17 November 2022

Picking Primacy over Procedural Autonomy

On 8 November, the Grand Chamber of the Court of Justice of the European Union (‘the Court’) decided that national courts are required to ascertain of their own motion whether detention of an illegally staying foreign national or asylum seeker is lawful. This judgment is an example of the ever-growing impact of EU law on national procedural rules, especially in the migration law area. The judgment is also noteworthy because of the difference in approach between, on the one hand, the Court and, on the other hand, the Dutch referring courts and AG Richard de la Tour. The latter approached the matter from the perspective of national procedural autonomy and Rewe effectiveness (in the absence of EU law harmonisation of procedural rules) while the Court concluded that there are common procedural EU standards and, hence, applied the Charter with an implicit ‘direct collision/ primacy’ logic in mind.

This blog will firstly discuss the Court’s striking approach, contrasting it with the AG’s Opinion. As will be explained, both the Court and the AG rely on a different frame, namely direct and indirect collision (section I). After having examined the implications of this judgment, we take the view that the Court’s approach probably has less far-reaching consequences for national procedural law than the one proposed by the AG (section II). Additionally, this blog addresses the intriguing dynamic between the two Dutch referring courts that led to the judgment in the Joined Cases (section III). The Cases involve a reluctant highest administrative court precluding ex officio review on the basis of Dutch administrative law and a confident district court judge maintaining a tough EU law-proof stance ‘no matter what’ (as she noted herself on Linkedin after the judgment of the Court).

The Court’s striking ‘direct collision’ approach

The preliminary reference questions dealt with detention of third country nationals in relation to three EU law instruments: the Return Directive (Article 15); Reception Conditions Directive (Articles 8 and 9) and the Dublin III Regulation (Article 28). These provisions provide the competences of Member State judges to verify the legality of such a detention. The Court took the question of the two Dutch courts together and examined whether a judicial authority must raise of its own motion any failure to comply with a condition governing lawfulness of detention which has not been invoked by the person concerned.

What is striking about the Court’s judgment is that the Court sidestepped the procedural autonomy discussion altogether. Just to recap, the Court has held, albeit not very consistently since Rewe that, in the absence of EU rules governing the matter, it is for each Member State, in accordance with the principle of the procedural autonomy, to lay down the detailed rules of administrative and judicial procedures governing actions for safeguarding rights which individuals derive from EU lawThose detailed procedural rules must, however, be no less favourable than those governing similar domestic actions (principle of equivalence) and must not render practically impossible or excessively difficult the exercise of rights conferred by EU law (principle of effectiveness). According to the Court, in this case there were, however, EU procedural standards (para 86). It therefore concludes: ‘Since the EU legislature requires, without exception, that supervision that the conditions governing the lawfulness of the detention are satisfied must be effected ‘at reasonable intervals of time’, the competent authority is required to carry out that supervision of its own motion, even if the person concerned does not request it’ (para. 85). This conclusion is surprising as AG Richard de la Tour determined that there are common rules on judicial review of detention but not the scope of review (paras. 71-73). The referring Dutch highest administrative court, the Council of State, held the same assumption (paras. 3.9 and 4).

How did the Court then arrive at this quite surprising conclusion? The Court pointed to the gravity of detention as an interference with the right to liberty as laid down in Article 6 of the Charter as well as the strict circumscription of the power of the competent national authorities (paras. 72-75). It subsequently noted that common EU standards regulate the conditions pertaining to detention, prescribing that a person should be released immediately when these conditions are no longer satisfied (paras. 76-80). Complementary common EU standards on judicial protection in relation to detention, that give concrete form to the right to effective judicial protection safeguarded in Article 47 of the Charter, provide for ‘speedy’ judicial review as well as review or supervision ‘at reasonable intervals of time’ whether the conditions governing the lawfulness of the detention continue to be met (paras. 81-84).

The Court, thus, suggests that the required periodic review or supervision simply implies ex officio review. It further adds that the ‘imperative need to release such a person, where the conditions governing the lawfulness of detention are not, or are no longer, satisfied’ simply requires that the court is able to ‘rule on all matters of fact and of law’ and ‘under no circumstances be confined just to the matters adduced by the administrative authority’ (para. 87). The Court bolsters this conclusion by emphasising the importance of the right to liberty and the gravity of the interference (para. 88), the necessity to effectively guarantee judicial protection of the fundamental right to liberty in all Member States (para. 90) and the strict circumscription of (the continuation of) detention in EU law (para. 92). The Court also notes that ex officio review is already a reality in Member States where detention orders must be made by a judicial authority (para. 89). Only towards the end did the Court address the national procedural autonomy perspective, albeit rather implicitly, by determining that its ex officio case law alluded to by the referring Dutch Council of State does not invalidate its conclusion (para. 91).

By not scrutinising national procedural rules through the lens of national procedural autonomy the Court opted, albeit implicitly, for the ‘direct collision’ frame. In case of a direct collision, both the EU and national norm provide diverging, incompatible legal regimes for the same set of facts. Such collisions are typically ‘solved’ by the Court by relying on the primacy of EU law. Even though the current judgment avoids this language, the case at hand essentially boils down to a direct collision between common EU procedural standards and national procedural law. In case of an indirect collision, a national (procedural) rule hinders the effectiveness of a (material) EU norm, without prescribing different legal consequences for the same factual situation. In those instances, there is no ‘EU equivalent’ to the national norm and therefore the collision cannot be reconciled through the primacy of EU law. Instead, the Court then finds the solution in the national procedural autonomy of the Member State, setting limits for the national procedural rule utilising the Rewe principles of equivalence and effectiveness and/or Article 47 of the Charter of Fundamental Rights (CFR) (see here for a more in-depth discussion on the notion of (in)direct collisions).

Overly intrusive, or not? Consequences for national procedural rules

This judgment is an example of the growing influence of EU law on national procedural law, primarily via Article 47 CFR and especially in migration and asylum law (see for a discussion here). This includes judgments about the competence to review detention measures in the absence of a national power (FMS), the requirement of judicial instead of administrative remedy (El Hassani) and the possibility for courts to take into account circumstances subsequent to the transfer decision (H.A.).

In our view, the Court’s approach actually minimises the consequences of the judgment beyond the particular circumstances of the case involving the right to liberty in a context in which the applicants are vulnerable. It explicitly determines that detention is ‘not similar in every respect to administrative proceedings, in which the initiative and delimitation of the dispute lie with the parties’ (para. 92). Therefore, it to an extent distinguished detention cases from ordinary administrative law proceedings. In Dutch administrative law, parties typically delimit the dispute, with little to no room for ex officio review by the administrative judge (see Article 8:69 of the General Law on Administrative Law (Awb). This provision played a central role in the two Dutch orders for reference, especially the order of Council of State. The Court determines in essence that this provision does not apply to these cases, thereby (implicitly) giving the Council of State a rap over the knuckles for applying the provision in detention cases. This judgment thus does not necessitate the amendment of Article 8:69 Awb, because of this clear distinction. Nevertheless, we cannot rule out that ex officio review will be required in more situations in which important fundamental rights are at stake.

At first glance and at a more abstract level, the ‘direct collision’ approach of the Court seems more intrusive than the approach suggested by the AG. While the material outcomes of the judgment and the Opinion do not differ much, we, however, think that the AG’s approach would have more far-ranging and uncertain consequences. In the reasoning of the Court, the specific EU legal framework directly prescribes what the national court must do: review the legality of detention measures ex officio. So the Court’s approach only seems to overhaul particular procedural rules in case of detention measures, namely those relating to ex officio review. By contrast, the AG draws a (particularly narrow) ‘circle’ around national procedural rules by relying on effectiveness, substantiated by Article 6 and 47 CFR, concluding that a rule that does not allow for ex officio review falls outside the permissible ‘circle’ (para. 92-95). Through his ‘effectiveness circle’, narrowed down by Article 47 CFR and potentially other fundamental rights, other national procedural rules could be scrutinised and struck down as well.

Intriguing dynamic between the two Dutch referring courts

Another interesting dimension of the joined cases is the dynamic between the two referring courts. The first reference (C-704/20) of December 2020 was made by the highest Dutch administrative court, the Council of State, in a case that was decided in first instance by the same district court judge that made the reference in the second case a month later (C-39/21). This judge (Steffie van Lokven) has by now built a solid reputation of making very solid and elaborate references in the migration law field in relation to situations in which Dutch legislation and practice are in breach of EU law. The district court in essence forced the Council to refer when it conducted an ex officio review of detention in clear deviation from the prohibition in Dutch procedural law and the established case law of the Council supporting this prohibition. Even though CILFIT (2.0) does not force a highest court to refer when a lower court casts doubts upon its interpretation of EU law, it becomes more difficult for the highest court to maintain that there is no doubt and, hence, an acte clair (see also par. 7 of the Council’s order for reference).

The district court subsequently felt compelled to send additional questions to Luxembourg. The court noted rather critically in its order (a summary of which is available in English here) for reference that the Council had failed to notify all other Dutch administrative courts of its intention to refer pursuant to non-binding agreements (as discussed here). This enables other courts to indicate that they are considering asking similar questions so that the questions can be submitted to the CJEU jointly. The district court felt compelled to independently make an additional reference in order to be able to elucidate the whole picture to the CJEU (par. 4). This is, as such, not surprising and has happened before, especially in the migration law context (see YS in which a district court challenged the case law of the Council with a reference and the Council sent another reference to “defend” itself, as discussed here).

In its order, the district court responded quite directly, critically and extensively to the Council’s order for reference. It noted, for example, that the Council paid no attention to Article 47 of the Charter in its 25 page-long order aside from an implicit reference in passing (par. 6; par. 4 summary). The court also downplayed the Council’s emphasis on procedural autonomy (par. 7; par. 6 summary) and noted that the Council had not explicitly held that there is a competence to conduct an ex officio review (par. 9; par. 5 summary).

It would be simplistic to label the reference of the district court as strategic or activist (see here), because the court aims to arrive at the correct interpretation of EU law. Even though lower courts are not obliged to refer, as Union courts they are obliged to interpret and apply EU law in line with primary and secondary EU law. Thus, using these labels without sufficient regard to the wider context, does not do justice to the intentions of lower courts and could also lead to an unjustified frame of a reluctant highest court disrespecting EU law. At the same time, one should not overlook the fact that these detention cases were the result of strategic litigation on the side of various (other) actors. A prominent academic expert on immigration detention appeared before the Council in support of the third country nationals. The lawyer Piet Hein Hillen also played an essential role as noted by the district court judge who lauded his exceptional pleas. In addition, an expert opinion was written by the VU Migration Law Clinic at the request of the applicants’ lawyer.

All in all, the references led to highly anticipated judgment from Luxembourg in a salient case involving various actors and courts. According to us, the clarity from the Court is clearly desirable from the perspective of the uniform protection of fundamental rights in all EU Member States. It requires Dutch judicial and administrative authorities to provide the level of judicial protection as agreed in EU secondary law and already provided in several Member States. Primacy after all.