Return hubs – innovative lawmaking or a dangerous legal experiment?

 

 

By Jonas Bornemann, Assistant Professor of European Law at Rijksuniversiteit Groningen and re:constitution fellow 2024/2025 and Isabela Brockmann, Research intern at the Department of European and Economic Law, Rijksuniversiteit Groningen.

 

Photo credit: Konstantin von Wedelstaedt, via Wikimedia commons

 

1.     Legislative creativity: thinking outside the (legal) box?

 

In lawmaking, creativity is an asset. The ability to devise innovative solutions can break deadlock and help align previously irreconcilable positions. But even the most innovative idea must stay within the limits of legality. This is particularly true for sensitive areas like migration law, where ‘out-of-the-box’ thinking has recently become fashionable among political leaders in Europe to signal their determination to pursue a more restrictive migration policy.

 

The most recent illustration of such an out-of-the-box mentality figured in the Commission’s reform of the Return Directive. Among the changes proposed, the Commission seeks to introduce an option for Member States to create so-called return hubs – centres established in third states to facilitate the return of persons who are not or no longer authorised to stay in the territory of EU Member States. The term ‘return’ is used here, following the terminology used by the Commission, although it should be borne in mind that persons send to return hubs are usually expelled to places they have never been to before. Speaking of ‘return’ in such a context may therefore already be a euphemism. Be that as it may, the Commission’s proposal has been presented and hailed as an innovative way to increase the effectiveness of returns, offering an avenue of deporting persons from the territory of EU Member States who cannot, for practical or legal reasons, be returned to their country of origin. However, the fate of this innovative instrument is far from certain. Even if it would ultimately be accepted by the co-legislatures, several important aspects regarding the implementation of the proposed policy remain unclear. This blog post discusses the potential limits in EU primary law that circumscribe the way that return hubs can be established and run, and proposes refinements to the legal design of the currently discussed reform of the Return Regulation.

 

The post will do so by, first, placing the idea of return hubs in the broader scheme of initiatives that seek to offshore migration administration (2.). It will subsequently present the legal basis for the establishment of return hubs as proposed by the Commission (3.), before zooming in on the options regarding the practical implementation of these hubs by the Member States. Specifically, it will discuss possible limits to the geographical location of these hubs (4.), the risk of systematised detention (5.) and the paramount importance of an effective remedy (6.). It concludes by drawing attention to aspects that should be addressed during the legislative process, to circumscribe, with greater clarity, the mandate of Member States to establish and use return hubs (7.).

2. Return hubs – a piece in the puzzle of offshoring migration administration

 

The idea of return hubs is closely linked to related attempts of offshoring migration administration. The Rwanda scheme, for instance, attempted to outsource asylum examinations through the adoption of a bilateral agreement between Rwanda and the UK that would have enabled the latter to send asylum-seekers to Rwanda to have their claims processed by Rwandan officials. However, following successful legal challenges before the UK Supreme Court and a new government which subsequently repealed the Safety of Rwanda Act, the Rwanda scheme was  abandoned and now largely serves as a cautionary tale for efforts to externalise migration administration. The (voluntary) return of four asylum-seekers pales in contrast to the extreme financial costs of the scheme, estimated to amount to £700m. Costs included upfront payments to Rwanda, asylum processing and operational costs and funding for an integration package covering five years if the individual decided to stay. The results, however, remained extremely meagre.

 

More recently, the Italy-Albania deal followed a similar but not identical template. The deal took the format of a protocol concluded between the two states to allow for the transfer of asylum seekers to Albanian asylum facilities. This measure applied exclusively to persons intercepted in international waters, and would have allowed Italian authorities to process claims under Italian law, although physical access to Italian territory is prevented. After transfer to these newly established facilities had been halted by Italian courts, and legal challenges are currently pending before the Court of Justice, the government decided to ‘reactivate’ the now moot asylum facilities, turning them into so-called ‘repatriation hubs’. A similar initiative was announced by the British Prime Minister during his visit in Albania: the UK would establish ‘return hubs’ in Albania to facilitate return of persons whose asylum application had been rejected.

 

All this suggests that there is a functional connection between the attempts to offshore asylum processing and the emergence of ‘return hubs’, even though the categories of persons targeted by these initiatives differ. Return hubs have no role to play in the context of asylum processing, but rather during the return of third country nationals that are not or no longer authorised to stay in the territory of the Member States. This may relate to third country nationals who are not or no longer allowed to stay and cannot be returned to their state of origin. Return hubs could therefore equally be used to return persons whose application for international protection had been rejected. With a view to this group of persons, however, it should be borne in mind that the Commission’s recently proposed changes to the Asylum Procedures Regulation could give rise to situations where the third state in which a return hub is established will be designated as ‘safe third county’ and that would-be beneficiaries of protection would be returned to that third state without their case being assessed on the merits in the EU.

 

3.     Return hubs as a Member State project

 

Despite the fact that several Member States are currently considering the use of return hubs, and the Commission’s proposal would adjust the legal framework in EU law accordingly, the design – both in law and fact – of these hubs remains far from clear. The Commission’s proposal indicates that persons may be returned to a third country with which an agreement has been concluded, thereby effectively creating a legal basis in EU law for such agreements (see here at 6). However, the proposal only vaguely predetermines the elements that such an agreement should satisfy, thereby backloading potential legal problems to Member States’ implementation of return hubs.

 

First of all, the proposal suggests that return hubs may be established on the basis of either an ‘agreement or arrangement’. The reference to ‘arrangements’ may be read as permitting Member States to resort to forms of cooperation other than formal international agreements. This wording may be inspired by the blueprint of the original UK-Rwanda deal, which took the form of an informal Memorandum of Understanding rather than an international agreement and was termed an ‘asylum partnership arrangement’. Informal arrangements, however, would likely fall short of the requirements outlined by the Commission’s proposal, specifically the obligation to ensure that the third state would be ready to accept the returnee (on this point, see here at 17). While an informal agreement may, in principle, likewise practically ensure that the third state accepts entry of returnees (here at 148), it is characterised by weaker normativity than formal international agreements and appears less suitable to ensure the willingness of third states to accept entry of returnees or ensure respect for safeguards of migrant protection. The importance of formal rules is acknowledged not just by the UNHCR (here, point 3. v), but likewise by the EU legislature, as Art. 59 (7) of the Asylum Procedures Regulation, seems to rest on the view that a formal international agreement adopted under the procedure of Article 218 TFEU would ensure full respect of the principle of non-refoulement.

 

The Commission’s proposal spells out several requirements that the international agreement adopted with third states should satisfy. An agreement establishing return hubs would have to outline the procedure for transfer, the conditions of stay in the third state, including the responsibilities of the Member State and third state respectively, modalities of onward return and the consequences if onward return would not be possible. In addition, the Commission’s proposal clarifies that unaccompanied minors and families with minors shall not be returned to a return hub and insists, moreover, that the agreement would include an independent monitoring mechanism to verify the effective application of the agreement. Last, the agreement must make provision for cases in which the arrangement would be violated or a significant change had occurred that would adversely impact the situation of the third country.

 

While the Commission’s proposal therefore lists certain elements that agreements establishing return hubs should satisfy, it is safe to say that the Commission wishes to ensure that the establishment of return hubs is a Member State project. For EU lawmakers, the choice for national solutions may seem preferable for several reasons: first, it accepts a level of diversity, to the effect that Member States remain free to decide whether they wish to include return hubs as an element of their return efforts. In addition, a legal design that simply requires agreements to respect essential safeguards, such as respect for international law and human rights, may be viewed as a means to keep one's hands clean. Any violation of these safeguards would be attributable to Member State authorities rather than EU institutions. In this sense, the Commission can present its proposal as conforming with standards of EU and international law, whilst leaving the implementation of return hubs, and associated legal challenges, to national authorities.

 

 

4.     Does EU law accept arbitrary geographical choices?

 

The establishment of return hubs is legally operationalised through a broadening of the concept of the ‘country of return’. Pursuant to Article 4 (3) of the Commission proposal, a country of return includes not only the country of origin or transit of a third country national, but has been extended to any third state willing to host a return hub and has entered into an agreement to that end with one or several Member States. Leaving the issue of political feasibility at the side, this raises the question whether Member States’ choice for the location of return hubs would be subject to any standards of legality, or whether this would essentially allow national governments to make an entirely arbitrary geographical choice. Are Member States allowed to return persons to hubs in third states that are extremely remote from both the EU and the country to which the person should ultimately be returned?

 

Current legal standards limiting the geographic stretch of the deportation of asylum-seekers may not be applied by analogy when it comes to the return of third country nationals who are no longer authorised to legally stay in the EU Member States. Indeed, the ‘connection criterion’, a safeguard against arbitrary deportation which has aptly been referred to as the ‘anti-Rwanda’ rule, stipulates that asylum seekers may only be deported to a safe third state to which (s)he has a connection ‘on the basis of which it would be reasonable for him or her to go to that country’ (Article 59 (5) Asylum Procedures Regulation). While such a criterion may not be mandatory under international law and the Commission has recently proposed to allow Member States to derogate from it, one of the innovative aspects of the Commission’s proposal concerns the fact that this safeguard does not apply mutatis mutandis to persons who are not or no longer asylum seekers. More generally, the connection criterion does not apply outside refugee law, and is therefore a ground that cannot be relied upon to limit a Member State’s geographical choice for setting up return hubs.

 

In the absence of a legal standard such as the connection criterion, Member States appear, in principle, to be free to return third country nationals to hubs located in states to which the individual has no personal link whatsoever. However, it is not unreasonable to argue that this principled freedom should be limited by general principles of EU law, specifically the principle of proportionality. While the proposed Return Regulation would leave significant room for maneuver to Member States, it would nonetheless create a legal framework governing the use of return hubs, to the effect that Member States would act within the scope of EU law (for a discussion of the case law, see here at 141). If this is a correct reading of the proposed Regulation, return to a return hub in a region far removed from the country of origin of the person and to which that person has no connection may be viewed as violating this principle. While the principle of proportionality would not preclude Member States from establishing return hubs in third states, it may limit their geographical choice, nonetheless. As UNHCR rightly points out, return hubs would only further the effective return of a person if such hubs are geographically located in places from which individuals may actually be able to travel onwards, either by virtue of visa-free or other forms of mobility regimes. If this would not be the case, return hubs might not be suitable to attain the objective of effective return.

 

5.     Conditions in return hubs – towards systematised detention?

 

The Commission’s proposal suggests that an international agreement setting up return hubs must clarify the conditions of stay in the third state. However, it remains silent about the nature of these conditions. This has given rise to criticism by NGOs and scholars (see here, here, here as well as here, here and here), suggesting that return hubs may incentivise practices of arbitrary detention. On the one hand, it is not inconceivable to argue that an international agreement may allow third country nationals deported to a return hub in a third country to move freely in that country. However, existing practices suggest that cooperation with third states will most likely take the form of restricted mobility or ‘semi-carceral spaces’ (for this apt description, here at 34). This raises questions regarding the respect for fundamental rights in return hubs, specifically the right to liberty. Depending on the length and nature of stay, mobility restrictions in return hubs in third states may amount to detention. If it is presumed that return hubs will not magically resolve the obstacles of onward return to countries of origin, it is entirely reasonable to presume that these hubs will frequently amount to a restriction of liberty which would coincide with significant habeas corpus safeguards, particularly also the obligation to provide for periodic review of detention and to respect maximum detention periods.

 

The presumption that return hubs will frequently be designed as closed facilities follows from political preferences rather than legal necessity. Governments willing to host return hubs might accept that returnees are staying short-term before they are returned to their country of origin, but they might be much less enthusiastic about the prospect of returnees establishing themselves more permanently in that country. There is no obligation for Member States to ensure that third country nationals would remain at or in the proximity of return hubs. While Member State authorities (or the European Border and Coast Guard, as the case may be) may remain responsible for bringing the return of a third country national to her country of origin to a successful end (see here, point 14), there is no provision in EU law that would require Member States to limit the mobility of third country nationals in the country in which the return hub is located. Rather, with a view to persons who cannot be returned in the short- or medium-term to their country of origin, UNHCR advocates for arrangements that minimise restrictions to movement, and that allow for pathways to self-sufficiency in the host third country.

 

6.     Could return hubs increase the effectiveness of the EU’s return policy?

 

The creation of return hubs is driven by the ambition to increase the effectiveness of return. Although the Commission’s proposal does not explain how the establishment of such hubs would help attain this objective, there seems to be an implicit assumption that a person who is returned to a country in which (s)he does not like to reside would be more easily convinced to return. This assumption is problematic, not just because of the lack of empirical data supporting it. Such a conclusion equally fails to acknowledge the variety of reasons why persons leave their country of origin, and ignores the fact that return is often hampered by lack of cooperation on the side of the third state, not the individual.

 

In addition to these factual uncertainties, the political objective of increasing the effectiveness may be criticised for being based on fair-weather presumptions. It presupposes that Member States would be able to ensure that returnees will not be exposed to inhuman or degrading treatment in the third state to which they have been transported. On the one hand, it can surely be welcomed that the Commission’s proposal explicitly reminds Member States of this obligation, requiring them to respect “international human rights standards and principles (…) including the principle of non-refoulement” (Article 17 (1) of the Commission proposal). On the other hand, NGOs, scholars and curiously, in a 2018 working paper, even the Commission itself, had warned against the significant risks of refoulement associated with the establishment of return hubs. For example, a third country may be declared prima facie safe, however, this may still fail to account for intolerance towards certain groups, such as LGBTQ+ individuals or religious minorities. A de facto risk of violating non-refoulement therefore remains.

 

In this context, it is crucial that individuals have the possibility to challenge their deportation to a return hub. National courts may find deportations to return hubs to be unlawful if this would amount to a violation of migrants’ fundamental rights. One of the key questions for the future of the idea of establishing return hubs will therefore revolve around the availability and nature of judicial redress. While the Commission’s proposal clarifies that individuals would benefit from a right to an effective remedy with a possibility of suspensive effect, a document drafted by the Council Presidency before the publication of the Commission’s proposal indicated that ‘the prevailing position’ in the Council was that ‘judicial scrutiny [...] could put the implementation of this innovative solution at risk’ and should therefore be precluded. This proposition is highly problematic, as it seems to be at odds with EU primary law, specifically the right to an effective remedy, proposing to sacrifice this crucial constitutional safeguard in the name of innovative lawmaking, thereby exacerbating the risk of refoulement that is inherent in the idea of offshoring return.

 

7.     The uncertain future of return hubs

 

Return hubs are one of the key innovations proposed by the Commission. Hopes may be high that this will enable Member States to increase the effectiveness of return policies, thus marking what is presented as a significant change in EU migration policy. As this blog post has argued, however, the fate of return hubs is far from certain. It is unclear whether the co-legislatures may be able to find common ground on this element of reform or whether the proposed legal basis for return hubs in EU law will be kept as it is, amended or abandoned altogether. Specifically, it should be worth critically reflecting on some of the underlying presumptions of this policy instrument: would it really persuade third country nationals to move back to their country of origin? And could Member States in practice enter into a cooperation with third states that ensures that courts will accept, in a significant number of cases, deportation?

 

Presuming there is a political majority for the idea of return hubs, based on the preceding analysis, the co-legislature may wish to consider some of the following aspects that may allow them to define, more clearly, the legal option of Member States to rely on these return hubs. First, the Return Regulation should clarify that such return hubs could only be established on the basis of formal international agreements. Second, return hubs would have to be established in a third state from which a person may, factually or legally, be able to move to her or his country of origin. Third, the EU legislature should oblige Member States to include, in their cooperation with third states, rules that clarify the legal position of third country nationals who cannot return to their country of origin. After a reasonable period of time, these individuals would have to be authorised to move freely within the territory of that third state, in order to avoid a practice of indefinite and systematic detention. Lastly, unlike views reportedly raised in the Council, ideas to drop or undermine the effectiveness of judicial review should be abolished. The right to an effective remedy is crucial in the context of deportation to a third state, and likewise a firmly enshrined constitutional guarantee. While judicial review may undermine the idea of effective return, this is surely a price worth paying. The fate and success of return hubs as an innovative policy instrument will depend on these safeguards. Should return hubs emerge as an innovative policy tool, its legal design needs to be waterproof.

 

 

Advocate General Ćapeta’s Opinion in WS and Others v Frontex before the Grand Chamber: The End of Frontex’s Shielding? Joint Liability of Frontex and Member States in Return Operations

 

Antje Kunst*

* Antje Kunst is an international lawyer and barrister of Garden Court North Chambers, admitted to the Bar of England and Wales, and the Bar of Berlin, advising and representing individuals in a wide range of matters related to fundamental rights within the CFSP and other fields. She has appeared in numerous cases before both the Court of Justice and the General Court, within the Court of Justice of the European Union.  

Photo credit: Asurnipal, via Wikimedia Commons

 

Introduction

On 12 June 2025 Advocate General (AG) Tamara Ćapeta, following a hearing by the CJEU Grand Chamber in February 2025, rendered an important opinion (‘Opinion’) on the appeal of WS and Others v. Frontex,  in Case C-679/23 P against the General Court’s judgment of 6 September 2023.  The General Court’s judgment has been rightly heavily criticised (see for example here, and here). Two authors went as far as stating that the shielding Frontex from meaningful responsibility is not only harmful to victims of fundamental righta breaches committed by Frontex, but that it also undermines the institutional framework of EU agencies and affects the legal credibility of the Court (see here).

The clarity that was lacking in the General Court’s judgment is made up for by the AG, who addressed the issues directly and even offered two possible readings of what the General Court may have intended to find, including that it might have failed to attribute the unlawful conduct to Frontex. (see paras. 60-67 and 84 of the Opinion)

As outlined  here WS and Others v. Frontex involves a Syrian refugee family with four children, who are ethnic Kurds. They arrived on the Greek island of Milos in October 2016 by boat with the intention of claiming asylum. Even though they formally declared their interest in applying for international protection, just six days later, they were deported by flight to Turkey in a joint return operation conducted by Greece and Frontex. Because of the fear of being returned to Syria by the Turkish authorities (para. 59 of the GC’s judgment), the family decided to flee from Turkey to Iraq, where they reside to this day.

The applicants filed a claim for compensation, alleging that Frontex had violated its own obligations under the then-applicable Frontex Regulation (the 2016 Frontex Regulation) and several fundamental rights under the EU Charter during the return operation.

The significance of the case lies in the fact that, for the first time, the Court of Justice examines the joint liability of Frontex and Member States for breaches of fundamental rights arising from joint return operations. There is also an opportunity for the Court to develop its case-law on the causal link between the alleged unlawful conduct of an EU agency -specifically Frontex- and the damage suffered by individuals. This blog post focusses on these two issues.

 

Admissibility of the action

Any credible claim that fundamental rights have been violated deserves access to an effective judicial remedy. In cases like this, that remedy is a compensation claim under Article 340(2) TFEU to make good any damage caused by EU institutions, bodies and agencies.

Despite Frontex’s objections on four grounds, the AG rightly concluded that the claim is admissible (see paras. 42- 55 of the Opinion). This includes the General Court’s failure to assess relevant facts like the existence of a return decision which are arguments of points of law. If the Court of Justice agrees, which is expected, it would set an important precedent, opening the door for similar actions against Frontex to be heard in the future.

 

Frontex has its own obligation to verify under the 2016 Frontex Regulation

The AG makes it clear: Frontex has its own independent obligations under EU law when it takes part in joint return operations. That includes a duty to verify whether a valid return decision exists for each person put on the plane (paras. 68–83 of the Opinion).

Crucially, she stresses that Frontex -like all EU bodies- must uphold fundamental rights while carrying out its tasks. It is a legal obligation under Article 51(1) of the EU Charter of Fundamental Rights. So, when Article 34(1) of 2016 Frontex Regulation sets out Frontex’s duty, it is simply restating what EU law already requires (para. 72 of the Opinion).

The AG takes a close look at the 2016 Frontex Regulation and shows that Frontex’s responsibilities during return operations are clearly spelled out therein.

First, Article 28 of the Regulation ties Frontex’s operations directly to the 2008 EU Return Directive, which strictly prohibits returning individuals without an enforceable return decision. Second, Article 4(h) of the Regulation sets out the same point. This is supported by consistent case law from the Court of Justice. The idea is simple but crucial: the existence of return decision helps ensure that the return does not violate the principle of non-refoulement.

The AG makes an important -if obvious- point: the 2016 Frontex Regulation clearly intends that joint return operations coordinated or organised by Frontex must only involve individuals who are the subject of individual, enforceable return decisions.

She goes on to conclude that Frontex bears its own responsibility in this process- again an obvious point despite Frontex's consistent denials. By verifying that return decisions exist for everyone on board, Frontex is fulfilling its legal duty to ensure respect for fundamental rights, including the principle of non-refoulement, in the performance of its tasks, as required by Article 34(1) of the 2016 Frontex Regulation.

Consequently, the General Court erred in law that there could not be a causal link between the allegedly unlawful conduct of Frontex and the damage suffered ( see para. 83 of the Opinion)

 

Joint and several liability of Frontex and the host Member State?

In paragraphs 84–97 of the Opinion, the AG observes that the General Court may have erred in finding that Frontex and a Member State could not be held jointly and severally liable for the same damage. She clarifies the concept of joint and several liability: when more than one actor contributes to a single harm, each may be held fully liable, and the injured party has the right to claim compensation from any one of them for the entire damage (see para. 85 of the Opinion). The issue of shared liability is a much-debated topic in legal scholarship (e.g., here). Can both Frontex and a Member State be liable for separate breaches of EU law that give rise to a single harm? According to the AG, the answer is yes.

She goes on to clarify that under the EU’s integrated border management system, Member States can be held liable for any action or omission. Frontex, on the other hand, is accountable only for actions or failures that fall within the scope of its assigned tasks under EU law (para. 88 of the Opinion).

She offers an important clarification regarding the Court’s earlier ruling in Kočner v Europol (para. 90 of the Opinion), which concerns joint liability of an EU agency and Member States: that case did not establish a general principle that joint and several liability must be expressly provided for in EU legislation, as it was in Kočner.

In other words, joint and several liability can apply even when not explicitly laid down in legislation. At the same time, she distinguishes Kočner from the present case. In Kočner, it was not possible to attribute the unlawful conduct- namely, the data processing, to Europol or a Member State. By contrast, in the current case, she explains, it is possible to attribute the same omission - failing to ensure a valid return decision - to both Frontex and Greece (para. 91 of the Opinion).

She concludes correctly had either Frontex or the Member State conducted the necessary verification the Syrian family would have been excluded from the joint return operation, assuming there was no return decision. Consequently, responsibility for the damage in this case can be attributed to both, since either one had the capacity to prevent it.

 

Member State’s primary responsibility does not exclude Frontex’s liability

The AG rejected Frontex’s argument that the agency cannot be held liable simply because Member States bear primary responsibility for return procedures under the 2016 Regulation. She emphasizes that this cannot mean Frontex is absolved of liability for failures -here the omission to verify whether a valid return decision existed. She rightly pointed out that accepting such reasoning would effectively shield Frontex from accountability for unlawful actions or omissions during joint return operations. (para. 92 of the Opinion)

Ćapeta reaches this important conclusion: in situations where both Frontex and Member States share obligations in joint return operations, Frontex can be held liable for damage resulting from a breach of those obligations, even if a Member State may also be held liable for the same damage (para. 93 of the Opinion) This position should be adopted by the Court of Justice; otherwise, Frontex will continue to hide behind a Member State to evade liability.

 

Causal link - broken through own decisions?

The AG disagrees with the General Court’s conclusion that the harm suffered by the appellants -their forced relocation to Türkiye, flight to and eventual settlement in Iraq- was the result of their own choices breaking the causal chain, and not any action or omission by Frontex. She finds this correctly to be an error of law.

Regarding factual causality, the AG refers to the ‘but for’ test and to the General Court’s finding that the alleged unlawful conduct being a conditio sine qua non for the damage to arise is not sufficient to establish a causal link.  

The AG is of the view that the ‘but for test’ is fulfilled in the present case on the assumption that there was no return decision: had Frontex fulfilled its duty to verify whether a valid return decision existed, it might have discovered that none was in place. The appellants likely would not have been included in the joint return operation and would not have suffered the alleged damage. (see paras. 103 and 106 of the Opinion).  In this context the AG notes that, from an early stage, the Court’s case law has imposed limits on the ‘but for’ test when assessing liability, particularly by excluding the liability of EU institutions or bodies for damage deemed too remote (so-called legal causality) (para. 109 of the Opinion).

As set out here, in the present case, however, the damage is not too remote: Frontex’s operation constitutes the most proximate and direct cause of the harm suffered by the family.

As the AG explains, a finding of damage of being too remote often comes into play when others, sometimes even the injured party, have contributed to the harm (paras. 110-112 of the Opinion). In such cases, the causal chain may be considered broken, and liability may not attach. The AG refers to past case law where the injured party’s own actions or decisions were seen as breaking the chain of causation, predominantly in market cases where the injured parties were economic operators. But in this case this reasoning must not apply (paras. 114 to 116 of the Opinion).

A Syrian Kurdish family - two parents and their four children - fled Syria at the height of the war. What followed was a serious breach of their fundamental rights, most notably the principle of non-refoulement. As the AG points out, the family was undeniably in a position of vulnerability.

After being returned to Turkey, they eventually made their way to Iraq - a place they considered safer. While that move may have technically been their own decision, it cannot be called a ‘free choice’. As AG Ćapeta puts it, it was a ‘difficult life-choice decision’.

On that basis, the Court of Justice should follow Ćapeta’s reasoning and find that the factual chain of causation was not broken by the appellants’ actions. Without the unlawful omission by Frontex, they would never have been forced to make such a difficult decision in the first place.

 

Conclusion: The Shielding of Frontex Should End

The AG’s Opinion should mark the beginning of the end of Frontex’s shielding from accountability in EU joint return operations. It is now up to the Court of Justice to carry this principled stance forward. Based on the questions raised by the judges during the oral hearing, there is a strong indication that the Court will overturn the General Court’s judgment - though perhaps not entirely.

The Court may well agree with the AG that a final ruling is not yet possible and that the case should be referred back to the General Court for a fuller factual assessment. Nonetheless, a clear and thorough judgment will be essential - not only to guide the outcome of this case, but also to provide legal clarity for similar future cases. What emerges already from the facts presented and the Court’s existing case law is this: the EU judiciary must eventually take the necessary step of establishing Frontex’s non-contractual liability for the unlawful deportation of the family and award appropriate damages.

European Return Orders and the European Human Rights Convention: The Commission’s Proposal for a Return Regulation

 

 

Professor Elspeth Guild, University of Liverpool

 

Photo credit: Marcelx42, via Wikimedia Commons

 

In March 2025 the European Commission published a proposal for a re-cast return regulation. It has been met with mixed responses, a certain jubilation in some parts of Europe’s extreme right, a certain concern not least from a human rights perspective in other quarters. For an excellent analysis of this proposal see Majcher’s blog on this site. One of the innovations in the proposal is a European Return Order which any Member State may enforce without issuing its own expulsion decision (Articles 7-9). (‘Return’ is the EU term for ‘expulsion’ which is the term used in the European Convention on Human Rights and the International Covenant on Civil and Political Rights. I will use the term ‘expulsion’, consistent with European and international human rights law).

 

In this blog I will examine the human rights problems which Member States are likely to encounter where they seek to expel someone on the basis of a European Return Order issued by another Member State. I will focus in particular on the case law of the European Court of Human Rights regarding the human rights assessments which states must carry out before expelling a person. These requirements were recently reviewed, expanded and confirmed by the ECtHR in its 4 February 2025 judgment in A.B. and Y.W. v Malta in the context of Article 3 ECHR (the prohibition on torture, inhuman or degrading treatment or punishment). I will also examine the duty on states to carry out a human rights assessment in the context of Article 8 (the duty to respect private and family life). Although this is a qualified right, states must justify on grounds of necessity and proportionality any interference with it. The facts of each situation need to be considered and weighed against the interest of the person.

 

The European Return Order

 

According to the Commission’s explanatory memorandum, the European Return Order (ERO) will not replace each Member State’s own return decisions. Instead, the ERO will extend the scope of application of Member States' return decisions in order to make them actionable across the Union by Member States other than the one which made the order. The ERO will be in a common form which includes the key elements of the national expulsion decision. It will be made available through the Schengen Information System (the EU information sharing system for security and border management). In so far as technical changes may be required for this purpose, they will be addressed in amendments to SIS rules.

 

An implementing act will make it mandatory for all Member States to recognise and enforce an expulsion (return) decision issued by another Member State in the form of the ERO (with limited derogations). Among those derogations is a provision (Article 9(4)) which allows Member States not to recognise or enforce the expulsion decision of another Member State on ground that it is manifestly contrary to public policy. In EU law public policy does not expressly include fundamental or human rights grounds, although the similarly worded public policy derogation in the Brussels Regulation on the mutual recognition of civil judgments does include manifest human rights breaches, according to the case law of the CJEU (see the judgment in Real Madrid, for example).

 

More relevant may be Article 5 of the proposed regulation, which requires that Member States implement the regulation in compliance with the principle of non-refoulement, and with fundamental rights including the Charter, with relevant international law. However, there is no indication of how a Member State is to reconcile the duty to recognise and enforce expulsion orders made by other Member States with the Article 5 duties.

 

Again, according to the Commission, the Member State that issues an expulsion decision remains responsible for its enforcement. It must use all appropriate means to ensure effective expulsion of the person named therein. The new proposal is intended to be a mechanism for recognition of expulsion decisions issued by another Member State, underpinned and facilitated by the ERO. The explanatory memorandum states that the proposal clarifies the process and relevant procedural safeguards and remedies based on the lessons learnt from the implementation of the 2023 Commission Recommendation on mutual recognition of return decisions. 

 

The objective of mutual recognition of expulsion decisions among the Member States has long been on the EU agenda. In 2001, a directive (2001/40) was adopted to achieve this end. As noted by Peers, that directive does not apply to the process or substance of the decision and permits expulsion by another Member State as an option not an obligation. Under that directive, enforcement takes place under national law which must include the Returns Directive 2008/115 (which will be repealed if the proposed regulation is adopted in the current form). Thus, the main purpose of the 2001 directive appears to be to avoid a state having to send an irregularly present migrant back to the Member State which issued the expulsion decision. A system for financial compensation was adopted in 2004. The criticisms which Peers makes of the directive include: its application to persons with rights under international agreements agreed by the EU, inappropriate extension of Schengen (and the SIS II rules) to persons already present on the territory, inadequate application of human rights standards, opacity as regards as regards costs and inconsistency with the Returns Directive.

 

As for the Commission’s recommendation on mutual recognition of expulsion decisions, the Commission claimed that this would facilitate and accelerate the return processes for the Member State responsible for return and enhance cooperation and mutual trust between Member States by further increasing convergence between Member States on managing migration. This is a rather large claim as will be explained below. In particular, the Commission claimed that obstacles to cooperation and communication among national authorities responsible for the asylum and the return procedures represent a key structural challenge for a more efficient expulsion processes. This logic underpins the current proposal for an ERO.

 

The Human Rights Obstacles to EROs

 

The key problem, however, with mutual recognition of expulsion decisions among Member States is the duty in international and European human rights law that a state seeking to expel an individual must carry out an assessment which consists of two parts which are not necessarily part of the same proceedings depending on the national law of the state. First there must be an ex nunc rigorous assessment of any risk which the person might face if returned to the proposed country, for the purposes of ensuring that the prohibition on refoulement is fully complied with (para 72 A.B. and Y.W. v Malta). This is required by Article 3 ECHR and the subject of substantial caselaw. Secondly, there must be a full assessment of the proportionality of the expulsion decision in light of the right to respect for his or her private and family life (Article 8 ECHR). It is for the state to show that the decision is proportionate and the threat sufficient to outweigh the individual’s right to respect for his or her private or family life.

 

Assessing the risk of refoulement

 

On the first ground, the prohibition on refoulement, it is for the person to make the claim. But thereafter, the risk assessment must focus on the foreseeable consequences of the applicant’s removal to the country of destination, in the light of the general situation there and of his or her personal circumstances. It must be considered whether, having regard to all the circumstances of the case, substantial grounds have been shown for believing that the person concerned, if returned, would face a real risk of being subjected to treatment contrary to Article 3 ECHR. If the existence of such a risk is established, the applicant’s removal would necessarily breach Article 3, regardless of whether the risk emanates from a general situation of violence, a personal characteristic of the applicant, or a combination of the two (para 61 A.B. and Y.W. v Malta).

 

As regards the distribution of the burden of proof, the ECtHR has clarified that there is a shared duty between the applicant and the immigration authorities to ascertain and evaluate all relevant facts in asylum proceedings. This includes a procedural obligation (part of the Article 3 duty) to assess the risk before expelling the individual. There are two main consequences of the Article 3 ECHR requirements. First, the authorities of a Member State executing an expulsion order (or an ERO) must be satisfied that the expulsion will not constitute refoulement. This will have to be on the basis of its own assessment not that of another Member State as the difference in assessment of risk in countries of origin by different Member States vary widely (as regards the same country of origin and where the risk is the same: generalised violence or civil war). Thus, an assessment of a country of origin by one Member State may result in granting protection while the assessment of the same country by another Member State may result in no protection and the issue of an expulsion order. Secondly, the assessment must be ex nunc and carried out in close temporal proximity to the proposed date of expulsion. Where there is an extended delay between adopting an expulsion order and its proposed execution, the state must carry out a new assessment to ensure that the risk remains below the threshold required by Article 3. All assessments must be rigorous.

 

Assessing the right to respect for private and family life

 

The second human rights ground which may defeat an expulsion decision is the right to respect for the person’s private and family life (Article 8 ECHR). Expulsion will always be an interference with private and family life (there is no need for expulsion where the person wishes and chooses to leave the country thus exercising his or her own choice regarding private and family life). The grounds on which a state can interfere with the right are set out in Article 8(2) and in all cases must be in accordance with the law and necessary in a democratic society. The grounds are limited to the interests of national security, public safety or the economic well-being of the country, for the prevention of disorder or crime, for the protection of health or morals, or for the protection of the rights and freedoms of others. The ECtHR has elaborated an extensive list of elements which must be considered when assessing the proportionality of the state’s claim to interfere with the right which include: in all cases, the length of the applicant’s stay in the country from which he or she is to be expelled and the solidity of his, her or their social, cultural and family ties with the host country and with the country of destination.

 

In the case of expulsion on grounds of criminal conviction, the nature and seriousness of the offence committed by the person in the expelling state and the time elapsed since the offence was committed and the person’s conduct during that period must be assessed. Where family life is at issue, the assessment must include consideration of the nationalities of the person and any family members concerned, the family situation, such as the length of marriage, and other factors expressing the effectiveness of a couple´s family life, whether or not the spouse knew about the offence at the time when he or she entered into a family relationship, whether or not there are children of the marriage and, if so, their age(s), the seriousness of the difficulties that the spouse is likely to encounter in the country to which the applicant is to be expelled, the best interests and well-being of any children involved, in particular any difficulties they would encounter if they had to follow the person to the country to which he or she is to be expelled.

 

The assessment of the person’s entitlement to respect for privacy and family life in the context of the state’s claim to the necessity and proportionality of an interference with it must take place when the actual expulsion of the individual is planned to take place. It must be up to date. Where the state knew (or ought to have known) of the existence of the person’s relevant private and/or family life in accordance with Article 8, it is under a duty to undertake a thorough assessment, subject to procedural safeguards including a right of appeal.

 

In the context of an ERO, it is difficult to see how an expelling state which is other than that where the person’s private and family life is situated will be in a position to make such an assessment. Reliance on an assessment made in another state may be insufficient depending on the temporal relationship of the assessment and the proposed expulsion.

 

Similarly, while the objective is to ensure that persons named in an expulsion order by any Member State are expelled as quickly as possible to a third country or another Member State, it is not clear that this will be rapid. Nor is it self-evident that a person will not have developed family life in the expelling Member State. Such family life could have commenced in the issuing Member State then continued in that Member State which is responsible for the expulsion. There are myriad circumstances which occur in practice which blur the lines of where family and private life have developed and been exercised. It is worth remembering that in EU law as regards the expulsion of a third country national from one Member State who is in possession of a residence permit in another Member State, the CJEU has held that the person must be permitted to return the Member State which issued the residence permit and only expelled to a third country where the person refuses to do so. This relieves the Member State where the person is present from the obligation to carry out a human rights compliant private or family life assessment.

Conclusion

 

The Commission’s proposal for an ERO to facilitate the expulsion of third country nationals from the EU by requiring a Member State other than that which has taken the decision to expel the person raises substantial issues as regards human rights. International and European human rights require states to make an up to date risk assessment before expulsion where a person has claimed a fear of persecution, of the application of the death penalty or is otherwise at risk of deprivation of life, torture, inhuman or degrading treatment or punishment or arbitrary disappearance. This duty will apply to the expelling Member States in spite of the fact that the issuing state may have carried out such an assessment. This is particularly so where the original assessment is out of date or where there are substantial differences in protection rates for persons from the same country and in generally similar circumstances in the issuing state and the state which is to carry out the expulsion. The expelling state may also be required to carry out an assessment of the necessity, legality and proportionality of the expulsion decision in light of the person’s right to respect for their private and family life (Article 8 ECHR). In many cases, the elements of private and/or family life will be in the issuing state. The expelling state is likely to have difficulties as regards accessing of such elements and in the event that its assessment is that the interference is not proportionate, the fate of the person will be ambiguous. The required assessments may also be hampered by lack of access to and translation of documents. While Article 9(4) of the proposal permits a Member State to decide not to recognise another Member State’s expulsion order, this must be on the ground that to do so is manifestly contrary to public policy, and the proposal does not explicitly confirm that fundamental and human rights could be part of a public policy assessment.  

 

In sum, this proposal does not appear likely to speed up human rights’ compliant expulsions. Rather it is likely to result in lengthy procedures fraught with cross border problems for both states. A better approach to the issue of persons subject to expulsion decisions who move from one Member State to another is to facilitate their return to the first Member State. However, if the person claims a fear that the first Member State will expel them contrary to the prohibition on non-refoulement, the expelling state cannot simply disregard the claim without breaching its own human rights obligations.

 

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.