Tuesday, 29 December 2015

The Slovak Challenge to the Asylum-Seekers’ Relocation Decision: A Balancing Act

Zuzana Vikarska, Post-graduate research student, University of Oxford*

*This post is written in the author's academic capacity and does not represent the view of any of her past, present or future employers

On Wednesday 2 December, Slovakia filed an action for annulment to the Court of Justice (pending case C-643/15), challenging the legality of the Asylum-Seekers Relocation Decision (“the contested decision”) adopted on 22 September 2015. Just one day later, on Thursday 3 December, Hungary did the same (pending case C-647/15).
Steve Peers has already discussed some of the (anticipated) legal arguments against the validity of the challenged EU measure in his recent blog-post. In the following post, I briefly discuss the arguments on which the Slovak government’s case rests and I explore whether they are likely to succeed before the Court of Justice. Unfortunately, I do not have access to the full text of the Slovak action, nor to the arguments of the Hungarian government. I am therefore only commenting on the summary of the six arguments, which has been published on the webpage of the Slovak Ministry of Justice.

1. Division of competences and institutional balance
First plea in law, alleging breach of Article 68 TFEU, as well as Article 13(2) TEU and the principle of institutional balance: The Council by adopting the contested decision exceeding the previous guideline of the European Council, being therefore in contradiction to the mandate of the latter, infringed Article 68 TFEU as well as Article 13(2) TEU and the principle of institutional balance.
In its first claim, the Slovak government argues that the measure was adopted in contradiction to the guidelines set by the European Council, an institution made up of heads of state and government (as distinct from the Council, made up of national ministers),  whose role is to “define the general political directions and priorities” of the Union (Article 15 TEU), as well as to “define the strategic guidelines for legislative and operational planning within the area of freedom, security and justice” (Article 68 TFEU). Let us therefore have a look at the “strategic guidelines” determined by the European Council in its most recent meetings.
On 23 April 2015, the European Council stated that there was a need to “consider options for organising emergency relocation between all Member States on a voluntary basis” and to “set up a first voluntary pilot project on resettlement across the EU, offering places to persons qualifying for protection.” Furthermore, at its meeting of 25 and 26 June, the European Council agreed on “the temporary and exceptional relocation over two years from the frontline Member States Italy and Greece to other Member States of 40.000 persons in clear need of international protection, in which all Member States (except the UK) will participate,” as well as “the rapid adoption by the Council of a Decision to this effect; to that end, all Member States will agree by consensus by the end of July on the distribution of such persons, reflecting the specific situations of Member States.” No particular conclusions have been adopted as to the 120.000 further asylum-seekers whose situation forms the subject matter of the contested relocation decision.
a. European Council promising “that no quota would be adopted”?
In his recent explanation of this first plea, the Slovak Prime Minister said that the European Council had allegedly “stated that no quota would be adopted” and that the Council of Ministers had disrespected this agreement. Having read the conclusions of the European Council summarised above, the Prime Minister’s statement seems to be rather inaccurate. The Council has not acted contrary to the European Council’s conclusions, which makes the first part of the first claim moot.
Hypothetically, however, if there truly was a clear conflict between the conclusions of the European Council and the actions of the EU lawmakers (the Commission, the Parliament, and the Council), such situation could indeed raise some interesting questions concerning the institutional balance in the Union. What if the European Council had expressed a clear consensus that relocation of asylum-seekers was an undesirable, or even unacceptable way of addressing the current crisis, and despite such consensus, the Commission would subsequently propose a relocation measure (like the one adopted on 22 September), which would be passed by the Council and the Parliament?
It is necessary to realise that the conclusions of the European Council are endowed by political, rather than legal significance (see Article 15(1) TEU). That being said, if the Commission acted in direct opposition to the “general political directions” (Art 15 TEU) and “strategic guidelines” (Art 68 TFEU) adopted by the European Council, such action could violate the principle of institutional balance, since it would deprive the European Council’s decisions and conclusions of any effect. This shows that although the European Council’s conclusions are not legally binding, they should be endowed with political significance, therefore requiring the Commission either to act in line with them, or to provide an elaborate and politically acceptable explanation of any other action.
Still, this analysis remains hypothetical, since the quotations above suggest that the draft of the contested decision did not disrespect any of the political directions outlined by the European Council in the past couple of months.
b. European Council requiring that the vote in the Council be unanimous?
The second interesting question concerns the European Council’s wish that the relocation decision be reached by consensus, as expressed in the conclusions of 25 and 26 June. It is important to point out that this wish only concerned the first 40,000 asylum-seekers, and therefore remains irrelevant for the validity of the contested decision. That makes the second part of the first argument moot as well. However, let us see (once again, hypothetically) whether the European Council could, by its (political) decision, require a different threshold for adopting a decision in the Council, in contrast with the (legal) threshold required by the Treaties.
What happens if the procedure envisaged by Article 78(3) TFEU requires a qualified majority vote in the Council (which it does), while the European Council imposes a requirement of a unanimous vote? While the Treaties remain silent about a change of procedures from QMV to unanimity, they do include a provision governing a move in the opposite direction: from unanimity to QMV. Pursuant to Article 48(7) TEU, “where [TFEU] or Title V of [TEU] provides for the Council to act by unanimity in a given area or case, the European Council may adopt a decision authorising the Council to act by a qualified majority in that area or in that case.” A change from unanimity to QMV is therefore possible, but it comes with significant procedural guarantees, such as obtaining the consent of the European Parliament and giving national parliaments 6 months to block such action. It follows that changes of Treaty voting mechanisms require more than just a unanimous decision of the European Council. Furthermore, while a change from unanimity to QMV is at least foreseen by the Treaties, a change in the opposite direction is not. That leads to a conclusion that such an instruction given by the European Council would only have political significance, but would remain legally irrelevant.
Therefore, even if the requirement of a unanimous vote extended to all legal measures adopted under Article 78(3) TFEU, quod non, I hold the opinion that such an agreement would remain a gentlemen’s agreement, rather than a legally binding requirement. Therefore, the second part of the first plea put forward by the Slovak government does not seem to offer a good reason for the annulment of the contested decision, either.

2. Legislative vs. non-legislative acts
Second plea in law, alleging breach of Article 10(1 and 2) TEU, Article 13(2) TEU, Article 78(3) TFEU, Article 3 and 4 of the Protocol (No.1) and Article 6 and 7 of the Protocol (No.2), as well as the principles of legal certainty, representative democracy and institutional balance: Such kind of act as the contested decision cannot be adopted on the basis of Article 78(3) TFEU. Regarding its content, the contested decision is in fact of a legislative character and should therefore be adopted by legislative procedure, which, however, is not foreseen in Article 78(3) TFEU. By adopting the contested decision on the basis of Article 78(3) TFEU, the Council not only breached the latter, but it also interfered with the rights of national parliaments and the European parliament.
This second plea opens a very interesting question of the nature of “legislative acts” in EU law: what does a measure need in order to be “of a legislative character”? The Slovak government seems to invoke a material understanding of a legislative act, suggesting that some things are too important to be governed by a non-legislative act. This, however, is not how EU law seems to work. The concept of a legislative act seems to be a formal one, as follows from Article 289(3) TFEU: “Legal acts adopted by legislative procedure shall constitute legislative acts.
Although EU law surely doesn’t work with a material concept of “legislative acts”, it remains unclear what a legislative act really is. Two formal interpretations are still feasible: a narrow textual one and a procedural one.
According to a narrow textual interpretation, favoured e.g. by Lenaerts, by Craig & De Burca, as well as by Advocate-General Kokott in her opinion in C-583/11 Inuit Tapiriit Kanatami, the (non-) legislative nature of an act depends on one and only factor: whether its legal basis refers to a “legislative procedure” in its wording or not. Pursuant to this interpretation, an asylum measure adopted under Article 78(3) TFEU will be a non-legislative act, while a measure on diplomatic protection adopted under Article 23(2) TFEU will be a legislative act, although the procedure of their adoption is exactly the same: a proposal from the Commission and a qualified majority vote by the Council, after consulting the European Parliament. This interpretation seems to be supported by Article 289(2) TFEU, which states that special legislative procedure is only used “in the specific cases provided for by the Treaties,” which simply isn’t the case for measures adopted under Article 78(3) TFEU. Another argument for this narrow textual interpretation is the practical functioning of the EU institutions: Eur-Lex categorises the contested decision as an “NLE”, which stands for “non-legislative procedure”. On a practical level, this textual interpretation thus seems to be a well-established one.
The third thinkable alternative is a procedural interpretation (that seems to be preferred e.g. by Chalmers), according to which an act is to be labelled as “legislative” if both the Council and the Parliament are involved in its adoption, no matter whether the specific legal basis speaks of a “legislative procedure” or not. There are in fact a number of solid arguments for this broader interpretation of legislative acts. Firstly, Article 289(3) TFEU (technically) does not say anything about the nature of acts that are not adopted by a legislative procedure. In the terminology of formal logic, Article 289(3) is an implication; therefore, turning it around (and stating that “legal acts not adopted by legislative procedure shall constitute non-legislative acts”) would be a logical flaw. Secondly, the endorsement of the textual interpretation would lead to a situation where entire areas of EU law (such as competition law!) would not contain any legislative measures whatsoever, which is rather curious. Thirdly, the textual interpretation would have vast consequences for the procedural standing of non-privileged applicants under Article 263(4) TFEU who would be able to challenge a much broader category of acts without having to establish individual concern. Fourthly, if we were to link the legislative nature of a measure with its democratic legitimation, the procedural interpretation would make much more sense than the textual one. Yet, despite all these arguments (and despite the hopes of some academics, as well as practitioners), it seems that (at least at present), the narrow textual interpretation of “legislative acts” seems to prevail.
Be that as it may, both the narrow textual interpretation and the procedural interpretation are formal in their nature. It makes no sense to claim that the measure at stake is “of a legislative character” and that Article 78(3) TFEU is therefore not a correct legal basis for such measure, if it cannot lead to the adoption of a legislative act. The conditions of the said article have been fulfilled both materially (emergency situation, sudden inflow of third country nationals, benefit to the Member States concerned) and procedurally (proposal from the Commission, consulting the Parliament, QMV in the Council). Therefore, the second plea of the Slovak government seems to be unfounded as well.

3. Procedural conditions in case the contested decision is indeed a legislative act
Third plea in law, alleging breach of essential procedural requirements governing the legislative procedure, established in Article 16(8) TEU, Article 15(2) TFEU,  Article 78(3) TFEU, Article 4 of the Protocol (No.1) and  Article 6 and 7(1 and 2) of the Protocol (No.2), as well as Article 10(1 and 2) TEU, Article 13(2) TEU and the principles of representative democracy, institutional balance and sound administration: If the Court of Justice contrary to the submissions of the Slovak Republic within the second plea in law came to the conclusion that the contested decision was adopted by legislative procedure (quod non), the Slovak Republic in the alternative alleges the breach of essential procedural requirements, established in Article 16(8) TEU, Article 15(2) TFEU,  Article 78(3) TFEU, Article 3 and 4 of the Protocol (No.1) and  Article 6 and 7 (1 and 2) of the Protocol (No.2), as well as Article 10(1 and 2) TEU, Article 13(2) TEU and the principles of representative democracy, institutional balance and sound administration. In particular, the requirement of public discussion and voting within the Council was not respected, the participation of national parliaments in the process of adopting the contested decision was limited and the requirement of consultation of the European parliament was breached.
This is where the distinction between a textual and a procedural reading of “legislative acts” becomes relevant. If the Court endorses the narrow textual reading of the concept (which is, in my opinion, more probable), then the classification of the contested decision as a non-legislative act will be confirmed and the procedural guarantees invoked by the Slovak government will be inapplicable. Yet, should the Court wish to reassess the notion of “legislative acts” and change it to a procedural one, then the contested decision will have to be seen as a legislative act that has not fulfilled the requirements listed above, namely public voting in the Council (Article 16(8) TEU) and participation of national parliaments (Protocols 1 and 2), which would lead to its annulment.

4. Repeated consultation by the European Parliament
Fourth plea in law, alleging breach of essential procedural requirements, established in Article 78(3) TFEU and Article 293 TFEU, as well as Article 10(1 and 2) TEU, Article 13(2) TEU and the principles of representative democracy, institutional balance and sound administration: Before adopting the contested decision, the Council substantially amended the proposal of the Commission in several ways. When doing that, the Council breached essential procedural requirements, established in Article 78(3) TFEU and Article 293 TFEU, as well as Article 10(1 and 2) TEU, Article 13(2) TEU and the principles of representative democracy, institutional balance and sound administration. It is because the European parliament was not properly consulted and the Council did not decide on the amendments of the Commission’s proposal unanimously.
This fourth point has already been addressed by Steve Peers in his recent post and I agree that this indeed seems to be the strongest procedural argument against the validity of the contested decision. The original Commission draft concerned not just Italy and Greece, but also Hungary, which in the final version of the contested decision refused to be included in the group of external border States. As argued by Steve, this is a change of an essential element, which probably should have been subjected to a repeated consultation by the European Parliament. Possibly, the emergency nature of the contested decision could serve as an argument against an obligation to re-consult, which is, admittedly, a rather weak argument. However, as Steve has already argued, even if this argument were to succeed, it would only lead to a procedural redress (meaning that the outcome would be the same even if the European Parliament were to be re-consulted) while the contested decision would probably remain in force in the meantime. It therefore seems that the fourth plea will not suffice either to have the contested decision annulled.

5. The ‘provisional’ nature of the contested decision
Fifth plea in law, alleging breach of Article 78(3) TFEU by not fulfilling the conditions for its applicability: In the alternative to the second plea in law, the Slovak Republic argues that there was a breach of Article 78(3) TFEU because the conditions for its applicability, which concern the provisional character of the measures adopted, as well as the emergency situation caused by a sudden inflow of nationals of third countries, were not fulfilled.
The fifth plea can be seen from two points of view again: on the one hand, “the provisional character of the measures” can be assessed from a formal viewpoint, looking at the time period for which the measure remains applicable. On the other hand, the plea invites the Court to assess whether the EU finds itself in an “emergency situation”, facing a “sudden inflow” of migrants, which opens up a more substantive analysis of the contested measure.
As for the more formal part of the argument, the provisional character of the measures could be derived from the fact that the contested decision is limited in time - it only applies until 26 September 2017. Is a provision that remains applicable for two years “provisional” enough? The pre-Lisbon version of this provision was Article 64(2) TEC, which limited the duration of such provisional measures to six months. Compared to this past requirement, two years seem to be quite long. On the other hand, the current wording of the Treaties imposes no time limit whatsoever, so two years cannot be seen as a priori illegal. (Yet, this will be very similar to the data retention case: if 6 months are fine and 2 years are too much, how do we feel about one year? The criteria for assessment are not clear.) Still, the contested decision’s provisional character could also be derived from the fact that it only applies to those asylum-seekers who are already present on the European Union territory or who are likely to come in the very near future, while a long-term, sustainable solution is being sought by a legislative proposal, which is currently in the legislative process and which will presumably take quite some time.
As for the more substantive analysis, it is rather difficult to predict what criteria will be chosen by the Court to assess whether the EU really finds itself in an emergency situation. The current migrant crisis is undoubtedly one of the most serious crises that the EU has faced in the past couple of decades, but it remains to be seen how the Court tackles the interpretation of the notions in Article 78(3) TFEU.

6. Principle of proportionality
Sixth plea in law, alleging breach of the principle of proportionality: The contested decision is manifestly incompatible with the principle of proportionality, as it is manifestly neither suitable nor necessary to achieve the desired end.
The principle of proportionality can certainly serve as a very strong argument – it can open the door for a thorough assessment of the very core of any measure by the Court of Justice. Obviously, this assessment is much more likely to be political than legal, with the Court enjoying very wide discretion in balancing the values at stake. It is a huge pity that the Slovak government has not worked out this sixth plea in more detail, since any chance of success of this action is potentially hidden in this last argument. The action could have put forward an argument that the contested measure is not suitable to reach the desired aim (i.e. to relieve the burden borne by the external border states and to show “solidarity and fair sharing of responsibility between the Member States,” as outlined in the decision’s preamble), since relocating people is too difficult and their further movement is too unpredictable. Furthermore, under the heading of necessity, it could have been argued that a less restrictive measure could have been adopted in order to solve the problem, such as denying entry to migrants (although that could violate the basic principles of EU law, particularly EU asylum law), triggering the regime of the Temporary Protection Directive, or other forms of help (e.g. financial, material or personal help) to the affected states. Last but not least, the proportionality argument could have been coupled with other quasi-political arguments, such as a claim under the national identity clause in Article 4(2) TEU.
It can be concluded that the Slovak government has missed a great chance to actually make a point under the sixth plea, since a mere claim of manifest incompatibility with the principle of proportionality does not give the Court anything to build on. That being said, if the Court wants to annul the measure, it can always use this plea to build a strong argumentation under the principle of proportionality.

Based on the analysis above, it seems to me that the first three pleas have almost no chance of success (unless the Court decides to reinvent a new understanding of “legislative acts”), the fourth and the fifth pleas raise interesting issues that could lead to minor interferences, such as the need to re-consult the Parliament, a limitation of the contested decision’s application to a shorter period of time, or a very narrow interpretation of the concepts of “emergency situation” or “sudden influx of migrants”. The sixth plea, however, opens Pandora’s box and invites the Court to exercise broad discretion in its assessment of the current solution to the migration crisis. While proportionality leaves most scope for the creativity of the Court, it is hard to see it replacing its own judgment for the judgment of the Member States, especially in a situation where good, plausible alternatives are quite hard to find.

Barnard & Peers: chapter 3, chapter 5, chapter 26
JHA4: chapter I.5
Image credit: news.yahoo.com

Wednesday, 23 December 2015

Protecting Health, or Protecting Trade? A fine balance in the Scotch Whisky Association judgment

Angus MacCulloch, Lancaster University Law School 

The Court of Justice has now delivered its judgment in Case C-333/14 in relation to the lawfulness of the Scottish measure to introduce minimum alcohol pricing, or MUP for short. Both the Scottish Government and the Scotch Whisky Association, which brought the legal challenge, have “welcomed the ruling,” although I think that the SWA are probably a little happier than the Scottish Government as the case returns to the Inner House of the Court of Session, which had referred it to the CJEU. I’ve previously written about the AG’s Opinion and the Court has adopted a very similar approach, but in many ways the judgment leaves as many questions as it answers. It does appear to give quite a strong steer to the Court of Session that the CJEU would prefer the adoption of the “less restrictive” increase in general excise duties instead of the MUP, but it leaves the final decision on the proportionality of MUP to the Scottish court.

Is MUP caught by Art 34 TFEU?

Both parties to the dispute had accepted that MUP pricing was caught by Art 34 TFEU (the ban on measures with an equivalent effect to quantitative restrictions on imports), but there was little clarity as to how such a measure breached the prohibition. That at least has been clarified today. The Court followed the AG’s elegant solution of evading the complications of categorising a MUP as a “selling arrangement” and dealing with the matter under the Gourmet International style analysis, but rather preferring to use the Trailers “market access” test. A minimum pricing measure restricts access to the UK market as it prevents lower cost products from other Member States from exploiting that cost advantage in lower retail prices [32]. As the removal of the benefits of the cost advantage triggers the market access test there is no need to discuss whether there is any discrimination inherent within the scheme. This is another example of the Court preferring the flexibility of the new test to the more traditional Cassis and Keck line of decisions.

The Tricky Balancing Act in Proportionality

The majority of the ruling deals with the much more difficult question of the potential justification of the measure on health grounds and whether the restriction is proportionate. At first instance the Outer House of the Court of Session accepted that the measure was proportionate as it targeted ‘harmful and hazardous’ drinkers who tended to consume low price high alcohol products which were most effected by MUP, but in the CJEU ruling there is a different view taken as to the purpose of the measure. On the evidence presented to it the CJEU takes the view that MUP has a “twofold objective” [34], both targeting these “harmful and hazardous” drinkers, while also reducing general alcohol consumption in the wider population “albeit only secondarily”. It is this “ambiguity”, as the AG put it, which I think is at the heart of the problem in the Ruling. If one cannot clearly define what a measure is designed to achieve it is incredibly hard to come to a firm conclusion as to whether it is proportionate. The Court did accept, at [38], that the measure was a real attempt by the Scottish Government to address health problems within Scotland, but set out that it cannot go beyond what is necessary in order to protect health. The choice before the CJEU was between the Scottish Government’s preference for MUP, and the argument that the same health benefit could be obtained through an increase in the general excise duties applied to all alcohol products, as preferred by the SWA and the European Commission. The Court argued that increased taxation could be an effective heath protection measure, as it is in relation to tobacco, and that an increase in taxation:

is liable to be less restrictive of trade in those products within the European Union than a measure imposing an MPU. The reason is … that the latter measure, unlike increased taxation of those products, significantly restricts the freedom of economic operators to determine their retail selling prices and, consequently, constitutes a serious obstacle to access to the United Kingdom market of alcoholic drinks lawfully marketed in Member States other than the United Kingdom and to the operation of fair competition in that market.”
The contention that an increase in taxation would be less restrictive of trade, in comparison to MUP, is one of which I have never been convinced. Taxation affects all products, and MUP would only affect a limited number; on that simple basis I contend that MUP is arguably less restrictive in terms of the volume of trade impacted by the measure. Volume of trade affected has been seen as important in other Art 34 cases, see for example the Sunday Trading litigation of the 80s, but here the Court refers to this issue much more explicitly than before. It is not concerned with reducing the volume of trade impacted, but is much more concerned that the measure does not impact “fair competition” within the market; even if a greater number of products are affected. The Court refers to an argument made by the Lord Advocate questioning the relevance of the Court’s previous cases that dealing with minimum pricing in tobacco markets. The Court rejects that position, at [45], but I am nervous about simply reading across from those cases. Those cases centred on the Tobacco Harmonisation Directives, which were explicitly designed to enhance the single market integration by using price competition as a driver of integration. The direct protection of retail price competition is not usually seen so explicitly under Art 34 TFEU. It appears that the Court is now reading the protection of price competition into the prohibition. There is also, to my mind, another important distinction between the health problems associated with tobacco consumption and the health problems associated with alcohol - different problems will require different solutions. 

The final issue in the proportionality discussion relates to the vexed question of choosing the least restrictive of the two measures, and the intrinsically connected question of the balance between restrictiveness of a measure and its effectiveness at achieving its aim. Here we return to the “ambiguity” of the purpose of MUP. The Courts states, at [47]:

the fact that increased taxation of alcoholic drinks entails a generalised increase in the prices of those drinks, affecting both drinkers whose consumption of alcohol is moderate and those whose consumption is hazardous or harmful, does not appear, in the light of the twofold objective pursued by the national legislation at issue in the main proceedings … to lead to the conclusion that such increased taxation is less effective than the measure chosen”.
The Court appears to suggest that as taxation can achieve both the general and the specific aim it is as effective. I find that difficult to follow. One of the main reasons that MUP was adopted was it was targeted, in that it only impacted on cheap and strong products and would not have a wider impact on moderate drinkers or on-sales, which would generally be above the MUP floor. The Court is expressing a preference for the secondary aim of the measure and effectively side-lining its primary purpose. It describes this generalised impact as “additional benefits” [48], but I would argue this is not additional in any valuable sense if it removes the primary benefit, targeting, from the measure. The Court goes on to the usual statement that the final decision is, of course, for the referring court, once it has heard all the evidence and argument, but it is pretty clear where its preference lies. This preference for one aim over another does not sit well with the settled position, repeated at [35], that the Member State can decide on the degree of protection it requires.

On the Article 36 TFEU Derogation

The previous discussion was in relation to the ‘rule of reason’ within the Art 34 TFEU prohibition, but as health protection is one of the grounds for derogation in Art 36 TFEU it is also possible to justify MUP on that basis. The Court discusses Art 36 separately and while the questions are similar the Court appears to adopt a slightly more relaxed tone. It stresses the same proportionality test as above, and that it is the Member State’s responsibility to prevent the appropriate evidence, but also that:
that burden of proof cannot extend to creating the requirement that, where the competent national authorities adopt national legislation imposing a measure such as the MPU, they must prove, positively, that no other conceivable measure could enable the legitimate objective pursued to be attained under the same conditions”.
This appears to give some succour to the Scottish Government that the ball is now in their court, and that they must present the best evidence they can to convince the Court of Session. The alcohol policy evidence, including the Nuffield Report published yesterday, tends to suggest that there is a good case to be made for MUP. In that sense there is a still a lot for both sides to play for when the Court of Session comes back to this issue in 2016.


It is unfortunate that the Court has followed the reasoning of the AG and the weaknesses that it exhibited. We now have confirmation that price competition receives protection under Article 34 TFEU, and any attempt by Member States to interfere with the free setting of prices is likely to be scrutinised as a matter of EU law. The most disappointing aspect of the ruling is the lack of clarity in the Court’s discussion of proportionality, it has been described as “Delphic” by some commentators. I have explained some of my concerns, but the most troubling aspect is the Court’s apparent willingness to suggest that the Scottish Parliament picked the “wrong” health aim, and use proportionality analysis to “correct” that mistake. The Inner House of the Court of Session still has a lot of work to do in unpicking the Court’s Ruling.

Barnard & Peers: chapter 12

Tuesday, 22 December 2015

Standing up for children? The Directive on procedural safeguards for children suspected or accused in criminal proceedings

If we don't stand up for children, then we don't stand for much.
Marian Wright Edelman

Debbie Sayers, Legal Research Consultant, http://interalia.org.uk

Effective human rights protection is fundamental to any concept of fairness in the criminal justice system. Fairness, however, is relative: it may require different levels of protection in different circumstances.

Children require special measures of protection to take account of their particular vulnerability and needs (UN CRC Committee, General Comment 10, para 10). International standards confirm state obligations in this regard (e.g. UN Convention on the Rights of the Child (CRC), UN Standard Minimum Rules for the Administration of Juvenile Justice, the Council of Europe Guidelines of the Committee of Ministers of the Council of Europe on child-friendly justice.). The need for extra protection has also been confirmed by the ECtHR which has stated that the right to a fair trial under Article 6 requires that: “a child charged with an offence is dealt with in a manner which takes full account of his age/level of maturity and intellectual and emotional capacities and that steps are taken to promote his ability to understand and participate in the proceeding”(T v. UK, No. 24724/94, 16 December 1999, at [84]).

The Commission states that approximately 1 million children face criminal justice proceedings in the EU each year (around 12% of the total) (Commission Staff Working Document 2013). It has gathered data on child justice and its reports shows wide variability in practice and procedure between States. The EU has now agreed the text of a Directive to establish specific procedural safeguards for child suspects. This is the fifth in a series of six EU-specific standards, all in the form of Directives, which have been agreed under a Roadmap for strengthening the procedural rights in criminal proceedings (on the fourth measure, on presumption of innocence, see discussion here; on the sixth proposal, on legal aid, see discussion here). The Directives attempt to promote consistency in procedural protection within the criminal justice systems of EU Member States. Measure E of the Roadmap requires special safeguards to be created for vulnerable suspects. A Recommendation setting out procedural safeguards for vulnerable persons has already been published. This post deals with the recently agreed Directive on child suspects.

The Directive

The Directive is a lengthy and rather complex document with 39 Recitals and 25 operative Articles. The text has now been agreed subject to reservations by Romania (in relation Article 2(3) and Article 9 (1)) and by Poland (in relation to Article 6(8) (b)). Ireland, the UK and Denmark are not participating in the instrument. Member States will have three years to transpose the Directive from the date of its entry into force (Article 23). [Update; the Directive was officially adopted in April 2016].

The Directive’s purpose is “to establish procedural safeguards to ensure that children who are suspected or accused in criminal proceedings are able to understand and follow those proceedings, to enable such children to exercise their right to a fair trial and to prevent re-offending by children and foster their social integration” (Recital 1). Children already benefit from all the human rights guarantees available to adults but the Directive notes that “experience has shown that this in itself does not always provide a sufficient degree of trust in the criminal justice systems of other Member States” (Recital 3). The ECHR remains the baseline for any assessment of the efficacy of new standards. Recital 35 notes that a “higher level of protection should not constitute an obstacle to the mutual recognition of judicial decisions that those minimum rules are designed to facilitate. The level of protection should never fall below the standards provided by the Charter or by the ECHR, as interpreted in the case law of the Court of Justice of the European Union and the European Court of Human Rights.”

Legislation which promotes children’s rights is welcome. However, to be of genuine value, a separate document for child suspects should add to and build on existing standards in a child specific way. It must also address the obstacles which prevent current standards working: it must prioritise the best interests of the child and remove any vagueness and discretion which may facilitate non-compliance. Fundamentally, it must also be effectively implemented in practice. This post will consider some of the key elements of the Directive. It will conclude that the picture is not entirely rosy.


Article 3 defines a “child” as “a person below the age of 18 years”.  The relevant point at which age is assessed is the age at which the individual becomes subject to proceedings (Recitals 8 and 9) rather than the age at the time of the offence. In the case of uncertainty, there is a presumption that the person is a child. Under Article 2(3), the Directive also applies where a child comes of age during proceedings but only where the application of the Directive, or some of its provisions, “is appropriate in the light of all the circumstances of the case, including the maturity and vulnerability of the person concerned”. Even then. however, Member States may decide that this Directive does not apply when the person reaches 21. The Directive simply “encourages” Member States to apply the procedural safeguards where the person is older than 18 and below 21. The UN Committee on the CRC has stated in General Comment 10 that “every person under the age of 18 years at the time of the alleged commission of an offence must be treated in accordance with the rules of juvenile justice”. It is very disappointing that the Directive fails to follow this approach.

The Directive lays down “minimum rules” for children in criminal and in European Arrest Warrant (EAW) proceedings (Article 1). It does not affect national rules determining the age of criminal responsibility (Article 2(5)). It applies from the time children are suspected or accused in criminal proceedings until the final determination of the case including sentencing and appeal (Article 2 (1)).  The Directive also applies to children who are not suspects or accused persons but who, in the course of questioning by the police or by another law enforcement authority, become so (Article 2(4)). This mirrors Article 2(3) of the Directive on the right of access to a lawyer.

Article 17 confirms that the rights set out in Articles 4, 5, 6, 8, 10, 11, 12, 13, 14, 15 and 18 of the Directive also apply to EAW proceedings from the time of arrest in the executing Member State (Article 17). It should also be remembered that, although the Directive seeks to give children rights in EAW proceedings, Article 3(3) of the Framework Decision on the EAW actually provides a ground for mandatory non-execution of the EAW where a person, “owing to his age” cannot be held criminally responsible for the acts on which the arrest warrant is based under the law of the executing State.

Another concern is that the Directive restricts its application in respect of minor offences (Recitals 11a-11c). Article 2(5a) confirms that, in respect of minor offences, where Member States’ law provides for a sanction by to be imposed by “an authority other than a court having jurisdiction in criminal matters” and the imposition of such a sanction may be appealed or referred to such a court or where deprivation of liberty cannot be imposed as a sanction, the Directive only applies to the proceedings before a court having jurisdiction in criminal matters. A ‘minor’ offence is not defined and there is no penalty threshold. In C-60/12 Baláž, 14 November 2013 confirmed that ‘having jurisdiction in criminal matters’ means that a court “must apply a procedure which satisfies the essential characteristics of criminal procedure, without, however, it being necessary for that court to have jurisdiction in criminal matters alone” (at [36]). This is “an autonomous concept of Union law” (at [42]).

It is disappointing that the Directive is only guaranteed to “fully apply” where a child is deprived of liberty irrespective of the stage of the criminal proceedings. The restrictions on minor offences mirror those in other Directives (Article 1 (3) of the Directive on the Right to Interpretation and Translation, Article 2(2) of the Directive of the Right to Information, and Article 2 (4) of the Directive on the Right of Access to a Lawyer). But there is no justification for the imposition of this clause and no explanation of how it sits with current ECHR obligations. In terms of the exemption for ‘minor offences’, the ECtHR does not distinguish between different types of offences: the protection afforded by Article 6 is engaged as soon as a ‘criminal charge’ is brought against an individual. Engel and Others v The Netherlands No. 5100/71, 8 June 1976 confirms that, when determining whether a ‘criminal charge’ exists, specific criteria will be considered: (a) the classification of the offence under the domestic legal system; (b) the nature of the offence; and (c) the potential nature and severity of the penalty. Even a seemingly minor offence may have serious consequences for the child. It is disappointing to apply this unnecessarily restrictive approach to a document designed to protect vulnerable suspects.


The Directive sets out a series of rights.

Right to information: Article 4 creates a duty to inform child suspects “promptly” about their rights in Directive of the Right to Information and about general aspects of the conduct of the proceedings. The Letter of Rights required by Directive of the Right to Information should also set out a child’s rights under this Directive (Article 4(2)).

Article 4 distinguishes between those rights in the Directive which children are to be informed about “promptly” and those they are to be informed about at the at the “earliest appropriate stage in the proceedings”. There is no justification for such a distinction. Further, worryingly, under Article 4 (1a), this information can be given orally rather than in writing which undermines the protection. There is no requirement that either an appropriate adult or lawyer is present when this information is handed over or that the information is given in a child-friendly form (the Directive merely talks of “simple and accessible language”). The  Council of Europe Guidelines on child friendly justice require information and advice to be provided in a manner adapted to their age and maturity, in a language which they can understand and which is gender- and culture-sensitive. This is particularly important when the age of criminal responsibility varies wildly within the EU (e.g. it is 10 in England and 15 in Denmark).  Handing information to a child is no guarantee they will be able to understand or use it and a process to verify the child’s comprehension should have been established.

Right of the child to have the holder of parental responsibility informed: Article 3 defines the “holder of parental responsibility” (PR) to mean “any person having parental responsibility over a child”. Parental responsibility means “all rights and duties relating to the person or the property of a child which are given to a natural or legal person by judgment, by operation of law or by an agreement having legal effect. The term includes rights of custody and rights of access.”

Under Article 5, the state must ensure that the holder of parental responsibility is provided “as soon as possible” with the information that the child receives in accordance with Article 4 (see also Recitals 15 and 15a). However, Article 5(2), provides specific grounds for using another appropriate adult, albeit one nominated by the child and accepted as such by the competent authority. The grounds include:  if (a) it would be contrary to the best interests of the child to provide it to the holder of PR or (b) no holder of parental responsibility can be reached after reasonable efforts have been made to do so or his or her identity is unknown or (c) because, on the basis of objective and factual circumstances, providing information to the holder of PR could substantially jeopardise the criminal proceedings to provide it to the holder of PR.

Further, if the child has not nominated another appropriate adult, or if this nominated adult is not acceptable to the competent authority, “taking into account the best interests of the child”, the competent authorities may provide the information to another person, including someone from an authority or responsible for the protection or welfare of children. As soon as any of these grounds cease to exist, the information should be provided to the holder of PR. Provisions which too readily allow the state to appoint the adult supporting the child are of concern.

Right to assistance by a lawyer: The ECtHR has confirmed that, in order for the right to a fair trial to remain “practical and effective”, access to a lawyer should be provided from the first police interrogation (Salduz v. Turkey, No. 36391/02, 27 November 2008). Suspects are particularly vulnerable at the investigation stage and evidence gathered may determine the outcome of the case. The right of access to legal assistance is particularly important for vulnerable suspect such as minors (see for example, S.C. v. the United Kingdom, No. 60958/00, 15 June 2004).

Article 6 re-affirms the right of access to a lawyer. It also confirms that the right is about exercising the “rights of defence effectively” (Article 1a.). Under Article 6(2), this assistance should occur “without undue delay once [Member States] are made aware that they are suspected or accused of having committed a criminal offence” from whatever point is earliest: (a) before they are questioned by the police or by another law enforcement or judicial authority; (b) upon the carrying out by investigating or other competent authorities of an investigative or other evidence-gathering act in accordance with point (c) of paragraph 3; (c) without undue delay after deprivation of liberty; (d) where they have been summoned to appear before a court having jurisdiction in criminal matters, in due time before they appear before that court.

Under Article 6(3) (see also Recital 16a), assistance “includes” (so is not limited to): the right to meet in private and communicate with the lawyer representing them. The confidentiality of communications between the child and their lawyer is protected under Article 6(4) (see also Recitals 17c and 17d); the right to be assisted by a lawyer when they are questioned, and that “the lawyer can participate effectively during questioning” (as with other Directives this participation is “shall be in accordance with procedures under national law); and the right to be assisted by a lawyer during investigative or evidence-gathering act such as: (i) identity parades; (ii) confrontations; (iii) reconstructions of the scene of a crime.

However, Recital 16b also sets out the circumstances in which the right to a lawyer does not arise, e.g. including verifying the child’s identity, whether they have a weapon or taking fingerprints or photographs. These acts potentially carry significant consequences and it is unhelpful to have limited the scope in this way.

Further, despite the importance of this right, a series of derogations and discretionary provisions potentially undercut the protection. For example, there is a proportionality clause in Article 6 (5) (see also Recital 17) which permits Member States, provided that this is in conformity with the right to a fair trial, to derogate from the obligations set out in Article 6(2) where assistance by a lawyer “is not proportionate in the light of the circumstances of the case, taking into account the seriousness of the offence, the complexity of the case and the measures that could be  taken in respect of the alleged offence, it being understood that the best interests of the child shall always be a primary consideration”. Further, under Article 6(8), “in exceptional circumstances and only at the pre-trial stage”, Member States may temporarily derogate from the duty to provide the assistance of a lawyer “to the extent justified in the light of the particular circumstances of the case” on one of the following “compelling reasons”: (a) where there is an urgent need to avert serious adverse consequences for the life, liberty or physical integrity of a person; (b) where immediate action by the investigating authorities is imperative to prevent substantial jeopardy to criminal proceedings in relation to a serious offence. In applying this paragraph “the best interests of the child” must be taken into account. This can only be decided on “a case-by-case basis” either by a judicial authority or other competent authority on condition that the decision can be submitted to judicial review.

Finally, whereas the Directive on the right to a lawyer permits suspects or accused persons to waive their right, the original Commission proposal prohibited waiver in the case of children. It is disappointing this has been dropped. Ensuring every child gets legal representation would have been an important step forward in ensuring rights are protected.

Right to an individual assessment: Article 7 requires Member States to assess children individually to ensure that their specific needs concerning protection, education, training and social integration are taken into account. The costs are generally to be met by the state (Article 21).  The assessment shall “take into account the personality and maturity of the child, their economic, social and family background, as well as any specific vulnerabilities of the child” (Article 7(2) and Recital 19a). The nature and extent of the assessment will depend on the circumstances of the case (Article 7 (2a)). The assessment should “be carried out with close involvement of the child” and be conducted by “qualified personnel, following, as far as possible, a multidisciplinary approach and involving, where appropriate, the holder of parental responsibility or another appropriate adult and/or specialist professional” (Article 7 (5)). If circumstances change, an assessment may be updated (Article 7(6)).

The individual assessment, which should take place at the earliest appropriate stage of the proceedings and, before indictment (Article 7(4a)), will note information which might be of use by the competent authorities when making decisions about the child.  However, the absence of an early individual assessment will not stop an indictment being presented if this “is in the best interests of the child” provided that the individual assessment is available at the beginning of the trial (Article 7(4b) and Recital 19d). Additionally, derogation from this provision is possible if “warranted by the circumstances of the case, and provided that this is compatible with the child's best interests” (Article 7 (7) and Recital 19e). Again, the creation of rule accompanied by exceptions may compromise consistency.

Right to a medical examination: Children who are deprived of liberty have the right to a medical examination without undue delay (Article 8 and Recital 20). One concern here is that the Directive uses the term ‘undue delay’ rather than requiring the examination to be done promptly (compare this with that set out in in General Comment 10). The examination can be required by the competent authorities or by (a) the child, (b) the holder of PR responsibility or another appropriate adult as referred to in Article 5; or (c) the child’s lawyer (Article 8(2). The costs are generally to be met by the state (Article 21).

The examination is to assess the general mental and physical health of the child. The examination should be as non-invasive as possible and carried out by a physician or another qualified professional. It will be recorded in writing (Article 8 (3)) and may lead to medical assistance or may help determine the capacity of the child to face questioning, other investigative or evidence gathering acts or any measures taken or envisaged against the child (Article 8(1a)). It can be repeated where required (Article 8(4)).
The Directive should have guaranteed a right of access to medical treatment if directed by a doctor.

The restriction of this protection to those who are deprived of their liberty is a concern as medical examinations may be important to any child facing criminal proceedings. Deprivation of liberty is not defined but presumably is intended to relate solely to those detained in prison. In the juvenile justice system, children may also be obliged to reside with foster carers or have their liberty restricted in other very significant ways.  The ECtHR has said that the “difference between deprivation of and restriction upon liberty is … merely one of degree or intensity, and not one of nature or substance” Guzzardi v Italy. No. 7367/76, 6 November 1980. This issue is not addressed in the Directive.

Audio-visual recording of questioning: Member States will have to ensure that questioning of children by police or other law enforcement authorities is audio-visually recorded but only “when this is proportionate in the circumstances of the case, taking into account inter alia whether a lawyer is present or not and whether the child is deprived of liberty or not, it being understood that the child's best interests shall always be a primary consideration” (Article 9 (1) and Recital 21a).  The proportionality condition needlessly weakens this protection by introducing an unacceptably wide scope of discretion and consequently potential variability.

If there is no audio-visually recording, questioning is to be recorded in another appropriate manner, and including “by making written minutes, which are duly verified” (Article 9(2)). By whom? The child? This should have been more clearly set out.

Limitation of deprivation of liberty: Article 10 is a welcome re-confirmation that any “deprivation of liberty of a child at any stage of the proceedings shall be for the shortest appropriate period of time” and that it should be a “last resort”. It also confirms that the “the age and individual situation of the child, and of the particular circumstances of the case” should be taken into account and sets out the need for a reasoned decision and periodic review.  However, these really are the some of the most basic minimum requirements of lawful practice. Article 10 does not define deprivation of liberty, nor does it restate or refer to the clear requirements attached to any deprivation of liberty established by Article 5 ECHR and its case law which are also reflected in Article 48 of the EU Charter not least those concerning the need for prompt determination of the decision to detain.

Article 11 and Recital 25a confirm that competent authorities shall have recourse to alternative measures instead of detention but, again, this requirement is undermined by the requirement that this need only happen “where possible”.

Specific treatment in the case of deprivation of liberty: Article 12 details minimum requirements for children who are detained. They must be held separately from adults “unless it is considered in the child's best interest not to do so” (in accordance with Article 37(c) of the UN CRC). In police custody, there is an additional exception as children need not be held separately if “in exceptional circumstances, it is not possible in practice to do so, provided that this is compatible with the child's best interests” (Article 12 (1a) and Recital 26a). Additionally, Member States have a discretion to hold children separately when they reach the age of 18 years but they must still take into account the child’s individual circumstances as well as the best interests of children who are detained with them (Article 12(2)). However, General Comment 10 confirms that the ‘best interests’ test here should be interpreted narrowly and that “the child’s best interests” does not mean for the convenience of the States parties”. This requirement should have been replicated on the face of the Directive.

When in detention, Article 12(4) and Recitals 26c and 26d require Member States to take appropriate measures to: (a) ensure and preserve their health and their physical and mental development; (b) ensure their right to education and training, including for children with physical, sensory and learning disabilities; (c) ensure the effective and regular exercise of their right to family life; (d) ensure access to programmes that foster their development and their future integration into society; and (e) ensure respect for their freedom of religion or belief. These measures do not go far enough and do not match the standards set out in General Comment 10, paras. 85-89.

Children are also entitled to meet the holder of PR as soon as possible but only “where such meeting is compatible with investigative and operational requirements” and only where the provisions of Article 5 relating to the designation of another appropriate adult have not been applied (Article 12(5)). It is deeply disappointing that operational demands may be placed above a child’s right to see his/her parent/carer.

Timely and diligent treatment of cases: Article 13 requires that all appropriate measures should be taken to ensure that criminal proceedings involving children “are treated as a matter of urgency and with due diligence”. This is in line with ECtHR case law on the right to a trial within a reasonable time. Under Article 18a an effective remedy under national law should be provided in the event of a breach of Directive rights. 

The requirement that appropriate measures should also be taken to ensure that children are “always treated in a manner which protects their dignity and which is appropriate to their age, their special needs, their maturity and level of understanding, and bearing in mind any communication difficulties they may have” (Article 13 (2)) is welcome.

Right to protection of privacy: Article 14 protects the privacy of children during criminal proceedings. Court hearings involving children should ordinarily be held in the absence of the public, or courts or judges to decide to do so. This reflects current human rights standards in Article 6 ECHR (e.g., see T v. UK, No. 24724/94, 16 December 1999) and Article 47 of the EU Charter.

Right of the child to be accompanied by the holder of parental responsibility during the proceedings: Article 15 and Recitals 29 and 29 confirm the right of the child to be accompanied by the holder of PR during court hearings. This can be limited on the same conditions as Article 5(2). Article 15 similarly permits the state to appoint an appropriate adult where an alternative is not acceptable.

Children also have the right to be accompanied during other stages of the proceedings but only where the state considers that: (a) it is in the interest of the child to be accompanied by that person; and (b) the presence of that person will not prejudice the criminal proceedings. The state is given total discretion and the Directive should have been far clearer on a child’s rights in this regard.

Right of children to appear in person at, and participate in, their trial: Article 16 confirms a child’s right to be present at, and participate effectively in, their own trial. This includes giving them the opportunity to be heard and to express their views. If a child is not present at their trial, the Directive provides the right to a new trial, or another legal remedy, in accordance with and under the conditions set out in Directive on the presumption of innocence. The requirements of Article 6 ECHR and Articles 47 and 48 of the EU Charter would require nothing less.

Right to legal aid: Article 18 requires that national law in relation to legal aid guarantees the effective exercise of the right to be assisted by a lawyer as referred to in Article 6. The reference to national law does not help promote standards in view of the widespread inconsistency and variability (leading to the Commission’s Recommendation on the issue). The interests of justice test indicates that free legal assistance may be required for vulnerable groups such as minors (Quaranta v. Switzerland, No. 12744/87, 24 May 1991, para. 35. This could have been clearly restated.

Other provisions

Training: Article 19 deals with training for law enforcement authorities, staff of detention facilities, the judiciary, prosecutors and lawyers. This is a welcome and potentially practical step and should be supported by further EU wide guidance drawn from universal standards.

Data recording: Article 20 requires that Member States send data to the Commission five years after the Directive enters into force (and every three years thereafter) showing how the Directive has been implemented. The absence of reliable data on children’s rights in criminal proceedings has previously been noted by the Commission (see the report here) so this is a welcome obligation.


Clear standards protecting children’s rights are welcome but, based on the summary above, I have the following general concerns:

1.      The failure to mainstream these protections: they are placed in a separate document which will need to be compared with others. Should, they not run like a thread through all the standards? Developing safeguards incrementally may invite inconsistency.
2.      While some of the Directive’s provisions are new and child-specific, others seem to be little more than a reiteration (and sometimes a partial one) of the minimum requirements which exists in relation to all defendants.
3.      The Directive does not always reflect international standards of protection.
4.      The Directive is made potentially less progressive by a series of exceptional and discretionary provisions which may facilitate variable practice.

Additionally, in reality, to make a difference any new standards must actually be enforced in practice. We must ask: will these standards work where others have failed?

Barnard & Peers: chapter 25
JHA4: chapter II.4

Photo credit: http://chelseaclockwallpaper.blogspot.co.uk/2012/12/juvenile-justice-system.html