Restoring the Borderless Schengen Area: Mission Impossible? Summary of a new report

 

Professor Steve Peers, Royal Holloway University of London

Photo credit: BlueMars, via Wikimedia Commons

My European Policy Analysis report on the amendments to the Schengen Borders Code, published today by the Swedish Institute of European Policy Studies (SIEPS), assesses the likely impact of the recently agreed amendments to the Schengen Borders Code. This blog post is a summary of the main points of the report.

The context of the report is the concerns about the reintroduction of border checks at the internal borders of Schengen countries in recent years for long periods by a number of Member States. The EU has embarked upon a strategy to ‘save Schengen’, in part by amending the Schengen Borders Code to change the rules on internal border checks (among other changes), but also by agreeing major changes to EU asylum law alongside a number of other measures, for instance in the area of further police cooperation. The amendments to the Borders Code, along with most of the other proposals (including the asylum law changes), have now been agreed (on the asylum law changes, see the series of analyses on this blog beginning here). The report assesses the details of the Border Code amendments in the broader context, analysing whether they are likely to ‘save Schengen’ and whether they raise human rights concerns in the process.

Introduction 

The Schengen system, initially in the form of the Schengen Convention agreed in 1990 but largely replaced by EU measures since, is intended to abolish internal border checks between (most) EU Member States, as well as four associated non-EU countries. This abolition is linked to harmonised checks on external borders with non-EU countries, a common short-erm visa policy for non-EU visitors, and a Schengen Information System listing non-EU persons to be denied entry and objects and persons to be stopped or tracked.

The Schengen Convention rules on internal and external borders are now set out in an EU Regulation known as the Schengen BordersCode, the most recent version of which dates back to 2016. Although the Code does not abolish internal border checks absolutely, their reimposition is meant to be strictly limited. Nevertheless, there have been many resorts to internal border checks in recent years, in particular due to concerns about migration control and security.  

The response has been a plan to ‘save Schengen’, in particular by means of amending the Borders Code along with other changes to EU law. To what extent will the recently agreed amendments achieve that end – and do they raise human rights concerns in attempting to do so?

Internal Border Controls

The report details the current practice of Member States, which as noted already has entailed recent frequent resort to border checks for long periods. It then describes the current legal framework, including the case law.

In particular, the case law has taken a mostly deferential approach to Member States imposing checks on the territory near borders, as an exercise of ‘police powers’, if this does not have an ‘effect equivalent to border checks’.  According to the CJEU, it is sufficient if there are some safeguards in place to ensure that such checks are targeted, even if their purpose is to combat irregular migration.

The current Borders Code allows internal border checks as such for up to 30 days, or for a longer period if the duration of the relevant event is foreseeable, in the ‘event of a serious threat to public policy or to internal security’; but the ‘scope and duration’ of the reintroduced checks ‘shall not exceed what is strictly necessary to respond to the reintroduced checks’. The reintroduction of controls may be continued for further renewable periods of up to 30 days, ‘taking into account any new elements’. But the maximum time to reintroduce border controls is six months, or two years in ‘exceptional circumstances’, where there is an EU-wide threat (this was triggered in response to the refugee crisis, but the two year period has expired).

According to the CJEU’s judgment in NW, interpreting these provisions strictly as a derogation from the general rule of abolition of border controls, the six-month time limit on the reintroduction of internal border controls that applies where there are no extraordinary circumstances could only be triggered again where there was a serious new threat.  There are more specific rules, depending on whether the reintroduction of border checks is foreseeable, urgent, or constitutes ‘exceptional circumstances’

New Amendments

The recently agreed amendments to the Borders Code, likely to be formally adopted later this spring and apply soon after, aim to address Member States’ concerns. They address a number of issues besides internal border controls, although some of those other issues are related to the broader challenges facing the Schengen system, for example changes to the Borders Code as regards border surveillance, and rules on responses to future public health crises such as the covid pandemic.

This report looks in more detail at the changes on four key issues: the instrumentalization of migrants; the definition of border checks; fast-track returns to other Member States, and the reintroduction of border controls.

Cases of instrumentalisation

The ‘[i]nstrumentalisation’ of migrants is defined (by cross-reference to a recently agreed asylum law, discussed here) as ‘where a third country or a hostile non-state actor encourages or facilitates the movement of third country nationals or stateless persons to the external borders or to a Member State, with the aim of destabilising the Union or a Member State, and where such actions are liable to put at risk essential functions of a Member State, including the maintenance of law and order or the safeguard of its national security’.

The preamble clarifies the definition further, stating that ‘[s]ituations in which non-state actors are involved in organised crime, in particular smuggling, should not be considered as instrumentalisation of migrants when there is no aim to destabilise the Union or a Member State’. Furthermore, ‘[h]umanitarian assistance should not be considered as instrumentalisation of migrants when there is no aim to destabilise the Union or a Member State’.

The amendments provide that ‘in particular’ Member States can temporarily close border crossing points or limit their opening hours in instrumentalisation cases. However, any limitations must be ‘proportionate’, and must take ‘full account of the rights of’ those with free movement rights, non-EU citizens with a legal right to reside, and non-EU citizens ‘seeking international protection’. The revised code will also have a new rule, subject to the same guarantees, that ‘Member States may, where a large number of migrants attempt to cross the external border in an unauthorised manner, en masse and using force, take the necessary measures to preserve security, law and order’.

Defining internal border checks

The agreed amendments, taking account of the case law, take a deferential approach to the exercise of police powers on the territory, including in border zones and for the purpose of immigration control, along with checks on public health grounds.

Fast-track returns

A new clause will provide for the fast-track transfer to another Member State of non-EU citizens ‘apprehended in border areas’, where the non-EU citizen was ‘apprehended during checks involving the competent authorities of both Member States within the framework of bilateral cooperation’, which may include ‘joint police patrols’; and ‘there are clear indications that [they have] arrived directly from another Member State’, as further explained, if it is ‘established that the third‑country national has no right to stay on the territory of the Member State in which he or she has arrived’.

However, this process cannot be applied to people with international protection, or to applicants for asylum; according to the preamble, the Dublin rules (which are themselves about to be revised) ‘should apply’ to asylum seekers.

Where the new fast-track transfer rules will apply, as a derogation from the usual obligation in the EU Returns Directive (the law which sets out general rules on irregular migration) to issue a return decision, the Member State which stopped the person may transfer them immediately to the Member State from which they arrived ‘in accordance with’ a process set out in a new Annex to the Code. This Annex will require the authorities to give reasons for the transfer by means of a standard form handed to the person concerned. There will be a right to appeal the transfer, but it will not have suspensive effect, and in the meantime the person concerned will be transferred within 24 hours.

The reintroduction of internal border checks and controls

There will be a series of amendments to the existing rules on reimposing internal border checks in the Borders Code. In particular, the rules on reintroducing border controls in cases requiring ‘immediate action’ will be amended, referring instead to ‘unforeseeable’ events, and allowing border controls to be reintroduced for a one-month period with extensions up to three months, instead of the current ten days with extensions up to two months.

In ‘foreseeable’ cases, national decisions to reintroduce internal border checks could, under the agreed text, be renewed and apply for a maximum period of two years, rather than six months at present – although in a ‘major exceptional situation’, a Member State could in future also apply two further extensions of six months each.

In the event of a public health crisis, internal border checks can be indefinitely renewed for six-month periods. The current ‘exceptional’ procedure for border checks for up to two years due to an EU-wide crisis will remain without amendment.

Assessment and conclusions

The report notes that in part the changes will entrench the status quo, because they either take account of the case law issued beforehand or while the proposal was under negotiation (on checks on the territory and on public health).

But some of the agreed amendments are genuinely new, most notably fast-track returns of irregular migrants (although not asylum-seekers) between Member States, longer renewals of national internal border control, potentially indefinite border controls on public health grounds (if authorised by the Council), and the instrumentalisation of migration. In effect these amendments circumvent CJEU case law as regards the application of the Returns Directive when border controls are reintroduced (for instance, the recent ADDE judgment), which as it stands prevents the instant return of non-EU citizens to other Member States.

As for the new provisions on instrumentalization, could they circumvent the case law requiring the application of asylum law in such cases? At first sight, the prospect of closing border posts might evade the obligation to consider asylum applications, by making them impossible to  lodge. However, applications might still be lodged by those entering illegally, and in any event the prospect of closing border posts will be explicitly subject to a requirement to take full account of the rights of asylum-seekers. Member States’ power to take ‘necessary measures’ to respond to entry by force will also be subject to this requirement. Also, the entire Borders Code is ‘without prejudice’ to the rights of refugees and asylum-seekers. It requires Member States to act in ‘full compliance’ with the EU Charter, the Refugee Convention, and ‘obligations related to access to international protection […] in particular […] non-refoulement’ in both cases. Therefore there is no plausible argument that the new provisions will legalise illegal ‘push-backs’ by Member States (see both ECHR case law and CJEU case law on this). 

The changes to the rules, for instance allowing for longer periods of legally authorised reintroduction of border controls, raise the question of how limited these controls will be in practice. There are no specific benchmarks available for the abolition of reintroduced border controls, and even if there were it is likely that such abolition – like the extension of the Schengen zone itself – would be determined by political rather than legal factors; it might be more difficult politically to abolish internal border checks the longer they have been applied. While there are legal constraints on the maximum time limit of the reintroduction of those controls, as recently emphasised by the CJEU, it might be wondered – in light, for instance, of frequent allegations of illegal push-backs at the external borders, and the current apparent practice of exceeding the current time limits anyway – whether Member States are sufficiently concerned to observe the rule of law in this field.

Nevertheless, the report concludes that it might be useful to attempt to introduce such benchmarks, at least politically, if the intent is to give an impetus to the aim of ending internal border checks across the Schengen area. It also makes the case for the Commission to produce guidance to ensure that the application of the rules on external border controls are consistent with human rights and asylum law obligations, taking account of case law of the CJEU and the European Court of Human Rights.

The new Screening Regulation – part 5 of the analysis of new EU asylum laws

 

Professor Steve Peers, Royal Holloway University of London

Photo credit: Rock Cohen, via Wikimedia Commons

(Amended on 14 May and 10 and 18 June 2024 - changed sentences marked with an asterisk)

Just before Christmas, the European Parliament and the Council (the EU body consisting of Member States’ ministers) reached a deal on five key pieces of EU asylum legislation, concerning asylum procedures, the ‘Dublin’ system on responsibility for asylum applications (also known as the 'Asylum and Migration Management Regulation', or AMMR), the ‘Eurodac’ database supporting the Dublin system, screening of migrants/asylum seekers, and derogations in the event of crises. These five laws joined the previously agreed revised laws on qualification of refugees and people with subsidiary protection, reception conditions for asylum-seekers, and resettlement of refugees from outside the EU. Taken together, all these laws are a ‘package’ of new or revised EU asylum laws.* All the legislation was formally adopted on 14 May 2024, and published in the EU Official Journal on 22 May 2024.*

I’ve looked at all the new legislation on this blog in a series of blog posts, which are based on a forthcoming article.* This is the fifth post in the series, on the new Regulation on screening of migrants (mostly) at the external borders. The other blog posts in the series concern the new qualification Regulation (part 1), the revised reception conditions Directive (part 2), the new Regulation on resettlement of refugees (part 3), the revised Regulation on Eurodac (part 4), the revised Dublin rules on responsibility for asylum-seekers/AMMR (Part 6), the Regulation on asylum procedures (part 7), and the crisis Regulation plus general comments (Part 8).*

The 2024 asylum package joins the previous Regulation revising the powers of the EU asylum agency, which was separated from the package and adopted already in 2021.* (On EU asylum law generally, see my asylum law chapter in the latest edition of EU Justice and Home Affairs Law).

The Screening regulation: background

There have been two previous ‘phases’ in development of the Common European Asylum System: a first phase of laws mainly adopted between 2003 and 2005, and a second phase of laws mainly adopted between 2011 and 2013. The 2024 package is in effect a third phase, although for some reason the EU avoids calling it that.*

However, unlike most of the 2024 package of legislation, the Screening Regulation is entirely new – although to some extent it may provide a legal basis for things that were already going on in practice before its adoption. So unlike most of the other laws in the asylum package, there is no current version of the law to compare the new version to – and therefore no prior CJEU case law to consider either.

Having said that, the Screening Regulation amended a number of other EU measures, to ensure their consistency with it, namely the Regulations on: the Visa Information System; the entry-exit system; ETIAS (the travel authorisation system); and interoperability of databases.* Furthermore, a parallel Regulation amended two EU criminal law measures to ensure that they are also consistent with the main Screening Regulation.*

Why two parallel Regulations? Because the Screening Regulation, unlike the rest of the package of EU asylum law measures, is technically a law on external borders, not asylum. As such, it ‘builds upon the Schengen acquis’, and so is applicable in principle to the Schengen associates (Norway, Iceland, Switzerland and Liechtenstein) too.* Ireland must opt out (as it does not participate in Schengen) and Denmark is formally excluded (although it may apply the Regulation as a matter of national law). In contrast, the parallel amendment to EU criminal law is only relevant to Member States (but again, there will be an Irish and Danish opt out from it).

In this context, the preamble to the Regulation makes special provision for Cyprus, which has not yet fully applied Schengen; that country must apply the Regulation to those crossing the line separating the areas controlled by the Cypriot government and the Turkish Cypriot administration, even though it is not legally an international border from the perspective of EU law.  As for Denmark and Schengen associates, the preamble states that for them, references to the EU’s reception conditions Directive in the Screening Regulation should be understood as references to the relevant national law.   

As with all the new EU asylum measures, each must be seen in the broader context of all the others – which I discuss over the course of this series of blog posts.* Furthermore, the new Screening Regulation has links with the Schengen Borders Code, the main law governing crossing of external EU borders – although the Regulation did not formally amend the Code.* It will also link with (but again, not amend) the EU’s Returns Directive.

The legislative process leading to the agreed text of the screening Regulation started with the Commission proposal in 2020, as part of the attempt to ‘relaunch’ the process of amending EU asylum law, started back in 2016.* The proposal was subsequently negotiated between EU governments (the Council) and then between the Council and the European Parliament. But this blog post will look only at the final text, leaving aside the politics of the negotiations.

Like most of the other measures in the asylum package, the application date of the Screening Regulation will be about two years after adoption - namely 12 June 2026.* However, the provisions on queries of other EU information systems will only start to apply once those information systems enter into operation.

Scope

The Regulation applies to four categories of people, namely those who: 

without fulfilling the entry conditions [in the Schengen Borders Code], have crossed the external border in an unauthorised manner, have applied for international protection during border checks, or have been disembarked after a search and rescue operation

and of

third-country nationals illegally staying within the territory of the Member States where there is no indication that those third-country nationals have been subject to controls at external borders, before they are referred to the appropriate procedure.

The Regulation distinguishes between the first three categories, who are all connected with the external borders, and the fourth category (illegal staying where is there is no indication of having been controlled at external borders). For simplicity’s sake, this blog post refers to the first three categories as ‘external cases’, and the fourth category as ‘internal cases’. Both the first and third groups must be screened regardless of whether they apply for asylum or not.

Member States ‘may refrain’ from screening the fourth category of people (on the territory, having entered without authorisation), if they send the non-EU citizen back, ‘immediately after apprehension, to another Member State under bilateral agreements or arrangements or under bilateral cooperation frameworks.’ In that event, the other Member State must apply a screening process.

The Screening Process

For external borders cases, screening must be ‘carried out without delay’, and in any event completed within seven days of apprehension, disembarkation, or presentation at the border. For internal cases, the deadline is three days. Screening must end if the person concerned is authorised to enter the territory. Screening may end if the person concerned ‘leaves the territory of the Member States, for their country of origin or country of residence or for another third country’ to which they voluntarily decided to return to and were accepted by. In any case, screening ends once the deadline to complete it is reached.

Screening must take place at an ‘adequate and appropriate’ location decided by Member States; for external cases, that location should be ‘generally situated at or in proximity to the external borders’, although it could be at ‘other locations within the territory’. It must entail (referring in part to checks under other EU laws): checks on health, vulnerability, and identity; registration of biometric data ‘to the extent that it has not yet occurred’; a security check; and filling out a screening form.

For those who have made an asylum application, the registration of that application is governed by the asylum procedures Regulation. The preamble to the Screening Regulation explicitly states that an asylum application can be made during the screening process. Furthermore, the Screening Regulation is ‘without prejudice to’ the Dublin rules; and it ‘could be followed by relocation’ (ie movement to a Member State not responsible for the application) under the Dublin rules ‘or another existing solidarity mechanism’.

Member States are obliged to inform the persons being screened about the screening process itself, as well as asylum law and returns law, the Borders Code, national immigration law, the GDPR, and any prospect of relocation. Otherwise, there is no explicit reference to procedural rights. Conversely, the people being screened have procedural obligations: they must ‘remain available to the screening authorities’ and provide both specified personal data and biometric data as set out in the Eurodac Regulation. Finally, after screening ends, the person concerned should be referred to the appropriate procedure – either the asylum process or the returns process.

Treatment During Screening

As regards immigration law status during the screening process, external cases must not be authorised to enter the territory of the Member States, even though the screening might be carried out on the territory de facto. This is obviously a legal fiction, which is exacerbated by the prospect (under the procedures Regulation) of continuing that legal fiction under the ‘borders procedure’ for up to 12 weeks.

Moreover, Member States must provide in their national law that persons being screened ‘remain available to the authorities carrying out the screening for the duration of the screening, to prevent any risk of absconding and potential threats to internal security resulting from such absconding.’ This wording looks like a euphemism for detention, which the Regulation goes on to refer to more explicitly – providing that where the person being screened has not applied for asylum, the rules on detention in the Returns Directive apply.

For those who have applied for asylum, the reception conditions Directive applies to the extent set out in it. This cross-reference is potentially awkward because that Directive applies to those ‘allowed to remain on the territory’ with that status, whereas the Screening Regulation decrees that the people covered by it are not legally on the territory. Logically the reception conditions Directive must apply despite the non-entry rule of the Screening Regulation, otherwise that Regulation’s references to that Directive applying would be meaningless (the preamble to the Regulation also says that the detention rules in the reception conditions Directive ‘should apply’ to asylum seekers covered by the Regulation). Screening is not as such a ground for detention in the exhaustive list of grounds set out in the reception conditions Directive – so Member States will have to find some other ground for it from that list. The preamble to the Regulation sets out general rules on limits to detention, borrowing some language from the reception conditions directive.

As for other aspects of treatment, the Screening Regulation states that Member States ‘shall ensure that all persons subject to the screening are accorded a standard of living which guarantees their subsistence, protects their physical and mental health and respects their rights under the Charter [of Fundamental Rights].’ For asylum-seekers, this overlaps with the more detailed rules in the reception conditions Directive, but for non-asylum seekers, it in principle goes further than the Returns Directive – although the case law on that Directive has required some minimum treatment of people covered by it. Of course, for many people subject to screening, it will be the provisions on detention conditions under those two Directives which will be relevant in practice. There is a more specific provision on health care, stating that those being screened ‘shall have access to emergency health care and essential treatment of illness.’

The Regulation includes specific provisions on minors. The best interests of the child must always be paramount; the minor must be accompanied by an adult family member, if present, during the screening; and Member States must ensure the involvement of a representative for unaccompanied minors (overlapping with the relevant provisions of the reception conditions Directive).

Finally, as for contact with the outside world, ‘[o]rganisations and persons providing advice and counselling shall have effective access to third-country nationals during the screening’, although Member States may limit that access under national law where the limit is ‘objectively necessary for the security, public order or administrative management of a border crossing point or of a facility where the screening is carried out, provided that such access is not severely restricted or rendered impossible’. Presumably such access can help check that the rules on treatment are being applied, including possible challenges to detention and offering advice as regards subsequent asylum or returns procedures, or potential challenges to screening as discussed above.

Human Rights Monitoring

The Regulation sets out an overarching obligation to comply with human rights obligations, including the principle of non-refoulement (not sending a migrant to an unsafe country), as well as a requirement to have an independent human rights monitoring mechanism, which is specified in some detail. Member States must: ‘investigate allegations of failure of respect for fundamental rights’ as regards screening; ensure civil or criminal liability under national law ‘in cases of failure to respect or to enforce fundamental rights’; and create an independent mechanism to monitor human rights compliance during the screening, ensuring that allegations of human rights breaches are triggered or dealt with effectively, with ‘adequate safeguards’ to ensure its independence. The preamble points out that judicial review is not enough to meet these standards. (Also, these rules will apply to monitoring the borders procedure in the procedures Regulation)

Assessment

To what extent has this Regulation ensured a balance between migration control and human rights? It does aim towards a greater degree of migration control by imposing new legal obligations as regards many asylum seekers; but the key point as regards their rights is that the Regulation provides for a filtering process, not a final decision. In other words, the screening process does not entail in itself a decision on the merits or admissibility of an asylum claim, or a return decision. Whilst it is based on a legal fiction of non-entry, that process is strictly and absolutely limited in time, with no prospect of extending the short screening period even as a derogation under the Exceptions Regulation. (In contrast, the border procedure under the procedures Regulation lasts for longer, and can be extended in exceptional cases). And the legal fiction does not in any event mean that no law applies at all to the persons concerned; obviously at the very least, the screening Regulation itself applies, as do other EU laws which it makes applicable. (So does the ECHR: see Amuur v France) For instance, the Regulation refers to detention on the basis of the returns and reception conditions Directives, and although the lack of authorisation to enter means that the right to remain on the territory as an asylum seeker is not triggered as such, nevertheless the Regulation precludes Member States taking return decisions to remove asylum seekers, as it only provides for a filtering process.

Despite the absence of any express procedural rights in the Regulation, it is arguable that in light of the right to effective remedies and access to court set out in Article 47 of the Charter, it should at least be possible to challenge the application of the screening procedure on the basis that (for example) there is no legal ground for the screening at all, or that the screening has exceeded its permitted duration. In any event, the absence of express procedural rights should be seen in the context of the screening process not determining the merits of an asylum application.

The drafters of the Regulation chose instead to focus on the prospect of non-judicial processes to protect human rights in the context of the screening process. While non-judicial mechanisms of course play an important role in protection of human rights in general, it is useful if parallel judicial processes can be relied upon too. And one area where the Regulation should have explicitly provided for both judicial and non-judicial mechanisms is pushbacks from the territory – illegal not only under human rights law but also under EU law, as recently confirmed by the CJEU.

 

Responsibility in Joint Returns after WS and Others v Frontex: Letting the Active By-Stander Off the Hook

 

Melanie Fink and Jorrit J Rijpma

Melanie Fink is APART-GSK Fellow of the Austrian Academy of Sciences, Central European University and Assistant Professor, Europa Institute, Leiden University

Jorrit Rijpma is Professor of EU law, Europa Institute, Leiden University

Photo credit: Влада на Република Северна Македонија, via Wikimedia Commons

See also analysis of the human rights aspects of the judgment, by Francesca Romana Partipilo 

On 6 September 2023 the General Court delivered its long-awaited ruling in WS and others v Frontex. In a short and matter-of-fact judgment, it dismissed an action for damages by a Syrian refugee family against the European Border and Coast Guard Agency (Frontex). The family, escaping Aleppo at the height of the Syrian war in 2016, was returned to Turkey just days after their arrival in violation of the principle of non-refoulement. Their return was carried out as a joint return operation between Greece and Frontex. With the action brought before the General Court, they sought compensation from Frontex for its role in the violation of the principle of non-refoulement, as well as their degrading treatment in the return process.

After the many reports of fundamental rights violations at the external borders, including pushbacks, this was the first case in which Frontex came under judicial scrutiny for its role in potential violations. Earlier, a damning OLAF-report, demonstrating that the Agency had turned a blind eye to pushbacks in the course of operations it coordinated, had led to the resignation of its Executive Director.

Since its establishment, successive legislative amendments have consistently increased Frontex’s powers, short of transferring command and control over border guards and return officers. Yet, Frontex has always maintained that it cannot be held responsible for violations of fundamental rights as it merely acts as coordinator and facilitator in joint (return) operations. Wrongdoings in the context of joint operations, so Frontex, would be exclusively on the Member State in charge.

In an unsatisfactory judgment that fails to do justice to the plight of a refugee family that turned to the European Union for protection, the General Court now seems to confirm that stance. Doing so, it failed to acknowledge the role and obligations of Frontex during joint operations. Adopting an unreasonably and unnecessarily high threshold for the establishment of the causal link requirement, it also excludes almost any prospect of Frontex being accountable for any breaches of its obligations. After a brief overview of the judgment, we will discuss each of these points in turn. We refer to Regulation 2016/1624, which governed the activities of Frontex at the time of the return, even though it has been replaced with Regulation 2019/1896 in the meantime. However, the relevant provisions have not substantially changed.

 

The judgment

On admissibility, the Court rejected two arguments advanced by Frontex. First, it did not consider that it was called upon to make general statements of principle by ruling on the applicants’ damages claim. Second, it did not accept the argument that the applicants were barred from bringing an action for damages, as they could have brough an action for annulment against the letter of the Agency’s fundamental rights officers dismissing their complaint under the individual complaints’ mechanism. The Court held that these two actions do not preclude each other as they pursue different objectives, but explicitly left the question whether the actions of the Agency’s Fundamental Rights Officer within the framework of that administrative procedure constitute challengeable acts under Article 263 TFEU, which if they are would subject this procedure to judicial review by the Court.

On substance, non-contractual liability arises when three cumulative conditions are met: a sufficiently serious breach of a rule of EU law conferring rights on individuals, damage, and a causal link between the unlawful conduct and the damage. Reversing the order in which it assessed the conditions, the General Court dismissed the action based solely on the absence of a sufficiently direct causal link between the conduct of the Agency and the damage that was invoked. At the outset it had already recalled that the unlawful conduct would need to be the determining cause of the damage. It considered that the applicants wrongly departed from the presumption that without the alleged conduct by Frontex they would not have been returned. Here the General Court repeats Frontex’s mantra that it only provides technical and financial support. Most importantly, it emphasizes Frontex’s lack of competence to adopt a return decision or decide applications for international protection, leaving any liability with the responsible Member State.

The General Court skipped the question whether the return of the applicants and their treatment during the return procedure constituted a violation of EU law altogether. Although this may be interpreted as a sign of judicial economy, it is also a way to avoid having to pronounce itself on the behaviour of the Member State in question. In addition, the Court may have otherwise been required it to address the limits of its own jurisdiction under Article 276 TFEU, which precludes it from assessing the validity or proportionality of Member States’ law enforcement authorities.

 

The Role, Obligations, and Responsibility of Frontex

By virtue of Article 28 Regulation 2016/1624, Frontex is prohibited from ‘entering into the merits of return decisions’ because these ‘remain the sole responsibility of the Member States’. The Court rightly held that Frontex cannot be responsible for any potential unlawfulness of the return decision itself. As with any other national administrative decision, it would be for the Member State authorities to ensure its lawfulness.

Aside from the question whether a return decision was even taken under the Return Directive, and whether this decision was then lawful, the applicants’ allegations in the case go well beyond the decision itself. Frontex’s alleged wrongdoing concerns the implementation of the decision, despite clear indications of a risk of refoulement, and the degrading treatment of the applicants as the expulsion was carried out. This phase of the return process, i.e. the implementation of return decisions in the form of joint return operations, is a core competence of Frontex, which by virtue of Article 28(1) Regulation 2016/1624 renders ‘the necessary assistance’ to return operations and ensures their ‘coordination or […] organisation’.

This coordinating role comes with obligations. Concretely, Article 28(3) Regulation 2016/1624 explicitly states that ‘Agency shall ensure that the respect for fundamental rights, the principle of non-refoulement, and the proportionate use of means of constraints are guaranteed during the entire return operation’ (see also generally Article 34 Regulation 2016/1624). In addition, as an EU body, Frontex is bound by the EU Charter of Fundamental Rights, including the absolute prohibitions of refoulement in Article 19 and of inhuman or degrading treatment in Article 4. These rights are widely understood under European human rights law to include positive obligations that require authorities to actively ensure the protection of a right, for example by taking practical steps to protect a person against interferences by others. Frontex has a whole toolbox of means available to meet these obligations, including reporting and communication duties. As a last resort, Article 25(4) Regulation 2016/1624 requires the agency to withdraw, should violations of fundamental rights or international protection obligations occur that are serious or likely to persist.

Frontex conducts joint return operations together with the Member States. However, if it violates its own obligations under EU law, it bears responsibility that may be invoked through an action for damages. This responsibility is independent from any possible responsibility of the Member State who in turn bears responsibility for its own failures in the process (see also here). Not separating the return decision from its implementation, the Court failed to acknowledge the role of Frontex in the latter. In addition, shielding the agency from responsibility for a violation of its obligations in joint return operations emasculates these provisions, which also negatively affects the credibility of the EU as a rule of law advocate.

 

Causation and Joint Liability

An important complicating factor in this case, is the interplay between the actions of Frontex and the host Member State. Situations where more than one actor is involved in causing harm are not uncommon, but incredibly complex when it comes to allocating legal responsibility (see also here).

First, it might be unclear who is considered the ‘author’ of a violation, in other words, to whom the unlawful conduct is attributable. As a national administrative decision, the return decision is clearly attributable to the host state. Things are more complicated at the implementation level, where the actions of the host state and Frontex are more intertwined. However, since the Court did not separate the return decision from its implementation, the question of attribution played no role in the case.

The second difficulty concerns causation, that is the link between the unlawful conduct and the damage. The Court denies the existence of a sufficiently direct causal link between Frontex’s conduct and the harm complained of because Frontex lacks the competence to interfere with the return decision or grant international protection. In other words, in the Court’s view, the return decision is the cause for the applicants’ harm, not Frontex’s conduct. Underlying this argument seems to be an assumption that ‘exclusive’ causation might be required for liability to arise. This is also the view the General Court defended in the recent case Kočner v EUROPOL, a case currently under appeal with Advocate General Rantos suggesting the Court of Justice take a less restrictive approach to the causation requirement.

In the past, there have been cases in which the Court seemed accepting of the idea that the existence of an additional determining causes for a damage does not necessarily bar a finding of liability. In light of the coordinating nature of Frontex’s tasks, allegations of wrongdoing will usually, if not always, go hand in hand with (potential) wrongdoing by one or more Member States. If Frontex is not accountable simply because a Member State may have acted unlawfully too, this appears to exclude any reasonable prospect of Frontex being held accountable for breaches of its obligations. In fact, it would seem to stand in the way of joint liability between the Union and a Member State altogether, which has been recognised by the Court as early as 1967 and is a necessary means to ensure accountability in the EU’s multi-level administration (for more detail see here).

 

Conclusion

The Court, in limiting itself to an assessment of causality, failed to acknowledge a clear violation of one of the core tenants of EU refugee law, the prohibition of refoulement, as well as a range of safeguards laid down in EU secondary legislation. Frontex was present during this violation, and rather than intervened, contributed to it. All of this would not in itself have resulted in Frontex being held liable, but the argument that it is excluded because of a lack of competence regarding the decisions on return and international protection is flawed and lays bare a misconception of the practical reality of joint law enforcement operations as well as the role and obligations of Frontex under EU law in that context.

This judgment begs the question what Frontex’s fundamental rights obligations are worth in the absence of a meaningful way to enforce them. Even if a Member State could, at least in theory, be held responsible before the national judge, and ultimately before the ECtHR, that should not mean that the exercise of public power by a Union body should be allowed to escape judicial review. In a system of shared administration, which the management of the shared external borders has become, joint responsibility carries a need for joint liability.

This case shows how the ‘complete system of remedies’ fails to provide effective judicial control of public power in the EU's area of freedom, security and justice, which is characterised by integration through operational cooperation rather than law. Enforcement powers remain the Member States’ exclusive prerogative in name, but in practice are increasingly exercised jointly by the Member States and the EU. This judgment could have provided a welcome correction to this constitutional oversight. If upheld on appeal, it will reinforce the need for the long overdue accession of the EU to the ECHR.

 

Giving Back Control: British travel to the EU after Brexit

 

Professor Steve Peers, University of Essex*

*linked to research on the upcoming 5^th edition of EU Justice and Home Affairs Law

Photo credit: John Fielding, via Wikimedia Commons

 

One of the most Blindingly Obvious Things in the history of Blindingly Obvious Things is that one consequence of the UK leaving the EU is that travel to and stay in the EU by British citizens is now different – the obvious corollary of travel to and stay in the UK by EU citizens being different, as the Leave campaign specifically demanded.

In light of current disputes about delays crossing the border, what exactly does that mean in practice? The following blog post addresses the issues in Q and A format.

What were the previous rules?

UK citizens going to the rest of the EU, and EU citizens coming from the rest of the EU, were covered by EU free movement law – see Articles 4 and 5 of the EU citizens’ free movement Directive. This allowed travel across the border in both directions on the basis of showing an identity card or passport (although non-EU family members, while benefiting from free movement law generally, had to get a visa in some circumstances). 

This overlapped with the Schengen rules on external borders – dating back originally to the Schengen Convention of 1990 agreed between several Member States, applied from 1995, integrated into EU law in 1999 when the Treaty of Amsterdam came into force, and currently set out in the Schengen Borders Code, 2016 version. (Amendments to the Code have since been adopted or proposed).

The Schengen rules do not, as some may think, limit themselves to abolishing internal border controls between Schengen States. Because the abolition of such controls means that whoever enters (say) Italy from outside Schengen could travel to Austria without being checked at the border between those countries, there are common rules on external border controls, too. (There are also common rules on short-stay visas, criminal law and police cooperation, and a joint database, the Schengen Information System)

For EU citizens crossing the external borders (UK/France, for instance), the Borders Code requires a simplified check on their identity and the validity of the document (see Article 8(2)), including a check on security databases (see the 2017 amendment to the code, which requires further checks: the application of this delayed some travellers a few summers back).

The UK’s departure from the EU not only took it out of the scope of EU free movement law. Brexit also necessarily means that the UK is now treated as a non-EU country under the Schengen rules – because that is the status that the UK wanted to have. More on what that means in concrete terms in a moment.

Which countries apply these rules?

Free movement law applies not only to EU Member States, but also to Norway, Iceland, and Liechtenstein, which have signed up to those rules as part of the European Economic Area treaty, and to Switzerland, which signed up to similar (not currently identical) rules as part of a package of treaties with the EU.

The scope of the Schengen rules is more complicated. First of all, the immigration law part of the Schengen rules does not apply to Ireland (apart from rules on carrier sanctions and penalisation of smuggling), although Ireland has signed up to the criminal law parts of Schengen. Ireland has opted out of the immigration law part of the Schengen rules due to the Common Travel Area with the UK, as provided for in a Protocol attached to the EU Treaties. The Protocol still applies despite Brexit, and there is no sign that the Irish government is interested in fully joining Schengen, probably because that would mean border checks on the land border with Northern Ireland (unless Northern Ireland also joined Schengen, but it is hard to imagine any British government agreeing to that).

Among the other EU Member States, Schengen rules do not yet fully apply to Romania, Bulgaria, Croatia and Cyprus – although the external borders parts do apply already (see the accession treaty with Croatia, for example). Extending Schengen fully needs unanimous agreement of the existing Schengen States, which has been slow in coming (although a decision on extension to Croatia could be adopted soon). And Denmark applies Schengen rules, but in a legally different way from other EU Member States.

As for non-EU countries, Norway, Iceland, Liechtenstein and Switzerland have agreed treaties associating themselves to the Schengen rules. As with free movement law, this association is reciprocal – ie Norway or Switzerland must also let EU citizens visit and stay on the basis of free movement law, and drop border controls with Schengen States, not only the other way around.

Finally, the UK and the EU are negotiating a form of association of Gibraltar to Schengen. It remains to be seen whether those talks will be successful, and if so what will be agreed.

Is stamping British passports a fetish of Anglophobic, Brexit-hating, surrender-loving, French freaks?

No.

Stamping the passports of non-nationals is a normal part of immigration control worldwide. For France, and other Schengen countries, it’s an obligation in the Schengen Borders Code, Article 11:

1.   The travel documents of third-country nationals shall be systematically stamped on entry and exit. In particular an entry or exit stamp shall be affixed to:

(a)	the documents, bearing a valid visa, enabling third-country nationals to cross the border;
(b)	the documents enabling third-country nationals to whom a visa is issued at the border by a Member State to cross the border;
(c)	the documents enabling third-country nationals not subject to a visa requirement to cross the border.

 

(There are exceptions, but they are fairly limited). The UK having chosen to become a non-EU country without free movement, British citizens are necessarily covered by the stamping obligations after the end of the Brexit transition period. The code makes clear (see Article 8(3)) what the point of the stamp is – to enable checks as to whether the passport holder has not overstayed the permitted period of stay.

Some Brexit supporters had argued for a continuing close relationship with the EU post-Brexit, in which case stamping of documents would not have applied (see Article 2(6) of the Code, which defines ‘third-country nationals’ in the Code as not applying to citizens of countries with a free movement agreement with the EU). But those Brexit supporters who wanted to end free movement, and who argued explicitly against treating EU citizens coming to the UK better than non-EU citizens coming to the UK, won that argument. So it is hypocritical for them to complain about the obvious reciprocal consequence: that the EU is now treating UK citizens coming to the EU the same as other non-EU citizens coming there.

What are the other main features of the Schengen borders rules?

Leaving aside the bits on abolishing internal border checks, which are subject to many possible exceptions (Articles 22-35), the main points, besides stamping of documents, are:

-          the code is ‘without prejudice’ to free movement law, as well as the ‘rights of refugees and persons requesting international protection, in particular as regards non-refoulement’

-          Member States must comply with fundamental rights when they implement it

-          external borders must be crossed at official places and times, with penalties if this is breached (again ‘without prejudice’ to ‘international protection obligations’; note that Article 31 of the Refugee Convention provides for refugees crossing borders without authorisation, subject to certain conditions)

-          entry conditions: 90 out of 180 days stay in the Schengen area as a whole; justifying the purpose of the intended stay and having sufficient subsistence; a visa if EU law requires it; not being the subject of an alert in the Schengen Information System; not being a threat to public policy etc

-          details of the border checks for EU and non-EU citizens (which are more stringent)

-          separate lanes for EU and non-EU citizens

-          border surveillance to try to stop unauthorised entry

-          refusal of entry if conditions are not met; there is a right to appeal a refusal

-          general rules on cooperation between Member States

There are limited exceptions to some of these rules. There’s also a rule on sufficient resources for border checks (Article 15), which could be relevant to France/Dover:

Member States shall deploy appropriate staff and resources in sufficient numbers to carry out border control at the external borders, in accordance with Articles 7 to 14, in such a way as to ensure an efficient, high and uniform level of control at their external borders.

Does Brexit mean that UK citizens can’t go to the EU courts?

No. Access to the EU courts is not limited to EU citizens. There are many cases brought by non-EU citizens about the application of EU immigration and asylum law that go to the CJEU. Indeed, I’ve been slogging through them for the latest edition of my book. You can find a compilation of the case law on the 2016 version of the Borders code here.

As with other EU law, the main way that an argument about EU immigration and asylum law gets to the EU courts is by disputing the application of that law in the national courts (ie, challenging a German border guard’s refusal of entry in a German court). Any national court or tribunal may then send questions about the interpretation or validity of that EU law to the CJEU.

It’s also possible to sue the EU institutions or agencies (but not the Member States) directly in the EU General Court. Unlike (say) competition law or trade mark law, this has limited relevance to immigration or asylum law, because it’s mainly national authorities that implement this field of EU law on the ground. There are cases pending against Frontex, the EU borders agency – but these are about its alleged role in pushback of migrants at sea, not ordinary border checks.

Does the EU have immigration databases?

Hell, yes. The EU is the poster child of the border security/industrial complex. In particular it currently has the Schengen Information System (updated in 2018, but the update is not applied yet). The SIS contains not only security alerts (such as being subject to a European Arrest Warrant) which apply to and are checked for everyone, but also immigration alerts, which apply only to non-EU citizens, now including Brits (but not only Brits!). These are issued following entry bans for breaches of immigration law or criminal law.

The 2018 changes will link this system to the issue of entry bans under the Returns Directive, which regulates the expulsion, detention etc of non-EU citizens who are not authorised to stay, now including Brits (but not only Brits!!).

In the next year or two the EU is likely to roll out its entry-exit system (which will keep a record of non-EU citizens coming in and out of the EU) and its travel authorisation system, ETIAS (like the US ESTA, a requirement for non-visa non-EU nationals to get an advance authorisation to travel – similar to the idea of a visa but not technically a visa; it will be simpler and cheaper to get and apply for longer).  Both will apply to Brits, but are not being rolled out – despite absurd suggestions to the contrary – solely for Brits(!!!), or because of Brexit. The entry-exit system will in principle replace stamping of documents at the borders by e-gates, but whether that speeds things up or slows things down remains to be seen.

There’s also the Visa Information System (revised in 2021, but the revised system is also not yet applicable). It was originally for applicants for Schengen visas (ie short-term visas), so doesn’t apply to Brits, but the revised version will apply to Brits (but not only to Brits!!!!), as it will cover long-term visa and residence permit holders.

The Eurodac system keeps a record of asylum seekers and irregular border crossers. It is – you guessed it! – being revised and extended, so will also (among other things) apply to those on the territory without authorisation, which could include post-Brexit Brits (but not only Brits!!!!!).

These databases apply to Schengen States only, except Eurodac applies to all Member States and Schengen associates. Ireland participates in SIS criminal law alerts (but not immigration alerts), and the Member States not fully in Schengen have some access to these databases too.

Tl;dr: as with the Hulk, Brits won’t like the EU immigration control system when it’s angry. 

Could the UK reach a special deal with EU on border control?

The EU’s border control law has special rules for European micro-states, and local border traffic with neighbouring Eastern European states, so it’s not impossible. You can see an example of simplified border checks, in spite of underlying immigration control, at the Canada/US border, for instance. But the EU has never before agreed exemptions from its core border control laws with a large non-EU country.

And would the UK – at least under its outgoing and upcoming Prime Minister – be interested anyway? Neither side suggested any such thing during the Brexit talks, and the UK has moved away from the sort of simplification the EU side would ask for, as it is no longer accepting ID cards for EU citizens at the border (there’s an exception for those covered by the withdrawal agreement). It’s very hard to believe that the EU would agree such a thing unilaterally, if at all – and it’s not only Leavers who fantasise that nevertheless it would or should. And then they wake up.

What about Brits living in the EU?

Those Brits in the EU before the end of the Brexit transition period may – like EU citizens in the UK at that point – be covered by the withdrawal agreement, which I annotated here. 

Otherwise Brits in the EU are in the same position as other non-EU citizens living in the EU. If they have a long-term visa or residence permit, they can travel to other Member States for 90/180 days (see Article 21 of the Schengen Convention). They will not be subject to the entry-exit law or the travel authorisation law. They can be subject to immigration alerts in the Schengen Information System, subject to complex rules about consultation between Member States (amended in 2018). When crossing external borders, they are not subject to all the conditions of entry. If they are the family member of an EU citizen who has moved between Member States, they are covered by EU free movement law: the immigration law equivalent of marrying a billionaire. But complicated things could happen if that family relationship breaks down before they obtain their own permanent residence permit.

What about the rest of EU immigration and asylum law?

That’s enough for today. But I summarised the rest of EU immigration and asylum law when the EU Commission tabled its proposed Immigration and Asylum Pact in September 2020 – so see that blog post, with further links. The main development since then is that the EU has adopted a new version of its Blue Card law on admission of highly-skilled non-EU workers.

At least tell me if Brits can apply for asylum in the EU?

Brits are no longer subject to the quasi-ban on EU citizens applying for asylum in other Member States – because they are no longer EU citizens.  But if you believe that, as things stand, Brits are likely to be considered to meet the definition of ‘refugee’ in the Refugee Convention (and EU asylum law) – having a well-founded fear of persecution on grounds of race, religion, nationality, political opinion, or particular social group – then imagine me shaking my head in Godwin’s Law.