One Step Forward, Three Steps Back? The Alace Judgment and the Future of the 'Safe Country of Origin' Practice

 

Karla Žeravčić, Researcher, EUI

Photo credit: governo.it, via Wikimedia Commons

Less than a year after the Tribunale ordinario di Roma (Ordinary Court of Rome) submitted a preliminary reference, and just four months after Advocate General (AG) De La Tour delivered his Opinion, the CJEU’s Grand Chamber handed down its judgment in Alace on 1 August 2025.

This is the third case from Luxembourg tackling the Safe Country of Origin (SCO) concept under EU law. It follows the CV judgment, in which the Court ruled out territorial exceptions from the principle under the 2013 Recast Asylum Procedures Directive (2013 APD). In Alace, the focus shifted to group exceptions, with the Court firmly stating that the 2013 APD precludes Member States (MSs) from designating a third country as “safe” if it does not satisfy the relevant safety conditions for certain categories of persons (para 109).

At first glance, this looks like a welcome confirmation of a restrictive reading of the SCO designation. The Court rejected the more politically pragmatic approach advocated by Advocate General (AG) De La Tour (para 72), who had argued for a broader margin of discretion for Member States in designating safe countries, including group-based exceptions (para 70). But what does this mean in the long run – particularly in the context of the 2024 Asylum Procedures Regulation (2024 APR), and recent Commission proposals to bring some of its provisions into force ahead of schedule?

A Quiet Court in a Noisy Debate

The Alace judgment concerns two Bangladeshi nationals who were intercepted at sea on their way to Europe by the Italian coastguard and detained in Gjadër, Albania, a reception centre outside Italy’s borders but under Italian jurisdiction per the Italy-Albania protocol.

After their applications for international protection were rejected on the basis that Bangladesh was considered an SCO, the applicants appealed. The Tribunale ordinario di Roma referred four preliminary questions to the CJEU:

1. Whether EU law precludes a Member State from designating a third country as an SCO by a legislative act of primary law;
2. Whether EU law requires the sources justifying the designation to be “accessible and verifiable”;
3. Whether national courts may rely on sources beyond those listed in Article 37 of the 2013 APD when reviewing a country’s “safety” designation;
4. Whether Member States may designate a third country as “safe” where the safety conditions are not met for specific groups.

This piece focuses on the fourth question.

When addressing it, the Court clearly stated that group exceptions are not permitted under the 2013 APD. Meaning that a country must be safe for the entirety of its inhabitants not just part of it (para 96). Its reasoning relied on existing case law, the legislative context, and the objective of the recast. While the Court did not directly refer to the 2009 legislative procedure, Article 33 of the Proposal for the 2013 APD made clear the intention to remove the possibility for both territorial and group exceptions.

Instead, the Court relied on a textual interpretation of the Directive. It highlighted the terms “country” and “third country” in Article 37, noting that these offer no indication that a designation could refer only to part of a population (para. 92). The Court even examined the wording across all official EU languages to explore how the terms “generally” and “uniformly” – which describe the required absence of persecution in Annex I – support this restrictive reading (paras 93–95).

Thus, the Grand Chamber confirmed the position taken in CV: the 2013 APD does not allow for group or territorial exceptions in SCO designations.

The Shadow of the New Pact: Legal Clarity with an Expiry Date?

AG De La Tour’s Opinion diverged significantly from the Court’s approach. He argued that inconsistencies between the language versions of the Directive allow for a more flexible interpretation, leaving space for Member States to apply group-based exceptions when determining whether a country is “safe” (paras 76-77).

While the Court declined to follow this line of interpretation, it has been clear for some time now that EU asylum and migration law has served as a tool of migration control at the expense of rights, and the CJEU has been noticeably reluctant to make waves (see Tsourdi and Costello). AG De La Tour’s Opinion makes this tension explicit – he calls on the Court to consider “the pressures that national asylum systems are currently under” and suggests a need to balance legal and non-legal considerations (para 72). Therefore, with the Alace judgment, countries like the Netherlands and Italy that possessed group exceptions within their SCO lists would need to re-adjust them.

The problem with treating Alace as progress is that it comes at a transitional moment. In 2026, the 2013 APD will be repealed and replaced by the 2024 Asylum Procedure Regulation (2024 APR), part of the New Pact on Migration and Asylum. Crucially, the 2024 APR explicitly reintroduces both territorial and group exceptions for SCO designation – effectively reversing the approach confirmed in CV and Alace.

Both referring courts in Alace and (the pending case of) Oguta made reference to the forthcoming APR, seemingly inviting the CJEU to weigh in on its provisions. But the Court has so far refrained from doing so, merely stating that the legislator had the ‘prerogative’ to ‘reconsider that choice, by striking a fresh balance’ (Alace (para 106) and CV (para 82)).

With the Commission now proposing to accelerate the implementation of parts of the APR, amongst others, the designation of SCO with exceptions,  into force before the planned date of June 2026, the practical value of the Alace judgment may already be fading.

The Future of the SCO Practice

Neither the 2005 APD nor the 2013 APD truly operated in favour of people on the move. Despite the exclusion of territorial and group exceptions in the 2013 recast, these practices have persisted. Moreover, the material conditions for designating a country as “safe” have never been immune to political considerations. Although SCO lists are, in theory, subject to review by the Commission and judicial authorities, such safeguards offer little assurance that no one will fall through the cracks of the system.

AG De La Tour in his Alace Opinion, reaffirmed that Member States must identify at-risk categories when assessing a third country’s general situation, in order to explicitly exclude them from the application of the SCO concept and its associated presumption of safety. However, as the 2005 APD already acknowledged, the very act of labelling a country “safe” is politically significant (para 19). There is no guarantee that governments will identify all at-risk groups nor refrain from designating a country as “safe” where political motivations, such as enlargement or migration control priorities, may dictate otherwise.

Problems with the SCO practices become even further highlighted in the 2024 APR, which reintroduces both territorial and group exceptions but without clearly defining or limiting their scope. This raises a critical question: can Member States designate as many categories of people as they wish while still maintaining the fiction of general safety? AG De La Tour cautioned that too many exceptions may render the safety designation meaningless: if a country is unsafe for many, can it be presumed safe at all?

This line of reasoning lies at the heart of the pending Oguta case, in which the Tribunale di Firenze (Ordinary Court of Florence) asked whether the Italian government’s identification of ten categories of vulnerable individuals undermines the designation of Nigeria as a safe country of origin. Under the Alace judgment, the issue in Oguta arguably becomes moot for the purposes of the 2013 APD, which (as interpreted by the CJEU) does not permit group exceptions. Yet, if the Court issues its decision after the 2024 APR enters into force – or even if it does not issue a decision in this particular case – the question will remain relevant, as the APR allows for such exceptions. Should the Court follow AG De La Tour’s reasoning (para 91), it may conclude that Nigeria’s designation cannot stand. Nonetheless, the APR’s vague treatment of exceptions leaves ample room for extensive lists to persist.

Going back to the CV case, AG Emiliou, in his Opinion, noted that both territorial and group exceptions are legally complex and insufficiently defined. He focused particularly on territorial exceptions, raising the question of how to delineate which parts of a country are “safe enough” and which are not (paras 101 and 109). But similar challenges arise with group exceptions, especially when they concern marginalised communities such as ethnic, racial, or sexual minorities. For instance, queer asylum seekers are routinely subjected to credibility assessments that require them to “prove” their queerness. As noted in the Oguta reference (para 9c), if national authorities already struggle to determine group membership under ordinary procedures, how can such assessments be conducted fairly and accurately under accelerated ones?

These legal ambiguities are not merely technical; they expose people on the move to arbitrary, rushed, and potentially life-threatening decisions. The continued reliance on the SCO concept, especially as reformulated in the 2024 APR, entrenches a system that privileges administrative convenience and political expediency over genuine protection needs.

Conclusion

In light of what is to come with the 2024 APR, Alace is not progress but a temporary pause. While the Court has reaffirmed a narrow reading of the 2013 APD, this version of EU asylum law will soon be obsolete – and with it, the legal clarity Alace provides. Yet returning to the 2013 framework is no solution either. Both the Directive and the forthcoming Regulation rest on the flawed logic of the Safe Country of Origin concept itself. If protection is to be meaningful, and procedures truly fair, the time has come not to refine the SCO practice – but to abolish it altogether.