to claim damages for loss suffered following infringements of competition law.^[50] Therefore the court reached the conclusion that the interests in favour of disclosure and the ones in favour of the protection of voluntary provided information by a leniency applicant have to be balanced against each other,^[51] and that this weighing exercise has to be conducted

on a case-by-case basis by national courts and tribunals.^[52] This line was confirmed and continued in Donau Chemie, in which the court was concerned with an Austrian provision that made access to documents, including leniency submissions, in Art 101 TFEU damage claims subject to the consent of the other party. The ECJ ruled that a national provision concerning access to documents, which does not leave any possibility for the courts to weigh the involved interests against each other is precluded by Art 101 TFEU in conjunction with the principle of effectiveness.^[53] The court did explicitly not follow Advocate General Jääskinen, who held that access to voluntarily provided leniency documents should principally not be granted,^[54] and that public policy reasons and fairness towards the leniency applicant must weigh heavily against any orders to disclose leniency statements.^[55] He clearly valued public over private enforcement and stressed that private actions should not be developed to a point in which they imperil an efficient public cartel enforcement.^[56] Therefore he proposed a legislative act which absolutely protects leniency documents but leaves the weighing exercise regarding other types of documents to the deciding courts.^[57] The court of justice however decided differently and distanced itself from the Advocate General’s opinion that leniency materials should have a special status by confirming its view from Pfleiderer that a case-by-case assessment has to be conducted.^[58]

[...]

^[1] Anneli Howard, ‘Too little, too late? The European Commission's Legislative Proposals on Anti-Trust Damages Actions’ (2013) 4(6) Journal of European Competition Law & Practice 455, 456

^[2] Renato Nazzini, ‘The objective of private remedies in EU competition law’ (2011) 4(4) Global Competition Litigation Review 131, 144

^[3] Caroline Caufmann, ‘The Interaction of Leniency Programmes and Actions for Damages’ (2011) 7(2) The Competition Law Review 181, 184

^[4] Damien Geradin and Laurie-Anne Grelier, ‘Cartel Damages Claims in the European Union: Have We Only Seen the Tip of the Iceberg?’ in William E Kovacic (ed), An Antitrust Tribute - Liber Amicorum, Volume II: Institute of Competition Law (2014); Louis S Morais, ‘Integrating Public and Private Enforcement of Competition Law in Europe: Legal Issues’, European Competition Law Annual 2011 : Integrating Public and Private Enforcement of Competition Law - Implications for Courts and Agencies (Hart Publishing Ltd 2014) 121

^[5] Caroline Kersting, ‘Removing the Tension Between Public and Private Enforcement: Disclosure and Privileges for Successful Leniency Applicants’ (2013) 5(1) Journal of European Competition Law & Practice 2, 2; Julian A Sanner, Informationsgewinnung und Schutz von Unternehmensgeheimnissen in der privaten Kartellrechtsdurchsetzung (Europäisches Wirtschaftsrecht Band 58, 1. Aufl) 503; Amit K Singh, ‘Disclosure of leniency evidence: Examining the Directive on damages action in the aftermath of recent ECJ rulings’ (2014) 7(4) Global Competition Litigation Review 200, 208; Nicholas Hirst, ‘Donau Chemie: national rules impeding access to antitrust files liable to breach EU law’ (2013) 4(6) Journal of European Competition Law & Practice 484, 486

^[6] Green, Jonathan, McCall, Iona, ‘Leniency and civil claims: Should leniency porogrammes extend to private actions?’ (2009) 8(8) Competition Law insight 3, 4

^[7] Assimakis Komninos, ‘The Relationship between Public and Private Enforcement: quod Dei Deo, quod Caesaris Caesari’ (UCL June 2011) 1

^[8] Geradin and Grelier (n 4) 260

^[9] Case C-453/99 Courage v Crehan (2001) ECR I-6297 [24,26]

^[10] Joined Cases C-295/04 to C-298/04 Vincenzo Manfredi and Others v Lloyd Adriatico Assicurazioni SpA and Others (2006) ECR I-6619 [59,61]

^[11] Case C-453/99 Courage v Crehan (n 9) 27

^[12] Joined Cases C-295/04 to C-298/04 Vincenzo Manfredi and Others v Lloyd Adriatico Assicurazioni SpA and Others (n 10) 63

^[13] Howard (n 1) 455

^[14] Wils, Wouter P. J. ‘Leniency in Antitrust Enforcement: Theory and Practice’ (2007) 30(1) World Competition: Law and Economics Review, 40

^[15] Howard (n 1) 460

^[16] Pfleiderer AG v Bundeskartellamt (2011) Case C-360/09 (European Court of Justice)

^[17] Bundeswettbewerbsbeh ö rde v Donau Chemie AG (2013) Case C-536/11 (European Court of Justice)

^[18] National Grid Electricity Transmissions v ABB Ltd and others (2012) EWHC 869 (Ch)

^[19] Roasted Coffee (2012) Case No B-4 KART 5/11 [No 63] (Dusseldorf Appeal Court)

^[20] Gaetane Goddin, ‘The Pfleiderer Judgment on Transparency: The National Sequel of the Access to Document Saga’ (2012) 3(1) Journal of European Competition Law & Practice 40, 41

^[21] Singh, ‘Disclosure of leniency evidence’ (n 5) 203

^[22] Bruno Lebrun and Laure Bersou, ‘Commission v EnBW Energie: Non-Disclosure of Leniency Documents’ (2014) 5(7) Journal of European Competition Law & Practice 462, 462

^[23] Angus MacCulloch and Bruce Wardhaugh, ‘The Baby and the Bathwater: The Relationship in Competition Law between Private Enfocement, Crimal Penalties, and Leniency Policy’ (Working Paper 14 June 2012) 5

^[24] Takis Tridimas, The general principles of EU law (Oxford EC law library, 2nd ed, Oxford University Press 2006) 418

^[25] Case C-261/95 Palmisani (1997) ECR I-4025 27

^[26] Case C-453/99 Courage v Crehan (n 9)

^[27] Veljko Milutinovic, ‘The ‘Right to Damages’ in a ‘System of Parallel Competences’: A Fresh Look at BRT v SABAM and its Subsequent Interpretation’, European Competition Law Annual 2011 : Integrating Public and Private Enforcement of Competition Law - Implications for Courts and Agencies (Hart Publishing Ltd 2014) 350

^[28] Nazzini (n 2) 143

^[29] Amit K Singh, ‘Pfleiderer: assessing its impact on the effectiveness of the European leniency programme’ (2014) 35(3) European Competition Law Review 110, 112

^[30] Kristina Nordlander and Marc Abenhaim, ‘The "Discoverability" of Leniency Documents and the Proposed Directive on Damages Actions for Antitrust Infringements’ 2014 CPI Antitrust Chronicle 1, 5

^[31] Nazzini (n 2) 132

^[32] Joined Cases C-295/04 to C-298/04 Vincenzo Manfredi and Others v Lloyd Adriatico Assicurazioni SpA and Others (n 10)

^[33] Nazzini (n 2) 140

^[34] Marc Hansen, Luca Crocco and Susan Kennedy, ‘Challenges to International Cartel Enforcement and Multi-Jurisdictional Leniency Applications - Disclosure of Leniency Applicant Statements and Materials’ (Vancouver 2012), 21 <http://www.lw.com/presentations/challenges-to-international-cartel- enforcement> accessed 27 July 2015

^[35] Nazzini (n 2) 144

^[36] ibid.

^[37] ibid.

^[38] ibid.

^[39] Singh, ‘Disclosure of leniency evidence’ (n 5) 207

^[40] Christopher Leslie, ‘Antitrust Amnesty, Game Theory, and Cartel Stability’ (2006) 31(2) Journal of Corporation Law 453, 461; Scott D Hammond, ‘The Evolution of Criminal Antitrust Enforcement Over The Last Two Decades’ (Miami 2010) <http://www.justice.gov/atr/speech/evolution-criminal-antitrust- enforcement-over-last-two-decades> accessed 27 July 2015

^[41] National Grid Electricity Transmission v ABB and Others (2013) EWHC 822 (Ch)

^[42] Pfleiderer (2012) 51 Gs53/09 (Bonn Local Court)

^[43] Singh, ‘Disclosure of leniency evidence’ (n 5) 204; Christina Hummer and Milosz Cywinski, ‘ECJ's judgement in "EnBW" and "Donau Chemie" and the unresolved problems of access to the file’ (2014) 7(2) Global Competition Litigation Review 115, 117

^[44] Martin Nettesheim, ‘EGV Art. 249’ in Martin Nettesheim (ed), Das Recht der Europ ä ischen Union (40th edn 2009) 11

^[45] Michael Potacs, ‘Effet utile als Auslegungsgrundsatz’2009 EuR 465, 467

^[46] Koenraad Lenaerts and others, Procedural law of the European Union (2nd ed. 2006) 6-002

^[47] Nettesheim (n 44) 11

^[48] Adrian Wood, ‘Case Comment: The disclosure of leniency related materials supplied during competition enforcement proceedings: battle lines drawn by the ECJ’ (2013) 18(1) Coventry Law Journal 112, 119

^[49] Pfleiderer AG v Bundeskartellamt (n 16) [25,26]

^[50] ibid 28

^[51] ibid 30

^[52] ibid 31

^[53] Bundeswettbewerbsbeh ö rde v Donau Chemie AG (n 17) [49,51]

^[54] ibid 55

^[55] ibid 56

^[56] ibid 62

^[57] ibid 64

^[58] ibid 10