Budapest Pride: banned? - Banning Pride as a violation of EU free movement

 

Attila Szabó, LLM in European law, Head of Legal Aid Service, Hungarian Civil Liberties Union

The author would like to thank Steve Peers and Máté Szabó, HCLU's professional director, for their ideas for the text!

 

Photo credit: Tételes istentagadó, via Wikimedia Commons

 

On 5 April, Renáta Uitz published a pinpoint blog post on Verfassungblog about how and why Budapest Pride, which has been held every year for 30 years in Hungary for the equality of LGBTQ+ people, was banned. Uitz describes in detail why this is contrary to Article 2 of the TEU and how this government action violates the European Court of Human Rights case law. The CEU (and Royal Holloway) professor also explains how Hungary got to this point.

 

I would like to add two things to this reflection. One is just brief: Hungarian civil society organisations are committed to organising and hosting Pride and the Mayor of Budapest also supported the organizers, despite the categorical legal ban, on the very same fundamental rights basis that Uitz explains: assembly and thus standing up for LGBTQ+ rights is a fundamental right and can only be restricted if it restricts another fundamental right. The abstract and unscientific paedophile arguments that the Hungarian government is operating with are not. Consequently, the restriction of rights is also contrary to the Hungarian Fundamental Law and goes against the ECtHR case law cited by Uitz.

 

It is another matter that the Hungarian legislator allowed for fines to be imposed on the basis of the face recognition system at banned rallies, such as the Budapest Pride, which was planned to take place despite the ban. The legal issues involved would require a separate blog post. A very careful Hungarian-language piece concludes that "the use of facial recognition technology to detect and punish offences, and thus to restrict assembly and free expression, may be objectionable on a number of points. Although (...) the prohibition rules (Article 5 of the EU AI Regulation) do not directly preclude the non-real-time use of facial recognition technology, there are nevertheless a number of fundamental rights concerns which render the use of the system unlawful. The primary avenue of redress may therefore be to invoke these fundamental rights violations."

 

My argument, which may be news to many, is that the restriction runs counter to one of the EU's four freedoms, the right of free movement and residence. This is what I want to convince readers of.

 

Freedom of movement: limited?

 

In my view, if a non-Hungarian EU citizen can enjoy the right to participate in an LGBTQ+ rights march in every EU country then she can not freely choose this given country.

 

Participation in Pride is a fundamental right that has become part of Atlantic and therefore European culture. All EU citizens have the right to express their support for LGBTQ+ people in any EU Member State within the framework of Pride. If an EU citizen is not allowed to participate in Budapest Pride because it cannot take place, he or she is not free to choose Hungary as a place of residence, as this means that he or she cannot freely exercise in that Member State the fundamental rights that he or she could exercise in another Member State. So, an EU citizen who is committed to LGBTQ+ rights cannot come to Hungary to work or study, because he cannot stand up for the rights he could stand up for anywhere else in any city.

 

One might think that restrictions on freedom of movement are only violated if hard barriers are put in place, but in my view this is not the case. If a Member State introduces an ideological, ideological, theoretical or even religious restriction that alienates other EU citizens from exercising their right to move and reside freely, then that right is infringed. This is particularly so in cases where the exercise of the right in question is consensual in the EU or protected by human rights court decisions and other international law standards. Participation in Pride is such a case.

 

In my view, this conclusion is true in all cases where an EU citizen wishes to stay in Hungary on a long-term basis, but is discouraged from doing so by this kind of unlawful restriction of a fundamental right, and also in cases where an EU citizen wishes to come to Budapest specifically for Pride. The latter case is not unprecedented either, as Budapest Pride in recent years has hosted many EU citizens who would not otherwise have stayed in Hungary for a longer period. So banning Pride is not only an obstacle to the free movement of those who would like to attend such an event but cannot, but also to those who attend a formally illegal event and face sanctions for doing so.

 

Based on Art 6 of Directive 2004/38, free movement also extends to people who make short-term visits, ie to participate in an event or demonstration. The Cowan and Bickel and Franz judgments show that free movement rights in the context of short-term visits go further than entry onto the territory – ie. equal treatment as regards victims’ rights and fair trial rights respectively.

 

Legal bases and arguments

 

The right to freedom of movement is a cornerstone of EU law, enshrined in Article 21 of the Treaty on the Functioning of the European Union (TFEU). Furthermore, the EU Charter of Fundamental Rights reinforces this right, particularly Article 45, which explicitly states that every EU citizen has the right to move and reside freely within the territory of the Member States. Although Article 51 of the Charter says that the Charter itself should only be taken into account in the application of EU law, I am not arguing that the Charter should be taken into account in relation to the right of assembly in the Member States. I say that the Charter should be taken into account in relation to Article 21 TFEU. Indeed, the restriction on the right of assembly restricts this EU right, guaranteed in TFEU Article 21, as follows.

 

The Court of Justice of the European Union (CJEU) plays a vital role in shaping and expanding the interpretation of EU freedom of movement, particularly in light of the EU Charter of Fundamental Rights. The CJEU increasingly considers the Charter when assessing national measures that may restrict freedom of movement. This means that restrictions must not only be justified under EU law but also comply with fundamental rights.

 

Cases involving the rights of LGBTQ+ individuals and their families have highlighted the CJEU's willingness to use the Charter to protect fundamental rights within the context of freedom of movement. The Charter has strengthened the protection of freedom of movement by providing a clear and comprehensive list of fundamental rights. It has also given the CJEU a stronger basis for challenging national measures that are deemed to be incompatible with these rights.

 

The Coman and Hamilton (C-673/16) judgment helps us to understand the context. This case dealt with the refusal of a Romanian authority to recognize the marriage of a Romanian citizen with a US citizen, a same-sex couple, celebrated in Belgium. The CJEU ruled that the term "spouse" in the context of EU freedom of movement includes same-sex spouses. The Court emphasized the need to respect the fundamental rights of EU citizens, including the right to respect for private and family life (Article 7 of the Charter) and the principle of non-discrimination (Article 21 of the Charter). This case significantly expanded the rights of same-sex couples in the EU, ensuring that their family life is protected when exercising freedom of movement.

 

Steve Peers wrote about the case that “[t]he Court added that any measure restricting free movement rights also has to comply with human rights guaranteed by the EU Charter of Rights, which has to be interpreted consistently with the European Convention on Human Rights. According to the case law of the European Court of Human Rights, “the relationship of a homosexual couple may fall within the notion of ‘private life’ and that of ‘family life’ in the same way as the relationship of a heterosexual couple in the same situation.”

 

In another case (C-808/21), the CJEU determined that EU citizenship, established by the Maastricht Treaty and articulated in Article 20 TFEU, implies a right for EU citizens to join political parties in other Member States, despite the absence of explicit provisions. The Court reasoned that denying this right undermines the effective exercise of voting rights in municipal and European Parliament elections, which are expressly granted to EU citizens. The CJEU emphasized the principle of equal treatment under Article 22 TFEU, stating that Member States cannot impose conditions on non-national EU citizens that are not applied to their own nationals. While acknowledging national identity and the Member States' authority over national elections, the Court asserted that EU values, such as democracy and equal treatment, limit national discretion. The Court's rulings enhance EU citizenship's political rights, reinforcing its role as a fundamental status within the EU legal order. Finally, the court linked the right to join political parties to Article 12 of the Charter, which concerns freedom of association.

 

Peers wrote in the Verfassungblog that “remarkably, the Court links the implied political rights of EU citizens in municipal and European Parliament elections not only to the express rights of EU citizenship, but also to democracy and equal treatment, as represented in Article 10 TEU and the EU’s values – putting them at the same rank as national identity. So national identity cannot constitute an exception to democracy or the rule of law et al – but must be reconciled with those values.”

 

I think that the EU understandings, based on the ECtHR rulings, is that restrictions on certain political freedoms are not only a violation of EU law if they discriminate against EU citizens, but also if they restrict those rights without discrimination. For, as I argued above, a Member State that falls below the standards of European fundamental rights is in fact restricting the free movement of its citizens by alienating the citizens of other States from the freedom of movement and residence. Who would like to study in Budapest without being able to choose to participate in the celebration of equality according to sexual orientation? I think it can be argued that far fewer people would do so than if this right were not restricted.

 

The EU’s Fundamental Rights Agency has a more detailed collection of relevant cases.  These cases demonstrate the CJEU's commitment to protecting fundamental rights within the context of freedom of movement. The EU Charter of Fundamental Rights provides a vital framework for the CJEU's interpretation of EU law in this area.

 

Obviously, the facts of these cases are different from those of the Budapest Pride ban, but they follow a similar logical structure: that is, national measures become relevant under EU law when, in the exercise of free movement, they deprive an EU citizen of fundamental rights that would be guaranteed in another Member State.

 

Conclusion: no pasarán

 

I argue that banning Budapest Pride is not only a violation of the Hungarian Fundamental Law, not only a violation of the European Convention on Human Rights, not only a violation of Article 2 of the TEU, but also a violation of the right to freedom of movement and residence. There is the same human rights minimum without which an EU Member State cannot be considered so free, so grounded in European values, that a citizen of any other Member State is free to choose to travel and reside there. Member States must not violate the privacy prohibitions on which freedom of movement and residence are based, and they must not, in my opinion, exclude citizens of other Member States from joining political parties by excluding everyone equally. In the same way, Member States cannot exclude EU citizens from lawful assembly by European legal standards. If they do so, they restrict freedom of assembly.

 

Of course, many things can deter an EU citizen from moving to another Member State: from the climate, to the quality of social and health services, to tax policy. However, fundamental rights deterrence is different from other types of deterrence because of the Charter and the relevance of ECtHR jurisprudence. Denial of fundamental rights, on the other hand, carries a normative message: 'You cannot exercise certain fundamental rights here'. And it is no longer a matter of preference for the person, but of the exercise of his or her citizenship status. It is up to the Court of Justice of the European Union to work out the yardstick for this.

The New EU “Common System for Returns” under the Return Regulation: Evidence-Lacking Lawmaking and Human Rights Concerns

 

Dr. Izabella Majcher, Independent Consultant

Photo credit: Alamy Stock Photo 

On 11 March 2025, the European Commission published a proposal for a Return Regulation to replace the 2008 Return Directive. This new proposal also supersedes the Commission 2018 proposal to recast the Directive (discussed here and here). Negotiations on the 2018 recast proposal spanned several years, incurring significant costs, yet they stalled at the European Parliament level before even reaching interinstitutional negotiations between the Parliament and the Council of the EU. This raises questions about the necessity of launching fresh negotiations, their financial implications, and ultimately whether a comprehensive revision is needed at all. Why will this proposal pass through the negotiations if the previous one got stalled? In addition, opting for a regulation rather than a directive could make negotiations even more complex, as Member States recognize that a regulation’s provisions would be directly applicable in their domestic legal systems.

The underlying aim of the proposal is to increase the number of people returned, as it regrets that only around 20% of those issued a return decision actually leave. The proposal links this objective to the effectiveness of return. This objective is to be achieved through two main contributions of the proposal.

First, the proposal aims to streamline and harmonise return procedures, attributing the inefficiency of returns at the EU level to divergent national approaches. To address this, the Commission proposes shifting from a directive to a regulation and restricting the ability of Member States to provide stronger safeguards at the domestic level. Notably, the proposal criticises “significant room” the Directive leaves for national courts to interpret EU rules and removes the option for “more favourable provisions” option under the Directive (Art. 4). In the effort to create “modern, simplified and common” procedures, there is a risk that essential human rights safeguards will be eroded. It is crucial to remember that return procedures must comply not only with the principle of effectiveness but also with human rights standards. Second, the proposal upgrades “common standards and procedures” of return to a “common system for returns.” Besides the return standards and procedures, the “common system for returns” includes mutual recognition of return decisions (see discussion below), necessary resources and personnel to be ensured by the Member States, cooperation between Member States, support from EU bodies and agencies, and digital management systems. These measures are neither groundbreaking nor novel enough to justify an entirely new legislative text—especially considering the costs of adopting a new law.

The lack of an ex-ante impact assessment makes these questions even more pressing. Citing urgency and existing studies and consultations (most of which have not been made public), the Commission chose not to conduct an impact assessment for this proposal. None of these explanations is convincing, especially in light of the importance of evidence-based law-making. Under its own Better Regulation Guidelines, for every initiative that is likely to have significant economic, environmental or social impacts, the Commission should conduct an impact assessment to demonstrate the added value of the proposed measures and their coherence with key criteria—including impact on fundamental rights.

The proposal introduces some stronger human rights protections compared to the Directive, including the judicial form of appeal (Art. 26(1)), expanded monitoring of forced returns (Art. 15), basic needs provision during removal postponement (Art. 14(6)), and age assessment for children (Art. 19). However, these safeguards are overshadowed by the proposal’s overall coercive approach and several provisions that weaken rights. This analysis examines five key measures that are particularly problematic from a human rights perspective, questioning not only their human rights compliance but also their effectiveness.

1) Mutual recognition of return decisions: Building upon the Commission Recommendation from March 2023, the proposal enshrines the mutual recognition of return decisions. Under this mechanism, a Member State that apprehended a person who has already received a return decision in another Member State may (and in the future shall) enforce that decision rather than issuing its own (Art. 9). This is made possible through a “European Return Order” (a form listing main elements of the return decision) made available among the states through the Schengen Information System (Regulation 2018/1860) or other information exchange channels. In theory, mutual recognition enhances efficiency, as only one state would conduct the return procedure. It is therefore unsurprising that the Commission presents it as a key element of the common system for return. However, this measure was already provided in a 2001 Council Directive, yet it does not appear to be regularly used.

Recognizing another state’s return decision comes with practical and legal challenges. Since grounds for legal stay are not harmonised at the EU level, a person may be in an irregular situation in one state but not in another. Although the proposal aims to harmonise procedures, Member States will still be allowed to grant residence permits on humanitarian, compassionate, or other grounds (Art. 7(9)). Since these residence permits are not harmonised, situations may arise where a state is required to remove a person who would otherwise qualify for a residence permit under its domestic law. Additionally, despite formal harmonisation, there will remain the scope for domestic interpretation of grounds for a legal stay (currently demonstrated by discrepancies between asylum recognition rates for the same nationality across the EU). So a state enforcing another’s return decision may face litigation, even though appeals against the decision would have to be lodged against the issuing Member State, according to the proposal. The enforcing state would also bear the costs of removal and potentially detention. Although the proposal provides for Frontex funding or compensation by the issuing state, accessing this financing may be cumbersome to organise. Mutual recognition is thus generally in the interest of the transit countries while disadvantageous for the destination countries.

Crucially, mutual recognition of return decisions and entry bans raises proportionality concerns as it effectively spreads across the EU the most restrictive approaches. As demonstrated above, despite the proposal's aim to harmonise procedures, there will still be room for domestic non-harmonised statuses (e.g., under Art. 7(9) of the proposal) as well as variations in the interpretation of harmonised statuses (e.g., refugee or subsidiary protection statuses). The Commission’s approach lacks coherence on this issue. It glorifies the mutual recognition of return decisions as a key efficiency measure, yet it makes no effort to introduce the mutual recognition of protection statuses—exposing a clear double standard in EU migration policy.  

2) Expansion of detention: the proposal significantly expands the legal basis for detention. Under the Directive, states may only apply detention unless other sufficient but less coercive measures can be applied effectively in a specific case. The proposal removes this requirement (Art. 29(1)-(2)), making detention the first resort measure, which is at odds with the CJEU case-law and HRC jurisprudence. The proposal introduces three new grounds for detention (Art. 29(3)), which are also questionable under international human rights law. Among the new grounds for detention is the need to determine or verify the person’s identity or nationality, even where there is no risk of absconding. This ground could lead to unnecessary and prolonged detention of individuals whose nationality is disputed, not recognized, or who are stateless.

Detention would henceforth also be allowed if the person poses “security risks,” defined as a threat to public policy or public or national security, the existence of serious grounds for believing that the person has committed a serious offence, or a clear indication of their intention to commit one (Art. 16). The proposal further provides for derogations from the detention regime otherwise applicable. States will be able to detain those covered by this ground in prisons (albeit separated from “ordinary prisoners”) rather than in dedicated detention centres, and for a period longer than the maximum permissible length under the proposal—potentially indefinite. Detention based on “security risks” blurs the lines between (administrative) immigration detention and criminal detention. Because it offers fewer guarantees to detainees, immigration detention is an exceptional measure that should only be applied on narrowly defined grounds related to imminent removal. Immigration detention is by no means intended to address security risks; rather, criminal laws should apply equally to anyone under the state's jurisdiction. Criminal pre-trial and post-conviction detention provide stronger due process guarantees and clearer time-limits. The proposal thus intends to sideline the CJEU case-law, as the court was adamant in stressing that detention on public order or safety grounds cannot be based on the Directive.

The proposal extends the maximum permitted length of detention in ordinary cases from 18 months to 24 months (Art. 32(3)). While states are not required to maintain detention for the maximum period, in practice, they often do. As a result, the proposal would lead to longer detention periods across the EU. This stands in stark contrast to the Commission’s previous stance, which claimed that the Directive had a beneficial effect on detention lengths in Member States by generally shortening them. Additionally, the proposal clarifies that these limits apply within a given Member State. So, if a person is transferred between states (for instance, under the mutual recognition mechanism), the detention period may start anew. The proposal also facilitates extended and unnecessary detention by removing the principle that detention shall only be maintained as long as removal arrangements are in progress and executed with due diligence. However, this requirement stems from the ECtHR's well-established case-law, so the Member States remain bound by it in any case.

The expansion of detention thus conflicts with several safeguards flowing from the right to liberty. Additionally, as pointed out by the EPRS, detention is neither necessarily effective (since there is no evidence to suggest that more detention leads to higher return rates) nor efficient (due to the considerable costs involved).

3) Generalized restriction on freedom of movement: the proposal introduces five freedom-restricting measures (including reporting obligations, residing in a specific place and electronic monitoring), labelling them as “alternatives to detention” (Art. 31). However unlike genuine alternatives to detention, these measures are not to be imposed instead of detention (i.e., when detention would otherwise be lawful) but in addition to detention – when detention is not or no longer justified (Art. 32). Labelling such measures as “alternatives to detention” is therefore misleading and creates confusion. Under the framework of detention and its so-called alternatives, individuals risk being either detained or subjected to these restrictive measures depending on the perceived level of the risk of absconding. Given the expansive definition of the risk of absconding (Art. 30), personal freedom would effectively become the exception rather than the rule. Although the proposal includes some human rights safeguards flowing from the right to freedom of movement within a State—such as individual assessment and proportionality requirements—these may not be sufficient to counterbalance the broad scope of restrictions.

Another set of freedom-restricting measures introduced in the proposal relates to the new obligation to cooperate and remain available for the return process (Art. 23). To ensure a swift, efficient, and effective return, the proposal subjects individuals to geographical restrictions (such as confinement to a designated area or residence at a specific address) and/or reporting duties for the duration of the return procedure. Notably, the proposal does not establish any additional conditions for imposing these measures. Subjecting every person in the return procedure to geographical restrictions is an indiscriminate and disproportionate measure. It fails to meet the necessity requirement under the right to freedom of movement, which demands that any restriction be necessary in the individual case for achieving the legitimate objective.

4) The downgrading of “voluntary” departure/return: The proposal rebrands "voluntary departure" (currently used in the Directive) as "voluntary return," which reduces clarity. A return that follows a return decision is not truly voluntary, as it is not based on the person’s informed and free consent. Even "voluntary departure" under the Directive is misleading, with "mandatory" or "accepted" return previously suggested as more accurate. The shift to "voluntary return" only deepens the confusion, likely aligning the language with assisted voluntary return and reintegration programs run by the International Organization for Migration.

Additionally, the proposal removes the priority currently given to "voluntary" departure/return, making removal the default option. The minimum 7-day departure period has also been eliminated, further downgrading the role of “voluntary” departure/return (Art. 12–13). This shift contradicts key legal principles and the Commission’s own position. The CJEU in Zh. and O. confirmed that “voluntary” departure/return is rooted in the principle of proportionality. Furthermore, this form of return is in states’ interests—it is cheaper and easier to organize. As the Commission acknowledged in its 2021 strategy, “voluntary” departure/return helps ensure more effective and sustainable outcomes. The explanatory memorandum applauds the increase in the uptake of “voluntary” departures/returns and, due to the absence of an impact assessment, it remains unclear why this form of return is now being restricted. While the proposal limits “voluntary” departures/returns, it simultaneously promotes them by introducing a new provision dedicated to return and reintegration assistance (Art. 46(3)). This inconsistency ultimately undermines both legal principles and practical considerations.

5) Externalisation of return: The proposal allows Member States to remove individuals—except unaccompanied children and families with children—to a third country with which they have an agreement or arrangement for return (Art. 17). This “return hub” would then be responsible for the further removal. In effect, the proposal establishes a legal basis in EU law for the externalisation of return. On paper, outsourcing returns may seem attractive to Member States. However, as a recent migration deal has shown, such measures often face legal challenges, operational hurdles, and high costs for the outsourcing state. The Commission itself considered “externally-located return centres” in 2018 but identified serious legal and practical challenges, including the risk of violating non-refoulement and conflicts with EU values.

Return hubs raise human rights concerns, particularly regarding detention and onward return. While the proposal requires third countries to uphold international human rights standards, including non-refoulement, it fails to specify how and by whom compliance would be assessed. Similarly, although independent monitoring is mandated, its scope and mandate remain undefined. Member States may not be able to escape responsibility for human rights violations in return hubs. Under the proposed framework, their involvement may be substantial enough to establish jurisdiction, as seen in recent UN Human Rights Committee decisions.

Concluding thoughts

As this analysis has shown, several measures in the Commission’s proposal risk violating fundamental human rights, including the prohibition of refoulement, arbitrary detention, and torture or ill-treatment. The generalised restrictions on freedom of movement and the downgrading of “voluntary” departure/return are not in line with the EU law principle of proportionality. At the same time, these measures are unlikely to improve the effectiveness of return, even when effectiveness is reduced to the sheer number of persons returned—let alone efficiency, which considers the human and financial resources required. In fact, detention does not necessarily lead to higher return rates, while being costly. Similarly, forced removal is more expensive and complex to organize than “voluntary” departure/return. The so-called “innovative solutions”—such as mutual recognition of return decisions and return hubs—lack clarity and may prove far more difficult to implement than envisioned on paper.

Moreover, the proposal fails to meet other key criteria for EU policymaking as outlined in the Commission’s Better Regulation Guidelines. The de-prioritisation of “voluntary” departure/return contradicts the criterion of sustainability. The criterion of coherence is also put into question. Internally, the downgrading of “voluntary” departure/return is inconsistent with the promotion of “voluntary” assisted return and reintegration. Externally, the use of “alternatives to detention” as standalone restrictive measures distorts the concept, which is meant for people who would otherwise be lawfully detained. Ultimately, the proposal disregards several core criteria of EU lawmaking—shortcomings that could have been identified through an ex-ante impact assessment.

The proposal fails to acknowledge that, despite its arsenal of measures to increase return numbers, not everyone in an irregular situation can or will leave EU territory. The obligation to issue a return decision for every person in an irregular situation—without a mandatory prior assessment of refoulement risks or other legal bars to removal (such as family and private life or health conditions)—inevitably creates a group of unreturnable individuals. This issue already existed under the Directive and is further exacerbated by the proposal. Currently, Member States may issue a residence permit for humanitarian or other reasons instead of a return decision, or withdraw/suspend an existing return decision for such reasons (Art. 6(4)). However, as the EPRS observed, these considerations are not automatically assessed within the return procedure.

Rather than addressing this gap, the proposal eliminates this possibility from the list of exceptions. Member States will only be able to withdraw or suspend a return decision (Article 7(9)); abstaining from issuing a return decision on such considerations will not be an option anymore. Impediments to return will be assessed only at the appeal stage of the return procedure, which undermines the effectiveness and efficiency of return policies let alone human rights compliance. Instead, an automatic assessment should be conducted before issuing a return decision, and a regular status should be granted to those who cannot be returned. Regular pathways for stay are widely recognized as a viable alternative to return and a normal feature of migration governance. A genuine "Common System for Returns" should at least recognize regularisation measures for unreturnable persons.

A legislative proposal based on coercion and restrictive measures—without any supporting evidence due to a missing impact assessment—would never see the light of day in the EU if its addressees were not people whose voices are typically unheard. People fall into irregular situations in many different ways and many have lived in and contributed to European societies for decades. They remain invisible, and those who defend their rights face increasing criminalisation. In this context, responding to the demands of some Member States, the Commission has enjoyed wide flexibility to propose unrestricted coercive measures—with little accountability for their impact on those affected. However, law-making without evidence sets a dangerous precedent. In the future, such an untransparent approach could extend beyond undocumented migrants, affecting asylum seekers, migrants in a regular situation, and ultimately EU citizens. Now, it is up to the European Parliament and the Member States in the Council to reaffirm the EU’s founding principles: democracy, the rule of law, and human rights.

Fast-track visa reimposition: a strengthened tool for EU immigration control and external relations policies?

 

Professor Steve Peers, Royal Holloway University of London

Photo credit: Sixflashphoto, via Wikimedia Commons 

How and why can short-term visa requirements be re-imposed, on a fast track basis? For the EU, this issue is decided collectively, on the basis of legislation that applies to all EU Member States (except Ireland), plus the Schengen associated countries of Norway, Switzerland, Iceland and Liechtenstein. A proposal to change those rules to reinforce EU immigration and asylum policy is moving closer to agreement; and as it does so, both Member States and the European Parliament seek to use the new law to reinforce EU external relations objectives more generally too.

Current legal framework

The EU has long had a law harmonising which non-EU States (and entities not recognised as States by some or all Member States) are – or are not – subject to visa requirement for their nationals to make short-term visits (90 days out of every 180 days) to visit the EU (Ireland has its own visa policy) and Schengen associated countries. The law was first amended to provide for fast-track reimposition of visa requirements on immigration and asylum policy grounds in 2013, to assuage Member States concerned about the impact of dropping visa requirements for most of the Western Balkans several years beforehand. These rules were amended in 2017, this time to assuage Member States concerned about dropping visa requirements for eastern neighbourhood countries (eventually visas were waived for Moldova, Ukraine and Georgia). Currently they are found in the codified version of the EU visa list Regulation, adopted in 2018.

At present, the countries whose nationals are not subject to EU short-term visa requirements (the ‘white-list’) comprise most non-EU European states (including the UK), North American and most Latin American countries, some wealthier Asian countries (such as Israel, the UAE, Japan and South Korea), Australia, New Zealand and a number of small middle-income islands in the Caribbean and the Indian and Pacific Oceans. All other countries (the ‘black-list’) are subject to visa requirements. Member States have some flexibility as regards some categories of people (such as holders of official passports), ie they can waive or impose visa requirements for these groups of people even if the non-EU countries as a whole are on the EU white-list or black-list.

For the countries on the black-list, the EU has agreed some visa facilitation treaties (usually in return for readmission treaties) which make it easier for their nationals to obtain visas. It has been willing to suspend these partly or wholly on political grounds (ie Russia and Belarus). The EU has also used its internal law (the visa code) in recent years to make (or threaten to make) visas harder and more expensive to get, if a country (in the EU’s view) does not cooperate sufficiently on readmission. A further proposal to restrict EU trade preferences for developing countries on these grounds has not yet been agreed (in part because it has been criticised for potentially breaching WTO law).

But our focus here is on the potential fast-track reimposition of visa requirements – ie, moving a country from the black-list to the white-list – in the first place. Under the current law, in addition to the separate possibility of a fast-track reimposition of visa requirements if a non-EU country does not waive its own visa requirements for EU citizens (a power that the EU has never used), visas can be reimposed on a fast track basis if there is ‘a substantial increase in the number of nationals of that third country refused entry or found to be staying in the Member State's territory without a right to do so’; or ‘a substantial increase in the number of asylum applications from the nationals of that third country for which the recognition rate is low’; or ‘a decrease in cooperation on readmission with that third country, substantiated by adequate data, in particular a substantial increase in the refusal rate of readmission applications submitted by the Member State to that third country for its own nationals’ (or nationals of other states who transited through that country, if a readmission treaty with the EU obliges the country concerned to readmit them); or ‘an increased risk or imminent threat to the public policy or internal security of Member States, in particular a substantial increase in serious criminal offences linked to the nationals of that third country’. The preamble to the Regulation defines some of these concepts further. A ‘substantial increase’ is more than 50%, although the Commission can use a lower threshold if it ‘deemed it applicable in the particular case’. A ‘low recognition rate’ for asylum applications is ‘around 3 or 4%’, although the Commission can deem a higher rate applicable in a particular case. There is no further definition of ‘serious criminal offences’, or any indications of what might otherwise comprise ‘an increased risk or imminent threat to the public policy or internal security of Member States’, given that the ‘serious criminal offences’ ground is only an example of such a risk or threat (‘in particular’). The reference period to assess these changes is a two-month period compared to the same period in the previous year, or compared with the last two months prior to dropping the visa requirement for a non-EU country. As for the process, after considering a complaint by a Member State for some issues, or on its own initiative for others, and holding discussions with the non-EU country concerned, the Commission can adopt an implementing act (ie subject to scrutiny by Member States’ representatives) suspending the visa waiver for some groups of that country’s nationals for nine months, followed (if the issues still persist) by a delegated act suspending the visa waiver for all nationals of that country for 18 months – which can be extended for six months if the Commission makes a legislative proposal (which would need to be agreed by the Council with a qualified majority of Member States, and the European Parliament) to make the move to the black-list permanent. As noted above, these rules were loosened in 2017, compared to the original 2013 version. The original version of the rules had: set a six-month reference period (instead of two months, with a seven-year time limit on the rule for countries moved to the white-list); required a ‘sudden’ increase creating an ‘emergency’ situation, and pressure on the asylum system as regards the ‘low recognition rate’ criterion’; and provided only for an implementing act reimposing visa requirements for all nationals of the country concerned for six months – possibly extended by 12 months if the Commission proposed legislation.   Despite this loosening of the rules, though, the EU has been reluctant to use them in practice. In fact, they have only been invoked once, to reimpose visa requirements for Vanuatu (see the implementing decision and the delegated act, later extended while the legislation was amended) after that Pacific micro-state introduced a form of ‘investor citizenship’, that the EU judged made it too easy for wealthy (and arguably dodgy) people from black-list countries to buy the nationality of the tropical island so they could visit the EU visa-free. (The question of whether a Member State – Malta – is breaching EU law by doing the same thing, is due to be decided by the CJEU next month). In the EU’s view, the ‘public policy’ ground for fast-track reimposition of visa requirements applied here. Proposal to change the rules After some period of generalised grumbling about the need to change the rules to reinforce the EU’s immigration and asylum policy – in particular to induce neighbouring countries to align their visa policies with the EU, to make it harder for nationals of black-list countries to reach the EU’s borders and claim asylum – the Commission tabled a proposed amendment to the visa list Regulation in 2023. The Council agreed its position on this proposed law in March 2024, and the European Parliament’s civil liberties committee agreed its position last week, as set out here.* (*updated 24 March 2025 to link to the final version of the committee's position)  The Commission proposal The Commission proposal keeps the existing four grounds for fast-track reimposition of visa requirements, with limited changes. There would have to be a significant risk to public policy, et al, rather than an increased risk (now expressly including ‘hybrid threats’, defined in the preamble as including ‘instrumentalisation’ of migrants, ie Belarus shoving them across borders). But there would be new grounds in addition: investor citizenship (which, as we have seen, is currently regarded as covered by the ‘public policy’ ground); non-alignment with visa policy; and non-compliance with commitments made to the EU when a country was moved to the white-list. The definitions of ‘substantial increase’ are moved from the preamble to the main text, and unchanged except for a power for the Commission to set them higher or lower than a 50% increase. For the asylum ground, the low recognition rate is set at 4%, but again the Commission can set it higher or lower than that. The Commission proposal also changes the details of how long visa requirements can be reintroduced for. An implementing act as regards some categories of people having to hold visas could apply for 12 months (instead of 9), and there would be a new power for the Commission to act urgently. A delegated act applying to all nationals of the country concerned could apply for 24 months, instead of 18. In either case, there would be a power to terminate the secondary measure in the event of a change in circumstances. The Council version The Council accepts most of the Commission’s proposal, but wants to make some changes. Its version would add ‘deficiencies in document security legislation or procedures’ to the public policy clause, and the preamble would refer to ‘terrorist offences and activities of organised criminal groups’ as regards public security. There would also be the possibility of reintroducing a visa requirement on external relations grounds:   a significant and abrupt deterioration in the Union’s external relations with a third country listed in Annex II, in particular when it relates to human rights and fundamental freedoms and is deriving from any of the following: (i) serious human rights violations and abuses; (ii) serious breaches of international law and standards, including human rights law and non-compliance with international court decisions and rulings   There would also be a fast-track possibility as regards ‘any other ground for suspension set out in a short-stay visa waiver agreement between the Union and a third country listed in Annex II, limited to the scope of application of such agreement’.   In the Council’s view, a ‘significant increase’ would be 30%, not 50%; and a low asylum recognition rate would be 20%, not 4% - matching the threshold for fast-tracking asylum applications under the recently agreed revised EU asylum procedures law.   The European Parliament   Like the Council, the European Parliament seeks to introduce a new external relations ground for fast-track reimposition of visas, although the Parliament’s text is worded differently:   a deterioration in the Union’s external relations with a third country listed in Annex II caused by: (i) serious breaches by that third country of the principles set out in the Charter of the United Nations; (ii) grave violations by that third country of the obligations deriving from international human rights law or international humanitarian law; (iii) violations by that third country of bilateral agreements between it and the Union; (iv) that third country carrying out hostile acts against the Union or Member States with the aim of destabilising or undermining society or institutions which are key for the public policy and internal security of the Union or the Member States; (v) non-compliance or non-alignment by that third country with relevant Union sanctions.

The preamble of the Parliament’s version states that the hostile acts ‘could result from foreign interference in political processes, economic coercion, cyber operations, economic espionage or the sabotage of critical infrastructure’.

Compared to the Council’s position, the ‘deterioration’ in external relations in the Parliament’s version need not be ‘significant and abrupt’, and is not solely focussed on human rights – although as the Council’s text is non-exhaustive (‘in particular’), the Council could encompass other issues too. In any event, the Parliament’s version mentions ‘serious breaches’ of the UN Charter, instead of violations of international law and non-compliance with judgments; it also differs by expressly referring to ‘international humanitarian law’, breaches of treaties with the EU, ‘hostile acts’ and non-compliance or non-alignment with EU sanctions. (It should be noted that non-EU countries are not legally obliged to comply or align themselves with EU sanctions; although there may be some cases where the EU and non-EU countries will be bound by sanctions because they were adopted by the UN Security Council, non-compliance with those sanctions would in that case be covered by the reference to non-compliance with the UN Charter)

Furthermore, the Parliament’s position would set 40% as the new definition of ‘substantial increase’, and drop any change in the threshold of low recognition rates for asylum. The Commission’s power to adopt urgent measures would be limited to cases of ‘significant risk or imminent threat to the public policy or internal security of a Member State’; and when visa obligations are reintroduced, Member States would lose their power to reintroduce the visa waiver for categories of people, such as holders of official passports. This last point may be a response to the situation created by the recent EU decision to partially suspend its visa waiver treaty with Georgia on human rights grounds, so there is no longer a visa waiver for holders of official passports; yet arguably Member States such as Hungary still have the option under the visa list Regulation to retain such waivers. The Parliament’s version of the law would address such issues. *The Parliament's version would also create a new power for the Commission to require (on a fast-track basis) Member States to reimpose visa requirements for official passport holders from black-list countries (ie Member States could no longer exempt them from the visa requirements) - either due to a 'significant and abrupt' deterioration in external relations (using the same definition of 'deterioration' as for white-list countries) or a failure to cooperate on readmission (as defined by the visa code). [last sentence updated 24 March 2025]    

Comments

Negotiations between the Parliament and the Council will likely start soon. Their two versions of the proposal are not profoundly far apart, differing in particular on: some of the details of a new external relations ground (but not the idea of such a new ground), as summarised above; the thresholds for ‘substantial increase’ and low recognition rates for asylum; limiting the new prospect of adopting a requirement to reimpose visa requirements urgently (but not the idea of an urgent procedure as such); and removing the power of Member States to waive visa requirements for official passport holders when visa requirements are reimposed. On the other hand, both institutions broadly agree on several aspects of the Commission proposal: the new grounds of investor citizenship, non-alignment with visa policy and non-compliance with commitments previously made to the EU; the lengthened periods for reimposing visa requirements via implementing or delegated acts; and a possible early end to the secondary measures.

It’s tempting to extrapolate which non-EU countries might be in the Council or Parliament’s mind as regards the new external relations ground. For instance, the Trump administration’s designs on Greenland (among other things) could, if put into the form of some economic pressure or military action, be seen as ‘hostile’; Israel is often accused of breaching international humanitarian law; and the EU has concerns about the UK’s compliance with Brexit deals. But the Commission will have a great deal of discretion using these powers, and is subject to scrutiny by representatives of Member States if it uses them. And it should be recalled that the Commission has been reluctant to impose visa requirements on the US in the context of non-reciprocity as regards some Member States for the EU’s visa waiver for the US (the European Parliament even sued the Commission on this point, and lost); the Commission and many Member States have been at most mildly critical of the Israeli government; and the EU has shown an established preference to settle disputes with the UK by negotiation, not escalation.  

There may be more enthusiasm for using new powers relating to non-alignment with EU visa policy – although on this point it should be noted that this would entail asylum seekers being prevented from taking a relatively safe journey to the EU border – following a visa-free flight to Serbia, for instance – and may undertake an unsafe sea journey via payment of smugglers instead. With enemies like the EU’s interior ministries, the smuggling gangs don’t need friends.

The existing powers as regards overstay rates, readmission cooperation and asylum recognition rates might be more widely used if the thresholds for their use are lowered. (Note that the European Parliament has already accepted the 20% threshold for asylum recognition, as regards fast-tracking asylum applications as part of the asylum procedures law). In each case, though, the Commission might prefer to use the threat of using the fast-track visa reimposition power to obtain concessions from non-EU countries, rather than move straight to using such powers. Speaking softly, and carrying (but rarely using) a big stick, has been the guiding principle of the EU’s approach to reimposing visa obligations – at least so far. Whether a bigger stick might be used more often remains to be seen.

The Italy-Albania protocol before the Court of Justice of the European Union – hearing of the CJEU

  

Matteo Zamboni (human rights lawyer working between Italy and the UK. He is a partner to the Immigration Law and Policy clinic of Goldsmiths, University of London)

 

Photo: The towers of the Court of Justice in the Luxembourg fog (c) Matteo Zamboni

 

On Tuesday 25 February 2025 I attended the Grand Chamber hearing before the Court of Justice of the European Union regarding the compatibility with EU law of the infamous Italy-Albania Protocol on asylum applications and return procedures as a partner to the Immigration Law and Policy clinic of Goldsmiths, University of London. Here is my take.

 

Introduction

 

On 25 February 2025 the Grand Chamber of the Court of Justice of the European Union (CJEU) held the much anticipated hearing regarding the compatibility with EU law of the Italy-Albania Protocol on asylum applications and return procedures. More specifically, the CJEU was called to give an answer to four preliminary questions referred by the Tribunal of Rome in November 2024.

 

Much of the reports appeared on Italian and international press focus on the “U-turn” of the European Commission, which, differently from what it had argued in its written observations, at the hearing stated that EU law grants Member States the possibility to declare a third country as a ‘safe country of origin’ within the meaning of Directive 2013/32 (the current EU Directive on asylum procedures) despite the existence of exceptions for specific at-risk groups of people for whom the country is, in fact, not safe.

 

However, there is an underlying legal issue which seems to go above and beyond this; i.e., the guarantee of uniformity in the application of EU law on asylum procedures across the European legal space.

 

Outline of the post

 

While waiting for the opinion of the Advocate General (due on 10 April 2025) and obviously for the judgment of the Court (expected before the Summer) the present post gives an account of the hearing by (a) giving a brief description of the content of the Italy-Albania Protocol; (b) summing up the procedural steps before Italian courts that prompted the referral to the CJEU; (c) reporting on the oral arguments made by the parties at the hearing and on (d) the questions asked by the Advocate General and the CJEU Judges.

 

In so doing, the contribution draws upon previous blogs which give detailed information as to (a) the relevant EU legislation (essentially directives 2005/85 and 2013/32, and regulation 2024/1348, applicable from June 2026); (b) the interpretation of the concept of ‘safe country of origin’ (SCO) given by the CJEU in judgment of 4 October 2024 (case C-406/22); (c) the questions referred to the CJEU for a preliminary ruling; and (d) the existence of other requests for referral stemming from the application of the Italy-Albania Protocol.

 

At the outset: the Italy-Albania Protocol

 

In short, and as already detailed by multiple sources, the Protocol concluded between Italy and Albania on 6 November 2023 allows Italy to process asylum applications through an accelerated border procedure from two centres built on Albanian territory (one in the port of Shengjin, the other in the city of Gjader). As clarified in the instrument of ratification (law no 14 of 21 February 2024) and in the SOPs of the Ministry of the Interiors on the implementation of the Protocol (obtained by lawyers from ASGI by means of a freedom of information request), the procedure applies only to ‘migrants’ coming from countries identified as SCOs by Italian law in accordance with relevant EU legislation, and with the exclusion of women, unaccompanied minors, and individuals affected by clear pathologies.

 

The selection of these ‘migrants’ is made by using a so-called ‘hub vessel’. In practice, eligible individuals rescued by vessels of the Italian navy on the high sea are transferred into a dedicated vessel (the hub vessel) and then brought to the centre in Shengjin for initial reception and screening procedures. They are then moved to the centre in Gjader, and therein detained while their asylum claim is processed. These individuals are then made to apply for international (or other forms of) protection directly from the centre in Albania, where they are interviewed by the Commissione Territoriale per il Riconoscimento della Protezione Internazionale di Roma (Rome Regional Commission for the grant of international protection). In case of refusal, they can appeal before the Tribunale ordinario di Roma (District Court of Rome). For this purpose, the asylum seekers are connected into the hearing via video link and represented by lawyers in Rome, who are chosen from a list. During the entire procedure, the asylum seekers remain in detention in the centre of Gjader pursuant to the reference to Article 28-bis legislative decree no 25 of 28 January 2008 contained in the ratification instrument (Article 3 law 14/2024).   

 

Finally, it needs to be underscored that, in Italian law, the list of SCOs was originally set out with a decree of the Minister of Foreign Affairs (lastly on 7 May 2024) adopted on the basis of Article 2-bis legislative decree 25/2008. However, following the CJEU judgment in case C-406/22 (which ruled out the designation of part of countries of origin as ‘safe’, under the current asylum procedures Directive), the Italian government, with the support of its parliamentary majority, passed decree-law no 158 of 23 October 2024, which directly provide an updated list of countries considered as SOCs by Italy. For our purpose, it is important to note that, along with other 18 countries, this list includes Bangladesh.

 

The case in Italy and the questions referred to the Court of Justice

 

The first transfer to Albania was performed on 14 October 2024 and concerned sixteen men – ten Bangladeshis and six Egyptians. Their requests for protection were rejected by the Regional Commission. However, ruling on an appeal brought by two individuals from Bangladesh (identified, in the procedure before the CJEU, with the fictitious names of Alace and Canpelli), the District Court of Rome did not validate their detention in the centres in Albania. Moreover, the District Court referred to the CJEU four questions regarding the concept of SCO under EU law.

 

The questions, which refer to Articles 36, 37, and 39 of Directive 2013/32 (concerning ‘safe countries of origin’), interpreted in the light of Article 47 of the EU Charter of Fundamental Rights (the ‘Charter’) and Articles 6 and 13 of the European Convention on Human Rights (‘ECHR’), ask the CJEU in Luxembourg whether EU law:

 

(1) precludes Member States from designating SCOs by means of legislative instruments (acts of Parliament); 

 

(2) requires national legislation to publish the sources relied on for the designation of a specific country as a SCO;

 

(3) allows national judges called to review the designation of SCOs to use information from sources other than those referred to in directive 2013/32; and

 

(4) precludes a non-EU country from being designated as a SCO where there are categories of people for whom the substantive conditions for such a designation laid down in Annex I to directive 2013/32 are not met.

 

The oral arguments made by the parties

 

This section resumes the arguments made by the parties at the hearing.  

 

The defence

 

The legal teams for Mr Alace and Mr Canpelli coordinated their oral pleadings. The defence thus opened by underlining that the designation of a country as a SCO bears critical consequences, as it triggers the application of the accelerated procedure, entailing serious hindrances to the right of the individual requesting international (and/or other forms of) protection, including shorter time-limits, the presumption (albeit rebuttable) that no protection is needed, and (under Italian law implementing the protocol) the detention pending the examination of the request.

 

That said, the defence moved very quickly on the first question, as it seems rather obvious that EU law does not mandate nor prevent Member States from resorting to specific acts (legislative or others; e.g., ministerial decrees) to implement the rules set out in secondary EU legislation. So, in this case, nothing prevents Italy from designating SCOs by means of legislation; i.e., acts of Parliament.

 

According to the defence, however, this should always guarantee the principle of legal certainty and the respect of the right to an effective remedy. As a consequence, with reference to the second and third question, the defence argued that Member States are under the obligation to make public the sources used to proceed to the designation of SCOs and that national judges cannot be prevented from performing their own research and use other sources to review the decision to consider one country as safe, in the specific case.

 

As regards the fourth question, the defence maintained that the letter of directive 2013/32, and especially its Annex I, read in the light of the subsequent amendments to the relevant EU legislation, leaves no doubt as to the absolute impossibility to designate a country as a SCO whenever there are elements to hold that the country is, in fact, not safe for specific categories of people.

 

The Italian government

The Italian government began the pleadings by addressing the fourth question, concerning the possibility to designate a country as ‘safe’ notwithstanding the existence of exceptions for certain categories of people. Relying on recital 42 in the preamble of directive 2013/32, the government argued that the expression ‘generally and consistently’ contained in Annex I is to be interpreted as meaning that a country is to be deemed safe when it is such for the majority of its nationals. Otherwise, it would be impossible to designate SCOs. In fact, in the view of the Italian government, most countries are in fact ‘un-safe’ for certain categories of persons. In this regard, the government added that the contrary interpretation, according to which a country can be designated as ‘safe’ under EU law only when it is such for the generality of its nationals, would be ‘desirable but unrealistic’ and overall ‘unreasonable’. In support of this argument, the government cited Article 59 § 2 of new regulation 2024/1348 (even though it does not apply yet).

 

In closing on this point, the government argued that the accelerated procedure still guarantees all of the substantive rights of the asylum seeker. The Italian government, thus, openly disavowed the interpretation given by the CJEU in case C-406/22 (see, in particular, § 71), where the Court held that, due to its exceptional nature, the interpretation of the rules concerning the accelerated procedure should always be a restrictive one. To the contrary, according to the Italian government, the accelerated procedure guarantees in full the right to asylum and the possibility to obtain judicial review, so that it should not be seen as ‘exceptional’ when compared to the ordinary one. The only difference, the Government said, is that the accelerated procedure is faster, which goes to the benefit of the State concerned as well as of the asylum seekers.

 

Finally, and this is the core of the Italian case, the government stressed that EU law leaves discretion to Member States when issuing the list of SCOs.

 

Turning to the other questions, the Government stated that Italy has a right to designate SCOs by an act of Parliament and that there is no need to make the sources used for the designation publicly available, as transparency is guaranteed by internal parliamentary rules stating that discussion in Parliament is public. In addition to more detailed arguments taken from the letter of directive 2013/32, the Government argued that EU law cannot impact on the constitutional system of Member States. Therefore, in accordance with the principle of the separation of powers enshrined in the Italian constitution, it is for the Legislature to designate SCOs in general terms, while the power of national judges is confined to evaluate whether, in a specific case, the designation applies to a given asylum seeker. At the most, the national judges can refer the question regarding the general designation of SCOs to the (Italian) constitutional court.

 

The other Member States

 

Thirteen Member States (Czech Republic, Germany, France, Cyprus, Hungary, Malta, the Netherlands, Austria, Poland, Slovakia, Finland, and Sweden) participated in the proceedings by filing written observations and seven Member States (Bulgaria, Czech Republic, Germany, Latvia, Lithuania, Hungary, and Sweden) were present at the hearing.

 

Generally speaking, all of these countries supported the position of the Italian Government. Among them, the arguments made by Germany, Hungary, and Sweden seem worth mentioning.

 

The German government focussed on the issue of which national authority has the power to designate a country as a SCO and which remedies may be enacted to challenge this designation at the domestic level. According to Germany, EU law (directive 2013/32) distinguishes between the general designation of a country as a SCO, on the one hand, and the application of the concept in a specific case, on the other hand. The former assessment is based on general considerations, while the latter is based on the specific circumstances of the asylum seeker concerned.

 

Against this background, the position of Germany is as follows: Every national court has the power to perform the second assessment (application of the general designation of a country as a SCO to a specific case). However, the first assessment (general designation of a country as a SCO) must remain separate from this, and reserved to the Legislature. This is so because EU law leaves ‘leeway’ to Member States in performing the first assessment. Therefore, in the opinion of the German government, a system that reserves the first (general) assessment to Parliament is not only compliant with EU law, but actually more transparent and democratic. This general assessment can be reversed by the constitutional court, which has the power to review the constitutionality of legislation. Instead, if a national court believes that the general assessment that prompted the designation of a country as a SCO goes against the Constitution and/or EU law, it will have to raise a challenge of constitutionality before the constitutional court.

 

The government of Hungary stretched this argument to the widest possible extent. In their view, only national constitutional courts have the power to review the (lack of) legality of the legislation establishing the list of SCOs. In reaching this conclusion, the Hungarian government insisted on the concept of ‘room for manoeuvre’, arguing that Member States enjoy wide discretion in the implementation of EU law. 

 

Finally, the pleadings of the government of Sweden differed from the previous ones in that they highlighted that the questions referred to the CJEU are relevant to the whole of the EU, since asylum procedures must be effective throughout the European Union legal space, without prejudice to the right of all individuals to effective judicial protection in terms of individual examination of their requests. Moreover, the Scandinavian State pointed out that, in its domestic legal system, there is no general list of SCOs. Instead, it is for each national court adjudicating on asylum claims to determine whether a legal principle that is being challenged is compatible with EU law.

 

All the same, in addressing the first two questions the Swedish government stood with Italy, arguing that the way in which the Italian Legislature decides to designate SCOs is not a matter for EU law. This position was justified by referring to the principle that it is incumbent on each Member State to organise its legal system in accordance with the overarching principles stemming from EU law and from its constitutional settlement.

 

The EU Commission

 

Finally, the lawyer for the Commission made the following points:

 

(a) in general terms, the national judge who finds that the definition of a third country as ‘safe’ is problematic must refer the question to a specific judicial authority; e.g., the constitutional court;

 

(b) in the specific case, the national judge has the power to find that the designation of a third country as a SCO does not apply to the individual(s) whose case is under examination, meaning that, for them and only for them, the country cannot be considered as safe;

 

(c) the fact that the exception concerning categories of people was established in directive 2005/85, but not replicated in directive 2013/32, does not in itself preclude Member States from providing for such exception in the national lists of ’safe countries’;

 

(d) Article 37 of directive 2013/32 empowers Member States to make a discretionary evaluation as to which countries to include in the list of SCOs; 

 

(e) consequently, nothing prevents Member States from including in the list of SCOs countries that may not be safe for a category of persons;

 

(f) it is irrelevant whether these categories include a large number of individuals (e.g. the category of ‘women’); what matters is that each category is clearly defined;

 

(g) the formula ‘generally and consistently’ contained in annex I to directive 2013/32 needs to be interpreted as meaning that a country cannot de deemed to be ‘safe’ under EU law when there are ‘systematic’ violations of human rights.

 

The questions from the Advocate General and the Court

 

The questions asked after the first round of pleadings give an indication as to the possible approach that the Court will take on the matter.

 

To begin with, the Judge rapporteur asked the Commission to clarify its position regarding the fourth question; i.e., the possibility to consider a country as a SCO notwithstanding the existence of specific at-risk categories. Interestingly, the question was reinforced by the President of the Court, which noted the possibility of a misunderstanding due to the very nuanced position taken by the EU institution. In its reply, the Commission clarified that, in its opinion, directive 2013/32 does not preclude this possibility.

 

As noted in the press, this position is different from the one stated by the Commission in its written observations, where it was argued that ‘Directive 2013/32 does not allow for the designation of a third country as a safe country of origin where a significant number of persons belonging to the same category are systematically persecuted or face a risk of suffering serious harm […]’.

 

Secondly, the Advocate General asked the Italian Government some specifics as to the review of constitutionality in the Italian system. Among other things, he enquired whether the Constitutional Court is empowered to review compliance of Italian law with EU law (including Annex I of directive 2013/32) and if a question of constitutionality can be raised by the local/district courts (courts of first instance).

 

The Italian government replied in the affirmative, holding that the question of constitutionality can be raised by all national judges and that the constitutional court can indeed evaluate compliance of national law with EU law. Moreover, the government reminded the CJEU that the constitutional court could, if it deems it useful, raise preliminary questions to the CJEU. In so doing, the Italian government unreservedly affirmed its view that the challenge of constitutionality provided for in the Italian legal system constitutes an effective remedy for the purpose of EU law.

 

In a second round of questions, the AG asked the Italian government about the periodic update of the Italian list of SCOs, that was due on 15 January but, as of the date of the hearing, was not published. In particular, the AG asked whether resorting to an act of Parliament for each and any update of the list of SCOs did not risk to excessively slow down the process.

 

In reply, the Italian Government informed the Court that the new list was about to be published and that the legislative process is not excessively burdensome (!) and thus not liable to cause undue delays.

 

Moreover, the AG questioned the reasons why the sources used by Parliament to designate a country as a SCO could not be made public and asked whether the Italian Government is perhaps using confidential sources.

 

The Government denied using confidential sources, but defended the position not to make them publicly available, trailing back to the argument that the legislative process is transparent, and that parliamentary discussion is public.

 

Finally, two Grand Chamber judges pressed the Italian government on the underlying issue of supremacy of EU law. In their questions, the judges asked whether a national judge that has no doubt over the non-compatibility of the national legislation designating SCOs with the criteria set out in Article 37 and Annex I of directive 2013/32 would, in any event, be forced to raise a question of constitutionality. This point was compounded by the apt reminder that, under the general principles of EU law as interpreted by the consistent case law of the CJEU, every national judge, including in first-instance proceedings, is empowered to assess the compliance of national law with the acquis communautaire.

 

Even there, however, the Italian government stuck to the position that it is only for the Constitutional Court to look at whether a manifest and general error was made in the general designation of a country as a SCO, while the power of lower courts is confined to the evaluation of the specific case. In the government view, it is to be reiterated that this is so because Member States enjoy discretion when making the general designation of SCOs.

 

Conclusion

 

While it is impossible to second-guess the outcome of the case, it is rather safe to assume that, contrary to what was held in press reports, the judgment on the four questions referred to the CJEU will not put an end to the litigation over the Italy-Albania protocol.

 

As noted by the CoE commissioner on human rights, future litigation may involve questions regarding the overall fairness of asylum procedures, the possibility of automatic detention without adequate judicial review, access to legal aid, and respect of the right to an effective remedy. On the same vein, in a report to the Italian Parliament of 30 November 2024, Professor Satvinder S. Juss from Kings College raised the question as to whether EU asylum law can be applied outside the EU.

 

Moreover, as already noted in the literature, the case insists on wider considerations concerning the ‘consistent interpretation’ of EU law.

 

Admittedly, during the hearing the issue remained in the background. The defence did not really push on it, while the Italian Government and the intervening Member States did so only implicitly, by heavily relying on the principle of procedural (and constitutional) autonomy (as evidenced, for example, by the pleadings of Sweden).

 

Yet, the questions of the Grand Chamber’s judges demonstrate that the issue is very much on the table.

 

Actually, it seems that this can be better dealt with if divided in two separate points:

 

-          first, whether the national (Italian) judge adjudicating in first instance on a request for international protection submitted by an individual coming from a country that has been designated as a SCO by Italian law is empowered to refer the question over the compatibility of national law with directive 2013/32 to the CJEU, pursuant to Article 267 TFEU, or whether this has to go, first, to the Italian Constitutional Court, who would then refer the question to Luxembourg;

-          second, whether the national judge is empowered to disapply the national legislation designating SCOs if he or she finds that this is contrary to directive 2013/32, and in particular to Articles 37 and 38 and Annex I thereto.

 

The first limb of the matter seems clearer, as it appears safe to assume that the argument of the Italian government (with support from the intervening Member States, especially the Hungarian government) on the need to raise a question of constitutionality contravenes the overarching principle that the principle of ‘the primacy of EU law […] requires […] national court[s] to be free to refer to the Court of Justice for a preliminary ruling any question that [they] considers necessary, at whatever stage of the proceedings [they] consider[…] appropriate’ (see joined cases C-188/10 and C-189/10 Aziz Melki, Sélim Abdeli).

 

The second limb of the question is not as straightforward.

 

As to the general principles, it is commonplace that, under the well-established case law of the CJEU,

 

(a) if a provision of EU law is directly effective, domestic courts must apply it in priority and are thus empowered to disapply conflicting provisions of national legislation without the need to request or await the prior setting aside of such provision by legislative or other constitutional means (see case C-106/77 Simmenthal and following case-law);

 

(b) specific provisions contained in EU directives are ‘capable of producing direct effects’ (see e.g. case C-9/70 Franz Grad and C-41/74 Van Duyn);

 

(c) the direct effect of provisions contained in a directive depend on whether they are ‘unconditional and sufficiently precise’ (see e.g. case C-236/92 Comitato di Coordinamento per la Difesa della Cava);

 

(d) if so, these provisions can be invoked by individuals appearing before national courts ‘whenever the full application of the directive is not in fact secured, that is to say, not only where the directive has not been implemented or has been implemented incorrectly, but also where the national measures correctly implementing the directive are not being applied’ (see case C-62/00 Marks and Spencer).

 

Therefore, the question is whether Articles 37, 38 and Annex I of directive 2013/32 are ‘unconditional and sufficiently precise’.

 

The existing jurisprudence is only relatively useful. True, the CJEU has already stated that, under these provisions, ‘it is for each Member State to designate safe countries of origin […] in accordance with the procedure laid down in Articles 36 and 37 and in Annex I to Directive 2013/32, namely, in particular, the adoption by the national legislature of a list of third countries in accordance with the criteria laid down in Annex I’ (see case C-404/17). Moreover, in the above-mentioned case C-406/22 the Court confirmed that, due to its exceptional nature, Article 37 must be interpreted restrictively. However, it does not seem that the Luxembourg judges had, so far, had the chance to decide over the question of direct effects of these provisions.

 

This seems indeed complicated.

 

On the one hand, the argument of the Italian Government (and of all the intervening Member States) according to which directive 2013/32 leaves discretion to States in designating SCOs seems to have some merits, and resonates with a stream of CJEU case law excluding direct effect of EU law provisions which are ‘subject […] to the to the taking of […] measure[…] by the Member States’ (e.g. case C-263/92 at § 9 and 12-14).

 

On the other hand, this stance can be said to have been overtaken by more recent case law that, in addition to their wording, focusses on the context and purpose of the EU law provisions at stake in determining whether they are capable of being directly effective. For example, this allowed the CJEU to hold that, although it required actions from the part of the Member States, the Annex to Directive 80/987 (on the protection of employees in the event of insolvency of their employer) regarding the categories of employee whose claims may be excluded from the scope of that directive within the meaning of Article 1(2) was, indeed, directly effective (see case C-441/99, in particular § 39 and 40, where the Court stated that ‘the Member State concerned has [made the designation required by the Annex]’ and therefore ‘made full use of the discretion which it enjoy[ed] […] in the implementation of the Directive’ (see case C-441/99 Riksskatterverket).

 

Against this background, the good news is that this question was referred to the CJEU by the district court of Bologna in another case concerning the application of the Italy-Albania Protocol (case C-750/24 Ortega).

 

It will no doubt be interesting to see the way in which this question will be addressed by the Luxembourg Judges. What is sure is that the case touches upon cornerstone principles of EU law and that specific expertise will be needed from the defence in order to stand up to the legal arguments made by Italy with the support of the EU commission and virtually all the other Member States.