The Dublin Regulation: Is the End Nigh? Where should unaccompanied children apply for asylum?

Steve Peers*

Two recent developments have raised controversy as regards the EU’s Dublin III Regulation, the set of rules which determines in which Member State asylum-seekers must make their asylum application. First of all, a British judgment yesterday stated that the UK was responsible for the asylum claims by unaccompanied children in France (in particular the Calais ‘Jungle’), who have a family member in the UK. Secondly, a press report indicated that the Commission is planning to propose a fundamental overhaul of the Dublin rules in the near future. Both developments have alarmed some commentators, but thrilled others. I will examine the legal and political context of each of them in turn.

Unaccompanied minors

Many describe the Dublin III Regulation as allocating responsibility to the ‘first Member State’ which an asylum-seeker entered. This is true for most asylum-seekers; technically it’s the first Member State which they entered without authorisation which has responsibility, but this amounts to the same thing. However, there are exceptions. In particular, since the very first version of the Dublin rules (the Dublin Convention), there have been special rules which apply where the asylum-seeker has a family member in one Member State. These were expanded in the Dublin II Regulation, among other things to add special rules where the asylum-seeker is an unaccompanied minor. Those rules were altered a little further in the Dublin III Regulation.

What are the rules for unaccompanied children? The priority is to place them in the same Member State as a family member. First of all, the Regulation gives responsibility to a Member State where they have a ‘family member’ (defined as a parent, spouse or child) or a ‘sibling’ who is ‘legally present’. Secondly, it gives responsibility to a Member State where they have a ‘relative’ (defined as an adult aunt, uncle or grandparent) who is ‘legally present’. In this second case, there are further conditions: there must be an individual examination to check that the relative is able to take care of the child, and the allocation of responsibility must be in the best interests of the child.

In the absence of a family member or relative, an unaccompanied minor – unlike any other asylum-seeker – in effect has a choice of which Member State to apply in. The CJEU has confirmed (in the case of MA) that this applies even after the child has already applied in one Member State. A subsequent proposal to confirm and extend this rule (which I discussed here) seems to be blocked for now. But this rule was not at issue in yesterday’s judgment.

Despite the consternation it caused in some quarters, that judgment is primarily straightforward. Once it’s clear that the asylum-seeker is a child who has family or relatives as defined by the EU rules in a particular Member State, the child must be transferred to that Member State to apply for asylum there. Sometimes it’s hard to prove the age of the child or the link to family or relatives, but it doesn’t seem like that was the case here (it’s hard to be certain, since the full text of the judgment is not reported yet).

One of the asylum-seekers in the case was not a child, but was dependent on a younger brother due to the effects of trauma suffered in Syria. His situation was covered by a separate clause in the Dublin III Regulation which says that ‘dependent persons’ should stay with a child, sibling or parent who can take care of them. This is a binding rule (‘shall normally keep or bring together’), as established in the CJEU judgment in Kand confirmed in the preamble to the Dublin III Regulation.  

Why was the judgment controversial? First of all, there is a particular legal point: the rules in the Regulation only take effect for those who have applied for asylum, following which the Member State where they have applied is obliged to contact the responsible Member State and arrange for the transfer to that State. The judgment appears to circumvent that process, simply requiring the UK to admit the four plaintiffs despite the absence of any action by the French authorities. It appears from press reports that at least one of the applicants did have some proof of having applied for asylum in France, but it is not clear if all of them did. This will probably be the basis of an appeal which the UK government might make – although the applicants will be allowed into the UK in the meantime.

On the face of it, this is a valid legal objection: the proper procedures were not followed. Having said that, the judgment is indisputably consistent with the substantive intention of the drafters of the law: to ensure that children (and dependent persons) who apply for asylum are with people who can look after them. Since all the plaintiffs are Syrian, there seems little doubt that they intend to apply for asylum (and perhaps had done already) – or that their application will be successful (the refugee recognition rate for Syrians being over 90%). 

More fundamentally, the plaintiffs alleged that the French government did not (or would not) process their asylum applications. If this is true, the French government is in breach of EU law, and it is arguable that its breach should not be allowed to stand in the way of applying the rules on asylum responsibility. But this line of argument raises complex legal questions about how to prove such fault and who has the burden of proving it – and whether such a fault justifies a procedural shortcut at all. It would be best if these issues are sent to the CJEU to clarify (it can use an emergency procedure to decide on cases involving children). That would also make it more obvious that these issues do not just concern the UK and France: there may be unaccompanied children seeking asylum in Greece who seek to join family or relatives in Sweden, for instance.

Secondly, there are political objections on the grounds that UK immigration law has been infringed. It should be noted that the UK chose to opt in to the Dublin III Regulation – while it opted out of almost every other EU immigration and asylum law of the last twelve years. The attraction for the UK was the possibility that some asylum-seekers could be sent back to other Member States – which they are. But the rules are reciprocal: sometimes they are bound to mean that the UK has to accept asylum-seekers from other Member States. In practice, most or all such cases will involve family members. And quite frankly, anyone who argues that in principle an unaccompanied child who is seeking asylum from war or persecution should remain in squalor in Calais or Dunkirk, rather than join a family member legally in the UK who can look after him or her, is devoid of basic humanity.

What is the impact of the ruling? It cannot affect anyone who has not got family members in the UK, subject to the conditions mentioned above. In some cases, it will be hard to prove that the applicant is a child, or that the person they seek to join is a family member. The procedural aspects still need to be clarified. But for anyone else, as I discussed last summer, the French government quite rightly remains responsible for dealing with their asylum applications and providing humane living conditions, or for returning those who have not applied for asylum or whose applications have failed to their country of origin.

Plans to amend the Regulation

The key feature in the reported plans to amend the Regulation is the intention to replace the ‘first country’ rule with a set of criteria allocating responsibility to Member States based on some kind of ‘fair shares’ principle. Presumably some rules allocating responsibility based on family members will remain, perhaps with amendments. No further details are known, and it should not be forgotten that the Commission proposal would have to be agreed by both the European Parliament and the Council. But I will focus on two key issues: the feasibility of the new system, and the impact on the UK.

First, the feasibility has to be assessed in light of the EU’s existing rules on ‘relocation’ of some asylum-seekers who reached Greece or Italy. Two measures were adopted in September, and I discussed them in detail at the time. Four months on, there are significant problems applying these rules in practice, as the Commission’s regular reportsindicate (see also the UNHCR’s assessment). A tiny fraction of the asylum-seekers have been relocated, due to the slow development of ‘hotspots’ for registering applications in Greece and Italy, as well as the reluctance of most Member States to receive asylum-seekers. Some Member States have flat-out refused to apply the system, and Slovakia and Hungary have challenged its legality (see the analysis of Slovakia’s challenge by Zuzana Vikarska here). The ‘hotspots’ are arguably not properly considering the asylum applications of many asylum-seekers who are not due for relocation (see analysis by Frances Webber here). While the Commission has proposed already to amend the Dublin III Regulation to make emergency relocation rules permanent, the Member States are objecting to this (see the leaked record of discussions here).

Overall, then, the relocation system is manifestly not working. It is therefore hard to see why the Commission could imagine that it could somehow work if it became the general rule, rather than the exception as at present (the relocation Decisions only apply to 160,000 asylum-seekers over two years, a small minority of the numbers that might be expected over that time).

Secondly, the UK, as already noted, has an opt-out from EU immigration and asylum law. This opt-out applies also to amendments to legislation that the UK already takes part in. Indeed, the UK has opted out of most of the ‘second phase’ of EU asylum legislation, even though it opted in to all the ‘first phase’ measures. So it is flagrantly not true to say that the UK could be ‘forced to accept more refugees’ under any new proposal. It could simply opt out.

However, that opt out could have consequences. The Protocol on the UK opt out says that if the opt out from an amendment to an existing law in which the UK participates makes the existing law ‘inoperable for other Member States or the Union’, that existing law can be revoked as regards the UK. This is a high threshold, and this clause has never been invoked to date. The UK was able to opt out of the relocation decisions, and of the recent proposal to amend Dublin III to add a permanent emergency system, without triggering this clause (as I discussed here). However, a complete overhaul of the Dublin system, replacing the core rules on the allocation of asylum responsibility, is likely to trigger it.

In that case, the UK would face a choice: (a) opt in to the new rules, and face more asylum-seekers as a result, or (b) opt out of the new rules, and face more asylum-seekers because the existing Dublin rules would be terminated for the UK. For those (like the UK government) who would not like to see more asylum-seekers, neither option is appealing. But it is likely that there would be a far smaller increase in asylum-seekers if the UK opts out. Recent statistics collated by Open Europe suggest that the UK only returns about 700 asylum-seekers a year to other Member States pursuant to the current Dublin Regulation in recent years; and yesterday’s ruling may mean that this may have to be offset against flows in the other direction. On the other hand, one estimate in The Independent suggests that a ‘fair shares’ rule could mean that the UK is responsible for 85,000 asylum-seekers a year.

Finally, what would be the impact of ‘Brexit’ on asylum-seeker numbers? Since the current Dublin rules are internal EU legislation (they ceased to take the form of a ‘Convention’ back in 2003), the UK would no longer be covered by them as from Brexit Day. The EU could sign a ‘Dublin’ treaty with the UK, but I rather doubt it would do so, for the reasons I discussed here). So Brexit would increase the numbers of asylum-seekers in the UK, assuming that the current Dublin rules are then still in force, and still result in a net removal of asylum-seekers from the UK. If there are new Dublin rules, and the UK has opted out of them, then Brexit would have no effect. Brexit would only reducethe number of asylum-seekers if the Dublin rules are still in force and have resulted in a net inflow of asylum-seekers to the UK.  In any event, Brexit would mean that unaccompanied children seeking asylum are left alone in other Member States rather than joining parents or other family members who are legally in the UK and able to look after them.

Barnard & Peers: chapter 26

JHA4: chapter I.5

Photo: children in the Calais ‘jungle’

Photo credit: DailyMail.co.uk

 *Disclosure: I have done some work as a sub-contractor for a contractor advising on the evaluation of the Dublin III Regulation and the impact assessment on its replacement. However, while I am bound to confidentiality as regards that work, I was paid for my independent advice and remain free to express my views on the current or future Dublin system. 

On knowledge as power: transparency of EU law-making procedures

Päivi Leino (University of Helsinki)

One of the great achievements of the Lisbon Treaty, flagged around its adoption and entry into force, was how the new Treaty would make EU law-making much more transparent. So far, this has remained an unfulfilled promise. Access to documents relating to the EU legislative procedure, in particular trilogues, the informal three-party meetings between the European Parliament, the Council and the Commission used at every stage of the EU legislative procedure, has become particularly topical during the past months.  First, the new Interinstitutional agreement (IIA) on Better Regulation, provisionally approved in December 2015, addresses this question.[1]Second, on 15 December 2015, Emily O’Reilly, European Ombudsman, launched a public consultation on transparency of trilogues,[2] stressing their role as the forum where the deals are done and the subsequent need to consider the proper trade-off between the Europeans’ right to open EU law-making processes and the space to negotiate. Finally, Emilio De Capitani, the previous head of the LIBE Committee Secretariat, has brought an appeal against the European Parliament’s decision to refuse full access documents relating to a legislative proposal.[3]De Capitani argues in essence that granting access to them would not specifically, effectively and in a non-hypothetical manner undermine the legislative decision-making process, and that notably after the Lisbon Treaty, legislative preparatory documents are subject to the principle of widest possible access.

The recent events raise a number of fundamental questions relating to how we understand the function of transparency in law-making that claims democratic foundations. Moreover, since EU law-making is currently not backed up by any shared interinstitutional space where documents and meeting schedules would be recorded in real time, following legislative procedures requires a serious amount of detective work for those with an interest in following them, and increases reliance on a culture of leaks. In brief, exercising your democratic rights should not be this difficult.

The legal framework of law-making in the EU

The main principle in the Treaty of Lisbon is clear: the Council and the European Parliament are to legislate in the open, and are under an obligation to ensure the publication of the documents relating to the legislative procedures. Regulation No 1049/2001 includes some references to legislative documents. Under Article 12(2), legislative documents meaning “documents drawn up or received in the course of procedures for the adoption of acts which are legally binding in or for the Member States” should be made directly accessible unless one of the exceptions under the Regulation is applicable. Therefore, making these documents publicly available should not presume specific requests but should instead take place automatically. These provisions have been subject to the Court’s landmark ruling in Turco concerning access to Council legal service opinions, which stresses how increased openness

enables citizens to participate more closely in the decision-making process and guarantees that the administration enjoys greater legitimacy and is more effective and more accountable to the citizen in a democratic system. Those considerations are clearly of particular relevance where the Council is acting in its legislative capacity […]. Openness in that respect contributes to strengthening democracy by allowing citizens to scrutinize all the information which has formed the basis of a legislative act. The possibility for citizens to find out the considerations underpinning legislative action is a precondition for the effective exercise of their democratic rights.[4]

In the ruling, the Court set the threshold high in referring to ‘all the information which has formed the basis for a legislative act’: it is clearly not only quantity that is of relevance, but also quality. The legislature needs to provide access to information concerning the key elements on the legislative agenda while the relevant process is on-going and there is a chance to influence it.

However, the institutions, and the Council in particular, have had difficulties with maintaining this standard and turned to a vast number of excuses that risk to water down the principle in rather fundamental ways.[5] In the Council’s view, the application of the Treaty provisions stipulating the publicity of its meeting are restricted to documents that are submitted to the Council for the preparation of the latter’s deliberation or vote.[6]This raises the rather fundamental question of the relevance of the stage of decision-making for the application of legislative transparency. The Council understanding would seem to deprive the principle of much of its democratic effet utile; If the publicity of legislative documents was indeed limited to only those documents that are issued a couple of days before a formal ministerial meeting where a legislative file is to be deliberated or approved, often without any discussion, then transparency would only be applied at the point when the file is to be closed and the process is over. This would effectively hamper access to documents produced during the time when the majority of questions are settled at lower decision-making levels between civil servants, ambassadors and the three institutions. The Council has also argued that the early stage of the legislative process between the institutions, namely first reading, is a point to be taken into consideration. [7] At the same time, a major part of legislative files are closed specifically at first reading: For example in 2009-2013, 83 % of legislative files were closed at first reading, 8 % during early second, 7 % at second, and only 3 % at third.[8]

The relevant provision in Regulation No 1049/2001 in this regard is the ‘space to think’ exception in Article 4(3), relating to situations where a decision has not yet been taken by the institution. The Parliament has repeatedly voiced the argument that the said provision is outdated as far as legislative matters are concerned,[9]and suggested both that no exception would ever apply to ‘documents transmitted within the framework of procedures leading to a legislative act or a non-legislative act of general application’ but also the deletion of Article 4(3) altogether.[10]The latter is a far-reaching proposal that might benefit from further reflection considering that the exception has been used for varying purposes outside the legislative context.

However, Article 4(3) was interpreted in the post-Lisbon legislative context by the Court when Access Info Europe, an NGO promoting freedom of information in the EU, requested access to a legislative document including footnotes indicating the positions of individual Member States. The central question was whether access to Member State positions distracts the effectiveness of decision-making and if yes, which one should take priority, effectiveness or openness. The Council lost the case in the General Court and appealed to the Court of Justice[11]arguing that the General Court’s reading had attached ‘undue and excessive weight to the transparency of the decision-making process, without taking any account of the needs associated with the effectiveness of that process’, and disregarded the balanced approach laid down both in primary law and secondary law between the two objectives. The Council argued that ‘its legislative process is very fluid and requires a high level of flexibility on the part of Member States so that they can modify their initial position, thus maximising the chances of reaching an agreement’. In the Council’s view, identifying the delegations was not necessary for ensuring a democratic debate. The CJEU rejected this with reference to how full access can be limited only if there is a genuine risk that the protected interests might be undermined. The high standard of proof required to establish that level of harm makes it almost impossible to rely on Article 4(3) in this context. In particular, according to the Court,

the various proposals for amendment or re-drafting made by the four Member State delegations which are described in the requested document are part of the normal legislative process, from which it follows that the requested documents could not be regarded as sensitive – not solely by reference to the criterion concerning the involvement of a fundamental interest of the European Union or of the Member States, but by reference to any criterion whatsoever (para 63).

So far, the Court has stood rather firm in promoting transparency of core legislative documents,[12]and is likely to have a word or two to say in the De Capitani case. However, its rulings are of a limited importance as long as it is accepted that the institutions may apply them selectively.[13]

Access to trilogue documents

Trilogues are increasingly taking over as the main forum for making legislative deals between the three institutions.[14]They are the prime example of informal decision-making, which is often glorified in the EU context as the tool guaranteeing efficient law-making.[15]In first reading deals trilogues function as the forum for interinstitutional agreement. The Treaties make no reference to this forum. In the joint declaration on practical arrangements for the co-decision procedure adopted in 2007, the three institutions argue that the trilogue ‘system has demonstrated its vitality and flexibility in increasing significantly the possibilities for agreement at first and second reading stages’. According to the declaration, trilogues ‘may be held at all stages of the procedure and at different levels of representation, depending on the nature of the expected discussion’.[16]The use of informal formats is not in any way restricted to technical, urgent or uncontested files.[17]Trilogues are an incredibly efficient format for accommodating institutional positions, and have led to a great majority of deals being closed early in the legislative procedure.

During the trilogue phase, the EU democratic process is in the hands of very few: the European Parliament rapporteur(s), the representatives of the Council Presidency and Secretariat and a few Commission officials. This phase largely escapes public scrutiny. The joint declaration quoted above stipulates that ‘trilogues taking place within the European Parliament and Council shall be announced, where practicable’, and ‘when conclusion of a dossier at first reading is imminent, information on the intention to conclude an agreement should be made readily available as soon as possible’. There is no mention of access to information while the process is in the substantive phase. Civil society representatives have, for a very good reason, pointed out how trilogues represent the victory of efficient law-making over the formal machinery of law-making, which results in a general lack of transparency.[18]

The informal arrangements surrounding trilogues not only affect public access, but also access within the institutions. Information is power, and the trilogue format empowers those that possess it at the expense of those who do not. Political scientists have demonstrated how within the Parliament, first reading agreements emphasise the discretion given to its own rapporteur(s)[19] and stress the role of responsible committee. Handing out information is in the hands of the rapporteurs, whose tweets commenting on the conduct of negotiations also create a practical avenue to information for many Member State governments. As far as the Council is concerned, first reading agreements mean in practice that deals are effectively concluded before they reach the ministers, since the Council position is settled by Coreper. In principle, feedback from trilogues is given to the delegations either through working parties or Coreper. National parliaments often experience difficulties following decision-making in trilogies, especially since amendments are made at great speed, which hinders their effective scrutiny at national level.[20]

Ombudsman inquiry and the new IIA

The European Ombudsman initiative is therefore timely, even if the objective of ensuring transparency in the legislative procedure reaches beyond its current scope. Last May, the Ombudsman approached the three institutions about their proactive disclosure policies on trilogue documents.[21]In his reply to the Ombudsman, Martin Schultz, President of the European Parliament, describes the current trilogue-related practices in the Parliament, but focuses on its internal transparency and accountability, which he sees as requiring that all political groups may follow and influence the negotiations. This speaks against an undue formalisation of the trilogue process. The Commission challenges the Ombudsman’s mandate to engage in the own-initiative inquiry altogether. In its view, trilogues are a

preparatory step in negotiations of a purely legislative nature and it is hard to see how any aspect of their conduct, such as whether or when to hold trilogue meetings, the choice of representatives by each institution to attend trilogue meetings, whether or which supporting documents to produce etc., can give rise to maladministration, except as regards questions related to public access to the documents used during trilogue meetings.

In its reply, the Commission stresses the internal nature of any possible briefing documents and lack of systematic records of what takes place during the trilogues. And the Council, not surprisingly, agrees with the Commission:

The Council is of the view that the exercise of legislative powers is not limited to the adoption of political choices on the merits of legislative files. It also includes the choices according to which the legislators decide to organise the legislative process itself. The organisation of the legislative process cannot be considered an administrative activity – and therefore cannot give rise to possible instances of maladministration – but ought rather to be regarded as an essential aspect of the exercise of the legislators’ prerogatives.

In other words, hands off legislative transparency, European Ombudsman. And yet, it is difficult to think of a matter where issues relating to access to documents and information in the broad sense would be more vital in a union that is tackling a permanent democratic and legitimacy deficit.  

In the new IIA, provisionally concluded in December 2015, the Commission proposal on how the ‘three institutions will ensure an appropriate degree of transparency of the legislative process, including of trilateral negotiations between the three institutions’[22]is strengthened. The three institutions now commit to ensuring ‘the transparency of legislative procedures, on the basis of relevant legislation and case-law, including an appropriate handling of trilateral negotiations’. For this purpose, they ‘improve communication to the public during the whole legislative cycle’ and ’undertake to identify, by 31 December 2016, ways of further developing platforms and tools to this end, with a view to establishing a dedicated joint database on the state of play of legislative files’.[23]While promises carry little weight if institutional attitudes remain unchanged, the latter commitment could contribute not only making more documents available earlier in the process and making their identification easier, also to shedding light on the vast number of legislative documents that are currently not made public while the process is pending.

In more than one way, the recent debates illustrate how knowledge is power; therefore, those that are in possession of it are unlikely to volunteer to distribute it. In this respect, what we need is a much more political debate on the way in which the exercise of democratic rights is effectively exhausted, often in practice through the choices of technocrats in the EU institutions, who make choices about how individual requests for documents are answered and which documents are placed on the web. Are the current institutional politics acceptable, or are they not? And more fundamentally, is efficiency such an overarching value that it triumphs in all circumstances over all other objectives, such as securing a transparent and democratic decision-making process?

Photo credit: pressgazette.co.uk

---

[1] Provisional text of the proposed interinstitutional agreement on better regulation, dated 16 December 2015, available at http://ec.europa.eu/smart-regulation/better_regulation/documents/20151215_iia_on_better_law_making_en.pdf, Article 28-28a.

[2] http://www.ombudsman.europa.eu/en/press/release.faces/en/61593/html.bookmark .

[3] Case T-540/15, De Capitani v European Parliament.

[4] Joined Cases C-39/05 P and C-52/05 P Kingdom of Sweden and Maurizio Turco v the Council, paras 45-46 

[5] See also Päivi Leino, “Transparency, Participation and EU Institutional Practice: An Inquiry into the Limits of the ‘Widest Possible’”, EUI Working Paper (LAW 3/2014). Available at http://cadmus.eui.eu/bitstream/handle/1814/30580/LAW_2014_03_Leino.pdf .

[6] See the Council reply to Confirmatory application No 05/c/01/12 by Ms Anneli Jäätteenmäki, MEP.

[7] For the Council’s reply to ClientEarth, see http://register.consilium.europa.eu/pdf/en/10/st12/st12068.en10.pdf . 

[8] The figures can be found on the Parliament’s website http://www.europarl.europa.eu/code/about/statistics_en.htm . 

[9] The ‘Hautala / Sargentini report’, Public access to documents 2009-2010 European Parliament resolution of 14 September 2011 on public access to documents (Rule 104(7)) for the years 2009-2010 (2010/2294(INI)) P7_TA(2011)0378. 

[10] Public access to European Parliament, Council and Commission documents (recast), P6_TA(2009)0114 A6-0077/2009 . 

[11] C-280/11 P Council v Access Info Europe. 

[12] Even if the General Court recently made a true bummer relating to impact assessments, which play a key role in the early stages of the legislative procedure; see T-424/14 ClientEarth v. the Commission.

[13] See Päivi Leino, “Transparency, Participation and EU Institutional Practice: An Inquiry into the Limits of the ‘Widest Possible’”, EUI Working Paper (LAW 3/2014); available at http://cadmus.eui.eu/bitstream/handle/1814/30580/LAW_2014_03_Leino.pdf

[14] ‘Shifting EU Institutional Reform into High Gear: Report of the CEPS High-level Group’, pp. 1-24, Report available at http://www.ceps.eu/book/shifting-eu-institutional-reform-high-gear-report-ceps-high-level-group .

[15] See in greater detail, Päivi Leino “The Politics of Efficient Compromise in the Adoption of EU Legal Acts” in Marise Cremona (Ed.), EU Legal Acts: Challenges and Transformations, Collected Courses of the Academy of European Law (Oxford University Press, forthcoming 2016.)

[16] Joint declaration on practical arrangements for the Codecision procedure. OJ 2007 C 145/02 .

[17] See Bressanelli, Héritier, Koop and Reh, ‘The Informal Politics of Codecision: Introducing a New Data Set on Early Agreements in the European Union’, EUI Working Papers RSCAS 2014/64.

[18] See e.g. Bunyan, ‘European Parliament: Abolish 1st [and 2nd] reading secret deals – bring back democracy ‘warts and all’’, Statewatch analysis available at: http://www.statewatch.org/analyses/no-84-ep-first-reading-deals.pdf

[19] Discussed in Hix and Hoyland, ‘Empowerment of the European Parliament’, 16 Annu. Rev. Polit.Sci (2013) 171, at 184. For rules concerning the Parliament’s representation and position-building, see European Parliament Rules of Procedure, Rule 70.

[20] House of Lords Report with Evidence. Codecision and national parliamentary scrutiny, HL Paper 125 (2009) 15-16.

[21] The opinions are available on the Ombudsman website, http://www.ombudsman.europa.eu/en/cases/correspondence.faces/en/61592/html.bookmark .

[22] Proposal for an Interinstitutional Agreement on Better Regulation, COM(2015) 216 final.

[23]  Provisional text of the proposed interinstitutional agreement on better regulation, dated 16 December 2015, available at http://ec.europa.eu/smart-regulation/better_regulation/documents/20151215_iia_on_better_law_making_en.pdf . 

Enforcement Priorities Paper on Article 102 TFEU: Is a Title Enough to Overtake Constitutional Rules and Fundamental Rule-of-Law Principles?

Konstantinos Sidiropoulos

DPhil Candidate at University of Oxford – Prof Steve Weatherill

Foundation for Education and European culture scholar

Introduction

There has been an intense debate around the optimal standard regarding abuse of dominance control in the EU. In an attempt to address the multiple complaints against the approach adopted to unilateral conduct control, in the early 2000s, the European Commission initiated a review of the law and practice of the relevant provision (see the EAGCP Report and the Discussion Paper). The recurring objective of the review was to inject more economic thinking into the enforcement of what is now Article 102 TFEU (see e.g. hereand here). In this context, the ‘traditional form-based analysis’ of the EU Courts was contrasted to a allegedly superior ‘more economic approach’ to abuse of dominance.

In December 2008, the Commission adopted a guidance paper declaring its enforcement priorities when dealing with exclusionary abuses under Article 102 TFEU (‘Enforcement Priorities Paper’). This represents the culmination of the debate that crystalised the Commission’s modern approach to abuse of dominance control.

Legal Status

The Enforcement Priorities Paper is a soft law instrument of sui generis nature which is said to operate as a statement of prosecutorial discretion. The said paper is of sui generis nature with regard to both its form and its substance.

From a formal perspective, it is a ‘novel instrument’ in that it has not been adopted in the form of a Notice or of Guidelines, but as a Communication containing ‘guidance on enforcement priorities’. This is neither a legal act envisaged in Article 288 TFEU nor an instrument recognised by the case-law as containing ‘rules of practice’ (see e.g. Dansk Rørindustri at para 209). Inevitably, this raises doubts as to whether the Priorities Paper may be taken into consideration pursuant to the Grimaldijudgment (para 18). 

            The Priorities Paper is also a novelty from a substantive perspective. Its content is said to be about enforcement priorities, and not about substantive guidelines, i.e. it allegedly explains where the Commission will focus its resources rather than interpreting the law (paras 2-3). This label is due to the fact that the Commission was restricted by the EU Courts’ jurisprudence in its attempt to evolve its policy with regard to Article 102 TFEU. Producing substantive guidelines was not an attractive option, because this would mean that it could do no more than describe the current state of the law (see the European Parliament’s reportat points K, L and N), and this would stand at odds with the Commission’s economics-based agenda. Under these circumstances, the adoption of a document that would claim to set enforcement priorities was seen as a wise way for the Commission to suggest that the existing law is unsatisfactory, while not directly challenging the interpretation of the jurisprudence.

At a superficial level, the constitutional question of whether the Commission has the authority to indicate ex ante which categories of practices it will prioritise when enforcing Article 102 TFEU was answered in the affirmative, by virtue of the judgment in Automec, where the GC held that ‘setting priorities within the limits prescribed by the law […]  is an inherent feature’ of the Commission’s administrative activity (para 77). Nevertheless, Automec cannot be regarded as an authority for the adoption of the Enforcement Priorities Paper for two reasons. To start with, its title is misleading because nothing is really prioritised. The Priorities Paper is drafted as substantive guidelines and mentions all the categories of exclusionary conduct that have been found to be abusive in the jurisprudence. Additionally, the recognition by the EU Courts of the Commission’s discretion to prioritise cases is not unlimited.  Above all, the Commission must act within the limits prescribed by the law; it must respect, inter alia, the interpretation given to Article 102 TFEU by the CJEU, since the latter is entrusted with the monopoly in the interpretation of EU law by virtue of Article 19(1) TEU in conjunction with Articles 267(3) and 344 TFEU. In this connection, the Priorities Paper is inconsistent with the case-law in several respects.

Consequently, the Priorities Paper is not about setting priorities in the way that it is understood in Automec; it is about providing substantive guidelines. Neither the caveat in para 3 of the Priorities Paper which states that it ‘is not intended to constitute a statement of the law’ nor the title of the document may affect this conclusion. Moreover, the attempt by certain distinguished lawyers in the Commission to reshape the objective of the Commission’s review, arguing that the case-law and the Priorities Paper do not overlap (see e.g. hereat p. 7), is not persuasive.

Even so, the Priorities Paper is deprived of legally binding force. Nonetheless, it could be argued that, being a soft law instrument, it stipulates ‘rules of conduct which are designed to produce external effects’ according to the Archer Daniels Midland judgment (para 91). These effects comprise two interrelated aspects, namely that they operate as instructions on the Commission’s administrative practice, and that the Commission cannot depart from these soft rules in an individual case without giving reasons for doing so. However, two aspects of the Priorities Paper prevent it from producing the legal effects of all other soft law instruments. First of all, it departs from the case-law. For such a non-binding text to produce legal effects, not only must it be in conformity with primary and secondary EU law (see e.g. Dansk Rørindustri at para 252), but it must also be consistent with established jurisprudence (see e.g. Dansk Rørindustri at para 261).

Secondly, the CJEU recognises legal effects to soft law instruments only to the extent that this serves the promotion of the protection of legitimate expectation and the principles of equal treatment and legal certainty (see e.g. Dansk Rørindustri at para 211). The Priorities Paper however does not produce legitimate expectations and is unable to serve the principles of equality and legal certainty. To start with, it cannot give rise to legitimate expectations, since it does not give precise assurances that a firm’s conduct will go unpunished if it does not fall within the scope of its provisions (para 3). Reliance on the principle of the protection of legitimate expectations is unacceptable if no precise assurances are given by the authorities that they will act in a particular way (see e.g. the GC’s judgment in Intelat paras 161-166). In addition, a dominant firm cannot invoke the right of equal treatment to protest that the Commission did not focus its resources on pursuing another’s firm exclusionary behaviour, because the announcement that intervention against certain practices will be an enforcement priority does not imply the lawfulness of other behaviours that have been found to infringe Article 102 TFEU according to the case-law. Finally, the Priorities Paper fails to serve the principle of legal certainty; if anything, it has led to more confusion than clarity.     

            All in all, the Priorities Paper has no particular legal status. It has no binding legal force, and cannot produce any legal effects either, because it does not serve the enforcement of any of the general principles of EU law. Therefore, neither the Automec judgment nor the Grimaldijudgment may act as authorities for the Priorities Paper to be taken into consideration, since it does not meet the requirements laid down in either of them.

Practical Significance

Despite the fact that the Priorities Paper lacks legal status, it may still be of decisive practical significance. In fact, the tension between its content and the case-law may be irrelevant for the undertakings and their legal advisers when deciding to consult it. This is because it indicates the Commission’s threshold for intervention. From a practical standpoint, if the Commission decides to pursue cases on the basis of a particular test, companies and their legal advisers would immediately attempt to understand it and bring themselves into line with it.

            Nevertheless, the practical significance of the Priorities Paper is obscured for several reasons. First, there are few bright-line rules. The principles are subject to significant exceptions, which are not sufficiently explained. Second, the paper provides no safe-harbours, which in turn reduces the overall level of guidance that it provides. Third, the purpose of guidelines is to enhance transparency and accountability, provide a clarification of the law, ensure consistency of enforcement and increase legal certainty (see e.g. Tréfilunion SA v Commission at para 142). This purpose was disregarded by the Priorities Paper, which is couched in terms of theoretical economics with minimal awareness of the practical consequences for companies. Finally, the Commission itself does not comply with its ‘guidance’ (see the Inteldecision).

Compatibility with the Principle of Loyal Cooperation

The principle of loyal cooperation lies at the heart of the European integration process, and the CJEU has from an early stage recognised it as being a general principle of Union law (see e.g. the ERTAjudgment at para 87). The principle governs the entire Union competence, including competition policy (see e.g. Case C-344/98 Masterfoods at para 56). Moreover, as the post-Lisbon Treaty formulation emphasises, the principle of loyal cooperation has a mutual nature (Article 4(3)(a) TEU), and governs both the relations of the Member States with the EU institutions, and the relations between the various EU institutions (Article 13 TEU).

The adoption of the Priorities Paper, as well as its content, run counter to both these manifestations of the principle of loyal cooperation.

Loyalty Between the EU Institutions

To the extent that the Commission’s new approach to abusive exclusionary conduct is incompatible with the case-law, the Commission disregarded its duty of loyalty toward the CJEU. By derogating from established jurisprudence by means of guidelines, the Commission exceeded the limits of the powers conferred on it by the Treaties. Thus, it did not practise mutual sincere cooperation as required by Article 13(2) TEU. Additionally, the choice of the Commission to name the document ‘enforcement priorities’ aggravates its infringement, in the sense that it was a manœuvre aiming at relaxing the tension between the approach adopted in the Priorities Paper and the one followed in the case-law.

Likewise, the Commission disregarded the principle of sincere cooperation as regards its relations with the Council. Specifically, the Commission’s Priorities Paper functionally amended the framework of Article 102 TFEU in several respects, thus circumventing the Council’s legislative responsibility in the area of competition law (Articles 103(1) in conjunction with 289(2) and (3) and 290(1)(b) TFEU). The Commission requires delegated authority to adopt acts in this area, which may only concern non-essential issues (Article 290(2) TFEU). Otherwise, the Commission acts ultra vires, contrary to Article 17 TEU, which sets out its responsibilities. In this context, the Commission also breached the principle of institutional balance.

Loyalty Between EU Institutions and Member States

The Commission also violated the duty of sincere cooperation in the context of its relations with the Member States. This is so, despite the fact that the Commission states in the Frequently Asked Questions accompanying the initial adoption of the Priorities Paper that the said paper has been discussed extensively with the NCAs (question 9).

This is because of the parallel competence that the Commission, the NCAs and the national courts share as regards the application of Article 102 TFEU (Articles 4-6 of Regulation No 1/2003), which entails that they must apply the relevant rule in close cooperation in order to avoid inconsistent and/or contradictory decisions (Articles 11(1) and 15 of Regulation No 1/2003). As such, in principle, NCAs and national courts must take the Priorities Paper into account (see Grimaldi at paras 18-19; Commission Notice at para 8). Yet, they must respect the EU Courts’ case-law pursuant to the principle of supremacy of EU law. Therefore, NCAs and national courts may take into account the Priorities Paper exclusively to the extent that its approach is in conformity with the CJEU’s jurisprudence. As if that situation were not complicated enough, the hybrid approach that the Commission follows in its post-Priorities Paper decisional practice further perplexes things, since NCAs and national courts cannot take decisions running counter to Commission decisions (see Article 16 and recital 22 of Regulation No 1/2003). This state of uncertainty endangers the uniform application of Article 102 TFEU at the national level.

Compliance with Fundamental Rule-of-Law Principles: Analysis of Rebates as an Example

Even more disturbing is the content of the Enforcement Priorities Paper, which at times does not integrate economic and legal analyses in a manner that would ensure compliance with fundamental rule-of-law principles, such as the legal certainty and the nullum crimen, nulla poena sine lege principles (Article 49 CFREU; Article 7 ECHR). The observance of these principles is of utmost importance in the context of the enforcement of Article 102 TFEU, which is performed through fines that may be skyrocketing, as was the case in Intel.

            The Priorities Paper’s approach to rebates granted by dominant firms offers a prime example of this (see paras 37-45). The ‘as-efficient competitor’ test (‘AECT’) advocated in that paper for the assessment of rebates is the epitome of a purely theoretical economic tool; although it is a sophisticated test which is grounded on robust economics, it cannot have any practical functionality (for a list of objections against this test, see here). In other words, it is a perfect test on paper that makes perfect sense with perfect numbers. The problem, however, is that there is no such thing as perfect numbers.

            For instance, this test requires a dominant firm to estimate ex ante the units that its client would potentially purchase from its competitors. The only possible way to do this is to ask its client, who is very likely to mislead the dominant firm. Similarly, under the test proposed by the Priorities Paper, a dominant firm must evaluate ex antethe pricing and range of products that a competitor has the capacity to produce. This depends on information on rivals’ costs and sales, which the dominant firm cannot be expected to possess. Furthermore, the AECT is assuming a single competitor and a single customer. But what if the dominant firm has, for example, thirty competitors; is the dominant firm required to conduct this complex analysis thirty times for each of its clients? This would be an impossible task. Hence, this test can only operate in retrospect and with access to confidential information. It is an unworkable test for a company that attempts to self-assess its practices.

            It appears that the Commission, in its zeal for injecting more economic analysis into the enforcement of Article 102 TFEU, was oblivious to the absurdity of bestowing a self-standing role for economics in the interpretation of this provision. Indeed, any argument in favour of a more economic approach to abuse of dominance is vulnerable insofar as it is not properly integrated with legal reasoning. In this connection, legal reasoning is, by definition, about categorical thinking, and hence, formal. Only formal rules can ensure that Article 102 TFEU will be enforced properly in light of its goal, namely the maintenance of effective competition within the internal market. Indeed, effective competition can only exist if the players in the market act in an environment where they can assess in advance and at reasonable cost whether their conduct violates Article 102 TFEU.

            All in all, no matter how theoretically sound the economic arguments for using the AECT may be, this test is not appropriate for the assessment of rebates under Article 102 TFEU. This is, inter alia, because the test fails to guarantee the observance of fundamental legal principles.

Conclusion

Overall, the Enforcement Priorities Paper is flawed, both constitutionally and as a matter of substance. Constitutionally, the Commission does not have carte blanche as regards the interpretation of Article 102 TFEU; that is, the Commission is constitutionally incompetent to alter the interpretation given to the law by the CJEU. Thus, the Commission, by adopting the Priorities Paper, acted outside its remit and breached a plethora of general principles of EU law. Moreover, the Priorities Paper’s approach to rebates illustrates that its substantive analysis is incompatible with fundamental rule-of-law principles. In this connection, the selected title cannot remedy or conceal these deficiencies. It seems to me that keeping the Priorities Paper alive would be akin to contempt of the CJEU and would cause unnecessary confusion.

Barnard & Peers: chapter 17 

Photo credit: http://ec.europa.eu/competition/court/antitrust.html