For a Wiser and Effective Management of Reintroducing Internal Border Controls: Comments on the NW Judgment

By Jean-Yves Carlier and Eleonora Frasca, Université catholique de Louvain (UCLouvain), members of the Equipe droits et migrations (EDEM)

    

This is a revised version of the commentary appeared in “Free movement of persons in the EU”, the yearly case law review published in French in the Journal de droit européen, No. 298, April 2023, pp. 198-213.

1. Interpreting or reforming Schengen?

In the Landespolizeidirektion Steiermark (NW) (C-368/20 & C-369/20) judgment of April 2022, the Court of Justice opted for a strict interpretation of the Schengen Borders Code (SBC) by prohibiting, in principle, Member States from reintroducing internal border controls for a total period of more than six months, even with the implicit approval of the European Commission. The Court was unusually firm and departed from the opinion of Advocate General Saugmandsgaard Øe, presented on the eve of the end of his term of office at the Court. The Advocate General tried to reach a conciliatory interpretation between the principle of the abolition of internal border controls and the exceptions on the grounds of public order or internal security. According to him, exceeding the maximum duration of six months could be subject to a refined proportionality test, thus leaving a certain margin of appreciation to the Member States. The Advocate General’s opinion was widely criticised (Leclercq), while the Court’s judgment was generally welcomed as particularly bold (Tometen; Bornemann; Cebulak and Morvillo). This common praise was tempered only by a few more in-depth comments that, while remaining positive, questioned the consequences of the judgment regarding reforms to the Schengen Code (Thym; Leboeuf; Marinai; Rom; Martins Pereira).

We are more critical of the Court and we believe that the conciliatory approach proposed by the Advocate General deserved greater attention and would have paved, in its reasoning if not in its opinion, a more pragmatic way of developing the Schengen Code, without the need for legislative reforms whose effectiveness seems inversely proportional to their frequency (see Proposal COM(2021) 891 final). The question of the effectiveness of the SBC is significant at a time when the Schengen area now includes Croatia and will eventually include Bulgaria and Romania.

2. Case summary

An Austrian citizen, NW, was checked twice on his way into Austria from Slovenia. NW refused to produce a passport or identity card to the border control officer, he asked for their registration number and questioned them about this control. Was it an identity check or a border check? He specified border checks were contrary to EU law. Having limited himself to producing a driving licence, NW received a 36 EUR fine and appealed to the Styrian Regional Administrative Court. While considering that the fine was an application of Austrian law, the Administrative Court asked the ECJ whether that national law was compatible with EU law.

Austria, like many other Member States, including Denmark, Germany, France and Sweden, which intervened in the procedure in support of Austria, has reintroduced controls at its internal borders since 2015, in the context of the migration “crisis” following the war in Syria.

Until 2017, these controls were extended based on Article 29 SBC. Article 29 SBC allowed the reintroduction of controls and their extension for up to two years in total where exceptional circumstances put the overall functioning of the area without internal border control at risk as a result of persistent serious deficiencies relating to external border control. Applying this exception requires a procedure with a Council recommendation on a proposal from the Commission.

After 2017, in the absence of a Commission’s proposal, subsequent extensions were no longer based on these exceptional circumstances related to the overall functioning of the Schengen area, but on a serious threat to public policy or internal security in the Member State concerned. The SBC covers two hypotheses: unforeseeable threats (Article 28 SBC) and foreseeable threats (Article 25(1) SBC). Subsequent extensions of controls were based on article 25(1) SBC at the time of the event, in 2019, and beyond. These extensions can be made for periods of 30 days, in compliance with the procedure set in Article 27 SBC which requires notification to the Commission and to the other Member States.

Austria had complied with this procedure. However, according to Article 25(4) SBC, “The total period during which border control is reintroduced at internal borders, including any prolongation provided for under […] this Article, shall not exceed six months”. In the present case, the six-month period was considerably exceeded.

In summary, according to Austria and the intervening States, on the one hand, a new serious threat to public order or internal security could lead to a new application of these periods and, on the other hand, a new assessment of the previous threat should also allow for further extensions. While the Court accepted the first hypothesis (new threat), it rejected the second (new assessment of a previous threat).

Through a literal, contextual and teleological analysis, the Court recalled that the principle was the absence of internal border controls in such a way that the exception of the reintroduction of controls must be strictly interpreted. In the event of a foreseeable threat, the total duration of the reintroduction of controls may not exceed six months.

Germany also invoked Article 72 TFEU, according to which the whole of Title V TFEU, concerning the Area of Freedom, Security and Justice, in which the SBC is included,

“is not to affect the exercise of the responsibilities incumbent upon Member States with regard to the maintenance of law and order (ordre public) and the safeguarding of internal security” (§ 85).

The Court replied that this was a derogation, also of strict interpretation. Article 72 TFEU

“cannot be read in such a way as to confer on Member States the power to depart from the provisions of EU law on the basis of no more than reliance on the responsibilities incumbent upon Member States with regard to the maintenance of law and order and the safeguarding of internal security” (§ 86).

The maximum duration, whether six months in the event of a threat to the public policy or security of a State or two years in the event of an exceptional event related to the overall functioning of Schengen,

“forms part of the comprehensive framework – established by the EU legislature […] – governing the way in which the Member States exercise the responsibilities incumbent upon them for the purpose of the maintenance of public policy and the safeguarding of internal security” (§ 87).

Consequently, the Court ruled that the SBC

“must be interpreted as precluding border control at internal borders from being temporarily reintroduced by a Member State on the basis of Articles 25 and 27 of that code where the duration of its reintroduction exceeds the maximum total duration of six months, set in Article 25(4), and no new threat exists that would justify applying afresh the periods provided for in Article 25” (§ 94).

3. Analysis

Our analysis will focus on three main points: the facts (3.1), the Court’s reasoning (3.2) and the potential consequences of this judgment (3.3).

3.1 Facts: An active EU citizen replacing the passive Commission

The reader will have noticed that Mr NW is no ordinary citizen. He is well-versed in EU law, which he invokes, showing to be aware of its details when he presents his driving licence to the officer, rather than his passport or identity card, just as Mr Wijsenbeek did twenty years ago while challenging the controls at Rotterdam airport. In fact, a well-informed young scholar tells us, in her relevant comment, that NW is an academic involved in “scholarly activism” (Rom). This is, therefore, a test case carried out by an active EU citizen, as it happened in other cases, including Coman. Nothing is wrong with this, it is just something one should know.

On the contrary, to this citizen’s merit, one shall acknowledge that he remedies the astonishing passivity of the Commission. Not only did the Commission fail to play its role as guardian of EU law, but did played both sides before and during the procedure. Before the procedure, it did not issue any opinion on the reintroduction of controls by Austria. According to the first subparagraph of Article 27(4) of the SBC, the Commission “may”, like any Member State, issue an opinion following notification of the reintroduction of controls, but the second subparagraph of the same Article states that

“If […] the Commission has concerns as regards the necessity or proportionality of the planned reintroduction of border control at internal borders, […] it shall issue an opinion to that effect”.

The words “it shall issue an opinion to that effect” indicate an obligation to issue an opinion in case of doubt. In these circumstances, silence could be interpreted as approval.

On the contrary, after this silence (which looks like a quasi-failure to act), the Commission intervened in the case before the ECJ in support of the applicant, taking the view that

“the reintroduction of internal border control […] for a period exceeding the maximum duration of six months […] necessarily results in that border control being incompatible with EU law” (§ 55).

This is a double standard. By omission and by action, this fault does the Commission little credit. The Advocate General and the Court have criticised the Commission’s failure to issue an opinion, with the Court noting that while this “does not, in itself, have any bearing on the interpretation by the Court of [the SBC’s] provisions”, it is nonetheless

“essential, in order to ensure proper operation of the rules established by the Schengen Borders Code, that, when a Member State wishes to reintroduce internal border control, both the Commission and the Member States exercise the powers conferred upon them by that code, in particular so far as concerns the exchanges of information and opinions, the consultations and, where appropriate, the mutual cooperation that are specifically provided for in Article 27 of that code” (§§ 92–93, and AG’s opinion, §§ 71–74).

Such a warning from the Court to the Commission echoes the Bosman judgment about free movement of persons. At that time, in connection with the Commission’s approval of the rules on nationality quotas applied by football associations, the Court recalled that

“except where such powers are expressly conferred upon it, the Commission may not give guarantees concerning the compatibility of specific practices with the Treaty […]. In no circumstances does it have the power to authorize practices which are contrary to the Treaty” (Bosman, § 136).

Conversely, in the Landespolizeidirektion Steiermark case, the Schengen Border Code required the Commission to take a position on a State’s behaviour contrary to EU secondary law so that that same State could legitimately consider the Commission’s self-restraint as a blank check.

This raises questions about the Court’s reasoning.

3.2 The Court’s reasoning: A frozen balance between free movement and security interests

The Court’s reasoning can be seen as attributing an absolute value to a time limit (six months) considered as an irrebuttable presumption of the lack of proportionality of a measure not respecting the fixed deadline. The Court rightly recalls the premise of the framework set by the Area of Freedom, Security and Justice. This framework, says the Court,

“strike[s] a fair balance between, on the one hand, the free movement of persons and, on the other, the need to safeguard public policy and internal security of the territory on which they move” (§ 72).

The Court adds that

“The objective pursued by the rule relating to the maximum duration of six months […] is thus a continuation of that general objective which consists in reconciling the principle of free movement with the Member States’ interest in safeguarding the security of their territories” (§ 76).

Indeed, it says,

“the EU legislature considered that a period of six months was sufficient for the Member State concerned to adopt, where appropriate in cooperation with other Member States, measures enabling such a threat to be met while maintaining, after that six-month period, the principle of free movement” (§ 77).

Consequently, “the period of a maximum total duration of six months […] is mandatory” (§ 78). In other words, the six-month period constitutes an irrebuttable presumption. Beyond this period, any extension of the checks is automatically deemed disproportionate.

This logic of the presumption period is not new in the Court’s case law. It also appears in various cases relating to EU citizens’ rights, in particular the social rights of jobseekers, with regard to the time limits for residence or economic activity laid down in Directive 2004/38.

For instance, in Alimanovic, the Court excludes an in concreto examination of proportionality to measure the excessive burden or not that the access of a given EU citizen to social rights at the expense of the host Member State would represent. In the context of what was considered a means of combating “social tourism”, departing from previous case law such as Grzelczyk, which founded EU citizenship as a fundamental status, and departing from the opinion of Advocate General Wathelet, the Court rejected the in concreto examination of proportionality, considering that

“no such individual assessment is necessary [since] Directive 2004/38 […] itself takes into consideration various factors […], in particular, the duration of the exercise of any economic activity” (Alimanovic, §§ 59–60).

In both cases, the time limits function as presumptions precluding any in concreto examination of proportionality. However, under Directive 2004/38, it was an obstacle to free movement while, in Landespolizeidirektion Steiermark, it is, on the contrary, a guarantee of free movement. Is this difference enough to criticise one Court’s reasoning and praise the other?

In both cases, the Court’s reasoning avoids, at least on surface, the search for a fair balance and favours an examination frozen by numbers. In both cases, the opinions of the Advocates General proposed more flexible approaches, based on an in concreto examination of proportionality. In the Landespolizeidirektion Steiermark case, at the outset of his opinion, the Advocate General sets out the same premise:

“the delicate balance […] between the freedom of EU citizens to move in an area without internal frontiers and the maintenance of security in that area” (AG’s opinion, § 1) and repeats it (§ 58).

However, stressing “The need to respect the Member States’ powers in public policy and internal security matters”, the Advocate General argued for an in concreto examination of the proportionality of any border control measure (opinion, heading (b) before § 50).

Moreover, this in concreto examination of proportionality is in line with the Court’s case law on internal border controls in Wijsenbeek, only cited by the referring court, and A., which the Court does not cite either, and for good reasons. Indeed, in that case, although the six-month time limit was not mentioned, the Court stated in a much more general way that

“the exercise by nationals of a Member State of their right to travel to another Member State is subject to the condition that they carry a valid identity card or passport” (A., § 52).

The Court did not deduced from this the possibility of systematic “checks” at internal borders, but the possibility of “verifications” of identity accompanied by proportionate sanctions. Hence Mr NW’s subtle question to the officer in the Landespolizeidirektion Steiermark case. However, this in concreto approach to proportionality differs from the presumptive approach adopted by the Court in Landespolizeidirektion Steiermark.

This in concreto approach was already developed by the same Advocate General Saugmandsgaard Øe in his opinion in case A (Public Health Care). As Advocate General Wathelet had already done on these questions of access to social rights concerning the unreasonable burden on the host State, Advocate General Saugmandsgaard Øe ruled out any “automatic refusal, in all circumstances” in favour of the

“need for an individual examination of the Union citizen’s situation in order to determine whether he constitutes an unreasonable burden” (AG’s opinion, heading c), before § 128).

At that time too, in A., the Court did not follow the Advocate General’s opinion. In the case A (Public health care), the Advocate General developed that same logic of an in concreto examination of proportionality in search for a balance between competing interests. However, there, he did so in favour of the individual interests of the EU citizen. Here, in Landespolizeidirektion Steiermark, he does so in favour of the collective security interests guaranteed by the State. Notwithstanding, here, in view of the risk of infringement of freedom of movement and the six-month time limit set by the SBC, the Advocate General proposes a “condition of enhanced proportionality” (AG’s opinion, § 68) for the continuation, beyond six months, of internal border controls based on the same renewed threat.

In any hypothetical case of reintroduction of controls, a proportionality test is required on the basis of Article 26 SBC. In the event of a renewal beyond six months due to a persistent threat, it would be necessary to assess “in even greater detail the strict proportionality of the measure in order to prove that the continuation of the controls is strictly necessary” (Opinion, § 67).

To do so, the State concerned should explain

“on the basis of specific, objective and comprehensive analyses, first, why the renewal of controls would be appropriate, by evaluating the degree of effectiveness of the initial measure reintroducing the controls. Second, it must make clear why it remains a necessary measure, by explaining the reasons why no other less coercive measure would be sufficient, such as, for example, the use of police control, […] intelligence, police cooperation at EU level and international police cooperation” (Opinion, § 67).

In conclusion, the Advocate General admitted that while this interpretation

“does mean that a control can, in certain circumstances, exceed the six-month period, that point is however counterbalanced, first, by the application of the principle of enhanced proportionality and, second, by the monitoring carried out by the Commission, which must verify scrupulously whether that condition is satisfied each time” (Opinion, § 74).

We support this in concreto examination of proportionality. Similarly, we subscribe to the idea of a strengthened or refined principle of proportionality, as we had already proposed, following the Bosman judgment, to distinguish the stricter proportionality criterion applied to overt or covert discrimination as obstacles to the free movement of persons, from the more flexible disproportionality criterion used for national measures that apply indiscriminately, as indistinctly applicable measures. However, we depart from the Advocate General’s opinion as to the consequences of his conclusion in the present case. At the end of the opinion, the Advocate General appears to endorse the measures adopted by Austria to extend checks and the penalties for refusing to produce an identity document. However, there is no indication that Austria had provided the necessary evidence for this enhanced proportionality check, merely asserting that there was a continuation of the threat.

A correct application of this proportionality test would have required the Court of Justice to lay a framework for it, as requested by the Advocate General, but also, in the light of the questions referred by the domestic court, to propose an understanding of the framework in the case in question, while at the same time delegating responsibility for the examination in concreto to the domestic court (see however, AG’s opinion, footnote 30). The national court seemed reluctant to accept the conformity of Austria’s measures with EU law.

An in concreto approach would probably have prevented the many questions that remain open after the Court’s judgment.

3.3 What next? The ongoing questions following the judgment

The consequences arising from this judgment mainly concern, in the short term, the SBC’s interpretations and, in the long term, its eventual reforms.

Short-term consequences of the judgment

In the short term, the judgment leaves many questions open. Concerning the time limit for the reintroduction of controls beyond the six-month period, the SBC is silent, and the judgment does not provide any answer. Of course, this question did not arise in the present case, but it may arise as due to the judgment. If Austria, learning from the judgment, removed the internal border controls (quod non), how much time would need to pass before it could reintroduce controls again if the threat proved to be persistent, despite the adoption of other measures? Thirty days, which is the period provided for by the renewal period? Less, to offset the risk of insecurity? More, i.e., the full six-month period? All this seems “wholly illogical” (AG’s opinion, § 52). In the case of a new threat, no delay would be necessary. The Court says so: in the case of a “new serious threat […], fresh periods […] may be regarded as beginning to run” (§ 79).

This raises the second question: what is a new threat? The Court attempts to define what a “new threat” would be. Using the terms in Article 27 of the SBC, the ECJ states that

“it is always by reference to those circumstances and events [which gave rise to the need to reintroduce control] that it is to be assessed whether […] the threat that the Member State faces remains the same threat or whether it is a new threat” (§ 81).

This seems obvious. Theoretically simple, this definition does not provide any practical answer. Of course, a terrorist or public health threat, such as the COVID pandemic, will be distinct “circumstances and events”. But what about a terrorist threat by group X, followed by a terrorist threat by group Y, or by the same group X, following other circumstances, such as the arrest or trial of one of its members? Respecting each other’s roles and competences, the ECJ states that it is “for the referring court to determine [whether] the Republic of Austria did not demonstrate the existence of a new threat” (§ 82). Rather than imposing on the national court a semantic study, distinguishing the new threat from the renewed threat, would it not have been more effective, in both the short and the long term, to invite the national court to carry out a strengthened proportionality test, requiring the State to produce an impact study providing, under the supervision of the Commission, proof of the absolute necessity of maintaining the controls and the impossibility of other less restrictive measures? If not, the practical effects of the judgment are likely to be limited. States will easily find – or fabricate – new threats. The first signs are already there.

In November 2022, Austria renewed the control period at its borders with Slovenia and Hungary. In addition to the already mentioned threat of migration and organised crime, it added the threat of the war in Ukraine.

In France, various associations, relying on the Landespolizeidirektion Steiermark judgment, have tried to obtain the annulment of the reintroduction of controls at all internal borders. However, the Conseil d’Etat has dismissed the application and consolidated its case law. France invoked threats linked to the risk of terrorism, the COVID pandemic, the secondary movement of migrants and the risks generated by the conflict in Ukraine. While the Conseil d’Etat accepted that migrant movements were not a new threat, it noted that the conflict in Ukraine was a threat that was “new by its nature”. For the threat linked to terrorism, the Conseil d’Etat considered the increased risk of combatants returning, the risks linked in particular to the expected verdict in the trial of the 2015 attacks in Paris. Finally, the arrival of new dominant variants was considered as a new threat of the COVID pandemic. All these elements represented “new circumstances and events”. Based on the Landespolizeidirektion Steiermark judgment, the Conseil d’Etat defines a new threat as follows:

“A threat may […] be regarded as new, within the meaning and for the application of these provisions, either when it is of a different nature from previously identified threats, or when new circumstances and events change its characteristics in such a way as to alter its actuality, scope or consistency. Such circumstances and events may relate, inter alia, to the purpose of the threat, its magnitude or intensity, its location and its origin” (French Conseil d’Etat, § 5).

Such a definition corresponds, in part, to the criteria envisaged by Advocate General Saugmandsgaard Øe for the exceptional renewal of checks beyond the six-month period, but with the essential difference that he added the procedural guarantee of the reinforced proportionality test, under the supervision of the Commission.

Long-term consequences of the judgment

In the long term, a further amendment of the SBC would probably be necessary to ensure that these guarantees are incorporated and that a relative balance is found. A proposal to amend the SBC had already been tabled by the Commission in December 2021. It was reviewed and substantially amended by the Council in June 2022. In contrast to the usual evolution of the texts, the text resulting from the Council is stricter and firmer than the Commission’s in its demands on the States to safeguard freedom of movement. In its typical style, the Commission had announced its proposal, that

“would aim to complete the range of tools necessary to ensure the proper functioning of the Schengen area in order to put the ecosystem of rules back into balance and restore and reinforce mutual trust between Member States” (Proposal, p. 1).

In practice, the proposal offers a wide margin of manoeuvre to the States. On the one hand, they would have the possibility to maintain controls beyond six months, “where the serious threat […] persists”, for “renewable periods of up to six months”, with a “maximum duration [which] shall not exceed two years” (Proposal, Article 25a(5)). On the other hand, States could also extend controls beyond two years, without any time limit, when they consider that “exceptional situations” justify it, in which case “the Member State shall substantiate the continued threat […], taking into account the opinion of the Commission” (Proposal, Article 27a(5)).

The Council introduced some modifications to the draft, on the one hand, by reinforcing the exceptional nature of unlimited extensions and the need for other measures, and, on the other hand, by strengthening the procedural monitoring role of the Commission.

These legislative reform proposals remain uncertain, in their adoption, in their content and in the interpretations that the States will make of them. In the meantime, far from having put an end to abuses, the only consequence of the Court’s ruling seems to be that States – and their supreme courts – will have to be more subtle in distinguishing among an old threat, a continuing threat, a renewed threat or a new threat.

Conclusion

On 6 December 2022, on the formal sitting for the 70^th anniversary of the ECJ, President Koen Lenaerts insisted on “preserving the harmonious cohabitation of values of the Union and national identities” which constitutes a “subtle exercise” of “delicate arbitration in the constant search for the best point of balance” (ECJ, Formal sitting: 1952–2022, at minute 29:19). In Landespolizeidirektion Steiermark, regarding the reintroduction of internal border controls in the Schengen area, the Court favoured a strict, even frozen, interpretation of freedom of movement rather than a concrete and enhanced assessment of the proportionality of the measures, as proposed by the Advocate General. All things considered, the conciliatory and pragmatic approach of an in concreto examination of enhanced proportionality, under the mandatory supervision of the Commission, seemed wiser and more effective than a strict interpretation of the six months maximum period. Sometimes, less may be more.