Thomas Sharpe QC and James Flynn QC invited me to speak to the Lincoln’s Inn Eurogroup on the relationship between public and private antitrust enforcement.

This is the third main section of the speech. The others are:
The Damages Directive and the protection of leniency submissions and settlement statements;
Publications and the content of decisions.

For those who prefer to read paper documents, I attach PDFs of the full prepared remarks and the handout.

In a selection of cases over the last couple of years the General Court and Court of Justice have grappled with what to do about cases where the Commission has found a particular company to have participated in part of a broader set of infringements, that broader set constituting a single and continuous infringement. We may, for example, be looking at a case where a company participated in respect of one product of a two-or-more product cartel, or where it was a fringe player in terms of either its participation, or its geographic reach.

Commission practice has tended to take an all or nothing approach to single and continuous infringement, secure in the knowledge that the fine imposed on any fringe player would properly reflect its fringe participation.

As a reminder, the finding of a single and continuous infringement has broadly three potential consequences for a case.

First prescription: if earlier conduct can be regarded as a single infringement with later conduct, then earlier conduct that might otherwise be prescribed would still be attackable.

Second, fines: if there is one single infringement, then there is one fine, one entry fee and one ten percent cap. If there are two infringements, then there can be two fines, with two entry fees, each subject to the ten percent cap.

Third, damages: participants in a single and continuous infringement are jointly and severally liable for the damage that they have caused.

It is this third point that makes the characterisation of a single and continuous infringement more important now than in the past.  The Commission can tailor the fine to the actual sales of the undertaking involved in the cartel – so that if it sold only one product in a cartel that encompassed several in a single infringement, the fine would be based on the sales of that one product. So for the fine, the finding of a single infringement does not necessarily increase the fine.

However a finding of a single infringment could substantially increase damages. If a maker of widgets was involved in a cartel which the Commission found to have covered both widgets and sproggets, then a finding of a single infringement would potentially expose that company to damages claims from purchasers of sproggets as well.

Once you take into account that a finding of participation in a single and continuous infringement brings with it joint and several liability for damages, then the position looks rather different. This, I think, is one of the main drivers of recent case law on single and continous infringement, and how the Commission has characterised it in its decisions.

Unfortunately the case law is not entirely clear as to how fringe players should properly be characterised.

In Coppens,[1] the Court of Justice found that a company had participated in one product of a two product cartel, and was not a part of the two-product single and continuous infringement. Nevertheless a finding of infringement in respect of that one product was upheld.

In Aalberts,[2] the Court took a different approach, finding that a company had only participated in some of the anti-competitive contacts. The Court – relying on the Commission’s characterisation of the contacts as a single indivisible infringement – annulled the entire decision.

Then more recently in Soliver,[3] the Court found that a fringe player carried out acts that fell short of full complicity in the more extensive single infringement carried out by other players, recognised that those acts were clearly anti-competitive, but nevertheless annulled the decision as a whole indicating that although the anti-competitive nature of those acts was clear, the party had not had an opportunity to respond to objections focussed solely on those anticompetitive acts during the administrative procedure. Soliver appears to change a substantive issue of the characterisation of the single infringement into a procedural one. The Court felt that it could not maintain the infringement in relation to those lesser acts without violating the party’s rights of defence.

I suspect that all of these cases were motivated at least in part by the concern of joint and several liability for possibly extensive follow on damages. The counter argument of course is that this can properly be resolved by a national court taking a view as the relative contributions of each party. But I suspect that many parties will want to avoid getting drawn into cases, rather than rely on their minor role being properly characterised once they are there. And there is, of course, always the possibility that one party may be bankrupt, potentially increasing the liability of the others.

So I think it inevitable that the Commission will have to deal with a single infringement in a more nuanced way in future decisions. Unfortunately the Courts’ rulings have not left us with clear guidance as to how this should be done.

I tend to the view that we should recognise more explicitly that findings of infringements can be asymmetric – that, for example, one party may be liable for a single and continuous infringement for a range of conduct, and another party may be liable for only a subset of that conduct, contributing to, but not being party to, the entire single infringement.

That is the type of reasoning used in the Commission’s decision in respect of Bananas. There the Commission did impose asymmetric liability. Weichert was found liable only for part of the single and continuous infringement for which the other parties were liable.

The General Court upheld this finding,[4] though the reasoning used does not seem entirely consistent with the other caselaw mentioned above. At paragraph 648, the Court held that: “…the fact that an undertaking has not taken part – like the undertaking comprising Weichert and Del Monte in the present case – in all aspects of an anti-competitive scheme or that it played only a minor role in the aspects in which it did participate is not material to the establishment of the existence of an infringement on its part. Such a factor must be taken into consideration only when the gravity of the infringement is assessed and if and when it comes to determining the fine”

How to draft a decision taking into account the differing participation of different firms? How many alternative characterisations of the infringement need to be set out in order to ensure that rights of the defence are properly respected? How to characterise a fringe player’s participation in a broader infringement? None of these points are, I’m afraid, easy to determine based on the case law. Yet given the importance of an SCI finding to damages claims, the Commission is going to be under increasing pressure on the point.

[1] Case C-441/11 P., Judgment of the Court (Fourth Chamber) of 6 December 2012, European Commission v Verhuizingen Coppens NV, ECLI:EU:C:2012:778

[2] Case C-287/11 P., Judgment of the Court (Third Chamber) of 4 July 2013, European Commission v Aalberts Industries NV and Others, not yet reported.

[3] Case T-68/09, Judgment of the General Court (Second Chamber) of 10 October 2014, Soliver NV v European Commission, not yet reported.

[4] Case T-587/08, Judgment of the General Court (Eighth Chamber) of 14 March 2013, Fresh Del Monte Produce, Inc. v European Commission, not yet reported.

Thomas Sharpe QC and James Flynn QC invited me to speak to the Lincoln’s Inn Eurogroup on the relationship between public and private antitrust enforcement.

This is the second main section of the speech. The others are:
The Damages Directive and the protection of leniency submissions and settlement statements;
Single and continuous infringement.

For those who prefer to read paper documents, I attach PDFs of the full prepared remarks and the handout.

The Commission has a dual role in both proposing legislation to promote the effective application of competition law, as with the Damages Directive, and in directly enforcing the competition rules through its own decisions. And obviously there is an inevitable link between public enforcement and private enforcement because of follow on actions that, for the foreseeable future, are likely to account for the majority of private enforcement activity in the EU.

However the Commission’s public enforcement is a standalone responsibility; we should not, for example, alter our approach to the drafting and publication of decisions in light of the increasing prevalence of follow on actions. In other words we should not increase the amount of information available so as to help follow on claims, but nor should we limit the information available so as to avoid arguments with defendants who fear damages litigation.

The publication process

This does not mean that the publication of decisions proceeds in a vacuum. We have certainly seen an increase in defence lawyers contesting the content of public versions of decisions, and there is a clear tension between our desire to publish quickly and the parties’ desire which is often to avoid publication for as long as possible.

This extends to asking – successfully so far – the General Court to suspend publication of our decisions pending a full hearing on confidentiality issues.  One might speculate that part of the motivation for some of the court cases is not to win or lose the case, but simply to delay the publication of Commission decisions further.

I am sure that many of you are aware that this has led to some English judicial criticism of the Commission’s “molasses” like speed of publications. Certainly the situation is unsatisfactory.

But as with many aspects of damages litigation, until we have clearer judicial precedents, the European Commission’s ability to move this process along is limited given the legal constraints on our actions.

We now have a ruling of the General Court – Schenker[1] – that suggests that we should publish a preliminary non-confidential version of our decisions, essentially taking on board all of the claims of the parties. That will, as I am sure you have already realised, in practice lead to preliminary publications that are mostly blank.

We are considering unilateral action that may help the publication process, for example by possibly publishing greater public guidance on what is and is not a valid confidentiality claim. That may help to clarify what is and is not confidential, but it is not a panacea. It would not bind parties and they could still appeal to the General Court who would, on current practice, be likely to suspend the Commission’s attempt to publish.

The only way through this, I think, is for the General Court to issue judgments that clarify some of these publication issues. If that is to happen sooner rather than later, that will require the backlog of cases to be reduced, and it is not at all clear that a significant reduction is within the gift of the General Court acting alone.

Greater clarification in this area is, therefore, an exercise that is more likely to take years rather than months.

While this continues, we nevertheless continue to try and publish non-confidential versions of our decisions. The following are some principles that we apply:

First, that a particular fact or piece of evidence in the decision might lead to or facilitate damages actions does not mean that it is a business secret;

Second, although historically – before damages actions became common – there were no particular consequences to naming companies in our decisions even were the decision not addressed to them, that is clearly – after Pergan – no longer acceptable. If companies have been at best peripherally involved, or perhaps only indirectly implicated, in anti-competitive activity, then they should not be named at all unless the decision is in fact addressed to them. If they are named, then their names and any description of their actions should be confidential and should remain so.

Third, the public versions of decisions should not allow information provided by leniency applicants to be traced back to them. This falls under the general principle of the Commission needing to protect the system of leniency, so these are deletions that we will make on our own initiative, whether or not the parties have requested it (though they always will).

Ultimately however, we cannot publish decisions without the parties having identified their business secrets, and agreed to the final form of the publication. The scope for delay remains great.

This delay in making public versions of decisions available will, as many of you will be aware, lead to follow on actions being ready to proceed before a non-confidential version of a Commission decision is ready. Defendant addressees of the decision will have a confidential version in their possession, but how should the confidentiality issues be dealt with?

First, it is for the company having possession of the confidential document to ensure that it does not disclose confidential information; it could be liable for any unauthorised disclosure.

Second, if a national judge considers that dissemination of the confidential version is necessary for the litigation to continue, then it is for the national court to put in place appropriate mechanisms to safeguard the confidentiality of the relevant parts of the Commission decision.

That may take the form of ensuring appropriate redactions for “Pergan” evidence,[2] and putting in place confidentiality rings with appropriate sanctions should those rings be broken.

This mitigates some of the problems caused by the delay in publishing non-confidential versions of Commission decisions and may in time prove to be a more effective basis for the follow on litigation than waiting for the public version. Indeed, even if a non-confidential version of a decision has been published, that does not bind a claimant to using that version. It could always go to a national court and argue for further disclosure of the confidential parts of the decision in the context of, for example, a confidentiality ring. It may be that ultimately the Commission’s non-confidential version of its decision will prove to be less important than it currently appears.

The publication principles outlined above apply equally to contested and settlement decisions. But from the point of view of public enforcement, there are significant differences between what is legally required as to the content of contested and non-contested (settlement) cartel decisions.

The contents of contested and non-contested (settlement) decisions

Contested decisions are likely – in practice are certain – to end up in Court. They therefore need to paint a complete picture of the anti-competitive activity, and of each party’s participation in it.

As an aside, I personally think that even in a contested case, a competition enforcer should aim for the minimum amount of information needed to demonstrate its case, rather than the maximum; I do not think that a complete description of the evidence on the file is necessary if highlights from the file would be sufficient.  For a public authority to address its entire case-load efficiently means that it should focus its description of conduct on that which is necessary, which may fall short of that which is complete.

Particularly with leniency-fuelled cases the amount of factual evidence can sometimes be extensive. Documenting in the decision every weekly meeting of a cartel that lasted for ten years may not be necessary to prove the infringement, and may mean that resource limited enforcement agencies spend too much time on too few cases. It is of course a delicate balancing exercise, and the safe option will always be to add more. A public authority should guard against that.

But consistent with that principle, settlement decisions should be treated differently by a public authority. It is – or should be – a non-contested procedure, that is not appealed. The addressees have acknowledged the Commission’s Statement of Objections and are not likely to appeal the case to the General Court. The amount of detail necessary for the public enforcement of the decision is therefore substantially less than that needed for a decision where every fact stands to be checked in court. We have so far only had one appeal on a settlement decision, which raised issues relating to the calculation of the fine rather than the description of the infringement: I imagine the General Court would give short shrift to any appeals on settlement cases where the parties sought to appeal that which they had admitted during the settlement procedure on the basis that the decision was too short to prove the case.

I realise that these shorter settlement decisions make for slimmer pickings for claimant lawyers who want to glean as much information as they can from cartel decisions so as to aid discovery. That, however, is not the concern of the Commission when it is acting as a public enforcer.

We have proposed in the Damages Directive a system of discovery that – although its implications are limited here in the UK – will have major consequences in other jurisdictions around the EU. It is a system that should ensure that claimants get access, directly from the parties to the cartel, to pre-existing documents, including those on which the Commission has based its decision. It is in that way that follow on actions are secured, even in settlement cases.

Given the increasing importance of damages actions it is no surprise that defendants and their advisers are looking ever more closely at Commission decisions and pursuing arguments about confidentiality with more vigour.

I think the increasing reality of private damages follow on actions will have other effects on the way the Commission approaches its decisions. For example, that the Commission has changed how it drafts its decisions to exclude Pergan-type information is one example of how greater private enforcement is affecting Commission practice. It is now our practice not to name companies in cartel decisions unless they are an addressee. This will help to avoid some of the difficulties currently faced in the Air Cargo litigation.

[1] T-534/11, Judgment of the General Court (First Chamber) of 7 October 2014, Schenker AG v European Commission, not yet reported (not yet available in English).

[2] Case T-474/04, Judgment of the Court of First Instance (Third Chamber) of 12 October 2007, Pergan Hilfsstoffe für industrielle Prozesse GmbH v Commission of the European Communities, 2007 ECR II-04225.

Thomas Sharpe QC and James Flynn QC invited me to speak to the Lincoln’s Inn Eurogroup on the relationship between public and private antitrust enforcement.

This is the first main section of the speech. on the damages directive and the protection of leniency statements and settlement submissions. The others are:
Publications and the content of decisions;
Single and continuous infringement.

For those who prefer to read paper documents, I attach PDFs of the full prepared remarks and the handout.

In terms of setting the policy framework, as you all know the Damages Directive was agreed between the Council and European Parliament earlier this year and formally adopted earlier this month.[1] I will not go through all of the details, but want to highlight one important point: the absolute protection of corporate leniency statements and settlement submissions contained in Article 6 of the Directive.

This was a controversial area, in light of the Pfleiderer[2] and Donau Chemie[3] rulings of the Court of Justice that held that a national court had to carry out a balancing exercise in deciding what protection to give to corporate statements that were in the possession of the national competition authority. Both rulings made clear that this balancing exercise was necessary in the absence of EU law on disclosure. Nevertheless, some interpreted the rulings as setting out a principle that national judges should always be in a position to carry out a balancing exercise; this, notwithstanding that explicit reference in both rulings to the absence of EU legislation.

With the Damages Directive, the Commission proposal and the Council and Parliament final text have met the principled requirements of these rulings by carrying out a balancing exercise when looking at the file as a whole: for the Commission’s file, some documents – leniency statements and settlement submissions – are never, pursuant to Article 6, disclosable; some are disclosable, pursuant to Article 5, after the proceedings are finished.

And, of course, pre-existing documents are always disclosable from the parties/ third parties in accordance with the national rules in force. That a document has been submitted to the Commission as part of an immunity or leniency application does not give that document any special status or protection.

So even ignoring the letter of what the Court said in Pfleiderer and Donau about the absence of EU legislation – which the damages directive now provides – we have met the spirit of what the Court was concerned about by balancing the various interests at stake here – those of public enforcement and the maintenance of leniency on the one hand and those of damages claimants on the other.

Perhaps inevitably, one of the consequences of this absolute protection for corporate statements is a necessarily clearer definition in the directive of what can be included in a corporate statement:

A leniency application is defined at Article 2 of the directive in part as a voluntary submission “describing the undertaking’s or a natural person’s knowledge of a cartel and its role therein, which was drawn up specifically for submission to the authority with a view to obtaining immunity or a reduction of fines under a leniency programme”

A leniency programme is in turn defined as a programme where a participant “in a secret cartel, independently of the other undertakings involved in the cartel, cooperates with an investigation of the competition authority, by voluntarily providing presentations of his knowledge of the cartel and his role therein.”

The voluntary aspect mentioned in both of these definitions is important; these are not submissions that the European Commission – or any court or authority – could compel. So it is entirely appropriate that these wholly voluntary submissions should receive absolute protection.

It is also clear from the above definitions that a leniency application must be focussed on inculpatory information.

Pre-existing documents, including translations of documents in non-EU languages, should not be provided by way of a corporate statement.  Similarly product information and market information should be provided separately. Generally all other information relevant to the case that is not inculpatory should be provided to the Commission separately to the corporate statement itself.

We have published brief guidance – in October 2013[4] – on the practical aspects of making corporate statements, as well as what should and should not be included as a way of helping companies and their advisers, and this does make clear that general market information or other publicly available information should not be included.

The damages directive means that this is not an optional exercise for us.  In order to ensure continued protection of corporate statements, we have to make sure that the corporate statement procedure is used only for materials within the definition of the damages directive; essentially inculpatory information about secret cartels.

There is also a general efficiency point as well – the oral statement procedure is lengthy and time consuming. We have limited resources and we have to use them efficiently. In recent months I have had to tell lawyers not to use it to dictate a letter requesting an extension of deadline for a request for information, or to dictate a procedural waiver allowing us to co-operate with other authorities. Other colleagues have had to refuse to allow the procedure to be used for English translations of pre-existing documents in non-EU languages.

Turning to settlement submissions, essentially the same principles apply as they do to corporate statements, with one provision added to Article 5 of the Directive by the Council and Parliament to the Commission’s original proposal – the treatment of withdrawn settlement submissions.

If we offer the settlement route to parties in a Commission case, it is for the Commission first to set out its case, and if the parties agree, they then provide a settlement submission. The various settlement submissions are then incorporated into a statement of objections to which the parties must acknowledge their agreement.  There is therefore no scope in the procedure for a draft submission. Though were a party to submit a draft then, in my personal view, that should – given the spirit of the directive – be covered by the same protection afforded to final submissions.

The Council and Parliament introduced a provision into the adopted directive which provides that withdrawn settlement submissions are not given the absolute protection afforded to settlement submissions that are not withdrawn. I cannot comment on what implications this might have for settlement procedures at the national level, but for the European Commission, our procedure does not allow for withdrawn settlement submissions; we have never had such a case and do not, given our rules, expect one.

Technically, it is important, however, to distinguish the possibility that parties withdraw settlement submissions – again not permitted in our settlement system – from situations where the Commission decides that it does not accept the settlement submission, perhaps because the Commission believes it does not fully reflect the Commission’s understanding of the case. The Commission then will reject the option of pursuing a settlement case and will revert to a normal, contested, procedure. This is not a case of a settlement withdrawn by the party, however, but of one rejected by the Commission. In those circumstances, the Commission would regard the rejected settlement submissions as meriting the protection afforded by the directive. I say it is technically important to make this distinction, because in practice this has again never happened and is again never likely to.

If the Commission has set out its position on the case, and the parties have agreed and submitted a settlement submission, it is hard to see why a party would then want to withdraw it.

[1] DIRECTIVE OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL 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, of 10 November 2014, not yet published.

[2] Case C-360/09, Judgment of the Court (Grand Chamber) of 14 June 2011, Pfleiderer AG v Bundeskartellamt, 2011 ECR I-05161.

[3] Case C-536/11, Judgment of the Court (First Chamber) of 6 June 2013, Bundeswettbewerbsbehörde v Donau Chemie AG and Others, not yet published.

[4] DG Competition publication: Delivering oral statements, published 8 October 2013, available at: http://ec.europa.eu/competition/cartels/leniency/oral_statements_procedure_en.pdf.

A small point, but one worth repeating. Competition law works to benefit society, not necessarily investors in individual companies.

This story may be apocryphal but I heard of a headline in an – investor focussed – newspaper, after an air crash that focussed more on the share price of the airline than on the number of the dead. I can’t think of a better example of an investor perspective trumping a consumer one.

Competition law is on the side of consumers (as a whole) and society (as a whole), not necessarily of investors (in individual companies).

Competitive markets give consumers the best choice at the lowest price. Markets as a whole will grow, though individual companies will have to compete harder and may see their profits squeezed as a result with a consequent impact on their investors. Some companies may go out of business. But society as a whole benefits, and consumers benefit; investors in individual companies may not.

Warren Buffet famously looks to invest in companies with a “moat”. He wants companies that are buffered from the competition, not buffeted by it. (Sorry.) There’s nothing wrong with that. As an investment strategy it makes a lot of sense and has been famously successful for Mr Buffet. Nor is there anything wrong with the existence of a moat.

But the moat is not necessarily good for society. And some actions to strengthen the moat may be unlawful. That’s where competition law gets interesting.

The final text of the Council and Parliament Damages Directive protects corporate leniency statements and settlement submissions. There are two small caveats.

First, national judges have the right to verify that corporate leniency statements are in fact that, and are not concealing other information.

Second, settlement submissions that the parties subsequently withdraw are potentially accessible to claimants.

These caveats are minor, however, and unlikely to have any real impact in practice. No settlement submission has ever been withdrawn, and it’s not entirely clear what other information could be concealed in a corporate leniency statement.

Of more interest is the relationship between the Directive and the Court of Justice’s rulings in Pfleiderer and Donau Chemie.

At paragraph 23 of Pfleiderer, the Court indicated that:

Accordingly, even if the guidelines set out by the Commission may have some effect on the practice of the national competition authorities, it is, *in the absence of binding regulation under European Union law on the subject*, for Member States to establish and apply national rules on the right of access, by persons adversely affected by a cartel, to documents relating to leniency procedures. (emphasis added

Similarly at paragraph 25 of Donau Chemie:

In the absence of EU rules governing the matter*, it is for the domestic legal system of each Member State to lay down the detailed procedural rules governing actions for safeguarding rights which individuals derive from EU law.(emphasis added

Some have questioned whether the absolute protection in the directive is consistent with the Court of Justice’s rulings that judges must make an assessment on a case by case basis. I think it’s fairly clear that it is, for two reasons.First, and most practically, both rulings explicitly referred to the Court setting out rules in those cases in the absence of Community law on the subject. This strongly suggests that if the Community legislature were to set out rules, then the Court would give those deference even if they deviated from the Court’s rulings in Donau and Pfleiderer. The rulings, implicitly, do not set out fundamental rights, but merely the Court’s own balancing of interests in the absence of, and subject to, legislative guidance.

Secondly, the Court required that a national judge carry out a balancing exercise between the various interests at issue – in particular the interest in protecting effective leniency systems and the interest of victims of anti-competitive behaviour to claim damages.

The Directive explicitly carries out this balancing exercise in making clear that pre-existing documents – even if they were submitted to the Commission in the context of a leniency submission – remain accessible to victims. There is no leniency shield for such pre-existing documents. However documents created solely for the purpose of leniency applications or settlement submissions – and which would therefore never have been created if a leniency/settlement system were not in place – are shielded from disclosure.

This is not a case by case balance, but a legislative balance looking at the competition enforcement system as a whole. There is nothing in the Court’s rulings to suggest that such a balance is problematic.

Quite the opposite, given that in its recent rulings in relation to access to the Commission’s file under Regulation 1049/2001, the Court has put great weight on the importance of public enforcement, and on the risks to enforcement of widespread disclosure of documents on the Commission’s file. It is hard to think of a category of document the disclosure of which would lead to more damage to enforcement than leniency statements.

I spoke at IBC’s Advanced EU Competition Law event in Brussels in November, presenting an annual overview of cartel developments with Johan Ysewyn of Covington & Burling. Julie Adyns did the hard work of putting the slide pack together.